text
stringlengths 14
5.77M
| meta
dict | __index_level_0__
int64 0
9.97k
⌀ |
|---|---|---|
Q: cakephp 3.0 query data manipulation? I'm trying to learn cakephp 3 and the ORM functions, wicht is great so far. But know I'm comming to a point on wich I'm not certain how I can aproach it in the best way, so I was hoping that somebody can tell what is the best way.
I'm using the query builder to load one or more products. In the data that's loaded I have one field called price. This is the main price of the product, but there can be an optional discount for an user. I know wich user is logged in to the system so I have an variabele witch contains his discount, for example 1.20 (=20%).
After the query has been fired I could do an foreach and recalculate the price before sending is to the view, but because of the query builder function I suspect that I can do it there before I fired the query.
Below an example where an search is done on input name. The field price is now standard, but should be recalculated with the discount.This could be an foreacht example:
$price = round((($price/$User->discount)-$shipping),2);
SearchController:
public function search()
{
if ($this->request->is('post')) {
//$this->request->data);
$options = array();
$options['query'] = $this->request->data['Search']['query'];
$products = TableRegistry::get('Products');
$query = $products->find('search', $options);
$number = $query->count();
$products = $query->All();
$this->set(compact('products'));
$this->set('number_of_results',$number);
}
The ProductsTable:
public function findSearch(Query $query, array $options)
{
$query
->where([
'name LIKE' => '%'.$options['query'].'%',
"active" => 1
])
->contain(['ProductsImages'])
->select(['id','name','price','delivery_time','ProductsImages.image_url'])
;
return $query;
}
Is there an way to implement this recalculation in the query builder? I tried to find some info on the web but there isn't much information yet about the ORM options. I was thinking about maybe ->combine. Hopefully someone wants to put me in the right direction.
Thanks in forward.
Changed the controller function to:
public function search()
{
if ($this->request->is('post')) {
$options = array();
$options['query'] = $this->request->data['Search']['query'];
$discount = 2;
$products = TableRegistry::get('Products');
$query = $products
->find('search', $options);
$query->formatResults(function (\Cake\Datasource\ResultSetInterface $products) {
return $products->map(function ($row) {
$row['price'] = ($row['price'] + 10);
return $row;
});
});
$number = $query->count();
$products = $query->All();
$this->set(compact('products'));
$this->set('number_of_results',$number);
}
}
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 7,798
|
Q: pySerial can send data time to arduino I am trying to send date time to arduino using Python with serial communication, but I am having trouble sending the data. When I use var = raw_input ("input:")
the data is sent. However when I use var = str (time.asctime (time.localtime (time.time ()))) the data not sent to arduino.
Here's my Python code:
import serial, time
port = serial.Serial('COM4',9600)
var = str(time.asctime(time.localtime(time.time())))
if port.isOpen():
print ('Port Aktif')
while 1:
port.write(var)
time.sleep(1)
print port.readline()
else:
print 'port Tidak Aktif'
my code arduino :
String msg ="";
void setup() {
Serial.begin(9600); // set the baud rate
}
void loop() {
if(Serial.available() > 0){
while (Serial.available()>0){
msg += char(Serial.read());
delay(30);
}
Serial.println(msg);
}
}
A: see https://pyserial.readthedocs.io/en/latest/pyserial_api.html#classes
it says:
write(data)
[...]
Write the bytes data to the port.
This should be of type bytes (or compatible such as bytearray or memoryview).
Unicode strings must be encoded (e.g. 'hello'.encode('utf-8').
This means, use
var = bytes(time.asctime(time.localtime(time.time())).encode('utf-8'))
A: Instead of
In [1]: import time
In [2]: str(time.asctime(time.localtime(time.time())))
Out[2]: 'Sat May 6 14:39:17 2017'
use:
In [3]: time.asctime(time.localtime()).encode('utf-8')
Out[3]: b'Sat May 6 14:39:17 2017'
The latter will return a string of bytes, which is required when writing to a serial port.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 1,683
|
\section{Introduction}
In molecular dynamics, the overdamped Langevin equation in~$\R^d$ is often used for modeling the behavior of a large set of particles in a high friction regime. It is given by
\begin{equation}
\label{equation:Langevin}
dX(t)=f(X(t))dt +\sigma dW(t),\quad X(0)=X_0,
\end{equation}
where~$f$ is a smooth Lipschitz function (typically of the form~$f=-\nabla V$ for~$V$ a smooth potential),~$\sigma>0$ is a constant scalar, and~$W$ is a standard~$d$-dimensional Brownian motion in~$\R^d$ on a probability space equipped with a filtration~$(\FF_t)$ and fulfilling the usual assumptions.
If the particles are subject to smooth constraints~$\zeta\colon\R^d\rightarrow\R^q$, such as strong covalent bonds between atoms or fixed angles in molecules, the dynamics follow the constrained overdamped Langevin equation
\begin{equation}
\label{equation:projected_Langevin}
dX^0(t)=\Pi_\MM(X^0(t)) f(X^0(t))dt +\sigma \Pi_\MM(X^0(t)) \circ dW(t),\quad X^0(0)=X_0\in \MM,
\end{equation}
where the solution lies on the manifold~$\MM=\{x\in\R^d, \zeta(x)=0\}$ with codimension~$q$ thanks to~$\Pi_\MM\colon\R^d\rightarrow \R^{d\times d}$, the orthogonal projection on the tangent bundle of the manifold~$\MM$.
In physical applications, constrained systems are often used as a limit model for stiff equations. For instance, in the dynamics of a diatomic molecule, the distance between the two atoms oscillates around an average length, called the bond length (see, for instance,~\cite[Sect.\thinspace 1.2.1]{Lelievre10fec} on the interactions of particles). One can work with a simpler constrained dynamics where the distance between the atoms is fixed as a constraint, or with the original (possibly stiff) dynamics in~$\R^d$. We refer the reader to~\cite{Plechac09imm,Lelievre10fec}, and references therein, for discussions on the use of constraints and penalizations in molecular dynamics.
For overdamped Langevin dynamics~\eqref{equation:Langevin}, choosing a function~$\widetilde{f}=-\nabla \widetilde{V}$ with the potential
$$\widetilde{V}=V-\frac{\sigma^2}{4}\ln(\det(G))+\frac{1}{2\varepsilon}\abs{\zeta}^2$$
gives penalized Langevin dynamics~\cite{Ciccotti08pod} of the form
\begin{equation}
\label{equation:modified_Langevin_epsilon}
dY^\varepsilon(t)=f(Y^\varepsilon(t)) dt +\sigma dW(t) +\frac{\sigma^2}{4}\nabla\ln(\det(G))(Y^\varepsilon(t))dt -\frac{1}{\varepsilon}(g \zeta)(Y^\varepsilon(t))dt,
\end{equation}
where we fix~$Y^\varepsilon(0)=X_0$, the parameter~$\varepsilon>0$ is fixed with arbitrary size,~$f=-\nabla V$,~$g=\nabla \zeta\colon\R^d\rightarrow\R^{d\times q}$, and~$G=g^T g\colon \R^d\rightarrow\R^{q\times q}$ is the Gram matrix.
It was shown in~\cite[Appx.\thinspace C]{Ciccotti08pod} that the solution~$Y^\varepsilon$ of~\eqref{equation:modified_Langevin_epsilon} converges strongly to the solution~$X^0$ of the constrained dynamics~\eqref{equation:projected_Langevin} if~$X_0\in\MM$.
The additional term~$\frac{\sigma^2}{4}\nabla\ln(\det(G))$ is a correction term (called the Fixman correction) that is needed to obtain the convergence to the constrained dynamics~\eqref{equation:projected_Langevin} (see~\cite[Sect.\thinspace 3.2.3.4]{Lelievre10fec} and references therein).
Thus, for~$\varepsilon$ small, the trajectory of the solution of~\eqref{equation:modified_Langevin_epsilon} lies in the vicinity of the manifold~$\MM$. This penalization can also appear naturally when simulating Langevin dynamics with a stiff potential (see, for instance,~\cite[Sect.\thinspace 5.1]{Vilmart15pif}). One is then interested in numerical schemes that are robust with respect to the parameter~$\varepsilon$ and that lie on the manifold~$\MM$ in the limit~$\varepsilon\rightarrow 0$.
In this paper, we study the following similar penalized dynamics in~$\R^d$ to simulate trajectories in a vicinity of the manifold~$\MM$:
\begin{align}
\label{equation:Langevin_epsilon}
dX^\varepsilon(t)&=f(X^\varepsilon(t)) dt +\sigma dW(t) +\frac{\sigma^2}{4}\nabla\ln(\det(G))(X^\varepsilon(t))dt
-\frac{1}{\varepsilon}(g G^{-1}\zeta)(X^\varepsilon(t))dt,
\end{align}
where~$X^\varepsilon(0)=X_0$. It is a simpler version of~\eqref{equation:modified_Langevin_epsilon} that also evolves in a vicinity of the manifold~$\MM$ in the limit~$\varepsilon\rightarrow 0$.
One result of this paper is the strong convergence of the solution~$X^\varepsilon$ of~\eqref{equation:Langevin_epsilon} to the solution~$X^0$ of~\eqref{equation:projected_Langevin} if~$X_0\in\MM$.
We mention that in the deterministic setting, that is, when~$\sigma=0$, equation~\eqref{equation:Langevin_epsilon} is a singular perturbation problem, and it converges to a differential algebraic equation (DAE) of index two in the limit~$\varepsilon\rightarrow 0$ (see~\cite[Chaps.\thinspace VI-VII]{Hairer10sod}).
We propose in this article a method that is robust with respect to the parameter~$\varepsilon$ for solving equations of the form~\eqref{equation:Langevin_epsilon}, and we leave the creation of robust integrators for solving~\eqref{equation:modified_Langevin_epsilon} for future work for the sake of clarity.
There are different ways to approximate the solution of the dynamics~\eqref{equation:Langevin_epsilon}.
A strong approximation focuses on approximating the realization of a single trajectory of~\eqref{equation:Langevin_epsilon} for a given realization of the Wiener process~$W$.
A weak approximation approximates the average of functionals of the solution at a fixed time~$T$, that is, quantities of the form~$\E[\phi(X^\varepsilon(T))]$ for~$\phi$ a smooth test function.
In addition, under growth and smoothness assumptions on the vector fields in~\eqref{equation:Langevin_epsilon} (see, for instance,~\cite{Hasminskii80sso}), the dynamics~\eqref{equation:Langevin_epsilon} naturally satisfy an ergodicity property; that is, there exists a unique invariant measure~$d\mu_\infty^\varepsilon$ in~$\R^d$ that has a density~$\rho_\infty^\varepsilon$ with respect to the Lebesgue measure, such that for all test functions~$\phi$,
$$
\lim_{T\to\infty}\frac{1}{T}\int_0^T \phi(X(t)) dt= \int_{\R^d} \phi(x) d\mu_\infty^\varepsilon(x)\quad \text{almost surely}.
$$
An approximation for the invariant measure focuses on approximating the average of a functional in the stationary state, that is, the quantity~$\int_{\R^d} \phi(x) d\mu_\infty^\varepsilon(x)$.
This is a computational challenge when the dimension~$d$ is high, which is the case in the context of molecular dynamics where the dimension is proportional to the number of particles, as a standard quadrature formula becomes prohibitively expensive in high dimension.
We emphasize that the invariant measure~$\mu_\infty^\varepsilon$ becomes singular with respect to the Lebesgue measure on~$\R^d$ in the limit~$\varepsilon\rightarrow 0$, and tends weakly as~$\varepsilon\rightarrow 0$ to~$d\mu_\infty^0$, a measure that is absolutely continuous to~$d\sigma_\MM$, the canonical measure on~$\MM$ induced by the Euclidean metric of~$\R^d$.
In this paper, we propose weak convergence results for a new uniformly accurate integrator for solving~\eqref{equation:Langevin_epsilon}, and numerical experiments in the weak context and for the invariant measure, as we recall that a scheme of weak order~$r$ automatically has order~$p\geq r$ for the invariant measure (see, for instance,~\cite{Mattingly02efs}).
The most used discretization for solving~\eqref{equation:Langevin_epsilon} is the explicit Euler integrator in~$\R^d$ (see~\cite{Ciccotti05bms,Lelievre08adc,Lelievre10fec,Lelievre12ldw}, for instance),
\begin{equation}
\label{equation:Euler_EEE}
X_{n+1}=X_n+\sqrt{h}\sigma \xi+h f(X_n) +h \frac{\sigma^2}{4}\nabla\ln(\det(G))(X_n) -\frac{h}{\varepsilon}(g G^{-1}\zeta)(X_n).
\end{equation}
This integrator has weak order one of accuracy, but it faces some severe stepsize restriction due to its instability, typically of the form~$h\ll \varepsilon$, in order to be accurate in the regime~$\varepsilon\rightarrow 0$.
Since the solution~$X^\varepsilon(t)$ of~\eqref{equation:Langevin_epsilon} converges to the solution~$X^0(t)$ of~\eqref{equation:projected_Langevin} when~$\varepsilon\rightarrow 0$, one can use integrators for the limit equation~\eqref{equation:projected_Langevin} and apply them to solve the original problem~\eqref{equation:Langevin_epsilon} when~$\varepsilon$ is close to zero.
Indeed, one can prove that the solution $X^\varepsilon(t)$ of~\eqref{equation:Langevin_epsilon} stays at distance $\OO(\sqrt{\varepsilon})$ of the constrained solution $X^0(t)$ of~\eqref{equation:projected_Langevin} (see Theorem \ref{theorem:CV_penalized_constrained}).
Thus, if the timestep of the integrator is small enough and satisfies~$\varepsilon\ll h$, then this integrator is consistent for solving~\eqref{equation:Langevin_epsilon}.
The alternative for the discretization of~\eqref{equation:projected_Langevin} on the manifold of the explicit Euler scheme~\eqref{equation:Euler_EEE} is the constrained Euler scheme
\begin{equation}
\label{equation:Euler_Explicit_constrained}
X^0_{n+1}=X^0_n+\sqrt{h}\sigma \xi +hf(X^0_n) + g(X^0_n)\lambda^0_{n+1}, \quad \zeta(X^0_{n+1})=0,
\end{equation}
where~$\lambda^0_{n+1}\in \R^q$ acts as a Lagrange multiplier, and is entirely determined by the constraint~$\zeta(X^0_{n+1})=0$. This integrator has weak order one for solving~\eqref{equation:projected_Langevin} (see~\cite[Sect.\thinspace 3.2.4]{Lelievre10fec}) and it lies on the manifold~$\MM$. It is a consistent approximation of~\eqref{equation:Langevin_epsilon} if~$\varepsilon$ is close to zero and~$\varepsilon\ll h$.
This integrator is, however, not appropriate for solving~\eqref{equation:Langevin_epsilon} if the size of~$\varepsilon$ is of the order of one, since the exact solution does not evolve in a neighborhood of the manifold in this regime.
We mention a few other techniques to integrate numerically~\eqref{equation:Langevin_epsilon} or~\eqref{equation:projected_Langevin}.
In~\cite{Laurent20eab,Laurent21ocf}, high order Runge-Kutta methods are proposed for sampling the invariant measure in~$\R^d$ and on manifolds.
The paper~\cite{Lelievre19hmc} presents a constrained integrator based on the RATTLE scheme (see~\cite{Ryckaert77nio,Andersen83rav,Hairer10sod}) in the context of the underdamped Langevin dynamics.
Some of the previously cited discretizations can be combined with Metropolis-Hastings rejection procedures~\cite{Metropolis53eos,Hastings70mcs}, as done, for instance, in~\cite{Girolami11rml,Brubaker12afo,Lelievre12ldw,Zappa18mco,Lelievre19hmc}.
As for the Euler integrators~\eqref{equation:Euler_EEE}-\eqref{equation:Euler_Explicit_constrained}, all of the previously mentioned methods are consistent provided that the parameter~$\varepsilon$ and the timestep satisfy~$h\ll \varepsilon$ in the context of methods in~$\R^d$, and satisfy~$\varepsilon\ll h$ in the context of methods on the manifold~$\MM$. When applied in the regime where~$\varepsilon$ and~$h$ share the same order of magnitude, the accuracy of these methods quickly deteriorates, and they may face stability issues.
In past decades, different solutions were proposed for treating the loss of accuracy in the intermediate regime~$h\sim \varepsilon$ in the context of multiscale problems with the help of uniformly accurate (UA) methods.
These methods are capable of solving dynamics indexed by a possibly stiff parameter~$\varepsilon$ with an accuracy and a cost both independent of~$\varepsilon$. A uniformly accurate method is automatically asymptotic-preserving (AP), that is, it converges in the two regimes~$\varepsilon\rightarrow 0$ and~$\varepsilon\sim 1$, but the converse is not true in general.
We refer the reader to the review~\cite{Jin12aps} and references therein for examples of AP integrators for solving multiscale problems.
We mention in particular the paper~\cite{Plechac09imm}, which proposes a penalized Hamiltonian dynamics, and AP discretizations for solving it, and the paper~\cite{Brehier20oap} that gives an AP scheme for the approximation of a class of multiscale SDEs.
In~\cite{Cohen12cao,Vilmart14wso,Laurent20mif}, trigonometric and multirevolution integrators are considered for solving highly oscillatory SDEs (see also the deterministic works~\cite{Melendo97ana,Calvo04aco,Calvo07oem,Chartier14mrc}).
We mention the recent papers~\cite{Chartier15uan,Chartier20anc,Chartier20uan,Almuslimani21uas} (see also the references therein) that introduce uniformly accurate methods for solving a variety of multiscale problems.
There is a rich literature on AP and UA methods, but, to the best of our knowledge, the problem we study here and the techniques we consider are new.
We propose in this paper a new consistent integrator with uniform accuracy and uniform cost for solving penalized Langevin dynamics, that is, a method for solving~\eqref{equation:Langevin_epsilon} whose accuracy and cost do not depend on the parameter~$\varepsilon$.
The article is organized as follows.
Section~\ref{section:main_results} is devoted to the presentation of the new integrator and to the main convergence results.
In Section~\ref{section:proofs}, we build a weak asymptotic expansion of the solution of~\eqref{equation:Langevin_epsilon} that is uniform in~$\varepsilon$, and we use it for proving the uniform accuracy of our integrator.
We compare in Section~\ref{section:numerical_experiments} the new integrator with the explicit Euler scheme in~$\R^d$~\eqref{equation:Euler_EEE} and the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained} on~$\MM$ in numerical experiments on a torus and on the orthogonal group to confirm its order of convergence in the weak context and for sampling the invariant measure.
Finally, we present some possible extensions and future work in Section~\ref{section:future_work}.
\section{Uniformly accurate integrator for penalized Langevin dynamics}
\label{section:main_results}
In this section, we present the new uniformly accurate integrator and the main convergence results of this paper. The proofs are postponed to Section~\ref{section:proofs}.
Let us first lay down a few notations and assumptions. We assume in the rest of the article that~$\MM=\{x\in\R^d, \zeta(x)=0\}$ is a compact and smooth manifold of codimension~$q\geq 1$ embedded in~$\R^d$, where the constraints are given by the smooth map~$\zeta \colon \R^d\rightarrow\R^q$. We write~$g=\nabla \zeta\colon\R^d\rightarrow\R^{d\times q}$ and we assume that the Gram matrix~$G(x)=g^T(x) g(x)\in\R^{q\times q}$ is invertible for all~$x$ in~$\MM$. With these notations, the projection~$\Pi_\MM$ on the tangent bundle is given by~$\Pi_\MM(x)= I_d-G^{-1}(x) g(x) g^T(x)$.
The test functions typically belong to a subspace of~$\CC^\infty_P(\R^d,\R)$, the vector space of~$\CC^\infty$ functions~$\phi(x)$ such that all partial derivatives up to any order have a polynomial growth of the form
$$\big|\phi^{(k)}(x)\big|\leq C(1+\abs{x}^K),$$
where the constants~$C$ and~$K$ are independent of~$x\in \R^d$ (but can depend on~$k$), and where we denote by~$\abs{x}=(x^Tx)^{1/2}$ the Euclidean norm in~$\R^d$.
Similarly, we denote by~$\CC^p_P(\R^d,\R)$ the space of~$\CC^p$ functions whose partial derivatives up to order~$p$ have polynomial growth.
Letting~$\varphi\colon \R^d\rightarrow \R^{d\times k}$, we use the following notations for differentials: for all vectors~$x$,~$a^1$, \dots,~$a^m\in\R^d$, we denote
$$\varphi^{(m)}(x)(a^1,\dots,a^m)
=\sum_{i_1,\dots,i_m=1}^d \frac{\partial^m\varphi}{\partial x_{i_1}\dots\partial x_{i_m}}(x) \thinspace a^1_{i_1}\dots a^m_{i_m}.$$
For the sake of clarity, we will often drop the coefficient~$x$, and if~$m=1$, we also use the notation~$\varphi'$ for the Jacobian matrix of~$\varphi$.
Moreover, for~$(e_i)$ the canonical basis of~$\R^d$, we write
$$
\Delta \varphi(x)=\sum_{i=1}^d \varphi''(x)(e_i,e_i)\quad \text{and} \quad (\Div \varphi(x))_j=\sum_{i=1}^d \varphi_{i j}'(x)(e_i).
$$
In the rest of the paper, we make the following assumption in the spirit of the regularity assumptions made in~\cite[Appx.\thinspace C]{Ciccotti08pod} and~\cite[Chap.\thinspace 2]{Milstein04snf}.
\begin{ass}
\label{assumption:regularity_ass}
The map~$f$ is bounded, is Lipschitz and lies in~$\CC^3_P$. The maps~$g$ and~$g'$ are bounded in~$\R^d$, and there exists~$c>0$ such that, for~$x$,~$y\in \R^d$,
\begin{equation}
\label{equation:assumption_G_modified}
\abs{G_y(x)}\geq c(1+\abs{y})^{-1},\quad \text{where} \quad G_y(x)=\int_0^1 g^T(x+\tau y)d\tau g(x).
\end{equation}
In addition, there exists a smooth change of coordinate
\begin{align*}
\psi\colon&\R^d\rightarrow \R^d\\
& x \mapsto \begin{pmatrix}
\varphi(x)\\\zeta(x)
\end{pmatrix}
\end{align*}
where~$\varphi\colon\R^d\rightarrow\R^{d-q}$ satisfies~$\varphi'g=0$.
The map~$\psi$ lies in~$\CC^5_P$ and is invertible,~$\psi'$ and~$\psi''$ are Lipschitz and there exist two constants~$c$,~$C>0$ such that~$c\leq\abs{\psi'(x)}\leq C$.
\end{ass}
\begin{remark}
Assumption~\ref{assumption:regularity_ass} is almost the same as the one given in~\cite[Appx.\thinspace C]{Ciccotti08pod} to prove the strong convergence of the dynamics~\eqref{equation:modified_Langevin_epsilon} to the constrained dynamics~\eqref{equation:projected_Langevin}. The difference lies in the additional estimate~\eqref{equation:assumption_G_modified} that replaces the weaker assumption that the Gram matrix~$G(x)=G_0(x)$ is invertible on~$\MM$.
We use the estimate~\eqref{equation:assumption_G_modified} for obtaining a uniform expansion of the Lagrange multipliers in the new method and for proving that the new method evolves in a neighborhood of the manifold (see Lemma~\ref{lemma:uniform_expansion_lambda}).
Note that the existence of the change of coordinate~$\psi$ is always valid in a neighborhood of the smooth manifold~$\MM$. The same goes for the estimate~\eqref{equation:assumption_G_modified} for~$x$ in a neighborhood of~$\MM$ and~$y$ in a ball centered on zero.
Assumption~\ref{assumption:regularity_ass} is valid in particular if~$\MM$ is a vector subspace of~$\R^d$, but it is quite restrictive. It would be interesting to extend the results of this paper under simpler regularity assumptions made only on the manifold, as numerical experiments hint that the results presented in this paper still stand without global assumptions. This is matter for future work.
\end{remark}
Under Assumption~\ref{assumption:regularity_ass}, the problems~\eqref{equation:projected_Langevin} and~\eqref{equation:Langevin_epsilon} are well posed, and we obtain the strong convergence of the penalized dynamics~\eqref{equation:Langevin_epsilon} to the constrained dynamics~\eqref{equation:projected_Langevin}.
\begin{theorem}
\label{theorem:CV_penalized_constrained}
Under Assumption~\ref{assumption:regularity_ass}, the solution~$X^\varepsilon$ of the penalized dynamics~\eqref{equation:Langevin_epsilon} converges strongly to~$X^0$, the solution of the constrained dynamics~\eqref{equation:projected_Langevin}; that is, for all~$t\leq T$, there exists a constant~$C>0$ such that, for all~$\varepsilon>0$,
$$\sup_{t\leq T}\E\Big[\abs{X^\varepsilon(t)-X^0(t)}^2\Big]\leq C\varepsilon.$$
Moreover,~$\zeta(X^\varepsilon(t))$ satisfies
$$
\sup_{t\leq T}\E\Big[\abs{\zeta(X^\varepsilon(t))}^2\Big]\leq C\varepsilon.
$$
\end{theorem}
This result was first introduced in~\cite[Appx.\thinspace C]{Ciccotti08pod} for slightly different penalized dynamics. The proof is almost identical, but we present it in Appx.\thinspace~\ref{section:proof_CV_penalized_constrained} for the sake of completeness. In the deterministic setting with~$\sigma=0$, the convergence to the manifold is of order~$1$ in~$\varepsilon$ instead of order~$1/2$.
We introduce the following additional assumption.
\begin{ass}
\label{assumption:regularity_ass_stronger}
There exists~$c>0$ such that, for~$x$,~$y\in \R^d$,
$$
\abs{G_y(x)}\geq c,\quad \text{where} \quad G_y(x)=\int_0^1 g^T(x+\tau y)d\tau g(x).
$$
\end{ass}
Assumption~\eqref{assumption:regularity_ass_stronger} is a stronger version of the inequality~\eqref{equation:assumption_G_modified} and is in the spirit of the concept of admissible Lagrange multipliers~\cite{Lelievre19hmc}. It is always satisfied for~$x$ in a neighborhood of the manifold~$\MM$ and~$y$ in a ball centered on zero if we assume that the Gram matrix~$G(x)=G_0(x)$ is invertible on~$\MM$.
We emphasize that we do not need this assumption for proving the uniform accuracy property of the new method, but we use it for obtaining uniform estimates in the regime~$\varepsilon\rightarrow 0$ and on the numerical implementation of the uniformly accurate method.
We introduce the new integrator for approximating the penalized dynamics~\eqref{equation:Langevin_epsilon} with cost and accuracy independent of the parameter~$\varepsilon$, and a cost comparable to that of the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained} in terms of the number of evaluations of the functions~$f$,~$\zeta$,~$g$, and~$g'$.
\begin{algorithm}[H]
\renewcommand{\thealgorithm}{New Method}
\caption{(Uniform discretization of penalized overdamped Langevin dynamics)}
\begin{algorithmic}
\STATE~$X_0^\varepsilon=X_0$ $\in \MM$
\FOR{$n\geq 0$}
\STATE
\vskip-4ex
\begin{align}
X_{n+1}^\varepsilon&=X_n^\varepsilon
+\sqrt{h}\sigma \xi_n
+hf(X_n^\varepsilon)
+\frac{(1-e^{-h/\varepsilon})^2}{2}(g'(g G^{-1} \zeta) G^{-1} \zeta)(X_n^\varepsilon) \nonumber\\
\label{equation:UA_integrator}
&+\frac{\sigma^2 \varepsilon}{8}(1-e^{-2h/\varepsilon})\nabla\ln(\det(G))(X_n^\varepsilon)
+g(X_n^\varepsilon) \lambda^\varepsilon_{n+1},\\
\zeta(X_{n+1}^\varepsilon)&=e^{-h/\varepsilon}\zeta(X_n^\varepsilon)
+\sigma\sqrt{\frac{\varepsilon}{2}(1-e^{-2h/\varepsilon})}g^T(X_n^\varepsilon)\xi_n \nonumber\\
&+\varepsilon(1-e^{-h/\varepsilon})\big(g^T f+\frac{\sigma^2}{4} g^T\nabla\ln(\det(G)) +\frac{\sigma^2}{2}\Div(g)\big)(X_n^\varepsilon) \nonumber\\
&+\sigma^2\Big( \varepsilon(1-e^{-h/\varepsilon})-\sqrt{\frac{\varepsilon h}{2}(1-e^{-2h/\varepsilon})} \Big) \nonumber\\
&\times \Big(\sum_{i=1}^d (g'(gG^{-1}g^Te_i))^T gG^{-1}g^Te_i - \sum_{i=1}^d (g'(e_i))^T gG^{-1}g^Te_i \Big)(X_n^\varepsilon). \nonumber
\end{align}
\ENDFOR
\end{algorithmic}
\end{algorithm}
The new method works in a way similar to the constrained Euler integrator~\eqref{equation:Euler_Explicit_constrained}. Knowing the approximation~$X_n^\varepsilon$ of~$X^\varepsilon(nh)$, we project a modified Euler step on a modified manifold defined by the constraint given in~\eqref{equation:UA_integrator} (in place of~$\zeta(X_{n+1}^\varepsilon)=0$ for the constrained Euler integrator~\eqref{equation:Euler_Explicit_constrained}). We project the modified step in the direction~$g(X_n^\varepsilon)$ with the help of a Lagrange multiplier~$\lambda^\varepsilon_{n+1}$. For the implementation of the method, one can use, for instance, a fixed point iteration or a Newton method at each step to find the solution~$(X_{n+1}^\varepsilon,\lambda^\varepsilon_{n+1})$ of the implicit system of equations~\eqref{equation:UA_integrator}.
In order for the discretization~\eqref{equation:UA_integrator} to be well-defined, we use bounded random variables.
The~$\xi_n$ are independent and bounded discrete random vectors that have the same moments as standard Gaussian random vectors up to order four, in the spirit of~\cite[Chap.\thinspace 2]{Milstein04snf}, that is, for instance, that their components satisfy
\begin{equation}
\label{equation:discrete_rv}
\P(\xi_i=0)=\frac{2}{3} \quad \text{and} \quad \P(\xi_i=\pm\sqrt{3})=\frac{1}{6}, \quad i=1,\dots,d.
\end{equation}
Note that using truncated Gaussian random variables would also work.
\begin{remark}
The new integrator is related to the popular idea of backward error analysis and modified equations for SDEs (see, for instance,~\cite{Zygalakis11ote,Abdulle12hwo,Debussche12wbe,Kopec15wbea,Kopec15wbeb}).
The idea is to define a projection method (see~\cite[Sect.\thinspace IV.4]{Hairer06gni}) with a modified constraint in place of~$\zeta(X_n)=0$.
Instead of evaluating the stiff term~$\frac{h}{\varepsilon} g G^{-1} \zeta$ as in the Euler scheme~\eqref{equation:Euler_EEE}, we project a modified step of the explicit Euler scheme in~$\R^d$ on a manifold that is close to~$\MM$ when~$\varepsilon\ll 1$ and whose constraint is given by a truncation of a uniform expansion of~$\zeta(X^\varepsilon)$.
When~$\varepsilon\rightarrow \infty$, the expression of the constraint~$\zeta(X_{n+1}^\varepsilon)$ in~\eqref{equation:UA_integrator} tends to a truncated Taylor expansion in~$h$ around~$X_n^\varepsilon$, while for~$\varepsilon\rightarrow 0$,~$\zeta(X_{n+1}^\varepsilon)$ tends to zero, which enforces that the integrator lies on~$\MM$. These intuitions will be made rigorous in Section~\ref{section:proofs}.
\end{remark}
\begin{remark}
In the context of a manifold~$\MM$ of codimension~$q=1$, the Gram matrix~$G(x)$ and~$g(x)^Te_i=g_i(x)$ are real numbers, so that~$\nabla\ln(\det(G))=2G^{-1} g'(g)$ and
$$
\sum_{i=1}^d (g'(gG^{-1}g^Te_i))^T gG^{-1}g^Te_i
=G^{-1}(g'(g))^T g
=\sum_{i=1}^d (g'(e_i))^T gG^{-1}g^Te_i.
$$
The discretization~\eqref{equation:UA_integrator} thus reduces to
\begin{align}
X_{n+1}^\varepsilon&=X_n^\varepsilon
+\sqrt{h}\sigma \xi_n
+hf(X_n^\varepsilon)
+\frac{(1-e^{-h/\varepsilon})^2}{2}(\zeta^2 G^{-2} g'(g))(X_n^\varepsilon) \nonumber\\
\label{equation:UA_integrator_q_1}
&+\frac{\sigma^2 \varepsilon}{4}(1-e^{-2h/\varepsilon})(G^{-1} g'(g))(X_n^\varepsilon)
+g(X_n^\varepsilon) \lambda^\varepsilon_{n+1},\\
\zeta(X_{n+1}^\varepsilon)&=e^{-h/\varepsilon}\zeta(X_n^\varepsilon)
+\sigma\sqrt{\frac{\varepsilon}{2}(1-e^{-2h/\varepsilon})}g^T(X_n^\varepsilon)\xi_n \nonumber\\
&+\varepsilon(1-e^{-h/\varepsilon})(g^T f +\frac{\sigma^2}{2} G^{-1} g^T g'(g) +\frac{\sigma^2}{2}\Div(g))(X_n^\varepsilon). \nonumber
\end{align}
\end{remark}
We present in the rest of the section the uniform accuracy property of the discretization~\eqref{equation:UA_integrator}, and we show that the integrator converges to the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained} when~$\varepsilon\rightarrow 0$.
The different convergence results are summarized by the following commutative diagram, where~$T=Nh$ is fixed. Note that, as we present a convergence result in~$h$ that is uniform in~$\varepsilon$, the two arrows for the convergence in~$h$ rely on the same Theorem~\ref{theorem:uniform_consistency_UA_scheme}.
$$
\begin{tikzcd}[row sep=large,column sep = large]
\text{integrator } X^\varepsilon_N \text{ in } \R^d
\arrow[d,"h\rightarrow 0"',"(Thm.\thinspace~\ref{theorem:uniform_consistency_UA_scheme})"]
\arrow[r,"\varepsilon\rightarrow 0","(Thm.\thinspace~\ref{theorem:convergence_integrator_epsilon})"']
& \text{integrator } X^0_N \text{ on } \MM
\arrow[d,"h\rightarrow 0"',"(Thm.\thinspace~\ref{theorem:uniform_consistency_UA_scheme})"]
\\
\text{solution }X^\varepsilon(T) \text{ of~\eqref{equation:Langevin_epsilon}}
\arrow[r,"\varepsilon\rightarrow 0","(Thm.\thinspace~\ref{theorem:CV_penalized_constrained})"']
& \text{solution }X^0(T) \text{ of~\eqref{equation:projected_Langevin}}
\end{tikzcd}
$$
We now state the main result of this work, that is, the uniform accuracy of the discretization~\eqref{equation:UA_integrator} for approximating the solution of the penalized Langevin dynamics~\eqref{equation:Langevin_epsilon}.
\begin{theorem}
\label{theorem:uniform_consistency_UA_scheme}
Under Assumption~\ref{assumption:regularity_ass}, the integrator~$(X_n^\varepsilon)$ given by~\eqref{equation:UA_integrator} is a consistent uniformly accurate approximation of the solution~$X^\varepsilon(t)$ of the penalized Langevin dynamics~\eqref{equation:Langevin_epsilon}; that is, for a given test function~$\phi\in \CC^5_P$, there exist~$h_0>0$,~$C>0$ such that for all~$\varepsilon>0$,~$h\leq h_0$, the following estimate holds:
\begin{equation}
\label{equation:uniform_global_consistency}
\abs{\E[\phi(X_n^\varepsilon)]-\E[\phi(X^\varepsilon(nh))]}\leq C\sqrt{h},\quad n=0,1,\dots,N,\quad Nh=T.
\end{equation}
\end{theorem}
\begin{remark}
\label{remark:discussion_order_UA_method}
Note that Theorem~\ref{theorem:uniform_consistency_UA_scheme} states the uniform consistency, but not the uniform weak order one, as one could expect.
The discretization~\eqref{equation:UA_integrator} has weak order one if~$\varepsilon=\varepsilon_0$ is fixed or in the limit~$\varepsilon\rightarrow 0$, but an order reduction occurs in the intermediate regime~$0<\varepsilon<\varepsilon_0$, and the integrator only has weak order~$1/2$ with respect to~$h$ in general.
For the sake of simplicity, we leave the creation of uniformly accurate integrators of higher weak order for future works.
\end{remark}
We present the proof of Theorem~\ref{theorem:uniform_consistency_UA_scheme} in Section~\ref{section:proofs}.
It relies on a weak expansion in~$h$ of the solution of~\eqref{equation:Langevin_epsilon} that is uniform in~$\varepsilon$.
One could directly use this uniform expansion as an explicit numerical integrator for solving~\eqref{equation:Langevin_epsilon}. It would also yield a uniformly accurate scheme and would not require one to solve a fixed point problem.
However, in the limit~$\varepsilon\rightarrow 0$, this integrator would almost surely not stay on the manifold.
The crucial geometric property that the integrator lies on the manifold when~$\varepsilon\rightarrow 0$ is satisfied for the new method, as stated in the following result.
\begin{theorem}
\label{theorem:convergence_integrator_epsilon}
Under Assumption~\ref{assumption:regularity_ass}, the integrator~$(X_n^\varepsilon)$ in~\eqref{equation:UA_integrator} converges to the Euler scheme on the manifold~\eqref{equation:Euler_Explicit_constrained} when~$\varepsilon\rightarrow 0$; that is, for~$h$ and~$N$ fixed such that~$T=Nh$, there exists a constant~$C_h>0$ that depends on~$h$ but not on~$\varepsilon$ such that
$$
\abs{\zeta(X^\varepsilon_n)}\leq C_h\sqrt{\varepsilon},\quad n=0,1,\dots,N.
$$
In addition, if Assumption~\ref{assumption:regularity_ass_stronger} is satisfied, then, for~$h_0$ small enough, for~$h\leq h_0$ fixed, there exists a constant~$C_h>0$ such that
\begin{equation}
\label{equation:weak_CV_epsilon_0_strong}
\abs{X_n^\varepsilon-X_n^0}\leq C_h \sqrt{\varepsilon},\quad n=0,1,\dots,N,\quad Nh=T.
\end{equation}
\end{theorem}
\begin{remark}
In the deterministic context (i.e.\thinspace when~$\sigma=0$), the uniform accuracy of the discretization~\eqref{equation:UA_integrator} still holds, and the speed of convergence to the manifold~$\MM$ of both the exact solution and the integrator are in~$\OO(\varepsilon)$.
To the best of our knowledge, the integrator given by~\eqref{equation:UA_integrator} is the first integrator with the uniform accuracy property for solving the singular perturbation problem~\eqref{equation:Langevin_epsilon} with~$\sigma=0$.
However, similar expansions that are uniform with respect to~$\varepsilon$ are presented in~\cite[Chaps.\thinspace VI-VII]{Hairer10sod} and references therein.
\end{remark}
\begin{remark}
Another widely used scheme on the manifold is the Euler scheme with implicit projection direction,
\begin{equation}
\label{equation:Euler_Implicit_constrained}
X^0_{n+1}=X^0_n+hf(X^0_n) +\sqrt{h}\sigma \xi_n + g(X^0_{n+1})\lambda^0_{n+1}, \quad \zeta(X^0_{n+1})=0,
\end{equation}
where the Lagrange multiplier~$\lambda^0_{n+1}$ is determined by the constraint~$\zeta(X^0_{n+1})=0$.
The uniformly accurate discretization given in~\eqref{equation:UA_integrator} can be modified so that it converges to the integrator~\eqref{equation:Euler_Implicit_constrained} when~$\varepsilon\rightarrow 0$.
It suffices to replace the first line of~\eqref{equation:UA_integrator}
\begin{align*}
X_{n+1}^\varepsilon&=X_n^\varepsilon
+\sqrt{h}\sigma \xi_n
+hf(X_n^\varepsilon)
-\frac{(1-e^{-h/\varepsilon})^2}{2}(g'(g G^{-1} \zeta) G^{-1} \zeta)(X_n^\varepsilon) \nonumber\\
&+\frac{\sigma^2}{4}\Big(\sqrt{2\varepsilon h(1-e^{-2h/\varepsilon})}-\frac{\varepsilon}{2}(1-e^{-2h/\varepsilon})\Big)\nabla\ln(\det(G))(X_n^\varepsilon)
+g(X_{n+1}^\varepsilon) \lambda^\varepsilon_{n+1},
\end{align*}
and to keep the same expansion for the constraint~$\zeta(X_{n+1}^\varepsilon)$. The methodology for the uniform expansion of the integrator that we present in Section~\ref{section:expansion_num_sol} extends to this context, so that the convergence results persist.
Similarly, one could change the direction of projection~$g(X_n^\varepsilon)$ into~$g(Y_n^\varepsilon)$, where~$Y_n^\varepsilon$ is any consistent one-step approximation of~$X_n^\varepsilon$, in the spirit of the class of projected Runge-Kutta methods presented in~\cite{Laurent21ocf}.
Finding a class of uniformly accurate discretizations that converge to a more general class of Runge-Kutta methods on the manifold~$\MM$ is matter for future works.
\end{remark}
The uniform discretization~\eqref{equation:UA_integrator} is implicit and requires one to solve a fixed point problem at each step with, for instance, a fixed point iteration or a Newton method.
The following result, in the spirit of~\cite[Chap.\thinspace VII]{Hairer10sod} for deterministic DAEs and~\cite[Lemma 3.3]{Laurent21ocf} for the constrained dynamics~\eqref{equation:projected_Langevin}, confirms that the associated implicit system is not stiff, that is, that its complexity does not depend on the stiff parameter~$\varepsilon$.
\begin{theorem}
\label{theorem:uniform_fixed_point_problem}
Under Assumption~\ref{assumption:regularity_ass}, each step of the integrator~$(X_n^\varepsilon)$ given by~\eqref{equation:UA_integrator} can be rewritten as a solution of a fixed point problem of the form
$$
X_{n+1}^\varepsilon=F_h^\varepsilon(X_{n+1}^\varepsilon),
$$
where~$F_h^\varepsilon\colon\R^d\rightarrow\R^d$ depends on~$X_n^\varepsilon$,~$\xi_n$,~$h$, and~$\varepsilon$.
Moreover, if Assumption~\ref{assumption:regularity_ass_stronger} is satisfied, then there exists~$h_0>0$ independent of~$\varepsilon$ such that for all~$h\leq h_0$,~$F_h^\varepsilon$ is a uniform contraction, that is, there exists a positive constant~$L< 1$ independent of~$h$ and~$\varepsilon$ such that, for all~$y_1$,~$y_2\in\R^d$,
$$\abs{F_h^\varepsilon(y_2)-F_h^\varepsilon(y_1)}\leq L\abs{y_2-y_1}.$$
\end{theorem}
\section{Weak convergence analysis}
\label{section:proofs}
In this section, we present the uniform weak expansion and the stability properties of the solution~$X^\varepsilon(t)$ to the penalized dynamic~\eqref{equation:Langevin_epsilon} and of the uniform integrator~$(X_n^\varepsilon)$ given in~\eqref{equation:UA_integrator}. We then use these results to prove Theorem~\ref{theorem:uniform_consistency_UA_scheme}, Theorem~\ref{theorem:convergence_integrator_epsilon} and Theorem~\ref{theorem:uniform_fixed_point_problem}.
Let us begin the analysis with a few technical lemmas and notations.
\begin{lemma}
\label{lemma:estimate_uniform_zeta}
Let~$(X_n^\varepsilon)$ be given by~\eqref{equation:UA_integrator}, then, under Assumption~\ref{assumption:regularity_ass}, there exists a constant~$C_0>0$ independent of~$X_0 \in \MM$,~$\varepsilon$, and~$h$ such that, for all~$n\geq 0$,
\begin{equation}
\label{equation:convergence_zeta_numeric_lemma}
(1-e^{-h/\varepsilon})\abs{\zeta(X^\varepsilon_n)}\leq C_0\sqrt{h}.
\end{equation}
\end{lemma}
\begin{proof}
As~$f$,~$\xi_n$,~$g$, and~$g'$ are bounded, we obtain from the definition of the integrator given by~\eqref{equation:UA_integrator} that
$$
|\zeta(X^\varepsilon_{n+1})|\leq e^{-h/\varepsilon} |\zeta(X^\varepsilon_n)|
+C \sqrt{\varepsilon(1-e^{-2h/\varepsilon})},
$$
where we used that the function~$(1-e^{-x})/x$ is bounded for~$x>0$.
Thus, as~$\zeta(X_0)=0$,
\begin{align*}
|\zeta(X^\varepsilon_n)|
&\leq e^{-nh/\varepsilon}|\zeta(X_0)|
+C\sqrt{\varepsilon(1-e^{-2h/\varepsilon})} \sum_{k=0}^{n-1} e^{-kh/\varepsilon}\\
&\leq C \frac{\sqrt{\varepsilon(1-e^{-2h/\varepsilon})}}{1-e^{-h/\varepsilon}}
\leq C_0 \frac{\sqrt{h}}{1-e^{-h/\varepsilon}},
\end{align*}
where the constant~$C_0$ does not depend on~$X_0$,~$\varepsilon$,~$n$, and~$h$. This yields the estimate~\eqref{equation:convergence_zeta_numeric_lemma}.
\end{proof}
The estimate~\eqref{equation:convergence_zeta_numeric_lemma} is a direct consequence of our choice of using bounded random variables in~\eqref{equation:UA_integrator}.
We shall use this estimate extensively in the rest of this section.
Thus, we denote by~$\MM^\varepsilon_h$ the set of vectors~$x\in \R^d$ that satisfy the estimate
\begin{equation}
\label{equation:convergence_zeta_numeric}
(1-e^{-h/\varepsilon})\abs{\zeta(x)}\leq C_0\sqrt{h},
\end{equation}
where~$C_0$ is the constant given in Lemma~\ref{lemma:estimate_uniform_zeta}. The set~$\MM^\varepsilon_h$ is a closed subset of~$\R^d$ that contains~$\MM$. The numerical scheme given by~\eqref{equation:UA_integrator} takes values in~$\MM^\varepsilon_h$. We mention that the convergence results are still valid if the initial condition~$X_0$ of~\eqref{equation:UA_integrator} is chosen in~$\MM^\varepsilon_h$ instead of~$\MM$.
As we aim at writing uniform expansions, we introduce the convenient notation~$R^\varepsilon_h(x)$ for any remainder that satisfies at least~$\abs{\E[R^\varepsilon_h(x)]}\leq C h^{3/2}$, where~$C$ is independent of~$\varepsilon$,~$h$, and~$x$.
The following result serves as a technical tool for simplifying the calculations in the uniform expansions of Subsection~\ref{section:expansion_exact_sol} and Subsection~\ref{section:expansion_num_sol}. It is proved with elementary computations.
\begin{lemma}
\label{lemma:rewriting_fixman_correction}
The Fixman correction can be rewritten in the following way:
$$
\frac{\sigma^2}{4}\nabla\ln(\det(G))
=\frac{\sigma^2}{2}\sum_{i=1}^d g'(e_i) G^{-1}g^T e_i
=\frac{\sigma^2}{2}\sum_{i=1}^d g'(g G^{-1}g^T e_i) G^{-1}g^T e_i,
$$
where~$(e_i)$ is the canonical basis of~$\R^d$.
\end{lemma}
\begin{proof}[Proof of Lemma~\ref{lemma:rewriting_fixman_correction}]
Using that~$G=g^T g$ is a symmetric matrix, that~$g=\nabla \zeta$ is a gradient (which implies~$x^Tg'(y)=y^Tg'(x)$) and the standard properties of the trace operator~$\Trace$, we deduce that
\begin{align*}
\partial_j (\ln(\det(G)))
&=\Trace(G^{-1}\partial_j G)
=2\Trace(G^{-1}(g'(e_j))^T g)
=2\Trace(g'(e_j) G^{-1} g^T)\\
&=2\sum_{i=1}^d e_i^T g'(e_j) G^{-1} g^T e_i
=2\sum_{i=1}^d e_j^T g'(e_i) G^{-1} g^T e_i,
\end{align*}
that is,~$\nabla\ln(\det(G))
=2\sum_{i=1}^d g'(e_i) G^{-1}g^T e_i$. For the second equality, we have
\begin{align*}
\sum_{i=1}^d e_j^T g'(g G^{-1}g^T e_i) G^{-1}g^T e_i
&=\sum_{i=1}^d e_i^T g G^{-1} g^T g'(e_j) G^{-1}g^T e_i\\
&=\Trace(g G^{-1} g^T g'(e_j) G^{-1}g^T)
=\Trace(g'(e_j) G^{-1}g^T)\\
&=\sum_{i=1}^d e_i^T g'(e_j) G^{-1}g^T e_i
=\sum_{i=1}^d e_j^T g'(e_i) G^{-1} g^T e_i.
\end{align*}
Hence we get the result.
\end{proof}
\subsection{Uniform expansion of the exact solution}
\label{section:expansion_exact_sol}
We consider the exact solution~$X^\varepsilon(t)$ of the penalized Langevin dynamics~\eqref{equation:Langevin_epsilon}, with the initial condition~$X_0=x$, that we assume is deterministic for simplicity. Then,~$X^\varepsilon$ satisfies the following expansion in~$h$ that is uniform with
respect to~$\varepsilon$.
\begin{proposition}
\label{proposition:weak_uniform_expansion_penalized_dynamic}
Under Assumption~\ref{assumption:regularity_ass}, there exists~$h_0>0$ such that for all~$h\leq h_0$, if~$X^\varepsilon$ is the solution of the penalized Langevin dynamics~\eqref{equation:Langevin_epsilon} starting at~$x\in\MM^\varepsilon_h$, then, for all~$\phi\in \CC^3_P$, the following estimate holds:
\begin{equation}
\label{equation:weak_expansion_penalized_dynamic}
\abs{\E[\phi(X^\varepsilon(h))]-\E[\phi(x+\sqrt{h}A^\varepsilon_h(x)+h B^\varepsilon_h(x))]}\leq C(1+\abs{x}^K) h^{3/2},
\end{equation}
where~$C$ is independent of~$h$ and~$\varepsilon$ and where the functions~$A^\varepsilon_h$ and~$B^\varepsilon_h$ are given by
\begin{align*}
A^\varepsilon_h&=\sigma \xi
+\frac{e^{-h/\varepsilon}-1}{\sqrt{h}} g G^{-1} \zeta
+\sigma\Big(\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})}-1\Big)g G^{-1}g^T \xi,\\
B^\varepsilon_h&=f
+\Big(\frac{\varepsilon}{h}(1-e^{-h/\varepsilon})-1\Big)gG^{-1}g^T f
+\frac{\sigma^2}{2}\Big(\frac{\varepsilon}{h}(1-e^{-h/\varepsilon})-1\Big) gG^{-1} \Div(g)\\
&+\frac{\sigma^2 \varepsilon}{8h}(1-e^{-2h/\varepsilon})\nabla\ln(\det(G))
+\frac{\sigma^2 \varepsilon}{8h}(1-e^{-h/\varepsilon})^2 g G^{-1} g^T\nabla\ln(\det(G))\\
&+\frac{1}{2h}(e^{-h/\varepsilon}-1)^2 \Big(g'(g G^{-1} \zeta) G^{-1} \zeta
-g G^{-1} g^T g'(g G^{-1} \zeta) G^{-1} \zeta
\\&-gG^{-1}(g'(g G^{-1} \zeta))^T g G^{-1} \zeta\Big)
+\sigma^2\Big(1+\frac{\varepsilon}{h}(e^{-h/\varepsilon}-1)\Big)\sum_{i=1}^d g G^{-1}(g'(e_i))^T g G^{-1}g^T e_i\\
&+\frac{\sigma^2}{4} \Big(\frac{\varepsilon}{h}(e^{-2h/\varepsilon}-4e^{-h/\varepsilon}+3)-2\Big) \sum_{i=1}^d gG^{-1}(g'(g G^{-1}g^T e_i))^T g G^{-1}g^T e_i,
\end{align*}
with~$\xi$ a discrete bounded random vector that satisfies~\eqref{equation:discrete_rv}, and where the functions~$A^\varepsilon_h$ and~$B^\varepsilon_h$ are bounded uniformly in~$\varepsilon$ and~$h$ on~$\MM^\varepsilon_h$.
\end{proposition}
Note that for~$q=1$, we obtain the first step of the uniformly accurate discretization~\eqref{equation:UA_integrator_q_1} by gathering all the terms of the form~$gM$ with~$M\in\R^q$ of the weak approximation given in Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic} in a Lagrange multiplier~$g \lambda^\varepsilon_1\in \R^q$ and by adding the truncated expansion of~$\zeta(X^\varepsilon(h))$ as a constraint.
The proof of Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic} relies on the change of coordinate~$\psi$ given in Assumption~\ref{assumption:regularity_ass}.
Instead of discretizing directly the penalized dynamics~\eqref{equation:Langevin_epsilon}, we first apply the change of coordinate~$\psi$ and we derive an expansion in time of~$X^\varepsilon(t)$ that is uniform in the parameter~$\varepsilon$.
The following result is used for proving Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}.
\begin{lemma}
\label{lemma:strong_expansion_penalized_dynamic}
With the same notations and assumptions as in Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}, the following estimates hold for all~$h\leq h_0$ and all $x\in\MM^\varepsilon_h$:
\begin{align}
\label{equation:strong_expansion_0}
\E[|X^\varepsilon(h)-x|^2]^{1/2}&\leq C\sqrt{h},\\
\label{equation:strong_expansion_1/2}
\E[|X^\varepsilon(h)-(x+\sqrt{h}\widehat{A}^\varepsilon_h(x))|^2]^{1/2}&\leq Ch,\\
\label{equation:weak_auxiliary_expansion_penalized_dynamic_psi}
\abs{\E[\psi(X^\varepsilon(h))]-\E[\psi(x+\sqrt{h}\widehat{A}^\varepsilon_h(x)+h B^\varepsilon_h(x))]}&\leq Ch^{3/2},
\end{align}
where~$C$ is independent of~$h$ and~$\varepsilon$,~$B^\varepsilon_h$ is defined in Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}, and~$\widehat{A}^\varepsilon_h$ is given by
\begin{align*}
\widehat{A}^\varepsilon_h&=\sigma \frac{W(h)}{\sqrt{h}}+\frac{e^{-h/\varepsilon}-1}{\sqrt{h}} g G^{-1} \zeta
+\frac{\sigma}{\sqrt{h}}g G^{-1}g^T\int_0^h (e^{(s-h)/\varepsilon}-1) dW(s).
\end{align*}
\end{lemma}
\begin{proof}[Proof of Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}]
The uniform bounds on~$A^\varepsilon_h(X^\varepsilon(t))$ and~$B^\varepsilon_h(X^\varepsilon(t))$ are obtained straightforwardly by using Assumption~\ref{assumption:regularity_ass} and the fact that~$x\in\MM^\varepsilon_h$.
We prove the local weak order one of the approximation~$Y^\varepsilon(h)=x+\sqrt{h}\widehat{A}^\varepsilon_h(x)+h B^\varepsilon_h(x)$ given in Lemma~\ref{lemma:strong_expansion_penalized_dynamic}.
Let~$\phi\in \CC^3_P$ and~$\widetilde{\phi}=\phi\circ \psi^{-1}$; then a Taylor expansion around~$x$ yields
\begin{align*}
\Big|\E[\phi(X^\varepsilon(h))-\phi(Y^\varepsilon(h))]\Big|
&= \Big|\E[\widetilde{\phi}\circ \psi(X^\varepsilon(h))]-\E[\widetilde{\phi}\circ \psi(Y^\varepsilon(h))]\Big|\\
&\leq
\Big|\E[\widetilde{\phi}'(\psi(x))(\psi(X^\varepsilon(h))-\psi(Y^\varepsilon(h)))]\Big|\\
&+\frac{1}{2}\Big|\E[\widetilde{\phi}''(\psi(x))(\psi(X^\varepsilon(h))-\psi(x),\psi(X^\varepsilon(h))-\psi(x))\\
&-\widetilde{\phi}''(\psi(x))(\psi(Y^\varepsilon(h))-\psi(x),\psi(Y^\varepsilon(h))-\psi(x))]\Big|\\
&+C(1+\abs{x}^K)h^{3/2}\\
&\leq
\Big|\widetilde{\phi}'(\psi(x))\Big| \Big|\E[\psi(X^\varepsilon(h))-\psi(Y^\varepsilon(h))]\Big|\\
&+\frac{1}{2}\Big|\widetilde{\phi}''(\psi(x))\Big|
\E[|\psi(X^\varepsilon(h))+\psi(Y^\varepsilon(h))-2\psi(x)|^2]^{1/2}\\
&\cdot \E[|\psi(X^\varepsilon(h))-\psi(Y^\varepsilon(h))|^2]^{1/2}
+C(1+\abs{x}^K)h^{3/2}
\end{align*}
where we used~\eqref{equation:strong_expansion_0}, Assumption~\ref{assumption:regularity_ass}, and the bilinearity of~$\widetilde{\phi}''(\psi(x))$.
With Lemma~\ref{lemma:strong_expansion_penalized_dynamic} and the regularity properties of~$\widetilde{\phi}$ and~$\psi$, we get
\begin{equation}
\label{equation:weak_auxiliary_expansion_penalized_dynamic_proof}
\abs{\E[\phi(X^\varepsilon(h))]-\E[\phi(x+\sqrt{h}\widehat{A}^\varepsilon_h(x)+h B^\varepsilon_h(x))]}\leq C(1+\abs{x}^K)h^{3/2}.
\end{equation}
In the spirit of~\cite[Chap.\thinspace 2]{Milstein04snf}, we replace the random variable~$\widehat{A}^\varepsilon_h(x)$ by the random variable~$A^\varepsilon_h(x)$ that share the same expectation and covariance matrix.
Indeed, a calculation gives
\begin{align*}
\Cov(\widehat{A}^\varepsilon_{h,i}(x),\widehat{A}^\varepsilon_{h,j}(x))
&=\sigma^2 \delta_{ij}
+\frac{2\sigma^2}{h} (g G^{-1}g^T)_{ij} \int_0^h (e^{(s-h)/\varepsilon}-1)ds\\
&+\frac{\sigma^2}{h}\sum_{k=1}^d (g G^{-1}g^T)_{ik}(g G^{-1}g^T)_{jk} \int_0^h (e^{(s-h)/\varepsilon}-1)^2 ds\\
&=\sigma^2 \delta_{ij}+\sigma^2 (g G^{-1}g^T)_{ij} \Big(\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})-1\Big),
\end{align*}
where we used that $g^T g=G$ and the Itô isometry.
On the other hand, a similar calculation yields
\begin{align*}
\Cov(A^\varepsilon_{h,i}(x),A^\varepsilon_{h,j}(x))
&=\sigma^2 \delta_{ij}
+2\sigma^2 \Big(\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})}-1\Big) (g G^{-1}g^T)_{ij}\\
&+\sigma^2\Big(\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})}-1\Big)^2 \sum_{k=1}^d (g G^{-1}g^T)_{ik} (g G^{-1}g^T)_{jk}\\
&=\sigma^2 \delta_{ij}+\sigma^2 (g G^{-1}g^T)_{ij} \Big(\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})-1\Big).
\end{align*}
Replacing $\widehat{A}^\varepsilon_h(x)$ by $A^\varepsilon_h(x)$ in the weak expansion~\eqref{equation:weak_auxiliary_expansion_penalized_dynamic_proof} gives the estimate~\eqref{equation:weak_expansion_penalized_dynamic}.
\end{proof}
The main ingredient of the proof of Lemma~\ref{lemma:strong_expansion_penalized_dynamic} is the decomposition of the terms of the expansion in a part that stays on the tangent space and a part of the form~$gM$ with~$M\in\R^q$ that is orthogonal to the tangent space.
\begin{proof}[Proof of Lemma~\ref{lemma:strong_expansion_penalized_dynamic}]
As~$\psi^{-1}$ is Lipschitz, we have
\begin{align*}
\E[|X^\varepsilon(h)-x|^2]^{1/2}
&\leq C\E[|\psi(X^\varepsilon(h))-\psi(x)|^2]^{1/2}\\
&\leq C\E[|\varphi(X^\varepsilon(h))-\varphi(x)|^2]^{1/2}
+C\E[|\zeta(X^\varepsilon(h))-\zeta(x)|^2]^{1/2}.
\end{align*}
On the one hand, applying the Itô formula to~$\varphi(X^\varepsilon)$ yields
\begin{align}
\label{equation:varphi_Ito_formula_integral}
\varphi(X^\varepsilon(h))
&=\varphi(x)
+\sigma \int_0^h \varphi'(X^\varepsilon(s)) dW(s)\\
&+\int_0^h [\varphi'f+\frac{\sigma^2}{4}\varphi'\nabla\ln(\det(G))+\frac{\sigma^2}{2}\Delta \varphi](X^\varepsilon(s)) ds, \nonumber
\end{align}
where the term in~$\varepsilon$ vanishes as~$\varphi'g=0$ (see Assumption~\ref{assumption:regularity_ass}).
Assumption~\ref{assumption:regularity_ass} allows us to write the uniform strong expansion
$$\E[|\varphi(X^\varepsilon(h))-\varphi(x)|^2]^{1/2}\leq C\sqrt{h}.$$
On the other hand, for~$\zeta(X^\varepsilon)$, we have
$$
d\zeta(X^\varepsilon)=\sigma g^T(X^\varepsilon) dW+\Big[g^T f+\frac{\sigma^2}{4}g^T\nabla\ln(\det(G))+\frac{\sigma^2}{2}\Div(g)-\frac{1}{\varepsilon}\zeta\Big](X^\varepsilon) dt.
$$
With the variation of constants formula, it rewrites into
\begin{align}
\label{equation:zeta_Ito_formula_integral}
\zeta(X^\varepsilon(h))&=
e^{-h/\varepsilon}\zeta(x)
+\sigma \int_0^h e^{(s-h)/\varepsilon} g^T (X^\varepsilon(s)) dW(s)\\
&+\int_0^h e^{(s-h)/\varepsilon}\Big[g^T f+\frac{\sigma^2}{4}g^T\nabla\ln(\det(G))+\frac{\sigma^2}{2}\Div(g)\Big](X^\varepsilon(s)) ds. \nonumber
\end{align}
As the integrands in~\eqref{equation:zeta_Ito_formula_integral} are bounded (using Assumption~\ref{assumption:regularity_ass}), we get
$$\E[|\zeta(X^\varepsilon(h))-\zeta(x)|^2]^{1/2}
\leq C((e^{-h/\varepsilon}-1)^2 \abs{\zeta(x)}^2+h)^{1/2}
\leq C\sqrt{h},$$
where we used that~$x\in\MM^\varepsilon_h$.
We thus get the desired estimate~\eqref{equation:strong_expansion_0}.
The estimate~\eqref{equation:strong_expansion_1/2} is obtained with the same arguments by keeping track of the terms of size~$\OO(\sqrt{h})$ in the expansions.
We now prove the weak estimate~\eqref{equation:weak_auxiliary_expansion_penalized_dynamic_psi}.
We denote for simplicity~$Y^\varepsilon(h)=x+\sqrt{h}\widehat{A}^\varepsilon_h(x)+h B^\varepsilon_h(x)$.
Let us first look at the approximation of~$\varphi(X^\varepsilon(h))$.
On the one hand, applying the Itô formula to~$\varphi(X^\varepsilon(t))$ gives
\begin{align*}
\varphi(X^\varepsilon(h))
&=\varphi(x)
+h\varphi'f(x)+h\frac{\sigma^2}{4}\varphi'\nabla\ln(\det(G))(x)+h\frac{\sigma^2}{2}\Delta \varphi(x)
+R^\varepsilon_h(x),
\end{align*}
where we used~\eqref{equation:strong_expansion_0} and we put in~$R^\varepsilon_h(x)$ all the terms that are zero in average.
On the other hand, an expansion in~$h$ of~$\varphi(Y^\varepsilon(h))$ yields
\begin{align*}
\varphi(Y^\varepsilon(h))
&=\varphi
+\sqrt{h}\varphi'\widehat{A}^\varepsilon_h
+h\Big[\varphi'B^\varepsilon_h+\frac{1}{2}\varphi''(\widehat{A}^\varepsilon_h,\widehat{A}^\varepsilon_h)\Big]
+R^\varepsilon_h\\
&=\varphi
+h\varphi'f
+\frac{\sigma^2 \varepsilon}{8}(1-e^{-2h/\varepsilon})\varphi'\nabla\ln(\det(G))
+\frac{1}{2}(e^{-h/\varepsilon}-1)^2 \varphi' g'(g G^{-1} \zeta) G^{-1} \zeta\\
&+\frac{\sigma^2}{2}\varphi'' (W(h),W(h))
+\frac{1}{2}(e^{-h/\varepsilon}-1)^2\varphi''(g G^{-1} \zeta,g G^{-1} \zeta) \\
&+\frac{\sigma^2}{2}\int_0^h \int_0^h (e^{(s-h)/\varepsilon}-1) (e^{(u-h)/\varepsilon}-1)\varphi''(g G^{-1}g^T dW(s),g G^{-1}g^T dW(u)) \\
&+\sigma^2\int_0^h (e^{(s-h)/\varepsilon}-1) \varphi''(W(h),g G^{-1}g^T dW(s))
+R^\varepsilon_h,
\end{align*}
where we use that~$\varphi'g=0$, we omit the dependence in~$x$ for conciseness, and we put in~$R^\varepsilon_h(x)$ all the terms that are zero in average.
We now replace the random terms by their expectation,
\begin{align*}
\varphi(Y^\varepsilon(h))
&=\varphi
+h\varphi'f
+\frac{\sigma^2 \varepsilon}{8}(1-e^{-2h/\varepsilon})\varphi'\nabla\ln(\det(G))
+h\frac{\sigma^2}{2}\Delta\varphi \\
&+\frac{1}{2}(e^{-h/\varepsilon}-1)^2 (\varphi'(g'(g G^{-1} \zeta) G^{-1} \zeta)+\varphi''(g G^{-1} \zeta,g G^{-1} \zeta)) \\
&+\frac{\sigma^2}{2} \int_0^h (e^{(s-h)/\varepsilon}-1)^2 ds \sum_{i=1}^d \varphi''(g G^{-1}g^T e_i,g G^{-1}g^T e_i) \\
&+\sigma^2 \int_0^h (e^{(s-h)/\varepsilon}-1) ds \sum_{i=1}^d \varphi''(e_i,g G^{-1}g^T e_i)
+R^\varepsilon_h.
\end{align*}
Letting~$M\in\R^q$, we differentiate the equality~$\varphi'(gM)=0$. We obtain that for any~$M\in\R^q$ and~$v\in\R^d$ we have~$\varphi'(g'(v)M)+\varphi''(gM,v)=0$. We deduce that
$$\varphi'(g'(g G^{-1} \zeta) G^{-1} \zeta)+\varphi''(g G^{-1} \zeta,g G^{-1} \zeta)=0.$$
Applying Lemma~\ref{lemma:rewriting_fixman_correction}, we get by a direct calculation that
$$\varphi(Y^\varepsilon(h))=\varphi(x)
+h\varphi'f(x)+h\frac{\sigma^2}{4}\varphi'\nabla\ln(\det(G))(x)+h\frac{\sigma^2}{2}\Delta \varphi(x)
+R^\varepsilon_h(x),$$
which gives the desired estimate
\begin{equation}
\label{equation:weak_auxiliary_expansion_penalized_dynamic_varphi}
\abs{\E[\varphi(X^\varepsilon(h))]-\E[\varphi(x+\sqrt{h}\widehat{A}^\varepsilon_h(x)+h B^\varepsilon_h(x))]}\leq Ch^{3/2}.
\end{equation}
For the one-step approximation of~$\zeta(X^\varepsilon(h))$, the Itô formula and the variation of constants formula yield
$$
\zeta(X^\varepsilon(h))
=
e^{-h/\varepsilon}\zeta(x)
+\varepsilon(1-e^{-h/\varepsilon})(g^T f+\frac{\sigma^2}{4}g^T \nabla\ln(\det(G))+\frac{\sigma^2}{2}\Div(g))(x)
+R^\varepsilon_h(x).
$$
For~$\zeta(Y^\varepsilon(h))$, using~$g^T g=G$, we get with the same arguments as for~$\varphi(Y^\varepsilon(h))$ that
\begin{align*}
\zeta(Y^\varepsilon(h))
&=\zeta
+\sqrt{h}g^T \widehat{A}^\varepsilon_h
+h\Big[g^T B^\varepsilon_h +\frac{1}{2}(g'(\widehat{A}^\varepsilon_h))^T \widehat{A}^\varepsilon_h \Big]
+R^\varepsilon_h \\
&=
e^{-h/\varepsilon}\zeta
+\varepsilon(1-e^{-h/\varepsilon})g^T f
+\frac{\sigma^2}{2}\Big(\varepsilon(1-e^{-h/\varepsilon})-h\Big) \Div(g)+\frac{\sigma^2}{2} (g' (W(h)))^T W(h)\\
&+\frac{\sigma^2 \varepsilon}{4}(1-e^{-h/\varepsilon})^2 g^T\nabla\ln(\det(G))
+\sigma^2\Big(h+\varepsilon(e^{-h/\varepsilon}-1)\Big)\sum_{i=1}^d (g'(e_i))^T g G^{-1}g^T e_i\\
&+\frac{\sigma^2}{4} \Big(\varepsilon(e^{-2h/\varepsilon}-4e^{-h/\varepsilon}+3)-2h\Big) \sum_{i=1}^d(g'(g G^{-1}g^T e_i))^T g G^{-1}g^T e_i\\
&+\frac{\sigma^2}{2} \int_0^h \int_0^h (e^{(s-h)/\varepsilon}-1)(e^{(u-h)/\varepsilon}-1) (g'(g G^{-1}g^T dW(u)))^T g G^{-1}g^T dW(s) \\
&+\sigma^2 \int_0^h \int_0^h (e^{(s-h)/\varepsilon}-1) (g'(W(h)))^T g G^{-1}g^T dW(s)
+R^\varepsilon_h.
\end{align*}
We now replace the stochastic integrals by their expectations (putting the remainders in~$R^\varepsilon_h$), and we use Lemma~\ref{lemma:rewriting_fixman_correction} to simplify the expansion. It yields
\begin{align*}
\zeta(Y^\varepsilon(h))
&=e^{-h/\varepsilon}\zeta
+\varepsilon(1-e^{-h/\varepsilon})(g^T f+\frac{\sigma^2}{4}g^T \nabla\ln(\det(G))+\frac{\sigma^2}{2}\Div(g))
+R^\varepsilon_h,
\end{align*}
which implies
\begin{equation}
\label{equation:weak_auxiliary_expansion_penalized_dynamic_zeta}
\abs{\E[\zeta(X^\varepsilon(h))]-\E[\zeta(x+\sqrt{h}\widehat{A}^\varepsilon_h(x)+h B^\varepsilon_h(x))]}\leq Ch^{3/2}.
\end{equation}
Combining the inequalities~\eqref{equation:weak_auxiliary_expansion_penalized_dynamic_varphi} and~\eqref{equation:weak_auxiliary_expansion_penalized_dynamic_zeta} gives the desired weak estimate~\eqref{equation:weak_auxiliary_expansion_penalized_dynamic_psi}.
\end{proof}
To end this subsection, we recall the growth properties of~$X^\varepsilon$ that will be of use in Subsection~\ref{section:proofs_cv_theorems}.
For this particular result, we add the dependency in the initial condition~$x$ of the exact solution of~\eqref{equation:Langevin_epsilon} with the notation~$X^\varepsilon(t,x)$.
\begin{lemma}
\label{lemma:growth_properties_exact_sol}
Under Assumption~\ref{assumption:regularity_ass}, for~$\phi\in \CC^5_P$ and~$t\leq T$ fixed, the function~$\widetilde{\phi}^\varepsilon(x)=\E[\phi( X^\varepsilon(t,x))]$ lies in~$\CC^3_P$ with constants independent of~$t$ and~$\varepsilon$.
\end{lemma}
\begin{proof}
The standard theory (see, for instance, the textbook~\cite{Milstein04snf}) gives~$\widetilde{\phi}^\varepsilon\in \CC^3$.
With the regularity assumptions on~$\psi$ and its derivatives, it is sufficient to prove that
$
\E[\phi(Y^\varepsilon(t,y))]
$
is in~$\CC^3_P$, where~$Y^\varepsilon(t,y)=\psi(X^\varepsilon(t,\psi^{-1}(x)))$ (replacing~$\phi$ by~$\phi\circ \psi^{-1}$).
We recall from the proof of Lemma~\ref{lemma:strong_expansion_penalized_dynamic} that~$\varphi(X^\varepsilon(t,x))$ satisfies the integral formulation~\eqref{equation:varphi_Ito_formula_integral} and that~$\zeta(X^\varepsilon(t,x))$ satisfies~\eqref{equation:zeta_Ito_formula_integral}.
Putting together these equations, we deduce that~$Y^\varepsilon(t,y)$ satisfies an equation of the form
$$
Y^\varepsilon(t,y)=A^\varepsilon(t) y+\int_0^t A^\varepsilon(t-s) F(Y^\varepsilon(s,y))ds+\int_0^t A^\varepsilon(t-s) G(Y^\varepsilon(s,y))dW(s),$$
where~$F$ and~$G$ are in~$\CC^3_P$ and do not depend on~$\varepsilon$, and
$$A^\varepsilon(t)=\begin{pmatrix}
I_{d-q} &0\\
0&e^{-t/\varepsilon}I_q
\end{pmatrix}.$$
As~$\abs{A^\varepsilon(t)}\leq C$, the process~$Y^\varepsilon(t,y)$ satisfies~$\E[\abs{Y^\varepsilon(t,y)}^{2p}]\leq C(1+\abs{y}^K)$, with~$C$ and~$K$ independent of~$\varepsilon$.
Thus we have
$$\E[\phi(Y^\varepsilon(t,y))]\leq C(1+\E[\abs{Y^\varepsilon(t,y)}^K])\leq C(1+\E[\abs{y}^K]).$$
For the derivatives, we recall from~\cite{Gikhman07sde} that~$Z^\varepsilon(t,y)=\partial_y Y^\varepsilon(t,y)$ satisfies the equation
\begin{align*}
Z^\varepsilon(t,y)&=A^\varepsilon(t) I_d+\int_0^t A^\varepsilon(t-s) F'(Y^\varepsilon(s,y))Z^\varepsilon(t,y)ds\\
&+\int_0^t A^\varepsilon(t-s) G'(Y^\varepsilon(s,y))Z^\varepsilon(t,y)dW(s).
\end{align*}
Applying the same arguments as for~$Y^\varepsilon(t,y)$ yields that~$Z^\varepsilon(t,y)$ has bounded moments of all order and that~$\E[\phi(Z^\varepsilon(t,y))]\leq C(1+\E[\abs{y}^K])$.
The same methodology extends to~$\partial_y^2 Y^\varepsilon(t,y)$ and~$\partial_y^3 Y^\varepsilon(t,y)$.
\end{proof}
\subsection{Uniform expansion and bounded moments of the numerical solution}
\label{section:expansion_num_sol}
In this subsection, we show that the integrator given by~\eqref{equation:UA_integrator} has bounded moments of all order, that it lies on the manifold~$\MM$ in the limit~$\varepsilon\rightarrow 0$, and that it satisfies the same local weak uniform expansion as the exact solution of~\eqref{equation:Langevin_epsilon} (see Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}).
First, the integrator given by~\eqref{equation:UA_integrator} satisfies the following bounded moments property.
\begin{proposition}
\label{proposition:bounded_moments}
Under Assumption~\ref{assumption:regularity_ass},~$(X_n^\varepsilon)$ has bounded moments of any order along time; i.e., for all timestep~$h\leq h_0$ small enough such that~$Nh=T$ is fixed, for all integer~$m\geq 0$,
$$
\sup_{n\leq N} \E[\abs{X_n^\varepsilon}^{2m}] \leq C_m,
$$
where the constant~$C>0$ is independent of~$\varepsilon$ and~$h$.
\end{proposition}
The integrator~$(X_n^\varepsilon)$ given in~\eqref{equation:UA_integrator} also satisfies a uniform local expansion that is similar to its continuous counterpart presented in Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}.
\begin{proposition}
\label{proposition:uniform_expansion_integrator}
Under Assumption~\ref{assumption:regularity_ass}, there exists~$h_0>0$ such that for all~$h\leq h_0$, if~$(X^\varepsilon_n)$ is the numerical discretization given by~\eqref{equation:UA_integrator} beginning at~$x\in \MM^\varepsilon_h$ (assumed deterministic for simplicity), then, for all test functions~$\phi\in \CC^3_P$, the following estimate holds:
\begin{equation}
\label{equation:weak_expansion_numerical method}
\abs{\E[\phi(X^\varepsilon_1)]-\E[\phi(x+\sqrt{h}A^\varepsilon_h(x)+h B^\varepsilon_h(x))]}\leq C(1+\abs{x}^K) h^{3/2},
\end{equation}
where~$C$ is independent of~$h$ and~$\varepsilon$, and~$A^\varepsilon_h$ and~$B^\varepsilon_h$ are the functions given in Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic}.
\end{proposition}
To prove Proposition~\ref{proposition:bounded_moments} and Proposition~\ref{proposition:uniform_expansion_integrator}, we rely on the following lemma, whose proof is postponed to the end of this subsection.
We emphasize that an inequality of the form $\abs{\zeta(x)}\leq C$ does not imply in general that $x$ stays close to $\MM$. That is why we rely in Lemma \ref{lemma:uniform_expansion_lambda} on an estimate of the Lagrange multipliers (using the inequality \eqref{equation:assumption_G_modified}).
This estimate ensures that the method evolves in a neighborhood of the manifold.
\begin{lemma}
\label{lemma:uniform_expansion_lambda}
Under Assumption~\ref{assumption:regularity_ass} and if~$x\in\MM^\varepsilon_h$, there exists~$h_0>0$ such that, for all timestep~$h\leq h_0$, the one-step approximation~$X_1^\varepsilon$ and the Lagrange multiplier~$\lambda^\varepsilon_1$ in the discretization~\eqref{equation:UA_integrator} satisfy
$$
\abs{X_1^\varepsilon-x}\leq C\sqrt{h},
\qquad
\lambda^\varepsilon_1=\sqrt{h}G^{-1}(x)\lambda_{1,(1/2)}^{\varepsilon}+hG^{-1}(x)\lambda_{1,(1)}^{\varepsilon}+R^\varepsilon_h(x),
$$
where~$|\lambda_{1,(1/2)}^{\varepsilon}|\leq C$,~$|\lambda_{1,(1)}^{\varepsilon}|\leq C$, and~$|R^\varepsilon_h(x)|\leq Ch^{3/2}$ with~$C$ independent of~$\varepsilon$,~$h$ and~$x$.
\end{lemma}
For proving Proposition~\ref{proposition:bounded_moments}, we apply the change of variable~$\psi$ and we adapt the standard methodology presented in~\cite[Lemma 1.1.6 \& Lemma 2.2.2]{Milstein04snf}.
\begin{proof}[Proof of Proposition~\ref{proposition:bounded_moments}]
We derive from~\eqref{equation:UA_integrator} that
\begin{align*}
\abs{\E[\zeta(X_{n+1}^\varepsilon)-e^{-h/\varepsilon}\zeta(X_n^\varepsilon)|X_n^\varepsilon]}&\leq C h,\\
\abs{\zeta(X_{n+1}^\varepsilon)-e^{-h/\varepsilon}\zeta(X_n^\varepsilon)}&\leq C \sqrt{h}.
\end{align*}
We prove that~$\zeta(X_n)$ has bounded moments by induction on~$n$. The binomial formula yields
\begin{align*}
\E[\abs{\zeta(X_{n+1}^\varepsilon)}^{2m}]
&=\E[\abs{e^{-h/\varepsilon}\zeta(X^\varepsilon_n)+\zeta(X^\varepsilon_{n+1})-e^{-h/\varepsilon}\zeta(X^\varepsilon_n)}^{2m}]\\
&\leq e^{-2mh/\varepsilon} \E[\abs{\zeta(X_n^\varepsilon)}^{2m}]\\
&+C\E[\abs{\zeta(X_n^\varepsilon)}^{2m-1}\abs{\E[\zeta(X_{n+1}^\varepsilon)-e^{-h/\varepsilon}\zeta(X_n^\varepsilon)|X_n^\varepsilon]}]\\
&+C\sum_{k=2}^{2m} \E[\abs{\zeta(X_n^\varepsilon)}^{2m-k} \abs{\zeta(X_{n+1}^\varepsilon)-e^{-h/\varepsilon}\zeta(X_n^\varepsilon)}^k]\\
&\leq \E[\abs{\zeta(X_n^\varepsilon)}^{2m}]+C(1+\E[\abs{\zeta(X_n^\varepsilon)}^{2m}])h.
\end{align*}
Following~\cite[Lemma 1.1.6]{Milstein04snf}, as~$X_0\in\MM$ is bounded, it implies that~$\E[\abs{\zeta(X_n)}^{2m}]$ is bounded uniformly in~$n=0,\dots,N$ and~$\varepsilon$.
Using Lemma~\ref{lemma:uniform_expansion_lambda} and the equality~$\varphi'g=0$, a direct calculation gives
\begin{align}
\label{equation:bounded_moments_sufficient_condition_1}
\abs{\E[\varphi(X_{n+1}^\varepsilon)-\varphi(X_n^\varepsilon)|X_n^\varepsilon]}&\leq C h,\\
\label{equation:bounded_moments_sufficient_condition_2}
\abs{\varphi(X_{n+1}^\varepsilon)-\varphi(X_n^\varepsilon)}&\leq C \sqrt{h}.
\end{align}
Following the same methodology as for~$\E[\abs{\zeta(X_n)}^{2m}]$, the estimates~\eqref{equation:bounded_moments_sufficient_condition_1}-\eqref{equation:bounded_moments_sufficient_condition_2} imply that the quantity~$\E[\abs{\varphi(X_n)}^{2m}]$ is bounded uniformly in~$n=0,\dots,N$ and~$\varepsilon$.
Then, as~$\psi^{-1}$ is Lipschitz, we have
$$
\E[\abs{X_n^\varepsilon}^{2m}]\leq C(1+\E[\abs{\psi(X_n^\varepsilon)}^{2m}])\leq C(1+\E[\abs{\varphi(X_n^\varepsilon)}^{2m}]+\E[\abs{\zeta(X_n^\varepsilon)}^{2m}])\leq C.
$$
Hence we get the result.
\end{proof}
We obtain the uniform expansion of the numerical solution by writing explicitly a uniform expansion of the Lagrange multiplier~$\lambda^\varepsilon_1$, in the spirit of~\cite[Lemma 3.25]{Lelievre10fec}.
\begin{proof}[Proof of Proposition~\ref{proposition:uniform_expansion_integrator}]
Using Lemma~\ref{lemma:uniform_expansion_lambda} and Assumption~\ref{assumption:regularity_ass}, we obtain
$$
X^\varepsilon_1=x+\sqrt{h}\Big[
\sigma \xi
+(gG^{-1})(x)\lambda_{1,(1/2)}^{\varepsilon}
\Big]+R_h^\varepsilon(x),
$$
where the remainder satisfies~$\abs{R_h^\varepsilon(x)}\leq Ch$.
The constraint is then given by
\begin{align*}
\zeta(X^\varepsilon_1)
&=\zeta(x)+\sqrt{h}\Big[
\sigma g^T(x)\xi
+\lambda_{1,(1/2)}^{\varepsilon}
\Big]+R_h^\varepsilon(x).
\end{align*}
On the other hand, we get from the definition of the integrator~\eqref{equation:UA_integrator} that
\begin{align*}
\zeta(X^\varepsilon_1)&=\zeta(x)+\sqrt{h}\Big[
\frac{e^{-h/\varepsilon}-1}{\sqrt{h}}\zeta(x)
+\sigma\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})}g^T(x)\xi
\Big]+R_h^\varepsilon(x).
\end{align*}
By identifying the two terms in~$\sqrt{h}$ in the expansions of~$\zeta(X^\varepsilon_1)$, we deduce the value of~$\lambda_{1,(1/2)}^{\varepsilon}$, that is,
$$
\lambda_{1,(1/2)}^{\varepsilon}=
\frac{e^{-h/\varepsilon}-1}{\sqrt{h}} \zeta
+\sigma\Big(\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})}-1\Big)g^T \xi.
$$
The expression of~$\lambda_{1,(1/2)}^{\varepsilon}$ and Lemma~\ref{lemma:uniform_expansion_lambda} give
\begin{align}
X^\varepsilon_1&=x+\sqrt{h}A^\varepsilon_h(x)+h\Big[f(x)
+\frac{(1-e^{-h/\varepsilon})^2}{2h}(g'(g G^{-1} \zeta) G^{-1} \zeta)(x)\nonumber\\
\label{equation:expansion_1_X_proof}
&+\frac{\sigma^2 \varepsilon}{8h}(1-e^{-2h/\varepsilon})\nabla\ln(\det(G))(x)+(gG^{-1})(x)\lambda_{1,(1)}^{\varepsilon}
\Big] +R_h^\varepsilon(x),
\end{align}
where~$\abs{R_h^\varepsilon(x)}\leq Ch^{3/2}$.
We then compute the expansion of~$\zeta(X^\varepsilon_1)$, and we compare it with the definition of the integrator~\eqref{equation:UA_integrator} to obtain the expression of~$\lambda_{1,(1)}^{\varepsilon}$.
Inserting this expression in~\eqref{equation:expansion_1_X_proof} gives
$$
X^\varepsilon_1=x+\sqrt{h}A^\varepsilon_h(x)+hB^\varepsilon_h(x)+R_h^\varepsilon(x),
$$
where the remainder satisfies~$\abs{\E[R_h^\varepsilon(x)]}\leq Ch^{3/2}$ and~$\abs{R_h^\varepsilon(x)}\leq Ch$.
A Taylor expansion of~$\phi(X^\varepsilon_1)$ around~$x+\sqrt{h}A^\varepsilon_h(x)+h B^\varepsilon_h(x)$ yields the estimate~\eqref{equation:weak_expansion_numerical method}.
\end{proof}
The proof of Lemma~\ref{lemma:uniform_expansion_lambda} mainly relies on the estimates~\eqref{equation:assumption_G_modified} and~\eqref{equation:convergence_zeta_numeric}.
We refer the reader to~\cite[Lemma 3.3]{Laurent21ocf} and~\cite[Chap.\thinspace VII]{Hairer10sod} for similar proofs where explicit expressions of Lagrange multipliers are derived.
\begin{proof}[Proof of Lemma~\ref{lemma:uniform_expansion_lambda}]
Let~$x\in\MM^\varepsilon_h$. For brevity, we rewrite the discretization~\eqref{equation:UA_integrator} as
\begin{align*}
X_1^\varepsilon&=x+\sqrt{h}Y_{(1/2)}^\varepsilon(x)+hY_{(1)}^\varepsilon(x)+g(x)\lambda^\varepsilon_1,\\
\zeta(X_1^\varepsilon)&=\zeta(x)+\sqrt{h}\zeta_{(1/2)}^\varepsilon(x)+h\zeta_{(1)}^\varepsilon(x),
\end{align*}
where the functions~$Y_{(1/2)}^\varepsilon$,~$Y_{(1)}^\varepsilon$,~$\zeta_{(1/2)}^\varepsilon$, and~$\zeta_{(1)}^\varepsilon$ are given by
\begin{align*}
Y_{(1/2)}^\varepsilon &=\sigma\xi,\\
Y_{(1)}^\varepsilon &=f
+\frac{(1-e^{-h/\varepsilon})^2}{2h}(g'(g G^{-1} \zeta) G^{-1} \zeta)
+\frac{\sigma^2 \varepsilon}{8h}(1-e^{-2h/\varepsilon})\nabla\ln(\det(G)) ,\\
\zeta_{(1/2)}^\varepsilon &=\frac{e^{-h/\varepsilon}-1}{\sqrt{h}}\zeta
+\sigma\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})}g^T \xi,\\
\zeta_{(1)}^\varepsilon &=
\frac{\varepsilon}{h}(1-e^{-h/\varepsilon})\big(g^T f+\frac{\sigma^2}{4} g^T\nabla\ln(\det(G)) +\frac{\sigma^2}{2}\Div(g)\big) \\
&+\sigma^2\Big(\frac{\varepsilon}{h}(1-e^{-h/\varepsilon})-\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})} \Big) \sum_{i=1}^d ((g'(gG^{-1}g^Te_i))^T gG^{-1}g^Te_i) \\
&-\sigma^2\Big(\frac{\varepsilon}{h}(1-e^{-h/\varepsilon})-\sqrt{\frac{\varepsilon}{2h}(1-e^{-2h/\varepsilon})} \Big) \sum_{i=1}^d ((g'(e_i))^T gG^{-1}g^Te_i) .
\end{align*}
Using Assumption~\ref{assumption:regularity_ass} and the estimate~\eqref{equation:convergence_zeta_numeric}, the following uniform estimates hold:
$$
\abs{Y_{(i)}^\varepsilon(x)}\leq C,\quad \abs{\zeta_{(i)}^\varepsilon(x)}\leq C,\quad i\in\{1/2,1\}.
$$
The fundamental theorem of calculus yields
$$
\zeta(X_1^\varepsilon)-\zeta(x)=\int_0^1 g^T(x+\tau(X_1^\varepsilon-x))d\tau(X_1^\varepsilon-x)=\sqrt{h}\zeta_{(1/2)}^\varepsilon(x)+h\zeta_{(1)}^\varepsilon(x).
$$
Substituting~$X_1^\varepsilon-x$ then gives
$$
\int_0^1 g^T(x+\tau(X_1^\varepsilon-x))d\tau (\sqrt{h}Y_{(1/2)}^\varepsilon(x)+hY_{(1)}^\varepsilon(x)+g(x)\lambda^\varepsilon_1)
=\sqrt{h}\zeta_{(1/2)}^\varepsilon(x)+h\zeta_{(1)}^\varepsilon(x).
$$
Using Assumption~\ref{assumption:regularity_ass}, we get the following explicit expression of~$\lambda^\varepsilon_1$:
\begin{align}
\label{equation:expression_exact_lambda}
\lambda^\varepsilon_1&=\sqrt{h}G_{X_1^\varepsilon-x}^{-1}(x)
\Big(\zeta_{(1/2)}^\varepsilon(x)-\int_0^1 g^T(x+\tau(X_1^\varepsilon-x))d\tau Y_{(1/2)}^\varepsilon(x)\Big)\\
&+hG_{X_1^\varepsilon-x}^{-1}(x)
\Big(\zeta_{(1)}^\varepsilon(x)-\int_0^1 g^T(x+\tau(X_1^\varepsilon-x))d\tau Y_{(1)}^\varepsilon(x)\Big). \nonumber
\end{align}
Then, the growth assumption~\eqref{equation:assumption_G_modified} on~$G_y^{-1}(x)$ allows us to write
$$\abs{\lambda^\varepsilon_1}\leq C\sqrt{h} (1+\abs{X_1^\varepsilon-x}) \quad \text{and} \quad \abs{X_1^\varepsilon-x}\leq C\sqrt{h}(1+\abs{X_1^\varepsilon-x}).$$
Hence, for~$h\leq h_0$ small enough, we deduce that~$\abs{X_1^\varepsilon-x}\leq C\sqrt{h}$ and
\begin{equation}
\label{equation:expansion_1/2_lambda}
\lambda^\varepsilon_1= \sqrt{h}G^{-1}(x)
(\zeta_{(1/2)}^\varepsilon-g^T Y_{(1/2)}^\varepsilon)(x)+R^\varepsilon_h,
\end{equation}
where~$\abs{R^\varepsilon_h}\leq Ch$.
For the term of size~$\OO(h)$, we first deduce from~\eqref{equation:expansion_1/2_lambda} that
$$\abs{X_1^\varepsilon-x-\sqrt{h}(Y_{(1/2)}^\varepsilon+gG^{-1}
\zeta_{(1/2)}^\varepsilon-gG^{-1}g^T Y_{(1/2)}^\varepsilon)(x)}\leq Ch.$$
By using this estimate in~\eqref{equation:expression_exact_lambda}, a Taylor expansion yields the desired expansion of~$\lambda^\varepsilon_1$.
\end{proof}
\subsection{Proofs of the convergence theorems}
\label{section:proofs_cv_theorems}
Now that we have the local uniform expansion of the exact solution and the numerical scheme, as well as the stability property of Proposition~\ref{proposition:bounded_moments}, we are able to prove the main convergence theorems.
\begin{proof}[Proof of Theorem~\ref{theorem:uniform_consistency_UA_scheme}]
We derive the global weak consistency~\eqref{equation:uniform_global_consistency} with techniques similar to the ones presented in~\cite[Chap.\thinspace 2]{Milstein04snf}.
We denote by~$X^\varepsilon(t,x)$ the solution of the penalized dynamics with initial condition~$x$ and~$X_n^\varepsilon(x)$ the numerical solution with initial condition~$x$.
For~$x\in\MM^\varepsilon_h$, Proposition~\ref{proposition:weak_uniform_expansion_penalized_dynamic} and Proposition~\ref{proposition:uniform_expansion_integrator} yield
$$
\abs{\E[\phi(X^\varepsilon(h,x))-\phi(X_1^\varepsilon(x))|x]}\leq C (1+\abs{x}^K) h^{3/2},
$$
where~$\phi\in \CC^3_P$.
Lemma~\ref{lemma:growth_properties_exact_sol} gives that~$\phi_n(x)=\E[\phi(X^\varepsilon((n-1)h,x))|x]$ is in~$\CC^3_P$.
We rewrite the global error, given by~$E^\varepsilon_h=\abs{\E[\phi(X^\varepsilon(T,X_0))-\phi(X_N^\varepsilon(X_0))]}$, with a telescopic sum,
\begin{align*}
E^\varepsilon_h
&\leq \sum_{n=1}^N \abs{\E[\phi(X^\varepsilon(nh,X^\varepsilon_{N-n}(X_0)))-\phi(X^\varepsilon((n-1)h,X^\varepsilon_{N-n+1}(X_0)))]}\\
&\leq \sum_{n=1}^N \abs{\E[\phi_n(X^\varepsilon(h,X^\varepsilon_{N-n}(X_0)))-\phi_n(X^\varepsilon_1(X^\varepsilon_{N-n}(X_0)))]}\\
&\leq \sum_{n=1}^N \E[\abs{\E[\phi_n(X^\varepsilon(h,x))-\phi_n(X^\varepsilon_1(x))|x=X^\varepsilon_{N-n}(X_0)]}]\\
&\leq \sum_{n=1}^N C (1+\E[\abs{X^\varepsilon_{N-n}(X_0)}^K]) h^{3/2}
\leq C h^{1/2},
\end{align*}
where we used the bounded moments property of Proposition~\ref{proposition:bounded_moments} and~$X_n^\varepsilon\in \MM^\varepsilon_h$ (Lemma~\ref{lemma:estimate_uniform_zeta}).
\end{proof}
\begin{proof}[Proof of Theorem~\ref{theorem:convergence_integrator_epsilon}]
We obtain straightforwardly from the expression of the integrator~\eqref{equation:UA_integrator} that~$\abs{\zeta(X_{n}^\varepsilon)}\leq C_h\sqrt{\varepsilon}$.
Using this estimate and the notation introduced in the proof of Lemma~\ref{lemma:uniform_expansion_lambda}, we observe that
$$
\abs{Y_{(i)}^\varepsilon(X_n^\varepsilon)-Y_{(i)}^0(X_n^\varepsilon)}\leq C_h\sqrt{\varepsilon},\quad \abs{\zeta_{(i)}^\varepsilon(X_n^\varepsilon)}\leq C_h\sqrt{\varepsilon},\quad i\in\{1/2,1\},
$$
where~$Y_{(1/2)}^0(x)=\sigma\xi_n$ and~$Y_{(1)}^0(x)=f(x)$.
The Lagrange multiplier given by~\eqref{equation:expression_exact_lambda} therefore satisfies~$\abs{\lambda^\varepsilon_{n+1}-\widetilde{\lambda^\varepsilon_{n+1}}}\leq C_h \sqrt{\varepsilon}$, where
$$
\widetilde{\lambda^\varepsilon_{n+1}}=-G_{X_{n+1}^\varepsilon-X_n^\varepsilon}^{-1}(X_n^\varepsilon)\int_0^1 g^T(X_n^\varepsilon+\tau(X_{n+1}^\varepsilon-X_n^\varepsilon))d\tau(\sigma\sqrt{h}\xi_n+hf(X_n^\varepsilon)).
$$
Similarly to~\eqref{equation:expression_exact_lambda}, the Lagrange multiplier~$\lambda^0_{n+1}$ of the constrained Euler integrator~\eqref{equation:Euler_Explicit_constrained} is given by
$$
\lambda^0_{n+1}=-G_{X_{n+1}^0-X_n^0}^{-1}(X_n^0)\int_0^1 g^T(X_n^0+\tau(X_{n+1}^0-X_n^0))d\tau(\sigma\sqrt{h}\xi_n+hf(X_n^0)).
$$
Using that~$G_y^{-1}(x)$ is bounded, a straightforward calculation shows that~$G_y^{-1}(x)$ is Lipschitz in~$x$,~$y\in\R^d$, that is, there exists a constant~$L>0$ such that
$$
\abs{G_{y_1}^{-1}(x_1)-G_{y_2}^{-1}(x_2)}\leq L(\abs{x_1-x_2}+\abs{y_1-y_2}), \quad x_1,x_2,y_1,y_2\in\R^d.
$$
Thus, we get the estimate
$$
\abs{\lambda^\varepsilon_{n+1}-\lambda^0_{n+1}}\leq C_h (\sqrt{\varepsilon} +\abs{X_n^\varepsilon-X_n^0})+C\sqrt{h}\abs{X_{n+1}^\varepsilon-X_{n+1}^0},
$$
and, as~$\lambda^\varepsilon_{n+1}$ and~$\lambda^0_{n+1}$ are bounded uniformly in~$\varepsilon$, we have
$$
\abs{g(X_n^\varepsilon)\lambda^\varepsilon_{n+1}-g(X_n^0)\lambda^0_{n+1}}\leq C_h (\sqrt{\varepsilon} +\abs{X_n^\varepsilon-X_n^0})+C\sqrt{h}\abs{X_{n+1}^\varepsilon-X_{n+1}^0}.
$$
From the definitions of~$X_{n+1}^\varepsilon$ in~\eqref{equation:UA_integrator} and~$X_{n+1}^0$ in~\eqref{equation:Euler_Explicit_constrained}, we deduce that
$$\abs{X_{n+1}^\varepsilon-X_{n+1}^0}\leq C_h (\sqrt{\varepsilon} +\abs{X_n^\varepsilon-X_n^0})+C\sqrt{h}\abs{X_{n+1}^\varepsilon-X_{n+1}^0}.$$
If~$h\leq h_0$ is small enough, we obtain
$$\abs{X_{n+1}^\varepsilon-X_{n+1}^0}\leq C_h (\sqrt{\varepsilon} +\abs{X_n^\varepsilon-X_n^0}).$$
We deduce the estimate~\eqref{equation:weak_CV_epsilon_0_strong} by iterating this inequality for~$n\leq N=T/h$.
\end{proof}
\begin{proof}[Proof of Theorem~\ref{theorem:uniform_fixed_point_problem}]
We take over the notations and the expression~\eqref{equation:expression_exact_lambda} of the Lagrange multiplier~$\lambda^\varepsilon_1$ that we used in the proof of Lemma~\ref{lemma:uniform_expansion_lambda}. Without loss of generality, we concentrate on the first step of the algorithm with~$X_0=x$.
Replacing~$\lambda^\varepsilon_1$ by the explicit formula~\eqref{equation:expression_exact_lambda} in~\eqref{equation:UA_integrator} yields that~$X_1^\varepsilon$ is a fixed point of the following map:
\begin{align*}
F_h^\varepsilon(y)&=x
+\sqrt{h}Y_{(1/2)}^\varepsilon(x)+hY_{(1)}^\varepsilon(x)
+g(x)G_{y-x}^{-1}(x)
\Big[\sqrt{h}\zeta_{(1/2)}^\varepsilon(x)+h\zeta_{(1)}^\varepsilon(x)\\
&-\int_0^1 g^T(x+\tau(y-x))d\tau (\sqrt{h}Y_{(1/2)}^\varepsilon(x)+hY_{(1)}^\varepsilon(x))\Big].
\end{align*}
For~$y_1$,~$y_2\in\R^d$, Assumption~\ref{assumption:regularity_ass} gives
\begin{align*}
\abs{F_h^\varepsilon(y_2)-F_h^\varepsilon(y_1)}
&\leq C\sqrt{h}\abs{y_2-y_1},
\end{align*}
where we used that~$G_y^{-1}(x)$ is Lipschitz in~$x$,~$y\in\R^d$.
We deduce that~$F_h^\varepsilon$ is a uniform contraction for~$h\leq h_0$ small enough.
\end{proof}
\section{Numerical experiments}
\label{section:numerical_experiments}
In this section, we perform numerical experiments to confirm the theoretical findings, on a torus in~$\R^3$ and on the orthogonal group in high dimension and codimension, in the spirit of the experiments in~\cite{Zappa18mco,Zhang19eso,Laurent21ocf}.
\subsection{Uniform approximation for the invariant measure on a torus}
We consider the example in codimension one of a torus in~$\R^3$.
We apply the new method given by the discretization~\eqref{equation:UA_integrator_q_1} for sampling the invariant measure of~\eqref{equation:Langevin_epsilon} for different steps~$h$ and parameters~$\varepsilon$, and we compare it with the Euler integrators~\eqref{equation:Euler_EEE} in~$\R^d$ and~\eqref{equation:Euler_Explicit_constrained} on the manifold.
We recall that for dynamics of the form~\eqref{equation:Langevin_epsilon}, the weak convergence implies the convergence for the invariant measure (see~\cite{Talay90eot}).
The numerical experiments of this subsection hint that the uniform accuracy property also extends to the approximation of the invariant measure, but we leave the mathematical analysis for the invariant measure for future work.
We consider the constraint~$\zeta(x)=(x_1^2+x_2^2+x_3^2+R^2-r^2)^2-4R^2(x_1^2+x_2^2)$, with~$R=3$ and~$r=1$, and we choose the map~$f(x)=-25(x_1-R+r,x_2,x_3)$, with~$\sigma=\sqrt{2}$, the test function~$\phi(x)=\abs{x}^2$,~$M=10^7$ trajectories, the final time~$T=10$, and the initial condition~$X_0=(R-r,0,0)$.
Increasing the value of~$T$ does not modify the computed averages, which hints that we reached the equilibrium. The factor~$25$ in~$f$ confines the solution in a reasonably small neighborhood of the torus, which allows a faster convergence to equilibrium and to take fewer trajectories.
We compute the Monte-Carlo estimator~$\widebar{J}=\frac{1}{M}\sum_{m=1}^M \phi(X_N^{(m)}) \simeq \E[\phi(X_N)]$, where~$X_n^{(m)}$ is the~$m$-th realization of the integrator at time~$t_n=nh$, and~$N$ is an integer satisfying~$Nh=T$.
We compare this approximation with a reference value of~$\int_{\R^d} \phi d\mu_\infty^\varepsilon$ computed with the uniformly accurate integrator, by using a timestep~$h_{\text{ref}}=2^{-12}$.
In the case of the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained}, it amounts to comparing an approximation of~$\int_{\MM} \phi d\mu_\infty^0$ with the reference value of~$\int_{\R^d} \phi d\mu_\infty^\varepsilon$. We observe in Figure~\ref{figure:Plot_torus} that the accuracy of the constrained integrator~\eqref{equation:Euler_Explicit_constrained} for solving the unconstrained problem~\eqref{equation:Langevin_epsilon} deteriorates when~$\varepsilon$ grows larger, as~$\mu_\infty^\varepsilon$ deviates from~$\mu_\infty^0$.
The explicit Euler scheme~\eqref{equation:Euler_EEE} faces stability issues when~$\varepsilon\rightarrow 0$.
The accuracy of the new method for solving~\eqref{equation:Langevin_epsilon} does not deteriorate depending on~$\varepsilon$, and it shares a behavior similar to the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained} when~$\varepsilon\rightarrow 0$, which is in agreement with Theorem~\ref{theorem:convergence_integrator_epsilon}.
The right graph of Figure~\ref{figure:Plot_torus} shows that the behavior of the error in~$\varepsilon$ is the same for any fixed value of~$h$. This is a numerical confirmation of the uniform accuracy property of the discretization~\eqref{equation:UA_integrator} (in the spirit of the numerical experiments in~\cite{Chartier15uan}), as stated in Theorem~\ref{theorem:uniform_consistency_UA_scheme}.
For any fixed $\varepsilon$, the right graph of Figure~\ref{figure:Plot_torus} also shows that the error decreases when $h\rightarrow 0$. A plot of the error against $h$ for a fixed $\varepsilon$ (not included for conciseness) shows a slope of order one (see Remark \ref{remark:discussion_order_UA_method}).
\begin{figure}[ht]
\begin{minipage}[c]{.49\linewidth}
\begin{center}
\includegraphics[scale=0.5]{Images/Comparison_methods_torus}
\end{center}
\end{minipage} \hfill
\begin{minipage}[c]{.49\linewidth}
\begin{center}
\includegraphics[scale=0.5]{Images/Uniform_accuracy_torus}
\end{center}
\end{minipage}
\caption{Error for sampling the invariant measure of penalized Langevin dynamics on a torus in~$\R^3$ of the uniform discretization~\eqref{equation:UA_integrator} and the Euler integrators~\eqref{equation:Euler_EEE} and~\eqref{equation:Euler_Explicit_constrained} for different values of~$\varepsilon$ with~$h=2^{-9}T$ (left), and error curves versus~$\varepsilon$ of the uniformly accurate method for different timesteps~$h=2^{-i}T$ and~$i=6,\dots,10$ (right), with the final time~$T=10$, the maps~$f(x)=-25(x_1-R+r,x_2,x_3)$,~$\phi(x)=\abs{x}^2$, and~$M=10^7$ trajectories.}
\label{figure:Plot_torus}
\end{figure}
\subsection{Weak approximation on the orthogonal group}
We apply the uniformly accurate method on a compact Lie group (in the spirit of the numerical experiments in~\cite{Zappa18mco,Zhang19eso}) to see how it performs in high dimension and codimension. We choose the orthogonal group~$\Or(m)=\{M\in\R^{m\times m},M^T M=I_m\}$, seen as a submanifold of~$\R^{m^2}$ of codimension~$q=m(m+1)/2$.
We compare the explicit Euler scheme~\eqref{equation:Euler_EEE}, the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained}, and the new method on~$\MM=\Or(m)$ for~$m=2,\dots,5$ with the parameters~$\varepsilon=0.005$,~$T=1$, and~$h=2^{-7}$.
Note that, as~$h$ and~$\varepsilon$ share the same order of magnitude, the explicit Euler scheme~\eqref{equation:Euler_EEE} can face stability issues, and the solution does not lie on the manifold~$\MM$. Thus, we are in the regime where the convergence results for both Euler schemes~\eqref{equation:Euler_EEE}-\eqref{equation:Euler_Explicit_constrained} do not apply.
We choose~$f=-\nabla V$, where~$V$ is given by
\begin{equation}
\label{equation:potential_O}
V(x)=50\Trace((x-I_{m^2})^T(x-I_{m^2}))
\end{equation}
with the parameters~$\sigma=\sqrt{2}$,~$X_0=I_d$,~$\phi(x)=\Trace(x)$, and~$M=10^6$ trajectories.
The reference solution for~$J(m)=\E[\phi(X^\varepsilon(T))]$ is computed with the uniformly accurate integrator with~$h_{\text{ref}}=2^{-9}$.
With the factor~$50$ in the potential~\eqref{equation:potential_O}, the trajectories stay close to~$I_{m^2}$, and~$J(m)$ is close to~$\phi(I_{m^2})=m$.
This choice of factor permits one to explore a reasonably small area of~$\Or(m)$, in order to avoid zones close to~$\MM$ where the Gram matrix~$G$ has a bad condition number or is singular, and to reduce the number of trajectories needed.
We observe numerically that replacing the factor~$50$ by~$1$ in~\eqref{equation:potential_O} induces a severe timestep restriction.
We present the results of the experiment in Table~\ref{Table:numerical_results_O}. We omit the results for the explicit Euler scheme~\eqref{equation:Euler_EEE} as the method is inaccurate in this regime (error of size~$1$).
We observe that, in the regime where~$h$ and~$\varepsilon$ share the same order of magnitude, the uniformly accurate integrator performs significantly better than the Euler schemes~\eqref{equation:Euler_Explicit_constrained} and~\eqref{equation:Euler_EEE} for solving the problem~\eqref{equation:Langevin_epsilon}.
In this regime, the Euler method~\eqref{equation:Euler_EEE} faces stability issues, and it is inappropriate to use the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained} as the solution~$X^\varepsilon(t)$ of~\eqref{equation:Langevin_epsilon} is not close to the solution~$X^0(t)$ of~\eqref{equation:projected_Langevin}.
Moreover, the cost in time of the new method stays the same in average for any value of~$\varepsilon$ (results not included for conciseness).
This confirms numerically the uniform cost of solving the fixed point problem~\eqref{equation:UA_integrator}, as stated in Theorem~\ref{theorem:uniform_fixed_point_problem}.
\begin{table}[H]
\setcellgapes{3pt}
\centering
\begin{tabular}{|c||c|c||c||c|c||c|c|}
\hline
$m$ &~$\dim(\MM)$ &~$q$ & \multicolumn{1}{c||}{$J(m)$} &~$\widebar{J}_\text{UA}$ & \multicolumn{1}{c||}{error of~$\widebar{J}_\text{UA}$} &~$\widebar{J}_\text{EC}$ & error of~$\widebar{J}_\text{EC}$\\
\hhline{|=||=|=||=||=|=||=|=|}
$2$ & 1 & 3 &~$2.00934$ &~$2.00619$ &~$3.1 \cdot 10^{-3}$ &~$1.99165$ &~$1.8 \cdot 10^{-2}$\\
\hline
$3$ & 3 & 6 &~$3.01458$ &~$3.00821$ &~$6.4 \cdot 10^{-3}$ &~$2.97460$ &~$4.0 \cdot 10^{-2}$\\
\hline
$4$ & 6 & 10 &~$4.02050$ &~$4.00972$ &~$1.1 \cdot 10^{-2}$ &~$3.94846$ &~$7.2 \cdot 10^{-2}$\\
\hline
$5$ & 10 & 15 &~$5.02669$ &~$5.00842$ &~$1.8 \cdot 10^{-2}$ &~$4.91298$ &~$1.1 \cdot 10^{-1}$\\
\hline
\end{tabular}
\caption{Numerical approximation of~$J(m)=\E[\phi(X^\varepsilon(T))]$ for~$2\leq m \leq 5$ with the estimator~$\widebar{J}=M^{-1}\sum_{k=1}^M \phi(X_N^{(k)})$, where~$(X_n)$ is given by the uniform discretization~\eqref{equation:UA_integrator} for~$\widebar{J}_\text{UA}$ and by the constrained Euler scheme~\eqref{equation:Euler_Explicit_constrained} for~$\widebar{J}_\text{EC}$ with their respective errors. The average is taken over~$M=10^6$ trajectories with the potential~\eqref{equation:potential_O},~$\phi(x)=\Trace(x)$, the final time~$T=1$, the stiff parameter~$\varepsilon=0.005$, and the timestep~$h=2^{-7}$.}
\label{Table:numerical_results_O}
\setcellgapes{1pt}
\end{table}
\section{Conclusion and future work}
\label{section:future_work}
In this work, we presented a new method for the weak numerical integration of penalized Langevin dynamics evolving in the vicinity of manifolds of any dimension and codimension.
On the contrary of the other existing discretizations, the accuracy of the proposed integrator is independent of the size of the stiff parameter~$\varepsilon$.
Moreover, its cost does not depend on~$\varepsilon$, and it converges to the Euler scheme on the manifold when~$\varepsilon\rightarrow 0$.
Throughout the analysis, we gave an expansion in time of the solution to the penalized Langevin dynamics that is uniform in~$\varepsilon$, as well as new tools for the study of stochastic projection methods for solving stiff SDEs.
Multiple questions arise from the work presented in this paper, with many of great interest for physical applications.
First, it would be interesting to get convergence results with weaker assumptions or to develop uniformly accurate integrators for different penalized dynamics with the same limit when~$\varepsilon \rightarrow 0$ such as the original penalized dynamics~\eqref{equation:modified_Langevin_epsilon} (see~\cite{Ciccotti08pod}).
One could build integrators for penalized dynamics of the form
$$
dX^\varepsilon=f(X^\varepsilon) dt +\sigma dW+\frac{\sigma^2}{4}\nabla\ln(\det(G)) -\frac{1}{\varepsilon}(g G^{-1}\zeta_1)(X^\varepsilon)dt-\frac{1}{\nu}(g G^{-1}\zeta_2)(X^\varepsilon)dt,
$$
where~$\varepsilon$ and~$\nu$ do not share the same order of magnitude, or for constrained dynamics with a penalized term.
One could also build a uniformly accurate numerical scheme with high order in the weak context, or just in the context of the invariant measure (in the spirit of the works~\cite{BouRabee10lra,Leimkuhler13rco,Abdulle14hon,Abdulle15lta,Leimkuhler16tco,Laurent20eab,Laurent21ocf} where numerical schemes of high order for the invariant measure and weak order one were introduced).
Postprocessors~\cite{Vilmart15pif} proved to be an efficient tool for reaching high order for the invariant measure without increasing the cost of the method and could be used in this context.
Moreover, the order conditions presented in~\cite{Laurent20eab} for Runge-Kutta methods for solving Langevin dynamics in~$\R^d$ both in the weak sense and for the invariant measure do not match with the order conditions for solving Langevin dynamics constrained on the manifold~$\MM$, as presented in~\cite{Laurent21ocf}. It would be interesting to create a unified class of high order Runge-Kutta methods with the same order conditions in~$\R^d$, on the manifold~$\MM$ and in the vicinity of~$\MM$.
The discretizations presented in this paper could also be combined with Metropolis-Hastings rejection procedures~\cite{Metropolis53eos,Hastings70mcs}, in the spirit of the works~\cite{Girolami11rml,Brubaker12afo,Lelievre12ldw,Zappa18mco,Lelievre19hmc}, in order to get an exact approximation for the invariant measure with a rejection rate that does not deteriorate in the regime~$\varepsilon\rightarrow 0$.
\bigskip
\noindent \textbf{Acknowledgements.}\
The author would like to thank Gilles Vilmart for helpful discussions.
This work was partially supported by the Swiss National Science Foundation, grants No.\thinspace 200020\_184614, No.\thinspace 200021\_162404 and No.\thinspace 200020\_178752.
The computations were performed at the University of Geneva on the Baobab cluster using the Julia programming language.
\bibliographystyle{abbrv}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 9,254
|
Bibliografia
.
Nacque a Homberg-Meiersberg il 9 ottobre 1921. Arruolatosi nella Luftwaffe come fahnenjunker iniziò il suo addestramento a Pardubitz, Cecoslovacchia, in seno al Fliegerausbildungsregiment 32.
Il 1º aprile 1940 fu promosso gefreiter, il 1º giugno unteroffizier e il 1º settembre 1940 fähnrich. Dopo l'addestramento basico trascorse diversi mesi volando su velivoli monomotore e conseguì il brevetto di pilota il 21 settembre 1940. Il 1 febbraio 1941 fu promosso oberfähnrich e fu successivamente inviato a Zeltweg, Austria, presso la C-Schule per l'addestramento al pilotaggio di velivoli plurimotore. Qui volo sui cacciabombardieri Messerschmitt Bf 110. Al momento del conseguimento del brevetto di pilota di plurimotori, il 1 aprile 1941, fu promosso tenente. Il 18 maggio si offrì volontario per l'assegnazione alla caccia notturna, e dopo sei settimane di addestramento al volo senza visibilità, fu assegnato alla al 3° staffel dello Nachtjagdgeschwader 1 (NJG 1)), insieme al suo operatore radio/radar, il gefreiter Albrecht Risop.
Ottenne la sua prima vittoria aerea nella notte tra il 26 e il 27 marzo 1942, quando abbatté un bombardiere bimotore della Vickers Wellington della Royal Air Force. Dopo aver abbattuto un altro Wellington il 17 giugno, attaccò quindi un bombardiere quadrimotore Short Stirling della RAF ma il suo Bf 110 D-3 (W.Nr. 4224) "G9+FL" fu colpito dal fuoco di risposta dei mitraglieri del bombardiere, che uccise il suo operatore radio/radar e lo ferì alla gamba sinistra. Riuscito a uscire dall'aereo in fiamme, si lanciò con il paracadute salvandosi. Trascorrere molto tempo in ospedale per curare le ustioni e la ferita alla gamba, ritornando in servizio nel luglio 1942. Il 1º ottobre lo 3./NJG 1 fu rinominato 1./NJG 5, e nel mese di dicembre 5./NJG 5. Nel corso del 1943 conseguì ulteriori sei vittorie aeree, a spese di un Wellington, due Stirling, due Handley Page Halifax e un Avro Lancaster.
Nel gennaio 1944 conseguì altre sette vittorie, fra cui tre bombardieri quadrimotori Lancaster della RAF abbattuti nella notte tra il 27 e il 28 gennaio. Conseguì un'altra tripletta nella notte del 15/16 febbraio (16-18) a spese di altrettanti Lancaster. Il 3 marzo 1944 fu nominato Staffelkapitän del 6./NJG 5. Nella notte tra il 27 e il 28 aprile abbatté un bombardiere quadrimotore Lancaster della RAF. Preso contatto con un altro Lancaster sul Lago di Costanza diretto a ovest, verso la Svizzera, non esitò ad attaccare il bombardiere ma il suo Bf 110 G-4 (W.Nr. 740 055) "C9+EN" fu colpito dal fuoco del Lancaster e il motore di babordo si incendiò. Condotto verso terra dai proiettori della difesa contraerea svizzera, e con un motore spento, fu costretto ad atterrare all'aeroporto di Zurigo-Dübendorf, venendo internato insieme agli altri membri dell'equipaggio. Vennero rimpatriati alcuni giorni dopo in mezzo a molte manovre politiche e di spionaggio, in quanto il Bf 110 G-4 e il suo radar erano considerati modelli i cui segreti erano da proteggere ad ogni costo. Nominato Staffelkapitän della 8./NJG 6 il 10 maggio 1944, con sede in Ungheria, in poco più di due mesi conseguì ulteriori 11 vittorie, inclusi quattro bombardieri bimotori North American B-25 Mitchell sovietici. Il 23 luglio, promosso oberleutnant, venne insignito della Croce di Cavaliere della Croce di Ferro (Ritterkreuz) per aver conseguito 33 vittorie. Nell'autunno di quell'anno venne nominato Gruppenkommandeur del III./NJG 6, comandando questa unità fino alla fine della guerra. Fu promosso hauptmann nel febbraio 1945. Dopo la fine del conflitto riprese a studiare frequentando l'università e conseguendo una laurea in ingegneria. Nel 1952 lavorò con il professor Willy Messerschmitt prima di intraprendere con successo una propria attività nel campo dell'ingegneria edile Si spense a Überlingen il 7 febbraio 2002.
Onorificenze
Pubblicazioni
Duell unter den Sternen. Tatsachenbericht eines deutschen Nachtjägers 1941–1945, Flechsig, Würzburg, 2009, ISBN 978-3-8035-0003-8.
Note
Annotazioni
Fonti
Bibliografia
Periodici
Voci correlate
Leopold Fellerer
Wilhelm Joswig
Collegamenti esterni
Assi dell'aviazione tedesca della seconda guerra mondiale
Piloti da caccia della Luftwaffe
Cavalieri dell'Ordine della Croce di Ferro
Croci di Ferro di prima classe
Aviatori tedeschi della seconda guerra mondiale
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 5,748
|
Q: SQL Server - Breadcrumb Type Effect I have a set of data such as the following:
ID, Description, ParentID
1, Savoury, -1
2, Cheese, 1
3, Pastry, 1
4, Quiche, 1
5, Sweet, -1
6, Chocolate, 5
...
What I'm looking to do is write a stored procedure which will produce a set of results similar to a breadcrumb link in a web browser. Something like the following would be ideal...
ID, BreadCrumb
1, Savoury
2, Savoury >> Cheese
3, Savoury >> Pastry
4, Savoury >> Quiche
5, Sweet
6, Sweet >> Chocolate
...
Items such as Cheese or Pastry wouldn't be listed as items on their own (like Sweet is). So far, I have the following code which works but it lists everything, regardless of whether it has a ParentID
With BreadCrumb AS
(
SELECT CAST(a.Description AS VARCHAR(100)) AS Path, a.ID, a.ParentID
FROM FoodStuff a
UNION ALL
SELECT CAST(BreadCrumb.Path + ' >> ' + b.Description AS VARCHAR(100)) AS Path, b.ID, b.ParentID
FROM FoodStuff b
INNER JOIN BreadCrumb ON BreadCrumb.ID = b.ParentID
)
SELECT * FROM BreadCrumb
ORDER BY Path
I'd be grateful for a nudge in the right direction.
Thanks in advance,
Kev
A: You need a condition in the anchor portion of the recursive CTE in order to limit it to root items:
With BreadCrumb AS
(
SELECT CAST(a.Description AS VARCHAR(100)) AS Path, a.ID, a.ParentID
FROM FoodStuff a
WHERE a.ParentID = -1 -- select roots only
UNION ALL
SELECT CAST(BreadCrumb.Path + ' >> ' + b.Description AS VARCHAR(100)) AS Path, b.ID, b.ParentID
FROM FoodStuff b
INNER JOIN BreadCrumb ON BreadCrumb.ID = b.ParentID
)
You don't need a similar condition in the recursive half of the CTE as the inner join should take care of it, since presumably you don't have any records with ID = -1.
As a best-practices aside, I'd recommend using NULL instead of -1 for foreign keys which are allowed not to have a value, such as ParentID.
SQL Fiddle, courtesy of @Conrad Frix.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 4,484
|
\section{Introduction}
The private retrieval of information from public databases has received
significant attention already for several decades from researchers in the
computer science community (see, e.g., \cite{Chor_etal1998,
Yekhanin2010}). While this line of work, commonly known as private information
retrieval (PIR), is concerned with downloading individual messages in a
private manner from databases, a recently proposed generalization of this
problem \cite{Maddah-Aliarxiv2017, SunJafararxiv_2017} addresses the private
computation of functions of these messages. In accordance with \cite{Maddah-Aliarxiv2017} we denote this approach as private
function retrieval (PFR) in the following. In PFR a user has access
to a given number of databases and intends to compute a function of messages
stored in these databases. This function is kept private from the
databases, as they may be under the control of an eavesdropper. Both works \cite{Maddah-Aliarxiv2017, SunJafararxiv_2017}
characterize the fundamental information theoretic communication overhead
needed to reliably compute the given function and specify the corresponding
capacity and achievable rates as a function of the message size, the number
of messages, and the number of databases, respectively. Further, the authors
assume that the data is replicated on each database. Surprisingly, the obtained PFR capacity
result is equal to the PIR capacity of
\cite{Sun_Jafar2015}.
However, although repetition coding adds the largest amount of redundancy and
thus protects effectively against erasures, it is associated with a large
storage cost. A more general way to optimally trade-off the available
redundancy (or rate) versus the erasure correcting capability is given by MDS
codes. In particular, for an $(N,K)$ MDS code with $N$ code symbols and $K$
information symbols and rate $R_c=K/N$ $N-K$ erasures can be
recovered from any $K$ code symbols. Coded PIR has been
addressed in two different lines of work. Achievable schemes for MDS coded
PIR have been presented in \cite{Chanetal_ISIT2015, Tajeddineetal_ISIT2016}
and the capacity has been established in
\cite{BanawanUlukus_Globecom2017}. On the other hand, in \cite{Yaakoobi_ISIT2015} linear codes with $k$
different reconstruction sets for each code symbol have been proposed in
form of so called $k$-server PIR.
In this paper we propose coded PFR, which to the best of our knowledge has
not been addressed yet in the recent literature, with the notable exception
of the parallel work in \cite{Karpuk}, which is based on a fixed ($k$-)
server PIR scheme with the inclusion of colluding databases. Our scheme is
based on MDS codes which in contrast to \cite{Karpuk} minimize the storage
overhead and maximize the achievable download rate. In particular, we provide a
characterization of the achievable rate of MDS coded PFR if the user wishes
to compute an arbitrary linear combination of $M$ independent equal-sized
messages over some finite field $\mathbb{F}_q$, distributed over $N$
non-colluding MDS-coded databases. Surprisingly, our achievable rate matches
the capacity for MDS coded PIR in \cite{BanawanUlukus_Globecom2017}. This
demonstrates that, compared to the naive scheme, where $M$ coded messages are
downloaded and linearly combined offline at the user (requiring $M$-times
the coded PIR rate), downloading the result of the computation privately and
directly from the databases does not incur any penalty in rate compared to
the coded PIR case. Thus, our result strictly generalizes the achievable schemes in
\cite{Maddah-Aliarxiv2017, SunJafararxiv_2017} which represent special
cases of our proposed PFR scheme.
\vspace*{-0.5ex}
\section{Problem Statement} \vspace*{-0.5ex}\label{ProblemStatment}
In the following, we use {$[1:X]$} to denote the set $\{1,\dots,X\}$. Similarly, $X_{1:N}=\{X_1,\dots, X_N\}$.
\vspace{-0.5ex}
\subsection{System Model}\vspace{-0.5ex}
In coded PFR, a user wishes to privately retrieve a linear combination of the messages stored in the databases such that the coefficients of the linear combination are kept secret from each individual database. Consider a linear distributed storage system storing $M$ equal-sized messages on $N$ non-colluding databases. The message $W_m, m\in [1:M],$ is composed from $L$ symbols chosen independently and uniformly at random from the finite field $\mathbb{F}_{q}$ with
\begin{align}
H(W_1)=& \dots = H(W_M)=L \log q,\\
\hspace{-0.45ex}H(W_1,\!\dots\!,W_M)=& H(W_1)+\!\dots\! + H(W_M)=ML \log q. \vspace*{-0.5ex}
\end{align}
Each message is divided into $\tilde{L}$ segments, each of $K$ symbols, forming a $\tilde{L}\times K$ matrix, where $L=\tilde{L} K$. The messages are stored using an $(N,K)$ MDS code with the full rank generator matrix defined by
\begin{equation} \vspace*{-0.5ex}
{\bf G}=\big[ {\bf{g}}_1\quad {\bf{g}}_2\quad \dots\quad {\bf{g}}_N\big]_{K\times N},
\end{equation}
with ${\bf{g}}_n,\;n\in[1:N],$ denoting the $n$-th column vector of ${\bf G}.$
The generator matrix produces a code that can tolerate up to $N-K$ erasures by retrieving data from any set ${\mathcal{K}}\subset \{1,\dots,N\}$ databases, where $|{\mathcal K}|\geq K$. The encoding process for message $W_m$ is defined as follows:
\begin{equation}
\hspace{-4ex}\begin{aligned}
{\begin{bmatrix}{\bf{w}}_{m,t}\end{bmatrix}}_{1\times K} &{\begin{bmatrix} {\bf{g}}_1& {\bf{g}}_2& \dots& {\bf{g}}_N\end{bmatrix}}_{K\times N}\\
&\quad\qquad= {\begin{bmatrix} {\bf{g}}_1^T{\bf{w}}_{m,t}&\dots&{\bf{g}}_N^T{\bf{w}}_{m,t}\\ \end{bmatrix}}_{1\times N},
\end{aligned}
\end{equation}
where ${\bf{w}}_{m,t},$ $\forall m\in[1:M], \forall t\in[1:\tilde{L}],$ denotes the $K$-dimensional vector of symbols of the $t$-th segment from the message $W_m$. The resulting $N$ coded symbols for each segment are then distributed over the $N$ databases, and the code rate is given by $R_c=\frac{K}{N}.$
Consequently, the code symbols stored at each database $n\in [1:N]$ are given by
\begin{equation}\label{eq:codedDatabse}
{\bf W}_{DB_n}=
\begin{bmatrix}
{\bf{g}}_n^T{\bf{w}}_{1,1} & {\bf{g}}_n^T{\bf{w}}_{1,2} &\dots& {\bf{g}}_n^T{\bf{w}}_{1,\tilde{L}}\\
\vdots &\vdots& \ddots & \vdots\\
{\bf{g}}_n^T{\bf{w}}_{M,1} & {\bf{g}}_n^T{\bf{w}}_{M,2} &\dots& {\bf{g}}_n^T{\bf{w}}_{M,\tilde{L}}\\
\end{bmatrix},
\end{equation}
\noindent where we use ${\bf{W}}[t]$ to denote the $t$-th column, and $W_{m}(t)$ for the element of the $m$-th row and $t$-th column of the database, respectively.
In PFR, the linear combination $\nu$ the user intends to retrieve is represented as
\begin{align}
\widetilde{W}_\nu &= {\bf{v_\nu} }[W_1,\dots,W_M]^T \label{eq:v_msg}\\
&=v_{\nu}(1)W_1+\dots+v_{\nu}(M)W_M \label{eq:v_msg1}\\
&= \begin{bmatrix}{\bf{v_\nu} }{\bf{W}}[1]& \dots & {\bf{v_\nu} }{\bf{W}}[\tilde{L}]\end{bmatrix}\!, \label{eq:v_ms3}
\end{align}
where ${\bf v_\nu}$ is an $M$-dimensional non-zero coefficient vector of the linear combination (row vector) indexed by $\nu$, the coefficients $v_{\nu}(m), \;\forall m\in[1:M],$ are chosen from the finite field $\mathbb{F}_{q}$, and the addition ``$+$''~is done element-wise over the same field. We assume that the vector ${\bf v_\nu}$ is an element of the set ${\cal V}$ that contains all possible distinct $M$-dimensional vectors defined over ${\mathbb F_q}$ where $\nu \in [1:V],\;V= |{\cal V}|=\frac{q^M-1}{q-1}$.
In order for the user to retrieve the linear combination $\widetilde{W}_\nu$, while keeping $\nu$ secret from each database, it generates $N$ query matrices for the databases $\{Q_1^{[\nu]},\dots,Q_N^{[\nu]}\}$. Since the query matrices are generated by the user without prior knowledge of the realizations of the stored messages, the queries must be independent of the messages,
\begin{equation}\vspace*{-0.5ex}
I(Q_1^{[\nu]},\dots,Q_N^{[\nu]};W_1,\dots,W_M)=0, \quad \forall \nu\in[1:V].
\end{equation}
Upon the reception of the query $Q_n^{[\nu]}$, the $n$-th database generates an answer string $A_n^{[\nu]}$ as a deterministic function of the received query and the stored symbols from each message. Hence,
\begin{equation}\vspace*{-0.5ex}
\!H(A_n^{[\nu]}|Q_n^{[\nu]},{\bf W}_{DB_n})=0, \quad\! \forall \nu\in[1:V],\forall n\in[1:N]. \!\!
\end{equation}
To maintain user privacy, the query-answer function must be identically distributed for each possible linear combination $\nu\in[1:V]$ from the perspective of each database $n\in[1:N]$. In other words, the scheme's queries and answers strings must be independent from the desired linear combination index, therefore the following privacy constraint must be satisfied:\vspace*{-1ex}
\begin{equation}\label{privacy_const}
\begin{aligned}
I(A_n^{[\nu]},Q_n^{[\nu]},{\bf W}_{DB_n};\nu)=0, \!\quad\! \forall \nu\in[1:V].\!
\end{aligned}
\end{equation}
After the user receives all answer strings from each database, the user must be able to reliably decode the desired linear combination message $\widetilde{W}_\nu$ with a probability of error $P_e$ that goes to zero as the message size $L$ approaches infinity. Following Fano's inequality, this translates to the decodability constraint
\vspace*{-1ex}\begin{equation}\label{decoding_const}
H(\widetilde{W}_\nu |A_{1:N}^{[\nu]},Q_{1:N}^{[\nu]})= o(L),
\end{equation}
\noindent where $o(L)$ represents any function of $L$, $f(L)$, that satisfies $\lim_{L\rightarrow \infty} f(L)/L\rightarrow 0.$
The retrieval rate of the coded PFR scheme is characterized by the message length $L$, the query structure $Q,$ and the query-answer function, and is defined as the ratio between the size of the desired linear combination message and the total number of downloaded symbols in bits as
\vspace*{-1ex}\begin{equation}\label{eq:PFR_rate_def}
R=\frac{H(\widetilde{W}_\nu)}{\sum_{n=1}^{N} H(A_n^{[\nu]})} .
\end{equation}
A rate $R$ is said to be achievable if there exist a sequence of coded PFR schemes that satisfy the privacy and correctness constraints of \eqref{privacy_const}, \eqref{decoding_const} for $P_e\rightarrow 0$ as $L\rightarrow \infty$.
\vspace*{-0.5ex}
\section{Achievable Rate of MDS Coded PFR} \label{MainResults}
\begin{theorem}
For an $(N,K)$ coded distributed storage system with code rate $R_c=\frac{K}{N}$, $M$ messages and a set of $V$ linear combinations defined over the field $\mathbb{F}_{q}$, a PFR achievable rate is given as \vspace*{-2ex}
\begin{align}
\qquad \qquad R&\leq \frac{1-R_c}{1-R_c^M} \label{eq:capacity1}\\
&=\Big(1+\frac{K}{N}+\frac{K^2}{N^2}+\dots+\frac{K^{M-1}}{N^{M-1}}\Big)^{-1}. \label{eq:capacity}
\end{align}\vspace{-2ex}
\end{theorem}
\begin{remark}
This achievable rate generalizes the achievable rate of repetition coded PFR \cite{SunJafararxiv_2017} which corresponds to the special case of $K\!=\!1$.
Also, \eqref{eq:capacity1} is only a function of the distributed storage
coding rate $R_c$ and the number of stored independent messages $M,$ and is
universal in the sense that it does not depend on the number of linear
combinations $V$ defined over the finite field $\mathbb{F}_{q}$ nor on the
explicit structure of the code.
\end{remark}
\begin{remark}
If we consider each of the $V$ linear combinations of messages in \eqref{eq:v_msg} as a new \emph{virtual message} $\widetilde{W}_\nu$, and then apply the coded PIR scheme of \cite{BanawanUlukus_Globecom2017}, the scheme rate will be $\frac{1-R_c}{1-R_c^V}$ which is smaller than \eqref{eq:capacity1} since $M\leq V$.
\end{remark}
\begin{remark}
When the linear combination set ${\cal V}$ is reduced to the first $M$
linear combinations
(i.e.,~${\bf{v}}_{1:M} \in {\cal V}:
[{\bf{v}}_1\;{\bf{v}}_2\;\dots\;{\bf{v}}_M]={\bf I}_M$),
the achievable rate of \eqref{eq:capacity1} is tight. That is
because in this setting the problem of coded PFR is reduced to coded PIR
where the converse is implied from \cite{BanawanUlukus_Globecom2017}.
Also, we note that \eqref{eq:capacity1} is equivalent to the coded PIR
capacity \cite{BanawanUlukus_Globecom2017}, which has been observed in
\cite{SunJafararxiv_2017} for $K=1$. Thus,
downloading linear combinations of messages does not incur additional costs
over downloading individual messages.
\end{remark}
\begin{remark}
Eq. \eqref{eq:capacity1} is a strictly decreasing function in the number of messages $M$ for fixed $R_c$. As the number of messages increases $M\rightarrow \infty$, the achievable rate approaches $1-R_c$. Moreover, as $R_c \rightarrow 1$ in \eqref{eq:capacity}, $R \rightarrow \frac{1}{M},$ indicating that to maintain the privacy of the desired linear combination, the user must download all the messages and perform the computation off-line.
\end{remark}
\vspace*{-1ex}
\section{Proof of Theorem~1} \label{AchievableScheme}
\vspace*{-0.5ex}
\subsection{Query generation}
\vspace*{-0.5ex}
The generation of the queries is shown in Algorithm~1.
Let $B\in[1:V]$ be the block indicator and $R\in[1:K]$ be the repetition indicator, respectively. Let the $v$-sum be the combination of $v$ distinct elements out of $V$ elements. Since we have $V \choose v$ different combinations, we denote each different combination as a \emph{type} of the $v$-sum. Let the components of these combinations be symbols of the $V$ virtual messages. As mentioned above, we generate the query set for each database in blocks, where a block represents a group of all ${V\choose v}$ types of $v$-sums for all $v \in[1:V],$ resulting in $V$ blocks in total. To this end, we let the size of the \emph{dependent} virtual messages to be $L=KN^V$ (i.e.,~${\tilde{L}=N^V}$).
For a desired linear combination $\nu \in[1:V]$ we use the notation $Q^{[\nu]}(DB_B)$ to indicate the query set of the database $DB_B\in[1:N]$. This set is composed from $VK$ disjoint subsets $Q^{[\nu]}_{B,R}(DB_B)$ generated for each block $B$ and repetition $R$. We require $K^{V-B}(N-K)^{B-1}$ distinct instances of each type of $v$-sum for every set $Q^{[\nu]}_{B,R}(DB_B)$. Each block and repetition subset is further subdivided into two subsets: the first subset $Q^{[\nu]}_{B,R}(DB_B,{\cal M})$ consists of the $v$-sum types with symbols from the desired linear combination, and the second subset $Q^{[\nu]}_{B,R}(DB_B,{\cal I})$ contains only $v$-sum types with symbols from undesired linear combinations. The query sets for all databases are generated by Algorithm 1 with the following procedures.
\indent {\bf \textit{1) Index assignment}}: In the MDS-coded PIR scheme
\cite{BanawanUlukus_Globecom2017}, the user privately applies a random
permutation over the coded symbols of each message independently. The goal
is to make the coded symbols queried from each database to appear to be
chosen randomly and independently from the desired message. However, for the
PFR problem the linear function is computed element-wise, thus there is a
dependency across the symbols with the same index, which must be maintained
under a permutation. To this end, we modify the permutation to be fixed
across all messages. Let $\pi(\cdot)$ be a random permutation function over
$[1:\tilde{L}]$. We use the notation $U_\nu(t)$, where
\begin{equation}
U_\nu(t)\triangleq \sigma_t\widetilde{W}_\nu(\pi(t))=\sigma_t{\bf v}_{\nu}{\bf{W}}[\pi(t)],
\end{equation}
to indicate the permuted message symbol from the virtual message $\widetilde{W}_\nu$. The random variable $\sigma$ is used to indicate the sign assigned to each individual virtual message symbol, $\sigma_t \in \{+1,-1\}$ \cite{SunJafararxiv_2017}. Both $\sigma_t$ and $\pi$ are randomly selected privately and uniformly by the user.
\indent {\bf \textit{2) Block ${B=1}$}}: This block is described by Steps {3} to {10} of Algorithm 1, where we have $v=1$ for the $v$-sum.\\
\indent {\textit{Initialization:}}
In the initialization step, the user queries the first database {${DB_1=1}$} for $K^{V-1}$ distinct symbols from the desired linear combination $U_\nu(i)$. This is done by calling the function "$\text{new}(U_\nu)$" that will select a symbol from message $U_\nu$ with a new index $i$ each time it is called (Step 6).
\indent {\textit{Database symmetry:}}
Database symmetry is obtained via the ``For'' loop in Step {3}, resulting in a total number of $N K^{V-1}$ symbols over all databases.
\indent {\textit{Message symmetry:}}
In Step {7}, to maintain message symmetry, the user ask each database for the same number of distinct symbols of all other linear combinations $U_\theta(i),\;\theta\in\{1,\dots,V\}\!\setminus\!\{\nu\},$ resulting in a total number of $NVK^{V-1}$ symbols. As a result, the query sets for each database are symmetric with respect to all linear combination vectors in $[1:V]$. We associate the symbols of undesired messages in $K$ groups $G\in{[1:K]}$ to be exploited as distinct side information for different rounds of the scheme as shown in Step {7.}
\indent {\bf \textit{3) Side-information exploitation:}}
In Steps {11} to {20}, we generate the blocks {$B\in[2:V]$} by applying two subroutines ``Exploit-SI'' and ``M-Sym'', respectively. We first use the subroutine ``Exploit-SI'' \cite{SunJafararxiv_2017} to generate queries for new symbols of the desired linear combination $U_\nu$ by combining these symbols with different side information groups from the previous block associated with $N-K$ neighboring databases, as shown in Step {13}. This is required by our proposed MDS coded scheme to ensure privacy and is in contrast to \cite{SunJafararxiv_2017}, where the side information of previous blocks from \emph{all} databases is utilized.
Then, the subroutine ``M-Sym'' \cite{SunJafararxiv_2017} is used to generate side information to be exploited in the following blocks. This subroutine select symbols of undesired messages to generate $v$-sums that enforce symmetry in the block queries. For example in $B=2$, if we have the queries $U_\nu(i)+U_2(j)$, and $U_\nu(l)+U_3(r) \in Q^{[\nu]}_{2,R}(DB_2,{\cal M})$, this subroutine will generate $U_2(l)+U_3(i)$. As a result, we can show that the symmetry over the linear combinations and databases is maintained. By the end of this step we have in total $N{V\choose B} K^{V-B}(N-K)^{B-1}$ queries for each block from all databases.
\indent {\bf \textit{4) Generation of further query rounds:}}
We require further query rounds to obtain $K$ linear equations for each coded symbol to be able to decode. To this end, we circularly shift the order of the database at each repetition. The shift is done for the initial block, $B=1$, in Steps {22} to {25}. However, for the following blocks we only rotate the indices of desired messages $U_\nu$ and combine them with new groups of side information from the neighboring databases from the first round as seen in Steps {26} to {33}. This rotation and side information exploitation for $B\in[2:V]$ is done using the subroutine ``Reuse-SI'' (omitted in the interest of space).
\indent {\bf \textit{5) Query set assembly:}}
Finally, in Steps {35} to {37}, we assemble each query set from the queries disjoint subsets obtained in the previous blocks and rounds.
\begin{remark}
Note that the proposed scheme significantly differs from the one presented in \cite{SunJafararxiv_2017} in terms of how the side information is exploited due to coding. In particular, we distribute the side information over $K$ rounds such that every database is queried for each message and linear combination only once.
\end{remark}
\begin{table}[ht!]
\small
\begin{tabular}{p{0.965\linewidth}}
\specialrule{.1em}{.05em}{.05em}
\rule{0pt}{2.5ex}
\hspace{-0.5em}\noindent\textbf{Algorithm 1:} Query set generation algorithm\\
\specialrule{.1em}{.05em}{.05em} \specialrule{.1em}{.05em}{.05em}
\rule{0pt}{2.5ex}
\hspace{-0.35em}\textbf{Input:} $\nu, K, N, M,$ and $V.$\\
\textbf{Output:} ${Q}^{[\nu]}(1),\dots,{Q}^{[\nu]}(N)$\\
\begin{hangparas}{.15in}{1}1.~\textbf{Initialize:} All query sets are initialized as a null set $Q^{[\nu]}(1),\dots,Q^{[\nu]}(N)\leftarrow \emptyset$, the block counter $B=1,$ and repetition counter $R=1.$ Let number of neighboring databases $Nb= {N-K}$ \end{hangparas}
2.~\textbf{Let} repetition $R_{B}= K^{V-B}(N-K)^{B-1}\quad \forall B\in[1:V]$\\
3.~\textbf{For} first database block $DB_1=1:N$ \textbf{do}\\
4.\hspace{2ex}\textbf{For} side information group $G=1:K$ \textbf{do} \\
5.\hspace{4ex}\textbf{For} repetition group $RG=1:(R_{1}/K)$ \textbf{do} \vspace{-1ex}
\begin{equation*}\vspace{-0.5ex}
\begin{aligned}
\hspace{-2ex}\text{6.} \qquad\qquad Q^{[\nu]}_{1,R}(DB_1,{\cal M})\!\leftarrow& \{u_{\nu}\}, u_\nu =\text{new}(U_{\nu})\\
\hspace{-2.8ex}\text{7.} \qquad\qquad Q^{[\nu]}_{1,R}(DB_1,{\cal I}_{G})\!\leftarrow& \{\text{new}(U_{1}),\dots,\text{new}(U_{V})\}\!\setminus\!\{u_{\nu}\}
\end{aligned}
\end{equation*}
8.\hspace{4ex}\textbf{End For} (repeat within the same SI group)\\
9.\hspace{2ex}\textbf{End For} (repetition for SI groups)\\
\hspace{-1.1ex}10.~\textbf{End For}\\
\hspace{-1.1ex}11.~\textbf{For} block $B=2:V$ \textbf{do}\\
\hspace{-1.1ex}12.\hspace{2ex}\textbf{For} $DB_{B}=1:N$ \textbf{do} \vspace{-1ex}
\begin{equation*}
\hspace{3.7ex}
\begin{aligned}
\hspace{-6.4ex}\text{13.} \quad\quad Q^{[\nu]}_{B,R}(DB_{B},{\cal M})\leftarrow &\text{\bf{Exploit-SI}}\big(Q^{[\nu]}_{B-1,R}(DB_{B}\!+\!1,{\cal I}_{Nb})\\[-0.7ex]
&\hspace{1.5ex}\cup\dots\!\cup Q^{[\nu]}_{B-1,R}(DB_{B}\!+\!Nb,{\cal I}_{1})\big)
\end{aligned} \end{equation*}
\hspace{-1.1ex}14.\hspace{4ex}\textbf{For} side-information group $G=1:K$ \textbf{do} \\
\hspace{-1.1ex}15.\hspace{6ex}\textbf{For} $RG=1:(R_{B}/K)$ \textbf{do} \vspace{-1ex}
\begin{equation*} \vspace{-0.6ex}
\hspace{-6.1ex}\text{16.} \qquad\qquad Q^{[\nu]}_{B,R}(DB_{B},{\cal I}_{G})\leftarrow \text{\bf{M-Sym}} (Q^{[\nu]}_{B,R}(DB_B,{\cal M}) )
\end{equation*}
\hspace{-1.1ex}17.\hspace{6ex}\textbf{End For} (repeat within the same SI group)\\
\hspace{-1.1ex}18.\hspace{4ex}\textbf{End For} (repeat for SI groups)\\
\hspace{-1.1ex}19.\hspace{2ex}~\textbf{End For} (repeat for each database)\\
\hspace{-1.1ex}20.~\textbf{End for} (repeat for each block)\\
\hspace{-1.1ex}21.~\textbf{For} query round $R=2:K$ \textbf{do}\\
\hspace{-1.1ex}22.\hspace{2ex}\textbf{For} $DB_1=1:N$ \textbf{do}\vspace{-1ex}
\begin{equation*}\vspace{-0.5ex}\hspace{-5ex}
\begin{aligned}
\hspace{-6.2ex}\text{23.} \qquad\qquad& Q^{[\nu]}_{1,R}(DB_1,{\cal M})\leftarrow Q^{[\nu]}_{1,R-1}(DB_1-1,{\cal M})\\
\hspace{-6.2ex}\text{24.} \qquad\qquad& Q^{[\nu]}_{1,R}(DB_1,{\cal I}_G)\leftarrow Q^{[\nu]}_{1,R-1}(DB_1-1,{\cal I}_G)
\end{aligned}
\end{equation*}
\hspace{-1.1ex}25.\hspace{2ex}\textbf{End For} (initializing rounds) \\
\hspace{-1.1ex}26.\hspace{2ex}\textbf{For} block $B=2:V$ \textbf{do}\\
\hspace{-1.1ex}27.\hspace{3ex}\textbf{For} $DB_B=1:N$ \textbf{do}\\
\hspace{-1.1ex}28.\hspace{4ex}\textbf{For} side information group $G=1:K$ \textbf{do} \vspace{-1ex}
\begin{equation*} \vspace{-0.5ex}
\begin{aligned}
\hspace{-7.0ex}\text{29.} \quad\qquad Q^{[\nu]}_{B,R}(&DB_B,{\cal I}_{G})\leftarrow Q^{[\nu]}_{B,R-1}(DB_B-1,{\cal I}_{G}) \\
\hspace{-7.0ex}\text{30.} \quad\qquad Q^{[\nu]}_{B,R}(&DB_B,{\cal M})\leftarrow \text{\bf{Reuse-SI}}\big(Q^{[\nu]}_{B,R}(DB_B,{\cal I}_{G}),\\[-0.7ex]
&Q^{[\nu]}_{B-1,1}(DB_{B}+1,{\cal I}_{Nb+R-1})\cup\dots \\[-0.7ex]
&\quad\qquad\dots\cup Q^{[\nu]}_{B-1,1}(DB_{B}+Nb,{\cal I}_{R})\big)
\end{aligned}
\end{equation*}
\hspace{-1.1ex}31.\hspace{4ex}\textbf{End For} (SI groups) \\
\hspace{-1.1ex}32.\hspace{3ex}\textbf{End For} (repeating for each database) \\
\hspace{-1.1ex}33.\hspace{2ex}\textbf{End For} (repeating for each block) \\
\hspace{-1.1ex}34.~\textbf{End For} (repeating for each round) \\
\hspace{-1.1ex}35.~\textbf{For} $DB_B=1:N$ \textbf{do}\vspace{-1ex}
\begin{equation*}\hspace{1ex}\vspace{-0.5ex}
\begin{aligned}
\hspace{-2.1ex}\text{36.} \quad\quad\! &Q^{[\nu]}(DB_B)\!\leftarrow\!\!\bigcup\limits_{B=1}^{V} \!\bigcup\limits_{R=1}^{K}\!\! \big(Q^{[\nu]}_{B,R}(DB_B,{\cal I})\cup Q^{[\nu]}_{B,R}(DB_{B},{\cal M}) \big)
\end{aligned} \end{equation*}
\hspace{-1.1ex}37.~\textbf{End For} (assembling the query sets)\\
\specialrule{.1em}{.05em}{.05em}
\end{tabular}
\normalsize
\vspace*{-4ex}
\end{table}
\vspace{-1ex}
\subsection{Sign assignment and redundancy elimination}\label{sec:SignAssyment} \vspace{-0.5ex}
We carefully assign an alternating sign $\sigma_t \in [+1,-1]$ to each symbol in the query set, based on the desired linear combination index $\nu$ \cite{SunJafararxiv_2017}. The intuition behind the sign assignment is to introduce a uniquely solvable linear equation system from the different $v$-sum types. By obtaining such an equation system in each block, the user can opt from downloading these queries, compute them off-line, and thus reduce the download rate. Based on this insight we can state the following lemma.
\begin{lemma}[\hspace{-0.1ex}\cite{SunJafararxiv_2017}]
For all {$\nu\in[1:V]$}, each database $n\in[1:N]$, and based on the side information available from the neighboring databases, there are $V-M\choose v$ redundant $v$-sum types out of all possible types $V\choose v$ in each block {${v\in[1:V-M]}$} of the query sets.
\end{lemma}
Lemma~1 is also applicable when the desired linear function is performed over MDS-coded databases due to the fact that each MDS-coded symbol is itself a linear combination. That is, the MDS code can be seen as an inner code and the desired linear function as an outer ``code'' with respect to the databases. Hence, the redundancy resulting from the linear dependencies between messages is also present under MDS coding and we can extend Lemma~1 to our scheme. We now make the final modification to our PFR query sets. We first directly apply the sign assignment $\sigma_t$, then remove the redundant $v$-sum types from every block $B\in{[1:V]}$. Finally, we generate the query matrices $Q^{[\nu]}_{1:N}$ using a one-to-one mapping function $f,$ for which $Q^{[\nu]}(DB_B)$ is the preimage.
\begin{proof}
The proof of optimality for arbitrary $N,K,M$ and $V$ follows from the structure of the query and Lemma~1. The achievable rate is given as
\begin{equation*}\label{eq:Cap_proof}
\begin{split}
&R\stackrel{(a)}{\leq}\frac{KN^V}{KN\sum_{v=1}^{V}\Big( {V\choose v}-{V-M\choose v}\Big) K^{V-v}(N-K)^{v-1}}\\
&=\frac{N^V\big(\frac{N-K}{N}\big)}{\sum_{v=1}^{V}\Big( {V\choose v} K^{V-v}(N-K)^{v}-{V-M\choose v} K^{V-v}(N-K)^{v}\Big)}\\
&\stackrel{(b)}{=}\frac{N^V\big(\frac{N-K}{N}\big)}{(N^V\!-K^V)-\sum_{v=1}^{V-M} {V-M\choose v} K^{V-v}(N-K)^{v}}\\
&=\frac{N^V\big(\frac{N-K}{N}\big)}{(N^V\!-K^V)-K^M\sum_{v=1}^{V-M} {V-M\choose v} K^{V-M-v}(N-K)^{v}}\\
&=\frac{N^V\big(1-\frac{K}{N}\big)}{(N^V\!-K^V)-K^M\big( N^{V-M}-K^{V-M}\big)}\\
&=\frac{N^V\big(1-\frac{K}{N}\big)}{(N^V\!-K^V)-K^M N^{V-M}+K^{V}}\\
&=\frac{N^V\big(1-\frac{K}{N}\big)}{N^V\!-K^M N^{V-M}}=\frac{1-R_c}{1-R_c^M};
\end{split}
\end{equation*}
where (a) follows from the definition of the PFR rate
\eqref{eq:PFR_rate_def}; (b) follows from the fact that the second
term of the summation in the denominator is equal to zero for
$v>V-M$ and consequently we can change the upper bound of the summation; and the first term of the summation follows from the binomial theorem.
\end{proof}
\vspace{-1.5ex}
\subsection{Correctness (decodability)}\vspace{-0.5ex}
To prove correctness, we show that the user can obtain the desired linear combination $\widetilde{W}_\nu$ from the answers retrieved from $N$ databases. From the query answers $A^{[\nu]}_{1:N}$, we group the $K$ identical queries from different rounds and databases. Each group will result in $K$ linearly independent equations that can be uniquely solved. We decode, block by block, starting from block one, which we directly decode and obtain
$K N \big(\!{V\choose v}-{V-M\choose v}\!\big) K^{V-v}(N-K)^{v-1}$ decoded symbols. Now, using these symbols we regenerate ${V-M\choose v}$ redundant symbols according to Lemma~1 and obtain
{$K N {V\choose v} K^{V-v}(N-K)^{v-1}$} symbols in total. Out of these queries there are
$K N \big(\!{V\choose v}-{V-1\choose v}\!\big) K^{V-v}(N-K)^{v-1}$ symbols from $\widetilde{W}_\nu$.
Next, for blocks $B\in[2:V]$, we use the symbols obtained in the previous block $B-1$ to remove the side information associated with the desired linear combination symbols of the current block $B$, then the operations from the first block (decode and retrieve redundancy) are repeated. As a result, we obtain a total number of symbols equal to $K N\sum_{v=1}^{V}\big(\! {V\choose v}-{V-1\choose v}\!\big) K^{V-v}(N-K)^{v-1} =\frac{N}{N-K} \big( N^V- KN^{V-1}\big) = K N^V$ denoting precisely the number of symbols in $\widetilde{W}_\nu$.
\vspace{-0.5ex}
\subsection{Privacy}
\vspace*{-0.5ex}
Privacy is guaranteed by preserving an equal number of requests for any
linear combination $\widetilde{W}_\nu,$ where the requests are symmetric from the
perspective of the accessed virtual messages. As the MDS code can be seen as
an outer code, the arguments in \cite{Maddah-Aliarxiv2017,
SunJafararxiv_2017} apply here as well. In particular, each database is queried with precisely the same $v$-sum type
components, i.e., $U_\nu(t)$, which ensures symmetry. This can be seen from Step 16 in Algorithm 1 where the same subroutine ``M-Sym'' is used for each block and database. By selecting a permutation $\pi(t)$ and a sign assignment $\sigma_t$ uniformly at random, queries for code symbols are permuted in the same way over all databases. With other words, for any
$U_\nu(t)=\sigma_t {\bf v}_\nu {\bf W}[\pi(t)]$ there exist $\sigma_t,
\pi(t)$ such that $Q^{[\theta]}(DB_B) \leftrightarrow Q^{[\nu]}(DB_B) \quad
\forall \nu, \theta \in [1:V]$. Thus, $A_n^{[\nu]}$, and $Q_n^{[\nu]}$
are statistically independent of $\nu$ and \eqref{privacy_const} holds.
\vspace{-1ex}
\section{Example} \label{EX} \vspace{-0.5ex}
We consider $M=2$ messages stored using a $(3,2)$ MDS code. The user wishes to obtain a linear combination over the binary field (i.e.,~${\bf{v}_\nu}\in {\mathbb F}_2^M, \quad V=3$). Therefore, we have the linear combinations ${{\bf{v}}_1=[1 \;0], {\bf{v}}_2=[0 \; 1], {\bf{v}}_3=[1\; 1]},$ and each message must be of length $L=KN^V=54$ symbols. We simplify the notation by letting $a_t=U_1(t), \; b_t=U_2(t),$ and $c_t=U_3(t)$ for all $t\in[1:27]$. Let $\sigma_t=1\;\forall t$ and let the desired linear combination index be $\nu=3$.
\indent{\textit{Query set construction:}}
Algorithm~1 starts with $B=1$ by generating queries for each database and {$K^{V-1}=4$} distinct instances of $c_t$ (i.e.,~from database~1 query ${\bf{g}}_1^T(c_{1:4})\triangleq\{{\bf{g}}_1^Tc_1,\dots, {\bf{g}}_1^Tc_4\}$). By message symmetry this also applies for $a_t$ and $b_t$ to form two groups of side information sets to be used in the next block with $NK^{V-1}=12$ symbols in total from each linear combination. Next, one group of side information is queried jointly with a new instant of the desired message. For example, for database~1 and type $b+c$ we have ${\bf{g}}_1^T(b_{5:6}-c_{13:14}) \triangleq \{{\bf{g}}_1^Tb_5-{\bf{g}}_1^Tc_{13}, {\bf{g}}_1^Tb_6-{\bf{g}}_1^Tc_{14}\}$. The remaining blocks and rounds follow from Algorithm~1. After generating the query set for each database, we apply the sign assignment and remove the redundant queries.
\indent{\textit{Decoding:}}
The answer strings from each query are shown in Table~1. Note that there is
no $c_t$ in the first block as they are redundant and can be generated by
the user. To decode we start with Block 1, and we obtain ${\bf{g}}_1^Ta_1$
from database~1 and ${\bf{g}}_2^Ta_1$ from database 2. Thus, by the MDS code
properties we can decode and obtain the segment $a_1$; similarly for all
other queries in this block. Now for Block~2, we first remove the side
information from the types containing symbols of $c_t$. For example, for
${\bf{g}}_1^Tb_5-{\bf{g}}_1^Tc_{13}$ from database~1, we have $b_5$ from the
previous block. As a result, we obtain ${\bf{g}}_1^Tc_{13}$. Similarly we
obtain ${\bf{g}}_2^Tc_{13}$ from database~2, and $c_{13}$ can be recovered.
\input{Ex1_table}
\vspace{-2ex}
\section{Outer Bound for the Special Case $V=2$}
\vspace{-0.5ex}
In the special case of $V=2,$ $M$ independent messages, and any $(N,K)$ MDS
code an outer bound for the coded PFR problem is obtained by combining the
independence of answer strings from any $K$ databases \cite[Lemma
2]{BanawanUlukus_Globecom2017} with
\cite{SunJafararxiv_2017}. Thus, we can show that the retrieval rate is
upper
bounded as $R \leq
\frac{NH(\omega_1)}{KH(\omega_1,\omega_2)+ H(\omega_1)(N-K)}$, where the
joint distribution of $(\omega_1,\omega_2)$ is the joint distribution of
$(\widetilde{W}_{1,\ell}, \widetilde{W}_{2,\ell})$ for all $\ell \in [1:\tilde{L}]$ selected iid~with respect to the symbols of the messages.
\vspace{-1ex}
\bibliographystyle{IEEEtran}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 4,374
|
Knights Of St Mulumba Creates New Sub-Councils In Lagos Metro
Metro Religion Slider
The Knights of St Mulumba in the Lagos Metropolitan Council has created three new sub- councils for knights and Ladies of the Order namely, Victoria Garden City (VGC) Sub-Council from Lekki Sub-Council, Gbagada Sub-council from Maryland Sub-Council and Amuwo-Odofin Sub-Council from Festac Sub-Council.
Speaking at the event in Lagos, at the weekend, the Parish Priest, Our Lady Queen of Apostles, Ilupeju, Rev Fr. Paul Anyansi, said the event was aimed at creating new sub- councils of the Knights and Ladies of St Mulumba. Anyansi explained that the creation of new sub- councils which has Victoria Garden City (VGC) as one of them is aimed at strengthening and promoting the ideals of the Order through evangelization of the Catholic faith and fostering a Catholic conscience and outlook in their respective Sub-councils and apply Ccatholic principles to all phases of societal life through examples and enlightenment.
He remarked that knighthood requires members to give their talent, treasure and time, adding that the newly created VGC sub- council no doubt will strive to be a leading light in fostering good Christian life among its members and maintaining a union of practicing Catholics in fraternal charity and fellowship.
In his remarks, the Grand Knight, VGC Sub- Council, Ferdinand Odoemenem, pledged to foster fraternal association for the good and progress of the Church as well as the well-being of its members.
Odoemenem assured members of the sub- council of the commitment of the new leadership to promote cooperation between one sub- council and the other. He added that they want to put measures in place to sustain the relationship that they have between sub- councils. "Our mother Sub-council (Lekki Sub- Council) is precious to us and we will continue to bond together on many issues especially on evangelization. "We are going to step up evangelization within our geographical area, and promote brotherhood among us. We will endeavour to live exemplary Catholic lives and to grow in the grace of God through the virtues of charity, meekness, fidelity, courage, humility, modesty and holiness."
According to Ferdinand Odoemenem, with these virtues, we will be fired with a fervent zeal for the social apostolate required of a knight, while holding in high esteem professional competence and superior family values.
Grand Knight of the Lekki Sub-Council, the mother Sub-Council, Benji Ofodile, in his remarks described the new VGC Sub-Council as a joyful birth adding that both Sub-councils will continue to share and grow from each other's strength in order to fulfil the aims and objectives of the Order.
Oyo Defies Court Order, Demolishes Ayefele's Music House
SERAP To Oyo Gov Ajimobi: You're Guilty Of Executive Rascality
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 2,167
|
Q: Approval time for in app purchase product For adding in app purchase products, how long does apple normally take to approve one of them? I have like hundreds of in app products and i wonder how long it will take for all to be approved.
A: It usually takes anywhere from 1-2 weeks depending on how busy it is. If you have like hundreds it may take a few months to get them all approved but then again, apple doesn't check everything, just a lot of them so probably 1-3 months.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 8,942
|
"""
This package contains the actual wikicode parser, split up into two main
modules: the :mod:`.tokenizer` and the :mod:`.builder`. This module joins them
together into one interface.
"""
from .builder import Builder
from .errors import ParserError
try:
from ._tokenizer import CTokenizer
use_c = True
except ImportError:
from .tokenizer import Tokenizer
CTokenizer = None
use_c = False
__all__ = ["use_c", "Parser", "ParserError"]
class Parser:
"""Represents a parser for wikicode.
Actual parsing is a two-step process: first, the text is split up into a
series of tokens by the :class:`.Tokenizer`, and then the tokens are
converted into trees of :class:`.Wikicode` objects and :class:`.Node`\\ s
by the :class:`.Builder`.
Instances of this class or its dependents (:class:`.Tokenizer` and
:class:`.Builder`) should not be shared between threads. :meth:`parse` can
be called multiple times as long as it is not done concurrently. In
general, there is no need to do this because parsing should be done through
:func:`mwparserfromhell.parse`, which creates a new :class:`.Parser` object
as necessary.
"""
def __init__(self):
if use_c and CTokenizer:
self._tokenizer = CTokenizer()
else:
from .tokenizer import Tokenizer
self._tokenizer = Tokenizer()
self._builder = Builder()
def parse(self, text, context=0, skip_style_tags=False):
"""Parse *text*, returning a :class:`.Wikicode` object tree.
If given, *context* will be passed as a starting context to the parser.
This is helpful when this function is used inside node attribute
setters. For example, :class:`.ExternalLink`\\ 's
:attr:`~.ExternalLink.url` setter sets *context* to
:mod:`contexts.EXT_LINK_URI <.contexts>` to prevent the URL itself
from becoming an :class:`.ExternalLink`.
If *skip_style_tags* is ``True``, then ``''`` and ``'''`` will not be
parsed, but instead will be treated as plain text.
If there is an internal error while parsing, :exc:`.ParserError` will
be raised.
"""
tokens = self._tokenizer.tokenize(text, context, skip_style_tags)
code = self._builder.build(tokens)
return code
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,043
|
\section{Coupling to a single ion in free space} \label{sec:1}
Atom-photon interfaces are a key element for constructing a quantum network \cite{Cir97,Bri98,Dua01}. Here, the interface maps quantum information from stationary to flying qubits and vice versa. Usually photons are employed as flying qubits due to their robustness in preserving quantum information during propagation, while atoms are used for storing and computing the quantum information in stationary nodes. The efficient mapping of quantum information from atoms to photons and back demands controlled photon emission and absorption with a very high probability. This condition can be achieved in a strong coupling regime where the information is exchanged between atoms and photons several times before it decoheres. The standard way to achieve strong atom-photon coupling is by using either small high finesse cavities, which increase the interaction between a single atom and a photon \cite{Bru94,Pin00,Hoo00} as described in chapter by A. Kuhn, or large atomic ensembles for continuous variable quantum interfaces \cite{Pol04,Hau99,Phi01} treated in chapter by Chuu \& Du.
In free space, i.e. without an enhancing cavity, the coupling of a single atom and light is generally considered to be weak. Nevertheless, it can be significantly increased if the light covers a large solid angle, for instance by using large aperture lenses \cite{Sor07} or mirrors \cite{Son07}. In such a setup it is possible to observe effects where a single atom can notably modify the light field. For instance, several experiments have recently demonstrated that a single quantum particle, like a single rubidium atom \cite{Tey08}, a molecule \cite{Zum08,Ger07,Wri08}, or a quantum dot \cite{Vam07} can cause extinction of more than 10\,\%, and a phase shift of one degree \cite{Alj09} for the transmitted light. Also, a single molecule has been shown to act as a nonlinear switch \cite{Hwa09}. These experiments are first steps towards realizing photonic quantum gates and quantum memories with single atoms in free space.
In this chapter we will discuss several experiments in which single ions are placed at the focus of \keyword{high numerical aperture optics} for efficient atom-light interaction. In section \ref{sec:1} of this chapter we will review several experiments on direct \keyword{free space coupling} of a \keyword{single trapped ion} with light. Here, the high numerical aperture optics allow the observation of effects which are usually only observed with large atomic ensembles or single atoms in high finesse cavities, including electromagnetically induced transparency \cite{Slo10}, coherent back scattering \cite{Het11}, and Faraday rotation induced on a propagating laser field \cite{Het13}. Very similar ideas and experiments discussing the efficient absorption of single photons by single ions are described also in the chapters by Leuchs \& Sondermann and by Piro \& Eschner.
Later in section \ref{sec:2} we will treat probabilistic methods to exchange quantum information over a distance. Here, a projective measurement on the fluorescence photons scattered by two atoms projects them into an entangled state, which then for instance could be used to link distant registers in a quantum network using protocols like quantum teleportation or entanglement swapping.
\subsection{Electromagnetically induced transparency from a single atom in free space}
The controlled storage and retrieval of photonic quantum information from an atomic medium is often based on a phenomenon called \keyword{electromagnetically induced transparency} (EIT)~\cite{Fle05} or its excitation in the form of stimulated Raman adiabatic passage (STIRAP)~\cite{Ber98a}. This technique has been widely used to control the storage of weak light pulses or single photons in atomic ensembles \cite{Phi01,Hau99,Eis05} and high-finesse cavities \cite{Boo07,Mue10}. For EIT, atoms in a lambda-type three-level system are driven by a weak probe laser and a strong control laser in Raman configuration. Due to a destructive quantum interference effect the control laser suppresses the absorption of the resonant probe light. Consequently, by changing the control laser intensity it is possible to switch the medium between transmitting and absorbing the probe light. Seen from a different point of view, the control laser intensity changes the group velocity of the probe laser. Thus, adiabatic ramping of the control laser intensity can slow down and even stop a single probe photon. In this way the photon is stored in the long-lived atomic ground states of the medium and can be retrieved by a time-reversal of the storage process \cite{Cir97,Fle05}. An extension of this scheme can store a photonic quantum state, for instance encoded in the polarization of light, in superpositions of atomic ground states \cite{Rit12}, thus realizing a memory for the photonic quantum information.
Effective switching between transmission and absorption can only be achieved in optically thick media. Therefore, until recently the application of EIT was restricted to ensembles of many atoms \cite{Fle05}. In contrast, (discrete variable) quantum information processing is based on single well-defined qubits (for example single atoms or ions) where each qubit can be individually manipulated to perform quantum gates. A quantum network which combines these two technologies requires strong single atom-single photon interaction within the interface nodes to distribute quantum information over the nodes of the quantum network.
Trapped ions are at the moment one of the most advanced systems for quantum information processing \cite{Hae08}. Also, since ions of the same species are identical, they are very well suited as indistinguishable light sources \cite{Mau07,Ger09njp} at the distant locations of a quantum network. Furthermore, the precise control over the electronic and motional states of the ions in Paul traps makes them ideal to investigate the coupling of radiation to single absorbers. In the following, we will present first steps towards a free-space single ion quantum interface by demonstrating an extinction of a weak probe laser of 1.3\,\%, electromagnetically induced transparency from a single trapped ion, and the corresponding phase shift response.
\subsubsection{Extinction and phase shift measurements}
The experiments that we will describe in this chapter show the relation between the input and the transmitted light field in the presence of an atom \cite{Het13}. The following simple theoretical model can describe the basic properties of the \keyword{extinction} and reflection of a weak probe field from a single atom. This approach uses a perturbative \keyword{input-output formalism} to relate the input field, $\hat{E}_{\rm in}$, and the output field, $\hat{E}_{\rm out}$, through their interaction with the atom \cite{Koc94} as schematically depicted in Fig.~\ref{Fig:2Lscheme}. In Markov approximation, the output field in forward direction can be described as a superposition of the transmitted input field and the emitted field of the radiating atomic dipole as
\begin{equation}\label{EqInpOutp}
\hat{E}_{\rm out}(t)=\hat{E}_{\rm in}(t) + i\sqrt{2\gamma_{\rm in}}\hat{\sigma}(t),
\end{equation}
where $\hat{\sigma}(t)$ is the atomic coherence and $\gamma_{\rm in}$ is an effective coupling coefficient of the input field to the atom. The coupling coefficient can also be expressed by the total decay rate of the excited state $\gamma$ and the fraction $\epsilon$ of the full solid angle covered by the incoming field as $\gamma_{\rm in}=\epsilon\gamma$.
\begin{figure}[t]
\begin{center}
\includegraphics[width=0.6\columnwidth]{Slodicka_fig1}
\end{center}
\caption{Schematic view of a single atom irradiated by continuous laser light. The atom is illuminated from a fraction of the solid angle $\epsilon$ and radiates into full solid angle. The fluorescent light in general contains both elastic and inelastic components, nevertheless, for weak excitation the elastic component dominates. In the forward direction the elastically scattered part of the fluorescent light interferes with the transmitted laser beam ($\epsilon$). In the backward direction ($1-\epsilon$) the fluorescent light is observed.}
\label{Fig:2Lscheme}
\end{figure}
The atom reacts on the excitation by the input field, thus the atomic coherence $\hat{\sigma}(t)$ can be calculated by solving Bloch equations of the two-level atom in the weak excitation limit and in steady state, which gives
\begin{equation}
\hat{\sigma}=\frac{i \sqrt{2\gamma_{\rm in}}}{\gamma+i\Delta} \hat{E}_{\rm in},
\end{equation}
where $\Delta$ is the frequency detuning of the probe light from the excited state. Finally, the transmission of the intensity of the probe field $T=|E_{\rm out}/E_{\rm in}|^2$ in steady state reads
\begin{eqnarray}\label{Eqabs}
T(\Delta)=|1-2\epsilon\mathcal{L}(\Delta)|^2,
\end{eqnarray}
where the amplitude of the atomic coherence is proportional to $\mathcal{L}(\Delta)=\gamma/(\gamma+i\Delta)$ for a two level atom. Furthermore, the \keyword{phase shift} $\phi$ of the transmitted light field is
\begin{eqnarray}\label{ps}
\phi(\Delta)=\arg\big[1-2\epsilon \mathcal{L}(\Delta)\big].
\end{eqnarray}
Please note that this simple theory predicts perfect extinction of the transmitted probe field, and thus full reflection for a weak resonant input field covering half of the full solid angle ($\epsilon=0.5$). This results from the interference between the transmitted input beam and the radiated dipole field, which yields a considerable decrease in the forward mode amplitude \cite{Tey09,Zum08}. In the experiment described below, we use a lens with numerical aperture NA=0.4 (i.e $\epsilon=4\,\%$), so we expect a probe beam extinction of $16\,\%$ from Eq.~(\ref{Eqabs}) from this basic theoretical model. More refined models \cite{Venk04,Tey08,Zum08} include effects like the polarization of the input beam and the exact mode overlap between the transmitted beam and the dipole emission pattern which becomes especially important when the input beam is focused to the atom from a large solid angle with a high numerical aperture lens beyond the paraxial approximation. From the model of \cite{Tey08}, we expect an extinction of around $13\,\%$ for our experimental parameters. A higher extinction is possible for larger solid angles. Nevertheless, if one wants to use a single atom in free space as a quantum interface, for an efficient sender one will need to collect the emitted photon from full solid angle. Also, an efficient quantum receiver will demand reversal of the emission process, thus the single photon input mode will have to match the full dipole radiation pattern and the reversed temporal mode of the atomic emission as detailed in \cite{Son07} and chapter by Leuchs \& Sondermann.
\begin{figure}[ht!]
\centerline{\includegraphics[width=0.9\columnwidth]{Slodicka_fig2.pdf}}
\caption{a) Scheme of the experimental set-up used to measure EIT and the corresponding phase shift on a single atom~\cite{Het13}. The probe field is defined in horizontal polarization by passing through a polarizing beam-splitter (PBS). The laser beam is then expanded by a telescope and focused by a high numerical aperture lens in vacuum (NA=0.4) onto the ion. Depending on the intensity of the control laser, the single ion changes the transmission of the probe light and rotates its polarization, which after re-collimation is detected by polarimetry.
b) Level scheme of $^{138}$Ba$^+$ and probe and control laser fields used in the experiment. The input probe field at 493\,nm is decomposed in the two circular polarizations which excite two branches of the spin-half system with different detunings. The laser field at 650\,nm is used as the control field in the EIT measurements and for repumping population from the $5D_{3/2}$ level.}
\label{setup}
\end{figure}
In the following we will describe how we measure the single ion transmission, the phase shift, and the EIT effect in our experiment. The experimental setup consists of a high numerical aperture objective \cite{Esc01} to focus the probe field onto a single trapped ion and a second objective to collect the transmitted light onto a detector, shown in Fig.~\ref{setup}-a). The barium ion is trapped and cooled in a standard spherical Paul trap. As already mentioned before, good extinction of the probe field can only be achieved if the incoming probe beam and dipole emission pattern are carefully overlapped. This mode-matching is done using an expanding telescope and a custom-designed objective with a numerical aperture of NA=0.4 ($\epsilon=4\,\%$). A magnetic field of 5 Gauss applied along the probe beam propagation direction defines the quantization axis. The probe field is linearly polarized perpendicular to the quantization axis, collected after the ion using a second high numerical aperture lens, and then analyzed using a polarimetric set-up and photo-multiplier tubes (PMT). In practice, we detect the transmitted polarization alternately at +45 or -45 degrees with respect to the input polarization.
The level scheme of Ba$^{+}$ is shown in Fig.~\ref{setup}-b). The probe field is tuned to the $6S_{1/2}\rightarrow 6P_{1/2}$ transition and with our choice of quantization axis, its polarization can be decomposed onto left and right circularly polarized modes that drive $\sigma_-$ and $\sigma_+$ transitions in the ion, respectively. The two polarization modes do not have the same detuning from their respective transitions and thus may experience different indices of refraction. We set the intensity of the probe field well below saturation so that most of the light is elastically scattered. The probe field supplies only weak cooling to the ion. Therefore the ion is cooled additionally by a red detuned 493\,nm laser field perpendicular to the probe direction and a laser at 650\,nm, co-propagating with the cooling beam, for pumping out population from the $5D_{3/2}$ level. For the characterization of the Faraday rotation we use a lock-in method where the 650\,nm repumping laser is switched on and off at a rate of 5\,kHz to precisely measure the polarization rotation signal. When the repumping laser is off the population of the atom is pumped into the $5D_{3/2}$ state. Since the probe beam drives the transition between $6S_{1/2}$ and $6P_{1/2}$, it does not feel the presence of an ion that resides in $5D_{3/2}$. Thus, the modulation of the repumping laser also modulates the effect of the atom on the probe field. The photo-multiplier signal is demodulated and low-pass filtered with a time constant of 1s.
The intensity of the light at the PMT measured at +45 degrees with respect to the input polarization can be written as
\begin{eqnarray}
I_{45}=\frac{1}{2}|E^+_{\rm out} e^{i\pi/4}+E^-_{\rm out} e^{-i\pi/4} |^2,
\label{eqn:pol45}
\end{eqnarray}
where
\begin{eqnarray}
E^+_{\rm out} &=(1-2\epsilon \mathcal{L}^+) \frac{1}{\sqrt{2}} E_{\rm in}&\quad\mathrm{and} \\
E^-_{\rm out} &=(1-2\epsilon \mathcal{L}^-) \frac{1}{\sqrt{2}} E_{\rm in}&
\end{eqnarray}
are the corresponding output fields of $\sigma_+$ and $\sigma_-$ polarization, respectively. The real and imaginary parts of
\begin{equation}
\mathcal{L}^{\pm}=\frac{\rho_\pm\gamma}{\gamma+i\Delta^{\pm}}
\end{equation}
correspond to absorption and phase-lag of the two scattered circularly polarized field modes with regards to the input field, respectively. Here, $\Delta^{\pm}=\Delta \pm \Delta_{B}$ are the detunings of the $\sigma^+$ and $\sigma^-$ polarized fields from their respective transitions. The two ground state populations $\rho_+$ ($\rho_-$) of the $\sigma_+$ ($\sigma_-$) transition correspond to the states $6S_{1/2}$, $m_J=-1/2$ ($m_J=+1/2$), respectively. $\Delta$ is the probe laser detuning with respect to the $6S_{1/2} \rightarrow 6P_{1/2}$ transition without Zeeman shift, and $\Delta_{B}$ is the frequency offset of the two resonances due to the Zeeman splitting. The $\pm\pi/4$ phase shifts in Eq.~(\ref{eqn:pol45}) are due to the rotation of the polarization direction between input and detected polarization induced by the $\lambda/2$ waveplate that allows us to characterize the Faraday rotation.
To measure the Faraday rotation angle $\theta = \frac{1}{2} \arctan(s_2/s_1)$, here defined as half the rotation angle from horizontal polarization towards 45 degree polarization on the Poincar\'e sphere, we need to record the Stokes parameter $s_1=I_{0}-I_{90}$, and $s_2=2 I_{45}-s_0$, with $s_0=I_{0}+I_{90}$. Please note that for our small extinction values the Stokes parameter $s_1$ can be approximated by $s_1\approx s_0\approx I_{0}$, which can be measured directly by removing the polarizing beam-splitter. After re-inserting the beam-splitter and adjusting the waveplate accordingly, we can access the Stokes parameter $s_2\approx 2 I_{45}-I_0$. The Faraday rotation angle $\theta \approx \frac{1}{2} \arctan((2 I_{45}-I_0)/I_0)$ is directly related to the phase shift induced by the atom. It can be shown, using the approximation $\arg(1-2\epsilon z)\approx -2\epsilon {\rm Im}(z)$ in the limit of small $\epsilon$, that
\begin{eqnarray}\label{ps1}
\theta=\frac{1}{2}\arg\big[1-2\epsilon ( \mathcal{L}^+ - \mathcal{L}^- )\big],
\end{eqnarray}
which is half of the phase lag experienced by the output with respect to the input field. A measurement of $I_{45}$ and $I_{0}$ thus provides a measurement of the Faraday rotation of the light across the atom together with the phase difference acquired by the two circularly polarized modes.
\begin{figure}[!t!]
\hspace*{-2mm}\centerline{\scalebox{0.47}{\includegraphics{Slodicka_fig3}}}
\caption{a) Transmission $I_0$ (ii) and phase shift $\theta$ (i) of a probe field transmitted through a single trapped barium ion as a function of probe beam detuning~\cite{Het13}. The transmission spectrum is fitted by a Lorentzian profile with a width of 11\,MHz. The peak probe beam extinction is 1.35\,\%. b) Transmission (i) and phase shift (ii) of a probe field transmitted through a single trapped barium ion as a function of probe beam detuning close to a dark resonance.}
\label{absorption}
\end{figure}
We characterize the Faraday rotation of the probe field by measuring the phase shift $\theta$ and the transmission $I_0$, which are plotted in Fig.~\ref{absorption}-a) as a function of the probe frequency detuning $\Delta$. As can be seen from the measurement of $I_0$, an optical pumping mechanism by the cooling and repumping lasers causes a strong unbalancing between the two ground states populations. In principle, one would expect to observe two identical absorption lines at +5\,MHz and -14\,MHz (shifted with respect to the symmetric case due to dipole shifts) for the $\sigma_+$ and $\sigma_-$ transitions of the probe light from $6S_{1/2}$ to $6P_{1/2}$, respectively. Here, the optical pumping induced by the cooling and repumping lasers traps population in the $| 6 S_{1/2}, m_J=+1/2 \rangle$ level so that only the $\sigma_-$ transition can be detected by the probe field. This manifests itself in the 1.5\,\% extinction that is seen -14\,MHz red-detuned from the central line for the $\sigma^-$ and in the almost completely suppressed extinction for the other $\sigma^+$ mode at 5\,MHz. With this state preparation, trace (i) displays a clear dispersive profile across the resonance of the $\sigma_-$ transition and the $\sigma^+$ polarized mode is almost not phase shifted. Even with our small magnetic field, the pumping technique thus allows us to isolate a single two-level atom and to reach a maximum of 0.3 degrees phase-shift. Solid lines show the result of a fit of the data using the above four-levels calculations, with $\epsilon=0.8\,\%$, $\Delta_B=9$\,MHz, $\rho_-=0.9$ and $\rho_+=0.1$. With these parameters, good agreement is found with the experimental results.
\subsubsection{Electromagnetically induced transparency and associated phase shift with a single atom}
In the above measurements, the cooling and repumping beams were tuned to a dark resonance with the intention to pump the population of the ion into one of the $S_{1/2}$ levels and therefore minimize the population in the $D_{3/2}$ state which is not interacting with the probe laser. Nevertheless, the transverse cooling beam can also be turned off and the ion be cooled by the linearly polarized probe field itself.
In such a configuration the probe undergoes \keyword{electromagnetically induced transparency} (EIT)~\cite{Slo10} where the population in the excited state of the $\Lambda$ scheme (see Fig.~\ref{setup}-b)) is canceled due to a quantum interference between the two excitation pathways leading to the $P_{1/2}$ excited state.
Under weak probe excitation, the probe transmission as a function of the two-photon detuning $\delta=\Delta_g-\Delta_r$ can be found by solving the Bloch equations \cite{Ima89} and using the above input-output relations. We neglect here the angular dependence of the extinction (due to polarization). That is, we suppose that the probe has a polarization profile that matches the dipole field. This is a good approximation for the relatively small numerical aperture we use in this experiment. We can thus replace the function $\mathcal{L}$ by
\begin{eqnarray}\label{EITeq} \mathcal{L}_{\Lambda}(\delta)=\frac{\gamma(\gamma_0-i\delta)}{(\gamma_0-i\delta)(\gamma+i\Delta_g)+\Omega_r^2},
\end{eqnarray}
in Eq.~(\ref{Eqabs}), where $\Omega_r$ is the Rabi frequency of the red laser field, $\gamma_0$ the ground state dephasing rate, $\gamma$ the natural linewidth of the two transitions (assumed to be the same for simplicity). An important condition for EIT to take place is $\gamma\gamma_0 \ll \Omega_r^2$, i.e. the pumping rate to the dark state must be much faster than any ground state decoherence process. Independent frequency fluctuations of the two laser fields, magnetic field fluctuations, and atomic motion induced Doppler shifts, must be therefore reduced. When this is the case, extinction of the resonant probe can be completely inhibited, within a small range of control laser detuning $\Omega_r^2/\gamma$, creating an EIT window. This is what we observed in this experiment.
In the experiment we found that the motion induced decoherence yields broadening of tens of kHz, which reduced the EIT when the control and the probe were orthogonal to each other. However, the effect of Doppler shifts due to the ion motion could be eliminated when we used co-propagating control and probe fields. Since for optimum EIT conditions we could not use the cooling fields which would have reduced the transparency achieved through EIT, so the ion needed to be cooled by the probe itself. Consequently, the probe beam was more intense and red detuned, which resulted in reduced extinction efficiencies of about $0.6\,\%$. Additionally, we have to note here that due to the multi-level structure of barium, a single three level system can only be perfectly isolated from the others through optical pre-pumping. Therefore, Stark-shifts induced by the other levels and double-$\Lambda$ type couplings contribute to a slight reduction of the EIT contrast.
The results of the measurement of the probe transmission versus the two-photon detuning $\delta$ are shown in Fig.~\ref{absorption}-b) trace (i). In this EIT regime, a rapid change of the transmission is found as a function of the two-photon detuning and an almost complete cancellation of the transmission is measured at $\delta=0$.
Associated with such a steep change of the probe transmission, we also expect a fast roll-off of the phase. Fig.~\ref{absorption}-b)-trace (ii) shows the measurement of the phase $\theta$ of the probe field, using the same polarimetric technique as in the measurement described in the previous section. Here again, close to the dark resonance, the Faraday rotation angle yields the phase-shift induced by the atom. The clear dispersive shape of $\theta$ across the two-photon resonance is here a sign of the EIT induced phase-shift from the ion where a maximum phase lag of 0.3 degrees is observed. The solid lines show a fit to the experimental results using 8-level Bloch equations, consisting of the 2 $S_{1/2}$, the 2 $P_{1/2}$ and the 4 $D_{3/2}$ states. Here we replace the two-level atom Lorentzian functions $\mathcal{L}^{\pm}$ in Eq.~\ref{ps} by the newly found susceptibilities. The theory describes well the data with the repumping and probe field intensities as the only two free parameters. The asymmetry of the dispersion and transmission profiles that we measured is due to a slight overlap with neighboring dark-resonances and our detuned driving of the $\Lambda$ scheme. The distinctive feature of this interference effect is that the flipping of the phase shift sign occurs only over a couple of MHz. Increasing the slope steepness further can in fact be done by performing the experiment with smaller probe and repumping powers which can be implemented by appropriate switching of the laser cooling beams involved the experiment. Achieving a very steep phase shift dependence across the atomic spectrum would open the way for reading out the motional and internal energy of the atom.
Tightly focusing a weak, detuned, linear polarized probe field onto a single barium ion thus enables observation of both the direct extinction of a weak probe field and electromagnetically induced transparency from a single barium ion. Besides demonstrating further the potential of these effects for fundamental quantum optics and quantum information science, these experimental results will trigger interest for quantum feedback to the motional state of single atoms, as proposed in \cite{Rabl} using EIT, for dispersive read out of atomic qubits and for ultra-sensitive single atom magnetometery.
In the following we will now discuss another effect observed with a similar experimental apparatus where we show that a single atom can act as a mirror of an optical cavity.
\subsection{Single ion as a mirror of an optical cavity}
Atom-photon interactions are essential in our understanding of quantum mechanics. Besides the two processes of absorption and emission of photons, coupling of radiation to atoms raises a number of questions that are worth investigating for a deeper theoretical and thus interpretational insight. The modification of the vacuum by boundaries is amongst the most fundamental problems in quantum mechanics and is widely investigated experimentally. We here present the very first steps towards merging the field of cavity QED with free-space coupling, using an ion trap apparatus.
Here we report an experiment where we set up an atom-mirror system~\cite{Het11}. As shown Fig.~\ref{setup}-a), we place a mirror in the path of the probe beam in front of the ion. The idea of this geometry is to form a \keyword{atom-mirror cavity} system consisting of the normal mirror and the ion acting as the second mirror. We observe the modification of the probe transmission and reflection of this atom-mirror cavity. Here, the atomic coupling to the probe is modified by the single mirror in a regime where the probe intensity is already significantly altered by the atom without the mirror. In principle, in the limit of an even higher numerical-aperture lens, the mirror-induced change in the vacuum-mode density around the single atom could modulate the atom's coupling to the probe, the total spontaneous decay and the Lamb shift, so that the atom would behave as the mirror of a high-finesse cavity.
As before, for extinction of a laser field by the ion in free space, we use a very weak probe beam resonant with the S$_{1/2}(m_J=+1/2)$-P$_{1/2}(m_J=-1/2)$ transition. In the case of coherent reflection of a laser field by a single atom, the backscattered field must interfere with the driving laser. To verify this, we construct the system shown in Fig.~\ref{setup} a) by inserting a dielectric mirror 30\,cm away from the atom into the probe path, with a reflectivity $|r|^2=1-|t|^2=25\,\%$. We align it so that the ion is re-imaged onto itself and shine the resonant probe through it. Using the Fabry-P\'erot cavity transmissivity, and modeling the atom as a mirror with amplitude reflectivity $2\epsilon$~\cite{Koc94}, one can naively assume that the intensity transmissivity of the probe reads
\begin{eqnarray}\label{EQ1}
T=\Big|\frac{t(1- 2\epsilon) }{1- 2r\epsilon e^{i\phi_L} }\Big|^2,
\end{eqnarray}
where $\phi_L=2k_LR$, $R$ is the atom-mirror distance and $k_L$ the input probe wavevector. The finesse $\mathcal{F}=\pi 2\epsilon r/(1-(2\epsilon r)^2)$ of such a cavity-like set-up can in fact be made very large by using a high numerical aperture lens such that $\epsilon\rightarrow 50\,\%$ together with a highly reflective dielectric mirror. By tuning the distance between the dielectric mirror and the ion, one would therefore expect a dependence of the transmitted signal on the cavity length, provided that the temporal coherence of the incoming field is preserved upon single-atom reflection.
The operation of our ion-mirror system is shown in Fig.~\ref{Fig:back}~b), where we simultaneously recorded the reflected and transmitted intensity. As the mirror position is scanned, we indeed observed clear sinusoidal oscillations of the intensity on a wavelength scale. These results reveal that the elastic back-scattered field is interfering with the transmitted probe, and that the position of the ion is very well defined, meaning that it is well within the \keyword{Lamb-Dicke regime}. Reflected and transmitted intensity have opposite phase, as is predicted for a Fabry-P\'erot cavity response. The slight shift in the two sinusoidal fitting functions is within the measurement error bars.
\begin{figure}[t]
\centerline{\scalebox{0.2}{\includegraphics{Slodicka_fig4}}}
\caption{a) Experimental setup with an optical cavity formed by the mirror and the single ion. PMT 1 and 2 measure the transmitted and the reflected probe laser intensity, respectively. b) Reflection and transmission signal of the ion-mirror cavity as a function of the mirror position~\cite{Het11}.
}\label{Fig:back}
\end{figure}
We now investigate whether the naive Fabry-P\'erot interpretation that we used to describe our results is valid. One could indeed wonder how the modification of the quantum vacuum around the atom affects our results. It is clear that the dielectric mirror imposes new boundary conditions that change the \keyword{vacuum mode density} close to the atom, but it is less obvious how much this change contributes to the probe intensity modulation that we observe in this experiment. One can in fact show \cite{Het11} that solving the multimode Heisenberg equations in a time-dependent perturbation theory gives
\begin{eqnarray}\label{EQ3}
T=|t|^2\Big|1-\frac{2g_{\epsilon}\overline{g}^\ast}{\tilde{\gamma}+i\tilde{\Delta}}\Big|^2,
\end{eqnarray}
assuming the input probe to be resonant with the atomic transition. Here, $g_{\epsilon}$ denotes the atomic coupling strength in the probe mode, $\overline{g}$ is the mean coupling to all the modes, $\tilde{\gamma}$ and $\tilde{\Delta}$ are the decay and level shifts modified by the presence of the mirror, respectively. Their value can be calculated using the appropriate spatial mode function for this system \cite{het10} and we can then show that
\begin{eqnarray}\label{EQ5}
\frac{g_{\epsilon}\overline{g}^\ast}{\tilde{\gamma}+i\tilde{\Delta}}= \frac{\epsilon(1- r e^{i\phi_L}) }{1- 2r\epsilon e^{i\phi_L} }.
\end{eqnarray}
After combining this relation with Eq.~\ref{EQ3} we obtain the same transmissivity as was obtained by modeling the atom as a mirror with reflectivity $2\epsilon$ (Eq.~\ref{EQ1}). Interestingly, the QED calculations yield the same mathematical results as the direct Fabry-P\'erot calculation.
In this \keyword{QED} approach, it was not necessary to invoke multiple reflections off the atom for the Fabry-P\'erot like transmission to appear. The transmission of the probe through the single atom+mirror system is mathematically equivalent to a cavity, therefore the origin of the peaked transmission profile can be interpreted either as a cavity effect or as a line-narrowing effect due to the QED-induced changes of the spontaneous emission rate and level shift. In the second interpretation, the observed oscillations can be interpreted as a change of the coupling between the atom and the probe mode, due to the modification of the mode density at the position of the ion induced by the mirror. For very high numerical optics the change of the extinction contrast would be analogous to an almost complete cancellation and enhancement by a factor of two of the atomic coupling constant in the probe mode. Deviations from the sinusoidal shape due to line narrowing would already be visible for a lens covering a solid angle of more than 10\,\%.
\section{Probabilistic entanglement between distant ions} \label{sec:2}
Long-lived \keyword{entanglement} between distant physical systems is an essential primitive for quantum communication networks~\cite{Bri98,Dua01}, and distributed quantum computation~\cite{Jia07,Cir99,Got99}. There are several protocols generating entanglement between distant matter qubits \cite{Dua10}, like single atoms. The majority of them exploit traveling light fields as mediators of the entanglement generation process. A way to generate distant entanglement is based on the spontaneous generation of entanglement between an atom and a single photon during the emission process followed by the absorption of the photonic state in a second atom~\cite{Cir97}. This method can generate entanglement deterministically, if the photon collection and absorption processes are highly efficient. Nevertheless, photon losses in experimental realizations might render it necessary to first detect a successful photon absorption in order to herald the successful entanglement generation. Another approach generates the entanglement probabilistically by detecting single photons that where scattered by two atoms. The projective measurement on the photons heralds an entangled state of the two atoms~\cite{Cab99, Sim03, Dua01, Bri98}.
The realization of heralded entanglement between distant atomic ensembles \cite{Cho05,Cho07} was amongst the first major experimental achievements in this field. Probabilistic generation of heralded entanglement between single atoms~\cite{Sim03} was demonstrated using single trapped ions~\cite{Moe07} with an entanglement generation rate given by the probability of coincident detection of two photons coming from the ions~\cite{Zip08,Luo09}. More recently, single neutral atoms trapped at distant locations were entangled using the deterministic entanglement protocol described above~\cite{Rit12}. Nevertheless, the efficiency of this realization was still limited due to losses to approximately 2\,\%. A heralding mechanism will therefore be essential for efficient entanglement and scalability of quantum networks using realistic channels \cite{Dua10}. The distant entanglement could also be used for one-way quantum computation schemes~\cite{Rau01,Dua05}. Such schemes for distributed quantum information processing would require only projective measurements and single qubit operations to perform quantum calculation~\cite{Dua10}. For future quantum information applications it therefore will be important to realize \emph{\keyword{heralded}} distant entanglement with the possibility of \emph{single qubit operations} and with \emph{high \keyword{entanglement generation rate}} at the same time.
\subsection{Single-photon and two-photon protocols}
The main limitation for generation of heralded distant entanglement between single atoms with high rate is imposed by relatively small overall detection efficiencies $\eta$ of fluorescence photons emitted by atoms trapped in free space \cite{Zip08}. For state-of-the-art experimental setups employing high numerical aperture optics close to single trapped neutral atoms or ions, $\eta$ is on the order of $10^{-3}$~\cite{Str11,Str12,Dar05,Olm10,Tey09}. There is a large effort in the experimental quantum optics community towards increasing this number both by employing very high numerical aperture optics in the form of spherical~\cite{Shu11} or parabolic~\cite{Mai09,Sto09} mirrors and by developing single-photon detectors with high quantum efficiency. However, even with these improvements it will be hard to increase the overall detection efficiency by more than one order of magnitude in the near future.
We compare the efficiency of the two known heralded entanglement generation protocols based on the single-photon~\cite{Cab99} and two-photon~\cite{Sim03} detection. Both protocols are based on the atomic excitation, indistinguishability and \keyword{interference} of the emitted photons and on state-projective detection, as illustrated in Fig.~\ref{fig:protocolsEnt}.
\begin{figure}[t]
\begin{center}
\includegraphics[scale=0.35]{Slodicka_fig5}
\caption{Both one- and two-photon entanglement protocols use the same steps for generating heralded entanglement of distant atoms. After electronic excitation the atoms decay back to a ground state, while spontaneously emitting a photon. The scattered photons interfere to make it indistiguishable which of the two atoms has scattered the photon. Finally, the photons are detected and project the atoms into entangled state. For the two-photon protocol, two photons must be emitted from two distant atoms and the detection corresponds to the projection onto one of the Bell states in the photon basis. For the single photon scheme, only one photon has to be emitted and detected. Depicted energy levels correspond to the typical schemes employed for the two protocols with $|g\rangle$, $|i\rangle$ and $|e\rangle$ corresponding to the initial ground state, auxiliary excited state and final state after Raman process, respectively.}
\label{fig:protocolsEnt}
\end{center}
\end{figure}
We define two dimensionless measures crucial for the performance of any practical quantum information network~\cite{Zip08}. \keyword{Fidelity} between the generated state described by the density matrix $\rho$ and the desired maximally entangled two qubit state $|\psi\rangle$,
\begin{equation}
F = \langle \psi |\rho|\psi\rangle
\end{equation}
and success probability $P_{\rm s}$, corresponding to probability with which this state can be generated for given overall detection efficiency $\eta$.
\begin{figure}[!ht!]
\begin{center}
\includegraphics[scale=1.0]{Slodicka_fig6}
\caption{Success probability ratio of entanglement generation for the single-photon and two-photon protocols. For current ion-trapping experimental setups \cite{Olm09,Slo13, Str11,Shu11,Mai12} the overall collection and detection efficiency is limited to few percents. For such realistic detection efficiencies, the single-photon entanglement generation scheme has potential to be several orders of magnitude faster than the two-photon scheme. The three highest detection efficiencies values are from experiments where fluorescence was detected directly, without coupling to optical fiber. \label{Motivation}}
\end{center}
\end{figure}
Following the simplified model in the work of Zippilli et al.~\cite{Zip08}, the fidelity and success rate of the single-photon protocol are given by
\begin{equation}
F_1 \sim (1 -p_{\rm e})/(1 - \eta p_{\rm e})\,\,\,\,\,{\rm and}\,\,\,\,\,P_{\rm s,1} \sim 2 \eta p_{\rm e} (1 - \eta p_{\rm e}).
\label{effSingle}
\end{equation}
Here $p_{\rm e}$ is the probability of the successful excitation and emission of a single photon by a single ion. For a given value of $p_{\rm e}$, the fidelity increases with overall detection efficiency because the likelihood of detecting events where two photons are scattered increases. For a two-photon protocol, the effect of detection efficiency on generated state fidelity is negligible, because both atoms need to be excited and only coincidence detection events trigger entanglement, and thus the fidelity of the generated state with the maximally entangled state is assumed to be $F_2=1$. However, the rate and \keyword{success probability} of entanglement generation depend here quadratically on $\eta$,
\begin{equation}
P_{\rm s,2} \sim \eta^2,
\end{equation}
since the two photons need to be detected at the same time.
Fig.~\ref{Motivation} shows the ratio of success probabilities $R=P_{\rm s,1}/P_{\rm s,2}$ of the two protocols for fixed values of the generated states fidelities as a function of detection efficiency. For a given desired fidelity the two-photon scheme is faster only for high overall detection efficiencies. There is a large advantage in using the single-photon scheme for experimental setups with detection efficiencies below $10^{-2}$. For most of currently realized single-atom experiments, the theoretical gain in entanglement generation rate using the single-photon scheme thus corresponds to several orders of magnitude. In addition, even for unrealistically high detection efficiencies of more than 90\,\%, the single-photon scheme can give higher success rates of generated entangled states with high fidelities. This is due to the high detection probability of double excitations in this limit, which correspond to the fundamental source of infidelity in the single-photon protocol.
\subsection{Generation of entanglement by a single photon detection}
The entanglement of distant single atoms through the detection of a single photon was proposed in the seminal work of Cabrillo et al.~\cite{Cab99}. In this scheme, two atoms (A,B) are initially both prepared in the same long-lived electronic state $|gg\rangle$, see Fig.~\ref{fig:protocolsEnt}. Each atom is then excited with a small probability $p_e$ to another long-lived state $|e\rangle$ through a spontaneous \keyword{Raman process} ($|g\rangle\rightarrow|i\rangle\rightarrow|e\rangle $) by weak excitation of the $|g\rangle\rightarrow|i\rangle$ transition and spontaneous emission of the single photon on the $|i\rangle\rightarrow|e\rangle$ transition. Here $|i\rangle$ denotes an auxiliary atomic state with short lifetime. This Raman process entangles each of the atom's internal states $|s \rangle$ with the emitted photon number $|n \rangle$, so the state of each atom and its corresponding light mode can be written as
\begin{equation}
|s, n\rangle = \sqrt{1-p_e}|g,0\rangle e^{i \phi_{L}} +\sqrt{p_e}|e,1\rangle e^{i \phi_D}.
\end{equation}
The phases $\phi_L$ and $\phi_D$ correspond to the phase of the exciting laser at the position of the atoms and the phase acquired by the spontaneously emitted photons on their way to the detectors, respectively. The total state of the system consisting of both atoms and the light modes can be written as
\begin{equation}
\begin{split}
&|s_{\rm A},s_{\rm B}, n_{\rm A},n_{\rm B}\rangle = (1-p_e)e^{i(\phi_{\rm L,A}+\phi_{\rm L,B})} |gg,00\rangle + \\
& \hspace{+2.8cm} +\sqrt{p_e(1-p_e)}(e^{i(\phi_{\rm L,A}+\phi_{\rm D,B})}|eg,10\rangle+e^{i(\phi_{\rm L,B}+\phi_{\rm D,A})} |ge,01\rangle) + \\
& \hspace{+2.8cm} + p_e e^{i (\phi_{\rm D,A}+\phi_{\rm D,B})}|ee,11\rangle.
\end{split}
\end{equation}
Indistinguishability of the photons from the two atoms is achieved by overlapping their corresponding modes, for example using a beam splitter. Single photon detection then projects the two-atom state onto the entangled state
\begin{equation}
|\Psi^\phi\rangle=\frac{1}{\sqrt{2}}(|eg\rangle+e^{i\phi}|ge\rangle),
\label{stateCreated}
\end{equation}
with the probability of 1-$p_e^2$, where $p_e^2$ is the probability of simultaneous excitation of both atoms. The phase of the generated entangled state $\phi$ corresponds to the sum of the phase differences acquired by the exciting beam at the position of the two atoms and the phase difference acquired by the photons from the respective atoms upon traveling to the detector,
\begin{equation}
\phi=(\phi_{\rm L,B}-\phi_{\rm L,A})+(\phi_{\rm D,A}-\phi_{\rm D,B})).
\label{phi}
\end{equation}
The only limiting factor for the fidelity of the generated state with respect to the maximally entangled state emerging from the presented simplified model is the probability of simultaneous excitation of the two atoms $p_e^2$. However, this can be made arbitrarily small at the expense of entanglement generation success probability $P_{\rm s}$, as demonstrated in Fig.~\ref{Motivation}. The phase of the generated state depends on the relative length of the excitation and detection paths, which therefore need to be stabilized with sub-wavelength precision. Random changes of these path-lengths caused by atomic motion or air density fluctuations change the phase of the entangled state in Eq.~(\ref{stateCreated}) in an incoherent way, which can considerably reduce the fidelity of the generated state. In the experiment we stabilize the phase $\phi$ with interferometric methods to $\phi=0$. The heralded detection of a single photon should then generate the maximally entangled target state
\begin{equation}
|\Psi^+\rangle=\frac{1}{\sqrt{2}}(|eg\rangle+|ge\rangle).
\label{stateCreated}
\end{equation}
\subsection{Experimental realization}
\begin{figure}[t]
\begin{center}
\includegraphics[scale=0.102]{Slodicka_fig7}
\caption{Scheme of the experimental setup for entanglement generation by a single photon detection and relevant electronic level scheme of $^{138}{\rm Ba}^+$~\cite{Slo13}. Fluorescence of the two ions is overlapped using a distant mirror which sets the effective distance between them to $d=1$\,meter. A half wave plate (HWP), a polarizing beam splitter (PBS) and a single-mode optical fiber select the polarization and the spatial mode before an avalanche photodiode (APD1). A non-polarizing beam-splitter and an additional avalanche photodiode (APD2) can be inserted to form a Hanbury-Brown-Twiss setup.
\label{setupEnt}}
\end{center}
\end{figure}
For the experimental realization of the single-photon entanglement generation\linebreak scheme two barium ions are trapped in a linear Paul trap setup. As shown in Fig.~\ref{setupEnt}, laser light at 493\,nm is used to Doppler-cool the ions and to detect their electronic states by means of electron shelving, and a laser field at 650\,nm pumps the atoms back to the 6P$_{1/2}$ level from the metastable 5D$_{3/2}$ state. By carefully adjusting the cooling and trapping parameters, the ions are always well within the Lamb-Dicke limit so that the photon \keyword{recoil} during the Raman scattering process is mostly carried by the trap. This ensures that only minimal information is retained in the motion of the ion about which atom has scattered the photon during the entanglement generation process. The fluorescence photons are efficiently collected by two high numerical aperture lenses (NA $\approx 0.4$) placed 14\,mm away from the atoms. A magnetic field of 0.41\,mT is applied at an angle of 40\,degrees with respect to the two-ion axis and defines the quantization axis. After passing through a polarizing beam splitter that blocks the $\pi$-polarized light and lets $\sigma$-polarized light pass, the spatial overlap of the photons is guaranteed by collecting the atomic fluorescence of the first ion in a single mode optical fiber, whilst the fluorescence of the second ion is sent to a distant mirror that retro-reflects it in the same optical fiber. The fluorescence of the two ions (including the Raman scattered light) is then detected by an avalanche photodiode with a quantum efficiency of 60\,\%.
In order to produce a pure entangled state of two qubits, the phase $\phi$ of the generated state, defined in Eq.~(\ref{phi}), must be controlled with high precision. This is achieved by a measurement of the phase of the interference produced by the elastic scattering of the 493\,nm Doppler-cooling beam from the two ions. Scattered photons will follow the same optical paths as the photons scattered by the Raman beam just in opposite directions. Observation of their interference can be then used for stabilization of the relative phase of the exciting Raman beam at the position of the two ions.
Every experimental sequence of this measurement starts by Doppler-cooling of the two ions. Then the ion-mirror distance $d/2$ is stabilized by locking the position of the measured interference fringe to a chosen position. The electronic states of the ions are then prepared to the Zeeman substate $|6{\rm S}_{1/2, (m=-1/2)}\rangle = |g\rangle$ by optical pumping with a circularly polarized 493\,nm laser pulse propagating along the magnetic field. Next, a weak horizontally polarized laser pulse excites both ions on the S$_{1/2}\leftrightarrow$ P$_{1/2}$ transition with a probability $p_e=0.07$. From the excited state the ion can decay to the other Zeeman sublevel $|6{\rm S}_{1/2,(m=+1/2)}\rangle = |e\rangle$, see Fig.~\ref{sequenceCab}. The electronic state of each ion is at this point entangled with the number of photons $|0\rangle$ or $|1\rangle$ in the $\sigma^-$ polarized photonic mode. Provided that high indistinguishability of the two photonic channels is assured and that simultaneous excitation of both atoms is negligible, detection of a single $\sigma^-$ photon on the APD projects the two-ion state onto the maximally entangled state given by Eq.~(\ref{stateCreated}).
\begin{figure}
\begin{center}
\includegraphics[scale=0.13]{Slodicka_fig8}
\caption{Experimental sequence~\cite{Slo13}. Spontaneous Raman scattering from $|g\rangle$ to $|e\rangle$ triggers emission of a single photon from the two atoms. Upon successful detection of a $\sigma^-$ photon, state analysis comprising coherent radio-frequency (RF) pulses at 11.5\,MHz, and electron shelving to the 5D$_{5/2}$ level are performed. \label{sequenceCab}}
\end{center}
\end{figure}
Following the detection of a Raman scattered $\sigma^-$ photon, the two-atom state is coherently manipulated to allow for measurements in a different basis used for the estimation of the generated state. As shown in Fig.~\ref{sequenceCab}, this is done by first applying radio-frequency (RF) pulses that are resonant with the $|g\rangle \leftrightarrow |e\rangle$ transition of both atoms at transition frequency of 11.5\,MHz. Finally, discrimination between the two Zeeman sub-levels of the 6S$_{1/2}$ state is done by shelving the population of the 6S$_{1/2,(m=-1/2)}$ state to the metastable 5D$_{5/2}$ level using a narrowband 1.76\,$\mu$m laser. The fluorescence rate on the 6S$_{1/2}\leftrightarrow {\rm 6P}_{1/2}$ transition \cite{Slo13} allows distinguishing between having no excitation at all $\rho_{gg}$, a single delocalized excitation $\rho_{ge}$ or $\rho_{eg}$, and two excitations $\rho_{ee}$ in the two-atoms system. These events can be separated with 98\,\% probability, which enables efficient reconstruction of the relevant parts of the two-atom density matrix. The 614\,nm laser field then resets the ions to the 6S$_{1/2}$ state and the same experiment is repeated 100 times.
\subsubsection{Estimation of the generated state}
The success rate and fidelity of the generated entangled state of
distant ions can be estimated by measuring the overlap of the
generated state with the desired entangled state every time the
heralding photon is detected. It is sufficient to measure only
certain parts of the density matrix which contribute to this
overlap~\cite{Slo13}. The fidelity $F=\langle
\Psi^+|\rho|\Psi^+\rangle$ of the general two-qubit state $\rho$
with the desired maximally entangled state $|\Psi^+\rangle$ reads
\begin{equation}
F=\frac{1}{2} [\rho_{ge}+\rho_{eg}+ 2 {\rm Re}(\rho_{eg,ge})].
\label{fidelityRho}
\end{equation}
The fidelity thus depends only on the sum of diagonal populations $\rho_{ge}$ and $\rho_{eg}$ and on the real part of the off-diagonal term $\rho_{eg,ge}$ that expresses the mutual coherence between them. All these terms can be accessed using the collective rotations
\begin{equation}
\hat{R}(\theta,\phi)=\exp{\left[-i\frac{\theta}{2}\left(\cos\phi\hat{S}_x+\sin\phi\hat{S}_y\right)\right]},
\end{equation}
followed by the measurement of parity operator
\begin{equation}
\hat{P}=\hat{p}_{gg} +\hat{p}_{ee}-\hat{p}_{eg}-\hat{p}_{ge},
\end{equation}
where $\hat{p}_{ij}$ are the projection operators on states $|ij\rangle$, $i,j\in \{g,e\}$ in different bases~\cite{Sac00} and $\hat{S}_{x,y}=\hat{\sigma}_{x,y}^{(1)}\otimes I^{(2)}+I^{(1)}\otimes \hat{\sigma}_{x,y}^{(2)}$ is the total angular momentum operator in x- or y-direction for both ions. The rotation angle $\theta$ and rotation axis $\phi$ on the Bloch sphere are determined by the duration and the phase of the RF pulses, respectively.
For the state $|\Psi^+\rangle$, it can be readily shown that
\begin{equation}
{\rm Tr}\left[\hat{P}\hat{R}(\pi/2,\phi)|\Psi^+\rangle\langle\Psi^+|\hat{R}^\dagger (\pi/2,\phi)\right]=1
\end{equation}
for all $\phi$. A parity measurement on the $|\Psi^+\rangle$
entangled state is therefore invariant with respect to the change
of the rotation pulse $\hat{R}(\pi/2,\phi)$ phase $\phi$. In order
to measure the parity oscillations for this state, it first has to
be rotated by a global $\hat{R}(\pi/2,\pi/2)$ pulse, corresponding
to a $\hat{\sigma}_y$ rotation on both qubits with the pulse area
of $\pi/2$.
\subsubsection{Entanglement generation results}
\begin{figure}[!t!]
\begin{center}
\includegraphics[scale=2.0]{Slodicka_fig9}
\caption{Characterization of the entangled state~\cite{Slo13}. a)
Two-atom state populations after the detection of a $\sigma^-$
photon showing that the total probability of measuring the state
with a single excitation is (89$\pm 3$)\,\%. Spurious populations
of the $|gg\rangle$ state are caused by double excitations of each
ion (0.07) and dark count rate of the employed avalanche diode
(0.02). State $|ee\rangle$ is populated due to the simultaneous
excitation of the two ions. b) Parity measurements as a function
of the RF-phase. Trace (ii) corresponds to the measurement of the
atomic populations after two global rotations
$\hat{R}(\pi/2,\pi/2)\hat{R}(\pi/2,\phi)$. In the measurement of
trace (i) only a single global RF-pulse $\hat{R}(\pi/2,\phi)$ is
applied.\label{Parity}}
\end{center}
\end{figure}
The electronic state of two ions is analyzed after each heralding
photon detection. Fig.~\ref{Parity}-a) shows that in (89$\pm
3$)\,\% of the heralded events correspond to the events where only
one of the atoms was excited to the $|e\rangle$ state. This is in
good agreement with the excitation probability $p_{\rm e}=0.07 \pm
0.03$ of each ion and the measured dark-count rate of the employed
avalanche photodiode of 10\,counts/s.
Fig.~\ref{Parity}-b), trace (ii), shows the results of the parity
operator $\hat{P}$ measurements that are, as explained above,
preceded by two global RF rotations
$\hat{R}(\pi/2,\pi/2)\hat{R}(\pi/2,\phi)$ for estimation of the
quantum coherence of the generated state. The first applied pulse
$\hat{R}(\pi/2,\pi/2)$ performs the unitary rotation
$\hat{R}(\pi/2,\pi/2)|\Psi^+\rangle \rightarrow |\Phi^-\rangle$,
where $|\Phi^-\rangle=\frac{1}{\sqrt{2}}(|gg\rangle-|ee\rangle)$.
The second RF-pulse with same duration but with a phase $\phi$
then performs the rotation $\hat{R}(\pi/2,\phi)|\Phi^-\rangle$.
The measured parity signal clearly oscillates as a function of
phase $\phi$ with contrast of (58.0$\pm 2.5$)\,\% and period of
$\pi$, a proof that we indeed succeed in preparing an entangled
two-ion state close to $|\Psi^+\rangle$~\cite{Sac00}. The mean
value of the parity operator at zero phase
$\langle\hat{P}\rangle_{\phi\rightarrow 0}$ corresponds to the
difference between the inner parts and outer-most coherence terms
of the density matrix. The measured value corresponds to
$2\rm{Re}(\rho_{ge,eg}-\rho_{gg,ee})=0.38\pm 0.03$. To precisely
quantify the fidelity of the generated state with
$|\Psi^+\rangle$, the real part of the coherence term
$\rho_{ge,eg}$ needs to be estimated. This is done by measuring
the parity without the first RF rotation. Trace (i) of
Fig.~\ref{Parity}-b) shows the expectation value of the parity as
a function of the phase $\phi$ of the single RF-pulse. The
invariance of the measurement result with respect to the phase
$\phi$ proves that $\rho_{gg,ee}=0\pm 0.03$, so that indeed only
the coherence corresponding to the state $|\Psi^+\rangle$ is
measured. The estimated fidelity of the generated state with the
maximally entangled state is $|\Psi^+\rangle$ is $F=(63.5\pm
2)\,\%$. The threshold for an entanglement is thus surpassed by
more than six standard deviations. The coherence between the
$|ge\rangle$ and $|eg\rangle$ states of $(38\pm 3)\,\%$ is limited
by three main processes ~\cite{Slo13}. First, imperfect
populations of $|ge\rangle$ and $|eg\rangle$ states set a limit of
89\,\%~\cite{Shi06}. Around 4\,\% of the coherence loss can be
attributed to the finite coherence time of the individual atomic
qubits (120\,$\mu$s) due to collective magnetic field
fluctuations. Although the generated $|\Psi^+\rangle$ state is
intrinsically insensitive against collective dephasing~\cite{Kie01,Roo04},
a loss of coherence is indeed expected after a
rotation of $|\Psi^+\rangle$ out of the decoherence-free subspace.
The highest contribution to the coherence loss can be attributed
to atomic motion, which can provide information about which atom
emitted the photon. Around 55\,\% of the coherence is lost due to
the atomic recoil kicks during the Raman scattering.
The overall fidelity of the maximally entangled state
$|\Psi^+\rangle$ with the experimentally generated one is limited
mainly by the imperfect populations of the desired $|ge \rangle$
and $|eg\rangle$ states and coherence loss due to the atomic
recoil kicks during the Raman scattering. The effect of
motion-induced decoherence can be reduced by cooling the radial
modes to the motional ground state~\cite{Slo12} or by choosing a
forward Raman scattering scenario~\cite{Cab99}. Error bars in the
presented measurements results correspond to one standard
deviation and are estimated statistically from several
experimental runs each giving approximately 120 measurement
outcomes. Up to 60\,\% of the measurement error is caused by the
quantum projection noise. Additional uncertainty comes from slow
magnetic field drift with a magnitude of several tens of nT making
the RF-driving off-resonant by tens of kHz.
An important advantage of the single-photon heralding mechanism is
the possibility of achieving a high entanglement generation rate.
In the presented experiment, the entanglement generation rate has
reached (14.0$\pm 1.5$)\,events/minute with an experimental duty
cycle of 2.3\,kHz~\cite{Slo13}. The probability of successful
entanglement generation per each experimental trial estimated from
the single photon detection efficiency and the measured
probability of Raman scattering is $P_{\rm succ}=2 p_e \eta =
1.1\times 10^{-4}$ that corresponds to 15.4 successful
entanglement generation events/minute. The factor two here
corresponds to the probability of detecting a single photon from
one of two ions. The efficiency of detecting a single
Raman-scattered photon was estimated to be $\eta =
8\times\,10^{-4}$. It was derived from the detection probability
of a single Raman scattered photon given by the collection
efficiency of high-NA lenses ($\sim 0.04$), the single-mode fiber
coupling efficiency ($\sim 0.1$) and by the avalanche-photodiode
detection efficiency ($\sim 0.6$). Additional factors of 0.5 and
0.66 come from the polarization filtering of unwanted
$\pi$-polarized photons and from the probability for the ion to
decay back to the $|g\rangle$ state after the Raman pulse
excitation, respectively. The single ion excitation probability
was $p_e = (0.07\pm 0.03)$\,\%. For comparison, the two-photon
heralding entanglement scheme proposed by Simon et
al.~\cite{Sim03} would for the employed experimental setup give
approximately $P_{\rm succ} \approx 2 \eta^2 = 1.3\times 10^{-6}$,
so about two orders of magnitude smaller success probability of
entanglement generation. For simplicity, $p_e=1$ was assumed here
for the two-photon scheme and an additional factor of 2 accounted
from the two possible contributions to coincidence detection
events.
\subsection{Summary}
In this chapter we have reviewed two methods for free-space
coupling between single ions and photons: the coupling of a weak
probe laser to a single ion in free space (section~\ref{sec:1}),
and the probabilistic generation of entanglement between two ions
by detecting a single scattered photon (section~\ref{sec:2}). Both
method rely on high-numerical aperture optics in order to achieve
a sufficiently high efficiency.
In the experiments presented in section~\ref{sec:1} we have
investigated the free-space interaction of a single ion to a weak
near-resonant probe field. The coupling mediated by a single
objective covering 4\,\% of the full solid angle resulted in an
extinction of the probe field of 1.5\,\%, a phase shift of up to
0.3~degrees, and the observation of electromagnetically-induced
transparency from single atom. Current experimental efforts to
increase the numerical aperture should significantly improve the
interaction strength (see chapter by Leuchs \& Sondermann), which will likely lead
to a number of direct applications of these effects in the field
of quantum information, quantum feedback or single atom
magnetometry. Utilization of a single ion as an optical mirror in
a Fabry-P\'erot-like cavity set-up enabled the observation of
almost full suppression and enhancement by a factor of two of the
atomic coupling constant in the laser probe mode. Besides the
appealing quantum memory applications of such a
set-up~\cite{Wan12}, the single ion mirror has the potential to become
useful for the realization of an optical switch similar to the
single atom transistors using EIT. Furthermore, the presented
experiment enables to study the quantum electrodynamics in an
exciting regime where both the free-space coupling of the probe
beam and the modification of the vacuum mode density at the
position of an ion play an important role~\cite{het10}.
In section~\ref{sec:2} of this chapter we have summarized an
experimental realization of a proposal by Cabrillo et
al.~\cite{Cab99} where the detection of a single scattered photon
generates entanglement between two ions. This presents an
important step towards the realization of the quantum information
networks with ions and photons. The maximally entangled state
$|\Psi^+\rangle$ was produced with a fidelity of 63.5\,\% and with
entanglement generation rates of 14\,events/minute, which is more
than two orders of magnitude higher than the rate obtainable with
protocols relying on a two-photon coincidence events with the
presented experimental parameters. These results can be further
improved by cooling all of the involved motional modes close to
their ground state~\cite{Slo12} or by choosing a different
excitation direction to minimize residual which-way information.
There are some obvious questions regarding the technical difficulties related to phase stability requirements and photon recoil problems of the single-photon entanglement generation scheme. For generation of the entanglement between distant atoms, the paths of the excitation and detection channels need to be interferometrically stable. This issue has been addressed by the community developing fiber links for comparing remote optical clocks. Recently, coherent laser light transfer over more than 900 km has been shown with a precision exceeding the requirements of the single-photon protocol~\cite{Pre12}. The problem of a which-way information available due to the atomic recoil upon scattering of single photon can be eliminated by changing the geometry of the system. These improvements, together with the experimental results presented, have potential to enable efficient creation and distribution of entanglement between distant sites with well-defined and controllable atomic qubits.
\begin{acknowledgement} {
We would like to thank all our colleagues who were involved in this work over the course of the years, in particular Nadia R\"ock, Philipp Schindler, Daniel Higginbottom, Fran\c{c}ois Dubin, Alexander Gl\"atzle, Muir Kumph, Gabriel Araneda, Sebastian Gerber, Daniel Rotter, Pavel Bushev, Yves Colombe, and J\"urgen Eschner.
The work reported in this chapter has been supported by the Austrian Science Fund FWF (SINFONIA, SFB FoQuS), by the European Union (CRYTERION), by the Institut f\"ur Quanteninformation GmbH, and a Marie Curie International Incoming Fellowship within the 8th European Framework Program.
The writing of this chapter was supported by the European Research Council project QuaSIRIO.
} \end{acknowledgement}
\bibliographystyle{SpringerPhysMWM}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 6,993
|
Hardened Con Offered Shot at Redemption by Horse Whisperer
FINDING REDEMPTION: In "The Mustang," a violent criminal (Matthias Schoenaerts) learns to tame his anger by participating in a program that pairs inmates with wild mustangs. (Photo courtesy of Focus Features)
By Kam Williams
Roman Coleman (Matthias Schoenaerts) has too quick a fuse to think before he acts. That's why he's done a dozen years and counting in a maximum-security prison for impulsively delivering a brutal beating that left his victim permanently brain-damaged.
Even while incarcerated, Roman never learned to control his temper. Consequently, he's voluntarily spent the bulk of his time in solitary confinement.
A shot at rehabilitation arrives when Myles (Bruce Dern), a salty old horse whisperer, offers Roman a spot in his program pairing inmates with wild mustangs. The hope is that each participant will learn to tame his own raging inner soul while bonding with his stallion.
Roman grudgingly accepts the invitation, and is assigned to work with a bucking bronco he names Marcus. Under the watchful eye of the sage trainer, con and colt do gradually take to each other, although not without plenty of fits and starts.
Marking Laure de Clermont-Tonnerre's directorial debut, The Mustang is a character-driven masterpiece. Schoenaerts and Dern generate considerable chemistry in the course of delivering powerful performances against a variety of visually-captivating backdrops, ranging from the vast expanse of the Nevada desert to the claustrophobic confines of one of the state's penal institutions.
An emotionally engaging meditation on redemption inspired by a real-life program helping inmates turn their lives around.
Excellent (****). Rated R for profanity, violence, and drug use. Running time: 96 minutes. Production Companies: Legende Films/Cine+/Canal+. Distributor: Focus Features.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,741
|
El condado de Stewart (en inglés: Stewart County), fundado en 1830, es uno de 159 condados del estado estadounidense de Georgia. En el año 2008, el condado tenía una población de 4666 habitantes y una densidad poblacional de 3 personas por km². La sede del condado es Lumpkin. El condado recibe su nombre por Daniel Stewart.
Geografía
Según la Oficina del Censo, el condado tiene un área total de , de la cual es tierra y (0.98%) es agua.
Condados adyacentes
Condado de Chattahooche (norte)
Condado de Webster (este)
Condado de Randolph (sur)
Condado de Quitman (suroeste)
Condado de Barbour, Alabama (oeste)
Condado de Russell, Alabama (noroeste)
Demografía
Según el censo de 2000, había 5252 personas, 2007 hogares y 1348 familias residiendo en la localidad. La densidad de población era de 4 hab./km². Había 2354 viviendas con una densidad media de 2 viviendas/km². El 37.11% de los habitantes eran blancos, el 61.54% afroamericanos, el 0.25% amerindios, el 0.17% asiáticos, el 0.01% isleños del Pacífico, el 0.11% de otras razas y el 0.82% pertenecía a dos o más razas. El 1.50% de la población eran hispanos o latinos de cualquier raza.
Los ingresos medios por hogar en la localidad eran de $24 789, y los ingresos medios por familia eran $29 611. Los hombres tenían unos ingresos medios de $27 568 frente a los $19 035 para las mujeres. La renta per cápita para el condado era de $16 071. Alrededor del 22.20% de la población estaban por debajo del umbral de pobreza.
Transporte
Principales carreteras
U.S. Route 280
Localidades
Americus
Andersonville
De Soto
Leslie
Plains
Cobb
Referencias
Enlaces externos
"Stewart County", New Georgia Encyclopedia
Stewart County Georgia Community Web Pages
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 7,764
|
{"url":"http:\/\/openstudy.com\/updates\/4dcb1ae940ec8b0b5eb01117","text":"## anonymous 5 years ago a pool is shaped like a rectangle with a length 4 times its width w. What is an expression for the distance between opposite corners of the pool?\n\n1. anonymous\n\nCan anyone help?\n\n2. anonymous\n\n$\\text{diagonal}=4w \\text{Sec}\\left[\\text{ArcTan}\\left[\\frac{1}{4}\\right]\\right]$\n\n3. anonymous\n\nThats the answer? :|\n\n4. anonymous\n\nNotice that no dimensions are given, no numbers like the distance around the pool. There is not enough information to solve for x. The diagonal is a function of w. Another expression: $\\text{diagonal}=\\sqrt{w^2 + 16 w^2}$","date":"2017-01-23 14:44:47","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\": 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\": 0.61419278383255, \"perplexity\": 736.0555192996376}, \"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-2017-04\/segments\/1484560282932.75\/warc\/CC-MAIN-20170116095122-00068-ip-10-171-10-70.ec2.internal.warc.gz\"}"}
| null | null |
Ernst Arbman, född den 2 november 1818 i Undersåkers socken i Jämtlands län, död den 18 maj 1896, var en svensk präst. Han var son till Johan Olof Arbman samt far till Johannes och Olof Emanuel Arbman.
Arbman blev 1862 kyrkoherde i Sunne socken, Jämtland och var 1874–1893 prost över Jämtlands östra kontrakt. Han tillhörde ledarna inom den jämtländska väckelsen under 1800-talet.
Han var gift med Eva Eskilina Dillner (1820–1907), född i Brunflo socken i Jämtlands län som dotter till prosten Olof Dillner. De hade utöver ovannämnda söner även fyra döttrar, bland dem Rosa Sofia (1861–1919), som blev författarinna och folklorist.
Källor
Svensk uppslagsbok, 2:a upplagan, 1947
Leonard Bygdén, Hernösands stifts herdaminne
Noter
Vidare läsning
C.J.E. Hasselberg, Ernst Arbman och den jämtländska väckelsen på hans tid (1923)
Svenska kontraktsprostar
Präster i Svenska kyrkan
Personligheter inom pietismen
Personer från Undersåkers socken
Män
Födda 1818
Avlidna 1896
Svensk uppslagsbok
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 8,035
|
{"url":"https:\/\/solvedlib.com\/martin-company-applies-manufacturing-overhead,130899","text":"# Martin Company applies manufacturing overhead based on direct machine hours. Martin estimates manufacturing overhead and labor...\n\n###### Question:\n\nMartin Company applies manufacturing overhead based on direct machine hours. Martin estimates manufacturing overhead and labor for the year follows: Estimated manufacturing overhead $120,000 Direct labor hours estimated 5,000 Estimated direct labor costs$50,000 Estimated machine hours 6,000 Assuming the actual manufacturing overhead cost and the actual activities for Job 201 is as follows: Actual manufacturing overhead $12,000 Direct labor hours actual incurred 300 Actual direct labor costs totaled$ 9,800 Actual machine hours totaled 400 Instructions (a) Compute the predetermined overhead rate. Please show your calculation. (b) What is the amount of manufacturing overhead applied to Job 201? Please show your calculation.\n\n#### Similar Solved Questions\n\n##### I was able to finish some of this problem pls help with the unfinished portions Required...\nI was able to finish some of this problem pls help with the unfinished portions Required Information The following information applies to the questions displayed below.] Tony and Suzie graduate from college in May 2021 and begin developing their new business. They begin by offering clin...\n##### Write each expression a5 single logarithm:4. 2l0g3 U log3 log( (3? Bw + 2) _ 2log(z + 1) 6. 21 log3 Va + log: (9z?) og:\nWrite each expression a5 single logarithm: 4. 2l0g3 U log3 log( (3? Bw + 2) _ 2log(z + 1) 6. 21 log3 Va + log: (9z?) og:...\n##### The U.S. healthcare system is currently in a stage where consumer choices are re-shaping health care....\nThe U.S. healthcare system is currently in a stage where consumer choices are re-shaping health care. What are the resulting strategic responses to this new stage? Do you work for an organization that is responding in any of these ways? Explain....\n##### 1. The following circuit shows a discrete common source MOSFET amplifier. The MOSFTE is n-channel MOSFET...\n1. The following circuit shows a discrete common source MOSFET amplifier. The MOSFTE is n-channel MOSFET and early voltage (Va) is c. The transconductance of the amplifier (ga) is 3 mA\/V The frequency responses of the amplifier are as follows i) The three low break frequencies f 3Hz (caused by C).fi...\n##### Initially contains 300 liters of Huid on which there is dissolved 50 500 liter tank grams of sulfuric acid. Fluid containing 30 gramns per liter of sulfuric acid dows into the tank At 4 rate of liters per minute_ The mixture is kept unilorm by stirring;, and the stirred mixture simultaneously flows Qut of the tank al a rate 0f 2.5 literd per minute: How much of sulfuric acid is in the tank at Lhe instant it overflows\" 76\ninitially contains 300 liters of Huid on which there is dissolved 50 500 liter tank grams of sulfuric acid. Fluid containing 30 gramns per liter of sulfuric acid dows into the tank At 4 rate of liters per minute_ The mixture is kept unilorm by stirring;, and the stirred mixture simultaneously flows ...\n##### Described below are six independent and unrelated situations involving accounting changes. Each change occurs during 2018...\nDescribed below are six independent and unrelated situations involving accounting changes. Each change occurs during 2018 before any adjusting entries or closing entries were prepared. Assume the tax rate for each company is 40% in all years. Any tax effects should be adjusted through the deferred t...\n##### Studies of automobile demand suggest that unit sales of compact cars depend principally on their average...\nStudies of automobile demand suggest that unit sales of compact cars depend principally on their average price and consumers\u2019 real personal income. Consider the historical record of sales shown in the table. Estimate the point elasticity of demand with respect to price. (Be sure to choose two...\n##### Eyamaie -nentecrz \"neretne qrer curexe\"< Js_sgment frcm0) =0 (4,\nEyamaie -ne ntecrz \"nere tne qrer cure xe\"< Js_ sgment frcm 0) =0 (4,...\n##### Vector Potential _ Any sufficiently nice vector field F on R3 with vanishing diver- gence; thus such that F = 0, can be expressed as F = x G for some vector field We call G a vector potential of F. (In general, vector potentials are not unique: _ Find & vector potential of the form G = (C,m, 0} for F = (yz, 12, Iy)-\nVector Potential _ Any sufficiently nice vector field F on R3 with vanishing diver- gence; thus such that F = 0, can be expressed as F = x G for some vector field We call G a vector potential of F. (In general, vector potentials are not unique: _ Find & vector potential of the form G = (C,m, 0} ...\n##### Question no:tThe following data give the correlation coefficient means and standard deviation of rainfall and yield of paddy in a certain tractYield per hectare in KgsAnnual rainfall incmMean973.518.338.4Coefficlent of correlation 0.58 the most Iikely yield of paddy when the annual rainfall Is 22 Cm, other factors being assumed Estimate (25 marks to remaln the same:8 16\nQuestion no:t The following data give the correlation coefficient means and standard deviation of rainfall and yield of paddy in a certain tract Yield per hectare in Kgs Annual rainfall incm Mean 973.5 18.3 38.4 Coefficlent of correlation 0.58 the most Iikely yield of paddy when the annual rainfall...\n##### You wish to test the following claim (Ha) at a significance level of \u03b1=0.01. \u00a0\u00a0\u00a0\u00a0\u00a0 Ho:\u03bc=73.2...\nYou wish to test the following claim (Ha) at a significance level of \u03b1=0.01. \u00a0\u00a0\u00a0\u00a0\u00a0 Ho:\u03bc=73.2 \u00a0\u00a0\u00a0\u00a0\u00a0 Ha:\u03bc<73.2 You believe the population is normally distributed, but you do not know the standard deviation. You obtain the following sam...\n##### What are the critical values, if any, of f(x)=x^3?\nWhat are the critical values, if any, of f(x)=x^3?...\n##### 843 DCb4014b1\n843 DC b4 014 b1...\n##### Point) Leonhard Euler was able to calculate the exact sum of the p-series with p 2:24 = ;Use this fact to find the sum of each series 24 4 (n + 3)2 2 n= (Sn)2\npoint) Leonhard Euler was able to calculate the exact sum of the p-series with p 2: 24 = ; Use this fact to find the sum of each series 24 4 (n + 3)2 2 n= (Sn)2...\n##### Which of the following is NOT a reason why it is difficult to obtain absolute ages...\nWhich of the following is NOT a reason why it is difficult to obtain absolute ages from many rocks. A. Many rocks do not have minerals that can be used to date their crystallization, recrystallization or deposition b. Methods of absolute age determination are expensive c most rocks that can be dated...\n##### Suppose u is unit vector in R' and U = uuT. Which of the following is not correct inference?Uis symmetric.Ur is the orthogonal projection of x onto u_U2 = UUis orthogonal:\nSuppose u is unit vector in R' and U = uuT. Which of the following is not correct inference? Uis symmetric. Ur is the orthogonal projection of x onto u_ U2 = U Uis orthogonal:...\n##### Please help me know the right anwser and the process Consider a metal lattice with atoms...\nPlease help me know the right anwser and the process Consider a metal lattice with atoms separated b 0.143 nm. Electrons will diffract when passing through this crystal if their de Broglie wavelength matches the lattice spacing. What speed (in m\/s) of electrons is required for the de Broglie wavele...\n##### T00 Jaxq4 8 Ierilen sistemin M genel cozumunu lJ - 3 bulunuzs\"-[4--[4 r\nT0 0 Jaxq4 8 Ierilen sistemin M genel cozumunu lJ - 3 bulunuzs \"-[4--[4 r...\n##### QUESTION 4A gas sample has a volume of 1.50 L at a temperature of \u00e2\u20ac\u201c 35.0\u00c2\u00b0C. The gas is expanded to a volume of 3.50L. Assuming no change in pressure or amount of gas, whatis the new temperature in \u00c2\u00b0C?Provide answer with no decimal places. Do not write unit inthe box.QUESTION 5The volume of a gas with a pressure of 1.2 atm increases from1.0 L to 4.5 L. What is the final pressure of thegas in atm, assuming constant temperature andamount of gas?Provide answer with 2 decimal places. Do not wr\nQUESTION 4 A gas sample has a volume of 1.50 L at a temperature of \u00e2\u20ac\u201c 35.0 \u00c2\u00b0C. The gas is expanded to a volume of 3.50 L. Assuming no change in pressure or amount of gas, what is the new temperature in \u00c2\u00b0C? Provide answer with no decimal places. Do not write unit in the box. QUESTION 5 ...\n##### The compound K3[NiF6] is paramagnetic. Determine the oxidation number of nickel in this compound, the most...\nThe compound K3[NiF6] is paramagnetic. Determine the oxidation number of nickel in this compound, the most likely geometry of the coordination around the nickel, and the possible configurations of the d electrons of the nickel. Oxidation number ___-7-6-5-4-3-2-10+1+2+3+4+5+6+7 Geometry Co...\n##### Can someone help with this? [8 points] 8 a) On the unit cell drawn below, sketch...\nCan someone help with this? [8 points] 8 a) On the unit cell drawn below, sketch the location of atoms for a BCC crystal. Be sure to include those that are hidden behind the front faces. b) Derive the cubic dimension a as a function of the atomic radius, R. c) What is the planar density of the (001)...\n##### Itatl Tirulci &\"Illed nicdunuLul?MF cu UyLUT\nItatl Tirulci &\"Illed nicdunuLul?MF cu UyLUT...","date":"2023-04-01 00:26:43","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\": 2, \"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\": 0.702674925327301, \"perplexity\": 3675.8446156298}, \"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\/1679296949694.55\/warc\/CC-MAIN-20230401001704-20230401031704-00796.warc.gz\"}"}
| null | null |
\section{Introduction}
Due to recent advances in machine learning (ML), ML methods are increasingly use in real world scenarios~\cite{deeprl, deepsound, deepnlp, deepcv}. Especially, ML technology is nowadays used in critical situations like predictive policing~\cite{predictivepolicing} and loan approval~\cite{creditriskml}. In order to increase trust and acceptance of these kind of technology, it is important to be able to explain the behaviour and prediction of these models~\cite{molnar2019} - in particular answer questions like ``Why did the model do that? And why not smth. else?''. This becomes even more important in view to legal regulations like the EU regulation on GDPR~\cite{gdpr}, that grants the user a right to an explanation.
A popular method for explaining models~\cite{molnar2019, surveyxai, explainingexplanations, explainableartificialintelligence} are counterfactual explanations (often just called counterfactuals)~\cite{counterfactualwachter}. A counterfactual explanation states changes to some features that lead to a different (specified) behaviour or prediction of the model. Thus, counterfactual explanation can be interpreted as a recommendation what to do in order to achieve a requested goal. This is why counterfactual explanations are that popular - they are intuitive and user-friendly~\cite{molnar2019, counterfactualwachter}.
Counterfactual explanations are an instance of model-agnostic methods. Therefore, counterfactuals are not tailored to a particular model but can be computed for all possible models (in theory). Other instances of model-agnostic methods are feature interaction methods~\cite{featureinteraction}, feature importance methods~\cite{featureimportance}, partial dependency plots~\cite{partialdependenceplots} and local methods that approximates the model locally by an explainable model (e.g. a decisiontree)~\cite{lime2016, decisiontreecounterfactual}. The nice thing about model-agnostic methods is that they (in theory) do not need access to model internals and/or training data - it is sufficient to have an interface where we can pass data points to
the model and observe the output/predictions of the model.
However, it turns out that efficiently computing high quality counterfactual explanations of black-box models can be very difficult~\cite{counterfactualslvq}. Therefore, it is beneficial to develop model-specific methods - that use model internals - for efficiently computing counterfactual explanations. Whenever we have access to model internals, we can use the model-specific method over the model-agnostic method for efficiently computing counterfactual explanations. In this work we focus on such model-specific methods.
In particular, our contributions are:
\begin{itemize}
\item We review model-specific methods for efficiently computing counterfactual explanations of different ML models.
\item We propose model-specific methods for efficiently computing counterfactual explanations of models that have not been considered in literature so far.
\end{itemize}
The remainder of this paper is structured as follows: First, we briefly review counterfactual explanations (section~\ref{sec:counterfactualexplanations}). Then, in section~\ref{sec:computationcounterfactuals} we review and propose model-specific methods for computing counterfactual explanations. Finally, section~\ref{sec:conclusion} summarizes this papers. All derivations and mathematical details can be found in the appendix (section~\ref{sec:appendix}).
\section{Counterfactual explanations}\label{sec:counterfactualexplanations}
Counterfactual explanations~\cite{counterfactualwachter} (often just called counterfactuals) are an instance of example-based explanations~\cite{casebasedreasoning}. Other instances of example-based explanations~\cite{molnar2019} are influential instances~\cite{influentialinstances} and prototypes \& criticisms~\cite{prototypescriticism}.
A counterfactual states a change to some features/dimensions of a given input such that the resulting data point (called counterfactual) has a different (specified) prediction than the original input. Using a counterfactual instance for explaining the prediction of the original input is considered to be fairly intuitive, human-friendly and useful because it tells people what to do in order to achieve a desired outcome~\cite{counterfactualwachter,molnar2019}.
A classical use case of counterfactual explanations is loan application~\cite{creditriskml,molnar2019}:
\textit{Imagine you applied for a credit at a bank. Unfortunately, the bank rejects your application. Now, you would like to know why. In particular, you would like to know what would have to be different so that your application would have been accepted. A possible explanation might be that you would have been accepted if you would earn 500\$ more per month and if you would not have a second credit card.}
Although counterfactuals constitute very intuitive explanation mechanisms, there do exist a couple of problems.
One problem is that there often exist more than one counterfactual - this is called \textit{Rashomon effect}~\cite{molnar2019}. If there are more than one possible explanation (counterfactual), it is not clear which one should be selected.
An alternative - but very similar in the spirit - to counterfactuals~\cite{counterfactualwachter} is the Growing Spheres method~\cite{growingsphere}. However, this method suffers from the curse of dimensionality because it has to draw samples from the input space, which can become difficult if the input space is high-dimensional.\\\\
According to~\cite{counterfactualwachter}, we formally define the finding of a counterfactual as follows: Assume a prediction function $\ensuremath{h}: \set{X} \mapsto \set{Y}$ is given. Computing a counterfactual $\ensuremath{\vec{x}'} \in \mathbb{R}^d$ of a given input $\ensuremath{\vec{x}} \in \mathbb{R}^d$\footnote{We restrict ourself to $\mathbb{R}^d$, but in theory one could use an arbitrary domain $\ensuremath{\set{X}}$.} can be interpreted as an optimization problem:
\begin{equation}\label{eq:counterfactualoptproblem}
\underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}{\arg\min}\; \loss\big(\ensuremath{h}(\ensuremath{\vec{x}'}), \ensuremath{y'}\big) + C \cdot \regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}})
\end{equation}
where $\loss()$ denotes a loss function that penalizes deviation of the prediction $\ensuremath{h}(\ensuremath{\vec{x}'})$ from the requested prediction $\ensuremath{y'}$. $\regularization()$ denotes a regularization that penalizes deviations from the original input $\ensuremath{\vec{x}}$ and the hyperparameter $C$ denotes the regularization strength.
Two common regularizations are the weighted Manhattan distance and the generalized L2 distance. The weighted Manhattan distance is defined as
\begin{equation}\label{eq:weighted_l1}
\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) = \sum_j \alpha_j \cdot |(\ensuremath{\vec{x}})_j - (\ensuremath{\vec{x}'})_j|
\end{equation}
where $\alpha_j > 0$ denote the feature wise weights. A popular choice~\cite{counterfactualwachter} for $\alpha_j$ is the inverse median absolute deviation of the $j$-th feature median in the training data set $\set{D}$:
\begin{equation}
\begin{split}
& \alpha_j = \frac{1}{{\mad}_j} \\
& \text{where }\\
& {\mad}_j = \underset{\ensuremath{\vec{x}} \,\in\, \set{D}}{\median}\left(\left|(\ensuremath{\vec{x}})_j - \underset{\ensuremath{\vec{x}} \,\in\, \set{D}}{\median}\big((\ensuremath{\vec{x}})_j\big)\right|\right)
\end{split}
\end{equation}
The weights $\alpha_j$ compensate for the (potentially) different variability of the features. However, because we need access to the training data set $\set{D}$, this regularization is not a truly model-agnostic method - it is not usable if we only have access to a prediction interface of a black-box model.
Although counterfactual explanations are a model-agnostic method, the computation of a counterfactual becomes much more efficient when having access to the internals of the model. In this work we assume that we have access to all needed model internals as well as access to the training data set - we will only need the training data for computing the weights $\alpha_j$ in the weighted Manhattan distance Eq.~\ref{eq:weighted_l1}. We do not need access to the training data if we do not use the weighted Manhattan distance or if we use some other methods for computing the weights $\alpha_j$ (e.g. setting all weights to $1$).\\\\
A slightly modified version of Eq.~\ref{eq:counterfactualoptproblem} was proposed in~\cite{counterfactualguidedbyprototypes}. The authors claim that the original formalization in Eq.~\ref{eq:counterfactualoptproblem} does not take into account that the counterfactual should lie on the data manifold - the counterfactual should be a plausible data instance. To deal with this issue, the authors propose to add two additional terms to the original objective Eq.~\ref{eq:counterfactualoptproblem}:
\begin{enumerate}
\item The distance/norm between the counterfactual $\ensuremath{\vec{x}'}$ and the reconstructed version of it that has been computed by using a pretrained autoencoder.
\item The distance/norm between the encoding of the counterfactual $\ensuremath{\vec{x}'}$ and the mean encoding of training samples that belong to the requested class $\ensuremath{y'}$.
\end{enumerate}
The first term is supposed to make sure that the counterfactual $\ensuremath{\vec{x}'}$ lies on the data manifold and thus is a plausible data instance. The second term is supposed to accelerate the solver for computing the solution of the final optimization problem. Both claims have been evaluated empirically~\cite{counterfactualguidedbyprototypes}.
Recently, another approach for computing plausible/feasible counterfactual explanations was proposed~\cite{face}. Instead of computing a single counterfactual, the authors propose to compute a path of intermediate counterfactuals that lead to the final counterfactual. The idea behind this path of intermediate counterfactuals is to provide the user with a set of intermediate goals that finally lead to the desired goal - it might be more feasible to ``go into the direction" of the final goal step by step instead of accomplishing it in a single step. In order to compute such a path of intermediate counterfactuals, the authors propose different strategies for constructing a graph on the training data set - including the query point. In this graph, two samples are connected by a weighted edge if they are ``sufficient close to each other" - the authors propose different measurements for closeness (e.g. based on density estimation). The path of intermediate counterfactuals is equal to the shortest path between the query point and a point that satisfies the desired goal - this is the final counterfactual. Therefore the final counterfactual as well as all intermediate counterfactuals are elements from the training data set.
Despite the highlighted issues~\cite{counterfactualguidedbyprototypes, face} of the original formalization Eq.~\ref{eq:counterfactualoptproblem}, we stick to it and leave further investigations on the computation of feasible \& plausible counterfactuals as future research. However, many of the approaches for computing counterfactuals - that are discussed in this paper - can be augmented to restrict the space of potential counterfactuals. These restrictions provide an opportunity for encoding domain knowledge that lead to more plausible and feasible counterfactuals.
\section{Computation of counterfactuals}~\label{sec:computationcounterfactuals}
In the subsequent sections we explore model-specific methods for efficiently computing counterfactual explanations of many different ML models. But before looking at model-specific methods, we first (section~\ref{sec:generalcase}) discuss methods for dealing with arbitrary types of models - gradient based as well as gradient free methods.
Note that for the purpose of better readability and due to space constraints, we put all derivations in the appendix (section~\ref{sec:appendix}).
\subsection{The general case}\label{sec:generalcase}
We can compute a counterfactual explanation of any model we like by plugging the prediction function $\ensuremath{h}$ of the model into Eq.~\ref{eq:counterfactualoptproblem} and choosing a loss (eq. 0-1 loss) and regularization (e.g. Manhattan distance) function. Depending on the model, loss and regularization function, the resulting optimization problem might be differentiable or not. If it is differentiable, we can use gradient-based methods like (L-)BFGS and conjugate gradients for solving the optimization problem. If Eq.~\ref{eq:counterfactualoptproblem} is not differentiable, we can use gradient-free methods like the Downhill-Simplex method or an evolutionary algorithm like CMA-ES or CERTIFAI~\cite{counterfactualsgeneticalgorithm} - the nice thing about evolutionary algorithms is that they can easily deal with categorical features. Another approach, limited to linear classifiers, for handling contious and discrete features is to use mixed-integer programming (MIP)~\cite{mipcounterfactual}. Unfortuantely, solving a MIP is NP-hard. However, there exist solvers that can compute an approximate solution very efficiently. Popular methods are branch-and-bound and branch-and-cut algorithms~\cite{mipsolver}.
When developing model-specific methods for computing counterfactuals, we always consider untransformed inputs only - since a non-linear feature transformation usually makes the problem non-convex. Furthermore, we only consider the Euclidean distance and the weighted Manhattan distance as candidates for the regularization function $\regularization(\cdot)$.
\subsection{Separating hyperplane models}\label{sec:separatinghyperplane}
A model whose prediction function $\ensuremath{h}$ can be written as:
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \sign(\vec{w}^\top\ensuremath{\vec{x}} + b)
\end{equation}
is called a separating hyperplane model. Popular instances of separating hyperplane models are SVM, LDA, perceptron and logistic regression.
Without loss of generality, we assume $\set{Y}=\{-1, 1\}$. Then, the optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:sephyperplane}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \vec{q}^\top\ensuremath{\vec{x}'} + c < 0
\end{split}
\end{equation}
where
\begin{equation}
\vec{q} = -\ensuremath{y'}\vec{w}
\end{equation}
\begin{equation}
c = -b\ensuremath{y'}
\end{equation}
Depending on the regularization, the optimization problem Eq.~\ref{eq:cf:sephyperplane} becomes either a linear program (LP) - if the weighted Manhattan distance is used - or a convex quadratic program (QP) with linear constraints - if the Euclidean distance is used. More details can be found in the appendix (section~\ref{sec:appendix:sephyperplane}).
If we would have some discrete features instead of contious features only, we would obtain a MIP or MIQP as described in~\cite{mipcounterfactual}.
\subsection{Generalized linear model}\label{sec:glm}
In a generalized linear model we assume that the distribution of the response variable belongs to the exponential family. The expected value is connected to a linear combination of features by a link function, where different distributions have different link functions.
In the subsequent sections, we explore how to efficiently compute counterfactual explanations of popular instances of the generalized model.
\subsubsection{Logistic regression}\label{sec:logreg}
In logistic regression we model the response variable as a Bernoulli distribution. The prediction function $\ensuremath{h}$ of a logistic regression model is given as
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \begin{cases} 1 & \quad \text{if } p(y=1 \mid \ensuremath{\vec{x}}) \geq t \\ -1 & \quad \text{otherwise} \end{cases}
\end{equation}
where $t$ is the discrimination threshold (often $t=0.5$) and
\begin{equation}
p(y=1 \mid \ensuremath{\vec{x}}) = \frac{1}{1 + \exp(-\vec{w}^\top\ensuremath{\vec{x}} - b)}
\end{equation}
When ignoring all probabilities and setting $t=0.5$, the prediction function $\ensuremath{h}$ of a logistic regression model becomes a separating hyperplane:
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \sign(\vec{w}^\top\ensuremath{\vec{x}} + b)
\end{equation}
Therefore, computing a counterfactual of a logistic regression model is exactly the same as for a separating hyperplane model (section~\ref{sec:separatinghyperplane}).
\subsubsection{Softmax regression}\label{sec:softmaxref}
In softmax regression we model the distribution of the response variable as a generalized Bernoulli distribution. The prediction function $\ensuremath{h}$ of a softmax regression model is given as:
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\frac{\exp(\vec{w}_i^\top\ensuremath{\vec{x}} + b_i)}{\sum_k \exp(\vec{w}_k^\top\ensuremath{\vec{x}} + b_k)}
\end{equation}
In this case, the optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:softmaxreg}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \vec{q}_{ij}^\top\ensuremath{\vec{x}'} + c_{ij} < 0 \quad \forall\,j \in \set{Y}, j\neq i=\ensuremath{y'}
\end{split}
\end{equation}
where
\begin{equation}
\vec{q}_{ij} = \vec{w}_j - \vec{w}_i
\end{equation}
\begin{equation}
c_{ij} = b_j - b_i
\end{equation}
Depending on the regularization, the optimization problem Eq.~\ref{eq:cf:softmaxreg} becomes either a LP - if the weighted Manhattan distance is used - or a convex QP with linear constraints - if the Euclidean distance is used. More information can be found in the appendix (section~\ref{sec:appendix:softmaxreg}).
\subsubsection{Linear regression}\label{sec:lr}
In linear regression we model the distribution of the response variable as a Gaussian distribution. The prediction function $\ensuremath{f}$ of a linear regression model is given as:
\begin{equation}
\ensuremath{f}(\ensuremath{\vec{x}}) = \vec{w}^\top\ensuremath{\vec{x}} + b
\end{equation}
The optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:lr}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \vec{w}^\top\ensuremath{\vec{x}'} + c \leq \epsilon \\
& -\vec{w}^\top\ensuremath{\vec{x}'} - c \leq \epsilon
\end{split}
\end{equation}
where
\begin{equation}
c = b - \ensuremath{y'}
\end{equation}
and $\epsilon \geq 0$ denotes the tolerated deviation from the requested prediction $\ensuremath{y'}$.
Depending on the regularization, the optimization problem Eq.~\ref{eq:cf:lr} becomes either a LP (if the weighted Manhattan distance is used) or a convex QP with linear constraints (if the Euclidean distance is used). More information can be found in the appendix (section~\ref{sec:appendix:lr}).
\subsubsection{Poisson regression}\label{sec:poissonreg}
In Poisson regression we model the distribution of the response variable as a Poisson distribution. The prediction function $\ensuremath{f}$ of a Poisson regression model is given as:
\begin{equation}
\ensuremath{f}(\ensuremath{\vec{x}}) = \exp(\vec{w}^\top\ensuremath{\vec{x}} + b)
\end{equation}
In this case, the optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:poissonreg}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \vec{w}^\top\ensuremath{\vec{x}'} + c \leq \epsilon \\
& -\vec{w}^\top\ensuremath{\vec{x}'} - c \leq \epsilon
\end{split}
\end{equation}
where
\begin{equation}
c = b - \log(\ensuremath{y'})
\end{equation}
and $\epsilon \geq 0$ denotes the tolerated deviation from the requested prediction $\ensuremath{y'}$.
Depending on the regularization, the optimization problem Eq.~\ref{eq:cf:poissonreg} becomes either a LP (if the weighted Manhattan distance is used) or a convex QP with linear constraints (if the Euclidean distance is used). More information can be found in the appendix (section~\ref{sec:appendix:poissonreg}).
\subsubsection{Exponential regression}\label{sec:expreg}
In exponential regression we model the distribution of the response variable as a exponential distribution. The prediction function $\ensuremath{f}$ of an exponential regression model is given as:
\begin{equation}
\ensuremath{f}(\ensuremath{\vec{x}}) = -\frac{1}{\vec{w}^\top\ensuremath{\vec{x}} + b}
\end{equation}
Then, the optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:expreg}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \vec{w}^\top\ensuremath{\vec{x}'} + c \leq \epsilon \\
& -\vec{w}^\top\ensuremath{\vec{x}'} - c \leq \epsilon
\end{split}
\end{equation}
where
\begin{equation}
c = b + \frac{1}{\ensuremath{y'}}
\end{equation}
and $\epsilon \geq 0$ denotes the tolerated deviation from the requested prediction $\ensuremath{y'}$.
Depending on the regularization, the optimization problem Eq.~\ref{eq:cf:expreg} becomes either a LP (if the weighted Manhattan distance is used) or a convex QP with linear constraints (if the Euclidean distance is used). More information can be found in the appendix (section~\ref{sec:appendix:expreg}).
\subsection{Gaussian naive Bayes}\label{sec:gaussnb}
The Gaussian naive Bayes model makes the assumption that all features are independent of each other and follow a normal distribution. The prediction function $\ensuremath{h}$ of a Gaussian naive Bayes model is given as:
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\prod_{k=1}^d \N(\ensuremath{\vec{x}} \mid \mu_{ik}, \sigma^2_{ik})\pi_i
\end{equation}
where $\pi_i$ denotes the a-priori probability of the $i$-th class.
The optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:gnb}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \ensuremath{\vec{x}'}^\top\mat{A}_{ij}\ensuremath{\vec{x}'} + \vec{q}_{ij}^\top\ensuremath{\vec{x}'} + c_{ij} < 0 \quad\forall\,j\in\set{Y},j\neq i=\ensuremath{y'}
\end{split}
\end{equation}
where
\begin{equation}
\mat{A}_{ij} = \diag\left(\frac{1}{2\sigma_{ik}^2} - \frac{1}{2\sigma_{jk}^2}\right)
\end{equation}
\begin{equation}
\vec{q}_{ij} = \left(\frac{\mu_{j1}}{\sigma_{j1}^2} - \frac{\mu_{i1}}{\sigma_{i1}^2}, \dots, \frac{\mu_{jd}}{\sigma_{jd}^2} - \frac{\mu_{id}}{\sigma_{id}^2}\right)^\top
\end{equation}
\begin{equation}
c_{ij} = \log\left(\frac{\pi_j}{\pi_i}\right) + \sum_{k=1}^d \log\left(\frac{\sqrt{2\pi\sigma_{ik}^2}}{\sqrt{2\pi\sigma_{jk}^2}}\right) - \frac{\mu_{jk}^2}{2\sigma_{jk}^2} + \frac{\mu_{ik}^2}{2\sigma_{ik}^2}
\end{equation}
Because we can not make any statement about the definiteness of $\mat{A}_{ij}$, the quadratic constraints in Eq.~\ref{eq:cf:gnb} are non-convex. Therefore, the optimization problem Eq.~\ref{eq:cf:gnb} is a non-convex quadratically constrained quadratic program (QCQP).
We can approximately solve Eq.~\ref{eq:cf:gnb} by using an approximation method like the Suggest-Improve framework~\cite{park2017general}. Furthermore, if we have a binary classification problem, we can solve a semi-definite program (SDP) whose solution is equivalent to Eq.~\ref{eq:cf:gnb}. More details can be found in the appendix (sections~\ref{sec:appendix:gnb},\ref{sec:appendix:nonconvexqcqp} and~\ref{sec:appendix:qcqp1constraint}).
\subsection{Quadratic discriminant analysis}\label{sec:qda}
In quadratic discriminant analysis (QDA) we model each class distribution as an independent Gaussian distribution - note that in contrast to LDA each class distribution has its own covariance matrix. The prediction function $\ensuremath{h}$ of a QDA model is given as:
\begin{equation}
\ensuremath{h}(\vec{x}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\N(\vec{x} \mid \vec{\mu}_i, \mat{\Sigma}_i)\pi_i
\end{equation}
where $\pi_i$ denotes the a-priori probability of the $i$-th class.
In this case, the optimization problem for computing a counterfactual explanation Eq.~\ref{eq:counterfactualoptproblem} can be rewritten as:
\begin{equation}\label{eq:cf:qda}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.}\\
& \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{ij}\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + c_{ij} < 0 \quad\forall\,j\in\set{Y},j\neq i=\ensuremath{y'}
\end{split}
\end{equation}
where
\begin{equation}
\mat{A}_{ij} = \mat{\Sigma}_i^{-1} - \mat{\Sigma}_j^{-1}
\end{equation}
\begin{equation}
\vec{q}_{ij} = \mat{\Sigma}_j^{-1}\vec{\mu}_j - \mat{\Sigma}_i^{-1}\vec{\mu}_i
\end{equation}
\begin{equation}
c_{ij} = \frac{1}{2}\Big(\vec{\mu}_i^\top\mat{\Sigma}_i^{-1}\vec{\mu}_i - \vec{\mu}_j^\top\mat{\Sigma}_j^{-1}\vec{\mu}_j\Big) + \frac{1}{2}\log\left(\frac{\det(\mat{\Sigma}_i)}{\det(\mat{\Sigma}_j)}\right) + \log\left(\frac{\pi_j}{\pi_i}\right)
\end{equation}
Because we can not make any statement about the definiteness of $\mat{A}_{ij}$, the quadratic constraints in Eq.~\ref{eq:cf:qda} are non-convex. Thus, like in Gaussian naive Bayes (section~\ref{sec:gaussnb}), the optimization problem Eq.~\ref{eq:cf:qda} is a non-convex QCQP.
Like in the case of the previous non-convex QCQPs, we can approximately solve Eq.~\ref{eq:cf:qda} by using an approximation method. Furthermore, if we have a binary classification problem, we can solve a SDP whose solution is equivalent to Eq.~\ref{eq:cf:qda}. More details can be found in the appendix (sections~\ref{sec:appendix:qda},\ref{sec:appendix:nonconvexqcqp} and~\ref{sec:appendix:qcqp1constraint}).
\subsection{Learning vector quantization models}\label{sec:lvq}
Learning vector quantization (LVQ) models~\cite{lvqreview} compute a set of labeled prototypes $\{(\vec{\prototype}_i, \ensuremath{o}_i)\}$ from a given training data set - we refer to the $i$-th prototype as $\vec{\prototype}_i$ and the corresponding label as $\ensuremath{o}_i$. The prediction function $\ensuremath{h}$ of a LVQ model is given as:
\begin{equation}\label{eq:lvq_predict}
\begin{split}
& \ensuremath{h}(\ensuremath{\vec{x}}) = \ensuremath{o}_i \\
& \text{s.t. } \min\,\dist(\ensuremath{\vec{x}}, \vec{\prototype}_i)
\end{split}
\end{equation}
where $\dist()$ denotes a function for computing the distance between a data point and a prototype - usually this is the Euclidean distance:
\begin{equation}\label{eq:distfunction}
\dist(\ensuremath{\vec{x}}, \vec{\prototype}) = (\ensuremath{\vec{x}} - \vec{\prototype})^\top \mat{\mathbb{I}} (\ensuremath{\vec{x}} - \vec{\prototype})
\end{equation}
There exist LVQ models like (L)GMLVQ~\cite{gmlvq} and (L)MRSLVQ~\cite{mrslvq} that learn a custom (class or prototype specific) distance matrix $\distmat_{\prototype}$ that is used instead of the identity $\mat{\mathbb{I}}$ when computing the distance between a data point and a prototype. This gives rise to the generalized L2 distance:
\begin{equation}\label{eq:lvq_generaldist}
\dist(\ensuremath{\vec{x}}, \vec{\prototype}) = (\ensuremath{\vec{x}} - \vec{\prototype})^\top{\distmat}_{\prototype}(\ensuremath{\vec{x}} - \vec{\prototype})
\end{equation}
Because a LVQ model assigns the label of the nearest prototype to a given input, the nearest prototype of a counterfactual must be a prototype $\vec{\prototype}_i$ with $\ensuremath{o}_i=\ensuremath{y'}$. According to~\cite{counterfactualslvq}, for computing a counterfactual, it is sufficient to solve the following optimization problem for each prototype $\vec{\prototype}_i$ with $\ensuremath{o}_i=\ensuremath{y'}$ and select the counterfactual $\ensuremath{\vec{x}'}$ yielding the smallest value of $\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}})$:
\begin{equation}\label{eq:cflvq_general}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.} \\
& \dist(\ensuremath{\vec{x}'}, \vec{\prototype}_i) < \dist(\ensuremath{\vec{x}'}, \vec{\prototype}_j) \quad \forall\, \vec{\prototype}_j\in\set{P}(\ensuremath{y'})
\end{split}
\end{equation}
where $\set{P}(\ensuremath{y'})$ denotes the set of all prototypes \emph{not} labeled as $\ensuremath{y'}$. Note that the feasible region of Eq.~\ref{eq:cflvq_general} is always non-empty - the prototype $\vec{\prototype}_i$ is always a feasible solution.
In the subsequent sections we explore the type of constraints of Eq.~\ref{eq:cflvq_general} for different LVQ models.
\subsubsection{(Generalized matrix) LVQ}\label{sec:gmlvq}
In case of a (generalized matrix) LVQ model - all prototypes use the same distance matrix $\distmat$, the optimization problem Eq.~\ref{eq:cflvq_general} becomes~\cite{counterfactualslvq}:
\begin{equation}\label{eq:cf:gmlvq}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}})\\
& \text{s.t.} \\
& \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + c_{ij} < 0 \quad \forall\, \vec{\prototype}_j\in\set{P}(\ensuremath{y'})
\end{split}
\end{equation}
where
\begin{equation}
\vec{q}_{ij} = \frac{1}{2}\distmat(\vec{\prototype}_{j} - \vec{\prototype}_i)
\end{equation}
\begin{equation}
c_{ij} = \frac{1}{2}\left(\vec{\prototype}_{i}^\top\distmat\vec{\prototype}_{i} - \vec{\prototype}_{j}^\top\distmat\vec{\prototype}_{j}\right)
\end{equation}
Depending on the regularization, the optimization problem Eq.~\ref{eq:cf:gmlvq} becomes either a LP (if the Euclidean distance is used) or a convex QP with linear constraints (if the weighted Manhattan distance is used). More information can be found in the appendix (section~\ref{sec:appendix:lvq}).
\subsubsection{(Localized generalized matrix) LVQ}\label{sec:lgmlvq}
In case of a (localiced generalized matrix) LVQ model - there are different, class or prototype specific, distance matrices $\distmat_p$, the optimization problem Eq.~\ref{eq:cflvq_general} becomes~\cite{counterfactualslvq}:
\begin{equation}\label{eq:cf:lgmlvq}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\,\mathbb{R}^d}{\arg\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{ij}\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + c_{ij} < 0 \quad \forall\, \vec{\prototype}_j\in\set{P}(\ensuremath{y'})
\end{split}
\end{equation}
where
\begin{equation}
\mat{A}_{ij} = {\distmat}_{i} - {\distmat}_{j}
\end{equation}
\begin{equation}
\vec{q}_{ij} = \frac{1}{2}\left({\distmat}_j\vec{\prototype}_{j} - {\distmat}_i\vec{\prototype}_{i}\right)
\end{equation}
\begin{equation}
c_{ij} = \frac{1}{2}\left(\vec{\prototype}_{i}^\top{\distmat}_{i}\vec{\prototype}_{i} - \vec{\prototype}_{j}^\top{\distmat}_{j}\vec{\prototype}_{j}\right)
\end{equation}
Because we can not make any statement about the definiteness of $\mat{A}_{ij}$, the quadratic constraints in Eq.~\ref{eq:cf:lgmlvq} are non-convex. Thus, like in Gaussian naive Bayes (section~\ref{sec:gaussnb}) and QDA (section~\ref{sec:qda}), the optimization problem Eq.~\ref{eq:cf:lgmlvq} is a non-convex QCQP.
Like the previous non-convex QCQPs, we can approximately solve Eq.~\ref{eq:cf:lgmlvq} by using an approximation method. Furthermore, if we have a binary classification problem and each class is represented by a single prototype, we can solve a SDP whose solution is equivalent to Eq.~\ref{eq:cf:lgmlvq}. More details can be found in the appendix (sections~\ref{sec:appendix:lvq},\ref{sec:appendix:nonconvexqcqp} and~\ref{sec:appendix:qcqp1constraint}).
\subsection{Tree based models}\label{sec:treemodel}
Tree based models are very popular in data science because they often achieve a high predictive-accuracy~\cite{treebasedmodels}. In the subsequent sections we discuss how to compute counterfactual explanations of tree based models. In particular, we consider decision/regression trees and tree based ensembles like random forest models.
\subsubsection{Decision trees}\label{sec:decisiontree}
In case of decision/regression tree models, we can compute a counterfactual by enumerating all possible paths that lead to the requested prediction~\cite{decisiontreecounterfactual, interprettreeesembles}. However, it might happen that some requested predictions are not possible because all possible predictions of the tree are encoded in the leafs. In this case one might define an interval of acceptable predictions so that a counterfactual exists.
The procedure for computing a counterfactual of a decision/regression tree is described in Algorithm~\ref{algo:cf:tree}.
\begin{algorithm}[!htb]
\caption{Computing a counterfactual of a decision/regression tree}\label{algo:cf:tree}
\textbf{Input:} Original input $\ensuremath{\vec{x}}$, requested prediction $\ensuremath{y'}$ of the counterfactual, the tree model\\
\textbf{Output:} Counterfactual $\ensuremath{\vec{x}'}$
\begin{algorithmic}[1]
\State Enumerate all leafs with prediction $\ensuremath{y'}$
\State For each leaf, enumerate all paths reaching the leaf
\State For each path, compute the minimal change to $\ensuremath{\vec{x}}$ that yields the path
\State Sort all paths according to regularization of the change to $\ensuremath{\vec{x}}$
\State Select the path and the corresponding change to $\ensuremath{\vec{x}}$ that minimizes the regularization
\end{algorithmic}
\end{algorithm}
\subsubsection{Tree based ensembles}\label{sec:randomforest}
Popular instances of tree based enesmbles are random forest and gradient boosting regression trees. It turns out that the problem of computing a counterfactual explanation of such models is NP-hard~\cite{interprettreeesembles}.
The following heuristic for computing a counterfactual explanation of a random forest model was proposed in~\cite{ceml}: First, we compute a counterfactual of a model from the ensemble. Next, we use this counterfactual as a starting point for minimizing the number of trees that do not outpur the requested prediction by using a gradient-free optimization method like the Downhill-Simplex method. The idea behind this approach is that the counterfactual of a tree from the ensemble is close to the decision boundary of the ensemble so that computing a counterfactual of the ensemble becomes easier. By doing this for all trees in the ensemble, we get many counterfactuals and we can select the one that minimizes the regularization the most. This heuristic seems to work well in practice~\cite{ceml}.
Another approach for computing counterfactual explanations of an ensemble of trees was propsed in~\cite{interprettreeesembles} - although the authors do not call it counterfactuals, they actually compute counterfactuals. Their algorithm works as follows:
We iterate over all trees in the ensemble that do not yield the requested prediction. Next, we compute all possible counterfactuals of each of these trees (see section~\ref{sec:decisiontree}). If this counterfactual turns our to be counterfactual of the ensemble, we store it so that in the end we can select the counterfactual with the smallest deviation from the original input. However, it can not be guaranteed that a counterfactual of the ensemble is found because it might happen that by changing the data point so that it becomes a counterfactual of a particular tree, the prediction of other trees in the ensemble change as well. According to the authors, this algorithm/heuristic works well in practice. Unfortunately, the worst-case complexity is exponential in the number of features and thus it is not suitable for high dimensional data.
\section{Implementation}
The gradient-based and gradient free methods, as well as the model specific methods for tree based models are already implemented in CEML~\cite{ceml}. The implementation of the LVQ specific methods are provided by the authors of~\cite{counterfactualslvq}. The Python~\cite{van1995python} implementation of our proposed methods is available on GitHub\footnote{\url{https://github.com/andreArtelt/OnTheComputationOfCounterfactualExplanations}} and is based on the Python packages scikit-learn~\cite{scikit-learn}, numpy~\cite{numpy} and cvxpy~\cite{cvxpy}.
We plan to add these model-specific methods to CEML~\cite{ceml} in the near future.
\section{Conclusion}\label{sec:conclusion}
In this survey we extensively studied how to compute counterfactual explanations of many different ML models. We reviewed known methods from literature and proposed methods (mostly LPs and (QC)QPs) for computing counterfactuals of ML models that have not been considered in literature so far.
\section{Appendix}\label{sec:appendix}
\subsection{Relaxing strict inequalities}\label{sec:appendix:relaxingstrictinequlities}
When modeling the problem of computing counterfactuals, we often obtain strict inequalities like
\begin{equation}
g(\ensuremath{\vec{x}}) < 0
\end{equation}
Strict inequalities are not allowed in convex programming because the feasible region would become an open set. However, we could turn the $<$ into a $\leq$ by adding a small number to the left side of the inequality:
\begin{equation}
g(\ensuremath{\vec{x}}) + \epsilon \leq 0
\end{equation}
where $\epsilon > 0$ is a small number.
In practice, when implementing our methods, we found that we can often safely replace all $<$ by $\leq$ without changing anything else - this might be because of the numerics (like round-off errors) of fixed size floating-point numbers.
\subsection{Separating hyperplane}\label{sec:appendix:sephyperplane}
Recall that the prediction function $\ensuremath{h}$ is given as:
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \sign(\vec{w}^\top\ensuremath{\vec{x}} + b)
\end{equation}
If we multiply the projection $\vec{w}^\top\ensuremath{\vec{x}} + b$ by the requested prediction $y$\footnote{Note that we assume $\set{Y}=\{-1, 1\}$.}, the result is positive if and only if the classification $\ensuremath{h}(\ensuremath{\vec{x}})$ is equal to $y$. Therefore, the linear constraint for predicting class $y$ is given as
\begin{equation}
\begin{split}
& y\left(\vec{w}^\top\ensuremath{\vec{x}} + b\right) > 0 \\
& \Leftrightarrow \vec{q}^\top\ensuremath{\vec{x}} + c < 0
\end{split}
\end{equation}
where
\begin{equation}
\vec{q} = -y\vec{w}
\end{equation}
\begin{equation}
c = -b\ensuremath{y}
\end{equation}
\subsection{Generalized linear models}\label{sec:appendix:glm}
\subsubsection{Softmax regression}\label{sec:appendix:softmaxreg}
Recall that the prediction function $\ensuremath{h}$ is given as:
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\frac{\exp(\vec{w}_i^\top\ensuremath{\vec{x}} + b_i)}{\sum_k \exp(\vec{w}_k^\top\ensuremath{\vec{x}} + b_k)}
\end{equation}
Thus, the constraint for obtaining a specific prediction $\ensuremath{y'}$ is given as:
\begin{equation}\label{eq:softmaxreg:constraint}
\frac{\exp(\vec{w}_i^\top\ensuremath{\vec{x}} + b_i)}{\sum_k \exp(\vec{w}_k^\top\ensuremath{\vec{x}} + b_k)} > \frac{\exp(\vec{w}_j^\top\ensuremath{\vec{x}} + b_j)}{\sum_k \exp(\vec{w}_k^\top\ensuremath{\vec{x}} + b_k)} \quad \forall\,j\neq i=\ensuremath{y'}
\end{equation}
Holding $i$ and $j$ fixed, we can simplify Eq.~\ref{eq:softmaxreg:constraint}:
\begin{equation}
\begin{split}
\frac{\exp(\vec{w}_i^\top\ensuremath{\vec{x}} + b_i)}{\sum_k \exp(\vec{w}_k^\top\ensuremath{\vec{x}} + b_k)} &> \frac{\exp(\vec{w}_j^\top\ensuremath{\vec{x}} + b_j)}{\sum_k \exp(\vec{w}_k^\top\ensuremath{\vec{x}} + b_k)} \\
& \Leftrightarrow \\ \exp(\vec{w}_i^\top\ensuremath{\vec{x}} + b_i) &> \exp(\vec{w}_j^\top\ensuremath{\vec{x}} + b_j) \\
& \Leftrightarrow \\ \vec{w}_i^\top\ensuremath{\vec{x}} + b_i &> \vec{w}_j^\top\ensuremath{\vec{x}} + b_j \\
& \Leftrightarrow \\ \vec{q}_{ij}^\top\ensuremath{\vec{x}'} + c_{ij} &< 0
\end{split}
\end{equation}
where
\begin{equation}
\vec{q}_{ij} = \vec{w}_j - \vec{w}_i
\end{equation}
\begin{equation}
c_{ij} = b_j - b_i
\end{equation}
Therefore, we can rewrite Eq.~\ref{eq:softmaxreg:constraint} as a set of linear inequalities.
\subsubsection{Linear regression}\label{sec:appendix:lr}
Recall that the prediction function $\ensuremath{f}$ is given as:
\begin{equation}
\ensuremath{f}(\ensuremath{\vec{x}}) = \vec{w}^\top\ensuremath{\vec{x}} + b
\end{equation}
By introducing the parameter $\epsilon \geq 0$ that specifies the maximum tolerated deviation from the requested prediction - we set $\epsilon = 0$ if we do not allow any deviations - the constraint for obtaining the requested prediction $\ensuremath{y'}$ is given as
\begin{equation}\label{eq:lr:constraint}
\begin{split}
& |\ensuremath{f}(\ensuremath{\vec{x}'}) - \ensuremath{y'}| \leq \epsilon \\
& \Leftrightarrow |\vec{w}^\top\ensuremath{\vec{x}'} + b - \ensuremath{y'}| \leq \epsilon \\
& \Leftrightarrow |\vec{w}^\top\ensuremath{\vec{x}'} + c| \leq \epsilon
\end{split}
\end{equation}
where
\begin{equation}
c = b - \ensuremath{y'}
\end{equation}
Finally, we can rewrite Eq.~\ref{eq:lr:constraint} as two linear inequality constraints:
\begin{equation}
\begin{split}
& \vec{w}^\top\ensuremath{\vec{x}'} + c \leq \epsilon \\
& -\vec{w}^\top\ensuremath{\vec{x}'} - c \leq \epsilon
\end{split}
\end{equation}
\subsubsection{Poisson regression}\label{sec:appendix:poissonreg}
Recall that the prediction function $\ensuremath{f}$ is given as
\begin{equation}
\ensuremath{f}(\ensuremath{\vec{x}}) = \exp(\vec{w}^\top\ensuremath{\vec{x}} + b)
\end{equation}
The constraint for exactly obtaining the requested prediction $\ensuremath{y'}$ is
\begin{equation}
\begin{split}
& \ensuremath{f}(\ensuremath{\vec{x}'}) = \ensuremath{y'} \\
& \Leftrightarrow \exp(\vec{w}^\top\ensuremath{\vec{x}'} + b) = \ensuremath{y'} \\
& \Leftrightarrow \vec{w}^\top\ensuremath{\vec{x}'} + b - \log(\ensuremath{y'}) = 0 \\
& \Leftrightarrow \vec{w}^\top\ensuremath{\vec{x}'} + c = 0
\end{split}
\end{equation}
where
\begin{equation}
c = b - \log(\ensuremath{y'})
\end{equation}
Finally, we obtain the following set of linear inequality constraints:
\begin{equation}
\begin{split}
& \vec{w}^\top\ensuremath{\vec{x}'} + c \leq \epsilon \\
& -\vec{w}^\top\ensuremath{\vec{x}'} - c \leq \epsilon
\end{split}
\end{equation}
where we introduced the parameter $\epsilon \geq 0$ that specifies the maximum tolerated deviation from the requested prediction - we set $\epsilon = 0$ if we do not allow any deviations.
\subsubsection{Exponential regression}\label{sec:appendix:expreg}
Recall that the prediction function $\ensuremath{f}$ is given as:
\begin{equation}\label{eq:h:expreg}
\ensuremath{f}(\ensuremath{\vec{x}}) = -\frac{1}{\vec{w}^\top\ensuremath{\vec{x}} + b}
\end{equation}
The constraint for a specific prediction $\ensuremath{y'}$ is given as:
\begin{equation}\label{eq:expreg:constraint}
\begin{split}
& \ensuremath{f}(\ensuremath{\vec{x}'}) = \ensuremath{y'} \\
& \Leftrightarrow -\frac{1}{\vec{w}^\top\ensuremath{\vec{x}'} + b} = \ensuremath{y'} \\
& \Leftrightarrow \vec{w}^\top\ensuremath{\vec{x}'} + b + \frac{1}{\ensuremath{y'}} = 0 \\
& \Leftrightarrow \vec{w}^\top\ensuremath{\vec{x}'} + c = 0
\end{split}
\end{equation}
where
\begin{equation}
c = b + \frac{1}{\ensuremath{y'}}
\end{equation}
Finally, we obtain the following set of linear inequality constraints:
\begin{equation}
\begin{split}
& \vec{w}^\top\ensuremath{\vec{x}'} + c \leq \epsilon \\
& -\vec{w}^\top\ensuremath{\vec{x}'} - c \leq \epsilon
\end{split}
\end{equation}
where we introduced the parameter $\epsilon \geq 0$ that specifies the maximum tolerated deviation from the requested prediction - we set $\epsilon = 0$ if we do not allow any deviations.
\subsection{Gaussian naive bayes}\label{sec:appendix:gnb}
Recall that the prediction function $\ensuremath{h}$ is given as:
\begin{equation}\label{eq:h:gnb}
\ensuremath{h}(\ensuremath{\vec{x}}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\prod_{k=1}^d \N(\ensuremath{\vec{x}} \mid \mu_{ik}, \sigma^2_{ik})\pi_i
\end{equation}
We note that Eq.~\ref{eq:h:gnb} is equivalent to
\begin{equation}\label{eq:h:gnb2}
\ensuremath{h}(\ensuremath{\vec{x}}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\sum_{k=1}^d \log\left(\N(\ensuremath{\vec{x}} \mid \mu_{ik}, \sigma^2_{ik})\right) + \log(\pi_i)
\end{equation}
Simplifying the term in Eq.~\ref{eq:h:gnb2} yields
\begin{equation}
\begin{split}
\log(\pi_i) + \sum_{k=1}^d \log\left(\N(\ensuremath{\vec{x}} \mid \mu_{ik}, \sigma^2_{ik})\right) &= \log(\pi_i) + \sum_{k=1}^d \log\left(\frac{1}{\sqrt{2\pi\sigma_{ik}^2}}\right) +\\&\quad\quad \sum_{k=1}^d -\frac{1}{2\sigma_{ik}^2}\pnorm{(\ensuremath{\vec{x}})_k - \mu_{ik}}_2^2 \\
&= \log(\pi_i) + \sum_{k=1}^d \log\left(\frac{1}{\sqrt{2\pi\sigma_{ik}^2}}\right) -\\&\quad\quad \sum_{k=1}^d \frac{1}{2\sigma_{ik}^2}\Big((\ensuremath{\vec{x}})_k^2 + \mu_{ik}^2 - 2(\ensuremath{\vec{x}})_k\mu_{ik}\Big) \\
&= c_i - \ensuremath{\vec{x}}^\top\mat{A}_i\ensuremath{\vec{x}} + \vec{q}_i^\top\ensuremath{\vec{x}}
\end{split}
\end{equation}
where
\begin{equation}
c_i = \log(\pi_i) + \sum_{k=1}^d \log\left(\frac{1}{\sqrt{2\pi\sigma_{ik}^2}}\right) - \frac{\mu_{ik}^2}{2\sigma_{ik}^2}
\end{equation}
\begin{equation}
\mat{A}_i = \diag\left(\frac{1}{2\sigma_{ik}^2}\right)
\end{equation}
\begin{equation}
\vec{q}_i = \left(\frac{\mu_{i1}}{\sigma_{i1}^2}, \dots, \frac{\mu_{id}}{\sigma_{id}^2}\right)^\top
\end{equation}
For a sample $\ensuremath{\vec{x}}$, in order to be classified as the $i$-th class, the following set of strict inequalities must hold:
\begin{equation}\label{eq:gnb:inequalities}
\begin{split}
c_i - \ensuremath{\vec{x}}^\top\mat{A}_i\ensuremath{\vec{x}} + \vec{q}_i^\top\ensuremath{\vec{x}} > c_j - \ensuremath{\vec{x}}^\top\mat{A}_j\ensuremath{\vec{x}} + \vec{q}_j^\top\ensuremath{\vec{x}} \quad \forall\,j\neq i
\end{split}
\end{equation}
By rearranging terms in Eq.~\ref{eq:gnb:inequalities}, we get the final constraints
\begin{equation}\label{eq:gnb:finalconstraint}
\ensuremath{\vec{x}}^\top\mat{A}_{ij}\ensuremath{\vec{x}} + \vec{q}_{ij}^\top\ensuremath{\vec{x}} + c_{ij} < 0 \quad \forall\,j\neq i
\end{equation}
where
\begin{equation}
\mat{A}_{ij} = \mat{A}_i - \mat{A}_j = \diag\left(\frac{1}{2\sigma_{ik}^2} - \frac{1}{2\sigma_{jk}^2}\right)
\end{equation}
\begin{equation}
\vec{q}_{ij} = \vec{q}_j - \vec{q}_i = \left(\frac{\mu_{j1}}{\sigma_{j1}^2} - \frac{\mu_{i1}}{\sigma_{i1}^2}, \dots, \frac{\mu_{jd}}{\sigma_{jd}^2} - \frac{\mu_{id}}{\sigma_{id}^2}\right)^\top
\end{equation}
\begin{equation}
\begin{split}
c_{ij} &= c_j - c_i = \log(\pi_j) - \log(\pi_i) +\\&\quad\quad \sum_{k=1}^d \log\left(\frac{1}{\sqrt{2\pi\sigma_{jk}^2}}\right) - \frac{\mu_{jk}^2}{2\sigma_{jk}^2} - \log\left(\frac{1}{\sqrt{2\pi\sigma_{ik}^2}}\right) + \frac{\mu_{ik}^2}{2\sigma_{ik}^2} \\
&= \log\left(\frac{\pi_j}{\pi_i}\right) + \sum_{k=1}^d \log\left(\frac{\sqrt{2\pi\sigma_{ik}^2}}{\sqrt{2\pi\sigma_{jk}^2}}\right) - \frac{\mu_{jk}^2}{2\sigma_{jk}^2} + \frac{\mu_{ik}^2}{2\sigma_{ik}^2}
\end{split}
\end{equation}
Because we can not make any statement about the definiteness of the diagonal matrix $\mat{A}_{ij}$, the constraint Eq.~\ref{eq:gnb:finalconstraint} is a non-convex quadratic inequality constraint.
\subsection{Quadratic discriminant analysis}\label{sec:appendix:qda}
Recall that the prediction function $\ensuremath{h}$ is given as:
\begin{equation}\label{eq:h:qda}
\ensuremath{h}(\vec{x}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\N(\vec{x} \mid \vec{\mu}_i, \mat{\Sigma}_i)\pi_i
\end{equation}
We can rewrite Eq.~\ref{eq:h:qda} as
\begin{equation}
\ensuremath{h}(\ensuremath{\vec{x}}) = \underset{i \,\in\,\set{Y}}{\arg\max}\;\log\Big(\N(\ensuremath{\vec{x}} \mid \vec{\mu}_i, \mat{\Sigma}_i)\pi_i\Big)
\end{equation}
Working on the $\log$ term yields
\begin{equation}
\begin{split}
\log\Big(\N(\ensuremath{\vec{x}} \mid \vec{\mu}_i, \mat{\Sigma}_i)\pi_i\Big) &= \log\Big(\N(\ensuremath{\vec{x}} \mid \vec{\mu}_i, \mat{\Sigma}_i)\Big) + \log(\pi_i) \\
&= -\frac{d}{2}\log(2\pi) - \frac{1}{2}\log\big(\det(\mat{\Sigma}_i^{-1})\big) - \frac{1}{2}(\ensuremath{\vec{x}} - \vec{\mu}_i)^\top\mat{\Sigma}_i^{-1}(\ensuremath{\vec{x}} - \vec{\mu}_i)\\&\quad + \log(\pi_i) \\
&= -\frac{1}{2}\ensuremath{\vec{x}}^\top\mat{\Sigma}_i^{-1}\ensuremath{\vec{x}} + \ensuremath{\vec{x}}^\top\vec{q}_i + c_i
\end{split}
\end{equation}
where
\begin{equation}
\vec{q}_i = \mat{\Sigma}_i^{-1}\vec{\mu}_i
\end{equation}
\begin{equation}
c_i = -\frac{d}{2}\log(2\pi) - \frac{1}{2}\log\big(\det(\mat{\Sigma}_i)\big) -\frac{1}{2}\vec{\mu}_i^\top\mat{\Sigma}_i^{-1}\vec{\mu}_i + \log(\pi_i)
\end{equation}
For a sample $\ensuremath{\vec{x}}$, in order to be classified as the $i$-th class, the following set of strict inequalities must hold:
\begin{equation}\label{eq:qda:constraint}
-\frac{1}{2}\ensuremath{\vec{x}}^\top\mat{\Sigma}_i^{-1}\ensuremath{\vec{x}} + \ensuremath{\vec{x}}^\top\vec{q}_i + c_i > -\frac{1}{2}\ensuremath{\vec{x}}^\top\mat{\Sigma}_j^{-1}\ensuremath{\vec{x}} + \ensuremath{\vec{x}}^\top\vec{q}_j + c_j \quad \forall\,j\neq i
\end{equation}
Rearranging Eq.~\ref{eq:qda:constraint} yields
\begin{equation}\label{eq:qda:finalconstraint}
\frac{1}{2}\ensuremath{\vec{x}}^\top\mat{A}_{ij}\ensuremath{\vec{x}} + \ensuremath{\vec{x}}^\top\vec{q}_{ij} + c_{ij} < 0 \quad \forall\,j\neq i
\end{equation}
where
\begin{equation}
\mat{A}_{ij} = \mat{\Sigma}_i^{-1} - \mat{\Sigma}_j^{-1}
\end{equation}
\begin{equation}
\vec{q}_{ij} = \vec{q}_j - \vec{q}_i = \mat{\Sigma}_j^{-1}\vec{\mu}_j - \mat{\Sigma}_i^{-1}\vec{\mu}_i
\end{equation}
\begin{equation}
\begin{split}
c_{ij} &= c_j - c_i \\
&= -\frac{d}{2}\log(2\pi) - \frac{1}{2}\log\big(\det(\mat{\Sigma}_j)\big) -\frac{1}{2}\vec{\mu}_j^\top\mat{\Sigma}_j^{-1}\vec{\mu}_j + \log(\pi_j) +\\&\quad \frac{d}{2}\log(2\pi) + \frac{1}{2}\log\big(\det(\mat{\Sigma}_i)\big) + \frac{1}{2}\vec{\mu}_i^\top\mat{\Sigma}_i^{-1}\vec{\mu}_i - \log(\pi_i) \\
&= \frac{1}{2}\Big(\vec{\mu}_i^\top\mat{\Sigma}_i^{-1}\vec{\mu}_i - \vec{\mu}_j^\top\mat{\Sigma}_j^{-1}\vec{\mu}_j\Big) + \frac{1}{2}\log\left(\frac{\det(\mat{\Sigma}_i)}{\det(\mat{\Sigma}_j)}\right) + \log\left(\frac{\pi_j}{\pi_i}\right)
\end{split}
\end{equation}
The final constraint Eq.~\ref{eq:qda:finalconstraint} is a non-convex quadratic constraint because we can not make any statement about the definiteness of $\mat{A}_{ij}$.
\subsection{Learning vector quantization}\label{sec:appendix:lvq}
Note: The subsequent sections are taken from~\cite{counterfactualslvq}.
\subsubsection{Enforcing a specific prototype as the nearest neighbor}
By using the following set of inequalities, we can force the prototype $\vec{\prototype}_i$ to be the nearest neighbor of the counterfactual $\ensuremath{\vec{x}'}$ - which would cause $\ensuremath{\vec{x}'}$ to be classified as $\ensuremath{o}_i$:
\begin{equation}
\dist(\ensuremath{\vec{x}'}, \vec{\prototype}_i) < \dist(\ensuremath{\vec{x}'}, \vec{\prototype}_j) \quad \forall\, \vec{\prototype}_j\in\set{P}(\ensuremath{y'})
\end{equation}
We consider a fixed pair of $i$ and $j$:
\begin{equation}\label{eq:nearestprototypeconstraint}
\begin{split}
& \dist(\ensuremath{\vec{x}'}, \vec{\prototype}_i) < \dist(\ensuremath{\vec{x}'}, \vec{\prototype}_j) \\
& \Leftrightarrow \pnorm{\ensuremath{\vec{x}'} - \vec{\prototype}_i}_{\distmat_i}^2 < \pnorm{\ensuremath{\vec{x}'} - \vec{\prototype}_j}_{\distmat_j}^2 \\
& \Leftrightarrow (\ensuremath{\vec{x}'} - \vec{\prototype}_i)^\top{\distmat}_i(\ensuremath{\vec{x}'} - \vec{\prototype}_i) < (\ensuremath{\vec{x}'} - \vec{\prototype}_j)^\top{\distmat}_j(\ensuremath{\vec{x}'} - \vec{\prototype}_j) \\
& \Leftrightarrow \ensuremath{\vec{x}'}^\top{\distmat}_i\ensuremath{\vec{x}'} - 2\ensuremath{\vec{x}'}^\top{\distmat}_i\vec{\prototype}_i + \vec{\prototype}_i^\top{\distmat}_i\vec{\prototype}_i < \ensuremath{\vec{x}'}^\top{\distmat}_j\ensuremath{\vec{x}'} - 2\ensuremath{\vec{x}'}^\top{\distmat}_j\vec{\prototype}_j + \vec{\prototype}_j^\top{\distmat}_i\vec{\prototype}_j \\
& \Leftrightarrow \ensuremath{\vec{x}'}^\top{\distmat}_i\ensuremath{\vec{x}'} - \ensuremath{\vec{x}'}^\top{\distmat}_j\ensuremath{\vec{x}'} - 2\ensuremath{\vec{x}'}^\top{\distmat}_i\vec{\prototype}_i + 2\ensuremath{\vec{x}'}^\top{\distmat}_j\vec{\prototype}_j + \vec{\prototype}_i^\top{\distmat}_i\vec{\prototype}_i - \vec{\prototype}_j^\top{\distmat}_i\vec{\prototype}_j < 0 \\
& \Leftrightarrow \ensuremath{\vec{x}'}^\top({\distmat}_i - {\distmat}_j)\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top(-2{\distmat}_i\vec{\prototype}_i + 2{\distmat}_j\vec{\prototype}_j) + (\vec{\prototype}_i^\top{\distmat}_i\vec{\prototype}_i - \vec{\prototype}_j^\top{\distmat}_i\vec{\prototype}_j) \\
& \Leftrightarrow \frac{1}{2}\ensuremath{\vec{x}'}^\top({\distmat}_i - {\distmat}_j)\ensuremath{\vec{x}'} + \frac{1}{2}\ensuremath{\vec{x}'}^\top({\distmat}_j\vec{\prototype}_j - {\distmat}_i\vec{\prototype}_i) + \frac{1}{2}(\vec{\prototype}_i^\top{\distmat}_i\vec{\prototype}_i - \vec{\prototype}_j^\top{\distmat}_i\vec{\prototype}_j) < 0 \\
& \Leftrightarrow \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{ij}\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + c_{ij} < 0
\end{split}
\end{equation}
where
\begin{equation}
\mat{A}_{ij} = {\distmat}_{i} - {\distmat}_{j}
\end{equation}
\begin{equation}
\vec{q}_{ij} = \frac{1}{2}\left({\distmat}_j\vec{\prototype}_{j} - {\distmat}_i\vec{\prototype}_{i}\right)
\end{equation}
\begin{equation}
c_{ij} = \frac{1}{2}\left(\vec{\prototype}_{i}\top{\distmat}_{i}\vec{\prototype}_{i} - \vec{\prototype}_{j}\top{\distmat}_{j}\vec{\prototype}_{j}\right)
\end{equation}
\mbox{}\\
If we only have one global distance matrix $\distmat$, we find that $\mat{A}_{ij}=\mat{0}$ and the inequality Eq.~\ref{eq:nearestprototypeconstraint} simplifies:
\begin{equation}
\begin{split}
& \dist(\ensuremath{\vec{x}}, \vec{\prototype}_i) < \dist(\ensuremath{\vec{x}}, \vec{\prototype}_j) \\
& \Leftrightarrow \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + c_{ij} < 0
\end{split}
\end{equation}
where
\begin{equation}
\vec{q}_{ij} = \frac{1}{2}{\distmat}\left(\vec{\prototype}_{j} - \vec{\prototype}_{i}\right)
\end{equation}
\begin{equation}
c_{ij} = \frac{1}{2}\left(\vec{\prototype}_{i}^\top\distmat\vec{\prototype}_{i} - \vec{\prototype}_{j}^\top\distmat\vec{\prototype}_{j}\right)
\end{equation}
\mbox{}\\
If we do not use a custom distance matrix, we have $\distmat=\mat{\mathbb{I}}$ and Eq.~\ref{eq:nearestprototypeconstraint} becomes:
\begin{equation}
\begin{split}
& \dist(\ensuremath{\vec{x}}, \vec{\prototype}_i) < \dist(\ensuremath{\vec{x}}, \vec{\prototype}_j) \\
& \Leftrightarrow \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + c_{ij} < 0
\end{split}
\end{equation}
where
\begin{equation}
\vec{q}_{ij} = \frac{1}{2}\left(\vec{\prototype}_{j} - \vec{\prototype}_{i}\right)
\end{equation}
\begin{equation}
c_{ij} = \frac{1}{2}\left(\vec{\prototype}_{i}^\top\vec{\prototype}_{i} - \vec{\prototype}_{j}^\top\vec{\prototype}_{j}\right)
\end{equation}
\subsection{Minimizing the Euclidean distance}\label{sec:appendix:mineuclideandistance}
Minimizing the Euclidean distance (Eq.~\ref{eq:lvq_generaldist}) yields a \textit{quadratic} objective.
First, we expand the Euclidean distance (Eq.~\ref{eq:lvq_generaldist}):
\begin{equation}
\begin{split}
\pnorm{\ensuremath{\vec{x}'} - \ensuremath{\vec{x}}}_2^2 &= (\ensuremath{\vec{x}'} - \ensuremath{\vec{x}})^\top(\ensuremath{\vec{x}'} - \ensuremath{\vec{x}}) \\
&= \ensuremath{\vec{x}'}^\top\ensuremath{\vec{x}'} - \ensuremath{\vec{x}'}^\top\ensuremath{\vec{x}} - \ensuremath{\vec{x}}^\top\ensuremath{\vec{x}'} + \ensuremath{\vec{x}}^\top\ensuremath{\vec{x}} \\
&= \ensuremath{\vec{x}'}^\top\ensuremath{\vec{x}'} - 2\ensuremath{\vec{x}}^\top\ensuremath{\vec{x}'} + \ensuremath{\vec{x}}^\top\ensuremath{\vec{x}} \\
\end{split}
\end{equation}
Next, we note that that we can drop the constant $\ensuremath{\vec{x}}^\top\ensuremath{\vec{x}}$ when optimizing with respect to $\ensuremath{\vec{x}'}$:
\begin{equation}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}{\min}\; \pnorm{\ensuremath{\vec{x}'} - \ensuremath{\vec{x}}}_2^2 \\
& \Leftrightarrow \\
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}{\min}\; \frac{1}{2}\ensuremath{\vec{x}'}^\top\ensuremath{\vec{x}'} - \ensuremath{\vec{x}}^\top\ensuremath{\vec{x}'}
\end{split}
\end{equation}
\subsection{Minimizing the weighted Manhattan distance}\label{sec:appendix:minmanhattandistance}
Minimizing the weighted Manhattan distance (Eq.~\ref{eq:weighted_l1}) yields a \textit{linear} objective.
First, we transform the problem of minimizing the weighted Manhattan distance (Eq.~\ref{eq:weighted_l1}) into epigraph form:
\begin{equation}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}{\min}\; \sum_j \alpha_j \cdot |(\ensuremath{\vec{x}'})_j - (\ensuremath{\vec{x}})_j| \\
& \Leftrightarrow \\
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d, \beta\,\in\, \mathbb{R}}{\min}\; \beta \\
& \quad \text{s.t.} \quad \sum_j \alpha_j \cdot |(\ensuremath{\vec{x}'})_j - (\ensuremath{\vec{x}})_j| \leq \beta \\
& \quad \quad \quad \beta \geq 0
\end{split}
\end{equation}
Next, we separate the dimensions:
\begin{equation}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d, \beta\,\in\, \mathbb{R}}{\min}\; \beta \\
& \quad \text{s.t.} \quad \sum_j \alpha_j \cdot |(\ensuremath{\vec{x}'})_j - (\ensuremath{\vec{x}})_j| \leq \beta \\
& \quad \quad \quad \beta \geq 0 \\
& \Leftrightarrow \\
& \underset{\ensuremath{\vec{x}'}, \vec{\beta} \,\in\, \mathbb{R}^d}{\min}\; \sum_j (\vec{\beta})_j \\
& \quad \text{s.t.} \quad \alpha_j \cdot |(\ensuremath{\vec{x}'})_j - (\ensuremath{\vec{x}})_j| \leq (\vec{\beta})_j \quad \forall\,j \\
& \quad \quad \quad (\vec{\beta})_j \geq 0 \quad \forall\,j
\end{split}
\end{equation}
After that, we remove the absolute value function:
\begin{equation}
\begin{split}
& \underset{\ensuremath{\vec{x}'}, \vec{\beta} \,\in\, \mathbb{R}^d}{\min}\; \sum_j (\vec{\beta})_j \\
& \quad \text{s.t.} \quad \alpha_j \cdot |(\ensuremath{\vec{x}'})_j - (\ensuremath{\vec{x}})_j| \leq (\vec{\beta})_j \quad \forall\,j \\
& \quad \quad \quad (\vec{\beta})_j \geq 0 \quad \forall\,j \\
& \Leftrightarrow \\
& \underset{\ensuremath{\vec{x}'}, \vec{\beta} \,\in\, \mathbb{R}^d}{\min}\; \sum_j (\vec{\beta})_j \\
& \quad \text{s.t.} \quad \alpha_j (\ensuremath{\vec{x}'})_j - \alpha_j (\ensuremath{\vec{x}})_j \leq (\vec{\beta})_j \quad \forall\,j \\
& \quad \quad \quad -\alpha_j (\ensuremath{\vec{x}'})_j + \alpha_j (\ensuremath{\vec{x}})_j \leq (\vec{\beta})_j \quad \forall\,j \\
& \quad \quad \quad (\vec{\beta})_j \geq 0 \quad \forall\,j
\end{split}
\end{equation}
Finally, we rewrite everything in matrix-vector notation:
\begin{equation}
\begin{split}
& \underset{\ensuremath{\vec{x}'}, \vec{\beta} \,\in\, \mathbb{R}^d}{\min}\;\vec{1}^\top\vec{\beta} \\
& \text{s.t.} \\
& \mat{\Upsilon}\ensuremath{\vec{x}'} - \mat{\Upsilon}\ensuremath{\vec{x}} \leq \vec{\beta} \\
& -\mat{\Upsilon}\ensuremath{\vec{x}'} + \mat{\Upsilon}\ensuremath{\vec{x}} \leq \vec{\beta} \\
& \vec{\beta} \geq \vec{0}
\end{split}
\end{equation}
where
\begin{equation}
\mat{\Upsilon} = \diag(\alpha_j)
\end{equation}
\subsection{Solving a non-convex QCQP}
Solving a non-convex QCQP is known to be NP-hard~\cite{Boyd2004,park2017general}.
In section~\ref{sec:appendix:nonconvexqcqp} we discuss a method for approximately solving a non-convex QCQP and in section~\ref{sec:appendix:qcqp1constraint} we describe how to solve the special case of a non-convex QCQP having a single constraint.
\subsubsection{Approximately solving a non-convex QCQP}\label{sec:appendix:nonconvexqcqp}
Recall the non-convex quadratic constraint:
\begin{equation}\label{eq:nonconvexquadratic}
\frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{ij}\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + r_{ij} \leq 0
\end{equation}
In this paper, we always defined the matrix $\mat{A}_{ij}$ as the difference of two s.p.s.d. matrices $\mat{A}_{i}$ and $\mat{A}_{j}$:
\begin{equation}\label{eq:matrixdiff}
\mat{A}_{ij} = \mat{A}_{i} - \mat{A}_{j}
\end{equation}
By making use of Eq.~\ref{eq:matrixdiff}, we can rewrite Eq.~\ref{eq:nonconvexquadratic} as:
\begin{equation}
\begin{split}
& \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{i}\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + r_{ij} - \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{j}\ensuremath{\vec{x}'} \leq 0 \\
& \Leftrightarrow f(\ensuremath{\vec{x}'}) - g(\ensuremath{\vec{x}'}) \leq 0
\end{split}
\end{equation}
where
\begin{equation}
f(\ensuremath{\vec{x}'}) = \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{i}\ensuremath{\vec{x}'} + \ensuremath{\vec{x}'}^\top\vec{q}_{ij} + r_{ij}
\end{equation}
\begin{equation}
g(\ensuremath{\vec{x}'}) = \frac{1}{2}\ensuremath{\vec{x}'}^\top\mat{A}_{j}\ensuremath{\vec{x}'}
\end{equation}
Under the assumption that our regularization function $\regularization()$ is a convex function\footnote{The weighted Manhattan distance and the Euclidean distance are convex functions!}, we can rewrite a generic version of the non-convex QCQP Eq.~\ref{eq:cf:lgmlvq} as follows:
\begin{equation}\label{eq:dcp}
\begin{split}
& \underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}{\min}\;\regularization(\ensuremath{\vec{x}'}, \ensuremath{\vec{x}}) \\
& \text{s.t.} \\
& f(\ensuremath{\vec{x}'}) - g(\ensuremath{\vec{x}'}) \leq 0
\end{split}
\end{equation}
Because $\mat{A}_{i}$ and $\mat{A}_{j}$ are s.p.s.d. matrices, we know that $f(\ensuremath{\vec{x}'})$ and $g(\ensuremath{\vec{x}'})$ are convex functions. Therefore, Eq.~\ref{eq:dcp} is a difference-of-convex program (DCP).
This allows us to use the penalty convex-concave procedure (CCP)~\cite{park2017general} for computing an approximate solution of Eq.~\ref{eq:dcp}, yielding an approximate solution of the original non-convex QCQP. For using the penalty CCP, we need the first order Taylor approximation of $g(\ensuremath{\vec{x}'})$ around a current point $\ensuremath{\vec{x}}_k$:
\begin{equation}
\begin{split}
\hat{g}(\ensuremath{\vec{x}'})_{\ensuremath{\vec{x}}_k} &= g(\ensuremath{\vec{x}}_k) + (\nabla_{\ensuremath{\vec{x}'}}g)(\ensuremath{\vec{x}}_k)^\top(\ensuremath{\vec{x}'} - \ensuremath{\vec{x}}_k) \\
&= \frac{1}{2}\ensuremath{\vec{x}}_k^\top\mat{A}_{j}\ensuremath{\vec{x}}_k + (\mat{A}_{j}\ensuremath{\vec{x}}_k)^\top(\ensuremath{\vec{x}'} - \ensuremath{\vec{x}}_k) \\
&= (\mat{A}_{j}\ensuremath{\vec{x}}_k)^\top\ensuremath{\vec{x}'} + \frac{1}{2}\ensuremath{\vec{x}}_k^\top\mat{A}_{j}\ensuremath{\vec{x}}_k -(\mat{A}_{j}\ensuremath{\vec{x}}_k)^\top\ensuremath{\vec{x}}_k \\
&= \vec{\rho}_{jk}^\top\ensuremath{\vec{x}'} + \tilde{c}_{jk}
\end{split}
\end{equation}
where
\begin{equation}
\vec{\rho}_{jk} = \mat{A}_{j}\ensuremath{\vec{x}}_k
\end{equation}
\begin{equation}
\tilde{c}_{jk} = -\frac{1}{2}\ensuremath{\vec{x}}_k^\top\mat{A}_{j}\ensuremath{\vec{x}}_k
\end{equation}
In order to run the convex-concave procedure, we have to provide an initial (feasible) solution. We could either use the original data point as an initial \textit{infeasible} solution, some data point yielding the requested prediction as an initial \textit{feasible} solution or some other ``smart'' initialization.
As an alternative, we could use other methods for approximately computing a solution of the non-convex QCQP like the Suggest-Improve framework~\cite{park2017general} - actually, the methods we described in the previous paragraph is an instance of the Suggest-Improve framework.
\subsubsection{Solving a non-convex QCQP with just one constraint}\label{sec:appendix:qcqp1constraint}
We consider the general QCQP
\begin{equation}\label{eq:qcqp1constraint}
\begin{split}
&\underset{\ensuremath{\vec{x}'} \,\in\, \mathbb{R}^d}\min\;\ensuremath{\vec{x}'}^\top\mat{Q}\ensuremath{\vec{x}'} + \vec{q}^\top\ensuremath{\vec{x}'} + c \\
& \text{s.t.} \\
& \ensuremath{\vec{x}'}^\top\mat{A}\ensuremath{\vec{x}'} + \vec{b}^\top\ensuremath{\vec{x}'} + r \leq 0
\end{split}
\end{equation}
where $\mat{Q}, \mat{A} \in \mathbb{R}^{d \times d}$, $\vec{q}, \vec{b} \in \mathbb{R}^d$ and $c, r \in \mathbb{R}$.
If $\mat{Q}$ and $\mat{A}$ are not symmetric positive semi-definite, Eq.~\ref{eq:qcqp1constraint} is a non-convex QCQP. However, despite the non-convexity, we can solve Eq.~\ref{eq:qcqp1constraint} efficiently by solving the dual of Eq.~\ref{eq:qcqp1constraint} and observing that the duality gap is zero~\cite{Boyd2004} - under the assumption that Eq.~\ref{eq:qcqp1constraint} is strictly feasible\footnote{We can always achieve strict feasibility by moving a non-strict feasible point away from the decision boundary.}.
Therefore, solving Eq.~\ref{eq:qcqp1constraint} is equivalent to solving the following semi-definite program (SDP)~\cite{Boyd2004}:
\begin{equation}
\begin{split}
& \underset{\mat{X} \,\in\,\set{S}^d, \ensuremath{\vec{x}'}\,\in\,\mathbb{R}^d}{\arg\min}\;\trace(\mat{Q}\mat{X}) + \vec{q}^\top\ensuremath{\vec{x}'} + c \\
& \text{s.t.}\\
& \trace\left(\mat{A}\mat{X}\right) + \vec{b}^\top\ensuremath{\vec{x}'} + r \leq 0 \\
& \begin{pmatrix} \mat{X} & \ensuremath{\vec{x}'} \\ \ensuremath{\vec{x}'}^\top & 1 \end{pmatrix} \succeq 0
\end{split}
\end{equation}
where we introduced an additional variable\footnote{$\set{S}^d$ denotes the set of symmetric $\mathbb{R}^{d \times d}$ matrices} $\mat{X}$ that can be discarded afterwards.
\begin{footnotesize}
\bibliographystyle{unsrt}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 8,365
|
Kerm () may refer to:
Farsi word for worm
See also
KERM, a radio station
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 6,181
|
{"url":"http:\/\/math-related.blogspot.com\/2007\/06\/texmaker.html","text":"## Wednesday, June 27, 2007\n\n### Texmaker\n\nTexmaker is a free LaTeX editor, that integrates many tools needed to develop documents with LaTeX, in just one application.","date":"2018-07-18 16:33:47","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\": 0.9476929903030396, \"perplexity\": 8863.924283152981}, \"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-2018-30\/segments\/1531676590295.61\/warc\/CC-MAIN-20180718154631-20180718174631-00092.warc.gz\"}"}
| null | null |
Q: set style.listStyleImage with a var I am setting listStyleImage with a for loop.
This works
li.style.listStyleImage = 'url("../images/video/bcg.png")';
This does not
var _image = "../images/video/bcg.png";
li.style.listStyleImage = 'url(_image)';
nor this
var _image = "../images/video/bcg.png";
li.style.listStyleImage = 'url("_image")';
Can anyone help with this? Thanks...
A: You should escape _image varaible:
li.style.listStyleImage = 'url(' + _image + ')';
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 144
|
REVOLUTIONARY
Amateur Theater
and the Soviet State,
Cornell University Press
Ithaca and London
For Bob and Nora
## Contents
1. Preface
2. Introduction
3. 1 The Revolution Loves the Theater
4. 2 Small Forms on Small Stages
5. 3 From "Club Plays" to the Classics
6. 4 tram: The Vanguard of Amateur Art
7. 5 Shock Workers on the Cultural Front
8. 6 Amateurs in the Spectacle State
9. Conclusion
10. Glossary
11. Bibliography
12. Index
## Preface
THIS BOOK began as a study of the Leningrad Theater of Working-Class Youth, known by its acronym TRAM. But as my work on this theater and its Moscow affiliate continued, I discovered that these popular youth stages were simply the most visible representatives of a much wider phenomenon of amateur theater that blossomed in the early Soviet period. My research then grew to include the amorphous network of impromptu stages that inspired, imitated, and eventually outlasted TRAM. Expanding this project beyond TRAM, which has its own archives and extensive secondary literature, made this book much harder to write. I hope that the end product has more to say about the place of theater and the amateur arts in the cultural transformation begun with the October Revolution.
At the early stages of this project I was taken under the wing of two remarkable experts on Russian theater, Vladislav Ivanov and Maria Ivanova. They introduced me to a new field and gave me invaluable bibliographic assistance. What began as a professional relationship ended as a friendship. Their book-lined apartment was a haven for me on my trips to Moscow. The Internet has made the work of writing a little less solitary. I was bolstered by good-natured criticism and cyber pep talks from Louise McReynolds, who tried to hem in my natural tendencies toward social history. Susan Larsen has been a wonderful reader, both on the 'Net and off. Evgeny Dobrenko aided me with his broad knowledge of mainstream Soviet culture. Susanna Lockwood Smith shared her insights into the world of amateur music. Alice Fahs and I shared many conversations on theater, popular culture, and writing, all of which helped me to refine my thoughts. The friendship and bibliographic assistance of Joan Ariel and Ellen Broidy were very important to me. Two theater lovers, Carroll Smith-Rosenberg and Alvia Golden, helped me see this project in a broader perspective. I am particularly grateful to Alvia for thinking up the title.
Many colleagues have given me comments on individual chapters, including Jim von Geldern, Lewis Siegelbaum, and Anne Walthall. A convivial group of Russian scholars in southern California, including Robert Edelman, Choi Chatterjee, Arch Getty, Georg Michels, Elise Wirtschafter, and Mary Zirin, read versions of assorted chapters. Laurie Bernstein saw me through a research trip to Moscow. Stan Karas smoothed my way through Soviet newspapers of the 1930s. Viktorina Lefebvre helped me to compile the bibliography and the notes. My warmest thanks go to John Ackerman of Cornell University Press. I received research support for this project from the International Research Exchange Board (irex), the Social Science Research Council, and the University of California, Irvine.
It is customary for authors to thank their families, but I believe that I have special reasons to do so. Robert Moeller has always been my first and final reader. I am grateful for his erudition, patience, and most of all his sense of humor. And my daughter Nora deserves a special mention because she continues to teach me about the passion and commitment that go into amateur theater.
Portions of chapter 4 were previously published in "The Rise and Fall of the Soviet Youth Theater TRAM," Slavic Review 51, no. 3 (Fall 1992) and "Performing the New Woman: The Komsomolka as Actress and Image in Soviet Youth Theater," Journal of Social History 30, no. 1 (Fall 1996). I gratefully acknowledge the permission of the American Association for the Advancement of Slavic Studies and the Journal of Social History to reprint them here. Chapter 5 includes revised sections of two published articles. I thank the Ohio State University Press for allowing me to incorporate sections of my "Autonomous Theater and the Origins of Socialist Realism: The 1932 Olympiad of Autonomous Art," The Russian Review 52, no. 2 (April 1993). Copyright 1993 by Ohio State University Press. All rights reserved. The journal Russian History gave me permission to include parts of "Shock Workers on the Cultural Front: Agitprop Brigades in the First Five-Year Plan," Russian History 23, no. 1–4 (1996).
## Introduction
"THE YEAR since the last festival of the October Revolution will enter the history of Russian theater," proclaimed Adrian Piotrovskii, a prominent Leningrad scholar, cultural activist, and local bureaucrat, in 1924. "1t is the first year that the triumph of the mass movement known as amateur theater [samodeiatel'nyi teatr] has become apparent." Piotrovskii listed what he believed were amateur theater's significant accomplishments, including its influence over the most progressive professional stages. "Maybe this coming year will lead us to a long awaited, unified theatrical style, rooted in the 'amateur' performances of Soviet youth."1 For Piotrovskii, amateurs were the main source of creativity in Soviet theater and he was not alone in his convictions. Many observers expected a new, participatory socialist culture to emerge from the amateur stage.
This book examines amateur theaters as a distinctive medium of urban organization and entertainment in the first two decades of Soviet power. Amateur theaters served a variety of functions for their participants, audiences, and sponsoring institutions. They provided avenues of artistic and political self-expression for their mainly youthful actors. Housed in local gathering spots and workplaces, they offered a vital form of entertainment for urban neighborhoods. State agencies used them to disseminate political information and mark important celebrations. Finally, they constituted a popular forum to help shape a unified and widely accepted Soviet theatrical repertoire.
Focusing on the two capital cities, Moscow and Petrograd/Leningrad, I argue that the study of amateur theater allows us to trace crucial transformations in Soviet cultural life from the early revolutionary years to the late 1930s: the growing status and prestige of artistic experts; the articulation of a unified Soviet artistic canon; and efforts to control spontaneous forms of cultural expression by the lower classes. I aim to show that amateur theaters are an important—and understudied—form of cultural production and consumption in the Soviet Union.2 This work also adds to a lively conversation about how the population at large contributed to the formation of the Stalinist aesthetic doctrine of socialist realism.
Early Soviet amateur stages were extremely diverse. Some inhabited beautiful halls and offered elaborate training programs; others were fly-by-night operations. Some played home-made propaganda sketches; others staged Chekhov and Schiller. What united them was their community base and nonprofessional standing. Participants did not earn a living from their cultural work. The Russian language offers two common words for amateurism: liubitel'stvo, rooted in the verb to love (as amateur is rooted in the Latin amare); and samodeiatel'nost', literally translated as "doing-it-yourself."3 Although the words can be used interchangeably, Soviet writers increasingly distinguished between the two. Liubitel'skii teatr came to stand for all that was bad in amateur activities. Samodeiatel'nyi teatr, by contrast, represented all that was good in the Soviet approach, including collective interaction and productive social results.
As the Soviet state took shape, amateur theaters opened everywhere, created by soldiers, workers, peasants, students, and the unemployed. They were a path for participants to claim a public role. "The country had never been attacked by such violent theater fever as during the first years of the revolution," recalled the director and critic Pavel Markov. "Every district, every army unit, every factory had its own 'theater-circle,' watched over and developed with the greatest care and attention."4 By the late 1920s trade unions alone were supporting a national network of some twelve thousand amateur stages.5
Amateur theaters provided a venue where performers, the audience, and political overseers intersected. They were located primarily in clubs, initially impromptu gathering spots carved out of urban spaces. Soviet clubs offered participants a chance to meet their daily needs, gain valuable information about work opportunities and state regulations, and discover opportunities for relaxation and amusement. These multiple tasks—combining necessity with pleasure—quickly made them a focus of attention for those committed to creating a Soviet public sphere, a realm where private needs could be met in a shared collective space. Cultural activists called clubs a "public hearth" in a world where private hearths were linked to the old bourgeois past. They believed clubs had the power to nurture a sense of common purpose and common identity. In the words of the artist El Lissitzky, "The club ought to become a gathering place where the individual becomes one with the collective and where he stores up new reserves of energy."6
Clubs offered a wide range of amusements, but theatrical events of various kinds—classical plays, improvisations, recitations, and staged games—were among the most popular forms of participatory cultural activity in the early Soviet years.7 In the minds of club advocates, discovering the right kind of inclusive and engaging theatrical work was a crucial step in creating a successful gathering spot. Thus amateur theaters were doubly blessed (or burdened) with the tasks of community building; they were deemed the ingredient most effective in transforming the bare walls of an occupied storeroom into a new kind of public space.
After the first chaotic years of the revolution, Soviet state agencies devoted considerable resources to clubs. Trade unions, their primary sponsor, began to construct new buildings devoted solely to club activities. For Soviet architects, these new spaces offered special challenges; it was a chance to create an environment that could embody the collectivist principles of the Russian revolution. So important were these problems to Soviet designers that a club interior—Alexander Rodchenko's model workers' club—was chosen to represent the Soviet Union at the International Festival of Industrial Arts in Paris in 1925.8 Inventive club buildings, most constructed in the late 1920s, served as the premier examples of Soviet constructivist architecture. The size, form, and placement of the stage, so central to club activity, became a key issue for a new generation of architects.9
No matter where they were staged, amateur performances helped to legitimize the Soviet state. By seizing on the pressing issues of the day, many works encouraged army enlistment, mobilized participants for Soviet celebrations, and informed audiences about international events. Those stages that chose a repertoire of familiar prerevolutionary plays helped to provide viewers with the rudiments of cultural literacy. Because of their popularity and ability to transmit political and cultural values, amateur stages were extremely useful to the government. At a time when film equipment was scarce and illiteracy was high, theaters spread the political message of the revolution. They were also evidence that the revolutionary state was committed to a mission of enlightenment.
These humble stages even contributed to the Soviet Union's cultural influence abroad. This was especially true during the years of the First Five-Year Plan (1928–32), when Soviet industrial expansion provided a vivid contrast to capitalist nations trapped in a global depression. The openly agitational, politicized style of amateur performance in this period inspired emulation by many Communist theater groups, from the Chicago Blue Blouses to the Red Rockets in Berlin.10 Soviet amateurs offered their Western counterparts a model of cultural creation through confrontation with the past, an approach radically opposed to the ideas of the Social Democrats, who still tried to offer workers access to their cultural heritage.11 Communists everywhere proclaimed that their art would be a "weapon in the hands of the working class." Yet despite this common expression of theater's role, and despite the many similar forms employed, there is one striking difference between agitational theater in Western nations and the Soviet Union. Soviet theaters were agitating for a state that already existed; in the West they were fighting for a state that was yet to be.
Although most Soviet amateurs embraced their agitational tasks, they could still run afoul of the political apparatus. State institutions invested considerable resources to oversee their work, but ultimately it could not be completely controlled. In the 1920s, scripts were prepared and altered at the performance site by individual instructors and the actors themselves. Even government-approved plays could be staged in unexpected ways. One critic writing in the early 1930s was offended by a performance of Nikolai Gogol's satirical play poking fun at the tsarist bureaucracy, The Inspector General, because the theater dared to draw parallels with the Soviet bureaucracy.12 Amateur performances were public forums where participants could express their own political and social visions. As such they always carried the potential for subversion.
#### Pre-Revolutionary Origins
Soviet amateur theaters drew on the experience, repertoire, and personnel of the Russian popular theater movement. Efforts to democratize the theater and broaden its social base were evident all over Europe in the last decades of the nineteenth century.13 In Russia, popular theaters (narodnye teatry) were sponsored by government agencies, progressive intellectuals, and sometimes even factory owners who built stages at the work site. Supporters believed that edifying forms of entertainment would help to transform the tastes and habits of the lower classes. In addition, some were motivated by altruistic aims. For them Russian elite culture, including the works of Gogol, Anton Chekhov, and Alexander Ostrovsky, was a national treasure that should be shared with all the people.14
Advocates of peoples' theater hoped their efforts would counteract less enlightened forms of popular entertainment. They specifically targeted a long tradition of fairground theaters, erected during the spring and winter holiday seasons. Called balagany, these temporary wooden structures offered a wide range of works, from comic sketches, to peep shows, to Petrushka plays, the Russian version of Punch and Judy. Initially presented free of charge, with donations requested by performers, these festivities became increasingly commercialized by the late nineteenth century. Entrepreneurs added roller-coaster rides and sometimes even short films to the attractions. They also varied the theatrical repertoire, offering patriotic plays, pantomimes, and scenes from works common on professional stages.15 Often rowdy affairs enlivened with drinking bouts and fist fights, such entertainments were placed under ever more watchful control by the tsarist government in the last decades of the regime. Officials moved them away from the central urban areas and even banned alcohol to make them more respectable.16
After the revolution of 1905, which stimulated the growth of proletarian institutions, workers began to create their own theaters, often located in clubs and Peoples' Homes funded by the trade union movement. As Anthony Swift's work has shown, Russian workers' theaters did not aim for an experimental or self-generated repertoire. However, participants did insist on deciding for themselves which works would be presented. By and large, they chose from a store of Russian and Western European classics. In particular workers were drawn to plays they felt had a progressive message to convey, such as Ostrovsky's Poverty Is No Crime, interpreted by worker audiences as a critique of capitalism, and Gerhart Hauptmann's The Weavers, an homage to workers' rebellion.17
Rural Russia had its own forms of theatrical entertainments. Drawing on both pagan and Christian traditions, Russian peasants participated in a wide range of ritualistic dramas associated with planting, harvesting, and the important life passages of birth, marriage, and death. Peasants also engaged in scripted games and improvisations. At least by the nineteenth century, these improvisations had evolved into more developed plot outlines for non-ritualistic dramas. Often performed at Lenten festivals, these included short satirical works and longer plays distinguished by their episodic structure and their free relationship to historical material. The best known of these works, The Boat (Lodka) examining a trip down the Volga, and Tsar Maksimilian, offering a moving confrontation between a tsar and his son, served as bases for improvised entertainment well into the Soviet period.18
Soviet amateur theaters drew on these different traditions of popular theater. Many cultural circles designed to foster amateur theatricals continued their work almost unchanged after the revolution. The famous Ligovskii People's Home in Petrograd/Leningrad, founded in 1903, served as an educational base for a generation of worker authors. After the revolution, it continued to sponsor the well-known traveling theatrical troupe led by Pavel Gaideburov, a director devoted to spreading the classics of Russian and world theater to the lower classes. The set designer Vasilii Polenov assumed control of an organization to aid workers' and peasants' theater in 1912. His center continued its work after 1917, now under the auspices of the Soviet state's cultural ministry, Narkompros.19 These artists shared many Bolsheviks' respect for established high culture, as well as their contempt for the commercialism of the capitalist marketplace.20 Thus, amateur theater provided a way for sectors of the old intelligentsia to find an institutional home under the new regime.
#### Amateur Theater in Soviet Cultural Debates
As an art form directly involving the lower classes, the very population the Bolsheviks claimed to serve, amateur theaters found themselves at the center of controversies concerning the form and function of revolutionary culture. Bolsheviks were passionately committed to bringing about a wide-scale cultural transformation or, in their words, a "cultural revolution" that would solidify the gains of the political upheaval.21 But this was hardly a straightforward process, since they did not share a common vision of what the cultured Soviet citizen should be like.
Nonetheless, one assumption the new state's leaders did share was that theater would be an important tool in crafting this new individual. Theater was, in the words of Katerina Clark, "the cradle of Soviet culture."22 As an art form that could unify actor and audience, theater was believed to have special abilities to create shared community values. It did not rely simply on words but melded language with color, light, music, and movement. By integrating the intellect and the emotions, theater had the power to create new patterns of behavior. Acting on these premises, the Bolsheviks moved quickly to nationalize important professional theaters and monitor the work of impromptu stages.
Questions about performance space and repertoire drew amateur actors, often unwittingly, into the long-running controversy between advocates of realism and the theatrical avant-garde. Avant-gardists contended that amateur theaters were uniquely situated to accomplish one of their most cherished goals—to erase the division between performers on the stage and the passive viewer audience. Vsevolod Meyerhold, the nation's most famous experimental director, opened a "Club Methodological Laboratory" to train directors and writers for club theaters.23 His students endorsed stylized, episodic performances that could be altered through group participation. Many amateurs embraced this direction as a way to make their performances directly relevant to local struggles. These methods, they argued, provided them with an avenue for self-expression and self-determination.
For proponents of realism, the amateur stage had a different purpose, namely "to illuminate life, satisfy spiritual longing, and aid in further education," in the words of one commentator.24 Professionals from the Moscow Art Theater taught Stanislavsky's acting techniques in order to help amateurs create persuasive characters. Realist playwrights presented stories that would teach an inspiring history of the revolution, offering actors and viewers positive role models to emulate. Supporters of this direction believed their work nurtured a socialist consciousness, while the loosely structured plays inspired by the avant-garde only confused audiences. Their insistence on uplifting tales and stellar heroes, still very fluid in the 1920s, eventually solidified into the fixed rules of socialist realism.
As competitors for an urban audience, amateur theaters were also drawn into debates over the continuing existence of commercial culture in the Soviet Union. In the years of the New Economic Policy (1921–28), when limited capitalist enterprise was permitted, Russian cities saw a growth in restaurants, dance halls, and movie theaters playing foreign films. A segment of the cultural bureaucracy saw this as a threat to socialism, questioning the value and indeed the morality of "purely entertaining" forms of amusement. Their strict position was opposed by others who insisted that work designed for educational purposes alone would alienate audiences and drive them away. This controversy, which Denise Youngblood has called the "entertainment or enlightenment debate," raged for much of the 1920s.25 It was hardly unique to Soviet Russia, as left-leaning cultural leaders everywhere questioned how best to approach popular entertainments such as adventure films and variety shows that were generated in the capitalist marketplace.26
As self-styled educators of their audiences, amateur performers tried to mix enlightenment and entertainment. During the 1920s, potential viewers had many other choices for an evening on the town. Innovators tried to devise politically acceptable works that incorporated appealing elements of urban mass culture. They integrated slide shows into performances to make them approximate films. They used melodies popular in cafes and night clubs. These efforts met with derision from puritanical critics, who found them at best frivolous and at worst a dangerous concession to capitalist decadence.
In addition, amateur theaters were plagued by an even more fundamental problem that went to the heart of Soviet social organization: What should these theaters' relationship to professionals be? This question was highly politicized in the early Soviet years, not only in the arts but also in the army, trade unions, and education. Everywhere, the revolution provoked hard-fought battles over the status of experts and the significance of expertise, battles that sought to determine the meaning of social equality in the world's first socialist stated.27
Participants in amateur theaters had no easy answers to these accursed questions, proposing two contradictory models for the cultured Soviet citizen. Some practitioners were inspired by ideas reminiscent of the young Karl Marx and Friedrich Engels, who claimed: "The exclusive concentration of artistic talent in a few individuals and its consequent suppression in the large masses is the result of the division of labor.... In a communist organization of society there are no painters; at most there are people who, among other things, also paint."28 According to this ideal, if amateur theatrical work spread widely among the population, established stages might eventually be abandoned altogether. Others insisted, however, that the Soviet system would show its superiority by discovering talented workers who would pass through amateur theaters to a career on the professional stage. These models were based on conflicting ideas of theater itself—was it a participatory activity infusing all of life or a skilled profession to be learned?
"Do-it-yourself theater" aptly describes amateur activity during the Bolshevik revolution and the Civil War, the subject of chapter 1. Central control was weak, accounting in part for the remarkable diversity of repertoire during this chaotic period. Original agitational works were common, particularly in the influential theaters of the Red Army. In addition, prerevolutionary classical plays as well as less edifying potboilers found their actors and audiences. Civil War amateur theater was in large part a battle for public visibility—with actors seizing the right to new public roles. Unlikely urban environments were transformed into performance spaces—restaurants, basements, and, most symbolically, gathering spots for the former privileged classes. The quality of performances was usually indifferent; actors took little time to prepare and they had few props or costumes. But a polished presentation was not essential for audience or actors—the important thing was that the performance was taking place at all.
In the early years of the New Economic Policy (NEP), many amateur actors and directors wanted to continue what they regarded as the real accomplishments of the Civil War, especially its improvisational, politicized theatrical experiments. Chapter 2 examines efforts to make "small forms"—skits, mime, circus techniques, and loosely connected episodic works—the main focus of amateur work. Its proponents claimed that the amateur theater of small forms was more innovative and invigorating, and more closely tied to daily life, than anything performed on professional stages. A few took these ideas to extremes, rejecting any kind of professional involvement. Enthusiastic voices in favor of an amateur theater of small forms found a broad public forum in the early 1920s, as theater circles experimented with methods to educate and entertain audiences simultaneously.
By the late NEP period, however, there were clear signs that small forms were beginning to lose their constituency. Chapter 3 investigates a turn away from small forms after 1925. Criticism came in part from the viewing audience, especially select worker-reporters (called rabochie korrespondenty, or rabkory), who had grown tired of well-worn stereotypes and predictability.29 The debate over the repertoire of club theater, generally restricted to specialized journals in the early 1920s, began to emerge as a topic of national discussion. Cultural bureaucrats insisted that positive changes in professional theaters had made the oppositional stance of amateur stages obsolete; now they advanced the idea of a smychka, or union, between the amateur and professional arts.30 They advised amateurs to turn to larger works and perhaps even to try the same plays that were gaining audiences on professional stages.
Chapter 4 offers a case study of a particularly influential amateur theater, the Leningrad Theater of Working-Class Youth, or TRAM, sponsored by the Leningrad Komsomol. It garnered more national and international attention than any other Soviet amateur stage, in large part because of its original repertoire. Begun in the early 1920s, it initially staged small forms. By the mid-1920s, however, it moved to more sustained plays written by its own youthful members. Helped by the Komsomol press, it soon acquired a national following and local affiliates in other urban centers. Its members saw themselves as separate from and, indeed, superior to professionals in the theater. TRAM members developed what they believed to be the clearest articulation of samodeiatel'nyi teatr. They were not actors but rather activists who drew their material from the streets, factories, and dormitories. Their aim was to influence the behavior of viewers.
Yet for all their claims to a radical aesthetics, TRAM theaters still bore many identifying marks of established theater. They performed three-to five-act plays that offered a cohesive narrative. They presented their work on conventional stages, with sophisticated lighting, costumes, and set designs. Chapter 5 examines a much more extreme form of cultural experimentation, the agitprop brigades. These small, mobile, and politically motivated groups flourished during the years of the First Five-Year Plan. Brigades were composed of young enthusiasts from trade unions, clubs, and factories. Touring work sites and the countryside to drum up support for the industrialization and collectivization drives, they prepared agitational skits and short plays from the raw materials at hand— newspapers, public speeches, and production statistics. Brigade participants were distrustful of professional theater workers and playwrights, who allegedly had no knowledge of daily struggle at the workplace. Instead, they tried to rely on their own experiences as laborers and political activists. Using aggressive methods inspired by shock workers in production, the brigades aimed to root out old habits' and shame those who practiced them.31
But the dominance of agitprop brigades was brief. Viewers complained about their monotonous repertoire and lack of believable heroes. Perhaps more serious, critics began to call the political reliability of agitprop brigades into question. By the time that the first National Olympiad of Amateur Art was held in Moscow in the summer of 1932, this form of theatrical activism faced overwhelmingly negative criticism in the cultural press. Judges and journalists advised amateur circles to attempt works by contemporary Soviet playwrights and also to take on classical plays. When addressing political themes, amateur theaters had to learn to do so "artistically," which was only possible with the intervention of those trained in technique and familiar with the long history of Russian and world theater.32
The final chapter examines amateur stages in the 1930s. The evolving doctrine of socialist realism, which was applied to amateur as well as professional art, brought increased standardization. The methods of a few professional groups, particularly the Moscow Art Theater, were imported to amateur stages. Established theaters supplied directors and opened comprehensive training programs for amateur circles. New stages built in the 1930s looked like their professional prototypes, with proscenium stages and a dear division between the performers and the audience. Select clubs in the capital cities now had performance halls seating thousands, large stages, and healthy financial support, allowing them to mount elaborate productions.
The acceptable repertoire for amateur stages narrowed precipitously during the 1930s. At the 1938 Moscow competition of amateur art, the end point of this book, amateur circles performed a short list of contemporary works along with a limited assortment of prerevolutionary das-sics.33 Although participants and critics paid lip service to the independent role of samodeiatel'nyi teatr—its dose ties to audiences and ability to provide insights into everyday affairs—in fact, the biggest compliment that could be bestowed on an amateur stage in the late 1930s was that its work met professional standards.
#### Assessing the Amateur
In her study of amateur films in the United States, Patricia Zimmermann traces the emergence of amateurism as a concept important to the late nineteenth century, when professionalization became a dominant force in American public life. Both defenders and critics of amateurism examined this phenomenon through a language of stark dichotomies. The professional worked for money, while the amateur worked for pleasure; the professional labored in public, while the amateur moved largely in the private sphere; the professional expressed commonly shared, rational values, while the amateur injected spontaneous, localized elements into creation. She concludes, "Amateurism deflected the chaotic, the incoherent, and the spontaneous into leisure and private life so that public time could persist as methodical, controllable, and regulated."34
These reflections on amateurism's place within capitalism offer a chance to locate its distinctive features in the Soviet system. Certainly, neither Soviet participants nor observers believed amateur theatricals took place in the "private sphere." Clubs were, after all, a "public hearth," valued precisely for freeing participants from the narrow, philistine confines of their private homes. Soviet amateur theaters had very important public responsibilities—to educate their audiences, to mark public holidays, to take part in public demonstrations, and to voice the creative ideas of their constituents. Furthermore, the entire concept of a "private life" was denounced as a bourgeois construct in the early Soviet years. As Eric Naiman's work has shown, early Soviet social discourse had nothing but disdain and fear for the private sphere.35
Soviet amateurs were also not separate from the world of work in the same way as their capitalist counterparts. Western amateurs often justified their activities as a way to maintain a spark of individual expression within an increasingly regimented capitalist economy.36 By contrast, Soviet advocates argued that their style of amateurism revealed the superiority of the socialist system. In their free time, Soviet citizens turned to edifying activities that raised their cultural level and facilitated collective interaction. Citizens' performances were designed to inspire workers with civic pride and professional skills. Soviet skits and plays of the 1920s addressed the flaws and accomplishments of the work environment. And even in the 1930s, when models of amateur participation changed radically, amateur actors asserted that the discipline required for performances increased their labor productivity.
However, in one important respect Soviet and Western amateurism held something in common. They both illuminated the differences between regimented professional activity and the spontaneous, self-regulated work of the non-professional. The very term that the Soviets chose for the amateur—samodeiatel'nost'—can be translated as autonomous action.37 Soviet defenses of amateur theater put forward in the 1920s underscored precisely this aspect of amateur work—that it gave voice to the local and particular in a way that professionalism never could. Amateurism was the source of inspiration for stagnant professionals, whose training and repertoire could quickly become stale. Outspoken advocates of amateur theater in the capitalist world have expressed their ideas in much the same terms.38
It was precisely the homemade, unpredictable quality of amateur work that concerned Soviet regulators. They viewed the amateur stage as a potential purveyor of low cultural values, degraded language, sexual titillation, and dangerous political ideas. During the 1920s and the years of the First Five-Year Plan, power shifted back and forth between those favoring either more spontaneity or more control. However, by the 1930s many avenues for independent creation had been blocked. The repertoire was tightly regulated, stages were now constructed along highly conventional lines, and professional artists supervised essential elements of training programs. The overseers of amateur theaters took all possible pains to determine that amateur art would become methodical, controllable, and regulated, eliminating any stark contrast with the professional.
Yet even under Stalin, amateur theater at times proved difficult to control. One critic was appalled by a 1936 Moscow amateur production of Nikolai Pogodin's Aristocrats, a very popular play depicting the rehabilitation of criminals sent to build the White Sea Canal. According to this critic, the criminals were portrayed as romantic, tragic figures. By contrast, their secret police overseers, the intended heroes of the piece, looked wooden and unconvincing. How could the Moscow trade union leadership have allowed such a raw and unfinished work to be performed in public, queried the critic in a tone of moral outrage.39 Although the director meekly denied any illicit intentions, amateur performances still permitted subversive interpretations.
* * *
1. A. I. Piotrovskii, "God uspekhov," in his Za sovetskii teatr! Sbornik statei (Leningrad: Academia, 1925), 67–68. Unless otherwise indicated, all translations are my own. On Piotrovskii's career, see Katerina Clark, Petersburg, Crucible of Cultural Revolution (Cambridge: Harvard University Press, 1995); James von Geldern, Bolshevik Festivals, 1917–1920 (Berkeley: University of California Press, 1993); and E. S. Dobin, ed., Adrian Piotrovskii: Teatr, kino, zhizn' (Leningrad: Iskusstvo, 1969).
2. For overviews of Soviet amateur theater, see S. Iu. Rumiantsev and A. P. Shul'pin, eds., Samodeiatel'noe khudozhestvennoe tvorchestvo v SSSR. 2 v. (Moscow: Gosudarstvennyi institut iskusstvoznaniia, 1995); V. B. Blok, "Khudozhestvennoe tvorchestvo mass," in A. Ia. Zis', ed., Stranitsy istorii sovetskoi khudozhestvennoi kul'tury (Moscow: Nauka, 1989), 11–42; V. Ivashev, ed., Ot "zhivoi gazety" do teatra-studii (Moscow: Molodaia gvardiia, 1989); V. N. Aizenshtadt, Sovetskii samodeiatel'nyi teatr: Osnovnye etapy razvitiia (Kharkov: Khar'kovskii gosudarstvennyi institut kul'tury, 1983); and N. G. Zograf et al., eds., Ocherki istorii russkogo sovetskogo teatra v trekh tomakh (Moscow: Akademiia nauk, 1954–60), v. 1: 467–78, v. 2: 460–78. On the significance of amateur theater to postwar educational institutions, see Anne White, Destalinization and the House of Culture (London: Routledge, 1990), esp. 70–79.
3. E. Anthony Swift coined this inventive translation. See his "Workers' Theater and 'Proletarian Culture' in Pre-Revolutionary Russia, 1905–1917," Russian History 23 (1996): 94.
4. P. A. Markov, The Soviet Theatre (London: Victor Gollancz, 1934), 137.
5. Novye etapy samodeiatel'noi khudozhestvennoi raboty (Leningrad: Teakinopechat', 1930), 11.
6. El Lissitzky, Russia: An Architecture for World Revolution, trans. Eric Dluhosch (Cambridge: MIT Press, 1970), 44.
7. On clubs in general, see Gabriele Gorzka, Arbeiterkultur in der Sowjetunion: Industriearbeiter-Klubs, 1917–1928 (Berlin: Amo Spitz, 1990); John Hatch, "Hangouts and Hangovers: State, Class and Culture in Moscow's Workers' Club Movement, 1925–1928," Russian Review 53, no. 1 (1994): 97–117; idem, "The Formation of Working Class Culture Institutions during NEP: The Workers' Club Movement in Moscow, 1921–1923," Carl Beck Papers in Russian and East European Studies, no. 806 (Pittsburgh: University of Pittsburgh Center for Russian and East European Studies, 1990).
8. Selim O. Khan-Magomedov, Rodchenko: The Complete Work (Cambridge: MIT Press, 1987), 178–86.
9. On club architecture, see V. Khazanova, Klubnaia zhizn' i arkhitektura kluba. 2 v. (Moscow: Rossiiskii institut iskusstvoznaniia, 1994); and Frederick Starr, Melnikov: Solo Architect in a Mass Society (Princeton: Princeton University Press, 1978), 127–47.
10. On Soviet influence, see Richard Stourac and Kathleen McCreery, Theatre as a Weapon: Workers' Theatre in the Soviet Union, Germany and Britain, 1917–1934 (London: Rout-ledge, 1986); Richard Bodek, Proletarian Performance in Weimar Berlin: Agitprop, Chorus, and Brecht (Columbia, S.C.: Camden House, 1997); David Bradby, "The October Group and Theatre under the Front Populaire," in Politics and Performance in Popular Drama, ed. David Bradby et al. (Cambridge: Cambridge University Press, 1980); Stuart Cosgrove, "From Shock Troupe to Group Theatre," in Theatres of the left, 1880–1935, ed. Raphael Samuels (London: Routledge, 1984), 168–69; and Ira A. Levine, Left-Wing Dramatic Theory in the American Theatre (Ann Arbor: UMI Research Press, 1985), 100–104.
11. On Social Democratic approaches to theater in the 1920s, see Cecil W. Davies, Theater for the People: The Story of the Volksbühne (Manchester: Manchester University Press, 1977); Helmut Gruber, Red Vienna: Experiment in Working-Class Culture, 1914–1934 (New York: Oxford University Press, 1991), ch. 4; and W. L. Guttsman, Workers' Culture in Weimar Germany: Between Tradition and Commitment (New York: Berg, 1990), chs. 2 and 8.
12. A. Kasatkina, "Problemy klubnogo repertuara," Teatr i dramaturgiia 8 (1933): 52.
13. See David Bradby and John McCormack, The People's Theatre (London: Croom Helm, 1978), esp. 15–29.
14. On Russian people's theater, see Gary Thurston, The Popular Theatre Movement in Russia, 1862–1919 (Evanston, Ill.: Northwestern University Press, 1998); idem, "The Impact of Russian Popular Theatre, 1886–1915," Journal of Modern History 55, no. 2 (1983): 237–67; E. Anthony Swift, "Theater for the People: The Politics of Popular Culture in Urban Russia, 1861–1917" (Ph.D. dissertation, University of California, Berkeley, 1992); and G. A. Khaichenko, Russkii narodnyi teatr kontsa XIX-nachala XX veka (Moscow: Nauka, 1975).
15. On fairground theaters, see Catriona Kelly, Petrushka: The Russian Carnival Puppet Theatre (Cambridge: Cambridge University Press, 1990), 19–55 On their transformation in the late nineteenth century, see Al'bin M. Konechnyi, "Popular Carnivals During Mardi Gras and Easter Week in St. Petersburg," Russian Studies in History 35, no. 4 (Spring 1997): 52–91, esp. 72–82; and A. F. Nekrylova, Russkie narodnye gorodskie prazdniki, uveseleniia i zrelishcha (Leningrad: Iskusstvo, 1984), esp. 163–75.
16. Von Geldern, Bolshevik Festivals, 106–7; Swift, "Theater for the People," 259–60.
17. Swift, "Workers' Theater," 67–94. See also Mark D. Steinberg, Moral Communities (Berkeley: University of California Press, 1992), 241.
18. Elizabeth Warner, The Russian Folk Theatre (The Hague: Mouton, 1977), esp. 127–76; A. F. Nekrylova and N. I. Savushkina, "Russkii fol'klornyi teatr," in L. M. Leonov, ed., Narodnyi teatr (Moscow: Sovetskaia Rossiia, 1991), 5–20.
19. On Polenov's center, see "Sektsiia sodeistviia ustroistva derevenskikh, fabrichnykh i shkol'nykh teatrov," 15 February 1915, GARF, f. 628 (Tsentral'nyi Dom narodnogo tvorchestva im. N. K. Krupskoi), op. 1, d. 1, ll. 22–24, and "Shtaty Doma teatral'nogo prosveshcheniia im. V. D. Polenova" (1921), ibid., d. 104, 1. 2.
20. On the shared values of the prerevolutionary intelligentsia and the Bolshevik leadership, see Jeffrey Brooks, When Russia Learned to Read: Literacy and Popular Literature, 1861–1917 (Princeton: Princeton University Press, 1985), esp. 295–353.
21. By "cultural revolution," I do not mean the limited period of cultural radicalism during the First Five-Year Plan. Rather, I am referring to the broad Bolshevik project of cultural transformation that included literacy, cleanliness, and improved standards of health, as well as distinctly Soviet art forms. Michael David-Fox, "What Is Cultural Revolution?" Russian Review 58 (April 1999): 181–201, for a discussion of the history and significance of this concept.
22. Clark, Petersburg, 104. On the centrality of theater to early Bolshevik cultural projects, see also Julie Anne Cassiday, "The Theater of the World and the Theater of the State: Drama and the Show Trial in Early Soviet Russia" (Ph.D. dissertation, Stanford University, 1995), ch. 1.
23. On this studio, see Iurii Kobrin, Teatr imeni Vs. Meierkhol'da i rabochii zritel' (Moscow: Moskovskoe teatral'noe izdatel'stvo, 1926), 35, and ch. 2.
24. A. M., "Kakim dozhen byt' rabochii teatr," Novyizritel' 11 (1926): 5.
25. Denise J. Youngblood, Movies for the Masses: Popular Cinema and Soviet Society in the 1920s (Cambridge: Cambridge University Press, 1992), 35–49.
26. Gruber, Red Vienna, 123–35; Andreas Huyssen, "The Hidden Dialectic: Avant-garde—Technology—Mass Culture," in his After the Great Divide: Modernism, Mass Culture, Postmodernism (Bloomington: Indiana University Press, 1986), 3–15.
27. See Richard Stites, Revolutionary Dreams (New York: Oxford University Press, 1989), 124–44.
28. Karl Marx and Friedrich Engels, The German Ideology, in Karl Marx and Friedrich Engels, Literature and Art (New York: Progress Publishers, 1947), 76.
29. See, for example, V. Gomello, "Nuzhna-li p'esa rabochemu klubu?" Rabochii i teatr 7 (1925): 19.
30. "Itogi Vsesoiuznogo soveshchaniia pri Glavpolitprosvete," Zhizn' iskusstva 1 (1926): 2.
31. L. Tasin, "Blizhaishie zadachi dramkruzhkov," Zhizn' iskusstva 36 (1929): 2.
32. A. Kasatkina, "Iskusstvo millionov," Izvestiia 22 August 1932.
33. L. Subbotin, "Nekotorye vyvody iz smotra teatral'noi samodeiatel'nosti," Kul'turnaia rabota profsoiuzov 12 (1938): 72–77.
34. Patricia R. Zimmermann, Reel Families: A Social History of Amateur Film (Bloomington: Indiana University Press, 1995), 1–11, quotation 11.
35. Eric Naiman, Sex in Public: The Incarnation of Early Soviet Ideology (Princeton: Princeton University Press, 1997), esp. ch. 2.
36. This is a central theme of Wayne Booth's For the Love of It: Amateuring and Its Rivals (Chicago: University of Chicago Press, 1999).
37. Indeed, I translated the term this way in my first works on this topic. See Lynn Mally, "Autonomous Theater and the Origins of Socialist Realism: The 1932 Olympiad of Amateur Art," Russian Review 52 (April 1993): 198–212.
38. See, for example, the comments of Bonamy Dobrée, who determined that "the amateur has an extremely important, indeed vital part to play... in maintaining just that contact with the common apprehensions of life without which an art becomes stale or thin." (The Amateur and the Theatre [London: Hogarth Press, 1947], 6).
39. M. B., "Dekada samodeiatel'nykh teatrov," Klub 6 (1936): 29.
## 1
## The Revolution Loves the Theater
"THEATER IS the self-educator of the people," proclaimed a broadside published by the new state's cultural ministry in 1919. "The revolution loves the theater and in revolutionary times theater comes alive and blossoms."1 This statement attempted to explain the remarkable proliferation of theaters during the first years of the new regime. Some were sponsored by the central government, like the broad network of Red Army theaters; some had local institutional sponsors, such as regional soviets and city governments; and many were impromptu, spontaneous creations by factory councils, newly formed clubs, and informal groups of friends. Given these new groups' ephemeral nature, a precise count is impossible, but the back pages of local newspapers were filled with advertisements for amateur performances. Enthusiastic Bolshevik supporters used this explosion of theatrical activity as proof of the emancipatory power of the revolution. "Future historians will note that during a time of the most bloody and cruel revolution, all of Russia was acting," opined the art historian Piotr Kogan.2
This explosion in amateur theater work provoked anxiety as well as pride. Bewildered Russian intellectuals were at loss for an explanation, likening the phenomenon to fevers, epidemics, or even psychosis.3 Many professionals saw a threat to theater as an art form. They bemoaned the untrained actors, the impromptu repertoire, and the shoddy appearance of amateur performances that took place without their oversight. State cultural bureaucrats charged with supervising theaters worried about their ability to channel this frenetic activity that was only nominally under their control. As one high-ranking central official wrote with some despair about amateur stages, "As yet, we know very little about them."4
Although their voices are harder to capture, the participants in this rush to theater appeared to have different standards of judgment. For them, acting was a way to enter the public sphere—and thus to lay claim to a new community in which they would have a voice. Performing new works they had helped to shape gave articulation to their revolutionary visions. But even if the works performed were not new, acting meant seizing a public role. In their accounts of the period, both amateur actors and viewers seemed amazed that the performances were happening at all. Eduard Dune's vivid recollections of the first months of the revolution in a large factory on the outskirts of Moscow give a central place to theater as a builder of community. The wife of a skilled worker discovered her talents as a director, while the sets ("as good as those in any provincial theater") were designed by a factory painter: "It was a real eye-opener for many, who were seeing theater for the first time and were captivated by our simple entertainments."5
Dune's sentiments are echoed in the memoirs of theater group members at a Moscow textile factory: "The first performances we did on our own, without a leader. We put on small works from Chekhov and other authors. The very fact that workers were performing on stage, even without a leader, made a huge impression. People were proud and said, 'What a life!'"6 In these accounts, the amateur standing of the actors, people barely differentiated from the viewing audience, was central to the performance's appeal. As one newspaper critic determined, "These events eliminated the forced and destructive passivity of the viewer and turned the entire hall—both actors and viewers—into a unified, merged whole."7
The links that anthropologists and performance theorists have drawn between "aesthetic drama" and "social drama" offer insights into this enthusiasm for theater in a time of revolution. Aesthetic drama is what usually comes to mind when we think of theater. Its elements are almost entirely prearranged. Actors use a prepared text, perform in a fixed spot, and use established theatrical techniques of staging (lighting, scenery, props, etc.) and acting (declamation and movement). They are separate from their audience. The goal of aesthetic drama is to affect some sort of transformation in the consciousness of the viewers, even if that change is only temporary.8
Social dramas are sparked by real-life events. Victor Turner, who developed the concept of the social drama, applied this term to insurrections and revolutions. These moments of rupture arise in conflict situations, especially when groups try to occupy a new place in the social system. The participants act out a crisis in society, which finally results in a re-evaluation of the social order. In the social drama, there is no hard and fast distinction between the actors and the viewers. Both are altered through the process, and the change in this case can be permanent.9
Although aesthetic drama functions primarily on the stage and social drama can be played out anywhere, these two forms are nonetheless closely intertwined. As Victor Turner writes:
The stage drama, when it is meant to do more than entertain—though entertainment is always one of its vital aims—is a metacommentary, explicit or implicit, witting or unwitting, on the major social dramas of its social context (wars, revolutions, scandals, institutional changes.)... Life itself now becomes a mirror held up to art, and the living now perform their lives, for the protagonists of a social drama, a "drama of living," have been equipped by aesthetic drama with some of their most salient opinions, imageries, tropes, and ideological perspectives.10
In an attempt to concretize Turner's abstract language, we might ask what amateur actors were learning from the roles they adopted during the Russian revolution. How did they attempt to apply these lessons to the new life they believed the revolution would bring? Nikolai L'vov, a director of amateur groups before and after 1917, grappled with these issues when he tried to assess why so many people were turning to the amateur stage. He believed that both the theater and the revolution were the creators of new futures: "Now, as new ideas make their way through the population at large, and as people begin to see the possibilities of the new life, the broad popular classes immediately feel the call to the stage. Here they find an avenue for their desire for a brighter life. Here they have a chance to expand their spiritual life with new and unknown experiences."11
#### The Problem of Naming
Before the revolution, theater aimed at the lower classes was commonly called "popular theater" (narodnyi teatr). When the Bolshevik government began to oversee professional and amateur theatrical activity, one of the first struggles was over language. What was the new state's Commissariat of Education (known by its acronym, Narkompros) supposed to call the burgeoning impromptu theaters that it hoped to control? Anatolii Lunacharskii, the head of Narkompros, initially stuck with the old appellation.12 But the term "popular theater" struck many others as anachronistic. They began searching for a new descriptive terminology that bore fewer prerevolutionary connotations. Platon Kerzhentsev, an important cultural figure, tried out a number of alternatives in his influential book Creative Theater (Tvorcheskii teatr), which went through five editions in the first five years of the new regime. He referred to a vaguely defined "creative theater," a "socialist theater," and a "proletarian theater."13
In early 1919 Narkompros formed a division of "worker-peasant theater," which marked an initial attempt to find a new terminology for amateur efforts by the lower classes. Its first leader was the long-time activist in popular theater, Valentin Tikhonovich. For Tikhonovich, worker-peasant theater was a logical term to embody the art of the laboring classes. Despite the anti-peasant bias of many proletarian-based organizations like the Proletkult, Tikhonovich believed that these two social groups shared a lot in common: they both worked, many factory laborers were not that far removed from the land, and many peasants spent part of the year in a factory. Rather than turning their backs on society's largest social group, workers should collaborate with peasants to create a new theater.14
The first national conference on worker-peasant theater showed quite vividly, however, that many were not convinced by this line of argumentation. Difficulties emerged already in the planning stages. At a Petrograd meeting called in April 1919, the main speaker was the theater historian Vsevolod Vsevolodskii-Gerngross, who was engaged in compiling a detailed history of Russian folk theater. He bemoaned the separation between actor and audience created by the introduction of Western European theater in the eighteenth century; in his view the revolution's task was to reintroduce native theatrical traditions. This speech drew a hostile response from the audience, especially from one speaker who decried the notion of separating art according to class.15
After many delays, the worker-peasant theater conference finally opened in November 1919. It attracted representatives from a wide range of organizations sponsoring theatrical activity, including Narkompros, the Proletkult, the Red Army, trade unions, and cooperatives. Given this diverse constituency, it is hardly surprising that they reached no shared consensus. A majority position emerged at the conference, presenting what could be called an updated version of the mission of popular theaters before the revolution. Supported by provincial delegates, cooperatives, and professional theater workers, this position applauded efforts to improve theatrical quality, to make critical use of progressive elements from the prerevolutionary theatrical heritage, and to strengthen cooperation between workers and peasants.16
However, a vocal minority, made up of Communists, Red Army representatives, and members of the Proletkult organization, disputed all these points. They refused to endorse any suggestions passed by the majority, insisting instead on a separate set of resolutions.17 While the majority endorsed cooperation between its two constituencies, the Communist faction believed that peasants could be helpful only insofar as they subordinated themselves to workers. "Proletarian theater," read one of the minority resolutions, "is the task of workers themselves, along with those peasants who are willing to accept their ideology." Faction members also rejected the inclusive attitude the majority had formulated toward the artistic accomplishments of past generations. "[Workers' theater] must devote all its energies to formulating a new repertoire, without any borrowings from the past," they insisted.18
Perhaps most provocatively, the minority faction at the conference determined that proletarian theater should not embrace conventional forms of performance. Instead, they should aim for "mass action" (massovoe deistvo), by which they meant festivals, processions, demonstrations, and the celebration of new holidays.19 In essence, they were advocating social dramas that would affect both participants and viewers, as opposed to "old-fashioned" aesthetic drama. The conference chair, Tikhonovich, warned that while new forms were important, conventional plays were still needed. Focusing on festivals alone would mark the end of art. Many provincial delegates felt alienated from the rancorous discussions, complaining that they received little relevant help for the theaters they represented and came away without any clear understanding of what mass action was supposed to be.20
The fractious conference on worker-peasant theater ended the tenuous influence of the Worker-Peasant Theater Division over urban amateur stages. Only in existence for another year after the conference, the division spent its last months scrambling for more funds from Narkompros. Meetings were marked by the same ruptures that had split the conference.21 Without the authority or the staff to guide amateur stages, it continued mainly as an information-gathering body, compiling statistics about the social composition and repertoire of urban and rural amateur groups.22
After this debacle, theater activists began looking for a less explosive term for amateur theatrical activities. Tikhonovich and others championed samodeiatel'nyi teatr—translated literally as "self-activated" or "do-it-yourself" theater—as a more neutral and inclusive category for all nonprofessional stages. He made the switch himself during the Civil War. In December 1917, Tikhonovich finished Narodnyi teatr, an overview of amateur stages. A much amended version (which still had the same basic organizational structure), entitled Samodeiatel'nyi teatr was published in 1922.23 This new term, which I have translated as "amateur theater," was meant to sidestep old definitional quagmires. It was not class specific and thus could be applied equally to workers, peasants, white-collar workers, and students engaged in amateur activity. It removed the troublesome adjective narodnyi, associated with the efforts of "bourgeois" intellectuals before the revolution. Users also rejected another Russian term for "amateur," liubitel', rooted (as it is in English) in the verb "to love." In the early Soviet period, liubitel'skii teatr connoted all that was bad in amateur activities, such as posturing for good parts and wasting time in frivolous leisure-time pursuits. Liubitel'stvo took on all the negative connotations of dilettantism.24 By contrast, samodeiatel'nyi teatr came to stand for a new Soviet approach that would foster collective interaction and bring about productive social results.
Samodeiatel'nost' is an old Russian word for amateurism; it was hardly an invention of Soviet bureaucrats. Long before the revolution, amateur theaters had been called samodeiatel'nye teatry. Nonetheless, the term was imbued with new significance in the Soviet era. In its literal meaning of "self-activity," samodeiatel'nost' was claimed by the Soviet trade union movement as the embodiment of the spirit of the autonomous working class. The Proletkult also appropriated samodeiatel'nost' as one of its cardinal principles. This might help to explain why a term without class specifications was chosen at a time when most people were embracing class labels; workers' organizations had already appropriated this concept as their own.
With its connotations of autonomy and self-expression, samodeiatel'nost' carried a potential threat to higher authorities.25 State leaders wanted to encourage the ambitions and talents of the lower classes, particularly the working class, on which they based their legitimacy. These new historical actors needed to be able to "do things themselves." But what would happen if they acted in ways that offended or challenged the new government? What if their creative work proved difficult to guide and control? The possibility that self-activity might turn into dangerous spontaneity was a constant worry for early Soviet leaders. Spontaneity, stikhiinost', was a negative term in the Bolshevik lexicon, linked to anarchism, mindless rebellion, and ignorance. The lower classes, with their tendency toward spontaneity, needed to be led by the Communist Party toward consciousness.26
Autonomous self-activity was a slippery category on the continuum between spontaneity and consciousness. Advanced, "conscious" workers could be trusted to choose edifying pursuits; their activity bolstered the state's own arguments for power. When exercised by the unsophisticated, however, samodeiatel'nost' posed a potential threat to order. Thus the Bolsheviks were in the paradoxical position of continually encouraging self-activity while simultaneously trying to control it. Many of the conflicts surrounding Soviet amateur theater were embodied within the very name used to describe it.
#### The Locus of Performance
Amateur theaters were for the most part situated in clubs, a broad category that could describe anything from a well-appointed center built before the revolution to a requisitioned noble palace or converted storeroom.27 Although urban clubs for the laboring population existed well before the revolution, they expanded rapidly during 1917 and even more chaotically during the Civil War. Many different organizations were responsible for club formation. One short list of clubs in a Petrograd newspaper identified sponsors ranging from the local Communist party, regional city soviets, individual factories, and trade union organizations.28 In addition, strong local Proletkult organizations were particularly active in founding clubs, which would eventually pass to the leadership of trade unions and local government educational divisions at the end of the Civil War.29
The proliferation of clubs attracted the attention of many observers, who called them "social hearths" and "proletarian homes."30 During the tumultuous years of revolution and Civil War, many came to clubs because their own apartments had become unlivable. At a time when urban housing was often unheated, clubs that were lucky enough to have access to fuel served as a warm retreat. Club reading rooms operated as centers for crucial information on employment opportunities. Club buffets and cafeterias were a source of nourishment, and many observers noticed lines forming for club services only when buffets were about to open. Newspaper advertisements announcing club events used operating buffets as a way to entice patrons.31 While club advocates realized that many patrons were turning to these spaces out of necessity, they speculated that clubs would eventually begin to assume many of the functions of the private home, creating a new kind of public space.
Clubs also served as entertainment centers for their local communities. If at all possible, they offered a wide range of activities. Some clubs affiliated with trade unions had already amassed large libraries with tens of thousands of volumes.32 Even new clubs tried to open libraries and reading rooms. The Third International Club in Moscow, begun in 1918, had collected a library of some three thousand volumes by the following year.33 Clubs sponsored lectures on many topics, ranging from essential political issues of the day to more ethereal reflections on social thought. The First Worker-Peasant Club of Petrograd offered the following array of lectures during one week in the summer of 1918: "The Importance of Life-Long Learning," "The Meaning of Biogenetic Laws and the Human Spirit," and "Sigmund Freud's Theory of the Subconscious."34
Theatrical work was among the most popular club activity, although not all centers could support a theater group.35 Amateur actors flocked to workshops, and local audiences came to view their offerings. Well-endowed theater workshops with trained staff members could host a serious range of classes on diction, movement, and theater history. But most club theater circles did not have the time (or skills) for elaborate training programs; they instead tried to put on as many performances as possible for club audiences hungry for a constantly changing repertoire. Newspaper advertisements for club events during the Civil War announced upcoming "concert-meetings," with a list of activities to match this eclectic title. A typical club evening would feature some kind of recitation, an improvisation or play, along with lectures, music, and sometimes even dancing.36
Few clubs had access to large rooms with raised stages. A limited number of factories, run by enlightened capitalists who saw cultural work as a way of creating an educated labor force, built club spaces with stages before the revolution. After the 1905 revolution, newly empowered labor unions also opened clubs with performance spaces.37 But the majority of new centers founded after the revolution were opened in urban environments designed for other functions. Not only did they lack auditoria, they had no dressing rooms, space for props, or comfortable seating for the audience. Even when a club could boast an adequate hall for performances, many different groups laid claim on the space. Trade unions, the official sponsor of many clubs, needed auditoria for professional meetings and conferences. Music circles demanded room to practice and perform. This meant that theater groups had to discover other rooms—or hallways—for rehearsals.38 Some groups moved often to search for better accommodations. One Petrograd circle changed quarters four times in a two-year period.39
But raised stages, costumes, and assigned seating were not necessary ingredients of Civil War theater. "Just as a farm is a field where edible foods are grown, so a theater is a place where transformations of time, place, and persons... are accomplished," notes the director and theater historian, Richard Schechner.40 Spaces were made into theaters by means of the work that transpired there. Amateur actors and directors in Moscow and Petrograd transformed cafeterias, storerooms, bars, and basements into performance spaces. They moved into areas formerly reserved for the privileged classes before the revolution. The Komsomol club of the First City District in Petrograd, for example, was based in a gathering spot for city nobles called Russkoe Sobranie. The building had not been designed for theatrical performances, so the new inhabitants fashioned their stage in what had once been a spacious reading room. They made a curtain from the draperies and held them up by hand before performances. One factory theater in Moscow created costumes and sets out of contributions foraged from participants' apartments.41
Observers from the theatrical avant-garde found these innovations exhilarating because such steps seemed to follow their own suggestions to move away from naturalistic forms of presentation. The symbolist director Alexander Mgebrov, active in amateur theater circles, asserted that improvised spaces were superior to standard stages: "Seek your own arenas, my friends, and not moldy, stuffy, dusty boxes.... Your arena is everywhere and anywhere that you are.... Your arena is the whole world."42
Revolutionary architects saw the expansion of clubs as a chance to design new kinds of spaces that would facilitate collective interaction. In 1919 the Petrograd Department of Education launched an architectural competition for a "Workers' Palace" that would be a "completely new building, with no links to the past." The elaborate proposal included three different-sized performance spaces—a huge hall that could seat from two to three thousand, a smaller gathering spot for two to three hundred, and a room-sized stage "suitable for amateur performances."43 Although this plan was one in a long line of unrealized Soviet architectural projects, it showed that urban planners realized the importance of expanding the store of physical structures to house proliferating theatrical performances.
Some architects went even further, envisioning spaces where theatrical activity would not be set apart from the rest of club work. One activist in extracurricular education, A. Petrov, wrote extensively on club architecture. He was bitingly critical of conventional spaces where the theater dominated everything else, crowding out other work. "Where the theater begins," he wrote, meaning a conventional raised stage, "there the club ends."44 Instead, he advocated multipurpose rooms that would be suited for a variety of functions, including small improvisations and theatrical games. Petrov's proposals were the first in a long line of Soviet debates over how the size and shape of performance spaces would affect the life of the club.45
Problems of space were made even more daunting because no one could be sure just how many actors or viewers there would be. During the Civil War, amateur theater groups were plagued by constantly shifting memberships. Urban youth, especially young men, were the most common participants in club activities. One overview of club activities in Moscow determined that the most active participants were young men aged twenty to twenty-two.46 Not tied down by family responsibilities, they had the most free time. But this segment of the population was also most likely to volunteer, or be drafted—to the Red Army. "Unfortunately a common problem has greatly affected the continuation of our work," complained one cultural organizer in Petrograd. "Namely, the flow of the most active workers from Petrograd to the front or the provinces. The lack of workers can be felt most strongly in theater."47 It was on such shifting ground—with limited resources, physical impediments, and unreliable memberships—that the first Soviet amateur theaters took shape.
#### Oversight Agencies
During the early years of the Soviet regime, the new government quickly moved to centralize theatrical work by nationalizing the most important Russian theaters. By 1919, it also began to pass legislation limiting the independence of private stages. Narkompros created a central theatrical administration that claimed control over the buildings and property of all theaters, state-owned or private. This new bureaucratic body also reserved the right to control repertoire.48 This rapid intervention into theatrical life has caused many scholars to see the demise of local control over community-based stages already during the Civil War.49
It would be a mistake, however, to confuse the new government's ambition with its actual accomplishments. The burgeoning network of amateur theaters in Moscow and Petrograd encouraged a rapidly expanding web of local and national institutions that attempted to monitor their work. Since many different agencies claimed control over amateur theatrical activity, overlap and conflict between them were inevitable. Competition between government sponsors, known by the special term "parallelism," was a standard feature of early Soviet socialism. It was especially pronounced during the Civil War period, when the responsibilities of government agencies were not yet clearly defined and semi-independent groups like the Proletkult still had some range of independent action. Paradoxically, this competition proved advantageous for some groups. Savvy local circles learned to play state agencies against one another, gaining more funds, space, and staff in the process.
Until the formation of the Central Division for Political Education (Glavpolitprosvet) within Narkompros in late 1920, central state oversight for urban theaters was exercised through Extracurricular Education Divisions (Vneshkol'nye otdely). In the two capitals, these divisions provided a variety of services for amateur theaters, including training programs for club theater workers. The Petrograd division was particularly active. It intervened in struggles over space and also tried to set minimum quality standards for amateur stages. One pronouncement from the Petrograd division decreed that no group should be allowed to go in front of audiences with fewer than five rehearsals under its belt and that no club should sponsor more than two performances a week. These efforts at standardization yielded few results.50
The Red Army also wielded significant influence over amateur theaters. Its Political Section (PUR—Politicheskoe upravlenie voennogo revoliutsionnogo soveta) had a theater division that sent performing troupes to the front and also planned mass spectacles and agitational trials (see below). The head of PUR, Nikolai Podvoisky, was a close friend of the avant-garde director Vsevolod Meyerhold, who took a personal interest in the army's theatrical activities. In the course of the Civil War, the Red Army devoted considerable educational resources to clubs, which were seen as a way to fill soldiers' free time with edifying forms of relaxation. These clubs introduced new audiences to conventional theater and also encouraged amateur improvisation as a method of education. By 1920, Red Army sources claimed control over a thousand club theaters.51 The army proved an important training ground for individuals who would come to advocate a special agitational role for amateur theaters in the 1920s. These advocates included Adrian Piotrovskii and Sergei Radlov in Petrograd, who both were influential in the world of workers' clubs. Moscow's most radical voice for a new approach to the amateur stage in the 1920s was Vitalii Zhemchuzhnyi, who also directed amateur theaters in the Red Army during the Civil War.52
In Petrograd, the army opened its own set of theatrical training courses in 1919, called the Red Army Studio, which integrated influential figures from the world of prerevolutionary people's theater, as well as recruiting new activists to amateur theater. This division staged mass events, the examples of massovoe deistvo discussed at the worker-peasant theater conference. Its first and most successful work was The Overthrow of the Autocracy, a celebration of the February revolution. The division also offered training courses for theater instructors. The Baltic Fleet had its own theatrical studio, the Baltflot Theater, which provided models of repertoire and performance styles for other amateur theaters.53
The Proletkult organizations in Moscow and Petrograd controlled their own network of theatrical circles. Numerous city clubs were affiliated with the organization, most operating amateur theatrical studios. When possible, the central city organizations sent staff to oversee the work of these affiliates.54 In Petrograd, the Proletkult opened a central theater studio already in early 1918, which experimented with a variety of new works designed to inspire amateur stages. The Moscow Proletkult opened its central studio a few months later. Proletkult journals and affiliated publications were an important source of new theatrical works addressing the revolution and Civil War, as well as reviews covering the work of amateur stages. In addition, the Proletkult sponsored instructors' training courses and numerous seminars and classes on theater history and dramatic techniques.55
Trade unions also took a healthy interest in amateur stages. In early 1919, the national trade union organization founded a cultural division, which claimed oversight over union-affiliated workers' clubs and theaters. Each city had its own trade union cultural bureaucracy. Individual unions also opened cultural divisions in charge of educational and artistic activity, some already beginning work in 1917.56 Funds for trade union cultural projects came through membership dues, providing a funding source that was not directly dependent on government subsidies. Unions used these resources to renovate buildings for club activities and to train instructors to guide club work. Although the first priority of trade unions was technical education, that many urban amateur theaters operated out of trade union clubs gave the unions a measure of control. In Petrograd, the Railroad Workers' Union sponsored its own theater training courses and opened a union-sponsored theater workshop. The Moscow Railroad Workers' Union also sponsored training classes for local instructors, which included lectures on art and culture.57
The list of local institutions engaged in amateur theatrical activities hardly ends here. In Moscow, the city soviet took an active part in supervising amateur stages. Its theatrical-pedagogical section opened a variety of classes with the goal of improving the quality of instruction and performances.58 Cooperatives operated their own theaters, some with prerevolutionary roots. In addition, local Komsomol and Communist Party organizations also opened theaters. One Komsomol club in Petrograd, begun in 1919, eventually evolved into TRAM (Teatr rabochei molodezhi), the most influential amateur theater of the NEP and First Five-Year Plan eras.59
In late 1920, central state supervision of amateur theaters fell to the newly formed division of political education (Glavpolitprosvet) within Narkompros. At the local level, extracurricular education divisions were replaced by Politprosvet sections. This reorganization was partly intended to eliminate "parallelism" in the cultural field. Glavpolitprosvet tried to wrest control of cultural work away from the army and trade unions.60 It was also instrumental in implementing the Communist Party's newly articulated assault against the Proletkult.
Although competing institutions offered similar basic services to amateur stages, they often promoted very different aesthetic agendas. It would be a mistake to overemphasize their ability to shape local work, since all of them were understaffed and underfunded. Nonetheless, by hiring instructors and disseminating publications they could set the tone for their dependent circles. Thus the aesthetic proclivities of Adrian Piotrovskii, head of the Politprosvet division in Petrograd, influenced the agenda of groups dependent upon the division. Piotrovskii hired instructors who agreed with his general program, which was to turn away from conventional plays and toward improvisations and mass actions.61 Conversely, savvy groups learned to apply to many different agencies for funding. Thus trade union clubs, which received support from their local and national institutions, also petitioned the Proletkult bureaucracy for staff, money, and supplies.62 A trade union theater in Petrograd gained funding from the local Narkompros division to become a local touring troupe, mounting performances on club stages all over the city.63
The welter of state and local agencies provided employment opportunities for established theater activists with long prerevolutionary resumes. Venerable figures like Tikhonovich found a home in the upper echelons of Narkompros. Valentin Smyshliaev, trained at the First Studio of Moscow Art Theater, headed the central theater workshop of the Moscow Proletkult and was also employed in Narkompros' Division for Mass Spectacles.64 Other important figures with long pedigrees in the field of popular theater included Pavel Gaideburov, whose Traveling Popular Theater, founded in 1903, proved an important training ground for teachers and organizers of Soviet amateur theater circles. Gaideburov's group had toured the provinces before the revolution, with work aimed primarily at the local intelligentsia. With the advent of the First World War, his circle began to address more humble audiences, touring the front lines with works designed to educate and entertain the fighting forces. Gaideburov's base remained the Ligovskii People's Home, one of the most important urban institutions devoted to the education of the working-class intelligentsia. Here he insisted on a "classical" repertoire and steadfastly refused to change the content of performances to reflect the Bolshevik seizure of power. He also found work as the organizer and prime instructor for theater classes organized by the Petrograd division of extracurricular education, which was influential in recruiting energetic young people to the cause of popular theater.65
A new generation of theater activists found a calling (and a salary) in the maze of new Soviet bureaucratic structures. These included Dmitrii Shcheglov and Vitalii Zhemchuzhnyi, both born just before the turn of the century. Zhemchuzhnyi got his start in the army and eventually made his way into the cultural bureaucracy of the central Moscow Trade Union organization. Shcheglov's checkered career showed that instructors who were dissatisfied with the aesthetic direction of one agency could easily find support elsewhere. This skilled director found his first job in Gaideburov's Traveling Popular Theater. From this post he moved to Petrograd's newly founded Red Army Studio, where he helped to develop mass spectacles. The Petrograd Proletkult lured him away from that spot in 1919 with a promise of his own theatrical studio. When aesthetic conflicts developed there, he was recruited for the city's new theater division under the auspices of the local Politprosvet division, at that point headed by Piotrovskii. However, the contentious Shcheglov had troubles there as well, and he eventually ended up working for the cultural division of the provincial trade union bureaucracy.66 Thus the tangle of overlapping agencies helped to promote the aesthetic diversity—or perhaps one should say aesthetic chaos—of amateur stages.
#### The Debate on Repertoire: Old and New Plays
By far the most rancorous debates surrounding amateur theaters—both during and after the Civil War—were about repertoire. While circle participants themselves were perhaps most focused on the very fact that they were performing, those who oversaw the proliferation of club stages were vitally concerned with the content of performances. Through published government lists, instructors' training courses, and interventions at the level of club theaters, cultural activists tried to shape a distinctly "revolutionary face" for amateur stages.
Those theater circles with a prerevolutionary pedigree had an existing repertoire at hand. They chose the same plays that had been common in intellectual-sponsored people's theaters before the revolution, among them works by Alexander Ostrovsky, Leo Tolstoi, and Maxim Gorky. Trade union stages, most founded after 1905, also performed selected works from the Russian classics, along with foreign works that could claim a revolutionary imprimatur.67 New theater circles also drew on a prerevolutionary repertoire. The Northwestern Railroad Workers' club in Petrograd started operations with a standard supply of Ostrovsky plays. A study conducted by the Worker-Peasant Theater Division in 1920 determined that more than a third of all plays performed in some two hundred and fifty amateur theater circles around the country came from the classical repertoire.68
Government agencies encouraged the use of the prerevolutionary repertoire by assembling lists of suitable plays. These compendiums included some of the most popular works on factory and club stages before the revolution, including Ostrovsky's dramas. They also indicated, however, that some officials of the new regime wanted to redefine what the classics meant. While Anton Chekhov's were plays often performed before and after the revolution, they were missing from a list of suitable works put out by the Moscow soviet. The compilers might have held the view, common in the early Soviet period, that Chekhov expressed the alien vision of the dying upper class. This agency also hoped to add some new names to the common amateur repertoire, including Greek classics, William Shakespeare, and a variety of contemporary Western European authors, including the revolutionary Belgian playwright Emile Verhaeren.69
Individual theater activists were also important in drawing up lists of appropriate prerevolutionary works. Piotrovskii put together an ambitious list of works for military stages that drew heavily on the theatrical classics, including Shakespeare, Molière, and prerevolutionary works focusing on social themes. He also suggested prerevolutionary plays that might be used as the basis for improvisations, including Nikolai Gogol's Inspector General, a work that satirized prerevolutionary officials.70 In his influential volume Creative Theater, Platon Kerzhentsev also provided many suggestions for suitable prerevolutionary plays. Many of his suggestions had obvious political connotations, such as his endorsement of Gerhart Hauptmann's drama on a worker uprising, The Weavers, or Romain Rolland's heroic account of the French Revolution, Taking the Bastille.71
Those who agreed to stage the classics did not agree on the reasons. For some, they were simply a stopgap measure necessary until a new "revolutionary" repertoire could be developed.72 For others, a mastery of the classics would prove that the proletariat had laid claim to the cultural riches of the past. The head of Narkompros, Anatolii Lunacharskii, was an unwavering advocate of the latter view, one he insisted also represented the desires of most proletarian audiences.73 These opposing positions would be rehearsed again and again in the next decade and a half until the classics were finally permanently enshrined in the amateur repertoire in the 1930s.
Almost all cultural leaders despised the work of second-rate popular authors, whose plays had circulated before the revolution. One of the prime examples of this supposedly shoddy material were the plays of Sofia Belaia, whose melodramatic potboilers had already outraged the cultural intelligentsia before the revolution. Her prolific output of plays, almost all extolling the moral superiority of poverty over wealth, were favorites on amateur stages. Only the Communist critic Piotr Kogan had kind words for Belaia's work, remarking that she revealed the heartlessness of the exploiting classes.74 Ignoring the critical opinion against her, amateur stages turned to Belaia with enthusiasm. A 1920 study of amateur stages in Soviet-held territory listed her as the third most popular playwright, coming after Ostrovsky and Chekhov but before Gogol.75 Red Army and trade union clubs even used her most popular play, The Unemployed, to mark important festivals.76
One disgruntled observer believed that the most popular works were the most frivolous, including outdated melodramas like Two Orphans and The Fall of Pompeii. Such plays served mainly to keep viewers' minds off their troubles. Contemporaries attributed the popularity of this "hackwork" to the fact that many provincial actors, suddenly put out of work by the revolution, passed themselves off as directors for new theater circles and then foisted inferior plays upon them.77 Platon Kerzhentsev, the self-appointed doyen of worker amateur stages during the Civil War, bemoaned how amateur stages were performing the worst kinds of "bourgeois claptrap" in the most dilettantish fashion and lacked a serious approach to the task of building a new theatrical culture.78
While some government leaders were content to find an acceptable repertoire from prerevolutionary works, others were deeply troubled by this position. The longtime theater activist Nikolai L'vov believed that the revolution had made very little impact on the theater by 1920. Amateur circles were still largely copying the repertoire of professional stages, which had hardly changed in the preceding ten years. What evidence did theater offer to show that a revolution had taken place? Where were the new plays that could reflect the new reality, he queried.79 L'vov was not alone in his distress. For many advocates of a revolutionary theater, old plays could not meet the needs of the new society that was supposed to be taking shape.
To encourage new works, cultural bureaucrats proposed a variety of innovative methods. One of the most common was the playwriting competition, sponsored by many different institutions. Already in early 1918, the Proletkult announced prizes for the best dramatic work of a "new, original [samobytnyi], revolutionary proletarian character." The Narkompros division in Petrograd soon followed suit, opening a competition for a revolutionary melodrama, a form that both Lunacharskii and Maxim Gorky deemed particularly inspiring in the revolutionary age. Individual trade unions and Red Army divisions also sponsored playwriting contests.80
Competitions, however, were a cumbersome method to invent a new repertoire. They were not only time consuming—someone had to sort through the many entries—but they also posed formal restrictions on works. Many authors circumvented this process and came up with their own short works that were directly focused on the current crises of Civil War and social transformation. Called agitki (singular agitka—an abbreviation foragitatsionnaia p'esa or agitation play), these works addressed immediate social problems and attempted to sway audiences to support the Red cause. Often one-issue works with easily identifiable themes, they aimed to stem desertion, win peasant support, and even sketch out a hazy but optimistic future that would come with Bolshevik victory. They were very common on Red Army stages and in areas that were directly affected by fighting during the Civil War.
Agitki were usually very short, commonly with one act and only rarely more than two. They aimed to bring across a single point and to inspire political action. Thus detailed character development was not required. Characters were most often distinct and sharply juxtaposed social types—Red Army soldiers versus White Army soldiers, capitalists versus workers.81 Indeed, there is something very similar between a dominant style of Civil War posters, those portraying a world sharply differentiated between the good characteristics of the Bolsheviks and the evil characteristics of their opponents, and agitation plays. "The emphasis in this type of poster is on the caricatures themselves rather than on the narrative," notes Victoria Bonnell in her study of Soviet poster art.82 One could say the same for agitki. Like the posters, with their easily identifiable figures of the strong soldier in his peaked Red Army cap juxtaposed to the fat capitalist in his top hat, agitki used stereotypes to make their points. Perhaps these similarities should not be surprising, since agitki and many of the posters had similar goals—to inform often illiterate audiences quickly and inspire them to action. Later critics would criticize agitki for their plakatnost', or their poster-like quality.
With their clear delineation between good and evil, most agitki can be characterized as melodramas. Peter Brooks' influential work has shown the complex structure of melodramatic literature, belying the genre's reputation as overly simplistic. These mini-melodramas of the Russian Civil War reveal key elements of the genre, including a world polarized into moral absolutes. However, agitki characters do not give voice to deep feelings or reveal their innermost thoughts, a key element of bourgeois melodrama. Instead, they work out guidelines for proper political behavior in the new world created by the revolution. For Brooks, melodramas are the expression of a highly personalized sense of good and evil.83 Agitki, by contrast, attempted to formulate collective standards applicable to all.
Most of these short plays were written for the moment and thus did not gain wide viewerships. One exception was For the Red Soviets (Za krasnye sovety), by the Proletkult member Pavel Arskii. It is set in the home village of a peasant who had become a Red commander. White forces had seized the village and killed his wife and children. When the village is finally liberated by the Reds, the aggrieved commander issues a fiery appeal to join the Bolshevik cause in order to defeat the enemy. It was widely staged by the Red Army as a method to discourage desertion, a common theme in agitational plays.84 Other agitki sought to illustrate the kind of world the revolution was meant to create. An example would be a work by another Proletkult author, Pavel Bessal'ko, who wrote an allegorical play about a bricklayer who wanted to be an architect. The hero uses his skills to build a revolutionary tower of Babel that would end national divisions and inspire an international language.85 The play was popular both in club theaters and in theater studios of the Red Army. As was common for many of these short works, participants changed them to suit their tastes. In one Red Army version, led by the cultural activist Zhemchuzhnyi, performers added a scene after the completion of the tower, where representatives from all the nations of the earth bring presents to thank the builder for his efforts. This version ended with the singing of the Internationale.86
Writers also made use of Russia's folk heritage, instilling new content into old forms. The best example is the radical transformation of the traditional folk drama The Boat (Lodka). This loosely structured work can be traced back to the late eighteenth century. It began as a celebration of boatmen who worked on the Volga. By the nineteenth century, it came to feature fiery Cossack outlaw leaders who confronted rich landlords on their journey down the river. Soviet improvisations turned The Boat into a celebration of historic uprisings against the tsars, praising such Cossack rebels as Stenka Razin and Ataman Ermak.87 The traditional bad boy of the Russian puppet theater, Petrushka, was also enlisted during the early Soviet period for satirical agitki. In the process, he was transformed from a disrupter of social order into an advocate for the Red cause.88
When original scripts were not at hand, theater circles adapted nondramatic material for their purposes, creating a hybrid form known as the instsenirovka. Through this process, short stories, poetry, and even political speeches were transformed into staged performances. The most straightforward kind of instsenirovka was the adaptation of a single work of literature into dramatic form. Competition announcements often included lists of stories and novels that were deemed suitable for theatrical renditions. One call from the Moscow Proletkult, for example, proposed stories by Victor Hugo, Anatole France, and Jack London.89 Instsenirovki also wove together many works by a single author into a cohesive performance, a technique frequently used with poetry. The poems of Walt Whitman, Vladimir Maiakovskii, and Aleksei Gastev were all used as the basis for dramatic events during the Civil War. Instsenirovki were also shaped out of collections of different works, sometimes tied together by a narrator or individuals given specific parts. "Dawn of the Proletkult," composed by Vasilii Ignatov, was a compendium of different works by a number of proletarian poets. It introduced figures like "capital" and "the Russian Soviet Republic," who acted as narrators to give a common structure to the poems.90
#### Social Dramas
Despite their hasty composition, most new works written for performance during the Civil War still bore a strong resemblance to conventional plays. They presented actors with a list of characters; they offered a story with a clear beginning, middle, and end that had been predetermined by the author; and they clearly differentiated between the performers and the audience. As such, they still functioned within the world of aesthetic drama, taking their formal inspiration (if not their political messages) from standard performance practices. However, some works composed during the Civil War began to question standard notions of authorship and audience involvement. They gave performers a more central role in shaping the content of the work and integrated the audience in a more direct way. In the process, they blurred the lines between aesthetic and social drama.
Improvisations were one such method, a common practice in the theater studios of the Red Army and the Proletkult. Usually, with the help of an instructor, participants would choose a theme, like a mother's attempts to dissuade her son from joining the army or a wife's resistance to her husband's departure. From this base they would devise a rudimentary plot structure. According to one club leader, participants were eager to provide themes for improvisations. Some clubs even solicited suggestions from the audience and determined what they would perform by drawing lots at random out of a bag.91 The many pressures and divisions that the Civil War caused within families were a common topic of these sketches. One improvised story involved two brothers, one White and one Red, who both loved the same woman. In the end, the woman chose to link her future to the Bolshevik.92 Although the plot exaggerated the dilemmas most individuals faced in the Civil War, it illuminated the wrenching, life-altering choices demanded in a revolutionary period.
Another innovation was the living newspaper (zhivaia gazeta), a form that would come to have considerable influence in the Soviet Union and beyond in the next decade. This improvised genre grew out of efforts to present the news to audiences in an easily understandable form. Public readings to audiences without the ability to read or without easy access to newspapers or books had a long tradition in Russia. During the Civil War, when paper was a scarce commodity, the Soviet news agency, ROSTA, encouraged open readings of newspapers, with special times and places set aside for this activity. The readings were often augmented with music, scenery, and poetry. It was but a short step for agitational drama circles to act out the main characters in the feature stories.93 While the performance texts often had very clever writing, the broad topics they addressed were not determined by the authors. The inspiration for the content lay in current events, such as the current status of the Red Army, the international situation, or the food supply. Living newspapers aimed not simply to inform viewers but also to inspire action—to get audience members to fight harder, or turn in food speculators, or join the Bolshevik Party.
During the Civil War, the professional satirical troupe of the Red Army, the Theater of Revolutionary Satire, or Terevsat, did much to help popularize the methods of living newspapers. Terevsat considered its job to be providing humorous commentary on current political events and used newspaper reports to structure improvisations. Their performances followed a format similar to that of a newspaper—first commentary on the international situation, then on national events, and then satirical works based on local affairs. The Moscow Terevsat performed in clubs and factories, familiarizing audiences with its satirical style of improvised performance.94 With these examples before them, amateur circles had models to begin their own living newspapers.
Yet another revolutionary innovation were the agitational trials, or agitsudy. They first emerged as a widespread method of educational entertainment during the Civil War. Agit-trials, which put controversial issues to viewers for their judgment, also had prerevolutionary roots. Mock trials of matchmakers were a routine part of peasant courting rituals. The legal reforms of the nineteenth century encouraged the use of moot courts to instill an understanding of courtroom procedures. Peasant forms of rough justice, or samosud, through which rural communities devised their own form of punishment for offenders of shared values, also had some parallels to agit-trials.95
The Red Army was the major propagator of agitational trials during the Civil War. Soldiers stationed near Kharkov in late 1919 put the main character of the play The Vengeance of Fate (Mest' Sud'by) on trial, a wife who had killed her brutal husband. The participants were drawn by lots and the entire audience took part in the proceedings, which ended in an acquittal. Historical figures were also put on trial, including Rasputin and in one case even Lenin. Some of these events were massive affairs. In 1920 the Southern Army staged its first large-scale agitational court, "The Trial of Wrangel," before an audience of some ten thousand spectators. In this case, the outcome was preordained; Wrangel, the White general, was condemned to extinction by scripted witnesses who did not solicit the audience's response.
By late 1920 the Red Army political education workers focused much of their attention on agitational trials. These were staged in areas of heightened political tension and offered a way to articulate and resolve some of the problems of the unstable new regime. Any number of political enemies were brought to trial—among them the anti-Bolshevik Ukrainian nationalist, Simon Petliura, and the leader of Polish opposition, Josef Pilsudski. As "The Trial of Lenin" shows, the heroes of the moment were also brought to trial—and given a forum to dispute charges brought against them.96 Trials were expanded from cases against concrete individuals—something that could conceivably bear some similarities to a real courtroom trial—to broader spheres of activity. Unnamed enemies, such as the bearers of syphilis, army deserters, and global capital, were also called into the agitational courtroom.
Agit-trials within the army could be massive affairs, involving casts of thousands. The method was also employed in the more intimate context of club theaters, however. One club workshop staged its own trial of enemies, putting Pilsudski on trial in 1920. Club members in the Bauman district of Moscow decided to put the year 1920 on trial at the beginning of 1921. Audience members accused the old year of bringing them war and hardship. In the end, the year 1920 was able to acquit itself eloquently, insisting that it had paved the way to a more optimistic period.97 The broad range of subjects addressed in agit-trials during the Civil War prepared the way for their widespread use during the next decade, when they became an important genre of club performances.
The Red Army was also influential in staging and encouraging large mass spectacles, which emerged as an important venue for the new Bolshevik government to articulate its aims to the population at large. These events aimed to present a coherent history of the revolution from the Bolshevik point of view. Modeled in part on tsarist festivals and engaging the talents of some of the finest theatrical professionals in Russia, these elaborate events engaged literally a cast of thousands.98 Amateur theater groups from workers' clubs and the Red Army were integrated into mass scenes, working together with theater professionals. In Petrograd on May Day 1920, for example, army and navy amateur groups performed together with theater students and professionals in a mass event entitled "The Spartacus Rebellion."99
In a fashion similar to the large public displays, clubs staged festivities that aimed to fix a prehistory of the revolution. One Petrograd Proletkult studio, under the leadership of Dmitrii Shcheglov, focused its entire theatrical activity on dramatizing turning points in the Russian revolutionary movement, including peasant rebellions, the Decembrist uprising, and the revolution of 1905.100 May Day and November 7, the anniversary of the revolution, emerged as the two most important Soviet holidays. Festivities were held all over the two capitals and included worker districts. Factory workers joined festive processions in their districts, and humble local clubs were decorated with posters and slogans to mark the holiday. Celebrated with musical concerts, plays, and agitational works on the street, these pivotal holidays gave club members an opportunity to celebrate within their own neighborhoods, claiming local spaces as their own.101
If revolution loves the theater, what happens when the revolution ends? This question was posed by many observers and activists in amateur theaters as the Civil War was coming to an close. The very proliferation of stages, so astonishing to all observers, was in part a product of the upheaval. Not only were actors trying on new social roles, but impoverished neighborhoods were attempting to fashion some sort of entertainment with the meager resources at their disposal. The innovative staging methods devised during the Civil War made a virtue of necessity. They were flexible, allowed adjustment to local conditions, and could provide "educational" entertainment for the audience. They required neither long rehearsal time nor elaborate costumes to be effective.
Just as some political theorists saw the deprivations of the Civil War as a shortcut to real Communism, some cultural activists saw these emergency measures as the roots of a new revolutionary theater. But for others, the end of the Civil War meant that cultural work could return to "normal." Trade unions and factories could begin to build clubhouses with well-appointed stages. Theater circle members could devote themselves to classes and rehearsals. The result would be an amateur theater of higher quality, devoted to aesthetic drama.
A debate on the future of amateur theaters was aired in Petrograd's main cultural journal, Zhizn' iskusstva (The Life of Art) in March 1921, just as the Tenth Party Congress was meeting to announce the measures known as the New Economic Policy. The literary critic and social commentator Viktor Shklovskii published an article assessing the remarkable proliferation of theater circles. He suggested that their continued existence and expansion was not a sign of cultural creativity but rather a symptom of a serious social malaise. The rush to the theater was a kind of psychosis, an attempt to avoid the real difficulties of life. "These millions of circles should not be closed—one cannot forbid people their ravings. As a sign of sickness, they should be studied by sociologists. But we cannot use them to construct a new life."102
This passionate condemnation of amateur theaters in their current form drew a heated response from the Petrograd cultural activist Adrian Piotrovskii, who had emerged as one of the most important advocates of improvisational theater during the Civil War. Piotrovskii drew precisely the opposite conclusion from the proliferation of theater groups. They came from the best and strongest sources of Soviet life, from Red Army soldiers and Communist youth. They did not turn from life but rather embraced it, staging trials, mysteries, and improvisations. In addition, they had a fundamentally different aim than amateur theater groups of old; their purpose was not to imitate life but rather to transform it. "Now let one thing be clear," intoned Piotrovskii. "The thousands of theater circles spread across the republic are militant signs of how daily life is being revolutionized [revoliutsionizirovanie byta]."103
This exchange marked a controversy over the cultural legacy of the Civil War. To what extent would the innovations of the revolutionary period live on in a transformed social and political climate? Where would the nation look to find a path toward "new life"? While Shklovskii searched for some return to "normal," to a world not infested with innumerable theater circles, Piotrovskii and others like him hoped to continue and advance Civil War cultural strategies into an era of social compromise.
* * *
1. "Polozhenie o 1-m Vserossiiskom s"ezde po raboche-krest'ianskomu teatru," Iskusstvo kommuny 16 February 1919: 5.
2. P. Kogan, "Sotsialisticheskii teatr v gody revoliutsii," Vestnik teatra (henceforth cited as VT) 40 (1919): 3–4. See also N. Krupskaia, "Glavpolitprosvet i iskusstvo," Pravda 13 February 1921, cited in Pedagogicheskie sochineniia, v. 7 (Moscow: Izdatel'stvo Akademii pedagogicheskikh nauk, 1956), 56.
3. See, for example, P. A. Markov, The Soviet Theatre (London: Victor Gollancz, 1934), 137; Ilya Ehrenburg, People and Life, 1891–1921, trans. Anna Bostock (New York: Knopf, 1962), 321–22; and Serge Wolkonsky, My Reminiscences, trans. A. E. Chamot (London: Hutchinson, 1924), vol. 2, 219–20.
4. V. Tikhonovich, "Tochki nad i," VT 66 (1920): 2.
5. Eduard M. Dune, Notes of a Red Guard, trans. and ed. Diane Koenker and Steven Smith (Urbana: Illinois University Press, 1993), 40.
6. Klub, kak on est' (Moscow: Trud i kniga, 1929), 47.
7. N. R, "Teatr vchera i segodnia," Krasnaia gazeta 7 September 1919.
8. On the elements of aesthetic drama, see Richard Schechner, Performance Theory, rev. ed. (New York: Routledge, 1988), esp. 166–72.
9. On social drama, see Victor Turner, "Social Dramas and Ritual Metaphors," and "Hidalgio: History as Social Drama," in Dramas, Fields, and Metaphors (Ithaca: Cornell University Press, 1974), esp. 37–41, 99, 123. For explicit comparisons between aesthetic and social drama, see Schechner, Performance Theory, 171–72, 187–93, 232.
10. Viktor Turner, On the Edge of the Bush: Anthropology as Experience (Tucson: University of Arizona Press, 1985), 300–901, cited in Schechner, Performance Theory, 191.
11. Nikolai L'vov, "Tiaga na stsenu," VT 56 (1919): 8.
12. Lars Kleberg, " 'People's Theater' and the Revolution: On the History of a Concept before and after 1917," in A. A. Nilsson, ed., Art, Society, Revolution: Russia 1917–1921 (Stockholm: Almqvist and Wiksell International, 1979), 191–92.
13. P. M. Kerzhentsev, Tvorcheskii teatr, 5th ed. (Moscow: Gosizdat, 1923), passim.
14. V. Tikhonovich, "Chto takoe raboche-krest'ianskii teatr," VT 12 (1919): 3.
15. "Gorodskoe soveshchanie po voprosu o raboche-krest'ianskom teatre," Zhizn' iskusstva 3 April 1919.
16. V. Tikhonovich, "Nashi raznoglasiia," VT 45 (1919): 4–5; idem, Samodeiatel'nyi teatr (Vologda: Oblastnoi otdel gosizdata, 1922), 32.
17. N. L['vov], "Nedelia o s''ezde," VT 43 (1919): 4.
18. "Deklaratsiia fraktsii kommunistov po voprosu proletarskogo teatra, vnesennaia na sessiiu Soveta 28/IX 1919g." GARF, f. 628 (Tsentral'nyi dom narodnogo tvorchestva im. N. K. Krupskoi), op. 1, d. 2, l. 57.
19. Ibid.
20. L['vov], "Nedelia o s''ezde," 4–5; N. L['vov], "S''ezd po raboche-krest'ianskomu teatru," VT 44 (1919): 2.
21. "Vtoraia sessiia Soveta raboche-krest'ianskogo teatra," VT 57 (1920): 4–5.
22. See, for example, its report "Samodeiatel'nye teatral'nye kruzhki v 1919 i 1920 godakh," GARF, f. 2313 (Glavpolitprosvet), op. 1, d. 134, ll. 3–4.
23. V. V. Tikhonovich, Narodnyi teatr (Moscow: V. Magnussen, 1918); idem, Samodeiatel'nyi teatr (Vologda: Oblastnoi otdel Gosizdata, 1922); Kleberg, " 'People's Theater' and the Revolution," 192–94; V. Filippov, Puti samodeiatel'nogo teatra (Moscow: Gosudarstvennaia akademiia khudozhestvennykh nauk, 1927), 57. Filippov credits Tikhonovich with the popularization of the term.
24. See, for example, Adrian Piotrovskii's denunciation of liubitel'skii teatr in Krasnoarmeiskii teatr: Instruktsiia k teatral'noi rabote v Krasnoi Armii (Petrograd: Izdatel'stvo Petrogradskogo voennogo okruga, 1921), 4. In this study, I have translated liubitel'skii as "dilettantish" and samodeiatel'nyi as "amateur."
25. For an examination of the conflicting meanings of samodeiatel'nost', see James von Geldern, Bolshevik Festivals, 1917–1920 (Berkeley: University of California Press, 1993), 28, 126–27, 146, 209, 216; Lynn Mally, Culture of the Future: The Proletkult Movement in Revolutionary Russia (Berkeley: University of California Press, 1990), 36–44, 232–39; Rosalinde Sartorti, "Stalinism and Carnival: Organisation and Aesthetics of Political Holidays," in Hans Günther, ed., The Culture of the Stalin Period (New York: St. Martins, 1990), 57–61.
26. There is a large literature on spontaneity and consciousness as important categories in Bolshevik political theory. For a discussion especially relevant for cultural history, see Katerina Clark, Thee Soviet Novel: History as Ritual (Chicago: University of Chicago Press, 1981), 22–24.
27. On clubs during the Revolution and Civil War, see Gabrielle Gorzka, Arbeiterkultur in der Sovietunion: Industriearbeiter-Klubs, 1917–1929 (Berlin: Arno, 1990), 67–168.
28. "Spravochnyi otdel," Krasnaia gazeta 5 September 1918.
29. Mally, Culture, 183–91.
30. Mikhail Zverev, "Klub ili obshchestvennyi ochag?" Griadushchee 5/6 (1919): 23; M. N. Belokopytova, "Kluby rabochikh podrostkov," Vneshkol'noe obrazovanie 1 (1919): 57.
31. E. Lozovskaia, "O raionnom moskovskom Proletkul'te," Vestnik zhizni 6/7 (1919): 140–141; R. Myshov, "O rabote v proletarskikh klubakh," Gorn 2/3 (1919): 41–42; "Spektakli," Krasnaia gazeta 21 June 1918.
32. Evgeny Dobrenko, The Making of the State Reader: Social and Aesthetic Contexts of the Reception of Soviet Literature (Stanford: Stanford University Press, 1997), 44.
33. Questionnaire from the central Proletkult organization, 1 March 1919, RGALI, f. 1230 (Proletkul't), op. 1, d. 430, l. 1.
34. "Rabochaia kul'tura," Krasnaia gazeta 1 June 1918.
35. See an overview of select Moscow clubs conducted in 1919, "Kul'turno-prosvetiter-naia rabota moskovskogo proletariata," Gorn 5 (1920): 71–80.
36. See, for example, O. Zol', "Petrogradskii latyshskii rabochii teatr," Griadushchee 9 (1918): 22.
37. On factory theaters, see Eugene Anthony Swift, "Theater for the People: The Politics of Popular Culture in Urban Russia" (Ph.D. diss., University of California, Berkeley, 1985), 170–83; on clubs affiliated with the trade union movement, see Victoria E. Bonnell, The Roots of Rebellion (Berkeley: University of California Press, 1983), 328–34.
38. Gorzka, Arbeiterkultur, 152.
39. A. S. Bulgakov and S. S. Danilov, Gosudarstvennyi agitatsionnyi teatr v Leningrade, 1918–1930 (Moscow: Academia, 1931), 19–20.
40. Richard Schechner, "Toward a Poetics of Performance," in Performance Theory, 166.
41. Pavel Marinchik, Rozhdenie komsomol'skogo teatra (Leningrad: Iskusstvo, 1963), 19–20; "Kul'turno-prosvetitel'naia rabota moskovskogo proletariata," 79.
42. A. Mgebrov, "Proletarskaia kul'tura," Griadushchee 2/3 (1919): 23.
43. "Konkurs na 'Dvorets rabochikh,' "; "Dvorets rabochikh," Iskusstvo kommuny 19 January 1919.
44. A. Petrov, Narodnye kluby (Moscow, 1919), 11, cited in V. Khazanova, Klubnaia zhizn' i arkhitektura kluba, v. 1 (Moscow: Rossiiskii institut iskusstvoznaniia, 1994), 18.
45. Ibid., v. 1, 23–25, 32.
46. P. Knyshov, "O rabote v proletarskikh klubakh," Gorn 2/3 (1919): 42. See also E. Ozovaia, "O raionnom moskovskom Proletkul'te," Vestnik zhizni 6/7 (1919): 141; V. Mitiushin, "Tesnyi kontakt," Gudki 6 (1919): 16.
47. N. Noskov, "Na putiakh kul'turnogo stroitel'stva," Zhizn' iskusstva 101, 21 March 1919.
48. Sheila Fitzpatrick, The Commissariat of Enlightenment: Soviet Organization of Education and the Arts under Lunacharsky (Cambridge: Cambridge University Press, 1970), 142–46.
49. See Robert Thurston, The Popular Theatre Movement in Russia, 1862–1919 (Evanston, Ill Northwestern University Press, 1998), 279–80.
50. "Ob''edinenie rabochikh klubov," Krasnaia gazeta 2 September 1919.
51. A. A. Gvozdev and A. Piotrovskii, "Petrogradskie teatry i prazdnestva v epokhu voennogo kommunizma," in Istoriia sovetskogo teatra, v. 1 (Leningrad: Gosizdat, 1933), 225. On Red Army theatrical activity, see also "Teatral'naia samodeiatel'nost' v Krasnoi Armii," in A. Z. Iufit, ed., Russkii sovetskii teatr, 1917–1921 (Leningrad: Iskusstvo, 1968), 314–23; Mark von Hagen, Soldiers in the Proletarian Dictatorship (Ithaca: Cornell University Press, 1990), 111–14; and Elizabeth Wood, Performing Justice in Revolutionary Russia: Agitation Trials, Society, and the State (Berkeley: University of California Press, forthcoming), ch. 1 and 2.
52. Von Geldern, Bolshevik Festivals, 131–32; Iufit, Russkii sovetskii teatr, 320–21; David Zolotnitskii, Sergei Radlov: The Shakespearean Fate of a Soviet Director (Luxembourg. Harwood Academic Publishers, 1995), 5.
53. On the Red Army studio, see von Geldern, Bolshevik Festivals, 124–33. On the Baltic fleet studio, see Gvozdev, "Petrogradskie teatry," in Istoriia sovetskogo teatra, 227.
54. V. Smyshliaev, "O rabote teatral'nogo otdela moskovskogo Proletkul'ta," Gorn 1 (1918): 53.
55. "Kul'tumo-prosvetitel'naia rabota moskovskogo proletariata," Gorn 5 (1920): 71; D. Zolotnitskii, Zori teatral'nogo Oktiabria (Leningrad: lskusstvo, 1976), 296–344.
56. V. A. Razumov, "Roi' rabochego klassa v stroitel'stve sotsialisticheskoi kul'tury v nachale revoliutsii i v gody grazhdanskoi voiny," in Rol' rabochego klassa v razvitii sotsialisticheskoi kul'tury (Moscow: Izdatel'stvo "Mysl'," 1976), 20.
57. On the cultural policies of trade unions during the Civil War, see Gorzka, Arbeiterkultur, 84–94. On Petrograd and Moscow groups, see Bulgakov and Danilov, Gosudarstvermyi agitatsionnyi teatr, 19.
58. Filippov, Puti samodeiatel'nogo teatra, 16.
59. Marinchik, Rozhdenie, 16.
60. On fights between PUR and Narkompros over control of army clubs, see Wood, Performing Justice, ch. 1.
61. Gvozdev, "Petrogradskie teatry," in Istoriia sovetskogo teatra, 251–54.
62. There are numerous such petitions in the Proletkult archive. See, for example, the deliberations of the Proletkult Central Committee on 15 February 1919, rgali, f. 1230, op. 1, d. 3, l. 15.
63. Bulgakov and Danilov, Gosudarstvennyi agitatsionnyi teatr, 14–15.
64. Iufit, Istoriia sovetskogo teatra, 81, 341; Zolotnitskii, Zori, 331–56, passim.
65. Von Geldern, Bolshevik Festivals, 119–22.
66. See Dmitrii Shcheglov, "U istokov," in U istokov: Sbornik statei (Moscow: VTO, 1960), 11–179, esp. 11, 26, 59, 127.
67. On the repertoire in urban amateur theaters before the revolution, see E. Anthony Swift, "Workers' Theater and 'Proletarian Culture' in Pre-Revolutionary Russia," Russian History 23 (1996): 80–82.
68. Bulgakov and Danilov, Gosudarstvennyi agitatsionnyi teatr, 15–18; "Samodeiatel'nye teatral'nye kruzhki v 1919 i 1920 godakh," GARF, f. 2313, op. 1, d. 134, ll. 3–3 ob. See also Tikhonovich, Samodeiatel'nyi teatr, 37–38.
69. "Spisok p'es, redomendovannykh moskovskoi repertuarnoi komissiei," GARF, f. 2306 (Narkompros), op. 2, d. 357, ll. 38–38 ob.
70. Piotrovskii, Krasnoarmeiskii teatr, 14–16.
71. Kerzhentsev, Tvorcheskii teatr, 70–71.
72. This was a common position in the Proletkult. See V. I. Lebedev-Polianskii, ed., Protokoly pervoi Vserossiiskoi konferentsii proletarskikh kul'turno-prosvetitel'nykh organizatsii (Moscow: Proletarskaia kul'tura, 1918), 46.
73. Fitzpatrick, The Commissariat of Enlightenment, 146–47.
74. P. Kogan, "Sotsialisticheskii teatr v gody revoliutsii," VT 40 (1919).
75. "Samodeiatel'nye teatral'nye kruzhki v 1919 i 1920 godakh," GARF f. 2313, op. 1, d. 134, 1. 3 ob. On Belaia, see Dictionary of Russian Women Writers (Westport, Conn.: Greenwood Press, 1994), 72–73.
76. Kerzhentsev, Tvorcheskii teatr, 68. For an example of the play's staging, see "Teatry i kinematografy," Krasnaia gazeta 1 May 1918.
77. Sergei Orlovsky, "Moscow Theaters, 1917–1941," in Martha Bradshaw, ed., Soviet Theaters, 1917–1941 (New York: Research Program on the ussr, 1954), 13; Z. G. Dal'tsev, "Moskva 1917–1923: Iz vospominanii," in U istokov, 185–88.
78. Platon Kerzhentsev, Revoliutsiia i teatr (Moscow: Dennitsa, 1918), 48.
79. N. L'vov, "P'esa ili stsenarii," Vestnik rabotnikov iskusstv, 2/3 (1920): 53–54.
80. Iufit, Russkii sovetskii teatr, 358–60; Daniel Gerould, "Gorky, Melodrama, and the Development of Early Soviet Theatre," Theatre Journal 7 (Winter 1976): 33–44; L. Tamashin, Sovetskaia dramaturgiia v gody grazhdanskoi voiny (Moscow: Iskusstvo, 1961), 271–80.
81. On agitki in general see Tamashin, Sovetskaia dramaturgiia, 104–33; Robert Russell, Russian Drama of the Revolutionary Period (London: Macmillan, 1988), 34–36; idem, "The First Soviet Plays," in Robert Russell and Andrew Barratt, eds., Russian Theatre in the Age of Modernism (London: Macmillan, 1990), 152–54.
82. Victoria Bonnell, Iconography of Power: Soviet Political Posters under Lenin and Stalin (Berkeley: University of California Press, 1997), 200.
83. Peter Brooks, The Melodramatic Imagination: Balzac, Henry James, Melodrama, and the Mode of Excess (New Haven: Yale University Press, 1976), 11–12, 16.
84. For Pavel Arskii's play, see V. F. Pimenov, ed., Pervye sovetskie p'esy (Moscow: Iskusstvo, 1958), 489–99.
85. P. Bessal'ko, "Kamenshchik," Plamia 33 (1919): 2–7.
86. Iufit, Russkii sovetskii teatr, 320–21.
87. On Lodka before the revolution, see V. I. Vsevolodskii-Gerngross, Russkaia ustnaia narodnaia drama(Moscow: Izdatel'stvo Akademii Nauk SSSR, 1959), 38–79 and Elizabeth Warner, The Russian Folk Theatre (The Hague: Mouton, 1977), 127–40; for its use in the Red Army, see Piotrovskii, Krasnaia armiia, 7; and von Geldern,Bolshevik Festivals, 124–25.
88. On Petrushka's long history and many transformations, see Catriona Kelly, Petrushka: The Russian Carnival Puppet Theatre (Cambridge: Cambridge University Press, 1990), esp. 179–211. On agitki with Petrushka themes, see Tamashin, Sovetskaia dramaturgiia, 124.
89. "Konkurs Proletkul'ta," VT 70 (1920): 19.
90. "Vecher Uolta Uitmena," Krasnaia gazeta 24 July 1918; Iufit, Russkii sovetskii teatr, 320; Tamashin,Sovetskaia dramaturgiia, 44, 52.
91. Iufit, Russkii sovetskii teatr, 318–20; "Khronika Proletkul'ta," Proletarskaia kul'tura 15/16 (1920): 79.
92. N. Karzhanskii, Kollektivnaia dramaturgiia (Moscow: Gosizdat, 1922), 56–57, 73.
93. M. Slukhovskii, Ustnaia gazeta kak vid politiko-prosvetitel'noi raboty (Moscow: Krasnaia zvezda, 1924), 14, 22–27, 35.
94. See the memoirs of Terevsat leader M. Ia. Pustynin in Iufit, Russkii sovetskii teatr, 185–87; Zolotnitskii, Zori,195; and J. A. E. Curtiss, "Down with the Foxtrot! Concepts of Satire in the Soviet Theatre of the 1920s," inRussian Theatre in the Age of Modernism, 221. For the evolution of this form in the 1920s, see ch. 2.
95. On the prerevolutionary origins of agit-trials, see Julie Cassaday, "The Theater of the World and the Theater of the State: Drama and the Show Trial in Early Soviet Russia," (Ph.D. dissertation, Stanford University, 1995), 51–53; on its roots in folk drama, see Warner, The Russian Folk Theatre, 51.
96. On agit-trials in the army, see Wood, Performing Justice, ch. 1; Tamashin, Sovetskaia dramaturgiia 57–60; and von Geldern, Bolshevik Festivals 110, 172, 181.
97. Karzhanskii, Kollektivnaia dramaturgiia, 66–69; Wood, Performing Justice, ch. 1.
98. There is a large literature on these festivals. See A. I. Mazaev, Prazdnik kak sotsial'nokhudozhestvennoe iavlenie (Moscow: Nauka, 1978), esp. 277–300; V. P. Tolstoi, ed., Agitatsionno-massovoe iskusstvo: Oformlenie prazdnestv, 2 v. (Moscow: Iskusstvo, 1984); von Geldern, Bolshevik Festivals; Christel Lane, The Rites of Rulers (Cambridge: Cambridge University Press, 1981), 153–73; Rosalinda Sartorti, "Stalinism and Carnival"; Richard Stites, Revolutionary Dreams: Utopian Vision and Experimental Life in the Russian Revolution(New York: Oxford, 1989), 79–100. For a contemporary assessment, see Gvozdev, "Petrogradskie teatry," inIstoriia sovetskogo teatra, 264–90.
99. Tolstoi, ed., Agitatsionno-massovoe iskusstvo, v. 1, 108–9.
100. Tamashin, Sovetskaia dramaturgiia, 30.
101. Tolstoi, ed., Agitatsionno-massovoe iskusstvo, v. 1, 49–50, 71, 95, 115; "Teatral'nye studii," Gudki 2 (1919): 26; Gorzka, Arbeiterkultur, 143–45.
102. Viktor Shklovskii, "Drama i massovye predstavleniia," Zhizn' iskusstva, 9–11 March, 1921. This oft-cited article is reprinted in Viktor Shklovskii, Khod konia: Sbornik statei (Moscow: Gelikon, 1923), 59–63.
103. A. Piotrovskii, "Ne k teatry, a k prazdnestvu," Zhizn' iskusstva 19–22 March, 1921. This article is also republished in an excerpted form in Adrian Piotrovskii, Za sovetskii teatr! Sbornik statei (Leningrad: Academia, 1925), 25–26.
## 2
## Small Forms on Small Stages
THE CONCLUSION of the Russian Civil War brought new challenges to all those engaged in constructing a Soviet culture. Efforts to rebuild the shattered economic base of the country, begun in 1921, meant that there were substantially fewer state funds available for cultural projects. Optimistic plans to construct new club buildings and new stages for amateur theaters were put off for several years. In addition, in order to infuse life into the economy, the Soviet government allowed limited capitalist enterprise to start up again in the form of the New Economic Policy (NEP). This program was not only an economic threat to those who hoped for the rapid victory of socialism, but it also posed significant cultural dangers in the minds of many affiliated with amateur theater. Urban commercial life quickly revived, offering entertainment opportunities ranging from imported films to boulevard literature. Many Bolsheviks, as well as their allies from the prerevolutionary people's theater movement, saw these aspects of urban life as a threat to the creation of a healthy and edifying culture for the masses.
Even though funds were tight, state agencies at both the national and the local level realized the importance of launching cultural campaigns to win the population over to the Soviet cause. The Soviet state of the 1920s might be best called an "enlightenment state," in the words of Michael David-Fox, so focused was it on transforming the consciousness of its citizens. "By the early 1920s," writes David-Fox, "Bolshevik leaders across factional lines came to portray cultural transformation, educational work, and the creation of a Bolshevik intelligentsia as pivotal to the fate of regime and revolution."1 Non-professional theatrical groups, which had proven themselves effective propagandists during the Civil War, emerged as an important arena in the struggle to educate the broad population to become enthusiastic Soviet citizens.
While sponsoring agencies had high expectations for amateur stages, they rarely provided new funds or resources. How, under these circumstances, could amateur stages best fulfill their pedagogical tasks? In this chapter I investigate one answer put forward initially by a select group of cultural activists in Petrograd and Moscow. They proposed abandoning conventional repertoires for club stages altogether, replacing them with the improvisational methods that had already gained ground during the Russian Civil War. These methods were called "small forms" (malye formy), yet another redefinition of a prerevolutionary term. Before the Bolsheviks came to power, the theater of small forms referred to music halls and vaudeville theaters.2 Now small forms meant agit-trials, satirical sketches, and living newspapers.
The agitational theater of small forms satirized Soviet enemies and praised Soviet heroes. It was used to impart lessons on how Soviet citizens should live—what books they should read, what their hygienic habits should be, and how they should relate to the Soviet regime. These methods were in part a response to the new cultural offerings of the NEP era. Sponsors envisioned the healthy entertainments of Soviet clubs, among them amateur theatrical works, as an alternative to "decadent" forms of commercial culture made possible by NEP's restricted capitalism. The theater of small forms was intended to be engaging; many skits used humor and buffoonery. Some groups consciously employed elements of NEP culture in order to interest viewers, giving them what they hoped was a healthy socialist twist. Thus this didactic theater was intended both to educate and entertain.
Limited cultural funding facilitated the turn to small forms. These improvisational methods were for the most part not dependent on well-appointed stages and expensive production techniques. Performers often played characters very much like themselves and therefore did not require expensive costumes or make-up. Because small forms were conceived as a method to bring performers closer to audiences, the humble performance spaces of clubs, with their small or non-existent stages, were not the impediment that they would have been for more conventional productions. Some groups, like the Moscow Blue Blouse theater, commanded its followers to eschew complex costumes and sets, turning necessities into virtues. "Blue Blouse rejects all beautiful, realistic sets and decorations," read one manifesto. "There will be absolutely no birch trees or little rivers."3
Yet even while small forms gained ground, there were heated debates surrounding the eventual direction of amateur theater. Were these improvisational forms an end in themselves? Did they point toward the development of new kinds of "big" theater—new plays and operas with a revolutionary thematic and presentational style? Or were they temporary, stop-gap measures for poorly equipped club stages and poorly trained amateur actors, measures that could be phased out as conditions improved? These questions were debated within state agencies and trade union bureaucracies, among theatrical professionals interested in amateur work, and inside club theatrical groups themselves. Certainly, some club stage advocates believed that if a new, distinctive style of Soviet theater ever was going to take shape it would emerge from the shabby environs of workers' clubs and not from the glittering stages of the old theaters.
#### The Turn to Small Forms
The promotion of small forms on local stages came initially from local agencies in Petrograd/Leningrad and Moscow at the onset of the New Economic Policy. Their advocacy of a unique, politicized repertoire for amateur theaters found favor among select local groups. By 1923, the Communist Party endorsed the idea that club cultural work should be directly relevant to political and economic campaigns, a pronouncement interpreted as an endorsement of this direction on the amateur stage. The national trade union leadership soon followed suit. Although not all amateur theater groups abandoned prerevolutionary works and full-length Soviet plays in the early years of NEP, there was a definite swing to small forms. Local living newspaper groups were an especially popular manifestation of this aesthetic turn.
The first efforts to formulate a unique form of amateur performance took shape in Petrograd at the end of the Civil War. Petrograd Politprosvet activists devised a special organizational framework within clubs to structure cultural work, a model they called the "united artistic circle" (edinyi khudozhestvennyi kruzhok—often called by its acronym, ekhk). The goal was to make the life of the club revolve around Soviet festivals. Adrian Piotrovskii, the head of the Petrograd Politprosvet division, envisioned the united artistic circle as a way to continue and expand the agitational, propagandistic direction of club theatrical work begun during the Civil War. He believed that many amateur theater circles had already made theatrical festivals a central focus of their work.4 In Piotrovskii's view, the united artistic circle simply described and clarified the direction that theatrical work had already taken in factory and neighborhood clubs.5 The central idea was to make all club artistic and educational circles work together toward the same goal. Music groups, physical education circles, and literary circles would all participate in the creation of a mass theatrical "happening" (deistvo). The newly emerging festival days of the revolution were the perfect occasion for these events. All would contribute to a celebration of Bloody Sunday, May Day, and the October Revolution. This new direction emerged from popular tastes, wrote Piotrovskii in 1921: "There is no pull toward the 'spectacle' [zrelishche] of professional theater; instead, popular theatrical events, popular performances have burst forth into light."6
In Piotrovskii's description, local theatrical activity was spontaneously moving toward club festivals; the united artistic circle was a method to better coordinate that activity. Piotrovskii's focus was on spontaneity, local creativity, and (although he did not say so directly) local resources. As Katerina Clark has noted, the united artistic circle displayed in striking clarity the newly constrained economic circumstances of NEP.7 What was proposed was in essence a bargain-basement festival, removed from the main squares of the city to the humble confines of club stages and their immediate neighborhoods.
In addition to these fiscal attractions, the united artistic circle marked a significant turn toward greater uniformity and control. What was proposed was nothing less than a complete transformation of amateur theatricals. No longer would clubs devote themselves to performing classic or contemporary plays. Rather, they would focus their work on a festive calendar of revolutionary celebrations. Moreover, some descriptions of the united artistic circle significantly curtailed the element of spontaneity. It was the club's political circle that gained the responsibility for drawing up the guidelines and taking control of artistic work.8 Grigorii Avlov, part of the cultural division of the Petrograd Politprosvet and editor of the most widely distributed book on united artistic circles, made the central role of the political group even more pronounced. The model he chose was that of a factory, where all sectors cooperated in the creation of a final product. The function of central planner was given to the political circle.9
It is not hard to understand the appeal of the united artistic circle for government and trade union organizations. In its ideal form, all club cultural activity would propagate the principles and goals of the revolution, supervised by political organs within the club. Petrograd Politprosvet workers enthusiastically embraced this new direction. By 1921, the political education division had opened a "Central Agitational Studio" that experimented with forms of collective improvisation. Headed by a former member of Gaideburov's theater, V. V. Shimanovskii, this studio grew out of the agitational work of railroad unions during the Civil War.10 The following year Shimanovskii's studio served as the basis for a special provincial Politprosvet division in charge of amateur theater. It sent trained workers out to monitor and direct the work of city clubs and tried to coordinate their activities. Soon the city's education division formed a special Home of Amateur Theater (Dom samodeiatel'nogo teatra), again under Shimanovskii's guidance, which provided a central performance stage where local clubs could show their work.11
Politprosvet institutions in Petrograd served as a focal point for the new agenda of amateur theaters. They provided a training ground for the methods of the united artistic circle and designed repertoire for amateur stages. To take just one example, two participants in the Shimanovskii studio, Iakov Zadykhin and Vladimir Severnyi, composed scripts ofinstsenirovki to mark the 1923 May Day festival.12 The Home of Amateur Theater helped to coordinate festival celebrations, noting which dates of the Red Calendar deserved commemoration and drawing up lists of suitable repertoire.13
In Moscow, it was the Proletkult organization that initially called for a more focused agitational approach in club theaters. At the end of 1921, the city's main Proletkult studio adopted a plan that rejected the plays of Ostrovsky in favor of "improvisations, living newspapers, and agitational work."14 In a long overview of the city's club activities, the leader of the Proletkult club division, Raisa Ginzburg, called for an end to standard repertoire on club stages in favor of a more didactic direction.15 By late 1922, the Proletkult began to advance these ideas in terms very similar to cultural workers in Petrograd. They called for the creation of a "united studio of the arts" (edinaia studiia iskusstv). All sections of the club would serve a common purpose, focusing on the harmonious interaction of club members to achieve the improvement of proletarian life.16 As a consequence, "theater work inevitably will turn toinstsenirovki focused on the burning issues of the day, on agitki, evenings of scenarios, revolutionary cabarets, living newspapers, theatricized courts, etc."17 The Proletkult plan included a long list of classes in art and political education.
The turn to small forms got a real boost in April 1923, when the Twelfth Communist Party Congress determined that clubs should become active centers of mass propaganda designed to encourage the creative abilities of the working class.18 While the resolution also addressed the necessity of leisure-time activities in clubs, many national and local institutions interpreted it as a call for better-coordinated agitational work from club cultural circles. Accordingly, they began to formulate programs for club activity that followed the general direction set by the Politprosvet division in Petrograd: they embraced agitational, educational work as the main focus of club activities. Although there were differences in emphasis, all these programs shared basic assumptions for amateur theaters. No longer would their primary task be to practice and disseminate conventional theatrical skills and repertoire. Rather, their main goal was to serve the club community as a whole and to provide highly politicized and topical activities.
The national trade union bureaucracy gave a resounding endorsement to agitational methods in club artistic circles.19 The cultural division determined that all artistic groups, including theater circles, would no longer be cut off from the general activities of the club. Instead, they should direct their efforts toward agitation and education. Evoking the words of the Party congress, the national trade union convention on club work meeting in the spring of 1924 voted to tie the work of all club circles to political education aimed at the broad masses. All club activity was to be unified into a single complex plan, embracing politics, professional life, and culture.20
The most radical proposal for small forms was devised by a group of Moscow Politprosvet workers in the summer of 1923.21 They rejected any form of theatrical work that was set off from general club activity. They even rejected common nomenclature like "theatrical studio" or "theatrical circle," proposing instead the new name of "action circles" (deistvennye kruzhki) or "action cells" (deistvennye iacheiki) that would initiate mass activities in clubs.22 The supporters of this position, including the long-time theater activist Nikolai L'vov and club instructors M. V. Danilevskii and Vitalii Zhemchuzhnyi, formed the Association of Action Circle Instructors, which sought a radical transformation (some would say annihilation) of theatrical work in clubs. The motto of the action circle was: "Stop play acting and start organizing life."23 To show how this method marked a break from past approaches, the Moscow proponents suggested that the words "spectacle" (spektakl'), "actor" (akter), and "play" (igra) no longer be used. Instead, they would be replaced by "presentation" (vystuplenie), "performer" (ispolnitel'), and "action" (deistvie).24 This linguistic shift was meant to show that artistic work in clubs would be fundamentally different from professional artistic work. Clubs should never aim to imbue professional artistic techniques among their students. According to a manifesto written by Zhemchuzhnyi, "In its organization and methodology, artistic work in clubs should not be different from other club work. For this reason, the goals and methods of club artistic work is fundamentally different than the professional arts. The goal of clubs should never be to establish a professional artistic studio."25
Agencies that supported the turn to small forms created courses to train instructors in the new techniques. The Home of Amateur Theater in Leningrad offered classes for theatrical workers.26 The Moscow Proletkult designed a number of short training sessions, some with the collaboration of the Association of Action Circles.27 In 1924 the central Moscow trade union organization opened the Theatrical-Artistic Bureau, which approved a suitable repertoire for club stages and also attempted to provide technical assistance and coordinate the work of local groups.28
The shift to small forms brought fresh resources to club theaters. Instructors trained in the new methods put themselves at the disposal of amateur theaters. Government agencies began to publish collections of short plays and sketches that could be performed by local theaters. In addition, a number of new journals devoted at least in part to amateur stages and their repertoire made their appearance, including The Worker Viewer (Rabochii zritel') and The New Viewer (Novyi zritel'), both located in Moscow, and Worker and Theater (Rabochii i teatr)from Leningrad. These publications, along with the older Life of Art (Zhizn' iskusstva), gave extensive space to performances on club stages. Two other important journals with coverage of amateur stages began in 1924: Workers' Club (Rabochii klub) and Blue Blouse(Siniaia bluza), both of which published sample works that local groups were encouraged to alter for their own purposes.
These publications are filled with what one might call "conversion stories," illustrating the switch from conventional repertoire to small forms within individual clubs. The tales have a similar structure: a club theater labored away with heroic prerevolutionary plays or silly melodramas, accomplishing very little. Performances were rare and inadequate. They had nothing to do with other events in the life of the club. Then suddenly the direction changed. From this point on the theater circle began to produce works for club events and festivals, becoming happily integrated into the life of the club. The impetus for change was not uniform in these tales; sometimes they came from the trade union sponsoring club work, sometimes club leaders intervened, and sometimes amateur actors themselves took credit for the reorganization. The Moscow Transit Workers' Union decided to alter the methods of a regional club, inviting a director trained in small forms to take charge. Almost overnight, the repertoire changed from Ostrovsky plays to celebrations honoring International Women's Day. At the Northern Railroad Club, the chief administrator dispensed with the old expert in charge of theater. Members of a club at a Moscow metal-working factory decided to adopt the new methods on their own, since they could not afford to pay an instructor.29
Not all amateur theater circles embraced small forms. One instructor who had been to a training course in Moscow and altered his club's work according to the "new course" met resistance from viewers. They did not like the instsenirovka Bourgeois in Hell (Burzhui v adu) and asked for a play from the prerevolutionary repertoire.30Some groups produced mixed repertoires. The Nekrasov People's Home in Leningrad performed a homemadeinstsenirovka together with two acts of Denis Fonvizin's The Infant, a standard of the prerevolutionary Russian repertoire. According to one worker correspondent attending the event, the Fonvizin play did not compare well with the improvised work.31 The Moscow Perfume Factory "Freedom" ended up with two theater groups, one that followed an old repertoire and another determined to devise new works. Those who had chosen the new direction called themselves "conscious workers" (soznatel'nye) as opposed to their "dilettantish"(liubitel'skie) former colleagues. The supporters of plays were still harboring hurt feelings at the Red October Club a year after theater members switched to small forms.32 The changes sometimes caused considerable bad blood. Members who resisted small forms were pronounced guilty of "dramatism" (teatral'shchina).33 Old-style drama circles had "crippled the worker from the bench," in the opinion of an activist from a Moscow metalworkers' club, "evoking from him either the most mundane dilettantism or turning him into a bad actor."34
Not surprisingly, the rift between those embracing small forms and those who preferred a more conventional repertoire was often interpreted as a split between the young and old in clubs. In early NEP, as during the Civil War, young people were the main users of clubs and the primary participants in theater workshops. A 1924 survey of large clubs in Moscow determined that the approximately ninety percent of those in artistic circles were young people. "Workers clubs are becoming youth clubs," the author concluded.35 Both advocates and foes of small forms saw the marked transformation of club theater as a reflection of its youthful composition. For those in favor of the shift, it was a sign of the radical and experimental nature of young people. "Youth instinctively turns away from old forms," wrote the trade union cultural leader, Emil Beskin, "believing them to be a vessel for old feelings and thoughts."36 Critics saw things differently. They felt that the disjointed, iconoclastic repertoire was a sign of youthful inexperience and maintained that older workers were not interested at all.37 These radically juxtaposed positions lent an aggressive undertone to discussions about club performances. Small forms were not simply an aesthetic direction; they implied political choices as well. For those in favor of the shift, their opponents were guilty of holding suspect beliefs. "We have noticed that the drama circle used to suffer from 'petty bourgeois theater,' " observed worker correspondents in the journal New Viewer. "But it has made a good recovery from that illness."38
#### Festivals and Celebrations
Small forms tied the work of theater groups to the dates of the "Red Calendar," a fluid list of celebrations designed to supplant Russian religious festivals and transmit the values of the new regime.39 The two most central dates were May Day and the anniversary of the revolution. Other celebrations might include January 9, to mark the revolution of 1905; January 15, the death date of German revolutionary leaders Rosa Luxemburg and Karl Liebknecht; January 21, the anniversary of Lenin's death; February 23, Red Army Day; March 8, International Women's Day; March 12, the anniversary of the fall of tsarism; March 18, the day of the Paris Commune; and the anti-religious festivals of Komsomol Easter and Komsomol Christmas.40 New celebrations and special occasions were added to the list at the local level. Amateur theaters also participated in efforts to publicize local election campaigns, to celebrate the founding dates of clubs, and to entertain at their sponsoring trade union's annual convention. Participants and club leaders frequently complained that this extensive list of festivities made performance groups struggle from campaign to campaign, without time for adequate preparation or rehearsal.41
Festivals were designed as participatory events, allowing as many people as possible a chance to perform—resulting in mixed-media events that could last all through the night. Many clubs opened the festivities with "An Evening of Remembrances." Workers with a personal link to the holiday being celebrated, such as the revolution of 1905 or a May Day celebration before the revolution, got a chance to tell their stories.42 The entertainments could include recitations of favorite poems, along with musical interludes by the club choir and orchestra. Art circles were active making banners, posters, and decorations. Sometimes physical education groups got involved, presenting feats of skill for the audience.43 The British writer Huntley Carter, who visited Moscow in the early 1920s, offered this account of a May Day celebration:
The performance and room decorations and inscriptions were clearly designed to usher in May Day, just as a certain church service is designed to usher in New Year's Day in England.... The club room, which was crowded to suffocation, was festooned with evergreens, draped with red and hung with portraits of Lenin, Trotsky, and Marx, and with inscriptions.... The exhibition was an improvised revue designed to emphasize the importance of May Day and its implications. One might call it a family affair in honour of the October communistic revolution.44
The festive family spirit was also noticed by Soviet observers. One witness to a Petrograd May Day celebration determined that "there were no spectators—everyone was a performer, a participant."45
Leaving the cramped spaces of clubs behind, some theater groups took their performances out into the streets, mounting trucks and platforms or using nearby squares as their performance space. The 1925 May Day celebration on Vasileostrovskii Island in Leningrad was a three-day affair, with club circles taking part in outdoor festivities the first day and returning to their club stages for the second and third.46 Festival performances were sometimes very elaborate, staged like small versions of the huge events of the Civil War years. The Trekhgornyi Factory in Moscow acted out scenes from John Reed's book Ten Days That Shook the World for the 1924 October celebration. Divided into fifteen different parts, the dramatization involved some two hundred participants.47 The Lenin Workers' Palace in Moscow staged an instsenirovka for the 1924 October celebration that portrayed the history of the revolution from the start of the First World War to the Bolshevik takeover. The cast included three hundred civilians and one hundred soldiers.48
Annual trade union conferences were a popular venue for agitational performances. Meeting in large halls, the conferences provided a forum where theaters from several clubs could collaborate and perform for a captive audience of trade union delegates. The Red Woodworkers' Club of Moscow decided to act out their union's charter at the annual union convention in 1924. Their goal, according to one viewer, was "to give the rank-and-file trade union member a chance to familiarize himself with the dry language of the union charter by artistic means." This observer was especially impressed by the scene in which the membership rules were enacted. Doors on stage opened wide to include all workers, regardless of sex or nationality. The doors closed quickly, however, when former members of the tsarist police, capitalists, or priests tried to enter. These undesirable elements, dressed up as "wolves in sheep's clothing," were excluded from the union's ranks.49
With the emergence of journals and publications aimed at amateur stages, it was not necessary for club stages to devise their own texts. If they found the work appealing, club circles could use published scripts for a variety of celebrations. One example of this new material is a 1924 work, Hands off China (Ruki proch' ot Kitaia),published in the journal Blue Blouse. Using rhymed couplets, this text depicts the victimization of the Chinese peasantry by Western and Japanese imperialists. Only the Soviet Union intervenes to help the people of China assert their independence. "China, squeeze imperialism with your claws!" intones the character representing the Soviet Union. "Hands off China! Let's have a [Chinese] October!"50 This short play, which offered a Soviet interpretation of international events and celebrated the October revolution simultaneously, was a popular choice for October festivities in Moscow in 1924. Five different club theaters used it to mark the holiday.51
Groups that embraced small forms judged instructors by their ability to stage successful festival performances. The Timiriazev Club in Moscow underwent a long search for an instructor who could meet the rigorous schedule of Soviet celebrations. After unsuccessful experiences with two instructors from the Moscow Art Theater and one from the Meyerhold Theater, members finally turned to the Moscow Politprosvet department for help. It sent the action circle advocate, Nikolai L'vov, who in short order turned the group toward agitational productions. Shortly after L'vov was hired, the circle staged a self-created work for the 1924 May Day celebrations, In Honor of May Day(K vstreche pervogo maia).52
"Hands off China" (Ruki proch' ot Kitaia). A Blue Blouse performance. Harvard Theatre Collection, The Houghton Library. Reproduced with permission.
The use of improvisational methods allowed club theaters to prepare performances in a hurry—one of the chief advantages of small forms. Participants in the Ivan Fedorov Printers' Club in Moscow decided just two days before Christmas that they would like to stage an anti-religious event. They brainstormed together and came up with a plot tracing how a worker convinced his wife to use icons as fuel for the samovar. Forty-eight hours later, aninstsenirovka entitled "A Purpose for Icons" was staged for the club community. Introduced by a lecture on the scientific creation of the world and a short performance by the club's living newspaper group, the improvised text was presented to a largely sympathetic young audience.53
These efforts at creation from below and broad participation gave a new spin to the word samodeiatel'nost',which for many club participants came to mean "homemade." Those who enthusiastically supported the turn to small forms claimed that it empowered the participants to try their own hands at cultural creation. Surveying the preparations for the 1923 October festival in Leningrad, Grigorii Avlov praised the level of independent work. Some groups were using prepared texts, which they altered to suit their purposes. Others had works that were written by individual group members or group leaders. The most impressive circles were those who created their works collectively, revealing the creative potential of the theater of small forms.54
#### Investigations of Daily Life: Agit-Trials and Living Newspapers
Festivals by their very nature were special occasions, separate from the quotidian world. In addition, most early Soviet festivals had an overt political meaning; they honored turning points in the revolutionary struggle and marked important moments for the new state. As such, they did not have a direct influence on workers' daily lives. For many advocates of small forms, marking festivals was not enough. They believed that daily habits and social interactions (in Russian byt) could be analyzed and transformed through performance. Two styles of agitational theater were particularly suited to topics of daily life: the agitsud, or agit-trial, and the living newspaper.
Agit-trials emerged as an important form of amateur performance during the Civil War, when they were used to praise the heroes and excoriate the enemies of the revolution.55 In the 1920s, their subject matter was broadened significantly. Workers who would not join a union were put on trial to demonstrate proper behavior at the factory. Social problems were acted out in trials of alcoholics and prostitutes. A wide range of trials also examined historical issues, like "The Trial of Those Responsible for the First World War."56 Some agit-trials were easily integrated into festival celebrations; "The Trial of Father Gapon," for example, was often staged during the events marking the anniversary of 1905.57
One kind of agit-trial aimed for verisimilitude, attempting to follow the structure and atmosphere of a real trial as much as possible. These events enumerated the violated paragraphs of the legal code. Both prosecuting and defense attorneys took part, as well as witnesses for both sides. The Leningrad club activist Grigorii Avlov insisted that the more realistic the trial, the more effective it would be. It was better to try the hooligan than hooliganism. His own "Trial of Hooligans" was a case in point. The two young men on trial for misbehavior, Pavel Iudin and Ivan Karnauchov, were introduced with detailed information about their character and appearance. In the course of the trial, viewers learned about the social circumstances and political beliefs of the accused.58
But not all trials followed these guidelines. Some, such as "The Trial of Bourgeois Marriage," put abstract concepts on the witness stand. Even inanimate objects could win their day in court. In "The Political Trial of the Bible," the good book itself had a speaking role. In the course of questioning, the Bible was forced to admit to many inconsistencies about its authorship and contents. Unnamed worker witnesses asked tough questions that the Bible had difficulty answering: If people were made of clay, then how could they burn? How could two of all animals in the world fit on Noah's Ark with enough food for forty days? The Bible's one defender, an illiterate peasant girl, could say only that she believed the Bible was the word of God. She could not defend its contents in detail because she had never read them.59 While the basic structure of the event bore some resemblance to a trial—with a prosecutor, a defendant, and witnesses—clearly this scenario was intended more to make the Bible look foolish than to give the audience a sense of life in the courtroom.
Trials also varied according to their predetermined nature. Many texts for agit-trials were published as elaborate scripts, with the speeches of witnesses and attorneys set down and the verdict preordained. Although the printed speeches (some quite detailed) were presented as a base for improvisation, and although the texts sometimes offered a variety of possible verdicts, the general outcome was unmistakable. In the many agit-trials of hooligans, for example, the young offenders were never allowed to escape without punishment. Here the drama of the event was in the performance, not the outcome.
However, other trials were impromptu events, sometimes written by local participants and casting members of the audience as witnesses and jurors with the power to come to their own conclusions. This kind of trial could sometimes bring surprising results. A factory circle from Petrograd playing in the countryside in the summer of 1923 had a very difficult time convincing village audiences to convict the character of a corrupt priest.60 Isaac Babel's controversial collection of stories, Red Cavalry, was put on trial in a Moscow club in 1926. Although the speeches against the book were passionate, Babel himself made an appearance to argue in his defense. The assembled crowd not only acquitted Babel, but also judged his work to be a real service to the revolution.61
Agit-trials in the 1920s aimed to stamp out old habits and inculcate new ones. This purpose is strikingly evident in one trial written by the Moscow advocate of action circles, Vitalii Zhemchuzhnyi, called "An Evening of Books" (Vecher knig). This humorous work lampooned popular reading tastes, which tended toward religious works, detective novels, romantic potboilers, and Tarzan adventures. The judge in the trial, the American socialist author Upton Sinclair sent the work of the prerevolutionary romantic writer Anastasia Verbitskaia to the archive; he allowed God, a character in the drama, to go free since no one paid attention to him anymore; and he determined that all of Nat Pinkerton's books should be burned "to the last letter." Only Tarzan managed to escape judgment, since he escaped from the courtroom during a brawl. By the end, the audience was presented with a new, wholesome reading library, including Soviet adventure stories and, of course, the works of Upton Sinclair himself.62
Agit-trials also attempted to teach audiences new standards of public health. "The Trial of a Midwife Who Performed an Illegal Abortion" showed the risks of the abortion and the content of the Soviet abortion law.63 Venereal disease was a very common theme. In a Moscow typographers' club, a hypothetical case was brought for a hearing; a man had infected his wife and children with syphilis. The court had a magnanimous verdict in this case. Because of his ignorance, the man should be cured and educated, not punished.64 The traditional Russian scourge of alcoholism was another common theme. In one work, "The Trial of the Old Life" (Sud nad starym bytom), a frequently drunk worker was put on trial for beating his wife and keeping her from political work. Not only was he condemned to jail at the end of the trial, but so were the small shop owner and tavern keeper, who kept him supplied with alcohol.65
"An Evening of Books" (Vecher knigi). The photograph depicts Komsomol members fighting with the representatives of bad literature, including God, Anastasia Verbitskaia, and Nat Pinkerton. Bakhrushin State Central Theatrical Museum. Reproduced with permission.
While many works for amateur theaters lacked major parts for women, agit-trials offered them starring roles, frequently as the chief obstacles to the new life. Agit-trial titles, which read like the names of bad mysteries, give a sense of women's "crimes": "The Trial of the Woman Who Did Not Take Advantage of the October Revolution," "The Trial of the Illiterate Woman Worker," and "The Trial of the Mother Who Deserted Her Child."66 One script, "The Woman Worker Who Did Not Attend General Meetings," examined six offending women who avoided trade union gatherings. Although the women had very different reasons for their truancy, ranging from fear to boredom, they were all charged with cultural and political backwardness.67
Club life itself was a theme for agit-trials, as activists considered ways to draw more workers to club activities. "The Trial of the Old Club" featured a surprising list of witnesses. One by one the library, the piano, and the drama studio came forward to accuse the club of poor organization, poor equipment, and lack of space. In the end, the old club broke down in tears because its members were preparing to leave for a new building.68
By claiming the small conflicts of everyday life as a proper subject matter, agit-trials could potentially turn nasty, singling out members of the audience for shame and censure. "The Trial of Six Workers at the Red October Factory" focused on actual factory workers who were deemed to have undermined cultural work. Their specific "crimes"— spreading rumors about the club and preferring an evening at a pub to wholesome entertainment—were presented in some detail in the proceedings. Although the accused were allowed to defend themselves, their comments were limited to lengthy admissions of guilt and promises of reform.69 This use of agit-trials as a form of urban charivari, or samosud, would become much more common during the First Five-Year Plan, when many forms of agitational theater took an aggressive stance toward the audience.
Even more common than agit-trials were living newspapers, which began to appear in great numbers on amateur stages after the phenomenal success of a professional living newspaper circle from Moscow, Blue Blouse (Siniaia bluza).70 The unusual name stemmed from the group's basic costume element, a blue work shirt, which is called a bluza in Russian. Begun at the Moscow School of Journalism in 1923, Blue Blouse was headed by the energetic Boris Iuzhanin, who had prepared living newspapers for the Red Army during the Civil War. The group attracted the writing skills of some of the country's finest satirical authors, including Sergei Tretiakov, Argo (Abram Markovich Gol'denberg), and Vladimir Maiakovskii. According to theater historian František Deak, Blue Blouse was "the largest movement in the history of theatre in which the avant-garde participated."71
Presenting the news of the day in a vibrant mix of satirical songs, lively posters, dances, and pantomime, the Blue Blouse living newspaper soon won an enthusiastic audience in Moscow. A performance typically opened with a parade of the "headlines," followed by from eight to fifteen short vignettes on topics ranging from international affairs to local complaints about factory management.72 The actors amended their simple work clothes with exaggerated props to identify the role they were performing, such as a top hat for a capitalist or a large red pencil for a bureaucrat. Since the troupe did not need sophisticated stages or lighting, it could perform almost anywhere. In the early 1920s Blue Blouse played in clubs, cafeterias, and factory floors throughout Moscow and Moscow province.73
In 1924, Blue Blouse was incorporated into the Moscow Trade Union cultural division. With this increased financial backing, it was able to start its well-known journal, Blue Blouse, which stayed in publication until 1928. The journal published scripts for living newspapers, offered advice on costumes and staging, and printed scores for Blue Blouse songs. It served as an inspiration for local groups wanting to stage their own performances. They either adopted the printed material whole cloth or used it as a model for their own creations. After a popular Blue Blouse presentation at the Moscow Elektrozavod factory in 1924, drama club members voted to start their own living newspaper. "In the future the living newspaper 'Electrical Current' (Elektrotok) will direct its work toward the productive life of the factory and will only make limited use of the material in Blue Blouse," the club members determined.74 It was only one of many living newspapers formed from below, including the "Red Tie" (Krasnyi galstuk), the "Red Sting" (Krasnoe zhalo), "Red Coil" (Krasnaia katushka), and the "Red Scourge"(Krasnyi bich).75
A Blue Blouse troupe. Harvard Theatre Collection, The Houghton Library. Reproduced with permission.
The success of Moscow's Blue Blouse sparked emulation in Leningrad. In early 1924, the Leningrad Trade Union Organization founded its own professional living newspaper group, Work Bench (Stanok). The agitational theatrical studio of the city's Politprosvet organization began a living newspaper as well.76 Both of these professional circles served a similar function to Blue Blouse in Moscow; their performances encouraged the creation of living newspapers at the club and factory level, with colorful names like "Our Pencil," and "Factory Whistle."77
By the summer of 1924, living newspapers gained national backing as a potent agitational form. The national convention of club workers, sponsored by the trade union organization and Glavpolitprosvet, endorsed living newspapers as a "method of agitation and propaganda, serving the political, productive, and domestic [bytovye] tasks of the proletariat."78 By the following year, living newspapers were widespread on amateur stages in both capitals. The Leningrad journal Worker and Theater printed a separate page devoted to local groups, and one observer determined that "some people are talking about the 'triumphant procession' of living newspaper through factories and plants."79 In a survey of one hundred city clubs, the Moscow-based journal Worker's Club revealed that living newspapers were the most popular form of performance, drawing in larger audiences than any other theatrical events.80
Advocates gave living newspapers an almost magical power to attract and educate audiences. One viewer was supposedly so taken by a lively Blue Blouse performance that he forgot to leave the club as usual for the local bar. His wife was astonished when he came home sober.81 An enthusiastic supporter from the Moscow Construction Workers' Union claimed that Blue Blouse performances always generated lively discussions in workers' barracks and even inspired some workers to find out more about the event portrayed on stage. "When a worker sees [the British politician] Curzon or some other important political figure, he is very interested in this guy who has played such a funny role on stage. And afterwards, even if he hasn't understood everything, he begins to look around to find out more."82
It is not hard to see living newspapers' attraction for cultural activists and for many viewers. Their agitational and didactic function was selfevident, because they always included information about contemporary national and international politics. They were often humorous, offering comic relief to viewers used to much drier political fare. Posters, slides, and sometimes even film clips were included, providing information and visual stimulation. They also were easily changed and amended to meet local conditions; a section for "letters to the editor" could contain complaints and commentary about concrete problems at the club or work site.
A Blue Blouse demonstration at the Club of Foreign Office Employees. Harvard Theatre Collection, The Houghton Library. Reproduced with permission.
Those who wrote the texts for living newspapers attempted to put the traditions of folk theater to use in a new way, thus providing a familiar entree for urban audiences. A director from the Moscow Blue Blouse theater called his group a balagan, a Russian folk theater, and claimed that for precisely this reason Blue Blouse was comprehensible to worker audiences.83 Parts of the performance were called a "raek," or peepshow verse, an important element of fairground theater. Many living newspapers included a role for a carnival-like barker, called arupor or raeshnik, who introduced the action and tied the various small skits together. Works included humorous four-lined rhymed ditties, or chastushki, an integral part of Russian urban and rural folk culture since the nineteenth century. The naughty star of Russian puppet theater, Petrushka, sometimes made an appearance.84 The living newspaper script "Give us a New Life" (Daesh' Novyi Byt), published in 1924 and performed by several groups in 1924 and 1925, contained a part for a ryzhii, the traditional red-haired Russian clowns.85
Not only did living newspapers attempt to transform folk culture for agitational use, they also drew on familiar forms of urban popular culture as well. In one programmatic statement, the editors of Blue Blouse insisted that a living newspaper should be performed at a fast pace and look like a film to the audience.86 Music was a crucial part of a Blue Blouse performance, and participants drew on melodies popular in cafes and nightclubs. The opening march for all Blue Blouse performances, "We Are the Blue Blousists," was set to the tune of the popular song "We Are the Blacksmiths."87 Organizers made no excuses for this eclectic approach; they maintained that their task was to reach out to unsophisticated viewers, appealing to their emotions as well as to their intellect.
The rapid expansion of Blue Blouse influence won it enemies as well as fans. Some critics resented the group's professional standing, saying it was a theater for workers, not by workers. They charged that professional groups had influenced the creation of local living newspapers from the top down. "Where was the self-activation [samodeiatel'nost'] in this?" wondered one Leningrad union activist.88 Other viewers did not believe that satire was an effective educational tool. "Does anything remain in workers' heads but laughter after a Blue Blouse performance?" asked a trade union leader.89
In contrast to Blue Blouse and other professional troupes, local living newspapers were located close to their audience and could address issues of direct relevance to the life of the club and the community. Participants included concrete details about problems in their union or work place and could shape their repertoire to fit any special event. Local groups could also aim their performances toward particular audiences. Komsomol-based newspapers addressed the problems of youth; a living newspaper sponsored by educational workers included a section called "Teachers during NEP."90 It was the thrill of immediacy that supposedly won audiences over to living newspapers, with the most contemporary material drawing the most interest. At an anniversary celebration for the Central Sales Workers' Club in Moscow, a living newspaper addressed the accomplishments and failures of different club circles, much to the delight of the audience. Such methods brought about a strong bond between the performers and the viewers, insisted one advocate.91
An amateur performance at the Moscow Sales Workers' Club. Bakhrushin State Central Theatrical Museum. Reproduced with permission.
However, homebred living newspapers often faced complaints for following the patterns designed by professionals too closely. The vast majority of local groups imitated Blue Blouse models, claimed the passionate advocate of action circles, Vitalii Zhemchuzhnyi. They did not involve the entire club in their preparation and drew their performers from drama circles alone. Because participants did not write their own work, performances did not address the specific needs and interests of the audience.92
Because they attempted to educate and entertain simultaneously, both amateur and professional living newspaper groups found themselves caught up in what Denise Youngblood has called the "entertainment or enlightenment debate" that dominated the Soviet film industry in the 1920s. One faction of Soviet filmmakers looked to foreign films for their inspiration, trying to use popular elements—suspense, slapstick humor, happy endings—to draw in viewers while imbuing their films with a Soviet message. They were opposed by those who argued that such concessions to capitalist methods undermined the films' socialist content. Instead, directors should concentrate on edifying and didactic topics that could not be mistaken for bourgeois products.93
In their efforts to integrate elements of urban commercial culture, living newspaper groups found themselves in a similar situation to filmmakers who sought to make their movies entertaining. In the words of one advocate, living newspapers were "a political genre, but a light and cheerful kind that is good for workers who want to relax and have a good time."94 Sanctimonious critics complained that this was precisely the problem. Their work looked too much like bourgeois cabaret. "The performance has a theatrical character, along the lines of "Crooked Jimmy' [a prerevolutionary cabaret], but the copy is worse than the original," complained a representative from the Central Art Workers' Union.95 One worker correspondent claimed that the living newspaper "Red Scourge" offered an inauthentic analysis of hooliganism among Soviet youth—the hooligans looked more like Parisians than "our own." He also did not like the dances, the many jokes, and the frivolous portrayal of the Komsomol girl. The entire performance was "sloppy, superficial, and thoughtless."96 "If this is theater," wrote two worker critics in response to a Blue Blouse performance, "then it is an unhealthy kind.... Put an end to bourgeois elements in workers' theater."97
Living newspapers, both professional and amateur, attempted a difficult balancing act. They tried to merge political information of national and local relevance, presenting it in a witty and engaging style. Complaints focused on their inability to meet all of these requirements—their approach to politics was too frivolous, their coverage of local problems too superficial, and they made too many concessions to commercial popular culture in their efforts to engage audiences. Eventually, these charges would coalesce into a blanket condemnation of "Blue Blouse-ism," a term of approbation that criticized living newspapers' satirical approach to serious issues and their episodic presentational style. It was a critique that attacked the central premise of most living newspapers— that elements of urban popular culture were appropriate conduits for political education.
#### Small Forms and the Avant-Garde
With their episodic structure and non-naturalistic staging, amateur productions employing small forms bore a distinct similarity to the theatrical experiments of the avant-garde. This similarity was hardly surprising, since avant-gardists were deeply involved in what they saw as an exciting attempt to create a utilitarian theater that would not simply observe life but also change in.98 Students of Vsevolod Meyerhold, the doyen of avant-garde theater, took positions in amateur theater groups. Writers from the Left Front of the Arts (LEF) wrote scripts for Blue Blouse and amateur living newspaper circles.99 The involvement of these artists brought precious resources to struggling amateur stages; in addition, it linked the fate of small forms to the avant-garde.
Meyerhold was a significant supporter of the club theater of small forms. In early 1924 he opened a "Club Methodological Laboratory" as part of his training courses in Moscow. It prepared directors for club theaters, reviewed manuscripts for club performances, devised plans for mass spectacles, and debated the aesthetic principles of club theatrical work.100 The workshop endorsed a variety of small forms, including instsenirovki and living newspapers. Improvisations took precedence over ready-made works. "Only when there is no material and no time to prepare any should the circle turn to plays," read one instruction.101
Training club instructors was the studio's most important task. Meyerhold enrolled students with a wide range of experience. Some, like Olga Galaktionova, came with modest credentials, having worked only briefly in the provinces before arriving at the studio. Others had considerable experience and would go on to make big names for themselves in the world of Soviet culture. These included Nikolai Ekk, who was simultaneously a student in Meyerhold's directors' studio and the Meyerhold Theater. He eventually turned from theater to movies and directed the first Soviet sound film, The Road to Life (Putevka v zhizn') in 1931. Another studio participant with a similar trajectory was Ivan Pyr'ev, who began in the Proletkult and also studied acting and directing with Meyerhold. He turned to film already in the middle of the 1920s and by the next decade began to build a reputation as a director of filmed musical comedies. His biggest hits included Tractor Drivers (Traktoristy) andKuban Cossacks (Kubanskie kazaki), which gained a reputation as one of the worst examples of Stalinist culture during the Khrushchev era.102
Meyerhold's laboratory acted as an employment facilitator, fielding requests from local clubs for experienced instructors. "We have almost no money," read one query from a local chemical factory. "However, we do have a good group of young workers who are interested in the the-ater."103 Instructors were sent to clubs that were willing to pay their salaries. By late 1925, the club laboratory had provided instructors for more than forty different Moscow clubs. The sponsors included trade unions, the Komsomol, the Red Army, and five clubs under the control of the GPU, the Soviet secret police.104 Most students worked in the capital, but some received placements in the provinces. Pyr'ev ended up in Ekaterinburg in 1924.105
Wherever they went, these instructors encouraged the turn to small forms. One Moscow club, which had been performing Ostrovsky plays, changed entirely with the arrival of Goltsov, a Meyerhold student. Participants began to use different kinds of material, like satirical stories from the journalThe Godless (Bezbozhnik) as the basis for their improvised work. Ispolnev, another Meyerhold student, took charge of the club at the large Trekhgornyi textile mill in Moscow, where he concentrated his efforts on staging club festivals.106
The Meyerhold laboratory encouraged instructors to write their own material. In a 1926 questionnaire, Nikolai Ekk boasted that only five percent of the work he staged in his four years of club activity was written by others; the rest he devised himself, together with his students. Two of his works, The Red Eagles (Krasnye orliata) and Ky sy my were published and performed in numerous clubs. Boris Ivanter, a leader of several Moscow clubs, composed The Earth in Flames (Zemlia zazhglas') in 1924 together with his wife Vera. It tells the story of the Bolsheviks' rise to power and subsequent efforts by the bourgeoisie to subvert the revolution. This work played on numerous factory stages in 1924, including the Triangle Rubber Plant in Moscow and the Putilov factory in Leningrad.107 Nikolai Mologin devised a humorous work making fun of bureaucratic language and the Soviets' new-found love of acronyms. Entitled "Upruiaz" (an acronym for "Uproshennyi russkii iazyk"—Simplified Russian Language), it proposed that individuals should start using abbreviations and acronyms in their daily speech. In addition, Mologin's witty script suggested that people should leave off Russian's complicated grammatical endings and communicate only with word roots, a change that might have been welcomed by non-Russians struggling with the language. It was staged at several clubs, including the Central Printers' Club, where Mologin was in charge.108
Studio members also organized large-scale festivals that drew in a number of amateur theater groups. The laboratory worked together with the Red Army to prepare plans for a "Red Stadium." Although the structure was never built, a number of large outdoor events using amateur participants were staged at its proposed site at the Lenin Hills in Moscow.109 In addition, studio participants planned and executed elaborate neighborhood festivals. One such event in Moscow celebrated the history of the Sokol district Communist Party. Using masses of raw material sent by the Communist Party Committee, two studio members devised a scenario that examined significant events from the revolution until 1925, incorporating data and statistics about the Sokol region. It was staged by Nikolai Ekk to mark the 1925 anniversary of the revolution.110
Through their efforts, Meyerhold students believed they were bringing sophisticated professional techniques to a broad audience. For them, the Meyerhold Theater was the main inspiration for small forms on amateur stages. One statement from the Meyerhold laboratory determined that those wishing to compose compelling living newspapers should look to Meyerhold's production of The Trust D.E. for inspiration.111 This play, based on a novel by Ilia Ehrenburg, caused a sensation in the world of professional theater with its use of jazz, stylish dance numbers, and physical education routines. Theoretical statements issued by the laboratory charged that other groups supporting small forms, particularly the action circles led by Zhemchuzhnyi, were making use of Meyerhold's methods without giving him credit.112
Artists from Moscow's action circles and the Politprosvet division in Leningrad, however, believed that Meyerhold had gotten his inspiration from amateur theaters, and not the other way around. "In the struggle against naturalistic and psychological tendencies, professional theaters have produced their conventions of heroism and buffoonery, their synthetic methods integrating music, song, and dance, under the influence (pod znakom) of amateur theaters," declared Adrian Piotrovskii.113 Stefan Mokulskii, from Leningrad's State Institute of the History of Art, concurred. Professional theater is challenged in each historical epoch by amateur theatrical forms—and this was precisely what was happening in Soviet Russia. The proletariat was creating its own forms of art, daily life, and knowledge within workers' clubs and Komsomol circles, he determined.114
Given the affinities between the theater of small forms and the avant-garde, is it possible to find a constituency among urban youth and working people for experimental theater? Until now, scholars have routinely rejected the idea that the avant-garde had much of a following outside the educated population, a conclusion we can find echoed in one school of 1920s criticism.115 But at least some club participants appreciated Meyerhold's methods enough to imitate them. The Kalinin Club in Leningrad put on a self-generated work called "Path to Victory" (Put' k pobede) that was obviously influenced by Meyerhold, according to one critic. The Sapronov Club in Moscow copied Meyerhold's controversial interpretation of Ostrovsky's play The Forest (Les), performing it for club members and other groups.116
In addition, worker viewers frequented the Meyerhold Theater. A number of worker correspondents (rabkory)attending a discussion about The Trust D.E. found it stimulating and insisted that it offered a very critical look at the bourgeoisie. "I talked to the workers at my factory," said one reporter. "There were some who found deficiencies but in general they praised it."117 The rabkor Iurii Kobrin was an enthusiastic supporter of Meyerhold's methods. In his short pamphlet The Meyerhold Theater and the Worker Viewer, he insisted that this innovative stage had a large and enthusiastic proletarian audience. Evidence from the Moscow agency that distributed tickets to trade unions provides some verification for this claim. The most popular tickets were those for the Meyerhold Theater, which were "snatched up and never re-turned."118 Evidently, not all worker viewers were averse to theatrical experimentation.
Eventually the ties between amateur stages and the avant-garde would work against small forms. Critics who opposed this direction in clubs used the same terms to denounce amateur performances as those reserved for the avant-garde, namely, that the approach was alien to proletarian taste and "bourgeois" in inspiration. "The worker is a realist down to his bones," wrote one observer. "He doesn't like phrases. He cannot stand abstractions. And by the way, for him the symbolic, entertaining concoctions of a closed off clique of educators, such as talking factory whistles or cake-walking money, are just so much red tape."119
#### Spontaneity and Consciousness
After watching an exhibition of amateur theaters in Leningrad in 1925, the distinguished theater historian, Alexei Gvozdev, announced, "There now can be no doubt that a new theater will be created not from above but rather from below."120 Debates about who was responsible for the making of Soviet culture—the broad population, the intelligentsia, or the state—began with the revolution and continue on in scholarship to this day. For amateur theaters of the early 1920s, it was clearly a combination. Piotrovskii was inspired by the methods he observed on the club stages of revolutionary Petrograd. He then used his considerable influence as head of the city's Politprosvet division to ensure that these methods were refined and spread to as many amateur stages as possible. Small forms were appealing to at least a segment of actors and audiences because they were inclusive, opening up the stage to a large number of people. They also held the promise of conveying "local knowledge," with spaces to insert the small victories and heartfelt needs of those creating and viewing them. Their structure, comprised of many small skits and vignettes, made the process of creation easier, opening it up to untrained club members. Thus they appeared to offer proof of the spontaneous, self-generated creativity of the masses. "Self-activity—that is the most distinctive feature of workers' theater. This is its most important difference from professional theater. Here the workers are simultaneously the carpenters, actors, and authors," effused one observer of a performance at Leningrad's Nekrasov People's Home.121
Advocates of small forms believed that they undercut the power of "bourgeois" professionals, who still dominated the world of established theater. Soviet club theater was the negation of bourgeois theater in all of its forms, insisted Grigorii Avlov of the Leningrad Politprosvet organization.122 It was called into life because professional theaters were not addressing the needs of workers and peasants to reflect upon their political situation and engage in the construction of a new political order. Because professional theaters had refused that role, a new kind of theater needed to take shape with a politicized, openly utilitarian purpose.123
Many participants in the theater of small forms went a step further and rejected the idea of professionalization altogether. They denied that club theatrical circles should try to find talented individuals who could be prepared for and promoted to the professional stage. Trade union leader S. Levman believed that the task of amateur theater was to serve the club community as a whole and to engage members in mass work, not to prepare well-trained actors.124 One union activist endorsed the unified artistic circle because this method supposedly limited the influence of specialists.125 Because small forms allowed participants to use speeches, newspaper articles, and other commonly available materials for theatrical work, it allowed energetic circles to work on their own without professional intermediaries.
Yet despite these broad claims for self-determination, the amateur theater of small forms was also created by club instructors (some with impressive theatrical credentials) and sponsoring agencies. Even when amateurs wrote their own material, they were influenced by printed scripts from a variety of official sources. Reports of festival repertoire show considerable uniformity, a uniformity that was encouraged by state organizations overseeing cultural work. The Petrograd section of autonomous theater collected information on fifty-seven clubs performing works for the 1923 May Day celebrations; thirty-three used works prepared by the city's Politprosvet division.126 A circular drawn up by the Moscow branch of the Communist Party, Komsomol, the trade union organization, and the Politprosvet division for the first anniversary of Lenin's death in 1925 offered this advice: "Prepared texts should be used only in extreme circumstances. All work should be built around the autonomous activity of club members." At the same time, though, the brochure included a list of acceptable repertoire and also determined the official slogans for the celebration.127
The sponsoring agencies for clubs, particularly local trade union divisions, were important supporters of the turn to small forms. They interceded to reorganize club management and to find instructors sympathetic to new modes of theatrical presentation. Many club training programs, from the Meyerhold laboratory to the Petrograd/Leningrad Politprosvet division, turned out instructors who wanted to stage holiday celebrations and living newspapers rather than Ostrovsky plays. Regardless of claims to the contrary by the most passionate defenders of small forms, the guidance of these instructors was a significant factor in changing local repertoires.
The amateur theater of small forms offered a vehicle for self-expression for participants, but one that was heavily supervised. The spontaneity of participants was guided and directed by a number of agencies whose job it was to ensure that the final product was imbued with "consciousness," with slogans and programs endorsed by the Communist Party and trade union and political agencies. Its overt messages were almost always politically correct; performances directed viewers to mark the holidays of the new state, sober up, and avoid boulevard literature left over from the old regime. Thus, "do-it-yourself" theater was not entirely the creation of its actors or audience. As one advocate argued, samodeiatel'nost' did not mean doing whatever one wanted. Without any sense of irony he continued, "It is most correct to speak about organized and directed self-activity."128
Despite the transparent didacticism, supporters of this style of performance worked with their audience in mind. They tried to engage and interest viewers. It was precisely this attempt to mix education with pleasure that angered many critics of small forms. They argued that club works were too amateurish, too disjointed, and too close to commercial culture to be effective aesthetic or political tools. Perhaps most important, they insisted that club forms were simply too small to articulate the grand dreams and accomplishments of the victorious Soviet revolution.
* * *
1. Michael David-Fox, Revolution of the Mind: Higher Learning among the Bolsheviks, 1918–1929 (Ithaca: Cornell University Press, 1997), 4–5, quotation 5.
2. Laurence Senelick, "Theatre," in Nicholas Rzhevsky, ed., The Cambridge Companion to Modern Russian Culture (Cambridge: Cambridge University Press, 1998), 273.
3. "Prostye sovety uchastnikam sinebluznoi gazety," Siniaia bluza (henceforth cited as SB) 18 (1925): 4.
4. Petrogradskaia obshchegorodskaia konferentsiia rabochikh klubov (Petrograd: Petrogradskii otdel narodnogo obrazovaniia, 1920), 116–17. James von Geldern, "Nietzschean Leaders and Followers in Soviet Mass Theater, 1917–1927," in Bernice Glatzer Rosenthal, ed., Nietzsche and Soviet Culture (Cambridge: Cambridge University Press, 1994), 127–48, for the broad intellectual context that encouraged Piotrovskii's ideas.
5. Adrian Piotrovskii, "Rabota Nekrasovtsev," in Sbornik instsenirovok: Opyty kollektivnoi dramaturgii(Leningrad: Izdatel'stvo knizhnogo sektora Gubono, 1924), 3–4.
6. A. I. Piotrovskii, "Edinyi khudozhestvennyi kruzhok," in Za sovetskii teatr! Sbornik statei (Leningrad: Academia, 1925), 7–9, quotation 7.
7. Katerina Clark, Petersburg: Crucible of Cultural Revolution (Cambridge: Harvard University Press, 1995), 146–47. See also James von Geldern, Bolshevik Festivals, 1917–1920 (Berkeley: University of California Press, 1993), 216–19.
8. See M. Danilevskii, "K provedeniiu prazdnika v klube," in Oktiabr' v rabochikh klubakh (Moscow: Krasnaia nov', 1923), 4–5.
9. G. Avlov, "Samodeiatel'nyi teatr i rabota edinogo khudozhestvennogo kruzhka," in G. Avlov, ed., Edinyi khudozhestvernnyi kruzhok: Metody klubno-khudozhestvennoi raboty (Leningrad: Izdatel'stvo knizhnogo sektora Gubono, 1925), 11–23. See the graphic depiction of the united artistic circle, a pyramid with the political circle at the apex, 40–41.
10. A. A. Gvozdev and A. Piotrovskii, "Rabochii i krasnoarmeiskii teatr," in Istoriia sovetskogo teatra, v. 1 (Leningrad: Leningradskoe otdelenie Gosizdata, 1933), 252.
11. G-n, "Gosudarstvennyi dom samodeiatel'nogo teatra," ZI 22 (1923): 22.
12. "Po rabochim klubam," ZI 15 (1923): 18. Both writers would have long histories as authors of agitational works.
13. "K prazdnikam revoliutsii," ZI 10 (1923): 12.
14. "Zhizn' Proletkul'ta," Gorn 6 (1922): 151.
15. R. Ginzburg, "Rabochie kluby pri novoi ekonomicheskoi politike," Gorn 6 (1922): 101–2.
16. See the records of a November 1922 Proletkult plenum, "Vserossiiskii Proletkul't," Gorn 8 (1923): 237, and Raisa Ginzburg, "Moskovskii Proletkul't v klubakh profsoiuzov," ibid., 260.
17. Iskusstvo v rabochem klube (Moscow: Vserossiiskii Proletkul't, 1924), 12. See also V. Pletnev, Rabochii klub (Moscow: Vserossiiskii Proletkul't, 1925), 52–54.
18. Kommunisticheskaia partiia Sovetskogo Soiuza v rezoliutsiiakh i resheniiakh s''ezdov, konferentsii i plenumov TsK, v. 2:1917–24 (Moscow: Izdatel'stvo politicheskoi literatury, 1970), 456–57.
19. "Rabochii klub i ego zadachi," Prizyv 5 (1924): 174.
20. "Formy i metody klubnoi raboty," Prizyv 5 (1924): 175.
21. See Nikolai L'vov's personal archive, gtstm, f. 150, notebook 1, ll. 25–29.
22. S. Lugovskoi, "Teatral'naia studiia, kruzhok ili deistvennaia iacheika?" Rabochii klub (henceforth cited as RK) 1 (1924): 12–14.
23. "Za chto my boremsia," RK 3/4 (1924): 18–22, quotation 22. According to L'vov's archive, he was the author of this manifesto.
24. "Resoliutsiia priniataia na sobranii deistvennogo kruzhka," 15 May, 1924, gtstm, f. 150, no. 17, l. 10.
25. V. Zhemchuzhnyi, "Printsipy khudozhestvennoi raboty v klubakh," n.d., rgali, f. 963 (Gosudarstvennyi teatr im. Meierkhol'da), op. 1, d. 1120, l. 23, emphasis in the original.
26. "Gosudarstvennyi dom samodeiatel'nogo teatra," ZI 22 (1923): 22.
27. "Moskovskii Proletkul't," Gorn 8 (1923): 242; "Vechernie teatral'no-instruktorskie kursy Proletkul'ta,"Rabochii zritel' (henceforth cited as RZ) 32/33 (1924): 16.
28. See, for example, A. Iudin, "Teatral'naia rabota v raionakh," RZ 18 (1924): 9; Sukhanov, "Kak mgsps budet obsluzhivat' teatrom rabochie massy," RZ 19 (1924): 6.
29. "Raionnyi klub rabochikh transportnogo soiuza," RK 5 (1924): 64; "U zheleznodorozhnikov," RK 6 (1924): 60; "K svetlomu budushchemu," RK 8 (1924): 43.
30. E. Beskin, "Na novykh putiakh," Prizyv 5 (1924): 55–56.
31. Rabkor Chustov, "V dome prosveshcheniia im. Nekrasova," ZI 27 (1924): 18. See also "Klub Krasnyi lechebnik," ZI 29 (1924): 17.
32. "Dramkruzhok parfiumernoi fabriki 'Svoboda,' " RK 6 (1924): 61; O. Liubomirskii, "Klub 'Krasnyi Oktiabr',' " NZ 6 (1925): 2. Shcheglov mentions similar splits in Leningrad; see U istokov: Sbornik statei (Moscow: VTO, 1960), 102.
33. E. Beskin, "Na novykh putiakh," Prizyv 5 (1924): 53.
34. Petr Sibartsev, "Na perelome," RZ 22 (1924): 20.
35. S. R-ch, "Molodezh' i vzroslye v klube," 2 (1924): 50. also John Hatch, "The Formation of Working Class Cultural Institutions during NEP," Carl Beck Papers, no. 806 (1990): 11–12; and Diane Koenker, "Class and Consciousness in Socialist Society," in Sheila Fitzpatrick et al., eds., Russia in the Era of NEP (Bloomington: Indiana University Press, 1991), 49–50.
36. E. Beskin, "Teatral'noe segodnia," Prizyv 1 (1924): 96.
37. V. Bogoliubov, "O klubnoi p'ese," RK 23 (1925): 73.
38. "Krasnyi luch," NZ 8 (1924): 12.
39. For assessments of the Red Calendar, see A. I. Mazaev, Prazdnik kak sotsial'no-khudozhestvennoe iavlenie (Moscow: Nauka, 1978), 349–56; E. A. Ermolin, Materializatsiia prizraka: Totalitarnyi teatr sovetskikh massovykh aktsii 1920–1930kh godov (Iaroslavl: Iaroslavskii gosudarstvennyi pedagogicheskii universitet, 1996); Daniel Peris, Storming the Heavens: The Soviet League of the Militant Godless (Ithaca, NY: Cornell University Press, 1998), 86–90.
40. See M. Veprinskii, Khudozhestvennye kruzhki i krasnyi kalendar' (Moscow: Gosizdat, 1926), 40–47.
41. O. Liubomirskii, "Bedy dramkruzhkov," NZ 30 (1924): 11.
42. "Klubnye postanovki k Oktiabr'skoi godovshchine," ZI 44 (1923): 25–26.
43. "Svedeniia," March 1925, GARF f. 7952 (Istoriia fabrik i zavodov), op. 3, d. 228, l. 29.
44. Huntley Carter, The New Theatre and Cinema of Soviet Russia (London: Chapman and Dodd, 1924), 101–2.
45. VI. S., "Ne slovo, a delo," ZI 18 (1923): 19.
46. N. P. Izvekov, "1 maia 1925 g.," in Massovye prazdnestva (Leningrad: Academia, 1926), 107.
47. Utkes, "Gotovimsia k Oktiabriu," RZ 21 (1924): 23; Andrei Shibaev, "Na puti k samodeiatel'nomu teatru," RZ 28 (1924): 6.
48. V. I. O., "Instsenirovka 'Oktiabr',' " RZ 29 (1924): 19.
49. Derevoobdelochnik no. 5994, "lnstsenirovka ustava derevoobdelochnikov," RZ 25/26 (1924): 32.
50. "Ruki proch' ot Kitaia," SB 4 (1924): 4–10, quotation 10.
51. RK 10/11 (1924): 78–79; "Khronika klubov," RZ 25/26 (1924): 35.
52. "Kak dobilis' uspekha," RZ 27 (1924): 8.
53. P. R., "Rozhdestvo v klube," Prizyv 1 (1925): 97–98.
54. G. Avlov, "Oktiabr' v klubakh," ZI 43 (1923): 15.
55. Seech. i.
56. This taxonomy of agit-trials comes from Grigorii Avlov, Klubnyi samodeiatel'nyi teatr (Leningrad: Teakinopechat', 1930), 93–94.
57. See, for example, "Sud nad gaponovshchinoi," RK 2 (1924): 6i.
58. Avlov, Klubnyi samodeiatel'nyi teatr, 94; idem, Sud nad khuliganami (Moscow: Doloi negramotnost', 1927), 10.
59. "Politsud nad Bibliei," Komsomol'skaia paskha (Moscow: Novaia Moskva, 1924), 111–28. It was performed at the Tsindel' textile factory in Moscow in 1924 (RK 7 [1924]: 38).
60. P. P., "Vpechatleniia teatral'nogo instruktora," ZI 22 (1923): 21–22.
61. V. Zhurina, "Diskussiia o khudozhestvennoi rabote," 29 (1926): 56.
62. This humorous trial is translated in James von Geldern and Richard Stites, eds., Mass Culture in Soviet Russia (Bloomington: Indiana University Press, 1995), 74–84, quotation 84. For evidence of its performance, see "Rabochii klub na Volkhovstroe," ZI 35 (1924): 20.
63. "V Tsentral'nom klube pechatnikov," ZI 8 (1924): 16.
64. K. Glubinskaia, "Sud nad sifilitikom," RZ 19 (1924): 17–18.
65. Boris Andreev, Sud nad starym bytom (Moscow: Deloi negramotnost', 1926).
66. "Sittsenabivnaia fabrika byvsh. Tsindel'," RK 7 (1924): 38; ZI 24 (1925): 21.
67. "Rabotnitsa ne poseshchaiushchaia obshchikh sobranii," RK 20/21 (1925): 20–28.
68. A. Abramov, "Kak my 'sdelali' sud," RZ 22 (1924): 22.
69. "Sud nad klubom," RK 2 (1924): 30–32.
70. On Blue Blouse, see Claudine Amiard-Chevral, "La Blouse bleue," in Le Theatre d'agit-prop de 1917 à1932, v. 1, L'URSS (Lausanne: La Cite—L'Age d'Homme, 1977), 99–109; František Deak, "Blue Blouse, 1923–1928," Drama Review 17: no. 1 (1973): 35–46; and E. Uvarova, Estradnyi teatr: miniatiury, obozreniia, miuzik-kholly, 1917–1945 (Moscow: Iskusstvo, 1983), 95–96, 100–104.
71. Deak, "Blue Blouse," 46.
72. See the Blue Blouse's instructions on the structure of a successful living newspaper, "Eshche raz—kak stroit' zhivuiu gazetu na mestakh," SB 13 (1925): 60.
73. See, for example, Andrei Shibaev, "Sinebluzniki," RZ 19 (1924): 15–16.
74. "Zhivaia gazeta 'Elektrotok,' " RZ 20 (1924): 16.
75. F. Troshin, "Nashi gazety," RZ 25/26 (1924): 33; Rabkor Svoi," 'Krasnoe Zhalo' klub imeni Libknekhta," RZ 28 (1924): 20.
76. "Khronika," Rabochii i teatr (henceforth cited as RiT) 7 (1924): 19; V. Shimanovskii, "Zadachi tsentral'noi agitstudii," RiT 2 (1924): 15.
77. " 'Nash karandash' No. 2," RiT 12 (1925): 14; "Klub im. tov. Volodarskogo," RiT 16 (1925): 15.
78. "Zhivaia i stennaia gazeta v rabochem klube," Prizyv 5 (1924): 182.
79. B. Fedorovskii, "Vyvod," RiT 11 (1925): 14; see also Sergei Spasskii, "Pis'mo iz Leningrada," NZ 11 (1925): 7.
80. E. K., "Khudozhestvennaia rabota v klube," RK 24 (1924): 20–21.
81. Sh. Ia., "Pochashche by," RZ 19 (1924): 17.
82. "Zhivye gazety v klube," Prizyv 6 (1924): 110; see also 112.
83. "Zhivaia gazeta v klube," Prizyv 6 (1924): 103.
84. On the chastushki, see Steven Frank, " 'Simple Folk, Savage Customs?': Youth, Sociability, and the Dynamics of Culture in Rural Russia, 1856–1914," Journal of Social History 25, no. 4 (1992), 723–24. On Petrushka in the Soviet context, see Catriona Kelly, Petrushka: Thee Russian Carnival Puppet Theatre(Cambridge: Cambridge University Press, 1990), 179–211.
85. "Daesh' novyi byt!" Iskusstvo v rabochem klube, 72–95.
86. "Kak rabotaet 'Siniaia bluza,' " SB 7 (1925): 4; "Na sinebluzom fronte," SB 23/24 (1925): 85.
87. "Marsh sinei bluzy," SB 2 (1924): 94–95.
88. "Zhivaia gazeta," RiT 12 (1924): 6. For the response of Blue Blouse advocates, see "Teatralizovannaia gazeta—profakterskaia i klubnaia," SB 13 (1925): 3–4.
89. Prizyv 6 (1924):106–7, 111.
90. F. M., "U prosveshchentsev," ZI 14 (1925): 16; "Zhivaia gazeta Mossel'prom," NZ 2 (1925): 17.
91. G., "Godovshchina kluba," 5 (1924): 49; K., "Zhivaia gazeta v klube Pervoi obraztsovoi tipografii," RK 3/4 (1924): 71.
92. V. Zhemchuzhnyi, "Klubnaia zhivaia gazeta," RK 15 (1925): 26–27.
93. Denise Youngblood, Movies for the Masses (Cambridge: Cambridge University Press, 1992), esp. chs. 2 and 3.
94. "Zhivaia gazeta v klube," Prizyv 6 (1924): 105.
95. Ibid. For similar comments, see R. G., "Siniaia Bluza v mestnom prelomlenii," RK 15 (1925): 31–33, where she complains about the "cafè-chantant" musical style of Blue Blouse performances.
96. Rabkov F. Lev, "Zhivaia gazeta 'Krasnyi bich,' " RZ 31 (1924): 19.
97. M. Piatnitskaia and Gavrikov, "Ob oshibkakh 'Sinei bluzy,' " RZ 20 (1924): 12.
98. B. Arvatov, "Zhivaia gazeta, kak teatral'naia forma," ZI 44 (1925): 2.
99. O. Liubomirskii, "1919—Sovrabotnik—1924," NZ 43 (1924): 10.
100. Archival holdings for the studio are in rgali, f. 963, op. 1, dd. 1088–1141. For contemporary accounts of its work, see Ligov, "Kruzhkovody, organizuites'!" RiT 13 (1924): 17; L-ai, "Na pomoshch' klubam," RZ 30 (1924): 7; "Klubnaia rabota meierkhol'dtsev," RK 12 (1924): 47; Nesterov (a participant), "God raboty," ZI 2 (1925): 12–13.
101. "Klubno-metodologicheskaia laboratoriia pri teatre i masterskikh im. Vs. Meierkhol'da," rgali, f. 963, op. 1, d. 1058, l. 3.
102. "Klubno-metodologicheskaia laboratoriia—anketa [1926]," rgali, f. 963, op. 1, d. 1103, ll. 59, 66.
103. rgali, f. 963, op. 1, d. 1094, l. 5.
104. "Klubno-metodologicheskaia laboratoriia pri teatre i masterskikh im. Vs. Meierkhol'da," 5. XI 1925, rgali, f. 963, op. 1, d. 1058, ll. 2–3; L-ai, "Na pomoshch' klubam," RZ 30 (1924): 7.
105. O. L., "Ekaterinburg," NZ 38 (1924): 13.
106. "Raionnyi klub rabochikh transportnogo soiuza," RK 5 (1924): 64; O. L., "Klub Trekhgornoi manufaktury," NZ 10 (1924): 8.
107. B. and V. Ivanter, Zemlia zazhglas'. P'esa v trekh chastiakh (Moscow: Molodaia gvardiia, 1924). For evidence of its performance, see Pravda, 12 November 1924, and RiT 6 (1924): 15.
108. Nikolai Mologin, "Upruiaz," rgali f. 963, op. 1, d. 118, ll. 3–7; "Po rabochim klubam," Pravda 12 January 1924.
109. O. L., "Klubnaia rabota na Krasnom stadione," NZ 21 (1924): 14; on the project in general, see Susan Corbesero, "If We Build It, They Will Come: The International Red Stadium Society," unpublished manuscript.
110. "Tekst instsenirovannogo otcheta Sokol'n. raikom VKP," rgali, f. 963, op. 1, d. 1104, ll. 1–12; "Klubno-metodologicheskaia laboratoriia pri teatre i masterskikh im. Vs. Meierkhol'da," 5 November 1925, rgali, f. 963, op. 1, d. 1088, l. 17; "Oktiabr' v klubakh," Pravda 11 November 1925.
111. "Proekt tezisov o 'zhivoi gazete,' " 1 December 1925, rgali, f. 963. op. 1, d. 1089, l. 80. On the impact ofThe Trust D.E., see S. Frederick Starr, Red and Hot: The Fate of Jazz in the Soviet Union (New York: Limelight Editions, 1985), 50–52.
112. "Osnovnye polozheniia doklada tt. Zhemchuzhnogo i L'vova," rgali, f. 963, op. 1, d. 1120, l. 25 ob.
113. A. Piotrovskii, "Samodeiatel'nyi teatr i sovetskaia teatral'naia kul'tura: Vmesto predisloviia," in Avlov,Klubnyi samodeiatel'nyi teatr, 9; see also S. Mokul'skii, "Pobegi novogo iskusstva," ZI 21 (1925): 13.
114. S. Mokul'skii, "O samodeiatel'nom teatre," ZI 29 (1924): 6.
115. Dobrenko, Thee Making of the State Reader, 120–21; Régine Robin, "Popular Literature of the 1920s," inRussia in the Era of NEP, 253–67, esp. 261. For a small sample of contemporary views, see V. Vsevolodskii, " 'Levyi' teatr sego dnia," ZI 6 (1924): 6; S. Prokof'ev, "Tozhe 'revoliutsionery,' " RZ 10 (1924): 1.
116. M. N-n, "Klub imeni tov. Kalinina," ZI 20 (1925): 19; "V klube imeni Sapronova," NZ 33 (1924): 15, and NZ 30 (1924): 6.
117. L. Leonov, "Rabkory o D.E.," RZ 24 (1924): 12.
118. Iurii Kobrin, Teatr im. Vs. Meierklol'da i rabochii zritel' (Moscow: Moskovskoe teatral'noe izdatel'stvo, 1926), esp. 23; A. Tsenovskii, "Chto liubit rabochii?" RiT 25 (1925): 7.
119. Aleksei Gotfrid, "Nuzhna-li p'esa rabochemu klubu," RiT 11 (1924): 15.
120. A. A. Gvozdev, "Samodeiatel'nyi teatr," Molodaia gvardiia 8 (1925): 197.
121. Sergei Tomskii, "V rabochem klube," ZI 37 (1923): 23.
122. G. Avlov, "Samodeiatel'nost' ili bezdeiatel'nost'," ZI 24 (1923): 9.
123. G. Avlov, Klubnyi samodeiatel'nyi teatr, 22–24.
124. S. Levman, "Khudozhestvennaia rabota v klube," Prizyv 3 (1924): 29.
125. E. Beskin, "O dramaticheskikh klubnykh kruzhkakh," Prizyv 2 (1924): 57–58.
126. "Podgotovka k prazdnovaniiu 1-go maia po sektsii Samodeiatel'nogo teatra," ZI 17 (1923): 19.
127. "Plan provedeniia kampanii po godovshchine smerti V. I. Lenina i '9 ianvaria' v klubakh i krasnykh ugolkakh," RGALI, f. 963, op. 1, d. 1096, ll. 1–5, quotation l.4.
128. E. Beskin, "Khudozhestvennaia rabota v klube," Prizyv 6 (1924): 34.
## 3
## From "Club Plays" to the Classics
A 1926 editorial in the Moscow journal The New Viewer asserted that Soviet theater was undergoing a fundamental transformation. During the first period of revolutionary upheaval, amateur stages had been an important force in destroying old forms and challenging professional stages. That period, however, was over. Now the battle had begun for higher quality and a new kind of professionalism, a battle that all theaters could engage in together.1 These statements in a journal aimed at a working-class audience and covering amateur stages would have been inconceivable only a few years earlier. They showed that the radical anti-professionalism of the early NEP period was on the wane.
Katerina Clark has called NEP a period of "quiet revolution in intellectual life," when some of the most distinctive elements of Soviet culture began to take shape. Particularly during the second half of NEP, the period investigated in this chapter, intellectuals began to group themselves into ever broader and ideologically more diverse organizations that bore some similarity to the professional unions formed in the 1930s. At the same time, the Communist Party and Komsomol established an important place as the sponsors of critical journals devoted to politics and culture. The result was a radical simplification of cultural debate.2
On the surface, at least, there was no such simplification process in the world of amateur theater. Instead, the offerings on club stages became more diverse in the second half of the 1920s. Small forms, which predominated in Moscow and Leningrad a few years earlier, began to share stage time with special plays written for club theaters as well as works intended for the professional stage. Yet even while the offerings expanded, the discussions about the significance of amateur theater narrowed. Two large camps took shape, one supporting the innovations of small forms and the other advocating a larger, grander style. The journals devoted to amateur stages were filled with vituperative attacks on rival directions: some denounced the incomprehensible and unsatisfying "leftism" of small forms whereas others saved their venom for the reactionary "naturalism" of those who were copying the works they saw on the professional stage.
The economic recovery of the second half of NEP changed the social context of amateur performance. After several years of hardship, state resources began flowing to factories and trade unions again. These modest increases gave them a chance to consider building or renovating spaces for performance. Paradoxically, new resources infused more animosity into the struggle between small and large forms. The shape and placement of the stage in new structures indicated what kind of performances the builders expected to see. Economic recovery brought anxieties as well, since it was achieved through the semi-capitalist mechanisms of the New Economic Policy. For some groups, this very fact made prosperity clubious. They felt it was no time to relax their revolutionary vigilance, nor to give up the agitational tactics that reminded actors and audiences of their political duties.
There was a clear (if temporary) winner in this struggle over the form and content of amateur performance. At the 1927 conference on theater sponsored by the Communist Party's Agitprop division, the organizational principles of the theater of small forms, especially the united artistic circle, came in for heavy criticism. Small forms had become too predictable to interest broad audiences, conference organizers determined. Moreover, significant changes in the repertoire of professional theaters, which had begun staging plays addressed to the revolution, made their work more appealing. Agitprop and trade union leaders recommended a new spirit of cooperation between amateurs and professionals. At the same time, however, they articulated their views in such a way as to show that amateurs would be the junior partners in this collaboration. In the words of final conference resolutions, the "theories thought up in isolated offices" that opposed amateur and professional methods had nothing to do with Marxism.3
#### Small Forms Besieged
The heyday of small forms began to wane by the middle of the 1920s. Living newspapers and improvisations faced criticism from cultural consumers, who claimed that they had become too monotonous; from club activists, who worried that they were driving older workers from clubs; and from political organizers, who were concerned about their spontaneous and uncontrolled nature. Some supporters of small forms themselves argued that they led organically to a search for more complex works that would still convey an agitational message but do so in a more compelling fashion.
Criticisms of small forms came in part from the audience. The most influential viewers were worker correspondents (rabkory), self-taught critics from the lower classes who gained positions in journals and newspapers during NEP.4 While most of their attention was addressed to the professional stage, they also evaluated amateur performances. Although they did not always agree, many worker correspondents were skeptical of the improvisational theater of small forms, believing that it was not really designed with workers in mind. One rabkor sent to review a living newspaper performance at the Volodarskii Railroad Workers' Club in Leningrad had to admit that the audience loved the show. Viewers applauded wildly and even demanded encores. The critic made short work of their enthusiasm, however. "There were almost no workers in the crowd. Perhaps that explains the success of these completely trashy numbers [chisto khalturnykh nomerov]." Another rabkor insisted that clubs filled up for festival performances only because they were free: "It is true that sometimes the viewers applaud, but their applause is not meant for the work. Rather [it is meant for] the actors and for the revolutionary content of the instsenirovka."5 It was not enough to be proud of a performance just because workers had done it themselves, determined one critic in The New Viewer. Opinions like these showed little faith in the creative abilities of the working class.6
Worker correspondents were not the only ones to complain. The well-worn stereotypes and predictability of small forms irritated many others. One trade union leader argued that audiences were fed up with standardized depictions of cruel bureaucrats, honest workers, and brainless secretaries powdering their noses.7 Moreover, the villains of small forms did not really have a contemporary ring. Such predictable stereotypes simplified social reality, insisted the club instructor Dmitrii Shcheglov. "Not all Mensheviks are bastards and not all generals blood suckers.... Finally, the working class does not always function as a collective (or rather as a mass). It has its own distinctive figures (heroes)."8
Even those who found agitational works compelling maintained that a constant diet of small forms was more than they could stomach. In scattered accounts, reporters complained that viewers found living newspapers poorly executed, mundane, and uninspiring. "They are not interesting," one audience member at the Red October Club in Moscow claimed. "We can read newspapers ourselves."9 Other small forms met similar reactions, with at least some audience members finding them disorganized and episodic. One reviewer of an instsenirovka performed in honor of Bloody Sunday in a central Moscow club called it an "arsenal of effects and buffoonery" that in no way evoked the historical drama of the march on the tsar.10
The perceived link to bad economic times was also a mark against small forms. By 1925, key economic factors began to swing sharply upward, finally reaching prewar levels by the following year. The draconian measures of early NEP that had meant harsh budget cuts for cultural institutions slowly were rescinded, bringing funds for club construction and expansion. The improvisational theater of the Civil War and early NEP had made the most of scarce resources. Many club theaters were still making do with small spaces, at the same time that club memberships were growing. A Moscow study of more than one hundred clubs conducted in late 1924 discovered that only a few could accommodate performances for more than five hundred people. Large festival events sometimes needed to be held in shifts.11 With the improving economy, trade unions and factories began to discuss new space allocations. A club building boom began in the second half of NEP, although many new structures were not completed until the First Five-Year Plan.12 First on many lists was the construction of a large auditorium with a foyer, a stage, and dressing rooms, all of which would facilitate larger and more elaborate events.13
Some of the most innovative club designs were begun in this period, including the Moscow buildings of Konstantin Mel'nikov, which remain the among most famous examples of constructivist architecture. Mel'nikov's clubs featured a stark geometric exterior design and innovative interior spaces. His Rusakov Municipal Workers' Club in Moscow, begun in 1927, made all kinds of performances possible. Large doors opened from the street into the club, so that demonstrations could move easily from inside to outside. The auditorium featured moveable walls, allowing the interior space to be divided into six separate meeting rooms to accommodate both large and small productions.14
The building boom renewed the debate about interior space that had already begun during the Civil War. How elaborate should new clubs be? What kinds of stages should they feature? Advocates of small forms wanted theater work to be completely integrated into club activities. Thus, they objected to making large auditoriums with raised stages the focal point of club structures, since this would physically separate club performances and encourage a passive audience. Architects began with elaborate stages when they made their designs, remarked one commentator in the journal Workers' Club, a strong supporter of small forms. He insisted that special theatrical spaces were not necessary for a successful club structure. Instead, it was essential to make rooms designed for meeting and discussion the focal point of new buildings.15 One important leader of the Blue Blouse living newspaper troupe, Sergei lutkevich, advocated a style of theater that would require no stage at all.16 However, neither architects nor club users were very sympathetic to this plea for a small-scale architecture. The new structures being planned, with space for costumes and scenery, undermined the minimalist aesthetics of small forms.
A more important challenge to small forms was the question of their political reliability. The open-ended nature of improvised performance put control of the final product in local hands. The ultimate decisions about content were left to the actors and directors, some of whom, in the words of one observer, were "politically illiterate."17 More often than not, shoddy preparations were the result of haste; overburdened theater circles did not have time to give their works the care that they deserved and performed them before political circles had time to monitor their content. The end result was low-quality work, which could easily be seen as a sign of disrespect for the very institutions that theaters intended to celebrate. One critic was particularly offended by a club's poor performance at a celebration of the October Revolution in 1925.18 Although charges of intentional political subversion were rare (that would come later), political overseers objected to performances that departed from goals set by the Communist Party and trade unions. For the rabkor Alexander Shibaev, the solution was more oversight, including the use of prepared texts that had been closely examined by political authorities.19
Because small forms were linked to youth, their standing in clubs was further threatened when the political reliability of Soviet young people came under increasing scrutiny. Young people, especially students, were the most vocal backers of Leon Trotsky when Joseph Stalin began to consolidate his political power base in 1923–24. Trotsky had addressed himself directly to young people in widely distributed periodicals such as Pravda and the Komsomol's Young Guard, encouraging them to see themselves as the nation's most important political and cultural constituency. He had also been a vocal advocate of workers' clubs as a place where youth could gather and discuss their experiences. When Trotsky came under fire, young people who supported him began to face political difficulties.20 The Leningrad Komsomol was censured and reorganized because of the support it gave to Zinoviev and Kamenev, two other opponents of Stalin, in 1925. In late NEP, young people emerged as the most articulate supporters of the United Opposition, the brief alliance of Trotsky, Zinoviev, and Kamenev against the party secretary.21
While one segment of Soviet youth expressed suspect political loyalties, other groups seemed more interested in having a good time. Urban youth were also the most enthusiastic consumers of Western styles in clothing, film, music, and dance. Young people formed the largest single constituency for imported movies, sometimes sneaking into upscale theaters to see their favorites over and over again. A very visible youth subculture copied the hairstyles and clothing they saw depicted in films, appearing as "flappers" and "dandies." Urban clubs and houses of culture were a cheap gathering spot where they could practice the fox-trot and other dances linked to the decadent West.22
Because young people had a reputation for disruptive behavior, their continued dominance of club life began to be seen as a serious problem. Study after study conducted in the late 1920s revealed that young people were the leading constituency in almost all areas of club work, including artistic circles. The Sickle and Hammer factory in Moscow reported that seventy percent of club participants were young. The Red Putilov Club in Leningrad had the same high number.23 "It is true," stated the central trade union leader F. Seniushkin in 1925, "that the club lives and bustles with worker youth and pioneers. But this just goes to show that the club is not yet drawing in adult workers. Instead, it attracts the Komsomol, which of course is not bad.... But in addition to youth we have a huge layer of middle-level workers for whom we have to show some concern."24
The question "Why doesn't the adult worker go to clubs?" appeared almost simultaneously in many cultural journals. The issue was important to cultural organizers because it meant that the club could not really serve as a new kind of public space that could replace the isolated world of the home. Young people were singled out as the root of the problem. Club leaders charged young people with drunkenness, disruptive behavior, and the defacement of club property, all of which drove more respectable elements from club events. Some clubs even formed volunteer militia groups (druzhinniki) to keep young people in line.25 As Joan Neuberger's innovative work on prerevolutionary St. Petersburg has shown, anxieties about cultural cohesion and loosening public control easily translated into charges of "hooliganism."26 Even before the gang rape of a peasant woman by young Leningrad workers turned hooliganism into a national obsession in the fall of 1926, fears of young people's disruptive influence in clubs filled the writings of low-level bureaucrats.27 These fears affected the discussion about small forms because young people were considered to be the most enthusiastic supporters. Older workers did not go to clubs because there was nothing for them to do there, complained one union leader. There was no quiet place for them to relax, the corridors were filled with noise and fistfights, and living newspapers did not interest them. "Bearded" viewers yearned for more serious content and complexity.28
The rhetoric of social progress also worked against small forms. The years since the revolution had brought real improvements in the lives of average workers, insisted many club activists. These positive changes included a measurable growth in the sophistication of working-class tastes. The 1925 Communist Party decree on literature addressed the "huge rise in the masses' cultural demands."29 Echoing this language, theater critics insisted that the revolution had refined the tastes of the broad masses. According to one worker correspondent, "The worker viewer has increased his theatrical and cultural level and expects a more serious and complete performance from the theater and the club."30 Following this logic, small forms were equated with small minds. Small forms were much lower on the artistic scale than large ones, insisted one Leningrad cultural worker: "The push toward 'higher' scenic forms by worker artists is completely understandable. Only such forms can convey the emotional experiences of the working class."31
With the sources at hand, it is very difficult to verify the claim that audiences in general were becoming more sophisticated and thus turning away from the simple agitational style of improvised theater. Audience studies of amateur stages were rare in the 1920s. Certainly, the most vilified genre of agitational theater, living newspapers, continued to prosper; more than thirty-five local groups performed at a Leningrad competition in 1927.32 The journal Blue Blouse provided ample evidence of the proliferation of such groups throughout the nation. In his work on Moscow clubs in late NEP, John Hatch argues that political theater lost significant ground to films as a popular form of entertainment. Nonetheless, some of the records he uses for the Sickle and Hammer factory in Moscow reveal that events featuring living newspapers, especially when they were free, continued to draw sizeable crowds.33 What one can say with more certainty is that the turn to more diverse repertoires came from within amateur theater circles themselves. At least at the outset it was not imposed by trade union, Komsomol, or Narkompros organs. It was only at the end of 1925—well after heated debates were already underway in the journals devoted to amateur theater—that the head of the art division of Glavpolitprosvet, Robert Pel'she, announced that agitational forms were beginning to play themselves out and that amateur theaters needed to search for more complex modes of performance.34 And it was not until 1927 that this view received official codification at the Agitprop conference on theater. By that time, a number of amateur circles in Moscow and Leningrad had moved away from a repertoire limited to small forms alone.
#### Club Plays
The mounting criticism of small forms led club participants to consider different kinds of works that might engage a broader segment of viewers. The first steps away from small forms were themselves quite small, however. Club advocates began to call for a special form of "club play" that would take its inspiration from instsenirovki, agit-trials, and living newspapers. Similar to small forms, these would ideally be created from within the united artistic circle; they would also address themselves to the issues of contemporary life.35 Club plays would differ from agitational forms by adding complex characters and following a unified story line. Such works needed to have an agitational content without being overbearing, determined one Leningrad critic. They should use simple, clear language and convey a logically constructed narrative. Such an approach would result in richer and more satisfying works that could speak to a broad working-class audience.36
The club play was a hybrid genre. Not only did advocates insist that their inspiration and roots should remain in the agitational theater of small forms, but they were also intended to be tailored to suit the specific difficulties and limitations of amateur stages. They had fewer characters, props, and technical challenges than did plays written for professional theaters. Because most clubs wanted new works at regular intervals to interest their audiences and mark Soviet festivals, they were also supposed to be fairly easy to prepare. This blurred the line dividing club plays from small forms. One advocate insisted that club plays reject the principles of psychological realism, which he believed formed the foundation of most professional theater. Like small forms, club plays would still be based on the principle of "massism" (massovost'), which meant they would be addressed to the broad masses in the most inclusive way possible.37 In an overview of different types of Soviet plays, another author defined what he believed were the essential elements of a club play: it had to have a unified story while maintaining modest staging requirements and a contemporary theme. These works should not attempt naturalistic, psychologically complex portrayals of characters, but they should move beyond the simple stereotypes, or "masks," of living newspapers.38
Advocates of club plays stressed repeatedly that they should emerge through the united artistic circle and not be appropriated ready made from available printed works. These points were made most forcefully by the Leningrad Proletkult leader, Valerii Bliumenfel'd, who wrote a number of articles on club plays for Leningrad journals and the Moscow publication Workers' Club. He envisioned a club play emerging from collective efforts, thus remaining true to the structure embraced by the advocates of small forms. These works "would maintain the basic forms of instsenirovki and living newspapers, but at the same time possess a unified and satisfying subject." Purely agitational works presented slogans, not life; club plays could come closer to the portrayal of life and move from agitation (a simple message for the masses) to propaganda (a more complex message for the more sophisticated). In the process, clubs would be better able to combat the rush of viewers to films and prerevolutionary melodramas.39 I. Ispolnev, a founding member of the Moscow action circle who had initially rejected any kind of play, went even further than Bliumenfel'd. He used the dialectic to describe the cultural process underway. Initially, amateur theaters had copied the work of professional stages. Then they had rejected this position and insisted on their own original works, such as living newspapers. But now he suggested that there must be a synthesis of these two extremes. "We have arrived at a unique and original form of theater performance, at the 'club spectacle' which contains all new forms, but united into a single unified action, and even sometimes into a single intrigue."40 For both of these authors, club plays were a necessary, discrete evolutionary step beyond small forms.
This supposed synthesis of small and large forms met opposition from both the artistic left and right. Even these modest proposals were a threat to those who wanted to continue improvisation and collective work. For such individuals, club plays posed a danger because they marked a move toward more conventional repertoires. Once an amateur circle had turned to plays, advocates of small forms feared that further steps toward prerevolutionary works and pieces designed for professional theaters would necessarily follow. That raised the danger that frivolous hackwork and apolitical entertainment might come to dominate amateur stages. These objections were raised most forcefully by Vitalii Zhemchuzhnyi, the most outspoken member of the Moscow action circle. He insisted that it was impossible to meld the two modes together because they were based on different principles: "Let us not study from centuries-old dramatic work and absorb the stagnant and alien dramatic canon. Rather let us create new rules for performance art."41
But another faction of critics believed that club plays should distance themselves much further from the impromptu methods of small forms. "Workers have longed for a deeper approach to questions of production and daily life in all their living dialectic," determined one Moscow club leader. "And this has not taken place, and cannot take place, in improvisations and living newspapers."42 A group letter signed by the staff of two large Moscow centers, the Sverdlov and Sapronov Clubs, stated that the united artistic circle itself should be tossed out because it was removed from life and gave too much power to the drama and literature groups within the club. They endorsed club plays and also insisted that more attention be given to providing participants better training in acting.43
A number of short plays were generated from within club circles. Our Daily Life (Nash byt), written and performed by the Moscow Electrical Light Factory Club, depicted a member of the German Communist Party working incognito at the Moscow plant. He went from division to division, trying to discover how the factory worked and what difference the revolution had made in people's lives. The protagonist "together with the viewers uncovered the good and bad aspects of the factory, the day care center, the club and the dormitory," one audience member observed. The focus on local themes reportedly drew an enthusiastic crowd: "Hearing that the factory intelligentsia didn't come off too well, the whole office came to watch. Many people recognized themselves in the play." Another homemade play, presented at the Moscow Artamonovskii Tram Park, showed how an older worker became convinced that the revolution was a good thing and explained his new convictions to doubting peasants.44 Noting this turn to self-generated plays, a writer in the journal Soviet Art (Sovetskoe iskusstvo) pronounced that amateur art was entering a new and more sophisticated staged.45
But just as living newspaper circles faced problems creating their own texts, amateur theater groups did not always have the time or talent to devise successful club plays. One work, Face the Countryside (Litsom k derevne), by a Leningrad worker who was studying in special classes at the university, received very bad reviews. It was not really a play at all, wrote one worker correspondent. Instead, it was a collection of scenes from city and country life without cohesion or a common thread. A serious, unified work required more preparation than the month and a half that the factory circle had spent on this production, the critic chided.46 A play with the enticing title of Factory Love (Fabrichnaia liubov'), written by a Komsomol member and addressing the sexual mores of young people, was panned for its weak dramatic structure and poor language.47
To alleviate the problem of repertoire, a specialized class of professional writers began to compose plays designed specifically for clubs. Widely published club authors included Boris Iurtsev, G. Bronikovskii, Vladimir Severnyi, Dmitrii Shcheglov, and Iakov Zadykhin. Very few of their works were ever performed by professional theaters. Most of these authors had begun their careers composing short agitational works and later turned to plays. Only Dmitrii Shcheglov had insisted on the play form from the beginning, making his mark as the leading opponent of the united artistic circle in Leningrad.
Iakov Zadykhin was involved with the Agitational Theater in Leningrad and also took part in the central Komsomol club theater that would eventually evolve into the youth theater TRAM.48 A successful play written for the club stage was Zadykhin's Hooligan, which examined a popular theme of club plays in the mid-1920s—the transformation of rowdy youth into upright Soviet citizens. The play, published in 1925, depicts the change of a drunken, unemployed young man into a model worker. There are many twists and turns along the way. One complication involves his former liaison with the daughter of a prosperous NEPman. Another has to do with a serious theft in the factory where he ends up working, a crime for which he is initially blamed. Nonetheless, with the help of his Komsomol girlfriend, a worker correspondent, he eventually enters the ranks of productive proletarians, even managing to save his factory from industrial sabotage.49
Hooligan is a cheerful play that offers a socialist-style happy ending. Not only does the hero announce his pending marriage to the exemplary worker correspondent, but he is also named a hero of labor for saving the factory. The hooligan villains, his former friends, are not particularly unsavory characters. Although they drink and swear, their main disruptive act is to throw rocks through the windows of the local House of Culture. The funniest parts of the play satirize elements of NEP popular culture through the figure of Katka, the daughter of the NEPman. Her main goal in life is to attain the kind of romance she has seen depicted in imported films. Against the wishes of her parents, she quickly transfers her affections from the main character to one of his hooligan friends. This young man, interested in her father's money, wins her affections by declaring, "You have stolen my heart like the daughter of the thief of Baghdad," alluding to the popular Douglas Fairbanks film that was one of the biggest box-office hits in the Soviet Union during the 1920s.50 When the suitor delivers a flowery address while Katka is standing at her window, she exclaims, "It's just like in the movies."51 In a Moscow performance by the Central Collective of Textile Workers, this star-struck shopkeeper's daughter stole the show.52
Zadykhin's work reveals the hybrid nature of club plays. Unlike most instsenirovki and living newspapers, it is not made up of many short segments; instead, it is divided into four discrete acts. The characters have personal names and the beginnings of developed personalities. It also tells a cohesive story. The advocate of psychological realism, Shcheglov, called it "a completely realistic play about daily life."53 Nonetheless, the play still bears a strong resemblance to the agitational theater of small forms. The characters are easily recognizable social types, especially the family of NEPmen and the upright Komsomol heroine. Only the hero experiences any kind of transformation during the play; the rest are static figures. Zadykhin's effort to use elements of urban popular culture was typical of small forms. Moreover, this work was clearly written with an eye to the limited resources at the disposal of amateur stages. It had only ten speaking roles, with a few additional walk-on parts. Not much was demanded in the way of scenery, making it very easy to stage.
To aid club theaters in choosing suitable works, the national Politprosvet organization began to publish reference works that offered an overview of plays suitable for club stages. The Repertory Guide (Repertuarnyi ukazatel'), published in 1925, listed works according to their theme (class struggle, war and revolution, old and new life) and offered brief summaries. The Repertory Bulletin (Repertuarnyi biulleten'), a periodical beginning publication in 1926, was more elaborate. It not only had plot summaries but also noted the staging requirements, the number and gender of parts, and the price of publication.54 In the following year, the national and Moscow trade union organizations began the publication of Club Stage (Klubnaia stsena), the first Soviet journal devoted entirely to amateur theater. It gave an overview and critique of current practices and also published the texts of short plays. One of the first issues included a work by Vladimir Severnyi, Rotten Thread (Gnilaia priazha), which examined a historic textile workers' strike.55 With these resources at their disposal, club participants could choose works that matched their resources and abilities.
#### Reappraising Professionals
Small forms of the early 1920s were in part a negative response to professional theaters. Most had not changed their offerings significantly in the wake of the revolution. With the exception of the Meyerhold studio, prestigious, state-supported academic theaters had also not made any organized attempt to offer assistance to amateur stages. Thus, many club theaters proclaimed that they were the only ones interested in examining the great social changes the revolution had brought about; professional theaters had little to offer the average viewer.
By the mid-1920s, however, some of the most important professional theaters began to perform new works that introduced revolutionary themes. Pressure from Narkompros and the installation of new directors brought significant changes to academic stages. After 1925, institutions like the Malyi Theater in Moscow, which until that point had concentrated on the classics, undertook new plays about the revolution and its aftermath.56 For some amateurs, this shift meant that they no longer needed to justify their work in oppositional terms, which expanded their sphere of activity. Not only would they direct themselves to the pressing issues of the day, they would also try to prepare their audiences to view works in professional theaters. In addition, the spate of new works written on the theme of revolution made a more sophisticated repertoire available to amateurs. These changes minimized the difference between established and club stages—and raised the question of how club theaters would use the skills and repertoire of their professional colleagues.
Moreover, there was a marked shift in the discussion about the ultimate aims of the amateur stage. For those devoted to the agitational theater of small forms, amateur performance was an end in itself. Its tasks were to educate the viewing audience and build a new community centered on the club. But by the second half of NEP, some participants began to see amateur theater differently. It was a necessary but incipient building block toward a new kind of professional stage. This point of view was quite apparent in a heated discussion among worker correspondents in the Moscow journal The Worker Viewer in 1925 on the topic "What should a worker's theater be like?" Although one participant expressed doubts about any kind of cooperation with professionals, this was a minority view. Most insisted that amateur theater had to grow aesthetically to the point at which it could form the basis for a new professionalism. And for that to happen, amateurs needed to solicit the acting, directing, and writing skills of professionals. One participant insisted that amateurs could never hope to create a serious theater unless they reached out to include specialists.57
This idea that amateurism was something incomplete and rudimentary came even from factions of the artistic left, who had, by and large, been very sympathetic to the club stage. In a 1926 article in Blue Blouse, Osip Brik maintained that professional living newspapers groups were the only hope for a new, revolutionary theater in the Soviet Union. They alone had the flexibility and immediacy to interest a wide audience and make theater relevant to the broad population. By contrast, club theaters, even those staging small forms, served mainly an educational role. Brik asserted that it was impossible to build a new theater from amateurs alone. Blue Blouse needed to guide and inspire these stages, "to transform their chaotic self-activity [stikhiinaia samodeiatel'nost'] into productive methods." Blue Blouse "moved beyond dilettantish self-help [liubitel'skaia samopomoshch'] toward a new professionalism in acting."58 By choosing derogatory words to describe amateur activity, such as liubitel'skii and stikhiinyi, Brik underscored the unpredictable nature of club theater. And although his aesthetic solutions were very different than those of most worker correspondents, he also felt that amateur stages were in great need of tutelage.
The new affinity between the professional and amateur stage was most apparent in club repertoire. As a new generation of Soviet playwrights began to create plays about the revolutionary struggle, amateurs started using their work. One example of these new authors was Vladimir Bill-Belotserkovskii, a former sailor whose prerevolutionary adventures had taken him to the United States. His plays, including Echo, Port the Helm (Levo rulia), and Storm, examined the impact of the Russian revolution in the West and the tumultuous years of the Civil War. First performed at the Malyi Theater and the Moscow Trade Union Theater (Teatr MGSPS), they were quickly taken up by amateur stages. Storm, set in the Civil War, was a particular favorite in clubs. At the first Moscow competition of trade union club theaters in 1927, the metal workers' club, Aviakhim, won first prize for its rendition of this play.59 Another new author was Alexander Afinogenov, who began his writing career for the Central Proletkult Theater in the 1920s. Several club stages, particularly in Moscow, staged his plays. One popular work was Robert Tim, which depicted a revolt of weavers in England in the nineteenth century. Konstantin Trenev's Liubov' Iarovaia, first performed at the Malyi Theater in 1926, traced the involvement of a rural school teacher on the Bolshevik side during the Civil War. This play was soon taken up by amateur theaters as well, although its popularity did not peak until the 1930s.60
Following the satirical bent of many living newspapers and instsenirovki, comic works written for the professional stage were also adopted by amateurs. One writer, Nikolai Erdman, whose plays were closely tied to the Meyerhold Theater in the 1920s, found a following in clubs. His biggest hit was the raucous satire The Mandate (Mandat), which was both a critical and popular success. It told the story of an anti-Soviet family that attempted to achieve a "security warrant" by having the son join the Communist Party. This would not only protect them but make the daughter more attractive for marriage. After a number of twists and turns, which included the family cook being mistaken for the dead empress Alexandra, the social outcasts remained without the coveted document.61 The play's biting treatment of NEPmen and a variety of hopeless prerevolutionary types made it a popular hit on club stages.
Another professional playwright who found a following on amateur stages was Boris Romashov. He wrote both serious dramas of the revolutionary struggle and comedies that were staged by the Malyi Theater and the Theater of Revolution. His depiction of the Civil War in South Russia, Fedka Esaul, was performed frequently in club theaters. His satire of NEP life, The End of Krivoryl'sk (Konets Krivoryl'ska), was a popular if controversial work. It examined social change in a small Russian town, portraying a wide range of characters from counter-revolutionaries, to small tradesmen, to careerist officials. Unlike many NEP satires, it does not pit evil anti-regime elements against the brave representatives of the new state. Even the heroes of the piece, the Soviet officials and Komsomol members, have plenty of flaws. They drink too much, have numerous sexual liaisons, and forget their duties in order to rush off to Mary Pickford films.62 Some rabkor critics objected to the play, saying that the portrayal of Soviet youth was much too negative.63
A limited number of classical plays also began to reappear on club stages in late NEP, after all but disappearing in the early 1920s. One critic in Leningrad noted an emerging specialization among amateur theaters there. Although most clubs performed works that reflected contemporary themes, a few staged primarily prerevolutionary classics, including Shakespeare, Lope de Vega, Molière, Carlo Goldoni, and Alexander Ostrovsky. Hamlet played to a packed house at the Leningrad Enlightenment Club. The Central Construction Workers' Union chose Gogol's Inspector General.64 Select Moscow club theaters also turned to prerevolutionary plays. One stage, sponsored with the support of several city trade unions, included The Lower Depths by Gorky among its opening works. A competition of chemical trade union theaters in 1927 featured works by Gogol and Pushkin, along with contemporary plays.65 Surveying this trend, Robert Pel'she of Narkompros concluded that it showed the growing sophistication of the average Soviet viewer, who now wanted a broader education in the arts.66
Those who looked to amateur theater for innovation, however, drew a different message from this shift to the classics and professional works. They felt that amateurs were moving backward, embracing prerevolutionary models of amateurism whereby groups simply copied what they saw on established stages. In the process, they tried to recreate the acting and staging techniques of professionals, which were usually beyond their abilities. In order to master complex works, they focused on one play for many months, or even years, which meant they performed very infrequently in the local club. As a result, amateur performance was once again isolated from the life of the club, undertaken with no consideration of the work going on in other circles. These pernicious trends would lead to an apolitical repertoire performed for the pure entertainment of viewers, warned Valerii Bliumenfel'd, who wanted club plays to emerge from amateur circles themselves.67
Nikolai L'vov saw the turn to classical plays as simply the latest stage in the battle between conservative and progressive forces that had begun at the Worker-Peasant Theater Conference in 1919. There, the conservatives from the popular theater movement had recommended Sophocles and Tolstoi; the progressive wing insisted that the masses create their own repertoire. While all those involved with amateur theater agreed that purely agitational works had grown tiresome, some were turning back the clock. Rather than attempt new works, they staged a repertoire common on amateur stages before the revolution. What L'vov saw as the progressive faction was building on what they had learned through agitational works, infusing their plays with more depth and complexity. "The struggle between these two directions has only just begun," he concluded.68
Although the struggle was for the future of amateur theater, both sides expressed their views with reference to the professional stage. Those who wanted a unique repertoire accused their opponents of following the strictures of the popular theater movement. Their ideas about acting and staging were derivative, guilty of a reactionary "naturalism" that exemplified the values and individualism of the bourgeoisie. The methods of Konstantin Stanislavsky, founder of the Moscow Art Theater, loomed large in this critique. His psychological approach to acting was unfit to capture the spirit of the collective. Clubs should never turn to naturalistic sets and try to imitate the psychological actor, argued one commentator in The Life of Art. Workers easily grasped the key elements of "leftist" theater, from its focus on the collective to its constructivist set designs."69
Those opposed to the theatrical left were ready with accusations of their own. They charged that their opponents were under the thrall of a segment of the bourgeois intelligentsia who had articulated their avant-garde ideas long before the revolution. Their work was a luxury for the very few but out of reach for the mass audience. Workers were baffled and offended by fragmented structure and strange staging of their performances.70 Instead, these critics wanted more convincing characters, easily understandable plots, and a positive treatment of Soviet heroes. In short, they wanted what they called "realism." One outspoken rabkor insisted, "Workers' theater must be as realistic as possible. It should illuminate the life of the worker, satisfy his spiritual longings, and encourage his future growth. It should not include primitive folk forms [ne dokhodit' do lubka] but nonetheless be close and understandable to the working masses."71 Two Moscow trade union leaders, put it this way: "[The worker viewer] wants to see life as it is. By this, he does not mean a naturalistic copy of life, but rather its most typical, realistic reproduction."72 This quotation articulates in embryo some of the basic principles of socialist realism. As most established stages also moved toward realism, the aesthetic distance between amateur and professional diminished.
#### The "Smychka" in Theater
The idea of a smychka, or union, between the proletariat and the peasantry, two potentially hostile classes, was a central concept of the New Economic Policy. This bedrock principle was extended outside the sphere of economics and class relations to discuss fruitful interaction between other potentially opposing forces. In late 1925, Robert Pel'she broached the idea of a smychka in culture. "Now apparently a smychka between professional and amateur art is beginning to show itself," he asserted. "We must deepen this trend."73
Pel'she's comments indicate that the debate surrounding the repertoire of club theater, largely conducted in specialized journals in the early 1920s, began to emerge as a topic of national discussion. The Seventh National Trade Union Congress, meeting in late 1926, addressed the problem of club repertoire. Mikhail Tomskii, the head of the national trade union organization, endorsed greater cooperation between amateurs and professionals. He insisted that much of the literature made by workers themselves was of poor quality—vulgar in tone and ungrammatical. If appropriate new works could not be found, then it was better to perform old plays. The final resolutions indicated that trade unions were willing to look to professionals for aid: "It is imperative to attract the best artistic forces to create and rework good plays for club stages and to lead club circles."74 At a special meeting devoted entirely to trade union cultural work in early 1927, the head of the national cultural division, Nikolai Evreinov (not to be confused with the famous theater director Nikolai Nikolaevich Evreinov, who had emigrated to France), made the message even clearer. He insisted that clubs attract professional groups to give model performances and strengthen ties with established stages in order to solicit better instructors.75
The gathering with the most significant impact on club stages was the national convention on theater sponsored by the Agitprop Division of the Communist Party in the spring of 1927. The main purpose of this meeting, designed to parallel the 1925 Party conference on literature, was to assess government policies for professional stages, reviewing their financial support, repertoire, and organization.76 But because the conference was justified as a method to acknowledge the importance of theater to the laboring masses, amateur stages in urban centers and the countryside received considerable attention. At this congress, the notion that there should be a rapprochement (sblizhenie) between amateur and professionals received official codification. The conference determined that amateur theaters needed to learn artistic mastery from established stages. At the same time, professional theaters had to provide the mass viewer with a repertoire that reflected the problems and accomplishments of the contemporary era.
The assumption that audiences in amateur theaters were becoming more sophisticated and discriminating reached the level of truism at the conference, being repeated in every single speech and resolution. In his opening statements, the head of Agitprop, Vilis Knorin, linked workers' improving economic and political condition to their demands for better art.77 He sounded a recurring theme at the proceedings—namely that amateur theaters had an important role to play as the transmitters of artistic values and the theatrical heritage: "Club and rural theaters at the present time must and should learn artistic mastery and the skills of directing and acting from professional theaters; they should transmit what they have learned to the broad masses."78 The imbalance between amateur theaters, which were revolutionary but not artistic, and professional theaters, which were artistic but not revolutionary, had to come to an end.
The main speech on amateur theaters was presented by Evreinov from the central trade union bureaucracy. Such institutions were important, he began, because professional theaters could not begin to meet all the needs of the viewing public. By making reference to the several thousand amateur stages under trade union control, Evreinov argued that amateur stages served a larger segment of the population than professional theaters. Because they were closer to their audiences, amateur stages could more easily meet viewers' needs and could better serve as a conduit for socialist education.79
But while he recognized the strategic importance of amateur theater, Evreinov began with the assumption that these institutions were in a deep state of crisis. Bitingly critical of current offerings, he denounced the idea that performance works should only be generated from within the club circle. He also attacked what he called the "false theories" of the past, which had led club theaters astray. These included all of the ideas that had formed the theoretical underpinnings for small forms. The unified artistic circle denied the importance of the artistic heritage and subordinated all creative work within the club to a single circle.80 He was no kinder to the principles behind action cells, which he referred to as "action art" (deistvennoe iskusstvo). This approach repudiated any specific qualities for the art of the stage. Finally, in a rather confused coda, he denounced the concept of utilitarian art, a notion he traced to the Civil War but which still had adherents in the trade union movement. All of these ideas were cooked up in isolated offices by poorly qualified intellectuals who wished to establish a monopoly on conceptions of proletarian cul-ture.81 They resulted in the worst possible kinds of agitational performances, works without any artistic value.
Evreinov believed that the movement away from small forms was coming from below, from amateur theater groups themselves. The worker viewer had begun to say, "Enough agitation, enough homemade concoctions [samodel'shchina]—give me theater." As a result, club stages were beginning to choose large, important plays such as the opera Rusalka, Bill-Belotserkovskii's Storm and the Dutch classic The Good Hope. Protests against this shift in repertoire came not from viewers but rather from (unnamed) leaders who opposed artistic mastery and the cultural heritage of the past.82
But rather than discard agitational work altogether, Evreinov proposed a mixed repertoire for clubs. Amateur stages should continue to use small forms; they could have a positive political value and in general were easy to perform. However, these had to be combined with plays written for club stages, works intended for professional theaters, and select representatives from the classics. The only way to achieve such a diverse repertoire, Evreinov suggested, was to strip club theaters of some of their agitational responsibilities. Such suggestions had been made by union leaders periodically during the 1920s to protect drama circles from their burdensome performance schedules. Evreinov articulated this position in uncompromising terms. "Finally," he proclaimed, "we must relieve circles of a whole variety of obligatory appearances for any number of campaigns. They now have to appear at all political campaigns and holidays—on March 8, the day of the Paris Commune, May Day, and others. For this reason drama circles cannot do serious work."83 Evreinov's intention was to lighten the burden of drama circles so that they would have more time for serious preparation. Yet seen from another perspective, he actually heaped more responsibilities on them. In essence he charged them with two equally important but not necessarily complimentary tasks—political agitation (although in lesser amounts) and the artistic education of their audiences.
The wide-ranging discussion after Evreinov's speech rehashed the many controversies facing club theaters since the revolution. How were they supposed to attend to agitation and also create artistically satisfying work? queried one representative from Ivanovo-Voznesensk. Some club theaters had responded by creating two separate groups, one for agitation and one to tackle plays, but this put a further strain on scarce resources. Another delegate from the Urals commented that it was a good thing to lighten club theaters' heavy agitational load, which required them to perform for every festival and also for political campaigns of all kinds. Still, he questioned how appropriate an opera like Rusalka would be to mark the celebration of the October Revolution. Others insisted that if clubs were going to produce more effective artistic works, they needed significantly more resources. Most club stages were simply too small and poorly equipped to perform big plays. Thus, there remained no serious alternative to agitational works.84
Beneath these debates around repertoire, one can discern a real sense of anxiety about a production strategy that left significant control in local hands. Many speakers at the gathering complained about lack of vigilance over club stages, which resulted in the performance of prerevolutionary hack work and even works that had been officially banned by the government. But the most passionate exclamations of censure were reserved for living newspaper scripts, which some participants felt had become the purveyors of pornography. Evreinov quoted offensive lines from a work dedicated to International Women's Day, which he believed praised philandering and loose moral conduct. Other delegates had similar stories. According to one, living newspapers had become a forum for "salty and pornographic anecdotes."85 The final resolutions at the conference called not only for tightened vigilance over the text of works performed, but also increased monitoring of the performances themselves—a recognition that improvised forms could vary considerably from one show to the next.86
Despite the considerable attention brought to amateur stages at the widely publicized national event, the conference served mainly to fix their junior status in any cultural partnership with professional stages. Many speakers found amateur theaters useful only to the degree that professional theaters could not meet all the needs of the population. "Professional theater cannot completely satisfy the interest of the broad proletarian masses in performance," read the final resolutions. "Because of this the club stage has become and will continue to be a significant tool to meet workers' cultural demands."87 And although small forms were not abandoned entirely, the organizing methods that had supported their proliferation and their most widely used form, the living newspaper, were singled out for criticism. Evreinov denounced the theories of early NEP that claimed a special role for amateur theater in the creation of Soviet culture as so much radical nonsense.88 Small forms were discussed primarily in terms of their inadequacies. The conference delegates did acknowledge why club stages might employ them; they were easy to perform, adapted easily to agitational goals, and allowed some local creativity. But these positive points were undercut by the premise that the most important task of club stages was to introduce a serious repertoire and show the masses the best examples of artistic work.
The Agitprop conference offered unambiguous advice to local amateur stages: call in the experts to improve your work. Certainly, delegates gathered at a trade union conference in Leningrad only a few weeks later got this message. Konstantin Tverskoi, head of the city-wide artistic section, insisted that trade unions must aim to "bring art closer to the masses" (iskusstvo–blizhe k massam). In local discussions on theater, the most common demand he had heard was for higher quality work. The only way to achieve this was to bring in more professionals. He predicted that by the celebration of the tenth anniversary of the revolution, only a few months away, "the antagonism between professionals and amateurs will finally be eliminated."89
Tverskoi's prophesy was fulfilled through a joint project between trade union theaters and the Leningrad Bol'shoi Drama Theater, called "Ten Octobers" (Desiat' Oktiabrei). This mass spectacle, which included an estimated one thousand performers, celebrated ten years of Soviet power. Billed as a direct response to the Agitprop conference, the printed program even incorporated excerpts from the conference resolutions. The episodic script, with twenty-five separate sections, used literary works, songs, excerpts from plays, and political writings to tell the story of the revolution from Lenin's arrival at the Finland station until October 1927. It incorporated the performance and technical staff of the Bol'shoi Drama Theater, as well as thirty-five amateur theater groups, music circles, and living newspapers.90
It is worth stepping back for a moment to compare "Ten Octobers" to the mass festivals of the Civil War years, such as "The Storming of the Winter Palace" in 1920.91 Those earlier events also included amateur performers, but only as part of the crowd scenes. In this celebration, amateurs had speaking and singing roles. They were given voice, and this reflected a positive change in the status of amateur creation in the Soviet artistic pantheon. According to one observer, however, "Ten Octobers" was structured so that the professional performers provided the narrative elements that moved the story forward. Amateur circles provided the local color, the background atmosphere; they were not integrated as equals.92 This judgment was unintentionally confirmed by the main director of the spectacle, B. Andreev-Bashinskii from the Bol'shoi Drama Theater. He contended that the amateurs learned about professional discipline and mastery from the collaboration. For their part, the professionals were inspired by the immediacy of amateur theater and its direct political relevance to proletarian audiences.93 While he surely meant his comments to be appreciative of amateur performers, he nonetheless presented them as valuable raw materials that needed to be shaped and directed through professional intervention.
The second half of the NEP period was a time of great diversity for amateur stages. Although sharply criticized by some viewers, living newspapers and other small forms remained popular. Plays designed specifically for clubs appeared in cheap editions and were also distributed in the new journal Club Stage. New dramas intended for professional stages, especially those dealing with the history of the revolution and contemporary life, also were common choices. Some circles turned to classic plays that had been common on club stages before the revolution.
This broad range of performances reflected a fundamental shift in ideas on the value and significance of professional stages. Once shunned as irrelevant or dangerous for amateur performers, many club theaters now looked to professionals for repertoire and assistance. They adapted work from major theaters and even collaborated on joint projects. Such a change in attitudes was hardly unique to amateur stages. By the middle of the 1920s, many utopian projects that envisioned a society less structured by hierarchies of skill and training had fallen by the wayside. The dream of an all-volunteer militia, put forward by a faction of the Red Army, evaporated. School curricula designed to nurture the whole person and avoid excessive specialization were revised as Soviet industry called for better training and students demanded programs that would offer better chances for advancement. Everywhere, the egalitarian currents of the Russian revolution were weakening.94
Nonetheless, we should not assume that calls for better training and skill meant the same thing to everyone who voiced them. Supervisory cultural institutions, such as trade union administrations and the Agitprop Division, hoped to limit the unpredictable nature of amateur performance; they wanted additional controls over poor production standards and poor choices in repertoire, which had allowed dangerous material to make its way on stage. Some professionals, like the director of the Leningrad Bol'shoi Drama Theater, expressed the wish that amateur stages might help them gain better access to proletarian audiences. As the program for "Ten Octobers" stated, amateurs could serve as a bridge between professionals and the proletarian public.95
The motives of the participants and audiences of the amateur stage were even more complex. A segment of the viewing audience hoped that a more varied repertoire would make for more interesting viewing. The financial improvements of late NEP, which meant expanded club stages and more funds for props and costumes, allowed actors and directors to bring performance standards closer to those on the professional stage. For some advocates, this was evidence of the smychka in culture; they were finally in a position to collaborate with their professional counterparts. But for others, this provided the basis for a new aggressive onslaught. Improvements by amateurs meant that select theaters, with intrinsic ties to lower-class audiences, might finally be able to challenge the dominance of conventional stages.
* * *
1. "Litsom k novomu tea-professionalizmu," NZ 47 (1926): 3–4.
2. On realignments among professionals, see Katerina Clark, "The 'Quiet Revolution' in Soviet Intellectual Life," in Sheila Fitzpatrick et al., eds., Russia in the Era of NEP (Bloomington: Indiana University Press, 1991), 210–30.
3. S. M. Krylov, ed., Puti razvitiia teatra: Stenograficheskii otchet i resheniia partiinogo soveshchaniia po voprosam teatra pri Agitprope TsK VKP (b) v mae 1927 g. (Moscow: Kinopechat', 1927), 498.
4. On worker correspondents and professional theaters, see L. A. Pinegina, Sovetskii rabochii klass i khudozhestvennaia kul'tura (Moscow: Izdatel'stvo Moskovskogo universiteta, 1984), 176–79.
5. Rabkor A. Stepkoi, "Zhivaia gazeta TPO," ZI 1 (1925): 15; Rabkor I. Iankelevich, "Vspomnite ob iskusstve," RiT 17 (1925): 20.
6. A. Sh., "Boliachki klubnogo spektaklia," NZ 7 (1926): 10.
7. I. Isaev, Osnovnye voprosy klubnoi stseny (Moscow: VTSSPS, 1928), 11, 29.
8. D. Shcheglov, Teatral'no-khudozhestvennaia rabota v klubakh: Metodika i praktika (Leningrad: Gubprofsovet, 1926), 7.
9. O. Liubomirskii, "Klub 'Krasnyi Oktiabr',' " NZ 6 (1925): 12. See also "Pochemu vzroslyi rabochii ne idet v klub?" RK 12 (1925): 57.
10. V. Nikulin, "V tsentral'nom klube kommunal'nikov," NZ 5 (1925): 12.
11. E K., "O byte rabochego kluba," RK 2 (1925), 27.
12. For overviews of club construction in late NEP, see Gabriele Gorzka, Arbeiterkultur in der Sowjetunion (Berlin: Arno Spitz, 1990), 246–64; V. Khazanova, Klubnaia zhizn' i arkhitektura kluba, v. 1 (Moscow: Rossiiskii institut iskusstvoznaniia, 1994), 44–74
13. N. Vorontyneva, "Stroitel'stvo rabochikh klubov," RK 24 (1925): 14.
14. Gorzka, Arbeiterkultur, 258–61; S. Frederick Starr, Melnikov: Solo Architect in a Mass Society (Princeton: Princeton University Press, 1978), 134–40.
15. R. Begak, "Litso kluba," RK 30/ 31 (1926): 71–74.
16. Khazanova, Klubnaia zhizn', v. 1, 77.
17. Arnold Gal', "O rabochikh iumoristakh," RZ 32/33 (1924): 31.
18. Liubomirskii, "Klub 'Krasnyi Oktiabr',' " 12. See similar complaints about a May Day performance in I.D., "Pervomaiskaia instsenirovka v klube Gosznaka," RK 41 (1927): 58.
19. A. Sh., "Boliachki klubnogo spektaklia," NZ 7 (1926): 10.
20. On the links between Trotsky and Soviet youth, see Sheila Fitzpatrick, Education and Social Mobility in the Soviet Union, 1921–1934 (Cambridge: Cambridge University Press, 1979), 94–97; Michael David-Fox, Revolution of the Mind (Ithaca: Cornell University Press, 1997), 151–60; and Anne Gorsuch, Enthusiasts, Bohemians, and Delinquents Soviet Youth Culture, 1921–1928 (Bloomington: Indiana University Press, forthcoming). For Trotsky's views on youth and clubs, see Leon Trotsky, "Leninism and Workers' Clubs," [1924] in his Problems of Everyday Life (New York: Monad Press, 1973), 288–322, esp. 289–91.
21. On the Leningrad Komsomol, see Eric Naiman, Sex in Public (Princeton: Princeton University Press, 1997), 266–70, and Ralph Fisher, Pattern for Soviet Youth (New York: Columbia University Press, 1955), 116–18.
22. On young people and Western culture, see Gorsuch, Enthusiasts chs. 5 and 6. See also Denise Youngblood, Movies for the Masses (Cambridge: Cambridge University Press, 1992), 27, 52–54, and Christopher Gilman, "The Fox-trot and the New Economic Policy," Experiment 2 (1996): 443–75.
23. Kluby Moskvy i gubernii (Moscow: Trud i kniga, 1926), 29; V. Bliumenfel'd, "Melochi klubnogo byta," RK 3 (1926): 42–45.
24. F. Seniushkin, "Zadachi klubnoi raboty," Prizyv 1 (1925): 5, emphasis in the original.
25. M. Zel'manov, "U trekhgortsev," Prizyv 3 (1925): 123; R. Ginzburg, "Bor'ba s khuliganstvom," RK 27 (1926): 11–15; A. Krutov, "Bor'ba s khuliganstvom v rabochikh klubakh," RK 36 (1926): 30–33.
26. Joan Neuberger, Hooliganism: Crime, Culture, and Power in St. Petersburg, 1900–1914 (Berkeley: University of California Press, 1993), esp. 275–79.
27. On the national anti-hooliganism campaign, which even at the time was interpreted as a fear of young people's potential oppositional power, see Naiman, Sex in Public, 250–88, and Gorsuch, Enthusiasts, ch. 8.
28. Gibeman-Vostokov, "K voprosu o privlechenii vzroslykh v kluby," RK 27 (1926): 66.
29. "O politike partii v oblasti khudozhestvennoi literatury," in Voprosy kul'tury pri diktature proletariata (Moscow: Gosudarstvennoe izdatel'stvo, 1925), 215.
30. Rabkor Aleksandrov, "P'esa, a ne instsenirovka," RiT 26 (1925): 13.
31. M. Bystryi, "O malykh (nizkikh) i bol'shikh (vysokikh) teatral'nykh formakh," ZI 13 (1926): 4.
32. B. Filippov, "Klubno-khudozhestvennye konkursy," in Z. A. Edel'son and B. M. Filippov, eds., Profsoiuzy i iskusstvo: Sbornik statei s prilozheniem rezoliutsii Pervoi leningradskoi mezhsoiuznoi konferentsii po voprosam khudozhestvennoi raboty (Leningrad: Izdatel'stvo Leningradskogo gubprofsoveta, 1927), 46.
33. See John Hatch, "Hangouts and Hangovers: State, Class and Culture in Moscow's Workers' Club Movement, 1925–1928," Russian Review 53, no. 1 (1994): 107–13; and the Sickle and Hammer factory records for March 1925, GARF f. 7952 (Istoriia fabrik i zavodov), op. 3, d. 228, 1. 29, and for July–September 1927, d. 233, 1. 29.
34. "Itogi Vsesoiuznogo soveshchaniia pri Glavpolitprosvete," ZI 1 (1926): 2.
35. A. Borisov, "Put' k p'ese," RiT 10 (1925): 19; Rabkor A. Zharikov, "Nuzhna p'esa," RiT 11 (1925): 16.
36. Rabkor Nikolaev, "Kakoi dolzhna byt' klubnaia p'esa," RiT 27 (1925): 8.
37. Boris Andreev, "Kollektivnaia dramaturgiia," RiT 34 (1925): 13.
38. A. Borodin, "Tekhnika klubnoi dramaturgii," RK 48 (1927): 29.
39. V. Bliumenfel'd, "O klubnoi p'ese," RK 22 (1925): 48–52, quotation 48. See also idem, "Massovaia dramaturgiia: Na puti k klubnoi p'ese," ZI 42 (1925): 7–8, and "Klubnaia dramaturgiia," RK 30/31 (1926): 11–17.
40. I. Ispolnev, "Teatral'naia deistvennaia rabota v klube," RK 22 (1925): 53–54, quotation 54.
41. V. Zhemchuzhnyi, "Po porucheniiu Assotsiatsii instruktorov deistvennykh iacheek," RK 23 (1925): 70–72, quotation 72.
42. V. Bogoliubov, "Nasha tribuna," RK 23 (1925): 73.
43. "Rabotniki klubov im. Ia. M. Sverdlova i im. V. Sapronova," RK 23 (1925): 69–70.
44. Zbytko, "Sami pishem p'esy," RZ 24 (1924): 23; M., "Po Oktiabriu ravniaisia!" RZ 28 (1924): 20.
45. V. Rudin, "Na poroge novogo samodeiatel'nogo iskusstva," Sovetskoe iskusstvo 1 (1925): 15–16.
46. On the creation of this work, see El'f, "Po nemnogu, no uporno!" RiT 24 (1925): 5; for the critique, see Rabkor Shevalkov, "U tekstil'shchikov," RiT 47 (1925): 15.
47. D. Tolmachev, "Fabrichnaia liubov'," ZI 31 (1925): 16.
48. On Zadykhin's involvement in Leningrad theaters, see RiT 39 (1925) and A. S. Bulgakov and S. S. Danilov, Gosudarstvennyi agitatsionnyi teatr v Leningrade (Moscow: Academia, 1931), 156, 167.
49. Ia. L. Zadykhin, Khuligan (Leningrad: MODPiK, 1925).
50. Ibid., 20. On the popularity of The Thief of Baghdad, see Youngblood, Movies for the Masses, 20.
51. Zadykhin, Khuligan, 22.
52. Nikolai L'vov, "Tsentral'nyi kollektiv tekstil'shchikov," NZ 23 (1927): 9.
53. Dmitrii Shcheglov, "U istokov," in U istokov (Moscow: VTO, 1960), 175.
54. Repertuarnyi ukazatel': Sbornik otzyvov o p'esakh dlia professional'nogo i samodeiatel'nogo teatra (Moscow: Glavpolitprosvet, 1925); Repertuarnyi biulleten' 1 (1926).
55. V. Severnyi, "Gnilaia priazha," Klubnaia stsena (henceforth cited as KS) 2 (1927): 49–69.
56. Richard Thorpe, "Academic Art in Revolutionary Russia: State, Society, and the Academic Stage, 1897–1929," unpublished manuscript, ch.7.
57. "Kakim dolzhen byt' rabochii teatr," RZ 7 (1925): 6–9. See also "Rabteatr iz dramkruzhka," RZ 1 (1925): 7–8.
58. O. M. Brik, "Segodnia i zavtra 'Sinei bluzy,' " SB 47/48 (1926): 1–15, quotations 7, 10.
59. Nikolai L'vov, "Mezhsoiuzhnye sorevnovaniia po zrelishchnoi rabote," NZ 24 (1927): 6.
60. For an overview of dramatic works popular in the late 1920s, see Harold B. Segal, Twentieth-Century Russian Drama, rev. ed. (Baltimore: The Johns Hopkins University Press, 1993), 147–81.
61. Nikolai Erdman, The Mandate, in Two Plays, trans. Marjorie Hoover (Ann Arbor: Ardis, 1975), 11–94.
62. Boris Sergeevich Romashov, Konets Krivoryl'ska, in his P'esy (Moscow: Khudozhestvennaia literatura, 1935), 143–245.
63. M. D-ov, S. Antonov, and S. Dubrovin, "Konets Krivoryl'ska," NZ 15 (1926): 12.
64. Sergei Spasskii, "Pis'mo iz Leningrada," NZ 11 (1925): 7; "Teatr stroitelei," ZI 35 (1926): 24.
65. A. Zhuravlev, "Moskovskii rabochii teatr," NZ 41 (1926): 16; Am. Aks, "U khimikov," NZ 23 (1927): 9.
66. R. Pel'she, "Itogi sezona," Repertuarnyi biulleten' 4 (1927): 7.
67. V. Bliumenfel'd, "Samodeiatel'nost' ili professionalizm v klubnykh i khudozhestvennykh kruzhkakh," ZI 38 (1926): 9; see also idem, "Klubnaia dramaturgiia," RK 30/31 (1926): 12.
68. Nikolai L'vov, "Repertuar za desiat' let," NZ 41 (1927): 6.
69. B., "Klubnaia dramaturgiia," ZI 32 (1926): 7. See also Shagin, "O chem govorili na Pervoi leningradskoi mezhsoiuznoi khudozhestvennoi konferentsii," ZI 23 (1927): 4.
70. For the most extreme expression of these views, see Rabochie o literature, 45–52, 99–100. This work is a compilation of highly selective quotes from The Worker Viewer. For a searing critique of this book by a worker correspondent, see Turii Kobrin, Teatr im Vs. Meierkhol'da i rabochii zritel' (Moscow: Moskovskoe teatral'noe izdatel'stvo, 1926), 51.
71. Vdovin, cited in A. Sh., "Kakim dolzhen byt' rabochii teatr," NZ 11 (1926): 5.
72. N. Volkonskii and A. Borodin, "Vliianie profteatra na klubnuiu stsenu," KS 2 (1927): 17.
73. "Itogi Vsesoiuznogo soveshchaniia pri Glavpolitprosvete," ZI 1 (1926): 2.
74. Sed'moi s''ezd professional'nykh soiuzov (Moscow: Profizdat, 1926), 67–69, 773, quotation 773.
75. "Za kachestvo klubnoi stseny," NZ 11 (1927): 8.
76. On the background for this conference, see Sheila Fitzpatrick, "The Emergence of Glaviskusstvo," Soviet Studies 32 (October, 1971): 238–41, and Thorpe, "Academic Art," ch. 7.
77. V. G. Knorin, "Teatr i sotsialisticheskoe stroitel'stvo," in S. M. Krylov, ed., Puti razvitiia teatra, 5.
78. Ibid., 12.
79. N. N. Evreinov, "Stroitel'stvo samodeiatel'nogo teatra," ibid., 264–65.
80. Oddly, Evreinov attributes this idea to "Avilov" instead of Grigorii Avlov, who was only one of its proponents in Leningrad (ibid., 267–68).
81. Ibid., 268–69, 498.
82. Ibid., 271–73, quotation 271–72.
83. Ibid., 277.
84. Ibid., 278–79, 285–86, 296, 298.
85. For anxieties about "pornography," see ibid., 270–71, 301, quotation 293.
86. Ibid., 487.
87. Ibid., 490.
88. Ibid., 498.
89. Konstantin Tverskoi, "Teatral'naia rabota leningradskikh profsoiuzov," in Profsoiuzy i iskusstvo, 65.
90. Desiat' Oktiabrei (Leningrad: Bol'shoi dramaticheskii teatr, 1927).
91. See James von Geldern's detailed description of this event in Bolshevik Festivals, 1917–1920 (Berkeley: University of California Press, 1990), 199–205.
92. V. B., "Spektakl' 'Desiat' Oktiabrei,' " KS 1 (1928): 74–75.
93. B. Andreev-Bashinskii, "Tekhnika massovoi postanovki: Kak my rabotali nad massovym predstavleniem," KS 1 (1928): 72.
94. See Mark von Hagen, Soldiers in the Proletarian Dictatorship (Ithaca: Cornell University Press, 1990), 206–8; Sheila Fitzpatrick, Education and Social Mobility in the Soviet Union 1921–1934 (Cambridge: Cambridge University Press, 1979), 57, 63; and Richard Stites, Revolutionary Dreams (New York: Oxford University Press, 1989), 140–44.
95. Desiat' Oktiabrei, 7.
## 4
## tram: The Vanguard of Amateur Art
THE LENINGRAD Theater of Working-Class Youth (Teatr rabochei molodezhi), called by its acronym TRAM, was the best-known amateur stage of the NEP period, eventually achieving a national reputation. Its far-reaching claims for the creative potential of amateurs and the special interests of Soviet youth sparked contentious discussions in the cultural press. Its original repertoire, a large part of which was published, was performed on club stages throughout the country. When its members abandoned their day jobs for full-time theatrical work in 1928, TRAM challenged conventional understandings of what it meant to be a professional. Because of the theater's notoriety and long record of controversy, TRAM participants have left behind a rich archival and published record of creative works and memoirs that make it possible to investigate its internal dynamics.1 Thus, TRAM offers a unique opportunity to sketch out details about membership, creative decision making, and social interactions within the amateur theater movement.
During the years of the New Economic Policy, TRAM in many ways embodied the amateur art movement. Although it can hardly be considered typical, TRAM conformed to basic developmental patterns of amateur theaters. It began in a Komsomol club staging agitational works for a neighborhood clientele. By the mid-1920s, participants began to write their own plays. Their original creations touched on everyday problems and tackled a central dilemma for Soviet youth eager to find a place in the new society—the strained relationship between personal happiness and community responsibility. These works contained the fundamental elements of the club play. They had a clear didactic purpose but also vivid characters and compelling stories that were designed to spark discussion. The success of these plays turned the Leningrad TRAM into a model for factory and trade union stages. As new TRAM circles opened around the country, the Leningrad group presented itself as a vanguard organization for amateur club stages.
With its increasing success, the Leningrad TRAM turned professional in 1928, allowing its members to devote themselves entirely to acting and writing. This was a step open only to small number of amateur stages, which found support from political and trade union organizations. These newly minted specialists, fiercely loyal to their original audiences, proclaimed that they would be professionals of a new type who would not forget their social base. TRAM was most vociferous in these pronouncements, insisting that its aesthetic principles grew from the unique social and political position of club theater.
The grander TRAM'S claims became, the more it exposed itself to criticism. Those who admired established professional theater felt that its work remained too amateurish; it offered no stable system of acting or directing that could inspire other groups. Amateur circles charged that it had distanced itself from its local constituency and no longer could generate a contemporary, relevant repertoire. This criticism became increasingly dominant during the years of the First Five-Year Plan, when a new kind of amateur circle took shape based directly in the factory workshop. For these aggressive small groups, TRAM had become the establishment, scarcely distinguishable from the nationally supported stages that it rejected.
#### TRAM Takes Shape
Like many Soviet amateur stages, TRAM could trace its roots to the Russian Civil War. Its inspired director was Mikhail Sokolovskii, a railroad worker born at the turn of the century who became a Komsomol organizer during the revolution. By 1919 Sokolovskii had taken charge of the literary studio for the Komsomol club in the First City District of Petrograd. He used the studio to create texts for agitational performances, such as a Komsomol evening celebration in honor of the Paris Commune staged in March 1920.2 Sokolovskii's penchant for agitation intensified when he was sent to Murmansk by the Komsomol to aid in efforts to defend the railroad against British attacks. While on assignment, he became involved in the railroad union's agitational studio, the Shimanovskii troupe. This proved an important contact, because Shimanovskii's group reshaped itself into the Petrograd Politprosvet's Central Agitational Studio after the war was over and provided resources and inspiration for Sokolovskii's own efforts.3
Not until 1922 did the Komsomol theater circle regroup, once again under Sokolovskii's leadership. Now located in the Gleron House, a Komsomol club named after a young Komsomol leader killed during the war, the theater embraced agitational tasks. The Gleron House club staged its first performance, an instsenirovka entitled "Five Years of the Komsomol," in June 1922. As was typical for the times, the script for the event was concocted in a mere three days.4 With a mix of songs, poems, and mass scenes, the evening event's most stunning moment was when a young Red Army soldier rode a horse onto the stage.5
The theater studio at the Gleron House attracted a group of dedicated enthusiasts who would form the core of TRAM activists throughout the 1920s. At the center was a group of young writers who were in charge of generating scripts and living newspapers. This circle, which called itself "The Lever" (Rychag), included the Petrograd Komsomol and Politprosvet organizer Nikolai L'vov (not to be confused with the Moscow theater activist of the same name), Arkadii Gorbenko, Pavel Marinchik (the future memoirist of TRAM), Dmitrii Tolmachev, and Mura Kashevnik. They worked under the supervision of the writer A. V. Sventitskii, who earned the princely sum of thirty rubles a month for his role as the local "expert."6 These young men, all of whom had been active in the Komsomol during the Civil War, gave the group its youthful, militant, masculine slant. Although young women participated in skits and performances, they were not part of the inner circle of organizers and writers. By 1923, yet another collaborator joined the group—none other than Adrian Piotrovskii, the articulate advocate of united artistic circles. Well-placed in the city's cultural bureaucracies, Piotrovskii became a coworker, promoter, and protector for the theater.7
A sketch of Mikhail Sokolovskii. Leningradskii TRAM v Moskve (Leningrad: Izdanie Gostrama, 1928).
The Gleron Club was simultaneously the central Komsomol club for Petrograd as a whole and a neighborhood center serving the Moscow-Narvskii District, part of the city's industrial core. It attracted factory youth from the nearby plants, like the big Skorokhod factory, the intended constituency for the Komsomol. The appeal of club events did not stop there, however. Well-dressed young people in straight-legged pants, starched collars, and silk handkerchiefs appeared alongside scruffy street youth of the semi-criminal demi-monde at club events. Evening performances were sometimes rowdy affairs, with fistfights breaking out between the club's different constituencies. The narrow club theater could pack in close to four hundred viewers, with the audience sitting in the aisles, on window sills, and even on the stage.8
Participants in the Komsomol club had little or no training in the arts, but neither they nor Sokolovskii saw this lack as an obstacle. When the club was swamped with volunteers who wanted to take part in the first performance in 1922, Sokolovskii encouraged all comers. According to Pavel Marinchik, when volunteers asked what they could do, `Sokolovskii grinned and looked at them, 'You can become playwrights and actors.' "9 This irreverence toward conventional artistic training was part of the spirit of the times. Marinchik recalls that club members laughed about the exalted connotations of the word "artist." "We went to the Gleron House, wrote and performed instsenirovki, conducted club evenings, and thought that all of this was just another form of Komsomol work."10 Some members even initially rejected the appellation of "theater" for their circle, believing that it evoked conservative notions of performance separated from life. As Sokolovskii put it, "For us it was not always clear where life ended and the performances began."11
From 1922 until 1925, the Komsomol theater circle was actively engaged in agitational work, marking the dates of the Red Calendar. On May Day 1923, it put on a performance called "Hymn to Labor" and participated in the city's anti-religious festivities for Komsomol Easter. In 1924, it celebrated Red Army Day with a skit called "The Dream of the Illiterate Red Army Soldier." For the Day of the Paris Commune, it staged a work by Piotrovskii devoted to the holiday. The group drew in part on folk traditions for its scripts, staging a loose adaptation of the venerable folk play Tsar Maksimilian at a performance in honor of Komsomol Christmas, this time with the evil father figure recast as General Kolchak, an anti-Bolshevik leader during the Russian Civil War.12
Like many other clubs, the Gleron House started its own living newspaper, "Gleron's Pencil," which chose its topics close to home. Its satirical performances were devoted almost exclusively to life within the club and to the role of youth in the new society. The tensions between Communist-identified youth and the other youth cultures of NEP were favorite themes, as in one performance that mocked "phony" Komsomol members who liked to go to dances instead of studying.13 "Gleron's Pencil" developed a stock character, "Crazy Sashka" (Sashka Chumovoi), an undisciplined factory youth who was having a hard time adjusting to NEP. Bored by political education classes and the dull routine of daily life, Sashka had ties to hooligan elements. Crazy Sashka was featured in club brawls, factory settings, street scenes, and even on foreign trips. In a 1925 May Day performance of "Gleron's Pencil," he showed up unexpectedly at a meeting of German capitalists.14 Played by the club member Vasia Kopachev, Sashka appeared in hooligan attire—bell-bottom trousers, a striped sailor's shirt, and a cap worn slightly askew.15 Performances featuring Crazy Sashka won the circle an enthusiastic audience.
The Gleron circle also began to experiment with more tightly organized works, moving away from small forms and toward plays. Man in a Red Wig, written by group members and staged in 1923, examined the problem of intellectuals in the revolution. The next year the Gleron circle performed several works that were sustained investigations into a single theme, meeting the requirements for a club play. These included The Story of the Iron Battalion (Slovo o zheleznom batal'one), based on an Alexander Serafimovich story about the fall of the Provisional Government and An Intellectual Marriage (Zhenit'ba intelligentnogo), which dealt with the moderate socialist leader, Alexander Kerenskii, during the February Revolution.16 By this time, the theater studio had gained a number of helpers, including the acrobat Fedor Knorre, who taught movement, and the avant-gardist director from the Agitational Theater, Igor Terent'ev, who gave instructions in diction.17
Serious efforts to turn the club circle into a theater studio with its own budget for rehearsals, scripts, and props began in 1924. The city Politprosvet division first took charge of the discussion, but conflicts quickly emerged over just who would head the circle: Sokolovskii, who had been responsible for the work thus far, or the more experienced theater worker Grigorii Avlov, from the city's Politprosvet division. But the most difficult problem proved to be money. Even after both the city government and the Komsomol had given approval for the venture, no funds materialized. The members of the Gleron circle took matters into their own hands. Turning to their protector Piotrovskii, who also worked for the Leningrad division of Sovkino, the state film trust, they volunteered to play bit parts as petty thieves and hooligans in the film The Devil's Wheel (Chortovo koleso). This film, directed by Grigorii Kozintsev and Leonid Trauberg, had a scenario by Piotrovskii. Their collective earnings—four hundred rubles—was used as capital to launch their first full-length play.18
The Gleron circle theater was officially reincarnated as TRAM in the fall of 1925 with the premiere of Crazy Sashka, based on the popular character in TRAM'S living newspaper sketches. Still housed in the Gleron club, the group's original collective had been augmented by new forces chosen through auditions advertised throughout the city. By the opening, TRAM had assembled a support staff that significantly expanded its abilities to stage a polished production. The set designer Roman Ianov, with a newly minted degree from the Technicum of Stage Arts (the future Leningrad Institute of Theater, Music, and Cinema), was involved in creating the scenery. The constructivist set was built by the young artist Alexander Zonov. The musician Nikolai Dvorikov wrote the score, and the caretaker for the Gleron House, Ekaterina Strazd, known to participants as "Aunt Katia," was placed in charge of wardrobe.19 Certainly, in the modest posters for the production, which listed the director, set designer, music director, and costumer, it appeared that the impecunious Komsomol collective was following in the footsteps of more established theaters. But the theme of TRAM'S first play, which built on its living newspaper, revealed its origins in the theater of small forms.
#### Staging Soviet Youth Culture
From the outset, TRAM members made it clear that the joys and sorrows of Soviet youth would be their subject matter. That meant depicting a wide range of social types, from the hooligan demi-monde to the children of White Guardists to upright Komsomol members. However, TRAM plays did not stop here. They also examined the tensions within the Komsomol milieu itself, showing unsavory bureaucrats, self-absorbed young women, and predatory young men. In the words of one reviewer, TRAM saw itself as fighting "on the barricades for a new life."20
As the work of Anne Gorsuch has shown, the future of Soviet youth was a major preoccupation of the Soviet regime during the 1920s. On the one hand, young people were valued for their enthusiasm and supposed pliability, which would allow them to be educated into the first true generation of Soviet men and women. On the other hand, they were feared for their volatility and lack of rational maturity. Young people, for their part, were not so easily molded into ideal Soviet citizens. During NEP a range of conflicting urban youth cultures took shape. They included the dean-living Komsomol, with their anti-alcohol and anti-religious campaigns; alienated hard-core Communists, who saw NEP as a disappointing retreat into petty-bourgeois life styles; a hooligan underworld enamored of violence; those who used the consumption possibilities of NEP to pursue fashion and imported Western culture; and a segment who rejected the anti-religiosity of the Bolsheviks to found their own new communities based on religious faith.21
TRAM plays acknowledged the existence of these complex urban cultures beckoning to Soviet youth. Although their purpose was obviously to draw young audiences to the "correct" solution, the upstanding life of the Komsomol, the main characters in TRAM works are most often people caught in the middle between two worlds who feel the attraction of other possibilities. This characteristic distinguishes TRAM plays from earlier agitational works that show simple conflicts between good and evil. The entirely good Komsomol characters in TRAM plays are often not appealing; rather, the works showcase conflicted figures who eventually lose their doubts and recommit themselves to the Communist cause. Of course, this left open the possibility that audiences might sympathize with characters in their less-than-perfect form.
The first TRAM play, Crazy Sashka, depicted the confrontation between the hooligan underworld of NEP and the healthy, purposeful Komsomol. Sashka Chumovoi, the central figure, was part of both worlds. He was drawn to the hooligans because of old friendships and because his girlfriend, Klavka, remained in that milieu. At the opening of the play, the hooligans are holding a noisy party and also planning a smuggling venture. Their disreputable antics are contrasted with the activities of the Komsomols, headed by Sasha Delovoi (literally, "Energetic Sasha") and his girlfriend, Niura. Rather than engaging in a drunken brawl, the Komsomols are exercising, singing, and playing basketball. This contrast is an intentional one between dark and light—the "good-for-nothing, meaningless stupidity of the rabble [shpana]" opposed to the cheerful aspirations of Komsomol members.22
Sashka Chumovoi has promised to join in a Komsomol venture, to crew on a Soviet ship heading for China. Before shipping out, he comes back to his old crowd to pick up Klavka. This detour, which includes some fistfights with a rival for Klavka's attention, makes him late for the Komsomols' departure. Sashka is determined to make the boat and Klavka is equally determined to stay with him, so she decides to follow along, disguised as a boy. Unfortunately, she makes such an unconvincing boy—constantly powdering her nose and wearing her bell bottoms backward—that Sashka worries that they will be stranded somewhere on the journey. So he and Klavka intentionally pick a fight with their hosts, are placed in the hold, and escape. They wind up on the border of Russia and Estonia, which they somehow mistake for America. There they witness a smuggling attempt by Sashka's old pals, which convinces Sashka once and for all that they are a bad lot. In the meantime, Sasha Delovoi has followed the pair and brings the border guards to arrest the smugglers. In the grand finale, Crazy Sashka gives up his ties to the life of crime and embraces the Komsomol collective, this time without his troublesome companion, Klavka.
As was often the case in TRAM plays, Sashka Chumovoi is a much more arresting figure than his clean-living counterpart. Chumovoi, who had fought in the Civil War, is bored by the dull routine of daily life and spends much of his time "making up for lost sleep."23 At crucial intervals he recounts his war stories, adventures of questionable veracity in which he plays a leading role. In one such tale he saves Trotsky's life, recounting a scenario with detective-thriller qualities. Trotsky discovers himself alone in an exposed position on a bad horse facing five White soldiers. Suddenly, Sashka appears and Trotsky appeals to him as if to an old friend. "Hey, Sashka," he calls in the familiar voice, "They are overtaking me, come help!" "Even if I die," Sashka tells his hooligan friends, "I will save the army." Sashka puts Trotsky on his own horse, leading the Whites to chase Sashka instead. Trotsky and the army were rescued, but Sashka modestly refused credit, even turning down a medal.24
According to one TRAM member, Roman Ianov, who provided an assessment of the play in the Leningrad journal Worker and Theater (Rabochii i teatr), the character Sashka Chumovoi suffered from "pinketonovshchina," an exaggerated interest in detective fiction ("pinkertony," in Russian). It was Ianov's hope that the funny story and its edifying conclusion would lure worker youth away from popular films, which "very often poison young viewers with detective stories and petty-bourgeois venom."25 Of course, just the opposite might have been the case—Crazy Sashka could have won an audience because it drew on familiar forms of popular entertainment, with viewers focusing on the wild antics and improbable, humorous plot rather on the last minute transformation of the hero. Even a team of earnest rabkor reporters commented that the performance's greatest asset was that it was fun to watch.26
For whatever reason, because it was successful transformative propaganda or because it was a fun evening's amusement, Crazy Sashka was a big success for TRAM. It was performed over fifty times in TRAM'S first season—a huge run for a club performance.27 According to Pavel Marinchik, even the city's street youth were fans. Rumors had spread through the city that Sashka Chumovoi was "svoi"—one of their own. A gang of rowdy youth arrived on opening night, a special event for the local press and Komsomol dignitaries, and were miffed that they were not let in to see the play. They stormed the doors, broke windows, and threatened a major disturbance if they were not given seats. They were only placated when TRAM leaders offered a special performance just for them.28
After the debut of Crazy Sashka, TRAM members prepared a number of plays in the late 1920s that touched on contemporary problems of Soviet young people: the Komsomol in revolution and Civil War in Zor'ka (1926) and Rowdy Cohort (Buzlivaia kogorta, 1927); struggles in the workplace in Factory Storm (Fabzavshturm, 1926) and Call the Factory Committee (Zovi fabkom, 1928); and the alienation of young people unable to make peace with NEP in Work Days (Budni, 1926). Four of the plays— Meshchanka (1926), The Days Are Melting (Plaviatsia dm, 1928), Happy Hillock (Druzhnaia gorka, 1928), and The Thoughtful Dandy (Klesh zadumchivyi, 1929)—take love and marriage within the Komsomol as their central problem.
The threat of the Soviet underworld is a recurring subject of TRAM plays. In Crazy Sashka hooligans are more buffoonish than dangerous. However, in TRAM'S next work, Factory Storm, written by Dmitrii Tolmachev, hooligans are more disruptive. The bad boy in the play, a drunken worker who shows up late and steals factory materials, has ties to the local gang. When he is kicked out of his job, he arranges for his friends to beat up the Komsomol hero who had orchestrated his dismissal, nearly leading to the hero's death.29 An even more dangerous underworld emerges in Work Days. In this work a disaffected Komsomol member loses interest in an upright Komsomol worker and falls in love with the daughter of a specialist, whom his colleagues suspect of shady dealings. Their suspicions are correct; she is in fact the member of a White Guardist family. Both her father and brother are secretly trying to overthrow the Soviet regime. Although her sympathies lie with her boyfriend, she is unable to inform on her own family. In the process, she entangles her boyfriend in a conspiracy that threatens his membership in the Komsomol.30
A scene from the Leningrad TRAM play Work Days (Budni). Bakhrushin State Central Theatrical Museum. Reproduced with permission.
The appeal of materialist consumption in NEP culture is also an important subject in TRAM plays, usually presented through female characters. The troubled heroine of Meshchanka is revealed as an unreliable figure when her friends discover powder in her purse—her use of makeup showing that she could not be a politically conscious individual.31 She is not the only vain woman in the play; her friends' pursuit of style has led them to get their hair scorched by an incompetent beautician.32 While male characters often ridicule women's interest in fashion, numerous Komsomol men in TRAM plays look for affection from stylish women outside of their political milieu. This causes the rejected Komsomol leader in Work Days to worry that "perhaps we are ourselves to blame... we factory girls are just too plain."33
While the appeals of fashion tempt Komsomol women, drunkenness is the most serious problem for the men. Drink drives them to bad deeds, like spending their Komsomol dues or beating up their colleagues. Minor male Komsomol members in TRAM plays constantly challenge the prohibition on alcohol at social gatherings, insisting that drinking would make them even more pleasurable. In Meshchanka, one male character who is particularly critical of women's interest in finery tries to convince them that his taste for vodka does not qualify as a vice. A party with only tea is a "bourgeois affair." By contrast, "Vodka is not a bourgeois drink; it is loved by all classes."34 Very few female characters indulge in alcohol, but those that do are a particularly bad lot, leading men astray and leaving broken homes in their wake.
The widely available but, for TRAM members, disreputable commercial amusements of NEP culture are another important subtheme. When the Komsomol collective in Work Days want to find the hero and his frivolous girlfriend, someone suggests that they begin their search at several local restaurants; if they come up empty-handed, they should then case the yacht club.35 The evil woman in Meshchanka is said to be "as beautiful as Mary Pickford," a reference to the popularity of American film in NEP and its troubling influence.36 The extremely conflicted female heroine of The Thoughtful Dandy is drawn to a number of highly questionable activities. She secretly takes dancing lessons, where she learns foreign dance steps. She has considered suicide and even keeps a store of poison. In an act of desperation, she leaves her husband and home for a religious commune, an allusion to the strong religious communities that were gaining youthful followings in the 1920s.37
These conflicted heroines and heroes are tempted by the distractions of alternative youth cultures, yet in the end they are always drawn back to the Komsomol community. TRAM plays offer a sense of the Komsomol club as a model for a new kind of family, responsible for the nurturing, care, and discipline of its members. Pavel Marinchik's play Meshchanka is an example of a TRAM family drama in which the Komsomol family overcomes obstacles to create a new sense of commitment and group responsibility. It plucked its topic directly from the Komsomol press. In the mid-1920s, as part of a general investigation into why so few young women participated in the Komsomol, activists began drawing attention to the fact that women who had once taken part tended to drop out once they married. They became "petty bourgeois," giving their attention to home life rather than the public sphere. This turned them into "meshchanki," petty-bourgeois women, a very derogatory term in the Soviet lexicon. The play set out to analyze this problem without offering definitive solutions.38
In his memoirs, Marinchik reveals that he was inspired to write the play in part by the intervention of Komsomol women, who complained about men's unwillingness to help out in the home. He quotes a member of the Skorokhod factory TRAM group, Niura Petrunina, who had spoken at a local Komsomol debate: "It is very easy to spout off charges. It is very easy for you to call us meshchanki. But when you go home, who fixes your dinner? Who washes your shirt? Who feeds your child? Your mother or your wife. And openly admit it, you politically advanced [ideinye] comrades—do any of you help to surmount the unavoidable difficulties of family life?"39
In the play, two young activists meet in a Komsomol club and decide to marry. Niura Panova is an important organizer at her factory; Mitia Panov belongs to both the Komsomol and the Communist Party and also leads his factory committee. The marriage, embarked upon joyfully, quickly changes their lives for the worse. Overburdened by new responsibilities, Niura applies to leave her Komsomol cell. For this she is attacked by the (male) Komsomol leaders, publicly humiliated, and ostracized as a meshchanka. At the same time, her marriage has become shaky. Mitia does not think that Niura is doing a good job running the household, and he is also embarrassed by her transformation into a homebody. As a result, he quickly forms a romantic liaison with another woman. Faced with public ostracism and a failing marriage, Niura is brought to the brink of suicide.
No one party is given sole responsibility for this unhappy situation. Niura is shown to be easily undone by the demands of marriage, quickly (and inexplicably) changing from a lively activist to a distraught and passive wife. "I just cannot go on this way," she tells her best friend, explaining her decision to leave the Komsomol. "During the day we are at the factory, then at the club, in the collective, and we only see each other late at night. He works very hard and wants to relax. I must, I am obligated to help him."40 The husband also must share the blame. Instead of helping his wife, he demands a hot dinner and nicely ironed shirts, insulting her when they are not prepared to his liking. He flaunts his new girlfriend and stalks out when his wife is reduced to tears, saying he cannot stand her "petty-bourgeois scenes."41 Nor does the Komsomol organization come off very well. Members discover Niura's intention to leave by violating her privacy, rummaging through her purse. The Komsomol cell chief does not think it is necessary to investigate why her marriage might have driven her to take such a step. "She herself is to blame," he opines. "She turned into a baba, a meshchanka. We don't need her kind."42
The final scene, when Niura is saved from a suicide attempt at the last minute, offers an interesting critique of male sexual behavior. A former admirer of Niura's, a returning Red Army soldier, lectures not only Niura's friends and husband but also the heartless group leaders. He criticizes men who pretend to believe that women are their equals but at the same time brag about their sexual conquests. He then turns to the audience and asks, "And you gathered here, what do you say? Should we struggle to end the barbarous, slipshod relations between young men and women?" A unanimous "We should," obviously designed to include the audience, is the last line of the play.43
Meshchanka offers a clear answer to the question about how the family can be integrated into the community: it must be made porous to the Komsomol collective, subsumed by the Komsomol family. In the words of a reviewer in the journal Worker and Theater, "A sensitive and comradely collective, that is the first element of normal domestic relations."44 Collective oversight is needed to curb both male sexual excess and female masochism. The Komsomol's intervention made the husband regret his rash behavior; it saved the wife from suicide. But the price was a loss of privacy—opening domestic relations between wife and husband to public oversight.
In his fascinating study about sexual discourse in the 1920s, Eric Naiman argues that the preoccupation with sex in Komsomol political and artistic literature was part of an effort to gain greater control over the nation's youth. Debates about the "problems of daily life" fascinated young people, resulting in "a process in which discussion was first eroticized so that it could ultimately be more effectively politicized."45 Although TRAM works do not contain the same level of sexually explicit language as more sensational works from the 1920s, they too present situations where intimate personal quandaries are ultimately solved by the intervention of the political community, effectively erasing any realm of private decision making. TRAM plays abound with invasive practices— eavesdropping, rifling of wallets and purses, and the public reading of diaries. The secret police are invoked as heroes who know everything and keep a watchful eye on the supposedly private affairs of young people. Thus, the new family is presented as a panoptical device, coming to know all so that it can solve all.
With this said, it is difficult to judge what messages viewers took from TRAM plays. By presenting a wide range of youthful types, these works offered audiences a chance to identify with the anti-heroes or with the central characters before they made their final transformation. The plays showed not only the positive aspects of youthful togetherness but also offered insights into the vulgar, bossy, and cruel side of the Komsomol collective.46 It was Crazy Sashka—not his upstanding counterpart—who had a following among Leningrad street youth, and we can assume it was his brazen and swashbuckling side, not his last-minute transformation to the clean living Komsomol, that attracted his fans. Similarly, the new family structure proposed in TRAM works subordinated the individual to the invasive Komsomol community. However, some characters resisted integration and showed the high price of subordination. By taking on the complexities of youth culture, TRAM sketched out a variety of possible trajectories and thus ensured that a wide range of Soviet youth would see themselves and their friends reflected in its works.
#### "TRAMism"
The Leningrad TRAM presented itself as a theater of a new type, a "bridge" between amateur and professional theaters.47 Rooted in the agitational theater of small forms, it embraced topical themes, thus meeting its audience's demand for works that reflected on the challenges and struggles of contemporary life. Initially, TRAM also rejected the professionalization of its worker actors. Participants kept their factory jobs in order to remain close to the problems, enthusiasms, and language of contemporary youth. TRAM'S application form, distributed through the Komsomol, pointedly asked for employment information and evidence of union membership.48 TRAM'S proletarian persona was already apparent in the opening march of Crazy Sashka: "In the mornings we are always there by our machines, but in the evenings our job is TRAM!"49
Sokolovskii insisted that they would not follow the guidance of any established theatrical training program; instead, they would devise their own methods. Along the same lines, they would not use scripts created by professionals because such works could not capture the milieu of worker youth. Moreover, TRAM plays would be a collective creation, with input from all members. Crazy Sashka was an example of such a group process. Although the script bore Arkadii Gorbenko's name, the text was worked out in a series of late-night discussions between TRAM'S young writers and Adrian Piotrovskii.50 His two rooms in a communal apartment, filled with pictures of ancient Greek playwrights, became the unofficial meeting place where TRAM members read, discussed, and hammered out their new plays.51 Thus, while individual authors took final credit for TRAM plays, writers, participants, and observers alike insisted that these group discussions were their main creative force.
For Piotrovskii, TRAM was the embodiment of all that was positive in amateur theater engendered by the revolution. His perceptions of the new theater fit into a grander theory about the continual historic tension between amateur and professional theatrical forms. In his view, amateur theater emerged organically from the rituals and festivals of the lower classes; it then challenged the dominant modes of expression in established theaters of the ruling class. Such a challenge was currently taking place in Soviet Russia, as working-class groups criticized the presentational style and repertoire of professional theaters that had not changed significantly since the revolution. "By constructing daily life [byt], by organizing it in a festive way, the working class with its 'amateur,' 'autonomous' performances lays the foundation for a radical reexamination of theatrical forms and marks the way to a theater of the future," Piotrovskii insisted.52
Both Piotrovskii and the TRAM director Sokolovskii were convinced that TRAM had a unique contribution to make to Soviet cultural life. Together, they began to formulate the social and aesthetic principles that they believed distinguished Komsomol theater from other forms of amateur artistic creation. As opposed to many other amateur theaters where the aim was primarily entertainment, TRAM articulated a clear political goal. The group's purpose was to change daily life, rather than to describe it. Tellingly, they referred to the collective not as a theater but as the agitational arm of the Komsomol.53
From this followed new roles for the TRAM actor. Rather than seeing themselves as passive observers of Soviet life, TRAM participants conceived of themselves as activists who drew their subject matter from factories and dormitories and aimed to influence the behavior of viewers. Their goal was to recreate the language and movement of present-day youth and to pose the problems of young people to the audience. Although this approach demanded training, it was directly opposed to the methods endorsed by Konstantin Stanislavsky, who encouraged his actors at the Moscow Art Theater to enter the emotional world of the characters they represented on stage. In an intentional contrast to Stanislavsky's ideas, Sokolovskii maintained that the TRAM participant was more an agitator than an actor of the old school.54 The parts TRAM actors played were social masks, beneath which the faces of worker youth were clearly visible. "While acting on the stage," wrote Piotrovskii and Sokolovskii, "the TRAM Komsomol actor does not give up his basic character of a worker youth; he continues to express his social and class nature. He does not take a 'passive' position, embodying his role, but rather remains 'active,' as if analyzing and judging the character."55 TRAM actors offered a critical interpretation, without entering into the character's emotional world.
This method of acting, which Piotrovskii called "the denial of illusion," articulated in the late 1920s, bears a close resemblance to Bertolt Brecht's concept of alienation, first developed in the 1930s.56 James von Geldern has speculated that TRAM might have influenced Brecht indirectly; the great director could have heard about these concepts through his friend, the Russian avant-garde author Sergei Tretiakov.57 Although this remains speculation, certainly there was much in common between the two methods. Martin Esslin's description of Brechtian acting techniques could serve equally well for TRAM performers: "The character who is being shown and the actor who demonstrates him remain clearly differentiated. And the actor retains his freedom to comment on the actions of the person whose behavior he is displaying."58
TRAM works were synthetic creations uniting a variety of art forms, including music, song, dance, and acrobatics. These principles grew from Piotrovskii's conception of the united artistic circle. TRAM productions were sometimes composed directly during rehearsals, with songs and dances evolving as actors learned their lines. The literary script was not the most important element of the play—nor was the actor the primary bearer of meaning.59 TRAM works were not bound to a unilinear presentation of the story. They used flashbacks; they proposed alternative endings. "It is not hard to see the ties between TRAM'S many-layered presentation of events and film montage," declared Piotrovskii. Indeed, he saw TRAM as only one piece of evidence pointing to what he called the "filmization" (kinofikatsiia) of all the arts.60
TRAM authors did not offer simple stories with single definitions of good and evil. Instead, they opposed obvious endings and presented a "dialectical structure."61 According to Piotrovskii and Sokolovskii, the function of a TRAM play was neither to tell a simple narrative nor to reveal the inner thoughts and feelings of the characters; rather, it was to illuminate the contradictions inherent in Soviet life and to depict the internal tensions of characters themselves.62 "Episodes are not linked in the sequence of events," explained Piotrovskii, "but as elements of a unique 'polemic,' as supporting or opposing sides of an argument."63 This attempt to depict problems as "many-layered, capacious and many-sided" complicated TRAM'S agitational role. Too many Soviet plays, insisted the two leaders, made the tensions of Soviet life seem minor: "They often show observers' conclusions in a simplistic and onesided way compared to the contradictory complications of our reality."64 TRAM'S goal was to heighten those tensions and to make the audience face the difficult choices they often confronted in their lives.
During the waning years of NEP, the Leningrad TRAM won an enthusiastic following in its home city. Not only was the original TRAM theater at the Gleron House a popular site, but TRAM cells [iadra] opened in eight city factories, including the Skorokhod plant, the Vera Slutskaia factory, and the Elektrosil plant. It also inspired groups in Leningrad province.65 TRAM expanded to Moscow in 1927, first beginning in several small affiliates in different parts of the city. The Moscow Komsomol sponsored a central city organization in late 1927, under the leadership of a student from the Meyerhold Theater.66
As TRAM began to spread and attract attention in the national press, advocates began to define TRAM as a mass movement, as "TRAMism," something that would eventually lead to a new kind of professional theater. TRAM participants attempted to set themselves apart from traditional academic theaters, which they charged had not changed sufficiently since the revolution. TRAM members were most critical of Stanislavsky's theater and theatrical method. Sokolovskii referred to the Moscow Art Theater as the "Theater of Bourgeois Youth."67 It had played an important role in reorienting theater under the old regime but now had become isolated from the external world.
TRAM also tried to set itself apart from the theatrical avant-garde, which proved a more difficult task. By abandoning straightforward narrative structures and endorsing non-realistic acting and staging techniques, TRAM theater was clearly indebted to Meyerhold's methods. Nonetheless, TRAM members insisted that while the left avant-garde performed works with revolutionary themes, their presentational style was often difficult for new viewers to understand.68 TRAM plays, by contrast, contained the living language of worker youth. TRAM presentations were not unilinear; instead, they followed the Marxist method of dialectical materialism.
These grand claims did not go without criticism. In a discussion about TRAM'S role that erupted in the Komsomol press in 1927, many skeptics came forward to challenge the ambitious theater. They asked just how TRAM actors could hone their craft if they were still employed in the factory. The entire premise of TRAM organization was wrong, according to one author. Soviet young people were not particularly interested in special plays about youth; instead, they wanted good plays, pure and simple. The collective process—whereby all members participated in every aspect of the performance, no one was devoted to the process full time, and professional help was discouraged—could never produce first rate work.69 These critics claimed that Soviet theater would be much better served if talented young actors and directors entered existing training programs; in the process they would help to proletarianize all of Soviet culture.70
Beneath these aesthetic and organizational arguments were strong political undertones. Narkompros leader Robert Pel'she warned that the concept of TRAM "smelled like syndicalism," alluding to oppositional tendencies that threatened Communist Party control. The entire educational system was designed to serve worker youth. Why was a special group necessary?71 And with a reference to former battles within the Komsomol over its relationship to the Communist Party, another critic asserted that "there is not and cannot be a special path for youth."72
Those in favor of TRAM intoned that professional theaters were ignoring the problems of youth and that a new system was needed to represent their ideals. Because of its contemporary themes, TRAM was also significant in enlivening the political-educational work of the Komsomol as a whole. In the process, the theater had created a following of viewers who regarded TRAM as their own. Since the Komsomol leadership remained firmly behind the youth theater, TRAM'S future was not really in doubt. At the Fifth Komsomol Congress in March 1927, the organization resolved that it was essential to offer worker youth theaters their support.73 However, to some degree TRAM also integrated the most significant criticism raised in this debate. In the spring of 1928, the Leningrad TRAM professionalized, freeing its young members from their factory jobs.
Gaining professional status would seem to exclude TRAM from the purview of this book, where the focus is amateurs. However, the theater was only "professional" insofar as its actors now earned meager salaries. Financial support was so low that many took pay cuts to join TRAM full time.74 The theater was certainly not integrated into the prestigious network of academic stages, like the Moscow Art Theater, which received support directly from Narkompros. Nor did TRAM offer official training programs. In essence, TRAM actors were "praktiki," experts without credentials who had learned their skills on the job rather than through formal educational channels.75
Moreover, the Leningrad (and later Moscow) professional troupes remained important to amateur stages as the self-proclaimed link between professional and non-professional theater. Participants vowed to become "professionals of a new type." Piotrovskii even insisted that the term "professional actor" did not really describe their new roles. What the young people had become were professional agitators for the Komsomol, leading discussion groups and devoting themselves to political education.76 These declamations notwithstanding, TRAM actors in Leningrad had now removed themselves from one of the problems of the amateur circles from which they had emerged. They no longer had to struggle to balance their theatrical activity with their daily working lives.
#### The Vanguard of Amateur Art
TRAM achieved a national following during the early years of the First Five-Year Plan (1928–32), a period of rupture so extreme that Stalin called it the "Great Break."77 As the state launched programs of breakneck industrialization and coerced collectivization, radical cultural projects were initiated at a feverish pace. City planners turned against the idea of cities; legal experts envisioned the withering away of law. These iconoclastic dreams had a profound effect on the broad array of amateur stages, which began to use cafeterias and dormitories as stages to enact disruptive social dramas (see chapter 5). For TRAM, this period of experimentation brought it even more public attention. Its own oxymoronic vision of a professional theater that disdained professionals fit the spirit of social leveling initiated by the plan.
As many scholars have argued, the First Five-Year Plan was to some degree a youth rebellion.78 Given the general privileging of youth, it should come as no surprise that TRAM—a theater by and for young people—would gain notoriety. Its political guardian, the Komsomol, won much greater visibility as an initiator of cultural campaigns. Not only did the Komsomol attack conventional educational programs, it also waged a war of words against the citadels of high culture. Both the Bol'shoi Theater and the Moscow Art Theater faced stinging attacks in the Komsomol press for their generous government funding, non-proletarian social composition, and outdated, anti-revolutionary repertoires. As the Komsomol assaulted these remnants of the old world, it championed new institutions like TRAM and attempted to win them greater financial support.79
The Leningrad TRAM's shift to professional status gave it much more national visibility. With support from the Komsomol, the theater made its first tour to Moscow in the summer of 1928, which helped to popularize TRAM plays and methods. It took Call the Factory Committee, Happy Cohort, and a new play, The Days Are Melting (Plaviatsia dm), which examined the strains experienced by a young Komsomol couple when a child was born.80 The group was met by enthusiastic audiences everywhere it went: in Orekhovo-Zuevo, a textile town not far from Moscow, they had to add extra performances; in Moscow itself, nearly twenty-five thousand people showed up for TRAM productions.81 As the industrialization drive began in earnest, TRAM collectives spread throughout Soviet territory. Only a handful of TRAM circles existed outside of Leningrad before 1928 but some observers counted up to seventy by the end of the year and three hundred by 1932.82 The Leningrad TRAM and its repertoire inspired emulation across the nation, as new factory and city-based TRAM organizations took shape. The vast majority of these new theaters, spread from Baku to Magnitogorsk, were amateur circles.83
TRAM began to attract the attention of national cultural institutions. At the Conference on Artistic Work among Youth, which met in Moscow in the summer of 1928, Anatolii Lunacharskii, the head of Narkompros, called TRAM a model for building a new socialist theater.84 The following month, he praised TRAM'S work in the nation's newspaper of record, Pravda. In a long article, Lunacharskii traced the theater's history from a Komsomol club to its recent visit to Moscow, giving particular attention to The Days Are Melting. The commissar of the arts also gave TRAM an aesthetic endorsement by linking it to the work of Meyerhold: "It is no accident that the great master Meyerhold expressed his admiration for these performances, and also no accident that Mikhail Sokolovskii, TRAM'S organizer, willingly admitted that TRAM drew broadly on the work of Meyerhold."85
A sketch of the hooligan figure in the TRAM play Call the Factory Committee (Zovi fabkom). Leningradskii TRAM v Moskve (Leningrad: Izdanie Gostrama, 1928).
A scene from the Leningrad TRAM play The Thoughtful Dandy (Klesh zadumchivyi). Bakhrushin State Central Theatrical Museum. Reproduced with permission.
With the onset of the First Five-Year Plan, TRAM plays became even more topical, addressing the rapid changes in labor organization and daily life demanded by the industrialization drive. In early 1929, the Leningrad TRAM premiered The Thoughtful Dandy (Klesh zadumchivyi), which addressed the many opportunities and distractions for youth as the country underwent rapid industrialization.86 The play introduces facts and figures of industrial expansion, a new element in TRAM works but typical of the writing of this period. "Five years!" exclaims the Komsomol cell leader. "Five years of fast-moving life. And then there will be an army of 150,000 tractors in battle position!"87 At the center of the work is a seemingly model young couple that is almost tom apart by temptations that lead them away from the collective. The wife has been advanced from her factory to study at the university but does not like it there. She is drawn instead to the night life of the city. The husband is entranced by the prospect of earning a lot of money and winning back his partner with the lure of material possessions. Although the play has a happy ending—the young couple come back to their Komsomol group—the work examined some of the personal costs of the rapid reorganization of daily life.88
As the First Five-Year Plan heated up, the Leningrad TRAM also began to intensify the pace of its production and reach out for new subject matter. At the end of 1929, the theater for the first time performed a work by an author from outside the TRAM circle, Alexander Bezymenskii's The Shot (Vystrel), which showed Komsomol activists attacking bureaucratic shortcomings in the Communist Party. It also began to move outside of the urban world of Komsomol youth, staging two works on collectivization and class struggle in the countryside, Virgin Soil (Tselina) and The Roof (Krysha) in 1931. Piotrovskii's Rule Britannia (Prav', Britaniia, 1931) moved its attention to a foreign setting for the first time.
By 1929, TRAM had many branches throughout the country and began to make plans to establish a national organization framework. Local representatives came to Moscow for the first national TRAM conference, intending to articulate standardized organizational guidelines and set up a national governing board, the TRAM soviet.89 The gathering featured a congratulatory greeting from Lunacharskii, who was "made younger by 35 years and pronounced an honorary TRAM member."90 After this cheerful beginning, serious disagreements emerged as the rapidly expanding local groups tried to discover common principles that would unite them on a national level.
At the national gathering, the Leningrad TRAM organization began to feel some of the contradictory pressures emerging from its success. What kind of theater was it—the vanguard of the amateur theater movement or a new kind of professional stage? Sokolovskii, TRAM'S forefather, chose to stress its amateur heritage. "We are not a theater and do not want to be one," Moscow newspapers quoted him as saying. "Instead we are a part of the Komsomol aktiv, taking part in the active battle for our class ideals, for the creation of socialism."91 By appealing to the original mission of the Leningrad TRAM, he showed that he was not entirely comfortable with its new role as a professional stage.
Platon Kerzhentsev represented the Communist Party at the conference. This chameleon-like Soviet bureaucrat had argued for the superiority of amateur theater during the Civil War in his often-republished book, Creative Theater.92 Now he argued in favor of professional standards. He warned that TRAM should recognize that it had a lot to learn from professional stages, even from Stanislavsky.93 The final decisions of this first national event were formulated in the spirit of compromise. TRAM served simultaneously as a school, an agitational cell, and a theater. The militant Sokolovskii announced that TRAM was willing to consider elements of the cultural heritage but would not follow the methods of any single school. The compromise formulation that depicted TRAM as both an organized theater and an agitational group seemed to feed leaders' grandiose ambitions; they would claim a leading role over both professional and amateur stages.
The Leningrad not only expanded its influence over its national affiliates, it also hoped to increase its power over rival cultural groups. In November 1929 it formed a theatrical alliance with the Proletkult, Red Theater, and Agitational Theater in Leningrad. The aim was to give strength to the "socialist assault" against conservative academic stages. "We refuse to follow the formal traditions of the old theater slavishly," the declaration read. "We consider the most serious danger at the moment the idealist tradition emanating from the [Moscow] Theater, which until this point has kept its influence on theatrical activity."94 Endorsing TRAm's claims to more power and resources, the Komsomol's national newspaper, Komsomol'skaia pravda, called for a grand new structure to be built for the Leningrad organization that would house up to two thousand viewers.95
Under the leadership of the Leningrad organization, TRAM groups were ready to claim control over amateur circles as well. At the 1930 national conference, delegates passed a resolution claiming TRAM'S preeminent position. Sokolovskii pronounced, "Only as the vanguard [golovnoi otriad] of all amateur art, only as the active participant and leading brigadier of all restructuring of all armies of amateur artistic forms can TRAM truly become a new, socialist phenomenon in our art."96 TRAM'S links with agitprop brigades, mobile agitational groups sponsored by trade unions, were a dear indication of a shift in its cultural role. The only difference between a TRAM cell and an agitprop brigade, proclaimed Ivan Chicherov, head of the national TRAM soviet, was that the TRAM cell emerged from the factory Komsomol organization, whereas the agitprop brigade was generated by the factory committee. By embracing these burgeoning new circles, TRAM leaders aimed to solidify their claims to lead the amateur theater movement.97
TRAM'S influence was even spreading to other media, with the founding of groups in the visual arts (IZORAM), film (KINORAM), and music (MUZORAM).98 It began to move outside of its urban base and adolescent core constituency, establishing a Kolkhoz TRAM in Leningrad province and a Pioneer TRAM to cater to younger children.99 For the fifth anniversary of the Leningrad TRAM in 1930 the national Komsomol newspaper published a full-page celebratory assessment of its accomplishments, including greetings from German groups that had been inspired by its repertoire and fulsome praise from Meyerhold. According to one commentator, TRAM was one of the few theaters in the Soviet Union that was not suffering a crisis of repertoire because it made its own.100 For a brief and heady movement, all appeared to be within TRAM'S grasp.
#### From Vanguard to Rear Guard
Just as the TRAM movement seemed set to assume a dominant position within the fractious world of cultural politics during the First Five-Year Plan, critics began to line up against the Leningrad organization. Its unstable position between amateur and professional stages made it an easy target from both sides. In the course of this critical onslaught, the Leningrad theater was unceremoniously ousted from its position as the head of the TRAM movement, one of those rapid reversals that were a common feature of Soviet cultural conflicts. By the time the First Five-Year Plan was over, the Moscow circle was the most important TRAM theater in the nation, a status it achieved by renouncing the distinctive artistic positions formulated in Leningrad.
In 1931, the "third and decisive year" of the industrialization drive, the Leningrad TRAM began to face attacks from the amateur circles that it claimed to lead. Many groups, now staging highly agitational works often composed directly at the performance site, argued that TRAM plays were quickly outdated. In addition, they were not specific enough to illuminate local issues. Although TRAM claimed credit for leading agitprop brigades, its leadership was "entirely fictitious," chided one critic.101 TRAM works were also subjected to much stricter ideological criticism. At a discussion after the collectivization play Virgin Soil, one viewer objected to the portrayal of the decision to turn to total collectivization. It seemed to him more a negative step to destroy the well-to-do peasant (kulak) than a positive decision to transform the countryside.102 A representative at a TRAM gathering in 1931 demanded that The Thoughtful Dandy no longer be performed: "It takes place outside of real production, outside of a specific factory or a specific city.... Such plays cannot inspire cadres among viewers to fight for production in their industry or town because the play does not address concrete, specific problems."103 The humorous tone of many TRAM productions was offensive to austere worker critics, who believed that the Soviet Union was fighting a life-or-death battle in its struggle for industrialization.104
Other critics challenged TRAM'S claim to be a new kind of professional theater. Here the most important antagonist was the Russian branch of the Proletarian Writers' Association, known by its acronym RAPP, the most aggressive cultural organization during the First Five-Year Plan. Its members believed that TRAM was not moving in the direction of a professional theater that represented proletarian interests, citing its rejection of established professional training methods and its "formalist" aesthetic principles. The leadership of RAPP linked the TRAM movement to the ideas and practices of the Leningrad Litfront, a dissident wing of the national association of proletarian writers. Litfront members believed that literature had to become more closely tied to life. To do this, writers should abandon traditional plot structures and psychologically motivated characters; instead, they should turn to short sketches and documentation taken from the lives of workers and peasants.105 TRAM theater, with its emphasis on illuminating the problems of youth and its opposition to the techniques of professional theater, did indeed bear similarities to the spirit of the Litfront. The Leningrad TRAM had even performed a work of one of the most vocal members of the Litfront, the playwright Alexander Bezymenskii. By late 1930, RAPP had succeeded in turning the charge of "Litfrontism" into a dangerous offense. This group, RAPP leaders argued, was essentially nihilistic and incapable of creating psychologically convincing characters. Their aesthetic errors were linked to more serious political failings: through rather strained logic, the head of RAPP charged that the Litfront was part of a bloc of highly placed critics of Stalin's social and political policies.106
At a January 1931 conference on theater, RAPP turned these same charges against TRAM. While recognizing TRAM'S important position within the amateur theater movement, RAPP leaders charged that the theater had been led astray by the "rightist" ideas of Piotrovskii, an agent of "the now defeated Litfront."107 These views were elaborated in greater detail in RAPP's major statement on theater, "RAPP's Duties on the Theatrical Front," published in fall 1931. In this lengthy denunciation of all current tendencies in Soviet theater, RAPP charged that the TRAM movement, inspired by the Leningrad organization, was based on faulty and harmful principles. These included the idea that amateur theater was fundamentally different than professional theater, which led TRAM to deny the theatrical heritage. Other serious criticisms included its focus on "class-alien elements within the Komsomol" and its attempt to minimize the importance of the actor and the script in plays.108 The solution was for RAPP to assume control over TRAM and amateur theaters in general, "in order to strengthen the struggle with the contrivances, mechanical methods, and vulgar Marxism permeating the TRAM and amateur theater movements."109
The national TRAM leadership put up some resistance to these criticisms, protesting that RAPP did not understand their position on the art of the past or their relationship to amateur theaters. In addition, they claimed that RAPP, a literary organization, did not really understand the aesthetics of theater, an art form that integrated many different media.110 However, their main strategy was to try to isolate the national organization from the Leningrad group.111 Underscoring this point, the head of the national organization, Ivan Chicherov, laid all of TRAM'S ills at the feet of Piotrovskii. According to Chicherov, Piotrovskii had caused TRAM to abandon a linear plot structure in favor of confusing experiments and excessive improvisation.112 Chicherov also charged that Piotrovskii did not understand the concept of the dialectic at all, even though he had championed the so-called dialectical play. "The Thoughtful Dandy was based on just such a false, mechanical understanding of the dialectic," Chicherov charged. "In fact, it was completely incompre-hensible. Why did a good Komsomol girl, a good young woman, suddenly turn to sectarianism?... There are a whole number of completely schematic, unconvincing, and false elements in the play."113
Fighting to retain leadership, Sokolovskii did not defend his group or accept responsibility for its errors. Instead, he also blamed Piotrovskii. Responding to RAPP's criticism, Sokolovskii admitted that the Leningrad organization had been slow to unmask the errors of Piotrovskii, especially his ideas of dialectical materialism as the aesthetic principle behind its creative work.114 Other Leningrad TRAM members did the same, charging that Piotrovskii had furthered harmful bourgeois ideas, urging actors to ignore psychological depth and to portray social types instead.115
Piotrovskii was drawn into the vortex of self-criticism, confessing at the beginning of 1932 that his ideas about amateur theater were fundamentally flawed. His theories were based on reactionary bourgeois notions—inspired by Viacheslav Ivanov and even Nietzsche. They had reduced class struggle to a struggle between different theatrical schools. Piotrovskii regretted his attempts to isolate amateur theater from the influence of professionals and to undermine the role of the actor within theatrical productions, and he apologized for his endorsement of disjointed, plotless performances. "Instead of raising the mass art of worker youth to the standards of great Bolshevik art, my 'theory' simply impeded its growth. Therefore this idealistic, bourgeois theory served the politically dangerous cause of the class enemy," Piotrovskii con-cluded.116 He soon thereafter quit his position in TRAM.
By abandoning the ideas formulated in the late 1920s, the Leningrad TRAM was unfortunately left without much of a method at all. This became painfully obvious when Sokolovskii wrote and produced his play Unbroken Stream (Sploshnoi potok) in 1932, which was a real departure from earlier work. Based on material he had gathered on trips to new construction sites, it was a straightforward presentation of a production collective's struggle to complete the construction of a dam before the onset of the spring thaw. Stripped of any of the elements that had distinguished TRAM works in the past, the play lacked song, dance, and satire, and hardly any attention was paid to the specific problems of working-class youth.117 Valerii Bliumenfel'd, a Leningrad critic who had been very sympathetic to TRAM in the past, called the play "silent": it sparked no interaction with the audience at all, almost as if the actors played alone without viewers. Bliumenfel'd concluded that in trying to cut itself off from the influence of Piotrovskii and conceding to the criticisms of RAPP, TRAM had in fact rejected its whole heritage. It was turning to the style of Stanislavsky's Moscow Art Theater and isolating itself from the youthful worker audiences that had once been so enthusiastic about its plays.118 Other viewers noted that this play—without the songs and dances of the old productions—made the weaknesses of TRAM actors painfully apparent.119
The overt assault on the Leningrad TRAM gave the Moscow organization, until that point very much in the shadow of the founding group, a chance to establish its predominance. At a national TRAM meeting in July 1931, the leader of the Moscow theater charged that all of TRAM'S problems rested with Sokolovskii and the Leningrad organization. It was impossible to work in a comradely manner with Sokolovskii, he opined. Instead, the Leningrad leader wanted to dominate everything himself. Until 1931, the Moscow TRAM had been very influenced by Leningrad's repertoire. But as the original organization became more and more embattled, the Moscow central TRAM started to shape its own work. It formed a special section for playwrights under the leadership of Fedor Knorre, who had moved from Leningrad to Moscow. Knorre's work Alarm (Trevoga), on the possibility of a coming foreign war, played to very good reviews in 1931 and was the focus of the Moscow TRAM'S 1931 May Day celebrations.120 "Finally," remarked one critic, "this theater has offered a large work."121
The Moscow TRAM'S efforts to set itself apart from Leningrad intensified when the Communist Party moved to reshape cultural politics in April 1932. The famous party pronouncement summarily dissolved self-proclaimed proletarian cultural groups, including RAPP. The decree also marked a major shift toward a cultural policy that was extremely hostile to the original ideas of the Leningrad TRAM. All amateur theater groups were urged to overcome their opposition to professional directors and professionally written plays, to institute training programs that integrated theatrical history, and to provide a better education for actors.122 At a special plenum of the central TRAM soviet called to respond to the party decree, leaders listed TRAM'S past failings, including the idea that amateur and professional theater were inevitably opposed. TRAM must work to attract good professional playwrights and begin to perform classic plays. TRAM'S mistakes, insisted Chicherov, could be attributed to the "Menshevist, liquidationist" views of Piotrovskii.123 The Moscow TRAM used the party's announcement as the occasion for its own internal purge, spearheaded by the Moscow Komsomol.124 Those who supported TRAM'S old methods were ousted from the theater. After a protracted struggle that lasted several months, those in favor of reorganization won. At this point, the new chief administrator invited Il'ia Sudakov from the Moscow Art Theater to become the artistic director, embracing the very principles the Leningrad organization had originally opposed. This move was another step in the consolidation of cultural power in Moscow, a process so vividly depicted by Katerina Clarke.125 It also was an ironic turn for the movement that had once denounced Stanislavsky's stage as "The Theater of Bourgeois Youth."
TRAM was an organization that stirred up controversy. It faced aggressive criticism during the waning years of the New Economic Policy and again during the First Five-Year Plan. Its function as a lightning rod in both eras, demarcated so clearly by historians, indicates its liminal status. It also reveals that the firm distinctions historians have drawn between NEP culture and the beginnings of Stalinist culture are perhaps not so fixed. During NEP, the theater drew criticism from those opposed to its aggressive anti-professionalism and its swaggering cult of youth, both qualities that are usually assigned to the period of the "Great Break." Those close to the professional stage doubted its commitment to art; political overseers worried that its insistence on a unique youth culture was a coded form of political protest.
Criticism continued during the First Five-Year Plan. Now TRAM faced charges that it had not shed the traces of NEP quickly enough. Its heroes remained ambivalent when the times called for decisive action. They considered too many alternatives and were too easily tempted by decadent youth cultures. That is vividly evident in the attack on The Thoughtful Dandy, which addressed the pressures on youth in the industrialization drive. The characters' search for meaning was emphatically denounced as "schematic, unconvincing, and false."126 TRAM plays offered audiences too much room to make up their own minds. Indeed, the "dialectic method," touted by Piotrovskii and Sokolovskii, made it possible for the synthesis viewers worked out for themselves to be very different than what the authors intended. Moreover, TRAM authors were not able to keep up with the broad complex of rapidly shifting issues in this period of rupture. Their skills were limited to the urban milieu that they knew.
In 1928 and 1929, TRAM stood briefly on the cusp of radical change with its attacks on staid cultural institutions and its aggressive rhetoric aimed at specialists. But the Leningrad and central TRAM organizations were inept warriors in the cultural battles of the era. They could barely shape a shared institutional identity, let alone defend themselves from external criticism. When TRAM came under attack, it immediately began to tear itself apart in a frenzy of internal incriminations. Although the assault began with RAPP, the organization's most vicious critics were its own members, who accused one another of dire political failings. The central TRAM, eventually under Moscow leadership, survived only by jettisoning its original leadership and aesthetic principles.
Moreover, the decision of the Leningrad TRAM and other affiliates to turn to theater work full time alienated them from their amateur origins, despite their heated claims to the contrary. Professional TRAM circles had very little in common with newly radicalized amateur stages. As TRAM strove for financial parity with academic theaters, these new cirdes insisted that professionals writers and directors could not hope to express the ideas of the Soviet Union's workers. As TRAM struggled for adequate performance spaces, amateur circles took their works to cafeterias, dormitories, and onto the factory floor. While TRAM theorists dreamed of changing audiences' attitudes through performances, agitprop brigades actually tried to change their actions, extracting funds and pledges of work speed-ups. TRAM had made its reputation on radicalism, on being the most extreme manifestation of amateur theater, but it had to give up this reputation to new groups that called themselves shock workers on the cultural front.
* * *
1. On the TRAM movement, see N. Rabiniants, Teatr iunosti: Ocherk istorii Leningradskogo gosudarstvermogo teatra imeni Leninskogo komsomola (Leningrad: Gosudarstvennyi nauchno-issledovatel'skii institut teatra, muzyki i kinematografii, 1959); V. Mironova, TRAM: Agitatsionnyi molodezhnyi teatr 1920–1930kh godov (Leningrad: Iskusstvo, 1977); L. A. Pinegina, Sovetskii rabochii klass i khudozhestvennaia kul'tura, 1917–1932 (Moscow: Izdatel'stvo Moskovskogo universiteta, 1984), 204–9; Lynn Mally, "The Rise and Fall of the Soviet Youth Theater TRAM," Slavic Review 51, no. 3 (Fall 1992): 411–30; Katerina Clark, Petersburg, Crucible of Cultural Revolution (Cambridge: Harvard University Press, 1996), 266–78.
2. From Sokolovskii's account in "Vecher vospominanii rabotnikov TRAM'a ot 12 maia 1930g.," RGALI, f. 2723, op. 1, d. 534, l. 1; M. Sokolovksii, "U istokov tramovskogo dvizheniia," RiT 29/30 (1932): 11–12.
3. A. S. Bulgakov and S. S. Danilov, Gosudarstvennyi agitatsionnyi teatr v Leningrade (Moscow: Academia, 1931), 30–34.
4. Zograf, "Tvorcheskii put' Leningradskogo TRAM'a," RGALI, f. 2723, op. 1., d. 220, l. 17; N. L'vov cited in "Vtoroi vecher vospominanii," RGALI, f. 2723, op. 1, d. 534, l. 13 ob.
5. Pavel Marinchik, Rozhdenie komsomol'skogo teatra (Leningrad: Iskusstvo, 1963), 42–43.
6. Ibid., 61; "Vecher vospominanii," l. 5; Sventitskii's account in "Vtoroi vecher vospominanii," RGALI, f. 2723, op. 1, d. 534, ll. 9–10 ob.
7. On Piotrovskii's first work with Sokolovskii, see "Vecher vospominanii," 1. 2.
8. Marinchik, Rozhdenie 43–44; "Vecher vospominanii," 1. 2.
9. Marinchik, Rozhdenie, 38.
10. Ibid., 54.
11. "Puti razvitiia Leningradskogo TRAM'a," rgali, f. 941, op. 4, d. 66, ll. 1–2.
12. "Pervoe maia v rabochikh klubakh," ZI 17 (1923): 8; "Dom Glerona," ZI 10 (1924): 19; B. Prokofev, "Dom Glerona," ZI 14 (1924): 19; "Vecher vospominanii," l. 2.
13. A. Garin, "Shestaia konferentsiia rlksm v Moskovsko-Narvskom raione," ZI 7 (1925): 32.
14. A. Senzul', "Gleronovskii karandash," ZI 22 (1925): 21.
15. Pavel Marinchik, "Dalekoe-blizkoe," Neva 11 (1957): 171; idem, Rozhdenie, 62–63.
16. B. S., "Raionnyi klub RKSM im. tov. Glerona," ZI 26 (1926): 21; T. R., "Dom im. Glerona," RiT 9 (1924): 15; "Tsentral'nyi dom komm. vospit. im. Glerona," ZI 24 (1924): 18; N. Nikolaev, "Zhenit'ba intelligentnogo," ZI 29 (1924): 17.
17. "U stanka," RiT 6 (1925): 15. On Knorre's background, see Marinchik, Rozhdenie, 85–86.
18. Sokolovskii's account in "Vecher vospominanii," l. 6. See also the Moscow TRAM leader, Pavel Sokolov's, account in "O piatiletii Leningradskogo TRAM'a," RGALI, f. 2947, op. 1, d. 18, l. 20.
19. Marinchik, Rozhdenie, 68–72; "Vecher vospominanii," l. 6.
20. M. Skripil', "TRAm," ZI 20 (1926): 17.
21. Anne Gorsuch, Enthusiasts, Bohemians, and Delinquents: Soviet Youth Culture, 1921–1928 (Bloomington, Ind.: Indiana University Press, forthcoming).
22. A. Gorbenko, Sashka Chumovoi: Komsomol'skaia komediia, in A. Piotrovskii and M. Sokolovskii, eds., Teatr rabochei molodezhi: Sbornik p'es dlia kamsomol'skogo teatra (Moscow: Gosudarstvennoe izdatel'stvo, 1928), 91.
23. Gorbenko, Sashka Chumovoi, 111.
24. Ibid., 96. See a similar tale with Voroshilov, 111.
25. Roman Ianov, "TRAM," RiT 46 (1925): 13.
26. M. Kudriashev, Aleksandrov, and Mirotvorskaia, "TRAM," RiT 51 (1925): 20.
27. V. Mironova, TRAM, 26.
28. Marinchik, Rozhdenie, 73–74.
29. D. Tolmachev, Fabzavshtorm, in Piotrovskii and Sokolovskii, eds., Teatr rabochei molodezhi, 82–84.
30. P. Marinchik, and S. Kashevnik, Budni, ibid., 56.
31. Pavel Marinchik, Meshchanka (Leningrad: Teakinopechat', 1929), 26.
32. Ibid., 44.
33. Marinchik and Kashevnik, Budni, 35. For more on gender roles in TRAM, see Lynn Mally, "Performing the New Woman: The Komsomolka as Actress and Image in Soviet Youth Theater," Journal of Social History 30, no. 1 (Fall 1996): 79–95.
34. Marinchik, Meshchanka, 45.
35. Marchinik and Kashevnik, Budni, 39.
36. Marinchik, Meshchanka, 17. On American films and their following, see Denise Youngblood, Movies for the Masses: Popular Cinema and Soviet Society in the 1920s (Cambridge: Cambridge University Press, 1992), 50-67.
37. N. L'vov, Klesh zadumchivyi (Leningrad: Teakinopechat', 1930), 38–42, 58–62.
38. Marinchik, Rozhdenie, 121. For an overview of this issue, see Gorsuch, Enthusiasts, ch. 5, and E. Troshchenko, "Devushkav v soiuze," Molodaia gvardiia 3 (1926): 129–35.
39. Marinchik, Rozhdenie, 121.
40. Marinchik, Meshchanka, 33.
41. Ibid., 49, 51.
42. Ibid., 36.
43. Ibid., 64.
44. Eres, "Igrovoi teatr," RiT 49 (1926): 16.
45. Eric Naiman, Sex in Public: The Incarnation of Early Soviet Ideology (Princeton: Princeton University Press, 1997), 101.
46. Piotrovskii and Sokolovskii, "Dialekticheskaia p'esa," in N. L'vov Plaviatsia dni (Leningrad: Teakinopechat', 1929), 4.
47. A. Piotrovskii, "Teatr rabochei molodezhi," ZI 43 (1925): 13.
48. "V priemochnuiu komissiiu teatra rabochei molodezhi," TSKhDMO, f. 1 (Tsentral'nyi komitet VLKSM), op. 23, d. 396, l. 125.
49. Mironova, TRAM, 34.
50. Marinchik, Rozhdenie, 61, 69–70.
51. See the memoirs of these evenings by Piotrovskii's wife, Alisa Akimova, "Chelovek dal'nikh plavanii," in E. S. Dobin, ed., Adrian Piotrovskii: Teatr, kino, zhizn' (Leningrad: Iskusstvo, 1968), 362.
52. A. I. Piotrovskii, "Osnovy samodeiatel'nogo iskusstva," Za sovetskii teatr! (Leningrad: Academia, 1925), 73.
53. A. Piotrovskii, "TRAM: Stranitsa teatral'noi sovremennosti," Zvezda 4 (1929): 142–52; M. Sokolovskii, "V nogu s komsomolom," Komsomol'skaia pravda, 1 June 1928; Piotrovskii and Sokolovskii, "O teatre rabochei molodezhi," in Teatr rabochei molodezhi, 3–8.
54. See Sokolovskii's speech at a meeting of the Leningrad TRAM, 4 March 1929, rgali, f. 2947 (Moskovskii teatr imeni Leninskogo komsomola), op. 1, d. 4, ll. 4–23.
55. Piotrovskii and Sokolovskii, "O teatre rabochei molodezhi," in Teatr rabochei molodezhi, 4–5, quotation 4. See also Novye etapy samodeiatel'noi khudozhestvennoi raboty (Moscow: Teakinopechat', 1930), 98.
56. See Bertolt Brecht, "Neue Technik der Schauspielkunst," in Schriften zum Theater I (Gesammelte Werke 16) (Frankfurt am Main: Suhrkamp Verlag, 1967), 339–88.
57. James von Geldern, "Soviet Mass Theater, 1917–1927," in Bernice Glatzer Rosenthal, ed., Nietzsche and Soviet Culture (Cambridge: Cambridge University Press, 1994), 144.
58. Martin Esslin, Brecht: The Man and His Work (New York: W. W. Norton, 1971), 137, emphasis in the original.
59. See, for example, M. Sokolovskii, "Nemnogo o postanovke," in Zor'ka: V pomoshch' zriteliu (Moscow: Teakinopechat', 1929), 3; I. Chicherov, "O teatre rabochei molodezhi," in I. I. Chicherov, ed., Za TRAM: Vsesoiuznoe soveshchanie po khudozhestvennoi rabote sredi molodezhi (Moscow: Teakinopechat', 1929), 15–16.
60. A. Piotrovskii, "TRAM," Zvezda 4 (1929), 147. On "filmization," see his Kinofikatsiia iskusstv (Leningrad: Izdanie avtora, 1929), esp. 14–25.
61. A. P-skii, "O tramizme," ZI 50 (1927): 2.
62. A. Piotrovskii and M. Sokolovskii, "Dialekticheskaia p'esa," in Plaviatsia dni, 3–9; idem, "Spektakl' o sotsialisticheskom sorevnovanii," in L'vov, Klesh zadumchivyi, 3–4.
63. Piotrovskii, "tram," 147.
64. Piotrovskii and Sokolovskii, "Dialekticheskaia p'esa," 5.
65. V. G-ov, "V bor'be za tramovskoe dvizhenie," ZI 19 (1928): 9.
66. Chicherov, ed., Za TRAM, 28; I. Chicherov, "Za teatr rabochei molodezhi," NZ 28 (1927): 2; "Teatr rabochei molodezhi," ZI 29 (1927): 14; "Akt no. 2087," tsman, f. 2007 (Upravlenie moskovskimi zrelishchnymi predstavleniiami), op. 3, d. 184, l. 3.
67. Ivan Chicherov, Perezhitoe, nezabyvaemoe (Moscow: Molodaia gvardiia, 1977), 176–77. Chicherov was the Komsomol representative in Narkompros and later the head of TRAM'S national soviet.
68. "Teatr rabochei molodezhi: Byt-proizvodstvo-bor'ba," rgali, f. 963, op. 1, d. 1099, l. 4
69. A. Neks, "Protiv TRAM'a," Komsomol'skii agitproprabotnik, 13/14 (1927): 77–79.
70. This discussion lasted through eight issues of the journal, from number fourteen to twenty-one. Each issue presented pro and con positions. Contributors included the author Boris Ivanter; the head of the Narkompros' cultural division, Robert Pel'she; and the director of the Moscow Trade Union theater, Evsei Liubimov-Lanskoi.
71. "Nuzhen li tram? Iz doklada tov. Pel'she na kul'tsoveshchanii pri TsK vlksm," Komsomol'skii agitproprabotnik, 18 (1927): 40–41.
72. Semen Shor, "Tataram o TRAMe," Komsomol'skii agitproprabotnik 20 (1927): 47.
73. "O postanovke massovoi kul'turno-vospitatel'noi raboty sredi molodezhi," in Tovarishch komsomol: Dokumenty s''ezdov, konferentsii i TsK vlksm, v. 1, 1918–1941 (Moscow: Molodaia gvardiia, 1969), 301.
74. "Vtoroi vecher vospominanii rabotnikov TRAM'a ot 15/V 1930," rgali, f. 2723, op. 1, d. 534, l. 19.
75. On praktiki, see Sheila Fitzpatrick, Educational and Social Mobility in the Soviet Union 1921–1934 (Cambridge: Cambridge University Press, 1979), 125, 202.
76. Adrian Piotrovskii, "TRAM i teatral'nyi professionalizm," Rabis 26 (1929): 4.
77. Michael David-Fox argues eloquently that this period should be called the "Great Break" to emphasize its radical nature. I have used the more prosaic "First Five-Year Plan" because that is the appellation most common in my sources. See Michael David-Fox, "What Is Cultural Revolution?" Russian Review 58 (April 1999): 184.
78. Sheila Fitzpatrick has made the most persuasive case for the role of youth in the First Five-Year Plan. See her "Cultural Revolution as Class War," in The Cultural Front (Ithaca: Cornell University Press, 1992), 115–48; and idem, Education and Social Mobility in the Soviet Union, 136–57. See also William J. Chase, Workers, Society, and the Soviet State: Labor and Life in Moscow, 1918–1929 (Urbana: University of Illinois Press, 1987), 256–92, and Hiroaki Kuromiya, Stalin's Industrial Revolution: Politics and Workers, 1928–1932 (Cambridge: Cambridge University Press, 1988), 100–35.
79. See, for example, "Poltora goda na odnom meste," Komsomol'skaia pravda (henceforth cited as KP), 13 November 1928. On the Komsomol in theatrical politics, see Nikolai A. Gorchakov, The Theater in Soviet Russia, trans. Edgar Lehrman (New York: Columbia University Press, 1957), 283–85; Richard G. Thorpe, Academic Art in Revolutionary Russia, unpublished manuscript, ch. 7.
80. N. L'vov, Plaviatsia dni: Dialekticheskoe predstavlenie v 3-kh krugakh (Leningrad: Teakinopechat', 1929). For an investigation of this play, see Mally, "Performing the New Woman," 86–88.
81. On TRAM'S repertoire, see Leningradskii TRAM v Moskve (Leningrad: Izdanie Gostrama, 1928); on its reception, see "Golos rabochei molodezhi," KP 6 July 1928; "Na zavodakh," KP 13 July 1928.
82. V. G-ov, "Vbor'be za tramovskoe dvizhenie," ZI 19 (1928): 9; Mironova, 6.
83. Marinchik, Rozhdenie, 170, 207; F. Knorre, "Moskovskii TRAM," Rabis 26 (1929): 9.
84. A. V. Lunacharskii, "Iskusstvo molodezhi i zadachi khudozhestvennoi raboty sredi molodezhi," in R. A. Pel'she et al., eds., Komsomol, na front iskusstva! (Leningrad: Teakinopechat', 1929), 33–35.
85. A. Lunacharskii, "TRAM," Pravda 8 July 1928.
86. L'vov, Klesh zadumchivyi. Not only was this published as an individual work, but it was also distributed in the journal Materialy dlia klubnoi stseny, which included information on staging, lighting, and music. See Materialy dlia klubnoi stseny 7/8 1929.
87. L'vov, Klesh zadumchivyi, 13.
88. I. Beletskii, "O tvorcheskom puti Moskovskogo Tsentral'nogo Trama," Za agitpropbrigadu i TRAM 1 (1932): 22–24.
89. Novye etapy samodeiatel'noi khudozhestvennoi raboty, 100–104.
90. "Vsesoiuznaia konferentsiia TRAM'ov," KG 3 July 1929.
91. B., "Printsipy i metody TRAM'a," Vecherniaia Moskva (henceforth cited as VM) 3 July 1929.
92. See ch. 1, pages 20, 35, 36. For Kerzhentsev's destructive role in the 1930s, see ch. 6.
93. B., "Printsipy i metody TRAM'a."
94. "Revoliutsionnyi dogovor," ZI 44 (1929): 7. See also "Prazdnik molodogo revoliutsionnogo teatra," Krasnaia gazeta 4 November 1929; Clark, Petersburg, Crucible of Cultural Revolution (Cambridge: Harvard University Press, 1995), 266–73.
95. "Leningradskii TRAM zadykhaetsia v tesnote," KP 2 September 1930.
96. "Novyi tramovskii god," Sbornik materialov k tret'emu plenumu tsentral'nogo soveta tRAm'ov pri TsK VLKSM (Moscow: Teakinopechat', 1930), 9. See also O. Litovskii, "TRAM," Smena 2/3 (1931): 28.
97. For more on agitprop brigades, see ch. 5.
98. "V komsomol'skom kino," Krasnaia gazeta 18 December 1929; "Zasedanie rabochego soveta Moskovskogo tsentral'nogo TRAM' a ot 29 ianv. 1931," RGALI, f. 2947, op. 1, d. 22, l. 5; N. Chemberdzhi, "Muzykal'nyi front TRAM'a," Sovetskoe iskusstvo 27 May 1932.
99. V. Ipatov, "TRAM—udarnaia brigada Komsomola v iskusstve," Klub i revoliutsiia 21/22 (1931): 38.
100. "Piat' let Leningradskogo TRAM'a," KP 2 December 1930.
101. V. Golubov, "Bez rulia i marshruta," RiT 6 (1931): 11. See also E. Lishchinev, "TRAM—khudozhestvennoe orudie marksistsko-leninskogo vospitaniia," Iunyi kommunist 13 (1931): 36.
102. "Obmen mnenii posle prem'ery Tseliny," 10. V. 1930, RGALI, f. 2723, op. 1, d. 535, l. 33.
103. "Orgsoveshchanie TRAM'ov: Utrennee zasedanie 7/VII/ 31 g.," RGALI, f. 2723, op. 1, d. 536, l. 21.
104. V. Petushkov, " 'Krysha' v TRAM'e," RiT 4 (1931): 13.
105. See A. Kemp-Welch, Stalin and the Literary Intelligentsia, 1928–39 (New York: St. Martin's, 1991), 82–89.
106. Ibid., 89.
107. "Za proletarskii teatr!" Sovetskii teatr 2/3 (1931): 1.
108. "O zadachakh RAPP na teatral'nom fronte," Sovetskii teatr 10/11 (1931): 4–10, esp. 8–9.
109. "Za proletarskii teatr!" Sovetskii teatr 2/3 (1931): 1; "Ob'edinim proletarskie sily na teatral'nom fronte," RiT 6 (1931): 14.
110. "Tvorcheskii metod proletarskogo teatra," Sovetskoe iskusstvo 7 February 1931.
111. "Vo fraktsiiu sekretariata RAPP'a," RGALI, f. 2947, op. 1, d. 24, l. 1.
112. I. Chicherov, "Oshibki i nedostatki tramovskogo dvizheniia," 23 June 1931, RGALI, f. 2947, op. 1, d. 32, ll. 3–6. See also idem, "Za boevoi soiuz RAPP'a i TRAM'a," Za agitpropbrigadu i TRAM 1 (1931): 5–10.
113. Chicherov, "Oshibki," ll. 7 ob.-8.
114. "Obedinim proletarskie sily na teatral'nom fronte," RiT 7 (1931): 4; these points are elaborated in M. Sokolovskii, "TRAM na perelome," Sovetskii teatr 2/3 (1931): 17. also I. Chicherov, "Za boevoi soiuz RAPP'a i TRAM'a," Za agitpropbrigadu i TRAM 1 (1931): 5–6.
115. O. Adamovich, "TRAM'u shest' let," RiT 32/33 (1931): 5.
116. A. Piotrovskii, "O sobstvennykh formalistskikh oshibkakh," RiT 3 (1932): 10.
117. [Mikhail Sokolovskii], "Sploshnoi potok," rgali, f. 2723, op. 1, d. 531, ll. 110–61.
118. V. Bliumenfel'd, "Za propagandistskii stil' v TRAM'e," RiT 12 (1932): 14.
119. See Marinchik, Rozhdenie, 240; Zograf, "Puti Leningradskogo TRAM'a," rgali, f. 2723, op. 1, d. 220, 11. 103–4.
120. "Prakticheskii plan provedeniia pervomaiskoi kampanii Moskovskogo tsentral'nogo TRAM'a," rgali, f. 2947, op. 1, d. 26, l. 2.
121. I. Berezark, "Novoe v moskovskikh teatrakh," RiT 8/9 (1931): 12; see also S. Podol'skii, "Dva spektaklia o voine," Sovetskoe iskusstvo 12 March 1931.
122. "O perestroike tramovskogo dvizheniia. Rezoliutsiia TsS VLKSM po dokladu TsS TTRAM'ov," RGALI, f. 2723, op. 1, d. 423, ll. 7–11.
123. "Otkrylsia plenum TsS tRAm," Sovetskoe iskusstvo 3 June 1932; A. K., "tRAm na putiakh perestroiki," Sovetskoe iskusstvo 9 June 1932.
124. "Resheniia sekretariata mgk VLKSM po dokladu tov. Beletskogo," rgali, f. 2947, op. 1, d. 36, l. 1.
125. Clark, Petersburg, 297–307.
126. I. Chicherov, "Oshibki i nedostatki tramovskogo dvizheniia," 23 June 1931, RGALI, f. 2947, op. 1, d. 32, 11. 7 ob.-8.
## 5
## Shock Workers on the Cultural Front
DURING THE First Five-Year Plan, a new kind of amateur theater group emerged known as the agitprop brigade. These small, itinerant circles had a hostile relationship to the drama workshops of late NEP, many of which had begun to devote themselves to honing their performance skills. Agitprop brigades loudly and aggressively rejected professional models and guidance. They not only reclaimed the performance styles of the agitational theater of small forms born during the Civil War and perfected in the early 1920s, they applied them to utilitarian purposes that earlier activists had never envisioned. In the process, participants argued that they were creating new standards for artistic expression. "Not one measure, not one step that does not serve to implement the industrial plan," intoned the Moscow trade union leader, I. Isaev.1
Created by the industrialization drive, agitprop brigades combined voluntarism with coercion. Participants rejected standard theatrical training; their work was task-oriented, put together at the performance site and often without any political oversight. They claimed disdain for established forms of leadership and distrusted all "experts." At the same time, however, members of the brigades saw themselves as conduits for government programs, interpreters who transformed the bureaucratic jargon of state initiatives into a language that average people could understand. Their role as the state's messengers, as enforcers of state initiatives, gave them power over their audiences that they used in oppressive ways. In a public environment suffused with messages about the production drive—with posters, banners, and placards urging harder work and chastising slackers—agitprop brigades were yet another method to incite workers to fulfill the plan.
Not willing to wait for an audience to come to them, agitprop brigades hunted down viewers in factories, in dormitories, and on the street. They took their works to the countryside in a new effort by amateurs to reach beyond the neighborhoods where groups were formed. They even found an international audience, as leftists everywhere, inspired by Soviet industrial expansion in the midst of a global depression, began to emulate Soviet performance styles. Commenting on the predominance of agitational groups in Russia, the British Communist theater leader, Tom Thomas, explained: "Experience has thus shown that this flexible, vigorous, mobile, inexpensive form is the one best adapted to Workers' Theaters in capitalist countries, if they wish to play their part in the class struggle and to be more than working class dilettantes (curse the breed!)."2
To understand the aggressive performance style of these groups, it is useful to return to the concepts of aesthetic and social drama introduced in chapter one.3 Agitprop brigades rejected the conventions of aesthetic drama, the established modes of presenting a story to an audience. They eschewed makeup, elaborate costumes, sets, stages, rigorous training programs, and sometimes even scripts. More to the point, they felt the goal of aesthetic drama, namely to bring about a change of consciousness in viewers, was much too modest. Instead, they wanted to change their viewers' actions. Theirs was social drama of a particularly invasive kind; they demanded to see tangible, measurable changes in their viewers' behavior. The purpose of agitprop brigades, noted the club leader Sergei Alekseev, "was not to provide the laboring population with art, but to mobilize it through art to overcome the difficulties of socialist construction."4
#### The Formation of Agitprop Brigades
As the industrialization drive began in earnest in 1929, many amateur circles located in clubs and factories shaped small, mobile, and politically motivated touring performance groups. They set themselves apart from conventional theater circles, whose goals and methods they disdained. "The leading amateur circles, TRAM cells, and their leaders must struggle against the apolitical comfort and slavish imitation of dilettantish circles," read one of the first pronouncements in favor of these groups. "Artistic agitprop brigades, closely tied to their clubs and factories, should play a leading role in the transformation."5
Front cover of a program for an amateur art competition in Leningrad, May 1928. Harvard Theatre Collection, The Houghton Library. Reproduced with permission.
Back cover of a program for an amateur art competition in Leningrad, May 1928. The caption reads: "Club art is a mighty weapon of cultural revolution." Harvard Theatre Collection, The Houghton Library. Reproduced with permission.
The industrial shock work movement was a major inspiration for agitprop brigades. Begun in embryo in the waning years of NEP, shock work became a widespread phenomenon during the First Five-Year Plan. Young workers, often organized in the Komsomol, banned together to shake up old patterns of production in factories. They attacked the privileges and established labor methods of older skilled workers, developing methods to produce goods more quickly and efficiently. Operating at first without support of the factory management or union leadership, shock workers began to gain more publicity for their efforts by 1928, using Party support to raise overall production quotas.6
In the summer 1929, the national trade union leadership was reconstituted. The head of the organization during NEP, Mikhail Tomskii, had hoped to defend trade union autonomy during the industrialization drive. Perceived as an opponent of Stalin, he was ousted in a bid to make unions more pliable to the Party leader's vision of rapid industrialization.7 These high-level shifts had a direct effect on amateur theaters. Tomskii had supported closer ties to experts and more conventional repertoires. Not only did the new leadership, headed by Nikolai Shvernik, give its support to shock work campaigns at the factory level, it also endorsed a more utilitarian cultural policy. An April 1930 national trade union gathering determined that all cultural work should now be directed toward production. The stress on leisure and relaxation in clubs must come to an end.8 Speaking at the Party congress later that year, Shvernik denounced the apolitical and culturally conservative direction of Tomskii's leadership and tied current cultural activity directly to shock work.9
As shock work gained visibility and acceptance in Soviet factories, agitprop brigades began to make their appearance in significant numbers. Participants in agitprop brigades compared themselves to shock workers in the most literal sense, calling their activities "a method of artistic shock work."10 In the fall of 1929, the Leningrad city trade union organization was the first to embrace agitprop brigades as a method to put art to the service of the Five-Year Plan. "Instead of apolitical, imitative, dilettantish groups, we should embrace artistic agitprop brigades as the basic form of union work," read the final resolution of a city-wide cultural conference.11
Agitprop brigades were composed of young enthusiasts from trade unions, clubs, and factories. Touring work sites and the countryside to drum up support for the industrialization and collectivization drives, they prepared agitational skits and short plays from the raw materials at hand—newspapers, public speeches, and production statistics. Brigade participants were distrustful of professional theater workers and playwrights, who allegedly had no knowledge of daily struggle at the workplace. Instead, they tried to rely on their own experiences as laborers and political activists. With their stress on local experience and hostility to conventional repertoire and training, agitprop brigades clearly harkened back to the use of small forms on stages in the early period of NEP. The links were closest to the action circles of Moscow, with their anti-aesthetic, task-specific orientation. One commentator describing agitprop brigades employed categories used during NEP to distinguish the goals of action circles from conventional drama groups. While drama circles aimed to entertain, action circles and agitprop brigades strove toward "active, practical participation in socialist construction."12 Some of the advocates of action circles, like Moscow's Nikolai L'vov, again became very visible commentators and organizers of amateur groups during the First Five-Year Plan.
Still, there were crucial differences between the small forms of the NEP era and these new configurations. Agitprop brigades were not tied to specific club venues; they might perform on club stages, but they were ready and eager to take their shows on the road. As one advocate proclaimed, "The main arena for circles should not be the stage, but rather the factory shop, the dormitory."!13 Most brigades were fairly small, with around ten members. Club circles might have more than forty members. Although the club groups of NEP expressed open contempt for "entertainment," their work had to also be pleasurable to watch; otherwise, audiences would leave. They incorporated songs, dances, physical humor, and other methods to keep audiences entertained. By contrast, agitprop brigades often played to captive audiences. They concentrated on words to convey their messages and were not interested in their viewers' pleasure.14 Instead, they wanted results—more money for industrialization, higher production figures, more volunteers for political causes. Accounts of agitprop performances read much like industrial production reports, filled with a blaze of figures on the number of appearances, along with statistical evidence about the transformative power of the presentation.15
Agitprop brigades also had different methods and goals from the expanding TRAM movement. TRAM circles, for the most part, still performed plays on stages; they met for rehearsals and did not present works until they felt they were ready for audience consumption. Although TRAM authors tried to stay up to date with their material, their fairly conventional method of first finding or preparing a script, choosing actors for the roles, and then starting rehearsals meant that the subject matter of TRAM plays could not keep up with the rapid twists and turns of government policies. Agitprop brigades, by contrast, could change their scenarios in an instant, to incorporate the very latest news.
So widespread were agitprop brigades by the fall of 1930 that many performed at a competition sponsored by the Moscow trade union organization at the new Park of Culture and Leisure, soon to be renamed Gorky Park. It was timed to correspond with the Sixteenth Party Congress and to show how union artistic groups were turning to productive tasks. Chanting in unison, brigade members delivered a public statement of their goals: "We will transform amateur art into a weapon for our great social construction against the class enemies of the proletariat. We will restructure our work to turn our face to production.... Amateur art, a fighting weapon in the battle for the Five-Year Plan in four!"16
After the Party congress, agitprop brigades proliferated rapidly. Because the circles were ephemeral, there are no reliable composite figures on their numbers or social composition. However, judging from statistics compiled at local and national competitions, it appears that participants were primarily young shock workers.17 Many could trace their involvement in factory cultural work back to the years before the First Five-Year Plan.18 Very spotty figures on gender composition indicate that women constituted some thirty percent of the membership, similar to the overall figure for women shock workers.19 The leader of the brigade at the Sickle and Hammer factory in Moscow said he judged potential members by their standing in the factory—only shock workers with excellent records on the factory floor need apply. Their performance skills received no mention at all.20 Grigorii Avlov, the seasoned Leningrad cultural activist, insisted that agitprop brigades could not be simply renamed factory drama circles (although they often were). "It is quite obvious," he argued, "that an agitprop brigade, which aims to depict questions of socialist construction and necessarily must integrate actual local material, should orient itself primarily to production workers who are closely tied to their work."21
Agitprop advocates believed that their homemade performances, filled with details from factory life, gave their work an immediacy and relevance that material by professional authors could not possibly achieve. Moreover, they asserted that their viewers responded with enthusiasm to the portrayal of events from their daily experience. To prepare, brigade members spent time collecting local material from factory workers, union and party members on site, and even from managerial reports. One Leningrad brigade gathered information from local newspapers, factory wall newspapers, and the factory union leadership.22 Group preparation was prized, and a few brigades even organized communal living arrangements to increase their cooperation.23 It was not uncommon for the final script to be the work of the brigade leader or factory literary circle, with input from participants along the way. However, some circles found more innovative methods, dividing up the task of writing as if on an assembly line. One subsection would collect material, another would shape it into a rough outline, while still another was responsible for the final product. "A shock work tempo demands shock work methods," proclaimed one brigade leader.24
At a time when speed was valued over all else, brigades prided themselves on being able to convert local material into a performance with dazzling dispatch. In one competition among five brigades from the Moscow woodworkers' union, participants started gathering material at six in the evening; by nine they began to write, and by three in the morning they had a finished work!25 Some brigades had a rigorous performance schedule that even surpassed the fervor of club circles during early NEP. The Moscow Rusakov brigade, for example, only took shape in mid-September 1930. By the beginning of October, it had already performed sixteen times.26
If brigades did not have the skill or confidence to shape their own work from scratch, they could turn to outlines and examples published in the trade union and Komsomol press, which they were encouraged to alter for their own purposes. The journal Club Stage presented what it called "repertoire material," intended as an outline or "carcass" for local work, along with bibliographies explaining where interested readers might turn to find the material used in the scenario. One literary montage called "Face to the Industrial Plan," was composed by Aleksei Arbuzov, who would later become a famous playwright. He drew on speeches by Lenin, Stalin, and prominent economic leaders as well as bits of plays and poems by Vladimir Maiakovskii, Alexander Bezymenskii, Sergei Tretiakov, and Demian Bednyi.27 Some of these outlines suggested places where local material could be inserted, leaving relatively little to the imagination.
Ideally, brigades also engaged in "extra-stage" (vnestsennye) tasks. They organized political meetings, sponsored demonstrations, and helped to write factory wall newspapers, designed to keep workers apprised of local efforts to fulfill the industrial plan. To stamp out the problem of drunkenness, one brigade held a political rally, encouraging those in attendance to share their reasons for drinking. Moscow and Leningrad brigades took trips to the surrounding countryside and nearby construction sites, where they helped to collect fuel and organize new dormitories.28 One participant recounted that he could only keep up with his many responsibilities by drastically cutting back on sleep.29
The most important function of these impromptu groups was to convey the constantly changing demands economic and political organizations placed on the laboring population. According to one advocate, "Agitprop brigades closely link their activities to factory committees and party organizations, quickly reacting to problems in production and trade union work. They discuss all current political campaigns and serve as a megaphone [rupor] for the local proletarian public."30 Their production methods allowed them to respond almost instantly to new regime initiatives, transforming the official language of communiques into vivid examples of how new policies could potentially aid workers who cooperated and hurt those who did not. Numerous state agencies appreciated their services and inundated them with assignments to explain factory economy measures, help with the collectivization drive, aid in national defense, and increase worker responsibility for factory production. A 1930 textile union directive, for example, instructed cultural groups that they needed to provide explanations for the raw material shortage that was plaguing the industry.31 Thus, the spiraling demands of the industrialization drive fueled the frantic work schedule of the brigades—their tasks, it seemed, were never done.
#### The War on the Audience
Agitprop brigade members claimed close ties to their audiences. Since they collected material locally, viewers could recognize their own victories and defeats acted out on stage. This kind of verisimilitude made some viewers see the players as "svoi," as their own.32 When brigades praised local heroes and chastised unpopular figures like managers and corrupt trade union leaders, they got an enthusiastic response. In addition, many brigades addressed true-to-life local problems—dangerous work sites, unrepaired housing complexes, and surly salespeople in stores. Sometimes their performances helped to alleviate these very real woes.
Absent managers and dirty dormitories were not the only targets of agitprop performances, however. Audience members themselves were not exempt from censure. Brigades did not shy away from ridicule, shame, and even political threats to achieve their goals. This was utilitarian art of the most extreme kind, with success measured in improved industrial output and political participation. Aiming for their own kind of production statistics, agitprop brigades became much more aggressive and invasive than earlier amateur circles had ever been. Given the many coercive methods they employed to change the actions of their viewers, it is no exaggeration to say that they declared war on the audience.
Because they did not need stages, lights, or elaborate props, agitprop brigades could perform almost anywhere that workers congregated. Rather than trying to lure viewers to clubs after the work day ended, they brought their performances to factory workshops, cafeterias, barracks, apartment courtyards, and even to underpasses linking factories to tram and train lines. Scripts were usually short, intended to fit into a lunch break or work break on the job. While such methods certainly increased the number of viewers, it also meant that many people were exposed to these performances against their will. Using aggressive methods inspired by shock workers in production, the brigades aimed to root out old habits and humiliate those who practiced them.
Many agitprop performances took the form of an urban charivari, where poor laborers were publicly humiliated. According to one report by Nikolai L'vov, the audience for a performance by the Rusakov agitprop brigade in Moscow only became attentive when the participants began naming names: "When the brigade began to castigate real perpetrators of the evils of laziness and shirking, etc., then the mass of viewers became lively and the guilty members of the audience began to feel un-comfortable."33 One brigade posted pictures of workers who were chronically late, soliciting the help of the factory art circle. It began the performance by carrying in a coffin emblazoned with the name of the worst offender. Another group extended its censure beyond the performance site itself, putting up signs in front of the apartments or barracks of repeat offenders and even sending notes home to their wives.34
Public shaming rituals were an important part of agitprop performances. One common tactic was to prepare lists of exemplary and poor workers, known as the "red list" (krasnaia doska) and the "black list" (chernaia doska). Before performances, local leaders would provide the brigade with the names of individuals who were to be censured for infractions such as drunkenness, lateness, and disruption of labor discipline, who would find their names on the black list.35 According to one Western observer, these public shamings had an influence on behavior: "Every factory in Russia today has its red and black boards [the literal translation of doska].... Wives and children, friends and fellow workers, see there who has disgraced himself as a slacker. Children lecture their inefficient worker-fathers."36 One Moscow troupe gave a factory's worst idlers a badge of shame to wear—a flag made of woven straw (or bast, a material typical of the peasantry) emblazoned with an empty bottle.37 As might be expected, not all audiences were appreciative of these demeaning tactics. When one brigade performed a skit satirizing the poor quality of a factory cafeteria's offerings, some workers protested that it was none of the brigade's business.38
Agitprop brigades honed their aggressive style by performing in the countryside for audiences they assumed would be hostile. One of the first organized agitprop campaigns was an effort to help the 1930 spring sowing campaign. More than sixty groups from Moscow alone answered a call by trade union organizations to aid in this effort. They disseminated information about the collectivization program, organized (or in some cases reorganized) local drama groups, and took part in anti-religious agitation.39 In the process, they sometimes met with overt resistance; unhappy viewers created disturbances and even threw rocks and bottles to disrupt performances.40
Judging from the arrogant tone of the reports brigades sent back to the city, the agitators maintained that they could discover and isolate problems within a village after a few hours of hasty research and then eradicate them in the course of a performance. One brigade arrived at a newly established collective farm plagued with difficulties. The problem, according to the group leader, was the resistance of local kulaks, better-off peasants who were blocking collectivization. Brigade members quickly isolated these offenders: "With our chastushki and prepared material [karkasnyi material] we revealed to kolkhoz members those responsible for hindering their work."41 In another village, inhabitants had been avoiding a general meeting on the issue of dekulakization. Collaborating with local authorities, a visiting agitprop brigade from Moscow gathered the population together for what viewers believed would be an evening of song and entertainment. The brigade then interrupted their performance and forced a vote on the touchy issue. "As a result," noted a proud reporter, "the village voted unanimously to rid itself of kulaks."42
These coercive techniques were perfected on urban audiences when agitprop brigades used their performances to extract funds for the industrialization drive. Since the early 1920s, the Soviet government had raised revenue through bond programs. With the turn to rapid industrialization, the regime implemented mass subscription bonds with deductions taken directly from paychecks. An expansive effort to increase participation in the bond program was launched in the spring of 1931, and agitprop brigades embraced it with enthusiasm. Not only did brigadiers themselves contribute generously, but they attempted to make audience members pledge at least the equivalent of one month's salary to the program.43 Since the First Five-Year Plan was a period of sinking salaries and worsening living conditions for workers, getting viewers to take even more money out of their pockets was a difficult task.44
Agitprop brigades used considerable ingenuity in their efforts, using fear, praise, shame, and even charges of sabotage to get viewers to give part of their meager salaries. Scattered published scenarios give some idea of how brigades worked their audiences. The agitprop troupe of the Postroika Theater Collective from Moscow composed a montage of political speeches, newspaper reports, and local statistics to drum up support for the campaign. Their performance opened with the exhortation: "The Party of Lenin has a pledge every worker should make: Catch up and overtake." The players challenged cynics who did not believe that the Soviet Union could ever overtake America by presenting current production statistics from newspapers and political speeches. How will the nation move even further forward? performers then asked. "Who will give the money? The capitalists? No, they will not give it and we will not take it.... We, we ourselves will give the money for construction." After explaining how much money was expected from each audience member, participants distributed pledge forms urging viewers to sign up. They ended their performance with statistics on local compliance.45 Other groups used even more forceful methods to secure support. One brigade employed a five-step scale, from "the order of the airplane" (the best) to "the order of the two-humped camel" (the worst), determined according to the speed and enthusiasm with which workers had responded to the bond campaign. During performances they would assign each factory workshop its fitting appellation.46 Another brigade not only distributed pledge forms but also revealed the names of those who did not fill them out. Recalcitrant workers who did not bend to this pressure found their names printed on a black list. If these methods were unsuccessful, some brigades escalated their tactics still further, accusing shirkers of sabotage.47 Clearly, all thought of entertainment had vanished from agitprop agendas.
The agitprop brigade of the Rusakov Club in Moscow. The poster says: "Loan—Five-Year Plan in Four Years." Bakhrushin State Central Theatrical Museum. Reproduced with permission.
#### Agitprop Aesthetics
The main goals of agitprop brigades were to inspire action, not to produce works of lasting artistic value. Nonetheless, participants did articulate principles that we could call an incipient aesthetic, ideas for what would make performances have the maximum impact on the audience. These principles were influential in many other discussions, including those affecting the design of club structures. They even had an influence on professional theaters. Moreover, they were echoed in many other branches of art during the First Five-Year Plan.
Agitprop performances were primarily a method of criticism, based on the unspoken assumption that censure was a more effective means of motivation than praise. The repertoire did have its share of exemplary railway switchmen who saved valuable shipments and plucky young workers who found better ways to install foreign technology—what Katerina Clark has called the "little heroes and big deeds" of the First Five-Year Plan.48 But the emotional weight fell more heavily on villains than on such paragons of socialist virtue. The most ubiquitous evil figures were those in positions of authority on the shop floor; cynical factory managers who did not show any interest in their workers or their jobs, drunken and corrupt union leaders, and of course, highly trained experts like engineers, whose sympathies for the Five-Year Plan were under constant scrutiny.49 These little villains with their bad deeds were the stock and trade of agitprop productions.
Agitprop scripts ignored the sphere of private life almost entirely. This set them apart from TRAM plays, which stressed the interaction between work life and the home and examined new rules of romance among Soviet youth. All that mattered for agitprop brigades was accomplishment at the work site, something that can be seen already in the prosaic titles of agitprop works: On Cost Accounting, Down with the Wreckers, Face to Production.50 Brigades were not alone in their hostility to the home. The youth newspaper Komsomol'skaia pravda launched a noisy campaign to rid the home of all remnants of cozy "domestic trash." Avant-gardists designed austere, portable furniture for a nation on the move, and the artist El Lissitzky claimed that all any person really needed in life was a mattress, a folding chair, a table, and a gramophone.51
Since home and family were coded as female spheres in the Soviet imagination, agitprop performance works depicted worlds almost entirely without women. This probably also reflected the overwhelmingly male composition of the brigades. Many published agitprop scripts had no female parts at all or else portrayed women as standing entirely outside the production process. The railroad drama Counter-Plan (Vstrechnyi), for example, featured only a cameo appearance by an aggrieved mother in a world otherwise populated entirely by men.52 This was by no means a "realistic" depiction of Soviet production sites, because women entered the labor force in record numbers during this period.53
Participants in agitprop brigades made a virtue of the speed with which they were able to conduct their work. Many enthusiastic young people believed that conventional theater's inability to keep up with current events was one of its greatest weaknesses. "Our life goes too fast," remarked one young actress to the American communist Ella Winter. "By the time a drama reaches the stage its theme is already dead."54 Here agitprop brigades had a clear advantage: they often performed work that had just been written from material that had only recently been collected at the factory. In the opinion of one agitprop advocate, this meant that for the first time art would not suffer from "tailism," from reflecting on events after they had already happened. Instead, it could shape events.55 Of course, the need to keep up with current events guaranteed that work was quickly outdated. Brigades could not perform the same composition over a long period, perfecting their methods, because they constantly needed to change the subject matter and consider a different audience.
This acute attention to current events turned agitprop work into what Susan Suleiman has called a "perishable genre." Her reflections about highly ideological novels that attempt to prove the validity of a certain doctrine apply equally well to agitprop scenarios: "Written in and for a specific historical and social circumstance, the roman à thèse is not easily exported. And even in its native land, it becomes 'ancient history' as soon as the circumstance that founded it no longer holds."56 Agitprop performances were of vital interest only to the actors and audience for whom they were composed. Brigades made a cult of local specificity and detail as a central element of their performance. Audiences also demanded to see themselves reflected on the stage. One group of railroad workers sent an open letter to the journal Club Stage asking for works about their industry: "Playwrights! Give us a play about the life and times of railroad workers! Show their heroic struggle to fulfill the transport plan in three years with twin mountings and labor discipline!"57
The perishable genre of agitprop theater exerted a powerful influence over the practices of Communist amateur and semi-professional theater groups in Western Europe and the United States. During the Depression, left-wing theater groups applied these techniques to local situations, performing in impoverished working-class neighborhoods, at striking factories, and in city parks. Their homemade scripts, with titles like Fight against Starvation and Lenin Calls, cursed capitalists and praised worker solidarity. "Groups in the vicinity of strikes should seek to perform appropriate plays,' admonished one American journal. "Groups should participate in the raising of funds for strike relief."58 Like their Soviet counterparts, these mobile circles tried to inspire their audiences to action. However, they did not have the power of the state behind them.
No matter where they took place, agitprop performances were animated by their attention to concrete detail. In the Soviet Union, this put agitprop brigades in the same camp with avant-gardists gathered around the journal Novyi lef, who advocated what they called a "literature of fact." Writers like Sergei Tretiakov, Boris Arvatov, and Nikolai Chuzhak wanted authors to abandon traditional large forms like the novel and base their work on short sketches, diaries, travel notes, and reports. Such works, without central heroes, would be a kind of literary montage, where the reader would be required to make meaning from the work. Not only would such a method make art relevant to the contemporary period, ideally it also opened up the process of creation to a very broad public.59
But during the First Five-Year Plan, "facts" were hard to come by; they were replaced by idealistic plans, projections, propagandistic statements, and even dreams. The sophisticated theorists of "literature of fact" were well aware of the tensions between straightforward documentation and inspiring propaganda, or what Osip Brik called "protocol" and "proclamation."60 In the hands of inexperienced agitprop brigade authors, this meant that despite their works' grounding in a specific time and place, printed scripts have a remarkable similarity to them, with all industries apparently blessed with standard-issue shock workers and cursed with the same incompetent managers and loutish drunks.
Defenders of agitprop brigades were at pains to show that earlier criticisms of small forms did not apply to their efforts. Brigades did not suffer from "Blue Blouseism," a blanket term of abuse that had crystalized by the late 1920s. Those afflicted with this aesthetic disease were overly influenced by popular entertainment styles and depended on formal innovations. In addition, Blue Blouse scripts had been written by professional authors far removed from the production site. Whereas Blue Blouse groups let the form determine the content, insisted one advocate, in agitprop brigades the content determined the form.61 It was precisely the privileging of function over form that was the most distinctive element of agitprop brigades, determined Grigorii Avlov, one of the earliest advocates of the united artistic circle in Leningrad.62 In Avlov's opinion, agitprop brigades were much more than a new performance style; in fact, they were a new social and organizational form.63 Their goal was not a conventional performance but rather propagandistic action to solve a social-political problem. Through their action, they not only affected the lives of their viewers, they also molded the participants into conscious fighters for socialism.64
These mobile troupes challenged conventional understandings of performance space. Their ability to work anywhere helped to refine a persistent criticism of club architecture. Just as the club building boom, started in late NEP, had begun to make large stages and auditoriums available to amateur theaters, advocates of agitational art criticized their expansion. Such structures were expensive and only encouraged passive viewership at a time when amateur circles were trying to activate audiences, they argued.65 Big performance spaces meant that clubs felt required to fill them and often resorted to inviting professional theaters or traveling troupes.66 One critical reviewer of Konstantin Mel'nikov's structure for the Kauchuk Chemical Workers' Club insisted that the new auditorium was so large that it "ate up" the rest of the club.67 Such shiny new buildings did not result in new or innovative work, chided the youth newspaper Komsomol 'skaia pravda. Instead, they offered poor films and low-quality entertainment, often for a fee.68
To add to the problems of conventional clubs, the introduction of the continuous work-week meant that they could no longer balance their budgets by staging popular big-ticket events on weekends.69 This unpopular attempt to maximize factory use was introduced in the fall of 1929. The labor force for each factory was divided into five groups, each of which worked for four days, with a fifth day off in staggered sched-ules.70 This upset family leisure time, since members might have different free days. It also complicated preparation time for artistic groups and caused scheduling nightmares for club directors. Critics of large halls used it as further evidence that big performance spaces were no longer needed.
To solve these many difficulties, a group of radical architects articulated ideas on a new kind of club just as agitprop brigades were taking shape in 1929–30. They argued that clubs should not contain performance spaces at all. The vast new auditoriums in many new clubs discouraged other forms of creative work. Instead, clubs should be reconceptualized as centers for cultural activists, without spaces for passive spectatorship. Some Komsomol activists even insisted that clubs should be eliminated altogether and replaced by "social-political soviets."71 Although it is hard to know if these ideas were inspired by brigades, they certainly fit the agitprop aesthetic. Brigades did not need stages; they brought the performances to the audience. With their combative, interventionist style, they eliminated the problem of passive viewers. Instead, involvement was obligatory. If viewers did not take action on their own, their participation would be forced.
Agitprop brigades embodied radically egalitarian, anti-intellectual principles. Many refused the oversight of specialists, instead electing one of their own members to take a leadership position.72 For Sergei Alekseev, a club leader in Moscow, this showed that agitprop brigades were really do-it-yourself collectives, playing off the literal meaning of samodeiatel'nost'. Older drama circles were subject to the author of the play, to the artistic director, and to the heads of artistic organizations that oversaw their work. Agitprop brigades removed these troublesome intermediaries, making their own scripts and their own formal decisions.73 By writing their own work, concurred Avlov, brigades were able to define their own issues. "An agitprop brigade should not wait until someone writes a play," he insisted." [It] should find its own question, investigate it, pose it, work up the material, and write a text."74 This directive put a lot of responsibility on young brigade members, who often were not used to writing. One brigade participant at the Putilov factory in Leningrad bemoaned, "It is a torturous process for young people to take up the pen for the first time and write their own plays. There was a lot of insecurity, dissatisfaction, and disillusionment."75
Such egalitarian strains were widespread during the First Five-Year Plan, evident in the shock work movement itself and in all artistic media. Amateur visual artists also formed brigades and moved into the factory, creating site-specific posters and satirical drawings of those falling behind in the campaign for higher work quotas.76 Amateur photographers captured images of their co-workers on the job.77 Scores of new writers and worker journalists were recruited by newspapers and literary organizations to document the rapid changes in industry and agri-culture.78 In the view of some observers, the rapid growth of amateur artistic movements was evidence of a seismic historical shift, where the divisions between mental and physical labor would disappear. Echoing Marx's contention that socialism would end the separation of the arts from the rest of life, one writer claimed he was witnessing such a process in the work of amateur circles.79
Agitprop methods even influenced professional stages. During the 1930 spring sowing period, when agitprop circles took to the countryside, brigades from professional theaters joined them.80 They performed in urban areas as well, sometimes offering their standard repertoire when it had some relevance to the venue. A brigade from the Meyerhold Theater, for example, performed Bezymenskii's The Shot at a special performance at the Moscow factory Elektrozavod, which had close ties to the theater. Professional groups heading to the countryside often chose works that were originally conceived for amateur circles, such as the agitational play Red Sowing (Krasnyi sev), written hurriedly for the Moscow House of Amateur Art.81 Since professionals were now relying on work designed for amateurs, some observers concluded that amateur theaters had now gained the upper hand; professionals should come to learn from them.
Advocates of agitprop brigades adopted an arrogant pose toward the drama circles from which many of them had emerged. Brigades presented themselves as a "vanguard" whose task it was to transform the work of conventional circles. However, some amateur actors were not so easily convinced. At the Rusakov Club in Moscow, several members left when the drama circle was transformed into an agitprop brigade. The Putilov factory drama circle lost more than half of its members when it turned to agitprop performance.82 The Krasnyi Bogatyr factory, located in Moscow, had begun an agitprop brigade already in the fall of 1929. Initially, it included about fifteen people who broke off from the drama circle. They then attracted other interested members from the literature and music circles, all with shock worker status. Although the brigade won distinctions for its work in the factory, it did not replace the drama circle entirely. One reporter noted that "the old drama circle members, having 'played' for years, mustered considerable oppositional strength."83
The language agitprop circles used to distinguish themselves from their "apolitical" rivals was filled with threats and warnings, reflecting the turn toward purging then taking place in society at large. Groups that refused to change their repertoire should be ostracized, with their names placed on a black list, insisted one cultural worker. Another believed that such groups should be disbanded for harboring Trotskyist sympathies.84 Critics frequently blamed theater circle directors for inhibiting change, charging that they were drawn from "opportunistic elements" and even included class enemies. Agitprop supporters called for "rigorous internal discipline," "strenuous oversight of repertoire," and a "purge of alien elements."85 "We know of cases," wrote the editors of Club Stage, "where the class enemy has not only worked his way into our mines, factories, and other institutions (think of the Shakhty affair, the Industrial Party Trial, the Menshevik Trial, etc.). In addition, he has wormed his way into choirs, orchestras, and drama circles in workers' clubs."86
#### The Assault on the Small
The hegemony of agitprop brigades did not go without protest even in their heyday. Cultural bureaucrats from Narkompros, like Robert Pel'she, charged them with an appalling lack of skill. Representatives from RAPP, who admired the Moscow Art Theater, complained that they had fallen under the influence of Litfront and Blue Blouse.87 As the First Five-Year Plan drew to a close, critics also called the political reliability of agitprop brigades into question. Although they eagerly embraced the task of presenting official positions in a palatable form, these groups were the ones in control of the process of composition. Their ability to prepare material in a hurry was the very reason for their existence. This meant that brigades often performed before their work could be previewed or censored by political authorities. According to one local observer, "Political control over the content by factory committees is practically nonexistent. The circles are left to themselves."88 Since the brigades were more effective in presenting villains and bottlenecks than heroes and production triumphs, their work could easily be seen as an open assault on Soviet production methods. Commentators increasingly called for more oversight over the content of performances.
Perhaps the most important mark against agitprop brigades was their aggressive stance against hierarchies based on skill, whether in the factory or the artistic community. Expert at bashing experts, many brigades had a difficult time adjusting to the new line on specialists and the turn against egalitarianism launched by Stalin in the summer of 1931.89 At this point, the government officially rejected the anti-authoritarian spirit fostered by the First Five-Year Plan, rehabilitating once-maligned experts and establishing sizeable wage differentials based on skill. Agitprop troupes' negative portrayal of managers and factory directors began to concern critics. They also worried that these impromptu performances presented no real heroes that audiences could use as role models.90
By 1932, agitprop brigades had ample evidence that the cultural winds were shifting. In April the Communist Party issued its momentous decree "On the Restructuring of Literary and Artistic Organizations." This proclamation marked an important turning point in the direction of Soviet culture. With it, the Communist Party ended the aggressive dominance of self-proclaimed proletarian cultural groups that had gained considerable visibility during the First Five-Year Plan. According to the resolution, these organizations had become too narrow and sectarian, hindering the further development of Soviet culture. They would be replaced by national artistic unions open to all classes.91 This decision meant the end for well-established groups like RAPP, the Proletkult, and the Proletarian Musicians' Union.
The Communist Party's pronouncement on the arts might have given local cultural circles like agitprop brigades some cause for concern. After all, they usually claimed to be class-exclusive groups catering primarily to workers or working-class youth. They also could be charged with narrow and sectarian pursuits. Initially, however, no one interpreted the directive in this light. Cultural journals urged participants to consider the significance of the April pronouncement in their work, and some local clubs and factories held discussions about its possible implications, but there was no consensus about what it might mean for amateur artistic activity.92
For agitprop brigades, a more significant turning point marking the end of their predominance was a supposedly festive event meant to celebrate the achievements of amateur art. At the beginning of 1932, the central trade union leadership joined together with the Communist Party, Narkompros, and the Komsomol to begin plans for a national Olympiad (a term used for competitions) of amateur musical and theatrical groups recruited from all over the country. Organizers had numerous models to follow, since local unions and city cultural divisions had been hosting competitions of this kind for many years and an Olympiad of theaters from the Soviet national republics had already been staged in 1930.93 Since agitprop brigades had come to dominate amateur theatrical work, they were the most widely represented theater circles at the competition.
By February 1932 the organizing committee began publicizing the upcoming event in the national press and specialized cultural journals. Competitions were to be held at the local level, with the best groups performing in Moscow that summer. The winners from those competitions were promised a prominent spot at the festivities surrounding the fifteenth anniversary of the revolution in November. The committee also determined appropriate topics for the theatrical presentations, choosing subjects that mirrored the kinds of themes already common in agitprop brigades, including military preparedness, the struggle for collectivization, and the successful completion of the industrial plan.94
In August 1932, one hundred and two amateur theater groups, choirs, and orchestras from all over the country converged on Moscow's Park for Culture and Leisure. Opening ceremonies followed the pattern of other First Five-Year Plan festivals, with their displays of military might and technical achievement.95 Dirigibles and spotlights announced the beginning of festivities. A group of soldiers from the local garrison made its appearance, followed by a parade of the participants, all in different colored costumes to identify their geographic regions. A huge orchestra and mass choir inaugurated the cultural competition. Estimates of the crowd at the opening ranged from 60,000 to 100,000, with many viewers coming from local Moscow factories.96
The week-long festival was elaborately scripted and included visits to local museums and theaters, speeches by important cultural experts, and meetings with Moscow shock workers. Proponents of amateur theater had extremely high expectations for this event. A lead editorial in Worker and Theater predicted that the Olympiad would fulfill an ambitious artistic and ideological agenda. It would contradict Leon Trotsky, who believed that there could be no proletarian socialist culture and no socialism in one country. The Olympiad would also disprove the views of Mikhail Tomskii, the recently ousted trade union leader, because it would show that workers' cultural circles could accomplish more than rudimentary cultural training. Finally, it would expose the false position of the avant-garde, who argued for a total break with the culture of the past. "Under the leadership of the party and unions, amateur art will strike at rightist 'advisers' and 'leftist phrases.' "97
The politicization of theatrical groups in the course of the First Five-Year Plan was clearly apparent at the festival. Well over half of the theater circles represented were agitprop brigades. TRAM groups were the next most numerous. Only one group called itself a drama circle.98 The festival's theme was "For a Magnitostroi in Art," signaling that cultural construction should be as successful as industrial construction in the new city of Magnitogorsk.99 Factory motifs and the struggle for higher production quotas dominated festival repertoires. The transformation of proletarian consciousness through the struggle for production was another popular subject.
Participants in the Olympiad wielded impressive political and social credentials. According to several estimates, more than fifty percent of them were members of the Communist Party or Komsomol, and seventy percent called themselves shock workers, a term used with increasing frequency by 1932. Youth was another common characteristic, which is hardly surprising given the age composition of amateur art circles. A full sixty-five percent of those taking part were under the age of twenty-three, with just ten percent over thirty-five.100 Only twenty-nine percent of the performers were women, a figure well below standard estimates for female participation in union club theaters. No doubt the high percentage of, shock workers, who were overwhelmingly male, help to explain women's low participation rates.101 Despite efforts to reflect the geographic and cultural diversity of the Soviet Union, Moscow and Leningrad were overrepresented at the festival, sending eleven out of sixty-one theater groups.
The Olympiad received extensive coverage in the national press, in journals devoted to art and culture, and in local newspapers in Moscow and Leningrad. Some newspapers offered lengthy daily accounts of the festivities with elaborate descriptions of the opening ceremonies, many individual performances, and the final results of the week-long competition. The official jury included representatives from professional theaters, trade union organizations, TRAM, and Narkompros. Some familiar names showed up among the judges, including the long-time activists in amateur theater, Valentin Tikhonovich and Nikolai L'vov. One of the Communist Party's chief experts on cultural affairs, Iakov Boiarskii, also participated in the jury discussions.102
Most printed assessments on the event began with general words of praise. The festival had shown that the masses now had access to a world of culture that had once been denied them. "Those who before the revolution could not even dream of art, whose lot in life was only prison-like labor, now sang and acted," determined an author in Pravda. "The gods have been pulled down from Olympus. 'Holy Art' has become accessible to the broad masses of the country and has been put to work for socialism."103 However, these vague and formulaic words of praise were overshadowed by a barrage of complaints leveled at amateur theatrical groups in general and agitprop brigades in particular. While reporters did not always agree in their choice of good and bad performances and jury discussions were sometimes acrimonious, the overall critical consensus was remarkably similar: amateur theater needed fundamentally different aesthetic and organizational principles.104 Moscow and Leningrad circles were in no way exempt from this critical onslaught.
Critics attacked groups for their remarkably low level of writing. The Izhorsk factory collective from the Leningrad suburb of Kolpino was a special target. It presented a homemade play called Coal (Ugol'), which showed how older workers learned the value of modern machinery while the factory management learned to appreciate the enthusiasm of the work force. Jury members charged that it offered far too much local detail. One observer was even harsher, condemning its ungrammatical speech and wooden dialogue.105 Another work from Leningrad, The Victors Will Judge, written by original TRAM member Arkadii Gorbenko, received low marks for poor organization, a predictable plot, and unconvincing characters. The entire play did not have one living image, protested a critic in the trade union paper Trud. "From the very first act it was easy to predict the end."106
Critics discerned dangerous aesthetic trends and the influence of discredited approaches in many performances. Numerous Leningrad circles were accused of suffering from "Piotrovshchina," of not yet having overcome the evil influence of the Leningrad TRAM.107 "Blue Blouseism" was another common failing. This error was revealed in the Putilov factory work How Tom Gained Knowledge, where all the characters spoke in slogans and used ungrammatical language. For the Moscow central TRAM, performing The 10:10 Moscow Train, Blue Blouseism was evidenced in its schematic construction and murky central narrative.108
The hallmark of the agitprop aesthetic, the use of local facts, was quite evident in the works presented at the Olympiad. Critics determined that such details might interest viewers from a specific factory, but they could not engage a broader audience. "Not every fact taken from real life rings true in art," determined jury members in response to a play tracing the building of a ship at the Northern Wharf in Leningrad. Their work, RT 57, had emerged from documentation collected in the "History of Factories and Plants" project. In the jury's view, its "photographic method" had resulted in a "naive naturalism."109
The judges were also dissatisfied with the heroes presented in works at the Olympiad. Although the characters presented in the longer works by Moscow and Leningrad circles were superior to those in short sketches, they remained unconvincing. Their personal failings were poorly explained and their heroic accomplishments remained sketchy.110 Moreover, the single-minded attention to production themes to the exclusion of other parts of life had become monotonous. Boiarskii, head of the artists' union Rabis, complained, "In the final analysis, agitprop brigades' focus on the struggle with negative elements in production (slackers, absentees, loafers, etc.); although this [topic] has great importance, it alone cannot reveal the activity and initiative of the working masses in construction. And does the life of the contemporary, cultured worker end at the construction site?... What about friendship, love?"111
The distribution of the prizes reinforced these critical judgments. Groups presenting short agitational works were slighted; instead, top honors went to those offering full-fledged plays. The recently professionalized Leningrad construction workers' theater, Stroika, received first prize for its performance of We Are from Olonets (My Olonetskie). In an interview, Stroika members stated that they had purposefully turned away from "small forms" and attempted to create a work that looked more like a professionally written play, even enlisting the help of professional playwrights in the process.112 Second place went to the Moscow central TRAM. Although agitprop brigades had dominated the festivities, the most they could claim was a third-place prize.113
By the time the Olympiad was over, a critical consensus about the future of amateur theaters had emerged. Participants were advised to go back home and turn to the experts in order to diversify their repertoires and improve the quality of their productions. They should attempt works by contemporary Soviet playwrights and also take on classical plays, including the work of Alexander Ostrovsky, Molière, and the eighteenth-century Italian playwright Carlo Goldoni. When addressing political themes, amateur theaters had to learn to do this "artistically," which was only possible with the assistance of those trained in technique and familiar with the long history of Russian and world theater.114 All of these strictures undercut the methods and messages of agitprop brigades.
For the national trade union organization, the indirect sponsor of most of the performing groups, the Olympiad marked a significant watershed in the history of the amateur arts. In a slim volume commemorating the festival, union leaders announced a major restructuring of amateur theatrical work: "The artistic demands of workers in the leading centers of production, demands once encouraged by agitprop brigades, can no longer be satisfied by the current level of agitprop brigade theater."115 A lengthy list of resolutions outlined a general retreat from the militancy and intolerance of the First Five-Year Plan era. Agitprop brigades and drama circles should coexist, employing both large and small forms. All amateur circles needed to develop tolerance toward different artistic approaches; they should also strive to articulate a unique "creative personality" (tvorcheskoe litso). Older workers and non-Party members had to feel welcome in amateur circles, ending the social exclusivity of agitprop brigades. Perhaps the most striking recommendation was a new attitude toward the audience. The coercive tactics perfected by brigadiers were no longer appropriate. Union leaders suggested that amateur circles "fight for the creation of cheerful, joyful performances that can bring viewers pleasure and constructive leisure."116 The war on the audience was over.
In a long assessment of the results of the Olympiad, one of the judges, the club leader Sergei Alekseev, offered a very positive interpretation of its lessons. He saw the event as the final struggle in a long "battle of genres" between amateur drama circles performing longer works and the agitational theater of small forms. During late NEP, drama circles had dominated. Agitprop brigades achieved the upper hand in the years of the First Five-Year Plan, however, driving out any other kind of work. Alekseev—who had a few years before decried the use of professional plays on club stages—now believed that the time for a true synthesis had come: "[The Olympiad] will lead to the free growth of diverse forms of amateur work. [Groups] will take from the rich experience of professional theaters and discover an independent, young, amateur theater of the working class." While drama circles had ignored their social context, agitprop brigades had ignored training. Now both sides would learn from the other and bring forth something new.117
This call for synthesis and cooperation was common at the time, as cultural bureaucrats and artistic practitioners struggled to define the emerging doctrine of socialist realism, only just articulated in 1932. Many hoped that it would bring an end to the politicized, nasty cultural battles of the First Five-Year Plan. The practices of the avant-garde would merge with the realists. Professional and amateur theaters would enter into mutually enriching cooperative arrangements. Soviet artists would finally discover how to create works that were both politically correct and entertaining, ending the dysfunctional tendency toward one or the other extreme.118 The amateur arts would be open to a broad range of approaches and training methods, announced a trade union official at a cultural conference in late 1932.119
But there were also voices that questioned such a socialist happy ending. In its assessment of the Olympiad, the newspaper Sovetskoe iskusstvo asserted that the time had come for "big forms" to address "big problems." While the editorial called for a multiplicity of artistic methods, it also asserted that small forms were unreliable. They could not do justice to the grandiose achievements of the Soviet Union. To make matters worse, the agitational theater of small forms had been tainted by its association with utilitarianism and Litfrontism.120 By including these explicit warnings, the unnamed author indicated that the new system taking shape might not be so tolerant and inclusive after all.
In a now-famous article first published in 1978, Sheila Fitzpatrick delineated a unique period of "cultural revolution" within the First Five-Year Plan. Beginning with a highly publicized trial against engineers in 1928 known as the Shakhty Trial, this distinctive phase was marked by youthful dominance, attacks on established hierarchies, and a militant language of class war. While "unleashed" from above, these assaults were enthusiastically embraced by the Komsomol, student organizations, self-proclaimed proletarian cultural groups, and young workers eager to shake up factory management. It was not until Stalin's direct intervention in the summer of 1931, with a speech denouncing egalitarianism and defending expertise, that this anti-authoritarian phase in Soviet cultural life came to an end.121
Agitprop brigades embodied many of the values Fitzpatrick described, with their denunciations of experts and rejection of bourgeois culture, broadened to include all elements of aesthetic theater right down to the stage boards. Thus, it might seem strange that I have consciously chosen not to use the term "cultural revolution" in my analysis of these creations of the First Five-Year Plan. One reason is simple chronology; agitprop brigades do not fit neatly into Fitzpatrick's periodization. The Shakhty Trial made little impact on the repertoire of amateur stages. The big celebrations and competitions in the fall of 1928 in Moscow and Leningrad were almost no different from those staged a year before.122 For amateur theaters, the change in trade union leadership in the summer of 1929, combined with the spread of shock work, marked the real rift from the cultural practices of late NEP. Similarly, while Stalin's 1931 panegyric to experts and expertise found an echo in the speeches of cultural organizers, it did not markedly change the radicalized repertoire of amateur stages. Not even the April 1932 decree dissolving independent proletarian cultural organizations like RAPP and the Proletkult affected the greatest swing back to more conventional forms. Rather, the crucial turning point was the national Olympiad of amateur arts in the summer of 1932. There small forms were viewed, assessed, and ultimately discredited in the most widely distributed media in the nation.
Moreover, Fitzpatrick's concept of cultural revolution highlights assaults on established authority as the cardinal marker of the era. Drawing on the research of Michael David-Fox, I see the First Five-Year Plan era, which he calls "The Great Break," as a complex of contradictory impulses that are not so easily separated. Challenges to hierarchy were combined with the great expansion of party and state power; attacks on "bourgeois" culture went hand in hand with efforts to instill the "bourgeois" values of sobriety and punctuality.123 Agitprop brigades acted out these contradictions. They ridiculed shop managers but expected rigid compliance to new state initiatives. They disdained stages as remnants of the old world but mocked those who were unable to show up to work on time. In their attempts to instill good labor habits, agitprop brigades had similar goals to the amateur theater of small forms during NEP. What was fundamentally different about these troupes was their methods. Rather than using humor, references to mass culture, and persuasion, they chose preachy didacticism. Rather than seeing their work as the beginning of a dialogue, like TRAM authors did, these actors relied on the tactics of shame and extortion. Seeing their audience as potential enemies, agitprop brigades widened the scope of accusation and distrust fostered by state and Party agencies during this period of upheaval.
Agitprop brigades were an attempt to fuse the contradictory categories of spontaneity and consciousness in Soviet art. Performers presented the messages of the regime, but they arranged those messages themselves and decided how they were to be transmitted. In the long run, this strange fusion pleased neither the audience nor the government. At the end of the First Five-Year Plan, these intrusive groups— hostile to the very concept of pleasure—either disappeared or changed beyond recognition. Turning against brigades as anti-aesthetic, sponsoring agencies curtailed their interventionist tactics. However, in the process of fixing their artistic failings, these agencies also guarded against a resurgence of spontaneity. The era of the homemade play was over.
* * *
1. "Za vypolnenie lozungov TsK partii!" KS 9 (1930): 3.
2. Tom Thomas, "World Congress of Workers' Theatre Groups," New Masses, November 1930, 21.
3. See ch. 1, pp. 19–20.
4. S. Alekseev, "Za khudozhestvennye proizvodstvennye agitpropbrigady," Malye formy klubnogo zrelishcha 16 (1930): 1.
5. "Boi na fronte samodeiatel'nogo teatra," ZI 46 (1929): 1.
6. On shock workers, see Hiroaki Kuromiya, Stalin's Industrial Revolution: Politics and Workers, 1928–1932 (Cambridge: Cambridge University Press, 1988), 115–28; David R. Shearer, "The Language and Politics of Soviet Rationalization," Cahiers du Monde russe et soviétique 32 (1991): 581–608; Louis H. Siegelbaum, Stakhanovism and the Politics of Productivity in the USSR, 1935–1941 (Cambridge: Cambridge University Press, 1988), 40–53; idem, Soviet State and Society between Revolutions (Cambridge: Cambridge University Press, 1992), 209–13; and Chris Ward, Russia's Cotton Workers and the New Economic Policy: Shop-Floor Culture and State Policy, 1921–1929 (Cambridge: Cambridge University Press, 1990), 244–52.
7. On the intricacies of these high-level changes, see Kuromiya, Stalin's Industrial Revolution, 40–46.
8. V. Kirov, "Za rabotu po-novomu," Klub i revoliutsiia 8 (1930): 8–9.
9. Shestnadtsatyi s''ezd Vsesoiuznoi Kommunisticheskoi partii (B): Stenograficheskii otchet (Moscow: OG12-Moskovskii rabochii, 1931), 662–63. See also "Zadachi profsoiuzov v rekonstruktivnyi period," Pravda 21 May 1930.
10. "Puti razvitiia agit-propbrigadnogo dvizheniia. Material k pervoi oblastnoi konferentsii khudozhestvennykh agitpropbrigad moskovskikh profsoiuzov," RGALI, f. 2723, op. 1, d. 419, l. 104.
11. "Ko vsem rabotnikam klubnykh khudozhestvennykh organizatsii," ZI 46 (1929): 12. See also Grigorii Avlov, Teatral'nye agitpropbrigady v klube (Leningrad: Gosudarstvennoe izdatel'stvo khudozhestvennoi literatury, 1931), 9–14. I found the first reference to brigades in Moscow in November, "Kul'tpokhod zrelishchnykh kruzhkov," VM 16 November 1929.
12. B. Filippov, "Khudozhestvennye agitpropbrigady—v deistvii," KS 5 (1931): 3–5, quotation 4.
13. Georgii Polianovskii, "Za khudozhestvennuiu propagandu promfinplana," Pravda 14 February 1930.
14. For an analysis of the shift away from multimedia techniques, see M. V. Iunisov, "Agitatsionno-khudozhestvennye brigady," in S. Iu. Rumiantsev, ed., Samodeiatel'noe khudozhestvennoe tvorchestvo v SSSR, v. 1 (Moscow: Gosudarstvennyi institut iskusstvoznaniia, 1995), 250.
15. N. Gruzkov, "Po Ves'egonskomu i Volokolamskomu raionam," Za agitpropbrigadu i TRAM 1 (1932): 39–40.
16. E. M. Karachunskaia, "Moskovskii smotr 1931g.," in Khudozhestvennye agitbrigady: Itogi smotra i puti razvitiia (Moscow: Narkompros RSFSR, 1931), 12–13. See also S. Alekseev, "Smotr khudozhestvennykh sil moskovskikh profsoiuzov," Malye formy klubnogo zrelishcha 17 (1930): 1–5.
17. A number of sources estimates the brigades to have had at least seventy percent shock workers. N. Strel'tsov, "Agitpropbrigadnoe dvizhenie v novoi obstanovke raboty profsoiuzov," Za agitpropbrigadu i TRAM 2 (1931): 5; "Samodeiatel'nyi teatr na olimpiade," KS 9 (1932): 1; "VI Plenum TsK Rabis," RiT 28 (1931): 7; V. N. Aizenshtadt, Sovetskii samodeiatel'nyi teatr: Osnavnye etapy razvitiia (Kharkov: Khar'kovskii gosudarstvennyi institut kul'tury, 1983), 39.
18. E. Permiak, "Khudozhestvennye brigady piatogo dnia," Klub i revoliutsiia 14 (1932): 41.
19. "Brosim sily na ugol', torf v Boriki," Za agitpropbrigadu i TRAM 1 (1931): 52; Delegat, "Nasha stsena—khlopkovye polia," KS 9 (1932): 41.
20. D. Korobov, "Kak sozdavalas' i stroilas' serpomolotovskaia agitbrigada," Za agitpropbrigadu i TRAM 1 (1931): 11–12.
21. Avlov, Teatral'nye agitpropbrigady, 32.
22. M. Reznik, "Opyt vskrytiia tvorcheskogo metoda agitpropbrigady LRROP," Za agitpropbrigadu i TRAM 1 (1931): 16.
23. B. P., "Agitbrigada 'Krasnyi Putilovets,'" RiT 19 (1931): 9; Karachunskaia, "Moskovskii smotr," in Khudozhestvennye agitbrigady, 15.
24. I. Mazin, "O khudozhestvennoi agitpropgruppe," KS 1/2 (1931): 24. See also B. Filippov, "Khudozhestvennye agitpropbrigady—v deistvii," KS 5 (1931): 8–9.
25. B. Shmelev, "Agitbrigady—khudozhestvennyi tsekh zavoda," KS 7/8 (1931): 17.
26. "Raport o rabote agit-brigady kluba im. Rusakova soiuza kommunal'nikov," GTSTM, f. 150, d. 37, l. 7.
27. A. Arbuzov, "Litsom k promfinplanu," KS 9 (1930): 40–49.
28. N. Gruzkov, "Po Ves'egonskomu i Volokolamskomu raionam," Za agitpropbrigadu i TRAM 1 (1932): 39–40; Gabaev i Lapshina, "Na novostroikakh," KS 11 (1931): 67.
29. Iunisov, "Agitatsionno-khudozhestvennye brigady," 250.
30. M. Veprinskii, "Zrelishchnye kruzhki v period rekonstruktsii," Klub i revoliutsiia 13/14 (1930): 58.
31. "Material dlia khudozhestvennykh kruzhkov v kampanii otpusknikov," June 1930, GTSTM, f. 150, d. 36, l. 22.
32. V. Darskii, "Agitbrigada na lesozagotovkakh," Za agitpropbrigadu i TRAM 1 (1931): 61; Eidinov, Polonskaia, Lekitskaia, "Chetvertaia fabrika Moskvoshvei," KS 4 (1932): 42.
33. N. I. L'vov, "Raport o rabote agit-brigady kluba im. Rusakova soiuza kommunal'nikov," GTSTM, f. 150, d. 37, l. 7. See also A. Gol'dman and M. Imas, Sotsialisticheskoe sorevnovanie i udarnichestvo v iskusstve (Moscow: Gosudarstvennoe izdatel'stvo khudozhestvennoi literatury, 1931), 65.
34. K. Barinov, "Sryvaiushchikh rabotu pod obstrel," KS 11 (1931): 68; Rumiantsev, "Agitpropbrigady zheleznodorozhnikov v boiu za chetkuiu rabotu transporta," KS 4 (1931): 8.
35. V. Rudman, "Dva mesiatsa na zheleznodorozhnykh stroikakh," Za agitpropbrigadu i TRAM 2 (1931): 53–54.
36. Ella Winter, Red Virtue: Human Relationships in the New Russia (New York: Harcourt Brace, 1933), 56.
37. P. Surozhskii, "Na obshchestvennom prosmotre igro-plakata," KS 10/11 (1930): 11.
38. "Na rel'sy reorganizatsii," KS 11 (1931): 3.
39. Vas., "Khudozhestvennye agit-brigady v derevniu na pomoshch' posevnoi kampanii," KS 2 (1930): 19.
40. V. Savkin, "Na khlebnom fronte," KS 11 (1931): 70; I. Elik, "V bor'be s klassovym vragom," KS 4 (1931): 6.
41. Shorshe, "Kak my rabotali v derevne," KS 11 (1931): 71–73, quotation 73.
42. N. Strel'tsov, "Agitpropbrigadnoe dvizhenie v novoi obstanovke raboty profsoiuzov," Za agitpropbrigadu i TRAM 2 (1931): 6. See also S. Bardin, "Agitflotiliia MOSPS na propolochnoi kampanii v kolkhozakh Moskovskoi oblasti," ibid., 48–49.
43. On Soviet bond programs, see James R. Millar, "History and Analysis of Soviet Domestic Bond Policy," in Susan Linz, ed., The Soviet Economic Experiment (Urbana: Illinois University Press, 1990), 113–19.
44. On popular protest against government loan campaigns during the Second Five-Year Plan, see Sarah Davies, Popular Opinion in Stalin's Russia: Terror, Propaganda and Dissent, 1934–1941 (Cambridge: Cambridge University Press, 1997), 35–37, 64.
45. N. Sen, "S agitatsiei zaima po novostroikam," KS 7/8 (1931): 27–30.
46. These labels, with some variation, were widely used during the plan. See Valentin Kataev, Time, Forward!, trans. Charles Malamuth (Bloomington: Indiana University Press, 1976), 26–31; Winter, Red Virtue, 58–59; and the agitational films of Nikolai Medvedkin, depicted in Christopher Marker's documentary The Last Bolshevik.
47. A. Ovchinnikov, "Agitpropbrigady i realizatsiia novogo zaima," Za agitpropbrigadu i TRAM 1 (1931): 27–28; N. Sen, "Agitbrigada 'Postroika' v Bobrikakh," ibid., 29–33; "Daesh' zaem!" KS 7/8 (1931): 30–33.
48. Katerina Clark, "Little Heroes and Big Deeds: Literature Responds to the First Five-Year Plan," in Sheila Fitzpatrick, ed., Cultural Revolution in Russia, 1928–1931 (Bloomington: Indiana University Press, 1984), 189–206.
49. V. Pavlov, "V polose otchuzhdenii," KS 3 (1931): 12–15; D. Korobov, "Za udarnyi zavod," KS 6 (1931): 35–40; M. B. Reznik and A. A. Fedorovich, "Vstrechnyi," Za agitpropbrigadu i TRAM 2 (1931): 29–36. See also V. Shneerson, "Spetsialisty, kul'trabota i bor'ba s vreditel'stvom," Klub i revoliutsiia 21/22 (1930): 6–14.
50. B. V. Shmelev, "Ot liubitel'skogo dramkruzhka—k pervoi khudozhestvennoi kul'tbrigade," KS 7/8 (1931): 13.
51. Svetlana Boym, Common Places: Mythologies of Everyday Life in Russia (Cambridge: Harvard University Press, 1994), 8–9, 35–38.
52. M. B. Reznik and A. A. Fedorovich, "Vstrechnyi," Za agitpropbrigadu i TRAM 2 (1931): 35. For just one of many examples of scripts without female parts, see V. Pavlov, "V polose otchuzhdeniia," KS 3 (1931): 12–15.
53. Gail Warshofsky Lapidus, Women in Soviet Society (Berkeley: University of California Press, 1978), 95–103.
54. Winter, Red Virtue, 292.
55. V. Sibachev, "Stat' na golovu vyshe," Za agitpropbrigadu i TRAM 2 (1931): 4; see also Avlov, Teatral'nye agitpropbrigady, 27.
56. Susan Rubin Suleiman, Authoritarian Fictions: The Ideological Novel as a Literary Genre (Princeton: Princeton University Press, 1983), 147.
57. "Otkrytoe pis'mo ko vsem proletarskim pisateliam, poetam, dramaturgam, stsenaristam, kinorezhisseram, khudozhnikam," KS 3 (1931): 3.
58. B. Reines, "The Experience of the International Workers' Theatre as Reported at the First Enlarged Plenum of the I.W.D.U.," Workers' Theatre 1, no. 9 (December 1931): 4. There is a large literature on agit-prop theater in the West. See W. L. Guttsman, Workers' Culture in Weimar Germany: Between Tradition and Commitment, ch. 9 (New York: Berg, 1990); Daniel Hoffman-Ostwald and Ursula Behse, Agitprop, 1924–1933 (Leipzig: BEB Friedrich Hofmeister, 1960); Raphael Samuels, et al., eds., Theatres of the Left, 1880–1935: Workers' Theatre Movements in Britain and America (London: Routledge and Kegan Paul, 1985); Richard Stourac and Kathleen McCreary, Theatre as a Weapon: Workers' Theatre in the Soviet Union, Germany and Britain, 1917–1934 (London: Routledge and Kegan Paul, 1986).
59. Hans Gunther, "Einleitung," in N. F. Chuzhak, ed., Literatura fakta (1929; rpr. Munich, 1972), v–xvi; Fritz Mierau, ed., Sergei Tretiakov: Gesichter der Avantgarde (Berlin: Aufbau Verlag, 1985), 87–122.
60. Osip Brik quoted in Günther, "Einleitung," ix. See also Suleiman's assessment that highly politicized literature is an odd mixture of verisimilitude and didacticism, Authoritarian Fictions, 146.
61. I. Mazin, "O khudozhestvennoi agitpropgruppe," KS 1/2 (1931): 26.
62. G. Avlov, Klubnyi samodeiatel'nyi teatr: Evoliutsiia metodov i form (Leningrad: Teakinopechat', 1930), 17–19. Avlov saw this work as an update and correction of Tikhonovich's study on amateur theaters published during the Civil War, which had defined amateur theater as those venues where participants did not earn their living from their performances.
63. Avlov, Teatral'nye agitpropbrigady, 12, 25.
64. "Puti razvitiia agit-propbrigadnogo dvizheniia," GTSTM, f. 150, d. 37, l. 30.
65. See S. Tretiakov, "Esteticheskoe zagnivanie kluba," Revoliutsiia i kul'tura 11 (1928): 36–39; A. Perovskii, "Dvortsy kul'tury bez mass," Pravda 9 January 1929. For an assessment of the architectural discussion surrounding club construction, see Roanne Barris, "Chaos by Design: The Constructivist Stage and Its Reception," (Ph.D. Dissertation, University of Illinois at Urbana-Champaign, 1994), 395–400.
66. Boris Roslavlev, "Klubnyi vopros," Revoliutsiia i kul'tura 15 (1928): 52.
67. Ts. A., "V novom klube 'Kauchuk,' " NZ 20 (1929): 14.
68. "Net kul'tury v dvortsakh kul'tury," KP 24 April 1932.
69. Iurii Beliaev, "Klubnaia samodeiatel'nost' na zadvorkah," KS 6 (1930): 20; M. Korol'kov, "Nepreryvka v klube," NZ 42 (1929): 11.
70. On the continuous work week, see Lewis Siegelbaum, State and Society Between Revolutions, 1918–1929 (Cambridge: Cambridge University Press, 1992), 204–14; Sheila Fitzpatrick, Everyday Stalinism (New York: Oxford University Press, 1999), 239.
71. V. Khazanova, Klubnaia zhizn' i arkhitektura kluba, v. 1 (Moscow: Rossiiskii institut iskusstvoznania, 1994), 119–127; Gorzka, Arbeiterkultur, 260–64. For Komsomol proposals, see "Estafeta kul'turnoi revoliutsii," KP 25 July 1930.
72. Rumiantsev, "Agitpropbrigady zheleznodorozhnikov," 7.
73. S. P. Alekseev, "Agitpropbrigady—udarnaia brigada samodeiatel'nogo iskusstva," KS 4 (1931): 12; see also M. Reznik, "Opyt vskrytiia tvorcheskogo metoda agitprop-brigady LRROP," Za agitpropbrigadu i TRAM 1 (1931): 16.
74. Avlov, Teatral'nye agitpropbrigady, 24.
75. Lind, "Putilovtsy na perestroike," KS 7/8 (1930): 6.
76. D. Mirlas, "Na peredovykh pozitsiiakh," Za proletarskoe iskusstvo 3/4 (1931): 31–34.
77. See A. L. Sokolovskaia, "Fotoliubitel'stvo," in S. Iu. Rumiantsev and A. P. Shul'pin, eds., Samodeiatel'noe khudozhestvennoe tvorchestvo v sssR, v. 1 (Moscow: Gosudarstvennyi institut iskusstvoznaniia, 1995), 215–45.
78. Harriet Borland, Soviet Literary Theory and Practice during the First Five-Year Plan, 1928–32 (New York: King's Crown Press, 1950), 38–74.
79. Iu. Ianel', "Massovaia khudozhestvennaia samodeiatel'nost' i puti sotsialisticheskogo razvitiia iskusstva," Literatura i iskusstvo 3/ 4 (1930): 43, 59.
80. "Moskovskie teatry—v kolkhozy," Literaturnaia gazeta 24 February 1930; M. Imas, "Sotssorevnovanie i udarnichestvo v teatre," Sovetskii teatr 13/16 (1930): 22; A. Gladkov, "Brigada MKhATa v pokhode," ibid., 20.
81. "Udarnye brigady meierkhol'dtsev," Literaturnaia gazeta 29 September 1930; M. Imas, Iskusstvo na fronte rekonstruktsii sel'skogo khoziaistva (Moscow: Gosudarstvennoe izdatel'stvo khudozhestvennoi literatury, 1931), 23, 118.
82. Gabaev and Vronskii, "Nasha istoriia," KS 4 (1931): 25; Avlov, Teatral'nye agitpropbrigady, 26.
83. "Nachalsia smotr—sorevnovanie khudozhestvennykh agitbrigad," Malye formy klubnogo zrelishcha, 23/ 24 (1930): 2.
84. I. Isaev, "Partiinye lozungi na khudozhestvennom fronte," KS 3 (1930): 1; V. Vin, "K perevyboram sovetov," KS 2 (1930): 23.
85. L. Tasin, "Blizhaishie zadachi dramkruzhkov," ZI 36 (1929): 2. See also references to a "purge" of club leaders, in Pravda, "Kluby—opomye punkty na fronte kul'turnoi revoliutsii," 22 April 1930.
86. "Priblizhaemsia k IX s''ezdu profsoiuzov," KS 1/ 2 (1931): 2.
87. "Stenogramma metosektora po zaslushannomu doklad tov. L'vova o formakh agit-brigadnykh vystuplenii," GTSTM, f. 150, d. 58, l. 5; R. Pel'she, "K itogam smotra," in Khudozhestvennye agitpropbrigady, 5–11.
88. N. Bespalov, "Smotr povyshaet aktivnost'!" KS 6 (1932): 34. See also Iakov Vasserman, "Chto tormozit rost?," ibid., 37.
89. J. V. Stalin, "New Conditions, New Tasks in Economic Construction," in Problems of Leninism (Peking: Foreign Languages Press, 1976), 532–59, esp. 552.
90. L. Sokolov, "Za vysokoe kachestvo khudozhestvennoi samodeiatel'nosti," Za agitpropbrigadu i TRAM 2 (1932): 10.
91. "O perestroike literaturno-khudozhestvennykh organizatsii. Postanovlenie TsK VKP(b) ot 23 aprelia 1932 g.," Pravda 24 April 1932.
92. "Vsesoiuznyi smotr samodeiatel'nosti," RiT 20 (1932): 1–2.
93. "Olimpiada samodeiatel'nogo iskusstva," Sovetskoe iskusstvo 9 February 1932; "Samodeiatel'noe iskusstvo—litsom k proizvodstvu," Trud 22 February 1932.
94. "Smotr sorevnovanie khudozhestvennoi samodeiatel'nosti," Za agitprapbrigadu i TRAM 2 (1932): 4–5.
95. On First Five-Year Plan festivals, see Vladimir Tolstoi, Irina Bibikova, and Catherine Cooke, eds., Street Art of the Revolution: Festivals and Celebrations in Russia, 1918–1933 (London: Thames and Hudson, 1990), 167–229.
96. For reports of the opening celebration, see "Olimpiada zakonchilas', perestroiku na polnyi khod," RiT 24 (1932): 9; "Grandioznyi prazdnik iskusstva i tvorchestva mass," KP 8 August 1932; N. Khor'kov, "Simfoniia, kotoruiu sozdaet kollektiv," Trud 8 August 1932.
97. "Privet Vsesoiuznoi olimpiade samodeiatel'nosti," RiT 22 (1932): 1–2, quotation 2. See also "Vyshe ideino-khudozhestvennyi uroven' tvorchestva," KP 6 August 1932.
98. "Spisok kollektivov uchastvuiushchikh na pervoi Vsesoiuznoi olimpiade," GARF, f. 5451, op. 1, d. 789, l. 36.
99. "Zavtra otkryvaetsia olimpiada samodeiatel'nogo iskusstva," Trud 5 August 1932.
100. A. Fevral'skii, "Smotr samodeiatel'nogo teatra," Literaturnaia gazeta 17 August 1932; N. G. Zograf, "Zhivaia teatral'naia gazeta," RGALI f. 2723, op. 1, d. 425, l. 102; "Kto uchastvoval v olimpiade," Trud 17 August 1932; Iakov Grinval'd, "Vchera zakonchilas' olimpiada," 15 August 1932; "Sostav kollektivov uchastnikov pervoi Vsesoiuznoi olimpiady," GARF, f. 5251 (VTSSPS), op. 16, d. 789, l. 37.
101. Women were estimated to make up thirty-nine percent of the participants in union theater circles in 1929 (see Gabriele Gorzka, Arbeiterkultur in der Sowjetunion: Industriearbeiter-Klubs 1917–1929 ([Berlin: Arno Verlag, 1990], 298). On the gender composition of shock workers, see Kuromiya, Stalin's Industrial Revolution, 321, and Siegelbaum, Stakhanovism, 170–72.
102. "Spisok chlenov zhiuri," GARF, f. 5451, op. 16, d. 789, l. 61; "Stenogramma zasedaniia zhiuri," 10 August 1932, ibid., d. 793, 1. 116.
103. T. Kh., "Bogi stashcheny s olimpa," Pravda, 10 August 1932.
104. Disagreements were for the most part parochial—with critics defending their own local interests and organizational affiliation. Leningrad journals tended to defend Leningrad performances. Jury members from TRAM defended TRAM performances, etc.
105. "Agitpropbrigada zavoda 'Severnaia sudostroitel'naia verf'," GARF, f. 5451, op. 16, d. 767, l. 42; Al. Gladkov, "Luchshie sily professional 'nogo iskusstva na pomoshch' khudozhestvennoi samodeiatel'nosti," Sovetskoe iskusstvo 15 August 1932.
106. M. Iurenev, "Chemu ne aplodiruet zritel'," Trud 14 August 1932. See also Nik. Iu., "Otoiti ot shablonnykh 'agitmarshirovok,' " KP 11 August 1932.
107. GARF f. 5451, op. 16, d. 798, l. 94; Gladkov, "Luchshie sily"; Iu., "Otoiti ot shablonnykh 'agitmarshirovok.' "
108. A. Kasatkina, "Zametki o piatom dne," Izvestiia 12 August 1932. For the text of the play, see F. Knorre, Moskovskii 10:10 (Moscow: Gosudarstvennoe izdatel'stvo khudozhestvennoi literatury, 1933); on the Putilov troupe, see "Agitpropbrigada krasnoznamennogo zavoda 'Krasnyi Putilovets,' " GARF f. 5451, op. 16, d. 798, l. 121.
109. "Agitpropbrigada zavoda 'Severnaia sudostroitel'naia verf'," GARF, f. 5451, op. 16, d. 797, ll. 40–41; "Perestroiku na polnyi khod," RiT 24 (1932): 12–13.
110. "TRAm Moskovskogo raiona," GARF, f. 5451, op. 16, d. 797, ll. 36–38; "Agitpropbrigada Izhorskogo zavoda," GARF f. 5451, op. 16, d. 798, l. 2.
111. Ia. Boiarskii, "Znachenie olimpiady—na printsipial'nuiu vysotu," KS 7/8 (1932): 4.
112. "K sporam o Stroike," Tribuna olimpiady 13 August 1932.
113. "Peredoviki khudozhestvennoi samodeiatel'nosti," Trud 17 August 1932. The Izhorsk agitprop brigade won sixth place; the Moscow Trade Workers Brigade won seventh.
114. See, for example, A. Kasatkina, "Iskusstvo millionov," Izvestiia, 22 August 1932.
115. Pervaia Vsesoiuznaia olimpiada samodeiatel'nogo iskusstva (Moscow: Profizdat, 1932), 10.
116. Ibid., 11, 19, 24, quotation 19.
117. S. P. Alekseev, "Samodeiatel'nyi teatr na olimpiade," KS 9 (1932): 1–7, quotation 3.
118. On the broad range of the debate about the meaning of socialist realism in its formative period, see Régine Robin, Socialist Realism: An Impossible Aesthetic, trans. Catherine Porter (Stanford: Stanford University Press, 1992), chs. 1 and 2. On socialist realism and pleasure, see Richard Stites, Russian Popular Culture: Entertainment and Society since 1900 (Cambridge: Cambridge University Press, 1992), 64–97.
119. Ivan Chicherov, "Stenogramma Vsesoiuznogo soveshchaniia teatral'nykh rabotnikov profsoiuznykh teatrov," December 1932, GARF f. 5451, op. 16, d. 788, ll. 6, 16–18.
120. "Pervye itogi," Sovetskoe iskusstvo 21 August 1932.
121. Sheila Fitzpatrick, "Cultural Revolution as Class War," in Fitzpatrick, ed., Cultural Revolution, 8–40.
122. N. L'vov, "Kluby v dni Oktiabria," NZ 46 (1928): 9. See also I. V., "Kluby v oktiabr'skie dni," Pravda 11 November 1928.
123. See Michael David-Fox, Revolution of the Mind (Ithaca: Cornell University Press, 1997), 254–72; idem, "What Is Cultural Revolution," Russian Review 58 (April 1999): 181–201.
## 6
## Amateurs in the Spectacle State
IN HIS fascinating discussion of the differences between early Soviet culture and the 1930s, the architectural historian Vladimir Papemyi argues that Stalinism was fixated on visual display.1 The architecture of the period began first with the façade as a means of illustrating power. Interior spaces featured grand foyers and decorative objects designed to impress users. While many theaters in the 1920s tried to liberate the viewer and offer different perspectives on events, stages in the 1930s were designed to fix the spectators' point of view. Nor was this concern for display limited to architecture. The mass festivals common to the early Soviet years, dispersed throughout the streets and squares of urban centers, were now reserved for special locations—with events in Moscow's Red Square gaining preeminent importance. Spectacles of individual heroism, such as the thrilling exploits of long-distance pilots and polar explorers, were the focus of an enthusiastic national press coverage.2
The Stalinist emphasis on impressive forms of presentation led to the creation of what we might call a "spectacle state" in the 1930s, a polity in which power was conveyed through visual means. As James von Geldern has argued, Soviet citizens' knowledge of state policies and public affairs was communicated increasingly through spectacles during this decade, including show trials, parades, conventions, and highly publicized state initiatives soliciting public input. The state invested considerable funds into new venues for spectacles, building massive sports stadiums, movie theaters, workers' palaces, and public parks. In the process, contends von Geldern, the spectator became the model for the ideal Soviet citizen.3
This aesthetic and political turn had profound consequences for amateur theaters. Despite their many vicissitudes in the years up until the First Five-Year Plan, the ethos of amateur stages had been first and foremost participatory; it was more important what they did than how they looked. They were valued for the contributions they made to neighborhood collectives, club entertainments, and political campaigns. In the 1930s, however, standards of judgment changed radically. Amateur performances received credit primarily for their aesthetic impact on specta-tors—how impressive their stages and sets were, how polished their acting techniques, how well they approximated professional standards of performance. By setting professional performance standards for amateur theaters, cultural institutions aimed to raise their value as spectacle. They became visual evidence of the kul'turnost', or "culturedness," of the formerly rough urban lower classes. The introduction of the aesthetic system of socialist realism, murky as it was in its first formulations, further served to guarantee a merger of amateur and professional art. If there was only one acceptable artistic method, intoned critics, then the distance between different forms of artistic expression could not be great.4
To effect this transformation, many circles in the capital cities established close links to professional stages and engaged a permanent professional staff. Rehearsal time increased exponentially. Amateur circles became a path to channel talented participants on to a lifetime career on the professional stage. In the process, however, amateur theaters lost any special place they might once have held in the lives of their audiences. In the 1920s, the lines between actors and spectators had not been rigid, both because of the location of the stage and the fact that one day's viewer might be included in the next day's play. But the changing architectural standards of the 1930s distanced performers from the audience; so did the much more rigorous training programs. As a result, amateur theatricals turned into a profession of sorts, although it remained unpaid. It was no longer open to every enthusiast.
#### The Locus of Performance
Amateur stages began to approximate professional theaters in all aspects in the 1930s, including their physical environment. The club building boom of late NEP and the First Five-Year Plan had created spacious stages for select clubs in both capital cities. With performance halls seating thousands, large stages, and healthy financial support (sometimes stemming directly from ticket sales), these club theaters could mount elaborate productions with impressive sets. The KOR Railroad Workers' Club in Moscow accommodated an audience of twelve hundred viewers and had special seats set aside for Stakhanovites. The drama circle had a separate building for its activities, including a special library, an administrative office, and costume and set design studios. The Krasnyi Bogatyr factory club spent fifteen thousand rubles on the scenery for a single play, at a time when many club budgets were not much higher.5
Although the biggest building boom came in the late 1920s and the years of the First Five-Year Plan, money continued to pour into new club structures in the 1930s. Pravda proudly announced new club buildings at the Putilov and Sestropetskii factories in 1933. The Sickle and Hammer factory opened a new club building in the same year, with an auditorium seating one thousand.6 Two separate architectural completions during the First Five-Year Plan had produced a wide range of designs for a Palace of Culture in Moscow's Proletarian District. They finally came to fruition in 1933, when a theater space for twelve hundred viewers opened. This hall not only provided entertainment for local neighborhoods, it also obliterated key elements of the old culture, since it was built on the grounds of the Simonovskii Monastery.7
But while performance spaces expanded, theater circles paradoxically became less important to the life of the club. The massive new club buildings of the 1930s contained rooms for a multitude of activities that could take place simultaneously, unlike the serial use of space in smaller buildings. Their expanded budgets supported buffets, cafeterias, and sometimes even stores, together with a wide array of innovative amateur artistic projects, including photo circles and jazz orchestras.8 With such a range of activities available, some club users boasted that they spent all their free time there, coming right after work and staying until it was time to sleep. Expanding entertainment opportunities made the work of amateur theater groups less central to club activities. Rising standards of performance reinforced this fact, as drama circles radically increased their rehearsal time and decreased the number of works they staged each year. Some circles planned only two plays during a season.
The decline in agitprop brigades meant that far fewer drama circles were prepared to partake in quickly organized agitational work and join in the small festivities and holidays that marked Soviet public life. As a result, the huge new halls were frequently filled with other forms of entertainment. Professional theaters could now bring their elaborate productions into workers' clubs much more easily. In fact, many of the new venues were used primarily for that purpose. The impressive new stage at the Proletarian District Palace of Culture hosted the cream of Moscow productions, including works by the Malyi Theater, the Theater of the Revolution, and the Meyerhold Theater. These new stages also facilitated visits of professional writers to working-class neighborhoods, where they read their works to assembled club members.9
Clubs built in the 1930s exhibited the kind of "elegant" details that marked Stalinist architecture in general in this period. The nomenclature of "workers' palace" (rabochii dvorets), claimed by club activists since the revolution, began to be taken literally by both architects and club users. Architects emphasized fine details in the façade. They constructed entry spaces that looked similar to hotels, with marble walls, fountains, flowers, and paintings. Some wealthy new clubs ordered furniture specifically designed for their interiors. One Leningrad club even boasted a doorman!10 Although the discussions around clubs still emphasized their importance as public space, architects stressed the need to make the club cozy and inviting. The club became the image of an ideal home that individuals could not yet achieve on their own.11
Despite the expansive new buildings, club life continued to raise problems for those activists who had hoped they would significantly transform leisure time activities. Trade union cultural workers conducted periodic "sweeps," making surprise club visits to assess their cleanliness, comfort, and range of activities. They were invariably disappointed, finding club spaces dirty and inhospitable.12 When national newspapers opened their culture pages to club users, their complaints sounded strikingly familiar to discontents voiced during NEP. Letters in Pravda about the Kauchuk Club listed a number of problems. This club was one of the premier structures of the Stalin era, boasting a building designed by the famous architect Konstantin Mel'nikov and a theater staff headed by directors from the Vakhtangov Theater.13 But users were nonetheless dissatisfied. Tickets to evening events were not distributed according to a fair system, and the events themselves started too late to fit into the schedules of busy workers. "The club we are writing about is not bad," read one letter. "It has a new, good building, a big auditorium. It recently won a competition. Nonetheless it still has problems serving the cultural needs of workers."14 A number of Leningrad clubs were cited for providing poor cultural resources to their members. The club "Promtekhnika" was singled out in Krasnaia gazeta for its lack of comfort and limited programs: "In Leningrad, there are thousands of amateur artistic circles, there is an army [of cultural workers]. But with all these massive resources, this rich material base, we need to bring club work up to a level to meet the tasks of the moment." This club had fallen sadly behind, concluded the article, not giving workers a sense that they could go there to improve themselves after working hours.15
While some club participants complained that their shiny new buildings had still not met their needs, others were struggling with older and more familiar problems. Not all clubs were reconfigured in the 1930s building boom. Participants in the drama circles at the Elektrozavod factory in Moscow were not housed in grand spaces. They retained a very modest rehearsal room that was much too small to meet their needs. They had to wait for hours to gain access to an auditorium.16 Other clubs suffered from serious structural weaknesses, including leaking roofs, that made it difficult to continue work.17 The cultural distance between these ordinary gathering places and the new "workers' palaces" increased exponentially.
Tensions between older and younger workers, a frequent theme in the club literature of the 1920s, continued into the next decade. Older workers complained about overcrowded rooms used by young people for their "trysts." They felt excluded from gatherings that seemed designed for young people, such as dance evenings. "At such events it is rare to see people like us, older workers and their wives," complained a group of long-time members of the Tsiurupy Club in Leningrad. The club needed a special room where older workers could go in peace.18 Young workers brought their own complaints against the new club system of the 1930s. In order to finance the expensive new buildings, clubs began to charge fees for a variety of services, sometimes even for hanging up coats and using the library. Young people argued that this discriminated against youth, who did not have much income to spare. "The NEPman, commercial culture should be banned from houses of culture," complained young workers in a letter to Komsomol'skaia pravda.19
#### Enter the Experts
After the National Olympiad of Amateur Art in 1932, the status inversions of the preceding years, when amateurs were at times deemed superior or at least equal to professionals, ended abruptly. Both trade union and Narkompros bureaucracies urged amateur theaters to form solid links with professional stages. Up until 1936, it did not matter which professional method amateurs chose, as long as it provided a standardized training system. Teachers who tried to come up with their own methods, without a nod to one of the established theaters, could find themselves accused of "schematism."20 Choices narrowed when the "formalism controversy," discussed below, broke in early 1936. Then groups with ties to the Meyerhold Theater began to receive negative reviews.21
Numerous professional theaters established patronage relationships with amateur stages in the theater-rich cities of Moscow and Leningrad. The Central Trade Union Theater (Teatr VTSSPS) in Moscow took oversight over a number of groups, including the Moscow-Kursk Railroad Club and the Metro Subway Builders' Club. The Meyerhold Theater worked together with the Elektrozavod factory. The Malyi Theater claimed control over the amateur work of the Sickle and Hammer factory.22 In Leningrad, the most active patron was the Leningrad Oblast Trade Union Theater, which sent instructors to more than fifteen local groups.23
The meaning of "professionalization" changed in the 1930s. During the First Five-Year Plan, a number of formerly amateur groups found funding sources to free them from their day jobs. But this step was not enough to gain them widespread recognition as a professional circle. In addition, these groups had to embrace an established professional acting method. This was strikingly evident in the reorganization of the Moscow TRAM in 1933. This circle had technically turned professional in 1930: actors were paid and expected to work full-time in the theater. However, it was only when the artistic directorship was taken over by Il'ia Sudakov from the Moscow Art Theater that TRAM began to receive critical attention as a serious professional theater. The new training program at the Moscow TRAM theater introduced key elements from Stanislavsky's method, with actors learning to "live their parts."24 By 1934, some theater critics in Moscow were comparing the TRAM collective to the original Stanislavsky circle. "Isn't a new cherry orchard being planted with [TRAM'S] laughter and jokes?" queried one newspaper review, referring to one of the Chekhov plays that had established the Moscow Art Theater's reputation.25
As this review indicates, the Moscow Art Theater (MAT) came to hold an elevated status among Soviet theaters in the 1930s. As evidence of its prestige, MAT directors and actors gained more influence over amateur stages. In addition to the Moscow TRAM, numerous club stages could boast workers from the famous theater, including the Krasnyi Bogatyr' factory, the Kukhmisterov Drama Club, and the Iakovlev Tobacco Workers' Club. MAT actors worked informally to supervise training programs, including those at the Central Woodworkers' Club. Some very well-off factories even paid to send their best actors to train at MAT, as was the case for one amateur actor from the Stalin auto plant.26
Training regimens in amateur theaters became much more rigorous. The Kauchuk Theater's program, led by Vasilii Kuza from the Vakhtangov Theater, required mandatory five-hour rehearsals five nights a week. Rehearsals lasted up to eight months on a single play. A range of teachers were involved in instruction, including a balletmaster. The Vakhtangov Theater's own design studio provided costumes. Interviews with eager club actors indicated that they took the new approach very seriously and looked upon their previous efforts as a kind of "bad dream."27 Teachers from professional theaters also wrestled with the bad dream of earlier training, as they struggled to make students adapt to the rigor of new methods.28
With the turn to higher professional standards, many stages attempted to present offerings with greater social and historical authenticity. That meant visits to libraries and museums to learn about the social settings of chosen plays. To prepare for their performance of Gossiping Women (Bab'i spletni) by the eighteenth-century Italian playwright Carlo Goldoni, the actors at Leningrad's Iakovlev Club listened to Italian music and took trips to the Hermitage museum. Another Leningrad theater began correspondence with agents from the secret police in order to deepen their presentation of criminals transformed by the Soviet criminal justice system, a favorite theme of drama and literature in the 1930s.29
The cultural press presented work in amateur theaters as a social responsibility for artistic professionals. Interviews with directors often included earnest statements about the importance of providing assistance to club stages. Some directors offered their free time and even their apartments as rehearsal space.30 The British leftist director Andre Van Gyseghem was amazed at the sense of responsibility that professionals felt for amateur actors, a relationship much different from what he knew in the capitalist West.31 Whether their motivation came from idealism or material need, well-trained theater workers could earn considerable sums in amateur circles. The musician Juri Jelagin was able to triple the salary he earned at the Vakhtangov Theater by taking jobs in clubs. One Stanislavsky student, Andrei Efremov, supplemented what he earned as an instructor at the Moscow Theater Institute by participating in club training courses.32
These patronage relationships served to increase the distance between a few important club stages and the rest—not to mention exaggerating the difference between club theaters in the two main cities and the provinces. While some theaters in Moscow and Leningrad could boast palatial settings and an outstanding staff, others continued to struggle with supply problems reminiscent of the 1920s—with no place to rehearse and very few props.33 Provincial theaters were even worse off. Representatives to a trade union club conference in 1935 complained that they did not even have enough funds to buy an adequate supply of scripts, let alone be overly concerned about production techniques.34
Club directors had to assume much more responsibility for the quality of the finished product in the 1930s. Their work was put to the test in periodic amateur theater competitions. Those who evoked fine performances won accolades, but those who did not were singled out for their poor practices and sometimes could even lose their jobs. As one jury member at a Moscow club competition explained it, "The young people are not to blame. We have to blame the leadership."35 Directors who chose the "wrong" plays, like the Red Woodworker Club's production of a light-hearted Kataev work in 1935, were simply barred from higher levels of competition.36 One club director issued a long mea culpa after his group came in last in the final round of the Moscow city-wide competition in 1937.37 This atmosphere of insecurity and distrust made skilled drama instructors even harder to find, which in turn increased suspicion against them. A commentator in Pravda observed, "The leaders of drama circles are often self-seekers and amateurs. They are hounded out of one position but almost immediately find a place somewhere else."38
#### Reshaping the Repertoire
During the 1930s, the repertoire of amateur stages began to mimic that of professional theaters. Gone were the homemade works generated by club members themselves or minor authors who wrote strictly for club stages. Most contemporary plays by Western authors also van-ished.39 Like their professional counterparts, club theaters chose "hit" Soviet plays—like Konstantin Trenev's Liubov' Iarovaia and Vladimir Kirshon's The Miraculous Alloy (Chudesnyi splav); they supplemented these works with a limited list of Russian and Western European prerevolutionary classics.
Oversight of amateur theater activity in clubs fell largely to trade union organizations. Although local Narkompros affiliates and city soviets still exercised some oversight and provided occasional funding, national and city trade union bureaucracies were the most directly involved in the daily life of club theaters. Through periodic conferences of club workers, trade union leaders urged local theaters to take their cultural tasks more seriously and turn away from agitational methods.40 The cultural division of the national trade union bureaucracy founded the journal Klub in 1933, which offered the most detailed published accounts of amateur theatrical activity during the 1930s.41 The Central Trade Union Theater opened a special school to train actors and directors for club circles.42 In addition, both Moscow and Leningrad had their own professional trade union stages that premiered repertoire suitable for club theaters, performed in clubs, and encouraged talented workers to make their way to the professional stage.43
Some trade unions sponsored methodological seminars to help their affiliates move to a new kind of repertoire.44 Just as they had in the early 1920s, club leaders and members put forward compelling "transformation stories" tracing their about-face in aesthetic methods. This time, however, they were moving from small forms to large ones. In one such tale, the Communist Party decision of 1932 galvanized members of the Theater Collective of Red Woodworkers into action. "The big play has returned once again and occupies its former position," noted the club leader Iakov Mostovoi. "But work on big plays must follow a new path and become deeper and more serious."45 Agitprop brigades did not disappear entirely, although they lost their centrality to club life. Theaters affiliated with the Moscow Chemical and Rubber Workers' Union, which included the Kauchuk factory, sponsored several different levels of theatrical activity. At the top came the premier club theaters, now run by professional directors and teachers. However, they continued to support agitprop brigades that staged works with titles like The Loan Campaign in the Second Five-Year Plan. The factories also sponsored smaller drama circles that worked on both small and large forms, ranging from Gorky plays to "thematic literary evenings" on topics such as Bloody Sunday.46 Yet even when clubs and factories continued to support brigades, their performances gradually became interchangeable with drama circle offerings. A Leningrad brigade performed a Molière play for the May Day holiday in 1933, a choice that would have been unthinkable a few years before.47
Some critics in the early 1930s continued to defend the need for agitprop brigades because they could provide commentary on local workplace issues and respond quickly to state campaigns. The high-level trade union official, Dmitrii Marchenko, warned against the assumption that all agitprop brigades must someday turn into drama circles; all too often, amateur groups used the circles to avoid their social responsibilities altogether. "This theory is mistaken and politically dangerous," he intoned.48 But the turn toward a professional ethos undercut his own strictures; the professional models that amateurs were now emulating were hostile to the organizational principles of agitprop brigades.
One author sympathetic to agitational theatrical forms lamented on the shift in repertoire taking place in the course of 1933. Very few works were written specifically for club stages anymore, observed A. Kasatkina, a writer for Pravda. Some playwrights, such as Vladimir Bill-Belotserkovskii, were willing to adapt their work for amateur theaters. However, other big names in Soviet theater, like Alexander Afinogenov and Anatolii Glebov, ignored these venues. That meant that club repertoires looked strikingly similar to professional theaters. She estimated that twenty to thirty percent of the plays on club stages came from the classics, with fifty to seventy percent coming from the professional stage. Only ten to fifteen percent of the works performed included any local contributions. While she praised the turn away from poor-quality agitki, Kasatkina lamented that amateur stages were losing their separate identity and their ability to address local problems.49
Works written specifically for amateur stages did not disappear entirely before the anti-formalist campaign brought much tighter scrutiny over repertoire. Popular club authors from the 1920s, such as Vladimir Severnyi, still composed plays for club audiences. Konstantin Finn and Vitaly Derzhavin wrote works aimed primarily at amateur actors. The most prominent of these authors was Aleksei Arbuzov, who won praise in Kasatkina's, article. Arbuzov would become a famous playwright in the Khrushchev era; however, in the 1930s his works were mainly performed by amateur theaters or by professional theaters that had recently made the transition from amateur status. He got his start as a writer of agitprop pieces during the First Five-Year Plan. In the 1930s he made his own transition to longer works. His lighthearted Six Lovers (Shestero liubimykh) found an audience in amateur theaters, like the Moscow Tobacco Workers' TRAM. His more serious investigation into the building of the Moscow subway, The Long Road (Dal'niaia doroga), was also staged by the Moscow Central TRAM.50
Some 1930s playwrights composed crossover works popular on both amateur and professional stages. One of the best examples is Ivan Mikitenko's Girls of Our Country (Devushki nashei strany), first performed by the Leningrad Drama Theater early in 1933. Later the same year the Moscow Central TRAM picked this play as its debut performance under new leadership from the Moscow Art Theater. By the summer, several important amateur stages, including the Krasnyi Bogatyr factory in Moscow and the Vyborg House of Culture in Leningrad, chose this work for summer theater competitions.51 Addressing the transformation of gender roles and romantic relationships during the First Five-Year Plan, this work centered on the formation of a female shock work brigade to lay concrete for an electrical power plant. This popular play also offered evidence of the transformation of socialist culture toward more traditional models. In it, Komsomol members read Lermontov and play the cello. They are, in Mikitenko's words, already a working-class intelligentsia.52 Several critics objected to the portrayals of the young women in the play, who, according to one, "acted like private school girls [institutki] from Charskaia [the popular prerevolutionary writer of literature addressed to girls]."53
Another widely performed work on both amateur and professional stages was Miraculous Alloy (Chudesnyi splav) by Vladimir Kirshon. This play was a winner in a national competition for the best new work in 1934. During the 1934–35 season, it played concurrently at the Moscow Art Theater, the Theater of Satire, the Moscow TRAM, and several smaller professional theaters in the city. In addition, it was widely distributed to amateur stages. The journal Kolkhoz Theater sent out copies to all of its subscribers in early 1935.54 The play offers a humorous look at scientific work among Komsomol students who are searching for a new alloy for airplane construction. In the process, the disparate group of students who come from different social strata and geographic areas, forges itself into a tightly knit collective. At the conclusion of the play, the institute director announces to the group: "You yourselves are the miraculous alloy, my friends, the most steadfast material against corrosion."55 The play remained a favorite on amateur stages until Kirshon fell victim to the purges of 1937.
A scene from the Moscow TRAM'Sperformance of The Girls of Our Country (Devushki nashei strany). Teatr i dramaturgiia 4 (1934).
The 1934 play Aristocrats (Aristokraty) by Nikolai Pogodin was also popular on both professional and amateur stages.56 It examines the reeducation of prisoners sent to build the White Sea Canal, a Stalinist work project administered by the secret police. The play traces the rehabilitation of thieves, prostitutes, and "wrecker" engineers who had been imprisoned for sabotaging Stalinist industrial programs. Inspired by the benevolent models of their secret police overseers, these outcasts are turned into useful, patriotic citizens through their labor. This panegyric to forced labor was even performed at the White Sea Project itself by imprisoned amateur actors.57
As choices among contemporary plays narrowed, amateur stages turned to the prerevolutionary classics. The Soviet regime sponsored a select list of classic authors as part of its general effort to show the cultural achievements of socialism. Public celebrations of these literary giants were a central feature of the Stalinist spectacle state. Festivities included elaborate performances, lectures, newspaper articles, carnivals, and even special consumption items. Among those feted were Alexander Pushkin, Anton Chekhov, Maxim Gorky, Lev Tolstoi, and Lope de Vega.58 The celebration marking the centennial of Pushkin's death went on for more than a year, from late 1935 until early 1937. Pushkin's works were integrated into the school curriculum, and special Pushkin cakes went on sale.59 Amateur stages were drawn into the festivities. They staged a spate of short Pushkin works and even a few attempts at the opera Evgenii Onegin.60 When Gorky was commemorated, Moscow amateur theaters staged a ten-day festival honoring his work.61 The official reevaluation of Chekhov, who been rejected by many during the 1920s for offering a overly sentimental look at the dying aristocracy, brought a resurgence of his plays on club stages.62
By far the most popular Russian author was Alexander Ostrovsky, whose plays came to assume a greater prominence than they had enjoyed since the revolution. At a 1933 competition of amateur theaters in Leningrad, more than fifty percent of the works performed were from the classics, with Ostrovsky as the most popular playwright.63 His plays also dominated the offerings at Moscow club theaters in the summer of 1934.64 Amateur stages making the difficult transition from an agitational repertoire to more conventional plays often made an Ostrovsky work their first choice. When the Kukhmisterov Club theater was reorganized under new management in 1934, its opening work was The Ward (Vospitannitsa).65 Indeed, Ostrovsky's plays had made such inroads into amateur performances by 1936 that trade union club activists determined that his Not a Cent and Suddenly a Windfall (Ne bylo ni grosha, da vdrug altyn) and A Family Affair (Svoi liudy—sochtemsia!) were the most likely representatives of the classics in club repertoires.66 The fiftieth anniversary of Ostrovsky's death, celebrated in the summer of 1936, only served to heighten his popularity. "A broad program of mass performances of Ostrovsky's work is being carried out in plants, houses of culture, and red corners," reported a Leningrad cultural journal. All the major factories sponsored lectures and performances of works by the nineteenth-century playwright.67
Playwright Anatolii Glebov (who had gotten his start in the club theaters of the 1920s) tried to assess Ostrovsky's central place on amateur stages. In a long article published in 1937, he cited reams of statistics showing that Ostrovsky's plays constituted almost twenty percent of all those performed on kolkhoz stages; in the Soviet Union as a whole, they were performed by amateurs some 150,000 times a year.68 Glebov argued that the reason for the playwright's success was that he wrote for the broad masses: "The simplicity and strength of his composition, the richness and populism of his characters and language, and, most important, the deep understanding of the psychology of the people, his ability to think their thoughts—this is why Ostrovsky attracts the Soviet worker and collective farmer."69 We might add that Ostrovsky's plays were published in large editions in the 1930s, making them readily available. In addition, Ostrovsky's works were one constant element of the classical repertoire that could claim an unbroken lineage on the amateur stage going back to prerevolutionary days, giving the playwright the advantage of familiarity. One participant in the Kauchuk theater production of A Family Affair confessed that she had a long history in amateur theater: "I've performed for a long time, ever since childhood, in many Ostrovsky plays."70
Amateur repertoire was policed through periodic amateur theatrical Olympiads sponsored by trade unions. These gatherings were a method to bring the practice of individual groups under the scrutiny of trade union officials and state cultural functionaries, who dispensed prizes and sanctions for winning and losing circles. Leningrad unions had instituted annual summer Olympiads already in 1926. Moscow amateur theaters never had the same kind of regularized annual competitions, but they performed in a variety of union-wide and city-wide Olympiads and "dekady," ten-day festivals, throughout the 1930s. In both cities, competitions had a celebratory spirit, allowing theater participants to show their accomplishments and win the attention of the press. These competitions were small versions of the big Stalinist festivities of the 1930s, with grand opening ceremonies, parades, and mass gatherings in central parts of the city.71
The 1933 summer competitions in Leningrad and Moscow were the first chance amateur theater circles had to show that they had indeed been transformed by the decisions made in the wake of the 1932 national Olympiad. The turn to the classics at the Leningrad competition, where more than half of the plays were written before the revolution, was widely interpreted as an indication that amateur circles had decided to improve their repertoire.72 Groups in Moscow faced a rigorous panel of judges that included the writer Fedor Gladkov. These observers sharply criticized theater circles that had not practiced enough and censured works that appeared "as if they were two years old"—a reference to overly agitational plays that still bore the stamp of the First Five-Year Plan.73 At competitions the following year, some groups continued to disappoint critics. Five agitprop brigades performed in the annual Leningrad competition in 1934, showing what one commentator denounced as "1930-style work."74
Competitions vividly revealed the standardization of amateur repertoire. In the early 1930s there were still embarrassing gaffes at these events, such as when one factory circle chose to perform a long-forbidden play by Sofia Belaia, The Abandoned Children (Besprizornye) in a 1934 festival in Moscow's Proletarian District.75 By 1938, such errors had been eradicated. At a Leningrad competition that year, only fifteen plays were allowed in the final competition.76 The elaborate Moscow city competition of amateur theater in the spring and summer of 1938 featured a strikingly limited program. Half of the seventy-six theater groups that made it to the second round of the competition chose prerevolutionary works, and almost half of these were by Ostrovsky, the safest and most conventional choice.77
#### Aesthetic and Political Purges
By the second decade of the revolution, purges were a well-established fact of life within the Communist Party, which conducted periodic "cleansings" of its membership. The practice of denunciation and expulsion was integrated into universities in the 1920s. In addition, already during the First Five-Year Plan, warring artistic factions learned how to paint their opponents as political outlaws. Thus, the waves of purges that shook the nation in ever-widening circles in the 1930s were not without precedent. What was different was the extent and the consequences. Early in the decade, being "purged" from the Communist Party had significant but still non-lethal results. The victims, like the cultural director at the Kauchuk Club, lost his party membership and his job. His particular crime was "political illiteracy."78 As the decade progressed, however, purges drew in a much wider public and ruined the lives of those caught up in the whirlwind. By 1937, the start of the Great Purges, the accused were charged with wrecking, spying, and treason. They now faced imprisonment, exile, and execution.79
Soviet cultural establishments also underwent ideological purges during the decade, as the government tried to shape a unified cultural doctrine for the nation. An important channel of artistic standardization was the Soviet Writers' Union, inaugurated in 1934. In early 1936, the state took an additional step by forming the National Committee of the Arts to oversee theater, film, literature, music, and the visual arts. This new body waged an aggressive battle, known as the "anti-formalist campaign," against the remnants of pre-1930s cultural ideas. As the purges progressed, accusers conflated political and aesthetic errors, linking the charge "formalist" to that of "Trotskyist."
The Arts Committee was headed by none other than Platon Kerzhentsev, author of Creative Theater and once a passionate advocate of a separate path for the amateur stage. He had already shown his intellectual flexibility by spearheading criticisms of TRAM during the First Five-Year Plan. Perhaps because of Kerzhentsev, the committee extended its attention beyond the professional arts to amateur work as well. It claimed "general methodological leadership" over the amateur arts and promised to coordinate the work of trade unions with other organizations sponsoring amateur artistic activity.80
At the same time the committee took shape, the national newspaper, Pravda, published an anonymous editorial attacking the work of the musician Dmitrii Shostakovich, in the first salvo of the anti-formalist campaign. Leonid Maksimenkov has compellingly argued that Kerzhentsev, acting for the committee, was the author of the Pravda publication.81 Entitled "Muddle instead of Music," this document denounced Shostakovich's previously acclaimed opera, Lady Macbeth of the Mtsensk District, as a dissonant cacophony, "a 'leftist' muddle instead of authentic, human music." As evidence that the opera was an example of petty-bourgeois, futurist experimentation, the editorial cited its popularity among bourgeois audiences in the West. In a clear sign to those involved in theater, the article called Shostakovich's errors remnants of "Meyerholdism" transferred into music.82
Republished in a number of key cultural journals, the attack on Shostakovich provided a basic vocabulary for the anti-formalist campaign, with its censure of works that were overly experimental, abstract, complicated, and "Western." At a national meeting of the Art Workers' Union shortly after the editorial was published, Kerzhentsev announced that the lessons of the Pravda article should be applied everywhere, since there were similar failings in all areas of the arts.83 Various artistic disciplines organized public gatherings at which individuals added their own contributions. At a meeting of theater professionals in the summer of 1936, speakers not only criticized themselves but used the occasion to attack both Shostakovich and Meyerhold. The stage director Nikolai Okhlopk, for example, determined that he himself was not free of all the elements of petty-bourgeois aesthetics of the early Soviet years. But he certainly never had gone to the extremes that Meyerhold had. He never allowed a jazz band in a Gogol play or made Ostrovsky characters walk a tightrope. Sergei Radlov complained that the problem of formalism, and of Meyerhold, was essentially one of arrogance; it was a process by which the director placed himself above everything else in the play, including the actors and even the author. On this occasion, Meyerhold himself spoke and defended his right to artistic experimentation, rejecting the charge that he had neglected content in favor of form.84
Amateur theaters were quickly drawn into the formalism controversy. One of the foremost contributors and defenders of the theater of small forms in the 1920s, Adrian Piotrovskii, was engaged in a very direct way. He was denounced by name in a second Pravda attack on Shostakovich, "False Ballet." This renewed salvo directed at the composer was aimed at his ballet Bright Stream (Svetlyi ruchei), for which Piotrovskii had written the libretto. The newspaper charged that the ballet, which was set in the Kuban, had nothing local or specific about it. The libretto depicted none of the peoples of the Caucasus with any individuality.85
As the campaign gained strength, amateur stages faced even greater scrutiny, especially during competitions. Theaters attempting innovative stagings of prerevolutionary plays quickly ran into trouble. Leveling the charge of formalism, critics attacked theaters that sought to update the classics. Declared one Leningrad observer, "In amateur theater the remnants of 'Meyerholdism' have not been extinguished.... One can find many examples of vulgar sociological approaches, especially in performances of Chekhov and Ostrovsky."86 A substantial list of productions were singled out for their "distortions" in a long article in the trade union journal Klub. The Aviakhim club's staging of The Marriage of Figaro was censured in terms taken straight from Pravda. Many aspects of the opera were not combined in an organic way, leaving viewers with a "muddle." This same article denounced the Leningrad Iakovlev Club's setting of Shakespeare's Merry Wives of Windsor in the contemporary era and the Bolshevik factory's performance of Gogol's The Wedding, staged in an exaggerated grotesque style. The TRAM circle from Moscow's Elektrozavod factory, which had long-standing ties to the Meyerhold Theater, was chided for its obvious elements of "Meyerholdism." The article concluded with an ominous-sounding warning, "Amateur theatrical work and popular art [narodnoe tvorchestvo] is imbued with foreign influences [chuzhdye vlianiia] because of the low level of club directors and the weak leadership... exercised by trade unions."87
Directors of amateur theaters issued their own statements of self-criticism and denunciation, just like the professionals. One Moscow club director, A. S. Azarkh of the Financial Bank Workers' Club, confessed to elements of formalism and excessive exaggeration in his productions of a Chekhov work and a contemporary Soviet play, The Iron Stream (Zheleznyi potok). He promised to devote far more preparation to each play, only planning two different stagings a year.88 Amateur stages run by directors trained by Meyerhold and those few Moscow venues with ties to the Meyerhold Theater were subject to particularly close scrutiny. And some authors' work, like that of Arbuzov, came under attack. Glebov, one of the most prolific critics of the amateur stage in the late 1930s, accused Arbuzov's work of containing "formalist elements" and presenting an overly symmetrical, "geometric" view of Soviet life.89
The anti-formalist campaign began before the political purges had made significant inroads into the artistic community. However, the beginning of the big purge trials in the summer of 1936 created an atmosphere in which aesthetic errors could easily be turned into political crimes.90 The trial of Zinoviev and Kamenev included the theater critic and magazine editor Richard Pikel as one of the defendants. He had once worked as Zinoviev's secretary and now was branded as a member of the "Moscow Terrorist Center." In particular, his treachery in promoting "Trotskyist plays" was brought up as evidence against him.91
In 1937, Kirshon and Afinogenov were caught up in the purges, both accused of being Trotskyist agents. A special meeting of Moscow playwrights in April of that year was largely devoted to denunciations of their work.92 Since Kirshon's plays in particular were very important to clubs, this posed yet another problem for circles faced with a diminishing repertoire. Those who followed the accusations closely needed to find substitutions. Those who did not were themselves at risk of accusations for "Trotskyist" sympathies. As more and more Soviet playwrights came under the sweep of the purges, including Ivan Mikitenko, Sergei Tretiakov, and Nikolai Erdman, amateur stages that had performed their work faced ostracism or worse.93
The political purges brought even greater scrutiny over club leadership. In late 1937, the trade union bureaucracy conducted a "cleansing" of club theater workers, focusing primarily on their professional qualifications. But political loyalties entered into the investigation as well. Those conducting the investigation complained that club theater directors did not seem to know about the anti-formalist campaign and its attendant political ramifications. They were not aware of the exposure of Kirshon and Afinogenov as part of a Trotskyist conspiracy and still staged works by the popular authors.94 In the course of the investigation, trade union officials determined that only fifty-one of the two thousand people investigated possessed the highest qualifications for their work. Almost four hundred people were removed from their jobs. Even after the inspection, concluded one official, artistic work in clubs continued to attract "alien and accidental people."95
A few club directors lost their jobs for overt political actions. One investigation determined that a small circle taking part in training courses designed for club theater leaders in 1937 sang anti-Soviet chastushki, defended Karl Radek, currently on trial as a spy, and denounced the show trials as an elaborate confabulation, using the term instsenirovka for a theatrical display concocted out of a variety of non-literary sources.96 The ouster of factory leaderships during the purges often caused cleansings of the entire administrative structure, including club leadership. At the Red Dawn telephone factory in Leningrad, the directorship was unmasked as "Trotskyists" in 1937; the club leader, perhaps reading the handwriting on the wall, "simply disappeared."97
The political purges reinforced amateur stages' already existing tendency to retreat to the safety of the classics to avoid contemporary works. However, one notable exception was a purge hit entitled The Confrontation (Ochnaia stavka), written by Lev Sheinin and the Brothers Tur (a pseudonym for Leonid Tubelskii and Piotr Ryzhei) in 1937. Amateur groups in Moscow and Leningrad staged it many times, and it was the single most popular play at the Moscow amateur competition in 1938.98 Sheinin was himself a criminal investigator who had helped the notorious state prosecutor, Andrei Vyshinskii, in a number of show trials.99 His play offered audiences and actors a simple explanation for the political upheaval shaking the nation. The main villains are two spies trained by the Gestapo. These unsavory characters are unmasked through the vigilance of ordinary Soviet citizens, who immediately report unusual events to the secret police. A wily peasant notices strange movements along the border and manages to foil an illegal crossing. A young student who has just arrived to Moscow is given someone else's letter by mistake and takes it straight to the police. The spies come dangerously close to success because they have help from Trotskyist groups within the country. But in the end, they are no match for the collective power of the Soviet citizenry. When one of the spies warns that a massive network is ready to infiltrate the Soviet Union, the star of the show, a Soviet secret police agent, laughs off the threat. "We have one hundred seventy million willing helpers."100
The political message of The Confrontation is transparent—potential enemies are everywhere. The foreign infiltrators know Russian perfectly. They recite poems by Alexander Blok before dispatching Soviet citizens to their death. In addition, they have plenty of help from evil forces within the nation. The informants who bring their suspicions to the secret police are always correct; in each instance, their information proves crucial in cornering the spies. As one model citizen tells his doubting wife, "I am not going [to inform] in order to be thanked. I am going because it is necessary."101 An American viewer who saw the play in Magnitogorsk commented, "They may catch some spies now, but it will take a generation to live down the fear and suspicion being created."102
#### Actors and Spectators
Reporters presented the achievements of amateur theaters to the public in a new way in the years of the Second Five-Year Plan. A decade earlier, groups were assessed for their collective performances, and individuals were rarely mentioned. Now directors, actors, and even set designers gave extensive interviews. Participants traced their humble beginnings, charted their progress from supporting to major parts, and shared their tips for learning roles. One club star shared how he was inspired by a glimpse of a fellow bus rider to develop the proper posture for his part; another recommended attending museums to gain the necessary background for a Shakespeare play.103 The journal Klub offered a venue for amateur theater workers to tell their stories, publishing short biographies and interviews. These highly selective portraits show amateur actors engaged in a self-education process, using theater as a path to a broadly defined kul'turnost'. Here they learned not only the classics of world drama and the contemporary stage; they were also exposed to lessons in diction and fashion. Trips to libraries and museums established these institutions as resources for the cultured individual.
Using the categories of the theorist Pierre Bourdieu, we can see how amateur theatrical training in the 1930s helped participants to accumulate valuable cultural capital.104 It gave them access to a state-promoted list of cultural references, teaching them the importance of Chekhov and Lope de Vega. Theatrical training also offered participants cues on proper manners, speech, and deportment. In this sense, taking part in amateur theater could cultivate a sense of cultural refinement. For those who pursued professional stage careers, amateur training also became an avenue for social advancement.
This sense of the theater circle as a kind of cultural accumulation process is especially apparent in the accounts of recent migrants to the city. The 1930s was a period of heightened rural-urban migration, as industrial expansion attracted peasants from the impoverished country-side.105 One recent female recruit to factory life, a worker at Moscow's Klara Tsetkin factory, used work in the drama circle to spark her interest in cultural life in general. Only a few months after running from her first audition in fear, she won a sizeable role. Gaining confidence from that success, she joined a music circle and began to subscribe to a newspaper. All of this led her to become a stalwart member of the Komsomol.106 One "young country boy" attested to the importance of the club in general and theater in particular to his cultural development. He was amazed at the fine appearance of other club members, whom he at first mistook for engineers and not other workers like himself. He quickly learned that to attend the club, one had to dress well (chisto odevat'sia). The drama circle taught him the importance of cultured, accurate speech. He was then inspired to go to the library to read "not simply for enjoyment, but to understand life." The theater circle took him on field trips to museums, planetariums, and professional stages. He eventually ended up as a union activist and tripled his starting salary.107 One Georgian member of the Paris Commune factory TRAM attested to the importance of theater work for his Russian-language skills. Before taking part his Russian had been weak, but the theater circle had taught him to express himself freely. Another non-Russian worker, a Tatar, used amateur theater as a method to organize his fellow countrymen at the Rusakov club.108
These tales of the civilizing mission of the theater stressed the use value of theatrical training in participants' daily lives, teaching them labor discipline, new views on social interaction, and vital life skills. "Work in the [theater] circle has taught me a better attitude toward my own growth and my productive work," opined one member of the Proletarian Club in Moscow. "Now I am a member of the Party, a brigadier, a labor organizer, and I am always looking to improve my qualifications." A machinist at the Kukhmisterov club in Moscow claimed to be inspired by the positive heroes that he portrayed. "I want to become just like them."109
In these narratives of self-improvement, it is as if the dreams of prerevolutionary activists in popular theater, who saw drama as a method to tame the uneducated narod, had finally come true. Amateur actors testified to the theater's ability to make them well-dressed, punctual, and responsible, all part of the prerevolutionary agenda. But they also show that the Soviet stage was a means for amateurs to learn how to "speak Bolshevik," to use Stephen Kotkin's phrase.110 They valued their work in amateur circles not only for its own sake or for the contribution that it made to their personal development as a cultured individuals. In addition, they stressed its significance for the labor productivity of the Soviet Union.
Successful amateur actors gained tangible rewards from their activities. More serious training programs opened up the chance to move on to the professional stage. Even if they stayed amateurs, they could win modest fame in their local work environment. The Moscow Proletarii Club hung pictures of amateur performers next to those of shock workers.111 A few "stars" of the amateur world found notoriety in national journals, such as the Kauchuk actress, Klava Soloveva, praised in an article in Klub for her spirited portrayal of Kate in Shakespeare's Taming of the Shrew. In her own celebrity interview, Soloveva was able to discuss her progress to the top of the amateur theatrical world and to reveal how the experience of working on such a demanding part had increased her skills as an actor.112 A year later, the Kauchuk factory sent Soloveva to study full-time at the Vakhtangov Theater School.113
Viewers gained a store of references to prerevolutionary culture by attending amateur performances. In addition, contemporary plays offered them instruction about appropriate manners in the restructured Soviet workplace. They showed, for example, that responsible women workers were now a force to be acknowledged. In a marked contrast to agitprop works, plays in the Second Five-Year Plan gave women important roles. Girls of Our Country pronounces this fact already in its title. Miraculous Alloy features aspiring women scientists. Arbuzov's The Long Road places a woman in charge of shock brigade building the Moscow subway. Konstantin Trenev's Liubov Iarovaia, about a woman commander in the Civil War, was written in the 1920s but only became a staple of amateur stages in the following decade.
Amateur performers had a very different relationship to their audiences than they had in the 1920s. While plays still had an obvious political agenda, the propaganda was less aggressive than during the First Five-Year Plan. Amateur productions tried to give viewers positive role models and to instill common cultural standards, but the audience's status as observers was not assaulted. No one tried to extract funds or shock work participants from the viewers seated in newly designed stately auditoriums. The raised proscenium stages gave the audience anonymity—the actors were on display, not the viewer.
The Soviet press presented the changes in amateur stages as audience-driven; the more sophisticated tastes of viewers had moved amateur stages to abandon simple agitational works and disjointed performances. It is difficult to confirm or deny this assertion, since the voices of the viewers are hard to find. The scanty sources from the 1920s, such as audience response forms and interviews by worker correspondents, disappeared in the next decade. Judging only from the fact that the biggest clubs filled their large halls, these performances certainly found an audience. Yet, from the very structure of these events, we can also conclude that amateur theatricals served a different function in the lives of viewers than they had earlier. Performances were now a special event, staged infrequently by well-coached actors. The rowdy evenings of the 1920s, where actors were barely differentiated from the audience, were a thing of the past.
#### Effacing the Amateur
By the late 1930s, the concept of "samodeiatel'nost'" was as problematic as it had been in the first years of Soviet power. Then, leaders needed to extract the idea of the Soviet amateur from notions of worker self-determination and autonomy put forward by factions of the trade union movement, the Proletkult, and the Workers' Opposition. In the late 1930s, samodeiatel'nost' needed to be rescued from the theorists of the 1920s, who had proposed a different and separate path for amateur theater. Critics initially rejected these ideas as "leftist" and "constructivist" notions. By 1937, the political purges brought ominous new elements to the charges; the ideas were branded as "harmful" and sometimes even "Trotskyist."114 One of the most eloquent advocates of those earlier theories, Adrian Piotrovskii, was arrested as a Fascist agent in 1937.115
The spontaneous and imperfect aspects of amateur theaters were effaced in the verbal accounts of their achievements in the late 1930s. Samodeiatel'nost' now came to stand for the creative potential of the Soviet people, who, with proper diligence and training, could aspire to the artistic standards of professionals. At least some participants believed the word samodeiatel'nost' no longer fit their activities. They were actors striving for the highest artistic standards who just happened to have day jobs. Not coincidentally, the term "narodnyi teatr," once rejected as a defunct relict of prerevolutionary times, began to reemerge in discussions about the amateur stage.116
A 1936 photograph of Adrian Piotrovskii. E. S. Dobin, ed., Adrian Piotrovskii: Teatr, kino, zhizn (Leningrad: Iskusstvo, 1969).
That amateurs had abandoned any claims to a unique approach to theatrical performance was graphically illustrated in the Moscow competition of amateur theaters held in the spring and summer of 1938. Pravda celebrated the sheer numbers of people drawn to the festivities: "At the Aviakhim factory, 395 new participants in the amateur arts were included.... No one knew there were so many talented people until now."117 The first prize went to the Gorbunov factory theater's performance of Maxim Gorky's The Philistines (Meshchane). In a long article assessing the results of the festival, the theater critic Leonid Subbotin presented this group as a stellar example of the path amateur theaters had followed in the 1930s. It began its existence as a TRAM circle in 1932 and continued in that guise until 1935, when its work was finally put on the right track by two directors from the Theater of Revolution, who began training the group in Stanislavsky's methods. "Here we see the typical biography of a collective decisively turning in the direction of serious study and education," opined Subbotin. This transformation process had yielded outstanding results. Its performance of Gorky's play was "of unusual quality for an amateur group."118
Subbotin's lengthy account of this circle's accomplishments offered a vivid example of the aesthetic values now applied to amateur circles. The collective "presented itself as a lively, well-coordinated ensemble," he determined. "The performance was marked by the collective's sincere, moving relationship to theatrical art."119 In others words, any trace of "amateurism" had disappeared. Subbotin was not the only commentator to erase any references to the troubling imperfections of amateurism. In his review of the festival, the professional director Il'ia Su-dakov spelled out what he believed to be the task of directors of amateur theaters, namely "to teach people to discover the deep, authentic truth in art. Then they will work with a real fire in their spirit, with deeply felt creativity, and without any false elements or tricks."120
The 1938 festival also reveals the declining role of amateur theater within the Stalinist spectacle state. Although the number of amateur stages in general continued to rise, and the festival itself was a much larger event than the 1932 gathering, it hardly generated any publicity in the press and was covered primarily in specialized journals. Grander spectacles, including polar expeditions, technical feats, and the purge trials themselves, offered stiff competition. Stalinist culture, in Richard Stites' words, was "the greatest show on earth," attracting consumers with its range and diversity—from restaurants with stylish jazz bands, popular movies at accessible prices, and newly available radio entertainment that brought concert music, sports coverage, adventure stories, and political trials into people's homes.121 Amateur theater occupied only a small niche in this grandiose display.
At the end of his famous essay, "The Work of Art in the Age of Mechanical Reproduction," Walter Benjamin argues that the essence of Fascism was its aestheticization of politics. "Communism has responded by politicizing art," he concludes.122 Although his assessment of Fascism has been very influential, I believe that Benjamin's observations about Communism, at least the Soviet variety, require revision. Soviet Communists politicized art starting with the revolution; this tendency was hardly called forth by the rise of Fascism. And while the regime's approach to political art changed in the 1930s, if anything it became more "aesthetic." The Soviet arts establishment wanted to prove that their cultural products, including amateur performances, deserved to be considered as serious art.
To this end, cultural agencies overseeing amateur stages engaged in an extensive effort to reclaim amateur performance as aesthetic drama. They revamped repertoires, improved physical facilities, and brought in professional help. This process offered real advantages to viewers, who had a chance to see the same plays current on more expensive professional stages at a location probably more convenient to their homes. Although the acting might not have been as polished, the experience was similar. The performance was a separate event, marked off from other activities. "Here too, as in the central stages of the capital, you can hear the rustle of new theater programs at seven-thirty," wrote one commentator on amateur theatrical life in Moscow.123 The actors also had a chance to benefit from the change. Many offered written testimonials detailing the importance of rigorous training programs in their lives. They knew that they were learning special skills; they were not just propagandists who could be pulled directly from the production line.
Yet, in this era of purges, it is also possible to see the aestheticization of amateur theater as a process of cleansings. Amateur stages lost troublesome leaders, who represented what were now considered to be dangerous artistic trends. They severed contact to earlier ideas of amateurism, which had seen non-professionals as the main source of vitality for all of Soviet culture. They no longer generated their own works, a process that had provided them a chance, however modest, to reflect on the problems of their daily lives. In other words, they were left with very little space in which to "do it themselves."
* * *
1. Vladimir Papernyi, Kul'tura "dva" (Ann Arbor: Ardis, 1985).
2. Ibid., 71, 209, 101; see also James von Geldern, "Cultural and Social Geography in the Mass Culture of the 1930s," in Stephen White, ed., New Directions in Soviet History (Cambridge: Cambridge University Press, 1990), 66–67.
3. Von Geldern, "Cultural and Social Geography," 71.
4. D. Marchenko, "Ocherednye zadachi khudozhestvennoi samodeiatel'nosti," Klub 7 (1933): 48–49.
5. G. Goncharov, "Klub KOR," Klub 19 (1935): 53; "Nasushchnye voprosy khudozhestvennoi samodeiatel'nosti," Klub 14 (1937): 48.
6. "Novye shkoly, kino, kluby i dvortsy," Pravda 19 March 1933; "Po sssr" Pravda 18 May 1932.
7. V. G., "Dvorets kul'tury rabochikh Leninskoi slobody," Pravda 29 September 1933. On the competitions, see V. Khazanova, Klubnaia zhizn' i arkhitektura kluba, v. 1: 1917–1932 (Moscow: Rossiiskii Institut Iskusstvoznaniia, 1994), 127–37.
8. On the simultaneous use of space, see Lewis Siegelbaum, "The Shaping of Workers' Leisure," unpublished ms. On the expansion of club amenities, see Khazanova, Klubnaia zhizn', v. 2: 109.
9. Khazanova, Klubnaia zhizn', v. 1: 151, v. 2: 9; E. Gabrilovich, "Dvadtsat' minut vos'-mogo," Pravda 6 January 1935.
10. Khazanova, Klubnaia zhizn', v. 2: 11–25. On the importance of the façade in 1930s architecture, see Papemyi, Kul'tura "dva," 69–71.
11. On the new language surrounding clubs, see Khazanova, Klubnaia zhizn', v. 2: 23–31, 137–39. On clubs as a surrogate home, see Leon Feuchtwanger, Moskau 1937: Eine Reisebericht for meine Freunde (1937; rpr. Berlin: Aufbau Verlag, 1993), 23.
12. See, for example, the untitled 1933 report on a variety of Moscow clubs, Tsentral'nyi munitsipal'nyi arkhiv Moskvy (TSMAM), f. 718 (TSMAM), op. 8, d. 45, ll. 92–95.
13. "Teatr kluba 'Kauchuk' pod rukovodstvom vakhtangovtsev," Klub 6 (1936): 33–35. On the club design, see S. Frederick Starr, Melnikov: Solo Architect in a Mass Society (Princeton: Princeton University Press, 1978), 140–42.
14. Bor. Levin, "V rabochem klube," Pravda 30 October 1934.
15. "Za kul'turnyi klub," Krasnaia gazeta 6 October 1934.
16. "Na zadvorkakh," KP 23 August 1934; see also the description of rehearsal space in the Kalinin factory, "Beseda s rukovoditeliami khudozhestvennykh samodeiatel'nykh kruzhkov," GARF, f. 5451, op. 19, d. 422, l. 3 ob.
17. A. A. Nokhin, "Pochemu kluby na zamke?," Pravda 15 October 1936.
18. "Takoi klub nam ne nuzhen," Krasnaia gazeta 6 October 1934.
19. "Dom kul'tury ili torgovyi dom?," KP 15 October 1934; "Klub—mesto kul'turnogo otdykha," KP 15 October 1934. See also V. Kostin, "Vecherami v klubakh," KP 28 December 1936.
20. N. Engel', "Tvorcheskaia konferentsiia leningradskoi samodeiatel'nosti," Klub 3 (1934): 55–56.
21. S. Room, "Teatral'naia samodeiatel'nost' na novom etape," RiT 18 (1936): 5.
22. Engel', "Tvorcheskaia konferentsiia leningradskoi samodeiatel'nosti," 55–57; D. M., "Konferentsiia zritelei teatra VTSSPS," Pravda 23 October 1933; "Studiia Malogo teatra na zavode 'Serp i molot,' " Pravda 16 October 1935.
23. "Vstrecha tvorcheskogo aktiva teatrov profsoiuzov," 14 August 1934, GARF, f. 5451, op. 18, d. 509, l. 12.
24. V. Solov'ev, "Akter—vyrazitel' idei p'esy," Teatr i dramaturgiia 5 (1933): 30–32.
25. D. Zaslavskii, " 'Chudesnyi splav,' " Pravda 15 June 1934.
26. Ia. Moskovoi, "Na poroge odinnadtsatogo," Klub 13 (1933): 44–45; Malov, "Uchimsia u MKhATA," Klub 21/22 (1934): 43; M. Berliant, "Desiat' let," Klub 2 (1935): 49; Norris Houghton, Moscow Rehearsals: The Golden Age of Soviet Theatre (New York: Harcourt Brace, 1936), 32. For a recent assessment of the influence of the Moscow Art Theater, see A. P. Shulgin, "Samodeiatel'nyi teatr v gorode," in S. Iu. Rumiantsev et al., eds., Samodeiatel'noe khudozhestvennoe tvorchestvo v SSSR, v. 2 (Moscow: Gosudarstvennyi institut iskusstvoznaniia, 1995), 169–70.
27. "Rabochii teatr zavoda 'Kauchuk,' " Klub 2 (1932): 56–57; S. Persov, "Samodeiatel'noe iskusstvo v klube zavoda "Kauchuk,' " Klub 1 (1933): 39–40; V. Kuza, "Teatr 'Kauchuk' pod rukovodstvom Vakhtangovtsev," Klub 6 (1936): 33–35, quotation 34.
28. "Mkhatovtsy v teatral'noi samodeiatel'nosti," Kul'turnaia rabota profsoiuzov 14 (1938): 46.
29. V. Derzhavin, "Teamasterskaia kluba im. Iakovleva," Klub 19 (1934): 43; "Rabochii teatr Volodarskogo Doma Kul'tury," Klub 2 (1934): 55.
30. See, for example, Nikolai Okhlopkov, "Samodeiatel'noe iskusstvo," Pravda 3 November 1935; "Mastera iskusstv o khudozhestvennoi samodeiatel'nosti," Klub 20 (1937): 38–39; "Mkhatovtsy v teatral'noi samodeiatel'nosti," Kul'turnaia rabota profsoiuzov 14 (1938): 46.
31. Andre Van Gysegham, Theatre in Soviet Russia (London: Faber and Faber, 1943), 150–53.
32. Juri Jelagin, The Taming of the Arts, trans. Nicholas Wreden (New York: Dutton, 1951), 214; Margaret Wettlin, Fifty Russian Winters (New York: John Wiley, 1994), 103, 119.
33. "Beseda s rukovoditeliami khudozhestvennykh samodeiatel'nykh kruzhkov," 5 April 1935, GARF, f. 5451, op. 19, d. 422, ll. 2–3.
34. "Stenogramma soveshchaniia pri klubnoi inspektsii VTSSPs," GARF, f. 5451, op. 19, d. 4Q4, ll. 1–3.
35. "Protokoly," tsmam, f. 718, op. 8, d. 49, l. 44.
36. See "Stenogramma soveshchaniia pri klubnoi inspektsii VTSSPS," GARF, f. 5451, op. 19, d. 404, l. 54.
37. L. Subbotin, "Samodeiatel'nyi teatr," Teatr 6 (1938): 119.
38. I. Blinkov, "Pochemu skuchno v klube," Pravda 19 July 1935. See also Veprinskii, "Zrelishchnaia samodeiatel'nost' na moskovskom smotre 1933 goda," Klub 15 (1933): 38–39, and Kobra, "Za kulisami kluba," Literaturnyi Leningrad 7 (8 February 1936): 4.
39. Of course, there were exceptions to these broad trends. The Klub im. Iakovleva, for example, performed Itogi serdtsa, written by club leader V. P. Derzhavin in 1934. See V. Derzhavin, "Teamasterskaia kluba im. Iakovleva," Klub 19 (1934), 43.
40. "Tvorcheskaia konferentsiia po samodeiatel'nosti," RiT 35 (1933): 17; "Stenogramma i rezoliutsii tret'ego plenuma Moskovskogo gorodskogo soveta professional'nykh soiuzov," March 1933, TSMAM, f. 718, op. 8, d. 1, ll. 90–91; "Vstrechi tvorcheskogo aktiva teatrov profsoiuzov," 14 August 1934, GARF, f. 5451, op. 18, d. 509; "Stenogramma soveshchaniia pri klubnoi inspektsii VTSSPS," 1935, GARF, f. 5451, op. 19, d. 404, ll. 1–44.
41. Klub began publication in 1933; it was superceded by Kul'turnaia rabota profsoiuzov in 1938.
42. "Tekhnikum teatra VTSSPS," Pravda 21 May 1934.
43. A. B. Viner, "Teatr im. MOSPS i samodeiatel'noe iskusstvo," Klub 6 (1933): 50–51.
44. A. Iordanskii, "Samodeiatel'noe iskusstvo po soiuzy rabotnikov kooperatsii i gostorgovli," Klub 3 (1933): 44–45.
45. Iak. Mostovoi, "Na poroge odinnadtsatogo," Klub 13 (1933): 43. This issue of the journal Klub is filled with similar tales of restructuring.
46. "Otchet po khudozhestvennoi rabote mosgorkoma soiuza rezino-khimicheskoi promyshlennosti s 1-go ianvaria po 14 iiunia 1933 g.," tsmam, f. 718, op 8, d. 49, ll. 58–59.
47. A. Berlin, "Tvorcheskoe sorevnovanie," Klub 6 (1933): 46; see also reports at a national trade union club conference, where local leaders insist that their agitprop brigades perform "big" works by important Soviet playwrights like Kirshon and Bill-Belotserkovskii, "Stenogramma soveshchaniia pri klubnoi inspektsii VTSSPS," GARF, f. 5451, op. 19, d. 404, l. 27.
48. D. Marchenko, "Khudozhestvennaia samodeiatel'nost' millionov," Teatr i dramaturgiia 5 (1935): 15.
49. A. Kasatkina, "Problemy klubnogo repertuara," Teatr i dramaturgiia 8 (1933): 52–55. This article was sharply attacked a few months later for ignoring the principles of socialist realism; see N. Engel', "Tvorcheskaia konferentsiia leningradskoi samodeiatel'nosti," Klub 3 (1934): 58.
50. "Khronika khudozhestvennoi samodeiatel'nosti," Klub 8 (1935): 63.
51. "Otchet po khudozhestvennoi rabote," TSMAM, f. 718, op. 8, d. 49, l. 60; Klub 10 (1932): 48.
52. I. Mikitenko, "Plachu dolg," in Devushki nashei strany (Leningrad: Gosudarstvennyi teatr dramy, 1933), 10.
53. M. Berliant, "Teatral'naia samodeiatel'nost' Leningrada," Klub 19 (1933): 54.
54. L. Tamashin, Vladimir Kirshon: Ocherk tvorchestva (Moscow: Sovetskii pisatel', 1965), 183; N. Gorchakov, "Chudesnyi splav v moskovskikh teatrakh," Kolkhoznyi teatr 2 (1935): 15.
55. Vladimir Kirshon, Chudesnyi splav: Komediia v chetyrekh aktakh (Moscow: Iskusstvo, 1956), 93.
56. "Teatral'naia dekada," Klub 5 (1936): 47.
57. On the significance of the play, see Cynthia A. Ruder, Making History for Stalin: The Story of the Belomor Canal (Gainesville: University Press of Florida, 1998), 155–73, and Natalia Kuziakina, Theatre in the Solovki Prison Camp (Luxembourg: Harwood Academic Publishers, 1995), 121, 138.
58. V. N. Aizenshtadt, Sovetskii samodeiatel'nyi teatr (Kharkov: Khar'kovskii gosudarstvennyi institut kul'tury, 1983), 54.
59. Karen Petrone, Life Has Become More Joyous, Comrades: Celebrations in the Time of Stalin (Bloomington: Indiana University Press, forthcoming) ch. 4.
60. L. Volkov, "Teatry v pushkinskie dni," Pravda 1 February 1937; "Pushkinskie dni v Moskovskikh klubakh," Klub 2 (1937): 46–48.
61. M. Berliant, "Dramaturgiia A. M. Gor'kogo v samodeiatel'nom teatre," Klub 12 (1937): 30–34; S. Valerin, "Gor'kovskaia dekada samodeiatel'nykh teatrov," Klub 15 (1937): 46–48.
62. M. Troianskii and P. Valentinskii, "Sed'maia Olimpiada khudozhestvennoi samodeiatel'nosti," RiT 14 (1933): 20; V. Bliumenfel'd, "Klassiki v samodeiatel'nom teatre," RiT 23 (1933): 6; Lesli, "Dramkruzhok kluba im. Kukhmisterova," Klub 19 (1934): 39.
63. V. Bliumenfel'd, "Klassiki v samodeiatel'nom teatre," RiT 23 (1933): 6.
64. M. Berliant, "Obzor repertuara moskovskikh dramaticheskikh kruzhkov," Klub 9 (1934): 36.
65. "Dramkruzhok kluba im. Kukhmisterova," Klub 19 (1934): 39.
66. M. Berliant, "Itogi teatral'nogo soveshchaniia," Klub 4 (1936): 52.
67. "K 50-letiiu so dnia smerti Ostrovskogo," Literaturnyi Leningrad 27 (12 June 1936): 4.
68. A. Glebov, "Ostrovskii v teatral'noi samodeiatel'nosti," Teatr 4 (1937): 116. His figures came from the Dom samodeiatel'nogo iskusstva, which primarily monitored amateur work in the countryside. Glebov opens by arguing that Ostrovsky performances had fallen off in urban amateur theaters, but the rest of his article contradicts this.
69. Ibid., 118.
70. "Teatr Kauchuk pod rukovodstvom vakhtangovtsev," Klub 8 (1934): 34.
71. The Seventh Leningrad Olympiad, for example, opened with a gathering in the main square of the city and a mass performance of eight thousand circle members (Pravda 9 June 1933).
72. Gomello, "O nekotorykh tvorcheskikh itogakh VII leningradskoi olimpiady," Klub 10 (1933): 48; V. Bliumenfel'd, "Klassiki v samodeiatel'nom teatre," RiT, 23 (1933): 6.
73. "Protokoly," TSMAM, f. 718, op. 8, d. 49, l. 47.
74. A. Berlin, "Vos'maia leningradskaia olimpiada," Klub 17 (1934): 34; M. Veprinskii, "Teatral'naia samodeiatel'nost' na moskovskom smotre," Klub 5 (1934): 47.
75. M. Veprinskii, "Teatral'naia samodeiatel'nost' na moskovskom smotre," 47.
76. A. Liubosh, "Smotr leningradskoi samodeiatel'nosti," Iskusstvo i zhizn' 2 (1938): 31.
77. L. Subbotin, "Nekotorye vyvody iz smotra teatral'noi samodeiatel'nosti," Kul'turnaia rabota profsoiuzov 12 (1938): 77.
78. "Stenogramma soveshchaniia o minimume znanii dlia zavklubov ot 7/XII/1934," GARF, f. 2306, op. 39, d. 5, l. 39.
79. There is a vast literature on the purges of the 1930s. For a recent study based on Communist Party archival sources, see J. Arch Getty and Oleg V. Naumov, The Road to Terror: Stalin and the Self-Destruction of the Bolsheviks (New Haven: Yale University Press, 1999). Fitzpatrick's Everyday Stalinism address the purges in the artistic community, Everyday Stalinism, 190–217.
80. Aizenshtadt, Sovetskii samodeiatel'nyi teatr, 44.
81. Leonid Maksimenkov, Sumbur vmesto muzyki: Stalinskaia kul'turnaia revoliutsiia, 1936–1938 (Moscow: Iuridicheskaia kniga, 1997), 88–112.
82. "Sumbur vmesto muzyki," Pravda 28 January 1936. On this episode in Soviet culture, see Sheila Fitzpatrick, "The Lady Macbeth Affair: Shostakovich and the Soviet Puritans," in idem, The Cultural Front: Power and Culture in Revolutionary Russia (Ithaca: Cornell University Press, 1992), 183–215; Evgenii Gromov, Stalin: Vlast' i iskusstvo (Moscow: Respublika, 1998), 245–53; Maksimenkov, Sumbur vmesto muzyki.
83. "III Plenum TsK Rabis," Sovetskii teatr 3 (1936): 3.
84. "Protiv formalizma i naturalizma," Teatr i dramaturgiia 4 (1936): 196, 198; Meyerhold's contribution is on pages 207–10.
85. "Baletnaia fal'sh'," Pravda 8 February 1936.
86. S. Room, "Teatral'naia samodeiatel'nost' na novom etape," RiT 18 (1936): 5.
87. "O formalizme, bezgramotnosti i khalture v teatral'noi samodeiatel'nosti," Klub 9 (1936): 40–42, quotation 42. See also M. Berliant, "Shekspir na rabochei stsene," Narodnoe tvorchestvo 1 (1937): 36.
88. "Rabotniki khudozhestvennoi samodeiatel'nosti ob urokakh tvorcheskoi diskussii," Klub 13 (1936): 38.
89. A. Glebov, "Samodeiatel'nyi teatr i ego dramaturgiia," Teatr i dramaturgiia 6 (1936): 314.
90. Maksimenkov, Sumbur vmesto muzyki, 10–11.
91. "Prigovor po delu trotskistsko-zinov'evskogo terroristicheskogo tsentra," Teatr i dramaturgiia 8 (1936): 449–50; "Zorche vzgliad, tverzhe ruka!," ibid., 453–55. On Pikel's background, see Fitzpatrick, Everyday Stalinism, 197–98.
92. I. Lezhnev, "Sobranie moskovskikh dramaturgov," Pravda 29 April 1937.
93. On the effects of the purges in the theatrical world, see Nikolai A. Gorchakov, The Theater in Soviet Russia (New York: Columbia University Press, 1957), 364–65. For complaints about provincial circles that still staged plays by the "Trotskyists" Kirshon and Afinogenov in 1937, see A. Berlin, "Korennye nedostatki rukovodstva khudozhestvennoi samodeiatel'nost'iu," Klub 12 (1937): 25–27.
94. Berlin, "Korennye nedostatki rukovodstva," 24–25.
95. Ibid. For the results in Leningrad alone, see A. Liubotsh, "Deklaratsii i praktika," RiT 5 (1937): 58.
96. Glusov and Smirnov, "Dokladnaia zapiska. Sekretariu VTSSPS tov. Veinbergu, G. D.," GARF, f. 5451, op. 21, d. 135, ll. 23–26.
97. G. Mikhailov, "Zhivaia rabota zamerla," Klub 18 (1937): 63.
98. L. Subbotin, "Nekotorye vyvody iz smotra teatral'noi samodeiatel'nosti," Kul'turnaia rabota profsoiuzov 12 (1938): 77.
99. Arkady Vaksberg, The Prosecutor and the Prey: Vyshinsky and the 1930s Moscow Show Trials, trans. Jan Butler (London: Weidenfeld and Nicholson, 1990), 66, 68–69, 74–75.
100. Brat'ia Tur and L. Sheinin, Ochnaia stavka (Moscow: Iskusstvo, 1938), 79. See also Sheila Fitzpatrick's discussion of this play, Everyday Stalinism, 203.
101. Ibid., 60.
102. John Scott, Behind the Urals: An American Worker in Russia's City of Steel, ed. Stephen Kotkin (Bloomington: Indiana University Press, 1989), 203.
103. "Profsoiuznye kluby dolzhny pomogat' rostu rabotnits," Klub 5 (1935): 34–35; "Kak my ovladevali obrazami Shekspira," Klub 6 (1936): 39–41.
104. Pierre Bourdieu, Distinction: A Social Critique of the Judgement of Taste, trans. Richard Nice (Cambridge: Harvard University Press, 1984), 53–54, 114–15.
105. For migration to Moscow, see David L. Hoffmann, Peasant Metropolis: Social Identities in Moscow, 1929–1941 (Ithaca: Cornell University Press, 1994), 32–72.
106. N. Filippova, "V klube rastem my i nashi deti," Klub 5 (1934): 30–31.
107. V. Nikiforov, "My vyrosli v klube," Klub 23/ 24 (1934): 3.
108. "Chto nam daet klub," Klub 23/24 (1934): 7.
109. Malov, "Uchimsia u MKhATA," Klub 21/22 (1934): 43.
110. Stephen Kotkin, Magnetic Mountain: Stalinism as Civilization (Berkeley: University of California Press, 1995), 198–237.
111. M. Veprinskii, "Teatral'naia samodeiatel'nost' fabrik i zavodov," Kolkhoznyi teatr 7/8 (1936): 54.
112. M. Berliant, "'Ukroshchenie stroptivoi' v klube zavoda 'Kauchuk,' " Klub 6 (1936): 32; "Kak my ovladevali obrazami Shekspira," ibid., 39, 42.
113. "Rabotnitsa na stsene teatra," Kul'turnaia rabota profsoiuzov 3/4 (1938): 43–44.
114. Gr. Avlov, "Rost khudozhestvennoi samodeiatel'nosti," RiT 11 (1937): 52. In this remarkable article, Avlov does not once mention his own contribution to the ideas of the 1920s. See also L. Subbotin, "Samodeiatel'nyi teatr," Teatr 6 (1938): 114.
115. Untitled editorial, RiT 8 (1937): 1.
116. On the rise of "folk art" (narodnoe iskusstvo) in the 1930s, see Frank J. Miller, Folklore for Stalin: Russian Folklore and Pseudofolklore of the Stalin Era (Armonk, N.Y.: M. E Sharpe, 1990), and Susanna Lockwood Smith, "Soviet Arts Policy, Folk Music, and National Identity: The Piatnitskii State Russian Folk Choir, 1927–1945," (Ph.D. Dissertation, University of Minnesota, 1997), esp. 130–36. For its application to theater, see "Teatr narodnogo tvorch-estva gotovitsia k otkrytiiu," Pravda 12 March 1936.
117. "Smotr khudozhestvennoi samodeiatel'nosti v Moskve," Pravda 19 April 1938.
118. L. Subbotin, "Nekotorye vyvody iz smotra teatral'noi samodeiatel'nosti," Kul'tur-naia rabota profsoiuzov 12 (1938): 73–74.
119. Ibid., 73.
120. I. Sudakov, "Klubnyi spektakl'," VM 25 May 1938.
121. Richard Stites, Russian Popular Culture: Entertainment and Society since 1900 (Cambridge: Cambridge University Press, 1992), 64–97, quotation 94.
122. Walter Benjamin, Das Kunstwerk im Zeitalter seiner technischen Reproduzierbarkeit (Frankfurt am Main: Suhrkamp, 1972), 51. See Linda Schulte-Sasse, Entertaining the Third Reich (Durham, N.C.: Duke University Press, 1996), 17, for Benjamin's importance in the analysis of fascist culture.
123. E. Gabrilovich, "20 minut vos'mogo," Pravda 1 January 1935.
## Conclusion
IN THE 1927 film The House on Trubnaia Square (Dom na Trubnoi), a young peasant woman named Praskovia makes her way to Moscow. Instead of finding good employment, she ends up working as a maid for a couple quite contented with NEP. A trade union organizer discovers her and signs her up to join a union, also encouraging her to come to a theater performance at the local club. This marks the young woman's entry into urban public life. Even though the performance space is shabby and viewers sit on hard benches, club members flock to see the play, Romain Rolland's The Taking of the Bastille. Just about everything that could go wrong does. At first there are no wigs to complete the period costumes. The event starts very late because the main actor shows up drunk. Nonetheless, the crowd is riveted by the play. Praskovia is so engaged that when the hero is killed by an evil general, she jumps onto the stage and urges the onlookers to continue the revolution. Although the actors and the club director are horrified, the audience loves the unexpected ending and greets her with cheers. It is Praskovia's first happy moment in Moscow.
The hit film Volga, Volga, released in 1938, shows the amateur arts as the main occupation of the Soviet citizenry. It depicts the residents of a small town passionately engaged in artistic activities, complete with singing waiters, dancing mechanics, and a classical orchestra led by an accountant. The star of the film is once again a young woman, the postal carrier Strelka, who has composed a song about the beauty of life along the Volga under Soviet power. When a call comes from Moscow for the town to send a representative to a national Olympiad of amateur arts, Strelka's folk music group enters into a madcap competition with the local orchestra to be the first to arrive. The two groups eventually combine forces, writing down and orchestrating Strelka's creation. When native talent is combined with classical training, the result is a song that wins first prize.
The heroines in these contrasting films offer very different images of the amateur in Soviet society. In The House on Trubnaia Square, the victimized maid becomes an actor, and then an activist, by taking part in amateur theatricals. In the process, she challenges standard guidelines for artistic creation. Rebelling against the club director, Praskovia's entrance on stage turns a conventional performance into a mass activity, changing its form and meaning. The film celebrates how do-it-yourself theater, open to all, facilitates a sense of local community. In Volga, Volga, Strelka's many talents are not restricted to the confines of a club. She dances, sings arias, and declaims poetry in public for all to see. Her main occupation is not delivering the mail; instead, she devotes herself to revealing the innate talents of her fellow citizens. But Strelka, unlike Praskovia, is seeking national recognition of her gifts. She wants to meet, not question, prevailing artistic standards. In order for that to happen, she must work together with conventionally trained artists before she can perform on the national stage.1
These two films reveal how the meaning of amateurism changed from the 1920s to the 1930s. In the early Soviet period, amateur performances were open to all comers. Directors complained that their participants were constantly changing; it was almost impossible to stage a performance with the same group that had been in rehearsals from the beginning. There were no set training methods; indeed, there might have been little training at all. Praskovia goes onstage because she wants to act out the revolution, not follow a script. But by the twentieth anniversary of the revolution, the makeshift origins of Soviet amateur theater had been erased. Amateur actors now put in long hours in preparation under the direct or indirect supervision of serious artists. Strelka's long trip to Moscow along the Volga is a continual rehearsal session. Her work is subjected to tougher scrutiny because it is put on display to represent the talents of the nation.
Writing on the significance of amateur theater in the mid-1930s, Bertolt Brecht determined that one needed to distinguish between amateurs and dilettantes. "A dilettante is someone who copies professionals. An amateur must find his own art."2 Brecht's statement evokes the principles of early Soviet writers on amateur theater, such as Adrian Piotrovskii, who believed that amateurs had something new to bring to theatrical art. Early Soviet theorists, however, went further than Brecht, insisting that amateur art would serve as a source of inspiration and renewal for all of Soviet theater. But by the late 1930s, this concept of amateurism had been discredited. Amateurs were only valued insofar as they followed the lead of professionals and provided a credible surrogate for professional performances.
This ideal of the amateur as proto-professional was not entirely an invention of the Stalin era. Key elements were already articulated in the second half of NEP, when the trade union cultural leadership declared the participatory methods of the theater of small forms an artistic and political failure. At the 1927 Agitprop conference on theater, Party and union bureaucrats urged professionals to come to the aid of amateurs. Already at this point, the ideas of Piotrovskii and his colleagues were denounced by some as dangerous nonsense that was threatening the cultural advancement of the nation. Trade union leaders and a segment of the viewing public demanded a more tightly controlled repertoire, more of the classics, more lessons, and more skill. They wanted plays instead of agitational sketches, heroes instead of stereotypes. Key elements of Stalinist culture were set even before the massive upheaval of the First Five-Year Plan or the forceful imposition of state agencies in the early 1930s.3 Nonetheless, the ideal of the amateur in the 1930s was much narrower than the version proposed at the 1927 conference. A decade later, amateurs had a very limited repertoire at their disposal. And while their training had improved, it was also homogenized. The Stalinist version of Stanislavsky's approach, which one commentator called the "accepted laws of stage work," dominated both the professional and amateur stage.4
Stalinist cultural politics contributed mightily to the narrowing of the cultural horizons of the amateur. Even before the anti-formalist campaign was launched in 1936, the Soviet theatrical establishment had rejected almost all repertoire from Western Europe and the United States. Through any number of official anniversaries and celebrations, the government promoted a limited classical heritage of the great prerevolutionary playwrights and those foreign authors who (at least in theory) represented capitalism in its revolutionary stage. With the founding of the National Committee of the Arts, control of repertoire grew even tighter. In addition, critics turned against all forms of presentation they considered unrealistic, "formalist," or strange. There were no more surprise endings in amateur theaters. Even surprise stagings were extremely rare.
These constraints on amateur theater in the 1930s were not set only by the state. The improvised theater of small forms, which was itself in part imposed by cultural agencies, had its enemies from the outset. Although small forms offered many avenues for group participation, they were based on the principle that leisure time had to be made useful and serve as a conduit for political education. The relentless lessons about state and local policies proved tiring to audiences who were either uninterested or, as one viewer remarked, "could read newspapers themselves." Therefore, we can imagine that a sizeable segment of the viewing public was eager to see Ostrovsky's The Storm, with its compelling story of love, betrayal, and suicide, instead of Face to Production, compiled from Party speeches and official documents.
The early model of Soviet amateur theater, emerging from the participatory theater of small forms, realized one of the central ideas of the theatrical avant-garde, who attempted to dissolve the lines between performers and the passive viewing audience.5 Meyerhold's significant involvement in club theaters in Moscow shows that avant-gardists found the amateur stage a fruitful forum for experimentation. Their belief that theater should activate the audience was fulfilled to its tall nightmarish potential by agitprop brigades during the First Five-Year Plan. Then the audience had little choice but to take part in the coercive project of participatory theater, facing shaming rituals or censure if they refused. The extreme methods used to encourage audience participation discredited any merits of this approach. Although isolated critics during the Second Five-Year Plan lamented that amateur stages no longer addressed themselves to local issues, no one called for a rem to the Soviet-style charivari. The amateur stage that emerged after the Stalin revolution was in part created by the success, and then rapid rejection, of such repressive methods of audience engagement.
In a contentious essay on the history of Soviet art, Boris Groys has argued that avant-garde artists were not the innocent victims of Stalinism, as they have often been described in Western scholarship. Instead, he insists that socialist realism was in many ways a realization of the avant-garde's vision of organizing society into monolithic forms, using art as a method to transform reality.6 These conclusions, based on a narrow sampling of artistic manifestos by visual artists and writers, do not hold true for theater. Certainly, it is false to see the theatrical avant-garde as innocent victims, but it is difficult to see them as victors either. Their visions were not realized in the theater of the 1930s. The lines between actors and audience, which they had hoped to erase, were more sharply drawn than ever.
It is tempting to depict amateur theaters of the 1930s as a fulfillment of the ideas of activists in popular theaters before the revolution, another example of many prerevolutionary institutions and values reborn in the Stalin era.7 The Soviet state established and supported a wide network of stages devoted to educating viewers in the classics and providing them models of civilized behavior, a goal put forward long before the revolution. A significant part of the repertoire, based heavily on Ostrovsky, was very similar; so was the rhetoric of cultural enlightenment. The resurgence of the term "narodnyi teatr" (popular theater), abandoned in the early Soviet period, seems to settle the argument.
Nonetheless, I contend that the Soviet regime's commitment to nurturing and training amateurs sets it apart from prerevolutionary models. The network created by intellectuals before 1917 was designed to provide fine performances of respected work to the lower classes. The performing troupes they favored were not composed of amateurs at all. The people's theater movement began to direct more attention to amateurs in the last years of the Imperial regime, sponsoring local stages and aiding with repertoire and set designs. Nonetheless, the main goal of participants in the movement was to bring good art to the people, not to encourage the people to make it themselves.
Nor can the club stages of the 1930s be easily equated with their prerevolutionary factory and club precedents, although all had ties to trade unions. The oppositional tendencies of worker's theaters before 1917 vanished by the Stalin era. Before the revolution, amateur actors sought out plays that expressed an implicit critique of the dominant economic order. Such intentions were gone by the 1930s, although critics made hypersensitive by the political climate of the purges still discovered hidden oppositional messages. Live performance always carries the potential for unexpected interpretation; through gestures and intonation actors can insert subversive messages in their work. The overseers of Stalinist amateur theater struggled mightily to limit alternative meanings by controlling the repertoire, training methods, and venues where performances took place.
By the Stalin era, urban amateurs also attempted to sever any links to the theater of the fairground and the countryside. Gone were the variations on Tsar Maksimilian and The Boat. Gone were the references to carnival barkers and red-haired clowns, unifying features of living newspapers in the 1920s. Stages in the two capitals no longer tried to offer images that would resonate with new arrivals from the countryside. Instead, amateur actors used their training to shed all traces of country origins in their speech patterns, clothing styles, and deportment.
The Soviet government's determination to attract and train the amateur also distinguish it from other regimes that attempted to bring theater into the lives of their citizenry in the 1930s. In Germany the amateur theater movement had been closely associated with the Social Democratic and Communist opponents of Nazism. Rather than building new opportunities for amateurs, the Hitler state shut down many of the stages built by its political rivals. Instead, it expanded the number of traveling theatrical troupes and integrated amateurs as bit players in large theatrical festivals, the so-called Thingspiele.8 While the Italian Fascist government encouraged clubs that sponsored amateur theatricals, its greatest support went to touring theaters that brought the classics to remote villages. Its most distinctive theatrical product, a theatrical spectacle called "18BL," named after a Fiat truck, was performed by soldiers organized along military lines.9
Nor was the Federal Theatre Project in the United States devoted to training non-professional actors. This massive state cultural program, which flourished from 1935–39, gave work to unemployed professional actors whose livelihood had been undermined by the Depression. It was an innovation in American arts policies, remarkable as an attempt to provide contemporary drama in a wide variety of venues. The repertoire was new and challenging. The program distributed funds to all parts of the nation and worked to encourage theatrical expression among ethnic and racial minorities. It did provide some services to help amateur stages, including a list and synopsis of labor plays. Nonetheless, amateurs were explicitly excluded from the program. Despite its many innovations, the Federal Theatre Project was designed to rejuvenate and expand professional theater.10
The Soviet regime's commitment to the amateur arts was one of its most distinctive cultural features. Through direct and indirect state funding, it supported institutions where amateurs could train and perform. The government encouraged professionals to lend their services to amateur stages. Although the number of journals devoted specifically to amateurs declined in the 1930s, mainstream theater publications set aside considerable space to events on the amateur stage. When the National Committee of The Arts took shape in 1936, it included amateur production in its purview. As Andre Van Gyseghem, active in left-wing theater circles in Britain, observed, "The Soviet State considers the encouragement and increase of amateur acting groups an important part of the educational system and so allots it a definite place in its programme."11
Amateurs are typically marginal figures, argues one scholar of amateurism in the United States. They are isolated from others engaged in leisure-time pursuits because of their serious commitment to their craft. At the same time, they are isolated from professionals because they do not have the time to perfect their skills in order to raise them to the high standards they admire.12 It was precisely this marginal status that the Soviet regime, and the Soviet amateur, hoped to overcome. During the revolution and Civil War, amateurs drew attention to themselves by their enthusiastic embrace of artistic performance. In the 1920s advocates of the amateur arts placed themselves at the center of artistic debates, deprecating the accomplishments of professionals and pointing to amateurs as a unique source of artistic renewal. During the First Five-Year Plan, amateur theater gained a reputation as a particularly effective transmiter of political information, a way to act out the radical transformations underway in the country at large. And even in the 1930s, when professionalism was reinstated to its traditionally privileged status, amateurs were integrated into the professional system. They received praise and support as an important channel to promote a unified artistic culture.
By supporting amateurs, the Soviet government projected the image of a culturally enlightened state. Not only did it strive to share the fruits of prerevolutionary culture with the population at large, it also nurtured the creative potential of all its citizens, turning mail carriers into composers. However, the Soviet government's solicitous concern for the amateur can also be seen as a tacit recognition of the dangers of amateur creation. By bringing it in from the margins, Soviet cultural leaders tried to guarantee that amateur art would offer no surprises. This goal proved elusive. We can see this illustrated in a long 1938 article examining the past and present of Soviet amateur stages by Leonid Subbotin, head of the former Polenov House of Amateur Art, renamed the Center of Amateur Art. Published in the prestigious journal Teatr (Theater), the article presents an optimistic narrative of political and cultural progress for the Soviet amateur stage. Surviving through phases of constructivist and formalist influence, amateur theater had finally become a powerful force serving the healthy cultural instincts of the Soviet viewer, contended Subbotin. Recent competitions had shown that club stages chose to perform the best of contemporary works along with the finest representative of the classics. Serious collaboration with theatrical professionals had raised the quality of the best amateur work almost to the level of the professional stage.13
Yet, Subbotin's cheerful story was periodically interrupted by dire warnings about the necessity for heightened vigilance and control. Enemies of the people had wormed their way into select amateur circles and even into the leadership of the National Committee of the Arts. There was no effective central guidance, no single system of rules outlining training and repertoire, no rigid standards to train and test club instructors. "Only a single, authoritative leadership can raise amateur theater to the level where it can meet the demands of the Soviet viewer and allow it to develop its full, rich artistic potential," he insisted.14 Even at the height of Stalinism, amateur theater remained a possible source of danger for state authorities, a form of cultural expression that could slip off into the margins and evade central control.
This study ends with the period when Soviet amateur stages had accepted the repertoire, training methods, and oversight of professionals. However, they did not always remain in such an abject state. After Stalin's death, amateur theaters experienced a reinvigoration. Select stages became a site for cultural experimentation once more. Housed again on the margins of Soviet cultural life, in basements and warehouses, amateur studio theaters in Moscow and Leningrad solicited an original repertoire and attracted a distinctive audience. By the Gorbachev era, there was a lively network of amateur stages shaping a conscious alternative to state-supported professional theater.15 Soviet amateur theater's oppositional potential, initially employed against remnants of the old regime, now turned against established Soviet culture.
* * *
1. For a discussion of Dom na Trubnoi, see Denise Youngblood, Movies for the Masses (Cambridge: Cambridge University Press, 1992), 135–38. On Volga, Volga see Evgeny Dobrenko, "Soviet Film Comedy, or the Carnival of Authority," Discourse 17, 3 (Spring 1995): 49–57, and S. Iu. Rumiantsev and A. P. Shul'pin, "Samodeiatel'noe tvorchestvo i 'gosudarstvennaia kul'tura,' " in Samodeiatel'noe khudozhestvennoe tvorchestvo v sssR, v. 1 (Moscow: Gosudarstvennyi institut iskusstvoznania, 1995), 18–20.
2. Bertolt Brecht, "Uber den Beruf des Schauspielers," in Schriften zum Theater 1 (Gesammelte Werke 15) (Frankfurt am Main: Suhrkamp, 1967), Anmerkungen, 11.
3. Katerina Clark, "The 'Quiet Revolution' in Soviet Intellectual Life," in Sheila Fitzpatrick et al., eds., Russia in the Era of NEP (Bloomington: Indiana University Press, 1991), 226–27.
4. M. Berliant, Samodeiatel'nyi teatr (Moscow: Profizdat, 1938), 23.
5. For a sophisticated examination of the relationship between the stage and audience in different forms of Soviet theater, see Lars Kleberg, Theatre as Action: Soviet Russian Avant-garde Aesthetics, trans. Charles Rougle (London: Macmillan, 1993).
6. Boris Groys, The Total Art of Stalinism: Avant-Garde, Aesthetic Dictatorship, and Beyond, trans. Charles Rougle (Princeton: Princeton University Press, 1992), 9, 36.
7. This argument about the resurgence of prerevolutionary values and institutions has been made most forcefully in Nicholas S. Timasheff, The Great Retreat: The Growth and Decline of Communism in Russia (New York: E. P. Dutton, 1945), esp. chs. 9 and 10.
8. Cecil W. Davies, Theatre for the People (Manchester: Manchester University Press, 1977), 113–16; Jutta Wardetsky, Theaterpolitik im faschistischen Deutschland: Studien und Dokumente (Berlin: Henschelverlag, 1983), 79–99, 138–64.
9. On Fascist art policies, see Victoria de Grazia, The Culture of Consent: Mass Organization of Leisure in Fascist Italy (Cambridge: Cambridge University Press, 1981), 203–4; Mabel Berezin, "The Organization of Political Ideology: Culture, State, and the Theater in Fascist Italy," American Sociological Review 56 (1991), 639–51; and Jeffrey T. Schnapp, Staging Fascism: 18BL and the Theater of Masses for Masses (Stanford: Stanford University Press, 1996).
10. On the policy of the Federal Theatre Project toward amateurs, see Hallie Flanagan, Arena (New York: Duell, Sloan and Pearce, 1940), 15–16.
11. Andre Van Gyseghem, Theatre in Soviet Russia (London: Faber and Faber, 1942), 153.
12. Robert A. Stebbins, Amateurs: On the Margin between Work and Leisure (Beverly Hills: Sage Publications, 1979), esp. 257–72.
13. L. Subbotin, "Samodeiatel'nyi teatr," Teatr 6 (1938), 113–22.
14. Ibid., 120, 122, quotation 122.
15. On these theaters, see Susan Constanzo, "Reclaiming the Stage: Amateur Theater-Studio Audiences in the Late Soviet Era," Slavic Review 57 (Summer 1998): 398–424.
## Glossary
agitka (plural agitki)—agitational play
Agitprop Division—The Agitation and Propaganda Division of the Communist Party
agitsud—agitational trial or mock trial
Glavpolitprosvet—The Central Agency for Political Education within Narkompros
instsenirovka—material originally not intended for the theater that has been shaped into a performance text
Narkompros—People's Commissariat of Education
NEP—The New Economic Policy, 1921–28
Politprosvet divisions—local agencies for political education
Proletkult—Proletarian Culture Organization
rabkor (rabochii korrespondent) a worker correspondent supplying criticism and reportage for newspapers and journals
rapp—The Russian Association of Proletarian Writers
samodeiatel'nost'—amateurism; independent action
smychka—union or cooperation
tram (Teatr rabochei molodezhi)—The Theater of Working-Class Youth
Vneshkol'nyi otdel—The Division for Extra-curricular Education within Narkompros
## Bibliography
##### Archival Sources
Gosudarstvennyi Arkhiv Rossiiskoi Federatsii (garf)
f. 628. Tsentral'nyi dom narodnogo tvorchestva im. N. K. Krupskoi
f. 2306. Narkompros
f. 2313. Glavpolitprosvet
f. 5451. Vsesoiuznyi tsentral'nyi sovet profsoiuzov (vtssps)
f. 7952. Istoriia fabrik i zavodov
Gosudarstvennyi Tsentral'nyi teatral'nyi muzei im. A. A. Bakhrushina (gtstm)
f. 150. N. L'vov. Lichnyi fond
Rossiiskii Gosudarstvennyi arkhiv literatury i iskusstva (rgali)
f. 645. Glaviskusstvo
f. 941. Teatr Leninskogo Komsomola
f. 963. Gosudarstvennyi Teatr im. Meierkhol'da
f. 1230. Proletkul't
f. 2723. N. G. Zograf
f. 2947. Moskovskii Teatr im. Leninskogo Komsomola
Tsentral'noe khranenie dokumentov molodezhnykh organizatsii (tskhdmo)
f. 1. Tsentral'nyi Komitet Vsesoiuznogo Leninskogo Kommunisticheskogo Soiuza Molodezhi
Tsentral'nyi munitsipal'nyi arkhiv Moskvy (tsmam)
f. 718. Moskovskii gorodskoi sovet professional'nykh soiuzov (mgsps)
f. 2007. Upravlenie moskovskimi zrelishchnymi predstavleniiami (umzp)
##### Newspapers and Journals
Gorn
Griadushchee
Gudki
Iskusstvo kommuny
Iunyi kommunist
Klub
Klub i revoliutsiia
Klubnaia stsena
Kolkhoznyi teatr
Komsomol' skaia pravda
Komsomol 'skii agitproprabotnik
Krasnaia gazeta
Kul 'turnaia rabota profsoiuzov
Literatura i iskusstvo
Literaturnaia gazeta
Literaturnyi Leningrad
Malye formy klubnogo zrelishcha
Materialy dlia klubnoi stseny
Molodaia gvardiia
Narodnoe tvorchestvo
Novyi zritel'
Plamia
Pravda
Prizyv
Proletarskaia kul' tura Rabis
Rabochii i teatr
Rabochii klub
Rabochii zritel'
Repertuarnyi biulleten'
Siniaia bluza
Smena (journal)
Sovetskii teatr
Sovetskoe iskusstvo
Teatr
Teatr i dramaturgiia
Trud
Vecherniaia Moskva
Vestnik teatra
Vestnik zhizni
Vneshkol'noe obrazovanie
Za agitpropbrigadu i TRAM
Za proletarskoe iskusstvo
Zhizn' iskusstva
##### Primary Sources
Andreev, Boris. Sud nad starym bytom: Stsenarii dlia rabochikh klubov ko dniu rabotnitsy 8-go marta. Moscow-Leningrad: Doloi negramotnost', 1926.
Apushkin, Ia. Zhivogazetnyi teatr. Leningrad: Teakinopechat', 1930.
Avlov, Grigorii, ed. Edinyi khudozhestvennyi kruzhok: Metody klubno-khudozh-estvennoi raboty. Leningrad: Izdatel'stvo knizhnogo sektora Gubono, 1925.
——. Igry v klube: Trenirovochnye razvlecheniia. Leningrad: Teakinopechat', 1929.
——. Klubnyi samodeiatel'nyi teatr: Evoliutsiia metodov i form. Leningrad: Teakinopechat', 1930.
——. Sud nad khuliganami. Moscow: Doloi negramotnost', 1927.
——. Teatral'nye agitpropbrigady v klube. Leningrad: Gosudarstvennoe izdatel'stvo khudozhestvennoi literatury, 1931.
Benjamin, Walter. Das Kunstwerk im Zeitalter seiner technischen Reproduzier-barkeit. Frankfurt am Main: Suhrkamp, 1972.
Bergman, S. Teatral'naia rabota v klube. Kharkov: Proletarii, 1925.
Berliant, M. Samodeiatel' nyi teatr. Moscow: Profizdat, 1938.
Boiarskii, Ia., A. Vigalok, M. Lenau, and P. Segal, eds. Professional'nye soiuzy na novom etape: Sbornik materialov k XVI s' 'ezdu VKP(b). Moscow: Izdatel'stvo vtssps, 1930.
Brecht, Bertolt. Schriften zum Theater, v. 1. Frankfurt am Main: Suhrkamp, 1967.
Brown, Ben W. Theatre at the Left. Providence, R.I.: The Booke Shop, 1938.
Bulgakov, A. S., and S. S. Danilov. Gosudarstvennyi agitatsionnyi teatr v Leningrade. Leningrad: Academia, 1931.
Carter, Huntley. The New Spirit in the Russian Theatre, 1917–28. London: Chapman and Dodd, 1924.
Chicherov, I. I. Perezhitoe, nezabyvaemoe. Moscow: Molodaia gvardiia, 1977.
——, ed. Za TRAM: Vsesoiuznoe soveshchanie po khudozhestvennoi rabote sredi molodezhi. Moscow: Teakinopechat', 1929.
Dal'tsev, Z. G. "Moskva 1917–1923: Iz vospominanii." In U istokov: Sbornik statei. Moscow: VTO, 1960.
Desiat' Oktiabrei. Leningrad: Bol'shoi dramaticheskii teatr, 1927.
Devushki nashei strany. Leningrad: Gosudarstvennyi teatr dramy, 1933.
Diament, Kh. Organizatsionnye formy profsoiuznoi kul'traboty. 2nd ed. Moscow: Trud i kniga, 1927.
Dobin, E. S., ed. Adrian Piotrovskii: Teatr, kino, zhizn. Leningrad: Iskusstvo, 1969.
Dolinskii, S., and S. Bergman. Massovaia rabota v klube. Moscow: Rabotnik prosveshcheniia, 1924.
Dune, Eduard. Notes of a Red Guard. Trans. and ed. Diane Koenker and Steven Smith. Urbana: Illinois University Press, 1993.
Edel'son, Z. A., and B. M. Filippov, eds. Profsoiuzy i iskusstvo: Sbornik statei s prilozheniem rezoliutsii pervoi leningradskoi mezhsoiuznoi konferentsii po vo-prosam khudozhestvennoi raboty. Leningrad: Izdatel'stvo leningradskogo gubprofsoveta, 1927.
Ehrenburg, Ilia. People and Life, 1891–1921. Trans. Anna Bostock. New York: Knopf, 1962.
Erdman, Nikolai. Two Plays. Trans. Marjorie Hoover. Ann Arbor: Ardis, 1975.
Feuchtwanger, Leon. Moskau 1937: Ein Reisebericht für meine Freunde. 1937. Reprint. Berlin: Aufbau Verlag, 1993.
Filippov, V. Puti samodeiatel'nogo teatra: Ocherk. Moscow: Gosudarstvennaia Akademiia khudozhestvennykh nauk, 1927.
Flanagan, Hallie. Arena. New York: Duell, Sloan and Pearce, 1940.
Fülöp-Miller, Rene, and Joseph Gregor. The Russian Theatre. Trans. Paul England. 1930. Reprint. New York: Benjamin Blom, 1968.
Gol'dman, A., and M. Imas. Sotsialisticheskoe sorevnovanie i udarnichestvo v iskusstve. Moscow: Gosudarstvennoe izdatel'stvo khudozhestvennoi liter-atury, 1931.
Gorbenko, A. N. "Krysha." rgali, f. 2723, op. 1, d. 531, 11. 190–245.
——. Sashka Chumovoi: Komsomol'skaia komediia. In A. Piotrovskii and M. Sokolovskii, eds. Sbornik p'es dlia komsomol'skogo teatra. Moscow: Gosu-darstvennoe izdatel'stvo, 1928.
Griffith, Hubert, ed. Playtime in Russia. London: Methuen, 1935.
Gvozdev, A. A. "Klassiki na sovetskoi stsene." Literaturnyi sovremennik 6 (1933): 127–46.
——. Teatral'naia kritika. Ed. A. Ia. Al'tshuller et al. Moscow: Iskusstvo, 1987. Houghton, Norris. Moscow Rehearsals: The Golden Age of Soviet Theatre. New York: Harcourt Brace, 1936.
Gvozdev, A. A., and A. Piotrovskii. "Petrogradskie teatry i prazdnestva v epokhu voennogo kommunizma." In Istoriia sovetskogo teatra, vl. 1 Ed. V. E. Rafailovich. Leningrad, 1933.
Imas, M. Iskusstvo na fronte rekonstruktsii sel'skogo khoziaistva. Moscow: Gosu-darstvennoe izdatel'stvo khudozhestvennoi literatury, 1931.
Isaev, I. Osnovnye voprosy klubnoi stseny. Moscow: vtssps, 1928.
Iskusstvo v rabochem klube. Moscow: Vserossiiskii Proletkul't, 1924.
Iufit, A. Z., ed. Russkii sovetskii teatr, 1917–1921. Leningrad: Iskusstvo, 1968.
Ivanter, B. and V. Zemlia zazhglas': P'esa v trekh chastiakh. Moscow: Molodaia gvardiia, 1924.
Jelagin, Juri. The Taming of the Arts. Trans. Nicholas Wreden. New York: Dutton, 1951.
Kabo, E. O. Ocherki rabochego byta: Opyt monograficheskogo issledovaniia domashnego rabochego byta. Moscow: vtssps, 1928.
Kagan, A. G. Molodezh' poslegudka. Moscow: Molodaia gvardiia, 1930.
——. Sorok piat' dnei sredi molodezhi. Leningrad: Priboi, 1929
Kak ia stal rezhisserom. Moscow: Goskinoizdat, 1946.
Karzhanskii, N. Kollektivnaia dramaturgiia. Moscow: Gosizdat, 1922.
Kataev, Valentin. Time, Forward! Trans. Charles Malamuth. Bloomington: Indiana University Press, 1976.
Kerzhentsev, P. M. Revoliutsiia i teatr. Moscow: Dennitsa, 1918.
——. Tvorcheskii teatr. 5th ed. Peterburg: Gosizdat, 1923.
Khudozhestvennye agitbrigady: Itogi smotra i puti razvitiia. Moscow: Narkompros rsfsr, 1931.
Kirchon [Kirshon], V., and A. Ouspensky. Red Rust. Adapted by Virginia and Frank Vernon. New York: Bretano's, 1930.
Kirshon, Vladimir. Chudesnyi splav: Komediia v chetyrekh aktakh. Moscow: Iskusstvo, 1956.
Klub kak on est'. Moscow: Trud i kniga, 1929.
Kluby Moskvy i gubernii. Moscow: Trud i kniga, 1926.
Knorre, F. Moskovskii 10:10. Moscow: Gosudarstvennoe izdatel'stvo khudozh-estvennoi literatury, 1933.
Kobrin, Iurii. Teatr im. Vs. Meierkhol'da i rabochii zritel'. Moscow: Moskovskoe teatral'noe izdatel'stvo, 1926.
Kommunisticheskaia partiia Sovetskogo Soiuza v rezoliutsiiakh i resheniiakh s' 'ezdov, konferentsii i plenumov TsK. vol. 2: 1917–1924. Moscow: Izdatel'stvo politich-eskoi literatury, 1970.
Komsomol'skaia paskha. Moscow: Novaia Moskva, 1924.
Korev, S. Zhivaia gazeta v klube. Ed. R. Pel'she. Moscow: Novaia Moskva, 1925.
Korovkin, I., and S. Erkov. Zovi fabkom: P'esa v 3-kh deistviiakh. Leningrad: Teakinopechat', 1929.
Kriuchkov, Nikolai. "Khudozhestvennyi agitprop komsomola." Teatral'naia zhizn 14 (July 1970): 1–3.
Krupskaia, Nadezhda. Pedagogicheskie sochineniia. Moscow: Izdatel'stvo Akademii pedagogicheskikh nauk, 1956.
Krylov, S. M., ed. Puti razvitiia teatra: Stenograficheskii otchet i resheniia partiinogo soveshchaniia po voprosam teatra pri Agitprope TsK VKP(b) v mae 1927 g. Moscow: Kinopechat', 1927.
Lebedev-Polianskii, V. I., ed. Protokoly pervoi Vserossiiskoi konferentsii proletarskikh kul'turno-prosvetitel'nykh organizatsii. Moscow: Proletarskaia kul'tura, 1918.
Leningradskii TRAM v Moskve iiun' 1928g. Leningrad: Gostram, 1928.
Lissitzky, El. Russia: An Architecture for World Revolution. Trans. Eric Dluhosch. Cambridge: MIT Press, 1970.
L'vov, N. [F.] Klesh zadumchivyi: Dialekticheskoe predstavlenie v 3-kh krugakh. Leningrad: Teakinopechat', 1930.
——. Plaviatsia dni: Dialekticheskoe predstavlenie v 3-kh krugakh. Leningrad: Teakinopechat', 1929.
——. Zor'ka: P'esa v trekh deistviiakh. In A. Piotrovskii and M. Sokolovskii, eds.
Sbornik p'es dlia komsomol'skogo teatra. Moscow: Gosudarstvennoe iz-datel'stvo, 1928.
L'vov, N. [I.] "P'esa ili stsenarii.' Vestnik rabotnikov iskusstv 2/3 (1920): 53–54.
——. Postroenie agitzrelishcha. Moscow: Teakinopechat', 1930.
Maksimov, P., and N. L'vov. Druzhnaia gorka: Komsomol'skaia operetta v 3-kh deistviiakh. Leningrad: Teakinopechat', 1929.
Marinchik, Pavel. "Dalekoe-blizkoe." Neva 11 (1957): 169–173.
——. "Khudozhestvennyi agitprop komsomola." Neva 11 (1965): 202–205.
——. Meshchanka. Leningrad: Teakinopechat', 1929.
——. Rozhdenie komsomol 'skogo teatra. Leningrad: Iskusstvo, 1963.
——, and S. Kashevnik. Budni. in A. Piotrovskii and M. Sokolovskii, eds. Teatr rabochei molodezhi. Moscow: Gosudarstvennoe izdatel'stvo, 1928.
Markov, P. A. The Soviet Theatre. London: Victor Gollancz, 1934.
Marx, Karl, and Friedrich Engels. Literature and Art. New York: Progress Press, 1947.
Massovye prazdnestva. Leningrad: Academia, 1926.
Mierau, Fritz, ed. Sergei Tretiakov: Gesichter der Avantgarde. Berlin: Aufbau Verlag, 1985.
Mikitenko, I. K. Devushki nashei strany: P'esa v 4 deistviiakh. Leningrad: Izdanie Kul'tkabineta Gosdramy, 1932.
Na putiakh iskusstva: Sbornik statei. Moscow: Proletkul't, 1926.
Novye etapy samodeiatel'noi khudozhestvennoi raboty. Leningrad: Teakinopechat', 1930.
Oktiabr' v klube. Moscow: Trud i kniga, 1924.
Oktiabr' v rabochikh klubakh. Moscow: Krasnaia nov', 1923.
Orlovsky, Sergei. "Moscow Theater, 1917–1941," in Martha Bradshaw, ed. Soviet Theater 1917–1941. New York: Research Program on the USSR, 1954.
Pel'she, R., ed. Komsomol, na front iskusstva! Moscow: Teakinopechat', 1929.
——. Nasha teatral'naia politika. Moscow: Gosudarstvennoe izdatel'stvo, 1929.
Pervaia Vsesoiuznaia olimpiada samodeiatel'nogo iskusstva. Moscow: Profizdat, 1932.
Pervichnaia komsomol'skaia organizatsiia: Dokumenty i materialy s' 'ezdov, konfer-entsii komsomola, Tsentral'nogo Komiteta vlksm, 1918–1971. Moscow: Molodaia gvardiia, 1972.
Petrogradskaia obshchegorodskaia konferentsiia rabochikh klubov. Petrograd: Izdatel'stvo Leningradskogo gubprofsoveta, 1920.
Pletnev, V. Rabochii klub. Moscow: Proletkul't, 1925.
Pimenov, V. F., ed. Pervye sovetskie p'esy. Moscow: Iskusstvo, 1958.
Piotrovskii, A. I. 'K teorii 'samodeiatel'nogo teatra.'" Problemy sotsiologii iskusstva. Leningrad: Academia, 1926.
——. Kinofikatsiia iskusstv. Leningrad: lzdanie avtora, 1928.
——, ed. Krasnoarmeiskii teatr: Instruktsiia k teatral'noi rabote v Krasnoi Armii. Petrograd: Izdatel'stvo Petrogradskogo voennogo okruga. 1921.
——. "tram: Stranitsa teatral'noi sovremennosti." Zvezda 4 (1929): 142–52.
——. Za sovetskii teatr! Sbornik statei. Leningrad: Academia, 1925.
——, and M. Sokolovskii, eds. Teatr rabochei molodezhi: Sbornik p'es dlia komso-
mol'skogo teatra. Leningrad: Teakinopechat', 1928.
Politprosvetrabota i teatr. Moscow: Doloi negramotnost', 1927.
Rabochie o literature, teatre i muzyke. Leningrad: Priboi, 1926.
Ravenskikh, Boris Ivanovich. "Istochnik novoi energii." Teatral'naia zhizn' 24 (1964): 17–18.
Reines, B. "The Experience of the International Workers' Theatre as Reported at the First Enlarged Plenum of the I.W.D.U." Workers' Theatre 1, no. 9 (December 1931):1–4.
Repertuarnyi ukezatel': Sbornik otzyvov o p'esakh dlia professional'nogo i samod-eiatel'nogo teatra. Moscow: Glavpolitprosvet, 1925.
Romashov, Boris Sergeevich. Konets Krivoryl'ske. In P'esy. Moscow: Khudozh-estvennaia literatura, 1935.
Rostislavlev, N. Dai piat': P'esa v 3 krugakh i 10 kertinakh, Leningrad: Teakinopechat', 1930.
Savvin, I. D., ed. Komsomol i teatr. Moscow: mk vlksm, 1933.
Sbornik instsenirovok: Opyty kollektivnoi dramaturgii. Leningrad: Izdatel'stvo Knizhnogo sektora Gubono, 1924.
Sbornik materialov k III plenumu Tsentral'nogo soveta TRAM'ov pri TsK vlksm, Moscow: Teakinopechat', 1930.
Sbornik rukovodiashchikh materialov po rabote klubov. Moscow: Profizdat, 1930.
Scott, John. Behind the Urals: An American Worker in Russia's City of Steel. Ed. Stephen Kotkin. Bloomington: Indiana University Press, 1989.
Sed 'moi s' 'ezd professional'nykh soiuzov. Moscow: Profizdat, 1926.
Shcheglov, Dmitrii. Chetyre kepki. In Krasnyi teatr: Komsomol'skie p'esy. Moscow: Gosizdat, 1926.
——. Teatral'no-khudozhestvennaia rabota v klubakh: Metodike i praktika. Leningrad: Gubprofsovet, 1926.
——. "U istokov." In U istokov: Sbornikstatei. Moscow: VTO, 1960.
Shestnadtsatyi s' 'ezd Vsesoiuznoi Kommunisticheskoi partii (B): Stenograficheskii otchet. Moscow: OGIZ-Moskovskii rabochii, 1931.
Shishigin, F. E. "Khudozhniki, kotorykh sleduet vspomnit'," Teatral'naia zhizn'. 24 (1964): 15–16.
Shishigin, F. E., A. Piotrovskii, and M. Sokolovskii. "Zelenyi tsekh: Operetta v 3-kh deistviiakh." rgali f. 2723, op. 1, d. 531, ll: 246–94.
Shklovskii, Viktor. Khod konia: Sbornik statei. Moscow: Gelikon, 1923.
Skorinko, I. Buzlivaia kogorta: P'esa v 4 deistviiakh. Leningrad: Priboi, 1928.
Slukhovskii, M. Ustnaia gazeta kek vid politiko-prosvetitel'noi raboty. Moscow: Krasnaia zvezda, 1924.
Sokolovskii, M. "Sploshnoi potok." rgali, f. 2723, op. 1, d. 531, ll. 110–61.
——. Za novyi byt: Zhivaia gazeta. Moscow: Gosudarstvennoe izdatel'stvo, 1927.
Sostav rabochei molodezhi: Po massovym dannym profperepisi. Moscow: Molodaia gvardiia, 1931.
Stalin, J. V. Problems of Leninism. Peking: Foreign Languages Press, 1976.
Tikhonovich, V. V. Narodnyi teatr. Moscow: V. Magnussen, 1918.
——. Samodeiatel'nyi teatr. Vologda: Oblastnoi otdel Gosizdata, 1922.
——. Teatr i sovremennost.' Moscow: Doloi negramotnost', 1928.
Thomas, Tom. "World Congress of Workers' Theatre Groups." New Masses (November 1930): 21.
Tolmachev, D. Fabzavshturm. In A. Piotrovskii and M. Sokolovskii, eds., Teatr rabochei molodezhi. Moscow: Gosudarstvennoe izdatel'stvo, 1928.
Tolstoi, V. P., ed. Agitatsionno-massovoe iskusstvo: Oformlenie prazdnestv, 1917–1932. 2 v. Moscow: Iskusstvo, 1972.
Tolstoi, V. P. et al., eds. Street Art of the Revolution: Festivals and Celebrations in Russia, 1918–33. London: Thames and Hudson, 1990.
Tovarishch komsomol: Dokumenty s' 'ezdov, konferentsii i TsK vlksm 1918–1968, v. 1. Moscow: Molodaia gvardiia, 1969.
Trabskii, A. Ia., ed. Russkii sovetskii teatr, 1921–1926. Leningrad: Iskusstvo, 1975.
——. Russkii sovetskii teatr, 1926–1932. Leningrad: Iskusstvo, 1982.
Troianskii, A. V., and R. I. Egiazarov. Izuchenie kino-zritelia. Moscow: Gosu-darstvennoe izdatel'stvo, 1928.
Trotsky, Leon. Problems of Everyday Life. New York: Monad Press, 1973.
Tur, Brat'ia [pseud.], and L. Sheinin. Ochnaia stavka. Moscow: Iskusstvo, 1938.
Van Gysegham, Andre. Theatre in Soviet Russia. London: Faber and Faber, 1943.
Veprinskii, M. Khudozhestvennye kruzhki i krasnyi kalendar'. Moscow: Gosu-darstvennoe izdatel'stvo, 1926.
——. Zhivaia gazeta. Moscow: Doloi negramotnost', 1927.
Von Geldern, James, and Richard Stites, eds. Mass Culture in Soviet Russia. Bloomington: Indiana University Press, 1995.
Voprosy kul'tury pri diktature proletariata. Moscow: Gosudarstvennoe iz-datel'stvo, 1925.
Vsevolodskii-Gerngross, V. N. Istoriia russkogo teatra. Leningrad: Teakino-pechat', 1929.
——. Russkaia ustnaia narodnaia drama. Moscow: Akademiia Nauk, 1959.
Wettlin, Margaret. Fifty Russian Winters: An American Woman's Life in the Soviet Union. New York: John Wiley, 1994.
Winter, Ella. Red Virtue: Human Relationships in the New Russia. New York: Har-court Brace, 1933.
Wolkonsky, Serge. My Reminiscences. 2 v. Trans. A. E. Chamot. London: Hutchinson, 1924.
Zadykhin, Ia. L. Khuligan. Leningrad: Izdatel'stvo MODPiK, 1925.
Zamoskvoretskii, V. Klub rabochei molodezhi. Moscow: Novaia Moskva, 1924.
Zor'ka: V pomoshch' zriteliu. Moscow: Teakinopechat', 1929.
##### Secondary Sources
Aizenshtadt, V. N. Sovetskii samodeiatel'nyi teatr: Osnovnye etapy razvitiia. Khar kov: Khar'kovskii gosudarstvennyi institut kul'tury, 1983.
Amiard-Chevrel, Claudine "La Blouse Bleue." In L'Théâtre d'agit-prop de 1917 à 1932, v. 1. Lausanne: La Cite—L'age d'homme, 1977.
——. "Le theatre de la Jeunesse Ouvrière (TRAM)." In Le Théâtre d'agit-prop de 1917 à 1932, v. 1. Lausanne: La Cité—L'age d'homme, 1977.
Aston, Elaine, and George Savona. Theatre as Sign-System. London: Routledge, 1991.
Barris, Roanne. "Chaos by Design: The Constructivist Stage and Its Reception." Ph. D. Dissertation, University of Illinois, Champaign-Urbana, 1994.
Berezin, Mabel. "The Organization of Political Ideology: Culture, State, and Theater in Fascist Italy." American Sociological Review 56 (1991): 639–51.
Blok, V. B. "Khudozhestvennoe tvorchestvo mass." In A. Ia. Zis', ed., Stran-itsy istorii sovetskoi khudozhestvennoi kul'tury. Moscow: Molodaia gvardiia, 1989.
Bodek, Richard. Proletarian Performance in Weimar Berlin: Agitprop, Chorus, and Brecht. Columbia, S.C.: Camden House, 1997.
Bonnell, Victoria. Iconography of Power: Soviet Political Posters under Lenin and Stalin. Berkeley: University of California Press, 1997.
——. The Roots of Rebellion. Berkeley: University of California Press, 1983.
Booth, Wayne. For the Love of It: Amateuring and Its Rivals. Chicago: University of Chicago Press, 1999.
Borland, Harriet. Soviet Literary Theory and Practice during the First Five-Year Plan, 1928–32. New York: King's Crown Press, 1950.
Bourdieu, Pierre. Distinction: The Social Critique of the Judgement of Taste. Trans. Richard Nice. Cambridge: Cambridge University Press, 1984.
Bown, Matthew Cullerne. Art under Stalin. New York: Holmes and Meier, 1991.
Boym, Svetlana. Common Places: Mythologies of Everyday Life in Russia. Cambridge: Harvard University Press, 1994.
Bradby, David. "The October Group and Theatre under the Front Populaire," in David Bradby et al., eds. Politics and Performance in Popular Drama. Cambridge: Cambridge University Press, 1980.
Bradby, David, and John McCormick. People's Theatre. London: Croom Helm, 1978.
Bradshaw, Martha, ed. Soviet Theatres, 1917–1941. New York: Research Program on the USSR, 1954.
Braulich, Heinrich. Die Volksbühne. Berlin: Henschelverlag, 1976.
Brooks, Jeffrey. When Russia Learned to Read: Literacy and Popular Literature, 1861–1917. Princeton: Princeton University Press, 1985.
Brooks, Peter. The Melodramatic Imagination: Balzac, Henry James, Melodrama, and the Mode of Excess. New Haven: Yale University Press, 1976.
Brown, Edward J. The Proletarian Episode in Russian Literature, 1928–1932. New York: Columbia University Press, 1953.
Cassiday, Julie. "The Theater of the World and the Theater of State: Drama and the Show Trial in Early Soviet Russia." Ph.D. dissertation, Stanford University, 1995.
Chase, William J. Workers, Society and the Soviet State: Labor and Life in Moscow, 1918–1929. Urbana: University of Illinois Press, 1987.
Clark, Katerina. "Little Heroes and Big Deeds: Literature Responds to the First Five-Year Plan." In Sheila Fitzpatrick, ed., Cultural Revolution in Russia, 1928–1931. Bloomington: University of Indiana Press, 1984.
——. Petersburg, Crucible of Cultural Revolution. Cambridge: Harvard University Press, 1995.
——. "The 'Quiet Revolution' in Soviet Intellectual Life." In Sheila Fitzpatrick et al, eds., Russia in the Era of NEP. Bloomington: Indiana University Press, 1991.
——. The Soviet Novel: History as Ritual. Chicago: University of Chicago Press, 1981.
Corbesero, Susan. "If We Build It, They Will Come: The International Red Stadium Society." Unpublished manuscript.
Cosgrove, Stuart. "From Shock Troupe to Group Theatre." In Theatres of the Left, 1880–1935. Ed. Raphael Samuels. London: Routledge, 1984.
Costanzo, Susan. "Reclaiming the Stage: Amateur Theater-Studio Audiences in the Late Soviet Era." Slavic Review 57 (Summer 1998): 398–424.
Curtiss, J. A. E. "Down with the Foxtrot! Concepts of Satire in the Soviet Theatre of the 1920s." In Russian Theatre in the Age of Modernism. Ed. Robert Russell and Andrew Barratt. London: Macmillan, 1990.
David-Fox, Michael. Revolution of the Mind: Bolshevik Strategies of Higher Education. Ithaca: Cornell University Press, 1997.
——. "What Is Cultural Revolution?" Russian Review 58 (April 1999): 181–201.
Davies, Cecil W. Theatre for the People: The Story of the Volksbühne. Austin: University of Texas Press, 1977.
Davies, Sarah. Popular Opinion in Stalin's Russia: Terror, Propaganda, and Dissent. Cambridge: Cambridge University Press, 1997.
Deak, Frantisek. "Blue Blouse (1923–1928)." Drama Review 17: 1 (1973): 35–46.
De Grazia, Victoria. The Culture of Consent: Mass Organization of Leisure in Fascist Italy. Cambridge: Cambridge University Press, 1981.
Dictionary of Russian Women Writers. Westport, Conn.: Greenwood Press, 1994.
Dobrée, Bonamy. The Amateur and the Theatre. London: Hogarth Press, 1947.
Dobrenko, Evgeny. The Making of the State Reader: Social and Aesthetic Contexts of the Reception of Soviet Literature. Stanford: Stanford University Press, 1997.
——. "Soviet Film Comedy, or the Carnival of Authority." Discourse 17, no.3 (Spring 1995): 49–57.
Ermolin, E. A. Materializatsiia prizraka: Totalitarnyi teatr sovetskikh massovykh akt-sii 1920–1930kh godov. Iaroslavl: iagpu im. K. D. Ushinskogo, 1996.
Esslin, Martin. Brecht: The Man and His Work. Rev. ed. New York: Norton, 1971.
Filtzer, Donald. Soviet Workers and Stalinist Industrialization: The Formation of Modern Soviet Production Relations, 1928–1941. London: Pluto, 1986.
Fisher, Ralph Talcott, Jr. Pattern for Soviet Youth: A Study of the Congresses of the Komsomol, 1918–1954. New York: Columbia University Press, 1959.
Fitzpatrick, Sheila. The Commissariat of Enlightenment: Soviet Organization of Education and the Arts under Lunacharsky. Cambridge: Cambridge University Press, 1970.
——. The Cultural Front: Power and Culture in Revolutionary Russia. Ithaca: Cornell University Press, 1992.
——, ed. Cultural Revolution in Russia, 1928–1931. Bloomington: Indiana University Press, 1984.
——. Education and Social Mobility in the Soviet Union, 1921–1934. Cambridge: Cambridge University Press, 1979.
——. "The Emergence of Glaviskusstvo." Soviet Studies 32 (October 1971): 236–53.
——. Everyday Stalinism. New York: Oxford University Press, 1999.
Frank, Stephen. "Popular Justice, Community, and Culture among the Russian Peasantry, 1870–1900." Russian Review 46, no. 3 (1987): 239–65.
——. " 'Simple Folk, Savage Customs?' Youth, Sociability, and the Dynamic of Culture in Rural Russia, 1856–1914." Journal of Social History 25 (1992): 711–36.
Gerould, Daniel. "Gorky, Melodrama, and the Development of Early Soviet Theatre." Theatre Journal 7 (Winter 1976): 33–44.
Getty, J. Arch, and Oleg V. Naumov. The Road to Terror: Stalin and the Self-Destruction of the Bolsheviks, 1932–1939. New Haven: Yale University Press, 1999.
Gilman, Christopher. "The Fox-Trot and the New Economic Policy." Experiment 2 (1996): 443–75.
Golomstock, Igor. Totalitarian Art in the Soviet Union, the Third Reich, Fascist Italy, and the People's Republic of China. New York: Harper Collins, 1990.
Gorchakov, N. A. The Theater in Soviet Russia. Trans. Edgar Lehrman. New York: Columbia University Press, 1957.
Gorsuch, Anne. Enthusiasts, Bohemians, and Delinquents: Soviet Youth Culture, 1921–1928. Bloomington Indiana University Press, forthcoming.
——. "Soviet Youth and the Politics of Popular Culture." Social History 17, no. 2 (1992): 189–201.
Gorzka, Gabriele. Arbeiterkultur in der Sovietunion: Industriearbeiter-Klubs, 1917–1929. Berlin: Amo Spitz, 1990.
Gromov, Evgenii. Stalin: Vlast' i iskusstvo. Moscow: Respublika, 1996.
Groys, Boris. The Total Art of Stalinism: Avant-garde, Aesthetic Dictatorship, and Beyond. Trans. Charles Rougle. Princeton: Princeton University Press, 1992.
Gruber, Helmut. Red Vienna: Experiment in Working-Class Culture, 1919–1934. New York: Oxford University Press, 1991.
Günther, Hans. "Einleitung." In N. F. Chuzhak, ed., Literatura fakta. Munich: W. Fink, 1972.
Guttsman, W. L. Worker's Culture in Weimar Germany: Between Tradition and Commitment. New York: Berg, 1990.
Hatch, John. "The Formation of Working Class Cultural Institutions during NEP: The Workers' Club Movement in Moscow, 1921–1923." Carl Beck Papers in Russian and East European Studies, no. 806, 1990.
——. "Hangouts and Hangovers: State, Class and Culture in Moscow's Workers' Club Movement, 1925–1928." Russian Review 53 (1994): 97–117.
——. "The Politics of Mass Culture: Workers, Communists, and Proletkul't in the Development of Workers' Clubs, 1921–1925." Russian History 13, nos. 2/3 (1986): 119–48.
Hoffman-Ostwald, Daniel, and Ursula Bekse. Agitprop, 1924–1933. Leipzig: VEB Friedrich Hofmeister, 1960.
Hoffmann, David L. Peasant Metropolis: Social Identities in Moscow, 1929–1941. Ithaca: Cornell University Press, 1994.
Hoover, Marjorie. Alexander Ostrovsky. Boston: Twayne Publishers, 1981.
Huyssen, Andreas. After the Great Divide: Modernism, Mass Culture, Postmodernism. Bloomington: Indiana University Press, 1986.
Ivashev, V., ed. Ot "zhivoi gazety" do teatra-studii. Moscow: Molodaia gvardiia, 1989.
Kakhnovich, A. D. "Rol' khudozhestvennoi samodeiatel'nosti rabochikh v podgotovke kadrov professional'nogo iskusstva, 1933–1937." In Iz istorii sovetskoikul'tury. Moscow: Mysl', 1972.
Kelly, Catriona. Petrushka: The Russian Carnival Puppet Theatre. Cambridge: Cambridge University Press, 1990.
Kemp-Welch, A. Stalin and the Literary Intelligentsia, 1928–1939. New York: St. Martin's, 1991.
Kenez, Peter. The Birth of the Propaganda State: Soviet Methods of Mass Mobilization, 1917–1929. Cambridge: Cambridge University Press, 1985.
——. Cinema and Soviet Society, 1917–1953. Cambridge: Cambridge University Press, 1992.
Khaichenko, G. A. Russkii narodnyi teatr kontsa XIX-nachala XX veka. Moscow: Nauka, 1975.
Khan-Magomedov, Selim O. Rodchenko: The Complete Work. Cambridge: MIT Press, 1987.
Khazanova, V. Klubnaia zhizn' i arkhitektura. 2 vols. Moscow: Rossiiskii institut iskusstvoznaniia, 1994.
Kino: Entsyklopedicheskii slovar'. Moscow: Sovetskaia Entsiklopediia, 1986.
Kleberg, Lars. "The Nature of the Soviet Audience: Theatrical Ideology and Audience Research in the 1920's." in Robert Russell and Andrew Barratt, eds., Russian Theatre in the Age of Modernism. London: Macmillan, 1990.
——. " 'Peoples' Theatre' and the Revolution: On the History of a Concept before and after 1917." In A. A. Nilsson, ed., Art, Society, Revolution: Russia, 1917–1921. Stockholm: Almqvist and Wiksell International, 1979.
——. Theatre as Action. Trans. Charles Rougle. London: Macmillan, 1993.
Koenker, Diane. "Class and Class Consciousness in Socialist Society." In Sheila Fitzpatrick et al., eds., Russia in the Era of NEP. Bloomington: Indiana University Press, 1991.
Kolesova, A. K. "Prakticheskaia deiatel'nost' rabochego kluba v 1917–1920 go-dakh." Uchenye zapiski Moskovskogo instituta kul'tury 17 (1968): 231–49.
Konechnyi, Al'ban M. "Popular Carnivals during Mardi Gras and Easter Week in St. Petersburg." Russian Studies in History 35, no. 4 (Spring 1997): 52–91.
Kotkin, Stephen. Magnetic Mountain: Stalinism as a Civilization. Berkeley: University of California Press, 1995.
Kukaretin, V. M. Nasledniki "Sinei bluzy". Moscow: Molodaia gvardiia, 1976.
Kuromiya, Hiroaki. Stalin's Industrial Revolution: Politics and Workers, 1918–1932. Cambridge: Cambridge University Press, 1988.
Kuziakina, Natalia. Theatre in the Solovki Prison Camp. Luxembourg: Harwood Academic Publishers, 1995.
Lane, Christel. The Rites of Rulers. Cambridge: Cambridge University Press, 1981.
Lapidus, Gail Warshofsky. Women in Soviet Society. Berkeley: University of California Press, 1978.
Lebina, Nataliia Borisovna. Rabochaia molodezh' Leningrada: Trud i sotsial'nyi ob-lik, 1921–1925gg. Leningrad: Nauka, 1982.
Levina, L. P. and T. B. Chirikova. Russkii sovetskii dramaticheskii teatr: An-notirovannyi ukezatel' bibliograficheskikh i spravochnykh materialov, 1917–1973. Moscow: Ministerstvo kul'tury sssR, 1977–78.
Levine, Ira A. Left-Wing Dramatic Theory in the American Theatre. Ann Arbor: UMI Research Press, 1985.
Leyda, Jay. Kina: A History of the Russian and Soviet Film. Princeton: Princeton University Press, 1983.
Maksimenkov, Leonid. Sumbur vmesto muzyki: Stalinskeia kul'turnaia revoliut-siia, 1936–1938. Moscow: Iuridicheskaia kniga, 1997.
Mally, Lynn. "Autonomous Theater and the Origins of Socialist Realism: The 1932 Olympiad of Autonomous Art." Russian Review 52 (April 1993): 198–212.
——. Culture of the Future: The Proletkult Movement in Revolutionary Russia. Berkeley: University of California Press, 1990.
——. "Performing the New Woman: The Komsomolka as Actress and Image in Soviet Youth Theater." Journal of Social History 30, no. 1 (1996): 79–95.
——. "The Rise and Fall of the Soviet Youth Theater TRAM." Slavic Review 51, no. 3 (1992): 411–30.
——. "Shock Workers on the Cultural Front: Agitprop Brigades in the First Five-Year Plan." Russian History 23, no. 1/4 (1996): 263–75.
Mazaev, A. I. Prazdnik kak sotsial'no-khudozhestvennoe iavlenie. Moscow: Nauka, 1978.
Millar, James R. "History and Analysis of Soviet Domestic Bond Policy." In Susan Linz, ed., The Soviet Economic Experiment. Urbana: Illinois University Press, 1990.
Miller, Frank J. Folklore for Stalin: Russian Folklore and Pseudofolklore of the Stalin Era. Armonk, N.Y.: M. E. Sharpe, 1990.
Mironova, V. "Rezhisser i akter v spektakliakh Leningradskogo tram'a." In Teatr i dramaturgiia, v. 5. Leningrad: Iskusstvo, 1976.
——. TRAM: Agitatsionnyi molodezhnyi teatr, 1920–1930kh gg. Leningrad: Iskusstvo, 1977.
Naiman, Eric. "The Case of Chubarov Alley: Collective Rape, Utopian Desire and the Neutrality of NEP." Russian History 17. (1990): 1–30.
——. Sex in Public: The Incarnation of Early Soviet Ideology. Princeton: Princeton University Press, 1997.
Nekrylova, A. F. Russkie narodnye gorodskie prazdniki, uveseleniia i zrelishcha. Leningrad: Iskusstvo, 1984.
——, and N. I. Savushkina. "Russkii fol'klornyi teatr," in L. M. Leonov, ed., Narodnyi teatr. Moscow: Sovetskaia Rossiia, 1991.
Neuberger, Joan. Hooliganism: Crime, Culture, and Power in St. Petersburg, 1900–1914. Berkeley: University of California Press, 1993.
Papernyi, V. Kul'tura "dva." Ann Arbor: Ardis, 1985.
Peris, Daniel. Storming the Heavens: The Soviet League of the Militant Godless. Ithaca: Cornell University Press, 1998.
Petrone, Karen. Life Has Become More Joyous, Comrades: Celebrations in the Time of Stalin. Bloomington: Indiana University Press forthcoming.
Pinegina, L. A. Sovetskii rabochii klass i khudozhestvennaia kul 'tura. Moscow: Iz-datel'stvo Moskovskogo Universiteta, 1984.
Rabiniants, N. A. Teatr iunosti: Ocherk istorii Leningradskogo teatra imeni Lenin-skogo komsomola. Leningrad: Iskusstvo, 1959.
——. "Teatry, rozhdennye revoliutsiei: Leningradskii TRAM." In Teatr i zhizn':
Sbornik, ed. M. O. Iankovskii. Leningrad: Iskusstvo, 1957.
Razumov, V. A. "Rol' rabochego klassa v stroitel'stve sotsialisticheskoi kul'-tury v nachale revoliutsii i v gody grazhdanskoi voiny." In Rol' rabochego klassa v razvitii sotsialisticheskoi kul'tury. Moscow: Izdatel'stvo "Mysl'," 1976.
Robin, Régine. "Popular Literature of the 1920s." In Sheila Fitzpatrick et al., eds., Russia in the Era of NEP. Bloomington: Indiana University Press, 1991.
——. Socialist Realism: An Impossible Aesthetic. Trans. Catherine Porter. Stanford: Stanford University Press, 1992.
——. "Stalinism and Popular Culture." In The Culture of the Stalin Period, ed. Hans Gunther. London: Macmillan, 1990.
Ruder, Cynthia A. Making History for Stalin: The Story of the Belomor Canal. Gainesville: University Press of Florida, 1998.
Rudnitsky, Konstantin. Russian and Soviet Theatre: Tradition and the Avant-Garde. Trans. Roxane Permar. London: Thames and Hudson, 1988.
Rumiantsev, S. Iu., and A. P. Shul'gin, eds. Samodeiatel'noe khudozhestvennoe tvorchestvo v SSRR. 2 v. Moscow: Gosudarstvennyi institut iskusstvoznaniia, 1995.
Russell, Robert. "The First Soviet Plays." In Robert Russell and Andrew Bar-ratt, eds., Russian Theatre in the Age of Modernism. London: Macmillan, 1990.
——. "People's Theatre and the October Revolution." Irish Slavonic Studies 7 (1986): 65–84.
——. Russian Drama of the Revolutionary Period. London: Macmillan, 1990.
Samuels, Raphael, Ewan MacCall, and Stuart Cosgrove, eds. Theatres of the Left, 1880–1935: Worker's Theatre Movement in Britain and America. London: Routledge and Kegan Paul, 1985.
Sartorti, Rosalinda "Stalinism and Carnival: Organization and Aesthetics of Political Holidays." In Hans Gunther, ed., The Culture of the Stalin Period. New York: St. Martin's, 1990.
Savov, S. "Stanovlenie." In I. K. Sidorina and S. S. Sovetov, eds., Narodnye teatry: Sbornik statei. Moscow, 1962.
Schechner, Richard. Performance Theory. Rev. ed. New York: Routledge, 1988.
Schnapp, Jeffrey T. Staging Fascism: 18BL and the Theater of Masses for Masses. Stanford: Stanford University Press, 1996.
Schulte-Sasse, Linda. Entertaining the Third Reich. Durham, N.C.: Duke University Press, 1996.
Segel, Harold B. Twentieth-Century Russian Drama from Gorky to the Present. 2nd ed. Baltimore: Johns Hopkins University Press, 1993.
Senelick, Laurence. "Theatre." In Nicholas Rzhevsky, ed. Cambridge Companion to Modern Russian Culture. Cambridge: Cambridge University Press, 1998.
Shearer, David R. "The Language and Politics of Socialist Rationalization: Productivity, Industrial Relations, and the Social Origins of Stalinism at the End of NEP." Cahiers du Monde russe et soviétique 34, no. 4 (1991): 581–608.
Siegelbaum, Lewis. "The Shaping of Workers' Leisure." International Labor and Working Class History, Forthcoming.
——. Soviet State and Society between Revolutions, 1918–1929. Cambridge: Cambridge University Press, 1992.
——. Stakhanovism and the Politics of Productivity in the USSR, 1935–1941. Cambridge: Cambridge University Press, 1988.
Smith, Susanna Lockwood. "Soviet Arts Policy, Folk Music, and National Identity: The Piatnitskii State Russian Folk Choir, 1927–1945." Ph.D. dissertation, University of Minnesota, 1997.
Starr, S. Frederick. Melnikov: Solo Architect in a Mass Society. Princeton: Princeton University Press, 1978.
——. Red and Hot: The Fate of Jazz in the Soviet Union. New York: Limelight Editions, 1985.
Stebbins, Robert A. Amateurs: On the Margin between Work and Leisure. Beverly Hills: Sage Publications, 1979.
Steinberg, Mark D. Moral Communities: The Culture of Class Relations in the Russian Printing Industry, 1867–1907. Berkeley: University of California Press, 1992.
Stepanov, Z. V. Kul'turnaia zhizn Leningrada 20-kh, nachala 30-kh godov. Leningrad: Nauka, 1976.
Stephan, Halina. "Lef" and the Left Front of the Arts. Munich: Verlag Otto Sagner, 1981.
Stites, Richard. Revolutionary Dreams: Utopian Vision and Experimental Life in the Russian Revolution. New York: Oxford, 1989.
——. Russian Popular Culture: Entertainment and Society since 1900. Cambridge: Cambridge University Press, 1992.
Stourac, Richard, and Kathleen McCreery. Theatre as a Weapon: Workers' Theatre in the Soviet Union, Germany and Britain, 1917–1934. London: Routledge, 1986.
Suleiman, Susan Rubin. Authoritarian Fictions: The Ideological Novel as a Literary Genre. Princeton: Princeton University Press, 1983.
Swift, E. Anthony. "Fighting the Germs of Disorder: The Censorship of Russian Popular Theater, 1888–1917." Russian History 18, no. 1 (1991): 1–49.
——. "Theater for the People: The Politics of Popular Culture in Urban Russia, 1861–1917." Ph.D. dissertation, University of California, Berkeley, 1991.
——. "Workers' Theater and 'Proletarian Culture' in Pre-Revolutionary Russia." Russian History 23 (1996): 67–94.
Tamashin, L. N. Sovetskaia dramaturgiia v gody grazhdanskoi voiny. Moscow: Iskusstvo, 1961.
——. Vladimir Kirshon: Ocherk tvorchestva. Moscow: Sovetskii pisatel', 1965.
Thorpe, Richard G. "Academic Art in Revolutionary Russia." Unpublished manuscript.
Thurston, Gary. "The Impact of Russian Popular Theatre, 1886–1915." Journal of Modern History 55 (June 1983): 237–67.
——. The Popular Theatre Movement in Russia, 1862–1919. Evanston, Ill.: Northwestern University Press, 1998.
Timasheff, Nicholas. The Great Retreat: The Growth and Decline of Communism in Russia. New York: Dutton, 1946.
Turner, Victor. Dramas, Fields, and Metaphors. Ithaca: Cornell University Press, 1974.
Uvarova, E D. Estradnyi teatr: Miniatiury, obozreniia, miuzik-kholly, 1917–1945. Moscow: Iskusstvo, 1983.
——, ed. Russkaia sovetskaia estrada, 1930–1945. Moscow: Iskusstvo, 1977.
Vaksberg, Arkady. The Prosecutor and the Prey: Vyshinsky and the 1930s' Moscow Show Trials. Trans. Jan Butler. London: Weidenfeld and Nicolson, 1990.
Van Erven, Eugene. Radical People's Theatre. Bloomington: Indiana University Press, 1988.
Von Geldern, James. Bolshevik Festivals, 1917–1920. Berkeley: University of California Press, 1993.
——. "Cultural and Social Geography in the Mass Culture of the 1930s." In New Directions in Soviet History, ed. Stephen White. Cambridge: Cambridge University Press, 1992.
——. "Nietzschean Leaders and Followers in Soviet Mass Theater, 1917–1927." In Bernice Glatzer Rosenthal, ed., Nietzsche and Soviet Culture: Ally and Adversary. Cambridge: Cambridge University Press, 1994.
von Hagen, Mark. Soldiers in the Proletarian Dictatorship. Ithaca: Cornell University Press, 1990.
Ward, Chris. Russia's Cotton Workers and the New Economic Policy: Shop Floor Culture and State Policy, 1921–1929. Cambridge: Cambridge University Press, 1990.
Wardetsky, Jutta. Theaterpolitik in faschistischen Deutschland: Studien und Doku-mente. Berlin: Henschelverlag, 1983.
Warner, Elizabeth. The Russian Folk Theatre. The Hague: Mouton, 1977.
White, Anne. Destalinization and the House of Culture. London: Routledge, 1990.
Wood, Elizabeth. Performing Justice in Revolutionary Russia: Agitation Trials, Society, and the State. Berkeley: University of California Press, forthcoming.
Youngblood, Denise. Movies for the Masses: Popular Cinema and Soviet Society in the 1920s. Cambridge: Cambridge University Press, 1992.
Zernitskaia, E. I., E. D. Loidina, and N. V. Skashkova. Molodezhnyi teatr v sssR, v. 1. Moscow: Gosudarstvennaia tsentral'naia teatral'naia biblioteka, 1968.
Zimmermann, Patricia R. Reel Families: A Social History of Amateur Film. Bloomington: Indiana University Press, 1995.
Zograf, N. G., Iu. S. Kalashnikov, P. A. Markov, and V. I. Rostotskii, eds. Ocherki istorii russkogo sovetskogo dramaticheskogo teatra v trekh tomakh. Moscow: Akademiia Nauk, 1954–1960.
Zolotnitskii, D. Sergei Radlov: The Shakespearean Fate of a Soviet Director. Luxembourg: Harwood Academic Publishers, 1995.
——. "Teatry revoliutsionnoi satiry." In Teatr i dramaturgiia: Trudy Leningradskogo gosudarstvennogo instituta teatra, muzyki i kinematografii. Leningrad: Iskusstvo, 1967.
——. Zori teatral 'nogo Oktiabria. Leningrad: Iskusstvo, 1976.
## Index
"Action circles," –, ,
Actors
activists vs.,
professional, ,
spectators and, –
terms for, –
Aesthetic drama,
agitprop brigades and,
social vs., –,
Aesthetics
agitprop, –,
bourgeois,
politics and, , –, –
Afinogenov, Alexander,
Agitational plays
agitprop brigades and, ,
for clubs,
cultural demands on,
endorsement of, –,
Evreinov on, –
goals of,
living newspapers and,
poster art and,
propaganda and,
Sokolovskii and, –
themes of, –
trade unions and,
tram and, ,
Zadykhin's, –
Agitki. See Agitational plays
Agitprop brigades, –, –
aesthetics of, –
amateur theater and, –,
audiences of, –
avant-gardists and,
Avlov on, –, ,
black listing by, –,
Blue Blouses and, –,
Communist Party and, –
criticism of, –
decline of, –, –
factory drama circles and,
formation of, –
methods of, –, –
origin of, –, –
performance spaces for, , –
professionalism and, , –,
purpose of,
repertoire of, , –
shaming rituals by, –
as shock workers, –
small forms and,
tram and, –
women in,
Agit-trials, –, –
clubplays and,
improvisations with, –
origin of, –
real conflicts as,
as small form,
about women, –
Alarm (Knorre),
Alcohol
agitprop brigades and,
agit-trial about,
in tram plays,
Alekseev, Sergei, , –,
Alienation, Brecht on,
Amateurism
assessment of, –, , –
capitalist vs. Soviet, –
effacement of, –
terms for, –, ,
Amateur theater
agitprop brigades and, –,
agit-trials as,
assessment of, –, , –
audiences of, –,
avant-garde and, ,
critics of, –,
cultural debates about, –13
cultural development from, –
cultural literacy and,
dilettantism of,
diversity of,
Evreinov on, –
funding of, –
goals of, ,
Gvozdev on,
improvisations by, –
journals for,
Kerzhentsev on,
Meyerhold and, –
New Economic Policy and,
oversight of, , –, –
bureaucratic, , –,
political, , –, –
Red Army, –
performance spaces for, , , –
Piotrovskii on,
prerevolutionary origins of, –8
professional vs., –, , –,
repertoire of, , –, , –,
"revolutionary face" of,
self-determination of, –
small forms in,
Social Democrats and,
Soviet influence on,
subversion and, ,
trade union support of, , ,
traditions of, –8
tram and, –, –
See also Small forms theater
Andreev-Bashinskii, B.,
Anti-formalist campaign, –, –
Arbuzov, Aleksei,
The Long Road of, ,
Six Lovers of,
Architecture
club, , –, –
constructivist, ,
Stalinist, –
Argo (Abram Markovich Gol'denberg),
Aristocrats (Pogodin), , –
Arskii, Pavel, –
Arvatov, Boris,
Audience, –,
agitprop, –
waron, –
Autonomy
amateur theater and, –
worker, –
Avant-garde
agitprop brigades and,
amateur art and, , , –
filmmakers and,
living newspapers and,
Meyerhold on, –,
realism vs., ,
small forms and, –
socialist realism and,
tram and,
Avlov, Grigorii, ,
on agitprop, –, ,
on club theater,
tram and,
"Trial of Hooligans" of,
Azarkh, A. S., –
Babel, Isaac,
Baltflot Theater,
Bednyi, Demian,
Belaia, Sofia,
Benjamin, Walter,
Beskin, Emil,
Bessal'ko, Pavel,
Bezymenskii, Alexander, ,
The Shot of, ,
Bill-Belotserkovskii, Vladimir,
Storm of, ,
Black listing, –,
Bliumenfel'd, Valerii, , ,
Blok, Alexander,
Blue Blouse (journal), , , ,
Blue Blouse (Moscow troupe), –
agitprop brigades and, –,
avant-garde and,
criticisms of, –
emulation of, –
manifestos of,
name for, –
performance spaces for,
professionalism of,
"Blue Blouseism," –, –
Blue Blouses (Chicago troupe),
The Boat, , ,
Boiarskii, I.,
Bol'shoi Drama Theater (Leningrad),
Bol'shoi Theater (Moscow),
Bonnell, Victoria,
Bourdieu, Pierre,
Bourgeois in Hell,
Brecht, Bertolt, ,
Bright Stream (Shostakovich),
Brik, Osip, ,
Bronikovskii, G.,
Brooks, Peter,
Call the Factory Committee, , ,
Carter, Huntley,
Central Agitational Studio, Petrograd Politprosvet's, –,
Central Division for Political Education, –,
Chekhov, Anton, , , , ,
government view of, ,
popularity of, ,
Chicherov, Ivan, , –,
Chuzhak, Nikolai,
Circus,
Clark, Katerina, ,
on agitprop brigades,
on New Economic Policy,
on united artistic circles,
Clowns,
Club Methodological Laboratory, , –
Club Stage (journal), , ,
Club theaters, –, –
agit-trials and,
architecture of, , –, –, –
Avlov on,
Evreinov on, –
festivals and, , –,
living newspapers and, –
methodology of,
Meyerhold and, –
plays of, –
professional writers for, –
propaganda and, –
repertoire of, –, , –
small forms in, –
training for, –
united artistic circles and,
young people and, –
See also Amateur theater
Coal,
Commissariat of Education. See Narkompros
Communist Party
agitprop brigades and, –
on artistic organizations, –
cultural politics and, –, –, –
German,
journal sponsorship by,
Komsomol and, ,
New Economic Policy and,
Proletkult and,
purges in, –
shock workers and,
Sokol district,
The Confrontation (Sheinin), –
Consciousness, spontaneity and, –
Constructivist architecture, ,
Consumption, materialist, –
"Crazy Sashka" (character), ,
Crazy Sashka (play), –,
Creative Theater (Kerzhentsev), , ,
Cultural revolution
club art and,
of First Five-Year Plan, –
Culture
commercial, –
Marx on, –
politics and, –, –, –, –
self-education and,
transformation of, –, –
See also Youth culture
Daily life
agitprop scripts of, –
capitalism and,
investigations of, –
public sphere in, ,
tram and,
David-Fox, Michael, –,
"Dawn of the Proletkult" (Ignatov),
The Days Are Melting, , ,
Depression, Great, –, , –
Derzhavin, Vitaly,
The Devil's Wheel (film),
Didactic art, –,
Dilettantism,
"Do-it-yourself" theater, –, , –
agitprop brigades and, –
concept of, –
origin of, –
small forms and,
Drama
aesthetic,
social vs., –,
classic, –, –,
Greek,
new vs. old, –
social, , –,
agitprop brigades and,
Turner on, –
traditional, –
small forms vs., –
See also Melodrama
"Dramatism,"
"The Dream of the Illiterate Red Army Soldier,"
Dvorikov, Nikolai,
The Earth in Flames (Ivanter),
Efremov, Andrei,
Ehrenburg, Ilia,
Ekk, Nikolai, ,
The Red Eagles of,
The End of Krivoryl'sk (Romashov),
Engels, Friedrich, –
Entertainment
enlightenment vs., , ,
living newspapers as, –
Erdman, Nikolai,
The Mandate of,
Esslin, Martin,
Evgenii Onegin (Pushkin),
Evreinov, Nikolai, –
Experts, debate about, , , –
Face the Countryside,
Face to Production,
Factory art circles,
Factory drama circles, ,
Factory Love,
Factory Storm (Tolmachev),
The Fall of Pompeii,
A Family Affair (Ostrovsky),
Family drama, –
Federal Theatre Project (U.S.),
Fedka Esaul (Romashov),
Festivals
agit-trials and, –
architecture for, –
calendar for,
club theaters and, ,
improvisations at, –
instsenirovka for, –, –
Komsomol, –
mass spectacles for, , –
May Day, , –
Red Army, –,
small forms and, –
"Ten Octobers,"
See also Spectacles
Films
avant-garde,
oversight of,
Piotrovskii on,
socialist model for,
Sovkino and,
theater vs.,
tram and,
Western, ,
The Devil's Wheel,
The House on Trubnaia Square, ,
Kuban Cossacks,
The Road to Life,
Tractor Drivers,
Volga, Volga, –
Finn, Konstantin,
Fitzpatrick, Sheila,
Five-Year Plan
First, –,
agitprop brigades and, –
art during,
cultural revolution of, –
shock workers for,
social rupture and, –
tram and, –, –
villains of,
Second, –
"Five Years of the Komsomol,"
Folk theater, , , –
living newspapers and, –
Fonvizin, Denis,
Formalism, –, –
For the Red Soviets, –
France, Anatole,
Freud, Sigmund,
Funding sources, –,
Gaideburov, Pavel, , –,
Traveling Popular Theater of,
Galaktionova, Olga,
Gastev, Aleksei,
Ginzburg,
Girls of Our Country (Mikitenko), , ,
"Give Us a New Life,"
Glavpolitprosvet (Central Division for Political Education), –,
on agitational plays,
living newspapers and,
Glebov, Anatolii, , –,
Gleron House, –,
The Godless (journal),
Gogol, Nikolai, ,
popularity of,
Inspector General of, , ,
The Wedding of,
Gol'denberg, Abram Markovich (Argo),
Goldoni, Carlo, ,
Gossiping Women of,
Gorbenko, Arkadii, , ,
Gorky, Maxim, , ,
on revolutionary drama,
The Lower Depths of,
The Philistines of,
Gorsuch, Anne,
Greek drama,
Groys, Boris,
Gvozdev, Alexei,
Hands off China, –
Happy Cohort,
Happy Hillock,
Hatch, John,
Hauptmann, Gerhart, ,
"Homemade" theater. See "Do-it-yourself" theater
Home of Amateur Theater,
Hooligan(s),
agit-trials of, –
in Call the Factory Committee,
"Crazy Sashka" as,
in Factory Storm,
in Work Days, –
young people and,
youth culture and,
Hooligan (Zadykhin), –
The House on Trubnaia Square (film), ,
How Tom Gained Knowledge,
Hugo, Victor,
"Hymn to Labor,"
Iakovlev Club, ,
Ianov, Roman, ,
Ideology
small forms and, –
theater and,
Ignatov, Vasilii,
Improvisations, –
with agit-trials, –
criticism of,
decline of,
at festival celebrations, –
small forms and, –
The Infant (Fonvizin),
Inspector General (Gogol), , ,
Instsenirovka
club plays and, ,
for festivals, –, –
political purges and,
as small form, ,
An Intellectual Marriage, –
The Iron Stream,
Isaev, I.,
Ispolnev, I.,
Iurtsev, Boris,
Iutkevich, Sergei,
Ivanov, Viacheslav,
Ivanter, Boris,
Jelagin, Juri,
Kalinin Club,
Kasatkina, A., –
Kashevnik, Mura,
Kauchuk Club, , , –
Kerenskii, Alexander,
Kerzhentsev, Platon,
Shostakovich and, –
Creative Theater of, , ,
Kirshon, Vladimir, , –,
Knorin, Vilis,
Knorre, Fedor,
Alarm of,
Kobrin, Iurii,
Kogan, Piotr, ,
Komsomol
Communist Party and, ,
festivals of, , –
First Five-Year Plan and, –
journals of, ,
Leningrad, ,
reorganization of, –
tram and, –
marriage and, –,
newspaper of, ,
1932 Olympiad and,
theater support by,
tram and, ,
women in, –
youth culture and, , –
Kotkin, Stephen,
Kozintsev, Grigorii,
Kuban Cossacks (film),
Kukhmisterov Club, , ,
Kul'turnost',
Kuza, Vasilii,
Lady Macbeth of the Mtsensk District (Shostakovich),
Left Front of the Arts (LEF),
Lenin, V. I., ,
Leningrad Drama Theater,
Leningrad Institute of Theater, Music, and Cinema,
Leningrad Theater of Working-Class Youth. See tram
Lenin Workers' Palace, – Lermontov, Mikhail,
Levman, S.,
Liebknecht, Karl,
Life of Art (journal), ,
Ligovskii People's Home, ,
Lissitzky, El, ,
Literacy
cultural,
political,
"Literature of fact,"
Litfront (writers' group), –,
Liubitel'stvo ("amateurism"), ,
Liubov' Iarovaia (Trenev), –, ,
Living newspapers, –
avant-garde and,
club plays and, ,
criticisms of, –,
decline of,
as entertainment, –
folk theater and, –
in late 1920s,
names of, –
pornography and,
Red Army and, –
as small form, , ,
tram and, ,
Zhemchuzhnyi on, –
The Loan Campaign in the Second Five-Year Plan,
London, Jack,
The Long Road (Arbuzov), ,
The Lower Depths (Gorky),
Lunacharskii, Anatolii,
on classic dramas, –
on revolutionary dramas,
on tram, –,
Luxemburg, Rosa,
L'vov, Nikolai (tram participant),
L'vov, Nikolai (theater activist), ,
agitprop brigades and, ,
on classic dramas, –
on revolutionary theater,
at Tirniriazev Club,
Maiakovskii, Vladimir, ,
Maksimenkov, Leonid, –
Malyi Theater, ,
The Mandate (Erdman),
Man in a Red Wig,
Marchenko, Dmitrii,
Marinchik, Pavel, , ,
Meshchanka of, –
Markov, Pavel,
Marriage
agit-trial about,
tram plays about, , ,
The Marriage of Figaro,
Marx, Karl, –
Mass events, –, , –,
Red Army, –,
"Massism," , . See also Spectacles
Materialist consumption, – May Day celebrations, , –, . See also Festivals
Mel'nikov, Konstantin (architect), , –,
Melodrama,
Brooks on,
propaganda and,
"revolutionary,"
See also Drama
Meshchanka (Marinchik), –
Meyerhold, Vsevolod, ,
amateur theater and, –
avant-garde and, –,
criticism of, –
productions of,
tram and, ,
Zhemchuzhnyi and,
Meyerhold Theater, , ,
The Meyerhold Theater and the Worker Viewer (pamphlet),
Mikitenko, Ivan, , ,
Mime,
The Miraculous Alloy (Kirshon), , –,
Mokulskii, Stefan, –
Molière, , , ,
Mologin, Nikolai,
Moscow Art Theater (MAT), , ,
critique of, , , ,
First Studio of,
Komsomol and,
Narkompros and,
tram and, ,
Moscow House of Amateur
Moscow Perfume Factory, –
Moscow Politprosvet, –
Moscow Rusakov brigade, ,
Mostovoi, Iakov,
Music halls,
Naiman, Eric, ,
Narkompros (Commissariat of Education),
on agitprop brigades,
drama prizes of,
Moscow Theater and,
1932 Olympiad and, ,
theater oversight by, , ,
theater support by,
tram and, –
worker-peasant theater of, –
National Committee of the Arts, , , ,
National Olympiad of Amateur Art, , –
Naturalism, –, . See also Realism
Nekrasov People's Home, , ,
Neuberger, Joan,
New Economic Policy (NEP), –,
Clark on,
commercialism of, –
popular culture and,
shock work in,
small forms and, ,
teachers and,
tram and, –
worker-peasant union and,
Newspapers. See Living newspapers
The New Viewer (journal), , ,
Nietzsche, Friedrich,
Not a Cent and Suddenly a Windfall (Ostrovsky),
Novyi lef (journal),
Okhlopkov, Nikolai,
Olympiad(s)
of Amateur Art, , –
trade union, –
Ostrovsky, Alexander, , , , ,
government view of, –
popularity of, , –
works of, , ,
Our Daily Life,
Oversight agencies, –, , –, , –
The Overthrow of the Autocracy,
Panopticon, family as,
Papernyi, Vladimir,
"Parallelism," ,
"Path to Victory,"
Pel'she, Robert
on agitational plays, ,
on classic drama,
on professional art,
on tram, –
People's Homes, , ,
Performance art,
Performance space
for agitprop brigades, , –
for amateur theater, , , –
for Blue Blouse,
Petrograd Politprosvet studio, –,
Petrushka plays, ,
The Philistines (Gorky),
Pikel, Richard,
Piotrovskii, Adrian, , ,
on amateur theater, –, ,
arrest of,
on classic dramas,
criticisms of, –, , , –
on "denial of illusion,"
on techniques,
methods of, ,
on tram, , , –,
on united artistic circles, ,
Rule Britannia of,
Playwriting contests, ,
Podvoisky, Nikolai,
Pogodin, Nikolai, , –
Polenov, Vasilii,
"The Political Trial of the Bible,"
Politics, cultural, –, –, –, –
Politprosvet
Leningrad,
Moscow, –
Petrograd, –, ,
Popular theater. See under Theater
Pornography, living newspapers and,
Poster art,
Private life. See Daily life
Professionalism
agitprop brigades and, , –,
amateurism vs., –, , –, , –
of Blue Blouse troupe,
club plays and, –
expertise and, , , –
Pel'she on,
reappraisal of, –
small forms theater and,
of tram, –,
Proletarian theater, ,
Proletarian Writers' Association (rapp), –, ,
Proletkult, –
clubs of,
Communist Party and,
drama prizes of,
improvisations and, –
theaters of,
tram and, –
Propaganda
agitation vs.,
See also Agitprop brigades
Public sphere, ,
Punch and Judy shows. See Petrushka plays
PUR (Politicheskoe upravlenie voennogo revoliutsionogo soveta),
"A Purpose for Icons," –
Pushkin, Aleksander, ,
Pyr'ev, Ivan,
Radek, Karl,
Radlov, Sergei, ,
Railroad Workers' Union, –, ,
agitprop brigades and,
RAPP. See Proletarian Writers' Association
Rasputin, Grigory,
Realism
agitprop brigades and, –
avant-garde vs., ,
didactic art of, –,
naturalism vs., –,
socialist, , , ,
Red Army
agitational plays and, –
agit-trials and, –
improvisations and, –
spectacles of, –,
Terevsat of, –
theater oversight by, –
theater support by, –,
Red Army Studio, –, –
Red Calendar,
Red Cavalry (Babel),
"Red Coil" (living newspaper),
The Red Eagles (Ekk),
Red October Club, ,
Red Rockets (Berlin theater group),
"Red Scourge" (living newspaper), ,
Red Sowing,
"Red Sting" (living newspaper),
Red Theater, –
"Red Tie" (living newspaper), –
Red Woodworkers' Club, , ,
Reed, John,
Religious communities, ,
Repertoires
agitprop brigade, , –
club theater, –,
debate on, –, –
politicized,
reshaping of, –,
small form,
tram, –, , , –
Repertory Bulletin,
Repertory Guide,
Revolutionary theater
competitions for, ,
critics of, –
The Road to Life (film),
Robert Tim (Afinogenov),
Rodchenko, Alexander,
Rolland, Romain, ,
Romashov, Boris,
The Roof,
ROSTA (news agency),
Rotten Thread (Severnyi),
Rowdy Cohort,
RT (play),
Rule Britannia (Piotrovskii),
Rusakov Club, , , , ,
Rusalka (opera), ,
Ryzhei, Piotr,
Samodeiatel'nost', , –, –
Samodeiatel'nyi teatr, –, –
concept of, –
See also "Do-it-yourself" theater
Samosud (mock trial),
Sapronov Club, ,
Schechner, Richard,
Schiller, Friedrich,
Seniushkin, F.,
Serafimovich, Alexander,
Severnyi, Vladimir, , ,
Rotten Thread of,
Shakespeare, William, ,
Hamlet of,
Merry Wives of, Windsor of,
Taming of, the Shrew of,
Shakhty Trial,
Shaming rituals, –
Shcheglov, Dmitri, –, ,
club plays of,
as psychological realist,
Sheinin, Lev, –
Shibaev, Alexander,
Shimanovskii, V. V., –
Shklovskii, Viktor, –
Shock workers, –
agitprop brigades as, –
in 1932 Olympiad, –
Shostakovich, Dmitri, –
Bright Stream of,
Lady Macbeth of the Mtsensk District of,
The Shot (Bezymenskii), ,
Shvernik, Nikolai, –
Sinclair, Upton,
Six Lovers (Arbuzov),
Small forms theater,
agitprop brigades and, ,
avant-garde and, –
club plays as,
criticism of, –,
decline of, –
definition of,
"dramatism" vs.,
for festivals, –
ideology and, –
Politprosvet on, –
professionalism and,
spontaneity and, –
stereotypes of, ,
tram and, ,
turn to, –
youth and, –
See also Club theaters
Smychka ("union"),
Smyshliaev, Valentin,
Social Democrats,
Social drama, , –,
agitprop brigades and,
Turner on, –
Socialist realism, , , , . See also Realism
Socialist theater,
Sokolovskii, Mikhail, –,
acting methods of, –,
criticism of,
Meyerhold and,
on professionalism,
on Stanislavsky method,
Unbroken Stream of, –
Soloveva, Klava,
Soviet Art (journal), , , –
Soviet Writers' Union,
Sovkino,
"The Spartacus Rebellion" (spectacle),
Spectacles, –
architecture and,
festival, , –
Red Army, –,
"The Spartacus Rebellion,"
"The Storming of the Winter Palace,"
"Ten Octobers,"
"Spectacle state," –. See also Festivals; Mass events
Spectators, actors and, –
Spontaneity
consciousness and, –
term for,
See also Improvisations
Stalin, Joseph, , ,
agitprop brigades and,
cultural politics of, , –
purges by, –,
rise of, –
Stanislavsky, Konstantin
acting techniques of, , , ,
critique of, , , ,
Stereotypes
of small forms, ,
in tram plays,
Stites, Richard,
Storm (Shtorm) (Bill-Belotserkovskii), ,
The Storm (Groza) (Ostrovskii),
"The Storming of the Winter Palace" (spectacle),
The Story of the Iron Battalion,
Strazd, Ekaterina, –
Subbotin, Leonid, ,
Sudakov, Il'ia, , ,
Suleiman, Susan,
Sventitskii, A. V.,
Sverdlov Club,
Swift, Anthony,
Syphilis, agit-trial about, –
The Taking of the Bastille (Rolland), ,
Talent, Marx on, –
"Teachers during NEP" (script),
Ten Days That Shook the World (Reed),
"Ten Octobers" (spectacle),
The 10:10 Moscow Train, –
Terent'ev, Igor,
Terevsat (Theater of Revolutionary Satire), –
Theater
agitational, , . See also Agitational plays
"bourgeois,"
Christian,
experimental. See Avant-garde
fairground,
films vs.,
folk, , , –
living newspapers and, –
journals on, –
popular, –,
living newspapers and,
terms for, –, –,
tradition of, –8
in public sphere, ,
revolutionary, –,
as "self-educator,"
utilitarian,
worker-peasant, –, , –
workers', , –
See also Amateur theater
Theater of Revolutionary Satire, –
Theatrical-Artistic Bureau,
Third International Club, –
The Thoughtful Dandy, , , ,
criticism of, –
Tikhonovich, Valentin, , ,
Narodnyi teatr of, ,
Timiriazev Club, –
Tolmachev, Dmitri, ,
Tolstoi, Lev, ,
Tomskii, Mikhail, , –,
Tractor Drivers (film),
Trade unions
agitational plays and, ,
cultural funding by, , , , –
Olympiads of, , –
shock workers and,
theater oversight by, –
tram (Theater of Working-Class Youth), , , –
acting method of, –
agitational plays and,
agitprop brigades and, –
alliances of, –
avant garde and,
Brecht and,
criticisms of, –, –
decline of, –
family drama and, –
First Five-Year Plan and, –, –
funding of, ,
at Gleron House, –,
Komsomol and, ,
living newspapers of, ,
Lunacharskii on, –,
Meyerhold and, ,
multimedia influence of, –
as national organization, –
in 1932 Olympiad, ,
origins of, –
Pel'she on, –
professionalism of, –,
Proletarian Writers' Association and, –
repertoire of, , , , –
rise of, –
training for,
women in, –
youth culture and, –
Crazy Sashka of, –,
Factory Storm of,
Girls of Our Country of, ,
Meshchanka of, –
The Thoughtful Dandy of, , , , , –
Tsar Maksimilian of,
WorkDays of, –
"tramism," , –
Trauberg, Leonid,
Trenev, Konstantin, –, ,
Tretiakov, Sergei, , , ,
at Blue Blouse,
"Trial of a Midwife Who Performed an Illegal Abortion,"
"Trial of Bourgeois Marriage,"
"Trial of Lenin,"
"Trial of the Old Life,"
Trotsky, Leon,
political purges and, –,
tram play about,
young supporters of, –
The Trust D.E., –
Tsar Maksimilian, ,
tram production of,
Tsiurupy Club,
Tubelskii, Leonid,
Turner, Victor, –
Tverskoi, Konstantin, –
Two Orphans,
Unbroken Stream (Sokolovskii), –
The Unemployed (Belaia),
Union, worker-peasant, . See also Trade unions
United artistic circles, –
Agitprop criticism of,
Clark on,
club plays and,
Piotrovskii on, ,
Shcheglov and,
"United studio of the arts,"
"Upruiaz" (Mologin),
Utilitarian theater,
Vakhtangov Theater, , ,
Van Gyseghem, André, ,
Vaudeville,
Vega, Lope de, , ,
Venereal disease, agit-trial about, –
The Vengeance of Fate,
Verbitskaia, Anastasia,
Verhaeren, Emile,
The Victors Will Judge (Gorbenko),
Virgin Soil, ,
Volga, Volga (film), –
Von Geldern James, , –
Vsevolodskii-Gerngross, Vsevolod,
Vyborg House of Culture,
Vyshinskii, Andrei, –
We Are from Olonets, –
"We Are the Blue Blousists" (song),
The Weavers (Hauptmann), ,
The Wedding (Gogol),
White Sea Canal,
Whitman, Walt,
Winter, Ella,
Women
in agitprop brigades,
agit-trials about, –
in Komsomol, –
during Second Five-Year Plan,
in tram, –
"Work Bench" (living newspaper), –
Work Days, –
Worker and Theater (journal),
on Crazy Sashka,
on living newspapers,
on Meshchanka, –
on 1932 Olympiad,
Worker-peasant theater, –, , –
Worker-reporters, –, ,
Workers' Club (journal), , ,
"Workers' palace," ,
Workers' theater, , –
The Worker Viewer (journal), ,
Youngblood, Denise, ,
Young Guard (periodical),
Youth culture
club system and, –,
films in, ,
First Five-Year Plan and, –
hooligans and,
Komsomol and, –
politicized,
small forms and, –
tram and, –
Trotsky and, –
Zadykhin, Iakov,
style of, –
Hooligan of, –
Zhemchuzhnyi, Vitalii, ,
on action circles,
on living newspapers, –
Meyerhold and,
on performance art,
"An Evening of Books" of, ,
Zhizn' iskusstva (journal),
Zimmermann, Patricia,
Zonov, Alexander,
Zor'ka,
Copyright © 2000 by Cornell University
All rights reserved. Except for brief quotations in a review, this book, or parts thereof, must not be reproduced in any form without permission in writing from the publisher. For information, address Cornell University Press, Sage House, 512 East State Street, Ithaca, New York 14850.
E-book edition 2016 by Cornell University Press
ISBN 978-1-5017-0697-4
Visit our website at www.cornellpress.cornell.edu.
The text of this book is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License: <https://creativecommons.org/licenses/by-nc-nd/4.0/>
|
{
"redpajama_set_name": "RedPajamaBook"
}
| 9,475
|
Prouds the Jewellers has been part of Australia's history over 110 years and their buying power means they can provide the best quality of jewellery. Prouds offers a complete range of gold and silver jewellery, watches, giftware for any occasion and is the largest jeweller in Australia. Prouds features a stunning collection of diamond jewellery and new season styles. Prouds have qualified staff to assist with al your needs and is the trusted name for jewellery.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,493
|
package relay
import (
"log"
"net/http"
"time"
)
// LoggerHandler creates a new FlatHandler using a giving log instance
func LoggerHandler() FlatHandler {
return func(c *Context, next NextHandler) {
start := time.Now()
req := c.Req
res := c.Res
addr := req.Header.Get("X-Real-IP")
if addr == "" {
addr = req.Header.Get("X-Forwarded-For")
if addr == "" {
addr = req.RemoteAddr
}
}
c.Log.Printf("Started %s %s for %s", req.Method, req.URL.Path, addr)
rw := res.(ResponseWriter)
next(c)
c.Log.Printf("Completed %s(::%s) -> %v %s in %v @ %s\n", req.URL.Path, req.Method, rw.Status(), http.StatusText(rw.Status()), time.Since(start), addr)
}
}
// Logger returns a new logger chain for logger incoming requests using a custom logger
func Logger(log *log.Logger) FlatChains {
return NewFlatChain(LoggerHandler(), log)
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 6,922
|
{"url":"https:\/\/www.rdocumentation.org\/packages\/R.utils\/versions\/2.10.1\/topics\/isSingle","text":"isSingle\n\n0th\n\nPercentile\n\nIdentifies all entries that exists exactly once\n\nIdentifies all entries that exists exactly once.\n\nUsage\nisSingle(x, ...)\nsingles(x, ...)\nArguments\nx\n\nA vector of length K.\n\n...\n\nAdditional arguments passed to isReplicated().\n\nValue\n\nA logical vector of length K, indicating whether the value is unique or not.\n\nInternally isReplicated() is used.","date":"2021-02-28 13:28:56","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\": 0.7342298626899719, \"perplexity\": 9014.173821230768}, \"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-2021-10\/segments\/1614178360853.31\/warc\/CC-MAIN-20210228115201-20210228145201-00398.warc.gz\"}"}
| null | null |
Q: Maximize $f(x) = x^3-3x$ subject to constraints I would like to understand more about how to maximise functions of one variable subject to constraints.
How can you find the maximum value of $f(x) = x^3 - 3x$ subject to $x^4+36 \leq 13x^2$?
The answer is apparently $f(x) = 18$ at $x=3$, but if this is true, how can you derive it without a computer?
A: The extreme value for a differentiable function can either come at a point where the derivative is zero or at the end of the interval of definition. You can take the derivative, getting $f'(x)=3x^2-3$ and find it is zero at $x = \pm 1$, then check those points and find $f(-1)=2, f(1)=-2$. Unfortunately, as we will see, these violate the constraint. Then you need to check the ends of the interval of definition. To have $x^4+36 \le 13x^2$ we need $(x-3)(x-2)(x+2)(x+3) \le 0$ so the function is defined on $[2,3]$ and $[-3,-2]$ You need to evaluate $f(x)$ at those four endpoints and take the greatest.
A: Hint: $x^4 - 13x^2 + 36 = (x^2 - 9)(x^2 - 4)$
Use this and the fact extrema for continuous functions defined on a closed interval occur at critical points or on the boundary of the interval
A: So the difficult part is the constrain:
$x^4-13x^2+36\leq0\iff(x^2-4)(x^2-9)\leq0\iff(x-2)(x+2)(x-3)(x+3)\leq0\ldots$
A: There are two parts to this.
The first part would be to rearrange the constraint equation to be $x^4-13x^2+36 \leq 0$. By using derivitves, we can find the zeroes of this function, and they occur at $x = \pm 2$ and $x = \pm 3$. Also by using derivatives, we can find the function is decreasing on the intervals $(-\infty, -\sqrt{\frac{13}{2}})$ and $(0, \sqrt{\frac{13}{2}})$ and increasing on the intervals $(-\sqrt{\frac{13}{2}},0)$ and $(\sqrt{\frac{13}{2}},\infty)$. Thus we know that the function satisfies the constraint on the intervals $[-3,-2]$ and $[2,3]$.
The second part is simply finding the maximum values of the function on these intervals, which you can do with derivatives, which you stated you know how to calculate. See here if you don't already know how to do this.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 1,056
|
Christian Alexander Oedtl (1655 Paznaun, Tyrolsko – 6. ledna 1731 Vídeň) byl rakouský barokní architekt a stavitel. Byl činný převážně v Dolních Rakousích a na Moravě.
Život
Oedtl působil ve Vídni jako stavitel pro císařský dvůr. V roce 1699 se zapojil do návrhu vídeňského paláce Questenberg-Kaunitzů. Počátkem 18. století byl jedním z největších vídeňských stavitelů a realizoval projekty architektů Johanna Bernharda Fischera z Erlachu a Johanna Lucase von Hildebrandt. Později Oedtl začal vytvářet i vlastní návrhy.
Na přání knížete Ditrichštejna vytvořil návrh přestavby mikulovského zámku, roku 1719 poničeného požárem. Posléze vybudoval nové třípodlažní jižní a západní křídlo zámku. Podle projektu Matthiase Steinla postavil Oedtl vídeňský Alserkirche. Vedl též stavbu farního kostela v Laxenburgu, podle návrhu Johanna Wiesera. Podle posledních poznatků je Oedtl považován za architekta vídeňského paláce Windisch-Graetzů, vybudovaného roku 1703.
Podle Oedtlových návrhů proběhla mezi lety 1750 až 1760 přestavba zámku Prštice. Spolu s Jakobem Prandtauerem vytvořil Oedtl plány pro Prandtauer-Hof ve Vídni.
Oedtl zanechal katalog svých realizací z let 1683 až 1726, které jsou situovány převážně v Dolních Rakousích.
Dílo
od 1692: palác Harrachů ve Vídni
(připsáno) 1702–1703: palác Windisch-Graetzů ve Vídni
od 1712: palác Trautsonů v Neubau, podle plánů Johanna Bernharda Fischera z Erlachu
1713–1720: přestavba zámku Marchegg
od 1714: zámek v Líšni
1717–1719: Geheime Hofkanzlei ve Vídni, podle plánů Johanna Lukase von Hildebrandt
1718: palác Batthyány ve Vídni
1719–1730: přestavba zámku Mikulov
od 1721: kostel Navštívení Panny Marie v Lechovicích
1722–1724: přestavba zámku Eckartsau a vybudování zámecké kaple
Reference
Literatura
Wilhelm Georg Rizzi: Der Tiroler Baumeister Christian Alexander Oedtl. In: Das fenster. Tiroler Kulturzeitschrift, číslo 28, 1981, 2821–2851
Christian Alexander Oedtl na Moravě. Nová připsání. 32, 1984, č. 3, s. 233–240
Walpurga Oppeker: Christian Alexander Oedtl und Franz Jänggl. Zwei Wiener bürgerliche Maurermeister an der Wende vom 17. zum 18. Jahrhundert. In: Jahrbuch des Vereins für Geschichte der Stadt Wien. 61. (2005), 99–152.
Externí odkazy
Rakouští stavitelé
Rakouští architekti
Barokní architekti
Narození v roce 1655
Úmrtí v roce 1731
Úmrtí ve Vídni
Muži
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 7,238
|
You're passionate about society, how we treat our vulnerable and needy citizens. You see a fulfilling role in a Social Care setting. Or you believe in the transformative power of early years education and quality childcare. You might be bringing your life experience to the Department of Social Science and Design as a mature learner, like 20% of our Social Care students who are 23 years old or older. You want the pioneering spirit of AIT, one of the leaders in the field of social care and social studies, with a track record of supporting research and progression to Masters-level studies and beyond. You will gain practical hands-on experience during extensive placements in residential agencies, youth-work and community-based projects.
Maybe your artistic flair is drawing you to a career in Design. You want to join our design graduates in the workforce in graphic design, advertising, publishing, film, TV or teaching. You know our design students have a reputation of success in local, national and international competitions. You want our guarantee of one-to-one attention from our lecturers working with you in our purpose-built environment, with excellent personal studio space and production facilities.
Or you see a career in teaching —graduates of our BA (Hons) in Design in Digital Media are eligible to apply for Postgraduate Diploma of Education programmes.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,822
|
Jenelle Kohanchuk (born October 3, 1990) was an ice hockey forward for the Boston University Terriers women's ice hockey program. She would compete for the Toronto Furies squad that captured the 2014 Clarkson Cup. She is a current player for Team Bauer (Boston) for the PWHPA (2021-22).
Playing career
Kohanchuk participated in the 2006 Mac's Midget Tournament with Balmoral Hall. The following year, she competed in the tournament with Notre Dame. She won a bronze medal with Manitoba at the 2007 National Women's Under-18 Championship. In the same year, she was part of the Female Midget AAA Hockey League championship with the Notre Dame Hounds. With Notre Dame, she was part of the gold medal victory at the 2007 Western Shield, where she was named top forward and MVP. At the 2007 Canada Winter Games in Whitehorse, Yukon, Kohanchuk won a silver medal with Team Manitoba. In 2007-08, Kohanchuk ranked second in scoring with the Balmoral Hall Blazers. In 2008, she won a gold medal at the Balmoral Hall tournament in 2008
NCAA
In 2008, she joined the Boston Terriers women's ice hockey program. She was named to the Hockey East all-rookie team and was named top rookie at the Hockey East championship in 2008-09. In the same season, she led the Terriers in scoring and was the co-leader among all Hockey East freshmen. The following season, she finished fifth on the Terriers scoring list in 2009-10, leading all sophomores. She was part of the Terriers squad that won its first-ever Hockey East championship in 2010. In addition, she participated with the Hockey East All-Stars in a game against the United States' Women's National Team on Nov. 22, 2009. She was in the Frozen Four in 2011, 2012, and 2013.
Hockey Canada
Kohanchuk's first experience with Hockey Canada came when she attended Canada's National Women's Under-22 Team selection camp in Toronto, Ont., in August 2008. On two separate occasions, she was a member of Canada's National Women's Under-22 Team for a three-game series vs. the United States. The first series occurred in Calgary, Alberta in August 2009 while the second was in Toronto, Ontario in August 2010. In addition, Kohanchuk won three gold medal's with Canada's National Women's Under-22 Team at the 2010, 2011, and 2012 MLP Cup in Ravensburg, Germany. She scored one of the six goals in the gold medal game of the 2011 MLP Cup. Kohanchuk was Centralized for the 2013/2014 Sochi Roster, but was one of the six players released from the final roster. In that year she won Gold with the National team at the 2013 Four Nations Cup, capturing 2 Goals and an assist in the gold medal game. The following year in Kamploops, she won another Gold at the Four Nations Cup, defeating Team USA in a shoot-out.
Career stats
Hockey Canada
NCAA
Awards and honours
2009 Hockey East all-rookie team
References
1990 births
Living people
Clarkson Cup champions
Canadian women's ice hockey forwards
Ice hockey people from Winnipeg
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 2,882
|
//
// SPTableContentDelegate.m
// sequel-pro
//
// Created by Stuart Connolly (stuconnolly.com) on March 20, 2012.
// Copyright (c) 2012 Stuart Connolly. All rights reserved.
//
// Permission is hereby granted, free of charge, to any person
// obtaining a copy of this software and associated documentation
// files (the "Software"), to deal in the Software without
// restriction, including without limitation the rights to use,
// copy, modify, merge, publish, distribute, sublicense, and/or sell
// copies of the Software, and to permit persons to whom the
// Software is furnished to do so, subject to the following
// conditions:
//
// The above copyright notice and this permission notice shall be
// included in all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
// EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
// OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
// NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
// HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
// WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
// FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
// OTHER DEALINGS IN THE SOFTWARE.
//
// More info at <https://github.com/sequelpro/sequelpro>
#import "SPTableContentDelegate.h"
#import "SPTableContentFilter.h"
#ifndef SP_CODA /* headers */
#import "SPAppController.h"
#endif
#import "SPDatabaseDocument.h"
#import "SPDataStorage.h"
#import "SPGeometryDataView.h"
#import "SPTooltip.h"
#import "SPTablesList.h"
#ifndef SP_CODA /* headers */
#import "SPBundleHTMLOutputController.h"
#endif
#import "SPCopyTable.h"
#import "SPAlertSheets.h"
#import "SPTableData.h"
#import "SPFieldEditorController.h"
#import "SPThreadAdditions.h"
#import "SPTextAndLinkCell.h"
#import <pthread.h>
#import <SPMySQL/SPMySQL.h>
@interface SPTableContent (SPDeclaredAPI)
- (BOOL)cancelRowEditing;
@end
@implementation SPTableContent (SPTableContentDelegate)
#pragma mark -
#pragma mark TableView delegate methods
/**
* Sorts the tableView by the clicked column. If clicked twice, order is altered to descending.
* Performs the task in a new thread if necessary.
*/
- (void)tableView:(NSTableView*)tableView didClickTableColumn:(NSTableColumn *)tableColumn
{
if ([selectedTable isEqualToString:@""] || !selectedTable || tableView != tableContentView) return;
// Prevent sorting while the table is still loading
if ([tableDocumentInstance isWorking]) return;
// Start the task
[tableDocumentInstance startTaskWithDescription:NSLocalizedString(@"Sorting table...", @"Sorting table task description")];
if ([NSThread isMainThread]) {
[NSThread detachNewThreadWithName:SPCtxt(@"SPTableContent table sort task", tableDocumentInstance) target:self selector:@selector(sortTableTaskWithColumn:) object:tableColumn];
}
else {
[self sortTableTaskWithColumn:tableColumn];
}
}
- (void)tableViewSelectionDidChange:(NSNotification *)aNotification
{
// Check our notification object is our table content view
if ([aNotification object] != tableContentView) return;
isFirstChangeInView = YES;
[addButton setEnabled:([tablesListInstance tableType] == SPTableTypeTable)];
// If we are editing a row, attempt to save that row - if saving failed, reselect the edit row.
if (isEditingRow && [tableContentView selectedRow] != currentlyEditingRow && ![self saveRowOnDeselect]) return;
if (![tableDocumentInstance isWorking]) {
// Update the row selection count
// and update the status of the delete/duplicate buttons
if([tablesListInstance tableType] == SPTableTypeTable) {
if ([tableContentView numberOfSelectedRows] > 0) {
[duplicateButton setEnabled:([tableContentView numberOfSelectedRows] == 1)];
[removeButton setEnabled:YES];
}
else {
[duplicateButton setEnabled:NO];
[removeButton setEnabled:NO];
}
}
else {
[duplicateButton setEnabled:NO];
[removeButton setEnabled:NO];
}
}
[self updateCountText];
#ifndef SP_CODA /* triggered commands */
NSArray *triggeredCommands = [SPAppDelegate bundleCommandsForTrigger:SPBundleTriggerActionTableRowChanged];
for (NSString *cmdPath in triggeredCommands)
{
NSArray *data = [cmdPath componentsSeparatedByString:@"|"];
NSMenuItem *aMenuItem = [[[NSMenuItem alloc] init] autorelease];
[aMenuItem setTag:0];
[aMenuItem setToolTip:[data objectAtIndex:0]];
// For HTML output check if corresponding window already exists
BOOL stopTrigger = NO;
if ([(NSString *)[data objectAtIndex:2] length]) {
BOOL correspondingWindowFound = NO;
NSString *uuid = [data objectAtIndex:2];
for (id win in [NSApp windows])
{
if ([[[[win delegate] class] description] isEqualToString:@"SPBundleHTMLOutputController"]) {
if ([[[win delegate] windowUUID] isEqualToString:uuid]) {
correspondingWindowFound = YES;
break;
}
}
}
if (!correspondingWindowFound) stopTrigger = YES;
}
if (!stopTrigger) {
if ([[data objectAtIndex:1] isEqualToString:SPBundleScopeGeneral]) {
[[SPAppDelegate onMainThread] executeBundleItemForApp:aMenuItem];
}
else if ([[data objectAtIndex:1] isEqualToString:SPBundleScopeDataTable]) {
if ([[[[[NSApp mainWindow] firstResponder] class] description] isEqualToString:@"SPCopyTable"]) {
[[[[NSApp mainWindow] firstResponder] onMainThread] executeBundleItemForDataTable:aMenuItem];
}
}
else if ([[data objectAtIndex:1] isEqualToString:SPBundleScopeInputField]) {
if ([[[NSApp mainWindow] firstResponder] isKindOfClass:[NSTextView class]]) {
[[[[NSApp mainWindow] firstResponder] onMainThread] executeBundleItemForInputField:aMenuItem];
}
}
}
}
#endif
}
/**
* Saves the new column size in the preferences.
*/
- (void)tableViewColumnDidResize:(NSNotification *)notification
{
// Check our notification object is our table content view
if ([notification object] != tableContentView) return;
// Sometimes the column has no identifier. I can't figure out what is causing it, so we just skip over this item
if (![[[notification userInfo] objectForKey:@"NSTableColumn"] identifier]) return;
NSMutableDictionary *tableColumnWidths;
NSString *database = [NSString stringWithFormat:@"%@@%@", [tableDocumentInstance database], [tableDocumentInstance host]];
NSString *table = [tablesListInstance tableName];
// Get tableColumnWidths object
#ifndef SP_CODA
if ([prefs objectForKey:SPTableColumnWidths] != nil ) {
tableColumnWidths = [NSMutableDictionary dictionaryWithDictionary:[prefs objectForKey:SPTableColumnWidths]];
}
else {
#endif
tableColumnWidths = [NSMutableDictionary dictionary];
#ifndef SP_CODA
}
#endif
// Get the database object
if ([tableColumnWidths objectForKey:database] == nil) {
[tableColumnWidths setObject:[NSMutableDictionary dictionary] forKey:database];
}
else {
[tableColumnWidths setObject:[NSMutableDictionary dictionaryWithDictionary:[tableColumnWidths objectForKey:database]] forKey:database];
}
// Get the table object
if ([[tableColumnWidths objectForKey:database] objectForKey:table] == nil) {
[[tableColumnWidths objectForKey:database] setObject:[NSMutableDictionary dictionary] forKey:table];
}
else {
[[tableColumnWidths objectForKey:database] setObject:[NSMutableDictionary dictionaryWithDictionary:[[tableColumnWidths objectForKey:database] objectForKey:table]] forKey:table];
}
// Save column size
[[[tableColumnWidths objectForKey:database] objectForKey:table]
setObject:[NSNumber numberWithDouble:[(NSTableColumn *)[[notification userInfo] objectForKey:@"NSTableColumn"] width]]
forKey:[[[[notification userInfo] objectForKey:@"NSTableColumn"] headerCell] stringValue]];
#ifndef SP_CODA
[prefs setObject:tableColumnWidths forKey:SPTableColumnWidths];
#endif
}
/**
* Confirm whether to allow editing of a row. Returns YES by default, unless the multipleLineEditingButton is in
* the ON state, or for blob or text fields - in those cases opens a sheet for editing instead and returns NO.
*/
- (BOOL)tableView:(NSTableView *)tableView shouldEditTableColumn:(NSTableColumn *)tableColumn row:(NSInteger)rowIndex
{
if ([tableDocumentInstance isWorking]) return NO;
#ifndef SP_CODA
if (tableView == filterTableView) {
return (filterTableIsSwapped && [[tableColumn identifier] integerValue] == 0) ? NO : YES;
}
else
#endif
if (tableView == tableContentView) {
// Ensure that row is editable since it could contain "(not loaded)" columns together with
// issue that the table has no primary key
NSString *wherePart = [NSString stringWithString:[self argumentForRow:[tableContentView selectedRow]]];
if (![wherePart length]) return NO;
// If the selected cell hasn't been loaded, load it.
if ([[tableValues cellDataAtRow:rowIndex column:[[tableColumn identifier] integerValue]] isSPNotLoaded]) {
// Only get the data for the selected column, not all of them
NSString *query = [NSString stringWithFormat:@"SELECT %@ FROM %@ WHERE %@", [[[tableColumn headerCell] stringValue] backtickQuotedString], [selectedTable backtickQuotedString], wherePart];
SPMySQLResult *tempResult = [mySQLConnection queryString:query];
if (![tempResult numberOfRows]) {
SPBeginAlertSheet(NSLocalizedString(@"Error", @"error"), NSLocalizedString(@"OK", @"OK button"), nil, nil, [tableDocumentInstance parentWindow], self, nil, nil,
NSLocalizedString(@"Couldn't load the row. Reload the table to be sure that the row exists and use a primary key for your table.", @"message of panel when loading of row failed"));
return NO;
}
NSArray *tempRow = [tempResult getRowAsArray];
[tableValues replaceObjectInRow:rowIndex column:[[tableContentView tableColumns] indexOfObject:tableColumn] withObject:[tempRow objectAtIndex:0]];
[tableContentView reloadData];
}
// Retrieve the column definition
NSDictionary *columnDefinition = [cqColumnDefinition objectAtIndex:[[tableColumn identifier] integerValue]];
// TODO: Fix editing of "Display as Hex" columns and remove this (also see above)
if ([[columnDefinition objectForKey:@"typegrouping"] isEqualToString:@"binary"] && [prefs boolForKey:SPDisplayBinaryDataAsHex]) {
NSBeep();
[SPTooltip showWithObject:NSLocalizedString(@"Disable \"Display Binary Data as Hex\" in the View menu to edit this field.",@"Temporary : Tooltip shown when trying to edit a binary field in table content view while it is displayed using HEX conversion")];
return NO;
}
// Open the editing sheet if required
if ([tableContentView shouldUseFieldEditorForRow:rowIndex column:[[tableColumn identifier] integerValue] checkWithLock:NULL]) {
BOOL isBlob = [tableDataInstance columnIsBlobOrText:[[tableColumn headerCell] stringValue]];
// A table is per definition editable
BOOL isFieldEditable = YES;
// Check for Views if field is editable
if ([tablesListInstance tableType] == SPTableTypeView) {
NSArray *editStatus = [self fieldEditStatusForRow:rowIndex andColumn:[[tableColumn identifier] integerValue]];
isFieldEditable = [[editStatus objectAtIndex:0] integerValue] == 1;
}
NSUInteger fieldLength = 0;
NSString *fieldEncoding = nil;
BOOL allowNULL = YES;
NSString *fieldType = [columnDefinition objectForKey:@"type"];
if ([columnDefinition objectForKey:@"char_length"]) {
fieldLength = [[columnDefinition objectForKey:@"char_length"] integerValue];
}
if ([columnDefinition objectForKey:@"null"]) {
allowNULL = (![[columnDefinition objectForKey:@"null"] integerValue]);
}
if ([columnDefinition objectForKey:@"charset_name"] && ![[columnDefinition objectForKey:@"charset_name"] isEqualToString:@"binary"]) {
fieldEncoding = [columnDefinition objectForKey:@"charset_name"];
}
if (fieldEditor) SPClear(fieldEditor);
fieldEditor = [[SPFieldEditorController alloc] init];
[fieldEditor setEditedFieldInfo:[NSDictionary dictionaryWithObjectsAndKeys:
[[tableColumn headerCell] stringValue], @"colName",
[self usedQuery], @"usedQuery",
@"content", @"tableSource",
nil]];
[fieldEditor setTextMaxLength:fieldLength];
[fieldEditor setFieldType:fieldType == nil ? @"" : fieldType];
[fieldEditor setFieldEncoding:fieldEncoding == nil ? @"" : fieldEncoding];
[fieldEditor setAllowNULL:allowNULL];
id cellValue = [tableValues cellDataAtRow:rowIndex column:[[tableColumn identifier] integerValue]];
if ([cellValue isNSNull]) {
cellValue = [NSString stringWithString:[prefs objectForKey:SPNullValue]];
}
if ([[columnDefinition objectForKey:@"typegrouping"] isEqualToString:@"binary"] && [prefs boolForKey:SPDisplayBinaryDataAsHex]) {
[fieldEditor setTextMaxLength:[[self tableView:tableContentView objectValueForTableColumn:tableColumn row:rowIndex] length]];
isFieldEditable = NO;
}
NSInteger editedColumn = 0;
for (NSTableColumn* col in [tableContentView tableColumns])
{
if ([[col identifier] isEqualToString:[tableColumn identifier]]) break;
editedColumn++;
}
[fieldEditor editWithObject:cellValue
fieldName:[[tableColumn headerCell] stringValue]
usingEncoding:[mySQLConnection stringEncoding]
isObjectBlob:isBlob
isEditable:isFieldEditable
withWindow:[tableDocumentInstance parentWindow]
sender:self
contextInfo:[NSDictionary dictionaryWithObjectsAndKeys:
[NSNumber numberWithInteger:rowIndex], @"rowIndex",
[NSNumber numberWithInteger:editedColumn], @"columnIndex",
[NSNumber numberWithBool:isFieldEditable], @"isFieldEditable",
nil]];
return NO;
}
return YES;
}
return YES;
}
/**
* Enable drag from tableview
*/
- (BOOL)tableView:(NSTableView *)tableView writeRowsWithIndexes:(NSIndexSet *)rows toPasteboard:(NSPasteboard*)pboard
{
if (tableView == tableContentView) {
NSString *tmp;
// By holding ⌘, ⇧, or/and ⌥ copies selected rows as SQL INSERTS
// otherwise \t delimited lines
if ([[NSApp currentEvent] modifierFlags] & (NSCommandKeyMask|NSShiftKeyMask|NSAlternateKeyMask)) {
tmp = [tableContentView rowsAsSqlInsertsOnlySelectedRows:YES];
}
else {
tmp = [tableContentView draggedRowsAsTabString];
}
#warning code inside "if" is unreachable
if (!tmp && [tmp length])
{
[pboard declareTypes:@[NSTabularTextPboardType, NSStringPboardType] owner:nil];
[pboard setString:tmp forType:NSStringPboardType];
[pboard setString:tmp forType:NSTabularTextPboardType];
return YES;
}
}
return NO;
}
/**
* Disable row selection while the document is working.
*/
- (BOOL)tableView:(NSTableView *)tableView shouldSelectRow:(NSInteger)rowIndex
{
#ifndef SP_CODA
if (tableView == filterTableView) {
return YES;
}
else
#endif
return tableView == tableContentView ? tableRowsSelectable : YES;
}
/**
* Resize a column when it's double-clicked (10.6+ only).
*/
- (CGFloat)tableView:(NSTableView *)tableView sizeToFitWidthOfColumn:(NSInteger)columnIndex
{
NSTableColumn *theColumn = [[tableView tableColumns] objectAtIndex:columnIndex];
NSDictionary *columnDefinition = [dataColumns objectAtIndex:[[theColumn identifier] integerValue]];
// Get the column width
NSUInteger targetWidth = [tableContentView autodetectWidthForColumnDefinition:columnDefinition maxRows:500];
#ifndef SP_CODA
// Clear any saved widths for the column
NSString *dbKey = [NSString stringWithFormat:@"%@@%@", [tableDocumentInstance database], [tableDocumentInstance host]];
NSString *tableKey = [tablesListInstance tableName];
NSMutableDictionary *savedWidths = [NSMutableDictionary dictionaryWithDictionary:[prefs objectForKey:SPTableColumnWidths]];
NSMutableDictionary *dbDict = [NSMutableDictionary dictionaryWithDictionary:[savedWidths objectForKey:dbKey]];
NSMutableDictionary *tableDict = [NSMutableDictionary dictionaryWithDictionary:[dbDict objectForKey:tableKey]];
if ([tableDict objectForKey:[columnDefinition objectForKey:@"name"]]) {
[tableDict removeObjectForKey:[columnDefinition objectForKey:@"name"]];
if ([tableDict count]) {
[dbDict setObject:[NSDictionary dictionaryWithDictionary:tableDict] forKey:tableKey];
}
else {
[dbDict removeObjectForKey:tableKey];
}
if ([dbDict count]) {
[savedWidths setObject:[NSDictionary dictionaryWithDictionary:dbDict] forKey:dbKey];
}
else {
[savedWidths removeObjectForKey:dbKey];
}
[prefs setObject:[NSDictionary dictionaryWithDictionary:savedWidths] forKey:SPTableColumnWidths];
}
#endif
// Return the width, while the delegate is empty to prevent column resize notifications
[tableContentView setDelegate:nil];
[tableContentView performSelector:@selector(setDelegate:) withObject:self afterDelay:0.1];
return targetWidth;
}
/**
* This function changes the text color of text/blob fields which are null or not yet loaded to gray
*/
- (void)tableView:(SPCopyTable *)tableView willDisplayCell:(id)cell forTableColumn:(NSTableColumn *)tableColumn row:(NSInteger)rowIndex
{
#ifndef SP_CODA
if (tableView == filterTableView) {
if (filterTableIsSwapped && [[tableColumn identifier] integerValue] == 0) {
[cell setDrawsBackground:YES];
[cell setBackgroundColor:lightGrayColor];
}
else {
[cell setDrawsBackground:NO];
}
return;
}
else
#endif
if (tableView == tableContentView) {
if (![cell respondsToSelector:@selector(setTextColor:)]) return;
BOOL cellIsNullOrUnloaded = NO;
BOOL cellIsLinkCell = [cell isMemberOfClass:[SPTextAndLinkCell class]];
NSUInteger columnIndex = [[tableColumn identifier] integerValue];
// If user wants to edit 'cell' set text color to black and return to avoid
// writing in gray if value was NULL
if ([tableView editedColumn] != -1
&& [tableView editedRow] == rowIndex
&& (NSUInteger)[[NSArrayObjectAtIndex([tableView tableColumns], [tableView editedColumn]) identifier] integerValue] == columnIndex) {
[cell setTextColor:blackColor];
if (cellIsLinkCell) [cell setLinkActive:NO];
return;
}
// While the table is being loaded, additional validation is required - data
// locks must be used to avoid crashes, and indexes higher than the available
// rows or columns may be requested. Use gray to indicate loading in these cases.
if (isWorking) {
pthread_mutex_lock(&tableValuesLock);
if (rowIndex < (NSInteger)tableRowsCount && columnIndex < [tableValues columnCount]) {
cellIsNullOrUnloaded = [tableValues cellIsNullOrUnloadedAtRow:rowIndex column:columnIndex];
}
pthread_mutex_unlock(&tableValuesLock);
}
else {
cellIsNullOrUnloaded = [tableValues cellIsNullOrUnloadedAtRow:rowIndex column:columnIndex];
}
if (cellIsNullOrUnloaded) {
[cell setTextColor:rowIndex == [tableContentView selectedRow] ? whiteColor : lightGrayColor];
}
else {
[cell setTextColor:blackColor];
}
NSDictionary *columnDefinition = [[(id <SPDatabaseContentViewDelegate>)[tableContentView delegate] dataColumnDefinitions] objectAtIndex:columnIndex];
NSString *columnType = [columnDefinition objectForKey:@"typegrouping"];
// Find a more reliable way of doing this check
if ([columnType isEqualToString:@"binary"] &&
[prefs boolForKey:SPDisplayBinaryDataAsHex] &&
[[self tableView:tableContentView objectValueForTableColumn:tableColumn row:rowIndex] hasPrefix:@"0x"]) {
[cell setTextColor:rowIndex == [tableContentView selectedRow] ? whiteColor : blueColor];
}
// Disable link arrows for the currently editing row and for any NULL or unloaded cells
if (cellIsLinkCell) {
if (cellIsNullOrUnloaded || [tableView editedRow] == rowIndex) {
[cell setLinkActive:NO];
}
else {
[cell setLinkActive:YES];
}
}
}
}
#ifndef SP_CODA
/**
* Show the table cell content as tooltip
*
* - for text displays line breaks and tabs as well
* - if blob data can be interpret as image data display the image as transparent thumbnail
* (up to now using base64 encoded HTML data).
*/
- (NSString *)tableView:(NSTableView *)tableView toolTipForCell:(id)aCell rect:(NSRectPointer)rect tableColumn:(NSTableColumn *)tableColumn row:(NSInteger)row mouseLocation:(NSPoint)mouseLocation
{
if (tableView == filterTableView) {
return nil;
}
else if (tableView == tableContentView) {
if ([[aCell stringValue] length] < 2 || [tableDocumentInstance isWorking]) return nil;
// Suppress tooltip if another toolip is already visible, mainly displayed by a Bundle command
// TODO has to be improved
for (id win in [NSApp orderedWindows])
{
if ([[[[win contentView] class] description] isEqualToString:@"WebView"]) return nil;
}
NSImage *image;
NSPoint pos = [NSEvent mouseLocation];
pos.y -= 20;
id theValue = nil;
// While the table is being loaded, additional validation is required - data
// locks must be used to avoid crashes, and indexes higher than the available
// rows or columns may be requested. Return "..." to indicate loading in these
// cases.
if (isWorking) {
pthread_mutex_lock(&tableValuesLock);
if (row < (NSInteger)tableRowsCount && [[tableColumn identifier] integerValue] < (NSInteger)[tableValues columnCount]) {
theValue = [[SPDataStorageObjectAtRowAndColumn(tableValues, row, [[tableColumn identifier] integerValue]) copy] autorelease];
}
pthread_mutex_unlock(&tableValuesLock);
if (!theValue) theValue = @"...";
}
else {
theValue = SPDataStorageObjectAtRowAndColumn(tableValues, row, [[tableColumn identifier] integerValue]);
}
if (theValue == nil) return nil;
if ([theValue isKindOfClass:[NSData class]]) {
image = [[[NSImage alloc] initWithData:theValue] autorelease];
if (image) {
[SPTooltip showWithObject:image atLocation:pos ofType:@"image"];
return nil;
}
}
else if ([theValue isKindOfClass:[SPMySQLGeometryData class]]) {
SPGeometryDataView *v = [[SPGeometryDataView alloc] initWithCoordinates:[theValue coordinates]];
image = [v thumbnailImage];
if (image) {
[SPTooltip showWithObject:image atLocation:pos ofType:@"image"];
[v release];
return nil;
}
[v release];
}
// Show the cell string value as tooltip (including line breaks and tabs)
// by using the cell's font
[SPTooltip showWithObject:[aCell stringValue]
atLocation:pos
ofType:@"text"
displayOptions:[NSDictionary dictionaryWithObjectsAndKeys:
[[aCell font] familyName], @"fontname",
[NSString stringWithFormat:@"%f",[[aCell font] pointSize]], @"fontsize",
nil]];
return nil;
}
return nil;
}
#endif
#ifndef SP_CODA /* SplitView delegate methods */
#pragma mark -
#pragma mark SplitView delegate methods
- (BOOL)splitView:(NSSplitView *)sender canCollapseSubview:(NSView *)subview
{
return NO;
}
/**
* Set a minimum size for the filter text area.
*/
- (CGFloat)splitView:(NSSplitView *)sender constrainMaxCoordinate:(CGFloat)proposedMax ofSubviewAt:(NSInteger)offset
{
return proposedMax - 180;
}
/**
* Set a minimum size for the field list and action area.
*/
- (CGFloat)splitView:(NSSplitView *)sender constrainMinCoordinate:(CGFloat)proposedMin ofSubviewAt:(NSInteger)offset
{
return proposedMin + 225;
}
/**
* Improve default resizing and resize only the filter text area by default.
*/
- (void)splitView:(NSSplitView *)sender resizeSubviewsWithOldSize:(NSSize)oldSize
{
NSSize newSize = [sender frame].size;
NSView *leftView = [[sender subviews] objectAtIndex:0];
NSView *rightView = [[sender subviews] objectAtIndex:1];
float dividerThickness = [sender dividerThickness];
NSRect leftFrame = [leftView frame];
NSRect rightFrame = [rightView frame];
// Resize height of both views
leftFrame.size.height = newSize.height;
rightFrame.size.height = newSize.height;
// Only resize the right view's width - unless the constraint has been reached
if (rightFrame.size.width > 180 || newSize.width > oldSize.width) {
rightFrame.size.width = newSize.width - leftFrame.size.width - dividerThickness;
}
else {
leftFrame.size.width = newSize.width - rightFrame.size.width - dividerThickness;
}
rightFrame.origin.x = leftFrame.size.width + dividerThickness;
[leftView setFrame:leftFrame];
[rightView setFrame:rightFrame];
}
#endif
#pragma mark -
#pragma mark Control delegate methods
- (void)controlTextDidChange:(NSNotification *)notification
{
#ifndef SP_CODA
if ([notification object] == filterTableView) {
NSString *string = [[[[notification userInfo] objectForKey:@"NSFieldEditor"] textStorage] string];
if (string && [string length]) {
if (lastEditedFilterTableValue) [lastEditedFilterTableValue release];
lastEditedFilterTableValue = [[NSString stringWithString:string] retain];
}
[self updateFilterTableClause:string];
}
#endif
}
/**
* If the user selected a table cell which is a blob field and tried to edit it
* cancel the fieldEditor, display the field editor sheet instead for editing
* and re-enable the fieldEditor after editing.
*/
- (BOOL)control:(NSControl *)control textShouldBeginEditing:(NSText *)aFieldEditor
{
if (control != tableContentView) return YES;
NSUInteger row, column;
BOOL shouldBeginEditing = YES;
row = [tableContentView editedRow];
column = [tableContentView editedColumn];
// If cell editing mode and editing request comes
// from the keyboard show an error tooltip
// or bypass if numberOfPossibleUpdateRows == 1
if ([tableContentView isCellEditingMode]) {
NSArray *editStatus = [self fieldEditStatusForRow:row andColumn:[[NSArrayObjectAtIndex([tableContentView tableColumns], column) identifier] integerValue]];
NSInteger numberOfPossibleUpdateRows = [[editStatus objectAtIndex:0] integerValue];
NSPoint pos = [[tableDocumentInstance parentWindow] convertBaseToScreen:[tableContentView convertPoint:[tableContentView frameOfCellAtColumn:column row:row].origin toView:nil]];
pos.y -= 20;
switch (numberOfPossibleUpdateRows)
{
case -1:
[SPTooltip showWithObject:kCellEditorErrorNoMultiTabDb
atLocation:pos
ofType:@"text"];
shouldBeginEditing = NO;
break;
case 0:
[SPTooltip showWithObject:[NSString stringWithFormat:kCellEditorErrorNoMatch, selectedTable]
atLocation:pos
ofType:@"text"];
shouldBeginEditing = NO;
break;
case 1:
shouldBeginEditing = YES;
break;
default:
[SPTooltip showWithObject:[NSString stringWithFormat:kCellEditorErrorTooManyMatches, (long)numberOfPossibleUpdateRows]
atLocation:pos
ofType:@"text"];
shouldBeginEditing = NO;
}
}
// Open the field editor sheet if required
if ([tableContentView shouldUseFieldEditorForRow:row column:column checkWithLock:NULL])
{
[tableContentView setFieldEditorSelectedRange:[aFieldEditor selectedRange]];
// Cancel editing
[control abortEditing];
// Call the field editor sheet
[self tableView:tableContentView shouldEditTableColumn:NSArrayObjectAtIndex([tableContentView tableColumns], column) row:row];
// send current event to field editor sheet
if ([NSApp currentEvent]) {
[NSApp sendEvent:[NSApp currentEvent]];
}
return NO;
}
return shouldBeginEditing;
}
/**
* Trap the enter, escape, tab and arrow keys, overriding default behaviour and continuing/ending editing,
* only within the current row.
*/
- (BOOL)control:(NSControl<NSControlTextEditingDelegate> *)control textView:(NSTextView *)textView doCommandBySelector:(SEL)command
{
// Check firstly if SPCopyTable can handle command
if ([control control:control textView:textView doCommandBySelector:command])
return YES;
// Trap the escape key
if ([[control window] methodForSelector:command] == [[control window] methodForSelector:@selector(cancelOperation:)]) {
// Abort editing
[control abortEditing];
if ((SPCopyTable*)control == tableContentView) {
[self cancelRowEditing];
}
return YES;
}
return NO;
}
#pragma mark -
#pragma mark Database content view delegate methods
- (NSString *)usedQuery
{
return usedQuery;
}
/**
* Retrieve the data column definitions
*/
- (NSArray *)dataColumnDefinitions
{
return dataColumns;
}
@end
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 1,446
|
Helianthus petiolaris est une espèce végétale nord-américaine de la famille des Asteraceae, communément appelée tournesol des prairies ou petit tournesol . Le naturaliste et botaniste Thomas Nuttall a été le premier à décrire le tournesol des prairies en 1821. Le mot petiolaris en latin signifie « avoir un pétiole ». L'espèce est originaire de l'ouest des États-Unis, mais s'est depuis étendue à l'est. Le tournesol des prairies est parfois considéré comme une mauvaise herbe.
Distribution
Helianthus petiolaris est originaire des prairies sèches du Minnesota, de l' Oregon, du Texas, des Dakota du Sud et du Nord, de la Californie et d'autres États de l'ouest et du centre des États-Unis. Il a depuis étendu sa distribution à tout l'est des États-Unis et dans le centre et l'ouest du Canada. C'est maintenant l'espèce de tournesol la plus répandue avec H. annuus .
Habitat et écologie
Les tournesols des prairies poussent couramment dans les zones sablonneuses. On les trouve aussi dans les sols argileux lourds et dans les prairies sèches. Ils sont incapables de pousser dans des zones ombragées, ils doivent être en plein soleil. Les tournesols des prairies ont besoin d'un sol sec à humide. Cette espèce de tournesol est annuelle, qui fleurit entre juin et septembre.
Morphologie
Le tournesol des prairies est une annuelle à racine pivotante. Il pousse jusqu'à de hauteur. Les feuilles semblent alternes et les fleurs ressemblent beaucoup au tournesol traditionnel. Les fleurs sont hermaphrodites, ce qui signifie que les fleurs contiennent à la fois des parties mâles et femelles. La tige de la fleur est dressée et velue. Les feuilles sont alternes, lancéolées, sont de texture rugueuse, de couleur bleu-vert et ont une longueur comprise entre 5 et 15 cm
Fleurs
Helianthus petiolaris a des capitules semblables à ceux d'un tournesol commun, H. annuus. Les fruits sont des akènes . Le capitule contient 10-30 fleurons ligulés, entourant 50-100 fleurons rouge foncé brun. L'arrière du capitule porte des bractées lancéolées, ou bractées involucrales, vertes. Le centre du capitule présente des reflets blancs dus à la présence de poils blancs sur les fleurons immatures. Les fleurs attirent les papillons et les abeilles pour la pollinisation.
Utilisations
Alimentaire
Les graines de la plante sont comestibles et peuvent être broyées en un tourteau huileux ou en un beurre.
Médicinal
Les feuilles en poudre peuvent être utilisées pour soigner des plaies et des œdèmes.
Sélection génétique
En 1969, Leclercq a découvert la stérilité mâle cytoplasmique (CMS) PET-1 dans une population de H. petiolaris, puis en 1984 les loci de restauration de fertilité Rf1 et Rf2 . Ce système PET-1/Rf1 est aujourd'hui très largement utilisé par les semenciers pour la production d'hybrides F1.
Sous-espèces
Helianthus petiolaris var. canescens A.Gray - Arizona, Californie, Nevada, Nouveau-Mexique, Texas, Chihuahua
Helianthus petiolaris var. var. fallax (Heiser) BLTurner - Arizona, Colorado, Nevada, Utah
Helianthus petiolaris var. var. petiolaris - la plupart de l'aire de répartition des espèces
Listes des références
Flore en Amérique du Nord
Statut UICN Préoccupation mineure
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 5,875
|
La teoría pura del derecho (en alemán: Reine Rechtslehre) es tanto un proyecto teórico, como el nombre de una obra del célebre jurista austríaco Hans Kelsen, quien dio el sustento más importante para el desarrollo de dicho proyecto.
Fundamentos
La idea subyacente en la teoría pura del derecho es la autonomización del Derecho de la Política, Sociología, Moral e Ideología. Esta autonomización busca otorgar al derecho unidad y carácter científico, lo consagra como una disciplina positivista. Kelsen en la teoría pura, opone el positivismo jurídico (o iuspositivismo) con el derecho natural. En la obra se identifica la predominacia absoluta del derecho positivo como orden normativo y las constante negaciones de supuestos dualismos como: el derecho natural/positivo, derecho público/privado, derecho/estado, etc.
La obra extirpa del análisis científico toda noción ajena a la producción jurídica (metajurídica) creada mediante medios, procedimental y formalmente, establecidos como la ley y los actos administrativos. Los móviles de la teoría pura del derecho son: En primer lugar, la cientificación del estudio del derecho y la desideologización del derecho.
Por otra parte, Kelsen sustenta un ordenamiento jurídico sobre la base de la jerarquía normativa (toda norma obtiene su vigencia de una norma superior). Esta jerarquía tiene su máxima representante en la Constitución; sin embargo, la Constitución tiene aún un sustento anterior conocido como Norma Fundante Básica.
Analizando la estructura de los sistemas jurídicos concluyó que toda norma obtiene su vigencia de una norma superior, remitiendo su validez hasta una Norma hipotética básica, cuyo valor es pre-supuesto y no cuestionado. Establece además la validez de la norma en su modo de producción y no en el contenido de la misma:
Críticas
Esta teoría fue objeto de muchas críticas tras la segunda guerra mundial; el motivo fue que dentro de ella, todos los abusos cometidos por los nazis —entre otros regímenes totalitarios—, eran actos "jurídicamente correctos" (eran legales, en su ordenamiento jurídico), y eso era de muy difícil aceptación. Sin embargo, las mayores críticas se refieren a la inaplicabilidad de su teoría en la realidad, a la imposibilidad de tratar una ciencia social como natural, al despropósito de eliminar los elementos metajurídicos del derecho, y al reducido espectro de estudio que posee la Teoría pura del Derecho.
Legado
El proyecto de Kelsen ha sido objeto de continuos ataques; sin embargo, la fuerza motivadora del proyecto, no ha dejado de marcar todas y cada una de las nuevas etapas de la teoría del Derecho, encumbrándose como uno de los proyectos con influencias más relevantes, no sólo para la modernización del Derecho, sino también para la modernización del análisis social y de la orientación de los fines del Estado.
Véase también
Filosofía del Derecho
Sociología política
Libros de derecho
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 1,655
|
\section{Introduction}
\label{sec1}
Radio pulsar B1933+16 is the brightest core-emission dominated pulsar
in the Arecibo sky, as the majority of such pulsars fall far in the
southern sky around the Galactic center. Despite its prominence, it
has received very little study and analysis. Sieber {\it et al.}\thinspace\ (1975)
early showed its profile evolution from a single form at meter
wavelengths to a tripartite one above 1 GHz, and this pattern of
evolution later became a defining characteristic of the core-single
profile class in the {\it Empirical Theory of Pulsar Emission} series
(Rankin 1983, 1993a; hereafter ET I, VIa). Other work (Gould \& Lyne
1998; Rankin, Stinebring \& Weisberg 1989; Weisberg {\it et al.}\thinspace 1999;
Hankins \& Rankin 2010) confirmed this profile evolution with even
better quality observations.
In the parlance of the ET papers, B1933+16 has been regarded as having
a core-single ({\bf S}$_t$) profile, meaning that it ``grows'' conal
``outriders'' only at high frequency. The star's core feature does
exhibit the antisymmetric circular polarization often seen under core
features, as we see in Fig.~\ref{fig1} below, and the angular
dimensions of the core and conal components seem to reflect the
angular size of the pulsar's polar cap emission region as expected in
ET VI. However, B1933+16's core ``component'' does exhibit a very
clear non-Gaussian structure---as do some other core features
(sometimes earlier referred to as ``notches'')---and its origin is one
of the questions before us below. At lower frequencies its core
component gradually broadens (ET II), in part because of scattering,
and by 100 MHz, the scattering tail occupies the entire rotation cycle
(Izvekova {\it et al.}\thinspace\ 1989).
Few pulse-sequence analyses had been carried out on pulsars with
dominant core emission. Exceptions are B0329+54 and the Vela pulsar
B0833--45, the study of this latter pulsar by Krishmohan \& Downs
(1983) being an early classic which has received far too little
followup. Our study of B0329+54's normal mode (Mitra, Rankin \& Gupta
2007) revealed a clear signature of intensity-dependent
aberration/retardation (hereafter A/R) in its core emission, which in
turn was interpreted as the presence of a linear amplifier within the
star's polar flux tube. A further study showing similar core
properties was conducted with pulsar B1237+25 (Smith, Rankin \& Mitra
2012).
Johnston {\it et al.}\thinspace\ (2005) first demonstrated that the ``fiducial''
polarization orientation of pulsar radio emission---at the central
longitude of the magnetic axis---exhibits preferential parallel or
orthogonal alignments with respect to their proper-motion (hereafter
$PA_v$) directions. In carrying this work further by considering only
pulsars with core emission, Rankin (2015) showed that the later core
emission shows an accurate orthogonal orientation reflecting the
extraordinary X propagation mode. The classic Johnston {\it et al.}\thinspace\ study
did include pulsar B1933+16; but, in their analysis its alignment was
indifferent. We have come to believe, however, that the longitude of
the pulsar's magnetic axis falls later than was then thought, and when
this is taken into account, B1933+16's late core emission seems to be
the X mode and, if so, the two polarization modes can be identified
throughout its profile. This will be further discussed below.
In this current analysis we have a few straightforward objectives that
follow from our earlier pulse-sequence analyses of core emission
above: a) we want to assess whether the star's core radiation exhibits
intensity-dependent A/R and, if so, interpret its physical
significance; b) we will attempt to understand how the pulsar's
polarization position-angle (hereafter PPA) traverse becomes distorted
from what would be expected from the rotating-vector model (hereafter
RVM; Radhakrishnan \& Cooke 1969); we want to understand the
significance of the pulsar's prominent circular polarization; further,
B1933+16 is one of the few bright pulsars available to us in which
core microstructure can be investigated, so we will attempt an
analysis similar to that of Mitra, Arjunwadkar \& Rankin (2015); and
finally we will assess how the pulsar's conal emission is connected to
its prominent core radiation.
\begin{figure}
\begin{center}
\mbox{\includegraphics[height=100mm,width=75mm,angle=-90.]{srB1933+16.57119l_pav2.ps}}
\mbox{\includegraphics[height=100mm,width=75mm,angle=-90.]{srB1933+16.57285c_pav2.ps}}
\caption{ Total polarization displays at 1515 MHz (top) and 4555 MHz (bottom),
giving (upper panels) the total intensity (Stokes $I$; solid curve),
the total linear ($L$ (=$\sqrt{Q^2+U^2}$); dashed red), the circular
polarization (Stokes $V$; dotted blue), the fractional linear
polarization ($L/I$; gray points) and a zoomed total intensity
(10$\times I $; cyan curve). The PPA [=${1\over 2}\tan^{-1} (U/Q)$]
densities (lower panels) within each 1$\degr$x1-sample cell correspond
to samples larger than 2 $\sigma$ in $L$, are plotted according to the
color bars at the lower right, and have been derotated to infinite
frequency. The average PPA traverses (red) are overplotted, and the
RVM fit to the 1515-MHz PPA traverse is plotted twice for the two
polarization modes (magenta dashed) on each panel. The origin is
taken at the fitted PPA inflection point.}
\label{fig1}
\end{center}
\end{figure}
In what follows, \S\ref{sec2} describes our observations. \S\ref{sec3}
presents our polarimetry, and \S\ref{sec4} discusses the results of
intensity fractionation. In \S\ref{sec5} we conduct analyses of the
segregated modes, \S\ref{sec6} interprets the PPA traverse
geometrically, and \S\ref{sec7} presents the microstructure
analysis. \S\ref{sec8} discusses the core and conal emission,
\S\ref{sec9} provides an aberration/retardation analysis, and \S\ref{sec10}
considers the frequency independence of the modal polarization.
\S\ref{sec:summary} then summarizes our results and \S\ref{sec:discussion}
considers their larger theoretical interpretation and significance.
\section{Observations}
\label{sec2}
The observations used in our analyses were made using the 305-m
Arecibo Telescope in Puerto Rico. The primary 1515-MHz polarized
pulse sequence (hereafter PS) was acquired with the upgraded
instrument using seven Mock Spectrometers on 2015 April 7 (MJD 57119)
and is comprised of 5017 pulses. Each spectrometer sampled a 25-MHz
portion of the 275-MHz band centered at 1515 MHz with 36.14 $\mu$s
sampling. The 4555-MHz observation was again carried out using seven
Mocks on 2015 September 20 (MJD 57284) with a total bandwidth of 7x150
MHz and 51 $\mu$s sampling. The Stokes parameters have been corrected
for dispersion, interstellar Faraday rotation and various instrumental
polarization effects and are ultimately derotated to infinite
frequency. Errors in the PPAs were computed relative to the off-pulse
noise phasor---that is, $\sigma_{PPA}
\sim tan^{-1 } (\sigma_{off-pulse}/L)$.
\begin{figure}
\begin{center}
\begin{tabular}{c}
\mbox{\includegraphics[height=78mm,angle=-90.]{B1933_lband_5levs.ps}}\\
\mbox{\includegraphics[height=78mm,angle=-90.]{B1933_cband_5levs.ps}}\\
\end{tabular}
\caption{Average profiles corresponding to five intensity fractions for the 1.5-GHz
(upper) and 4.6-GHz (lower) observations. Overall, this analysis
reflects the steadiness of the pulsar's emission. However, several
aspects are worthy of mention. Note the manner in which the leading
part of the core becomes progressively weaker relative to the trailing
region with increased intensity. Also, note that this leading part
becomes more linearly polarized.}
\label{fig2}
\end{center}
\end{figure}
\section{Confronting the Polarimetry}
\label{sec3}
The average 1515- and 4555-MHz polarimetry presented in
Fig.~\ref{fig1} is new and unique in several ways. First, the
pulsar is strong enough that a large proportion of the respective 36
and 51-$\mu$s samples show well defined polarization---and the linear
polarization angle (hereafter PPA) histograms corresponding to them
are plotted in the lower panels along with the average PPA
traverse. The histograms also have the effect of revealing the
polarized structure of the 1.5-GHz PPA ``notch'' at --5\degr\ much
more clearly, as in earlier profiles it is seen only a ``bump'' in the
average PPA traverse. The PPAs have been derotated to infinite
frequency, so that it is easy to see that several ``patches'' of PPAs
at the two frequencies correspond to each other.
The puzzling structure of the central core component is clear in the
total power profiles at both frequencies (solid black curves), and
while earlier noted as a ``notch'' possibly due to ``absorption'', we
will explore whether it has a different origin. The aggregate linear
polarization (dashed red) is only moderate at both frequencies across
the full width of the profile, and we see a consistent signature of
circular polarization (dotted blue). The conal outriding components
are more prominent at the higher frequency as expected, and PPA
``patches'' corresponding to their orthogonal polarization modes
(hereafter OPMs) fall at similar positions in the respective lower
panels.
Only at the lower frequency is it possible to fit a rotating-vector
model (hereafter RVM) traverse to the observed PPAs. The fit is
simultaneous to both modes, and the fitted PPA trajectory is shown by
a pair of magenta curves. The primary mode track is seen at a PPA of
+20\degr\ near the leading edge of the plot and rotates negatively
through about 160\degr\ to +41\degr\ on the far right of the plot.
The longitude origin of the plot is then taken at the inflection
point, and the PPA here is about --58\degr. This fitted track seems
very plausible in terms of the array of 1.5-GHz PPA ``patches'', but
it is important to resolve whether it is both unique and correct. Of
course, we have had to omit the non-RVM PPAs around --5\degr\
longitude from the fitting, and we clearly see pairs of other
``patches'' on the conal edges of the profile. We explored, for
instance, whether the core PPAs from about --9 to 0\degr\ might
represent the same mode as the main conal ``patch'' at 5-6\degr, and
we found not surprisingly that the total PPA traverse is not large
enough to fit these under any conditions. So, our conclusion is that
most of the core power represents one mode, {whereas the --5\degr\
region} and the brighter emission under the conal outriders the other.
The RVM fit to the 1.5-GHz PPA cannot constrain $\alpha$ and $\beta$
values due to the large correlation of about 98\% between them (see
for e.g. Everett \& Weisberg 2001). However, the fitted slope
$R_{PA}$ [=$\sin\alpha/\sin\beta$] of --39$\pm$3 \degr/\degr\ and PPA
at the inflection point of --58\degr$\pm$5\degr\ are well determined,
and the longitude origin is taken at this point in the plots. The
other way of estimating $\alpha$ comes from the core-width measurement
as was done in ET VI. Using this $\alpha$ (see section~\ref{sec6} and
~\ref{sec11}) as initial value to the fits we obtained $\alpha$ and
$\beta$ values of 125\degr $\pm$19\degr\ and --1.2\degr$\pm$0.5\degr,
where the errors correspond to the fit errors. We were unable to
carry out a similar fitting for the 4.6-GHz observation---as is pretty
obvious from the nature of the polarized profile. However, we were
able to ``anchor'' the lower frequency solution on its prominent
leading and trailing PPA ``patches'' and the PPA curve is plotted as
in the lower frequency profile.
That the fitted PPA inflection point falls so late under the core
feature initially surprised us, but we then recalled that we have seen
this behavior under core features in faster pulsars. With a rotation
period of 359 ms, we suspect that this is due to
aberration/retardation (hereafter A/R) which we will explore further
below.
\begin{figure}
\begin{center}
\begin{tabular}{c}
\mbox{\includegraphics[height=105mm,width=75mm,angle=-90.]{srB1933+16.57119l2p.ps}}\\
\mbox{\includegraphics[height=105mm,width=75mm,angle=-90.]{srB1933+16.57119l2s.ps}}\\
\end{tabular}
\caption{Polarization profiles and PPA distributions after Figure~\ref{fig1} for two-way
polarization-mode PS segregations of the 1.5-GHz observation per
Deshpande \& Rankin (2001); primary mode (top) and secondary mode
(bottom).}
\label{fig3}
\end{center}
\end{figure}
Similarly, we were perplexed by the orientation of the
inflection-point PPA ($PA_0$) as one of us had recently interpreted it
as --87\degr\ (Rankin 2015) on the basis of the Johnston {\it et al.}\thinspace's
(2005) average polarimetry. Surely this gives an object lesson
regarding the difficulties of interpreting average PPA traverses
without appeal to any pulse-sequence polarimetry. Here we find a
$PA_0$= --58\degr$\pm$5\degr, and we can use this value to estimate
the orientation of the pulsar's rotation axis on the sky so as to then
determine how the radiation is polarized with respect to the emitting
magnetic field direction. Pulsar proper motions have been found to be
largely parallel to a pulsar's rotation axis (Rankin 2015), and that
of B1933+16 was determined by VLBI to be $PA_v=$+176(0)\degr. A
similar value was given by Johnston {\it et al.}\thinspace\ as an improved estimate
over that determined earlier by Hobbs {\it et al.}\thinspace\ (2004). Therefore, the
alignment angle $\Psi = PA_v- PA_0$ is computed --4 -- (--58) =
+54\degr$\pm$5\degr, modulo 180\degr. Although this value is far from
orthogonal, we will interpret it as corresponding to the extraordinary
(X) propagation mode.
Indeed, the compact distribution of $\Psi$ values ({\it e.g.,}\thinspace Rankin 2015) is
surprising as there are a number of effects that would tend to
misalign a pulsar's rotation and proper motion direction even if the
natal ``kicks'' are initially aligned ({\it e.g.,}\thinspace Johnston
{\it et al.}\thinspace\ 2005). Noutsos {\it et al.}\thinspace\ (2012) in particular note the effects of binary disruption
which might impose a large velocity component orthogonal to the
rotation direction. We know little about B1933+16's specific origins
or natal binary system, but it is worth noting that there are four
known pulsars within a degree or so of it and at about the same
dispersion measure. So, the canted $\Psi$ value above should not
immediately be taken as evidence against our association of its
primary core emission with the X mode. We clearly have much yet to
learn both about why $\Psi$ values tend to close alignment and why
they depart from said.
\section{Intensity Fractionation of the Total Polarized Profiles}
\label{sec4}
Fig.~\ref{fig2} gives intensity-fractionated average profiles for
the two observations. The 1.5-GHz display in the upper plot exhibits
rather little change, showing that the pulsar's emission is very
steady from pulse to pulse. There is not enough intensity variation
in its pulses to show large differences. This said, we can see a
subtle but consistent pattern at both frequencies: the leading portion
of the core feature becomes slightly but progressively weaker relative
to the trailing portion with increased intensity. Note also that
while the aggregate linear power in the trailing portion of the
1.5-GHz profile depends little on intensity, the leading portion shows
an increase.
\section{Segregating the Polarization Modes}
\label{sec5}
In order to better study the physical emission properties, we next
segregate the modal emission in the 1.5-GHz PS into two sequences.
The procedures for the operations were discussed in Deshpande \&
Rankin (2001; Appendix), and here we first consider the two-way
segregation, which implicitly assumes that the depolarization is modal
in origin, and thus the procedure partially repolarizes the modal
sequences. Average profiles corresponding to the two modes are given
in Fig.~\ref{fig3}, where the primary polarization mode (PPM) is
plotted in the upper panels and the secondary (SPM) in the lower ones.
Dashed (magenta) RVM curves corresponding to the OPMs are indicated,
and the respective regions of polarized samples contributing to the
average profiles (exceeding a 2-$\sigma$ threshold) clearly shown.
\begin{figure}
\begin{center}
\begin{tabular}{c}
\mbox{\includegraphics[height=105mm,width=75mm,angle=-90.]{srB1933+16.57119l3p.ps}}\\
\mbox{\includegraphics[height=105mm,width=75mm,angle=-90.]{srB1933+16.57119l3s.ps}}\\
\mbox{\includegraphics[height=105mm,width=75mm,angle=-90.]{srB1933+16.57119l3u_mod.ps}}\\
\end{tabular}
\caption{Polarization profiles and PPA distributions after Figure~\ref{fig1} for three-way
polarization-mode PS segregations of the 1.5-GHz observation per
Deshpande \& Rankin (2001); primary mode (top), secondary mode
(middle) and unpolarized (bottom). PPA samples larger than 5$\sigma$
in the linear polarization are plotted in the histogram.}
\label{fig4}
\end{center}
\end{figure}
This polarization-modal segregation shows that the trailing portion of
the core feature plays the strongest role in the PPM and the leading
in the SPM. We also see that most of the conal power is in the SPM.
The two-way modal partial profiles in Fig.~\ref{fig3} are repolarized
following the assumption that the total profile is depolarized due
only to the effects of the two orthogonal polarization modes ;
however, the three-way modal segregations in Fig.~\ref{fig4} are
polarized similarly on their own, suggesting that the assumption is
not wrong.
This all said, it is the large negative `U'-shaped traverse of the PPA
in the 1.5-GHz profile at --5\degr\ that draws the eye. This feature
is much more prominent in this observation than any other previous one
because of its high resolution---and thus minimal mode mixing within
individual samples. Many other average profiles show only a ``bump''
at this point ({\it e.g.,}\thinspace see Johnston {\it et al.}\thinspace\ 2005). We here see that the
source of this abrupt shift in the average PPA is an aberrant
``patch'' of PPAs at about --50\degr, which falls close to, but
apparently just above, the SPM modal track. Moreover, the adjacent
regions are unusually prominent here just because of the
non-orthogonality of the --50\degr\ power, as linear polarization
mixing within even these highly resolved samples tends to result in
systematically progressive PPAs.
This interesting configuration is prominent in all of our analyses in
different ways: in the 1.5-GHz total average profile of
Fig.~\ref{fig1}, one can see that the depolarization around --5\degr\
longitude results from PPAs of almost every value. At 4.6-GHz the
situation is less clear because there are so many fewer qualifying PPA
values. In the two-way segregations of Fig.~\ref{fig3}, the
repolarization code produces an artifact primarily in the SPM. Only
in the upper two panels of the three-way segregations do we see a
fairly clear indication of the actual PPA populations here.
The three-way segregations thus give the clearest picture of the
polarization-modal structure of the profile in the --5\degr\ region:
the SPM power in the middle display fits precisely into the region
where PPM power is missing in the top display. Weak positive circular
polarization is associated with the trailing part of the core;
however, the leading SPM power in the middle display is almost
completely negatively circularly polarized.
\begin{figure*}
\begin{center}
\includegraphics[height=\textwidth,angle=-90.]{B1933+16-300-0p1.eps}
\caption{\label{fig5-1} Three examples of our analysis methods for extracting subpulse
timescales in the 1.5-GHz core emission, and the three columns
correspond to Stokes $I,$ $L$ and $V$. This is similar to MAR15's
fig. 5, but here only the smoothing bandwidth $h=0.1$ is shown. The
grey lines in the uppermost panels show the subpulse amplitudes, the
fits to them (black lines), the envelopes (red lines), and the
microstructure signal (blue lines), computed as the difference between
the subpulse amplitude and envelope. The second panel shows the ACF
of the microstructure signal. For a detailed description of the
technique refer to MAR15.}
\end{center}
\end{figure*}
\begin{figure*}
\begin{center}
\includegraphics[height=\textwidth,angle=-90.]{B1933+16-0p1-hist3.eps}
\caption{\label{fig5-2} The lower three panels are histograms of the estimated time scales
of the microstructure signals and similar to fig. 6 of MAR15. The
upper plots of these three panels show the relation between the first
dip and the first peak in the ACF, the middle shows the distribution
of the first dip in the ACF and the bottom panel shows the
distribution of the first peak in the ACF which is $P_{\mu}$. For a
detailed description of the technique refer to MAR15.}
\end{center}
\end{figure*}
\section{Interpreting the Emission Geometry}
\label{sec6}
Although the 1.5-GHz polarization display in Fig.~\ref{fig1} shows a
PPA distribution of unusual complexity (and one is initially at pains
to understand what the 4.6-GHz PPA ``patches'' have to do with the RVM
fit at all!), we believe the RVM fitting to the PPAs in the 1.5-GHz
observation is correct and consistent in the following ways: First,
exempting the region around longitude --5\degr, we can follow a PPA
trajectory under the core feature that begins near --9\degr, sweeps
strongly negative under the trailing edge of the core and includes the
weaker trailing conal PPA ``patch'' at some +45\degr. We already
noted that an RVM curve cannot be fitted if the brighter trailing
patch is included. The RVM fit then gives an inflection near
--58\degr. Second, note that the bright conal PPA ``patches'' both
fall near the SPM curve. And third, we see that the PPM inflection
point falls near the midpoint or average of the two SPM conal
``patches'' as it must if the RVM model is to be consistent for both
the core and conal regions.
At 4.6 GHz, we have much less to go on. If this were the only profile
available, no sound interpretation could be made. However, we note
that the three conal PPA ``patches'', one leading and the two
trailing, fall at nearly identical positions in the PPA distributions
of Fig.~\ref{fig1} that have been derotated to infinite frequency. So
it is the core region that is mainly different, but even here the PPA
``patch'' at about --1\degr\ falls at about the same place at the two
frequencies.
The effort in ET VI to interpret B1933+16's geometry confronted the
non-Gaussian shape of the pulsar's central feature and remarkably took
twice the peak-to-trailing edge distance, some 4.3\degr, as the ``core
width'' corresponding to the observed angular diameter of the polar
cap. $\alpha$, $\beta$ estimates of 72\degr\ and 1.1\degr\ followed
which are trivially different that the 125\degr\ (using the Everett \&
Weisberg convention) determined by the RVM fitting above. The paper's
discussion passed over in silence how the unresolved leading feature
might be interpreted.
The dramatic increasing prominence of B1933+16's conal outrider pair
leaves little doubt, then, that its high frequency profiles represent
a core/cone configuration, but it is less clear just how to classify
it. As we have seen in the 1.5-GHz profile, the conal power is very
broadly distributed and does not consist in neat components. Paper ET
VI then gave B1933+16 as a rare example of a core triple {\bf S$_t$}
pulsar with outer conal outriders, but its closer, better defined
conal components at 4.6 GHz defeat this interpretation. Notice the
very different shapes and extents of the conal components at the two
bands in Fig.~\ref{fig1}. Although no clear double-cone structure can
be discerned, the conal emission is so broad at 1.5-GHz (extending
over some 30\degr\ with half-power points around 22\hbox{$^{\circ}$}), that outer
conal power must dominate at this frequency. Further, broad conal
power can be seen in the recent 775-MHz profile of Han {\it et al.}\thinspace\ (2009),
such that the pulsar could be regarded as having a triple {\bf T} or
even five-component {\bf M} profile. However, these sometimes
``fuzzy'' 1-GHz profile classes all reflect and illuminate the
underlying core/double-cone structure of pulsar radio emission beams.
In any case, using the best determined values of $\alpha$, $\beta$ and
$R_{PA}$ above, the observed angular diameter of the polar cap would
be 5.0\degr, and the roughly 16\degr\ outside half-power width of the
conal outriders at 4.6 GHz definitely identifies them as having the
dimensions of an inner conal beam. Thus using this profile-width
information together with $R_{PA}$ and the quantitative geometry of
Paper ET VI permits us to confirm that the fitted alpha and beta
values were close to correct.
\section{Microstructure properties}
\label{sec7}
The bright single pulses of PSR B1933+16 in conjunction with the high
time resolution made possible by the Arecibo observations provide an
ideal context for studying the properties of microstructures in this
pulsar. Close visual inspection of single pulses at both 1.5 and 4.5
GHz reveals that the bright core component is rich in small timescale
fluctuations in all the Stokes parameters akin to the broad
microstructures recently studied by Mitra, Arjunwadkar \& Rankin
(2015, MAR15 hereafter). The top panels of Fig.~\ref{fig5-1}
and \ref{fig5-2} show the analysis steps of a single 1.5-GHz pulse as
a typical example of microstructure. A quantitative analysis of the
pulsar's microstructure properties as carried out by MAR15 was
performed on the strong core component, and the periodicities obtained
for the smoothing parameter $h=0.1$ were $P_{\mu}^{I} \sim 0.4
\pm 0.2, P_{\mu}^{L} \sim 0.47 \pm 0.2, P_{\mu}^{V} \sim 0.47 \pm 0.2$ msec, respectively.
MAR15 found a relation for $P_{\mu}$ that increases with increasing
pulsar period $P$ for $h=0.1$ as $P_\mu (msec) = 1.3 \times 10^{-3} P
+ 0.04$, predicting a value of about 500 $\mu$sec for PSR B1933+16,
which is in very good agreement with the estimated $P_{\mu}$. The
single pulses at 4.5 GHz were relatively weaker and the microstructure
analysis could only be done reliably for Stokes $I$ which yielded
$P_{\mu}^{I} = 0.35\pm 0.16$ msec for $h=0.1$.
\begin{figure}
\begin{center}
\begin{tabular}{c}
\mbox{\includegraphics[height=75mm,width=75mm,angle=-0.]{mod1.eps}}\\
\mbox{\includegraphics[height=75mm,width=75mm,angle=-0.]{mod2.eps}}\\
\end{tabular}
\caption{Longitude-longitude correlation displays for zero delay (top) and a one-pulse delay
(bottom). The regions on each side of the diagonal are identical in
the first plot, and in the second the upper region gives delay +1 and
the lower delay --1. Side and lower panels show the average profile
for reference. The 3-sigma error in the correlations is about 4\%.}
\label{fig6}
\end{center}
\end{figure}
\begin{figure}
\begin{center}
\mbox{\includegraphics[height=75mm,width=75mm,angle=-0.]{mod3.eps}}
\caption{Longitude-longitude cross-correlation display at zero delay between the three-way
separated primary and secondary OPM power. Side and lower panels show
the average model profiles (as in Figure~\ref{fig4}) for reference. The
3-sigma error in the correlations is again about 4\%.}
\label{fig7}
\end{center}
\end{figure}
One of the reasons that motivated the microstructure study was that
PSR B1933+16 is the only truly bright core dominated pulsar in the
Arecibo sky, and no published study of microstructure seems to
specifically address bright core emission. The comprehensive Arecibo
study of broad microstructure in MAR15 treats almost all
cone-dominated stars. However, we find a striking similarity between
the timescale observed for B1933+16's core component and those for
bright conal emission.
\begin{table}
\caption[]{A/R radio emission heights for pulsar PSR B1933+16, \\
computed as $r = -c P/4$ $\phi_c(\degr)/360$}
\begin{tabular}{cccccc}
\hline
{\bf Component} & $\nu_{obs}$ & $\phi_l$ & $\phi_t$ & $\phi_c$ & $r$ \\
& (GHz)& (\degr) & (\degr) & (\degr) & (km) \\
\hline
Core & 4.5 & --5.4 & 0.16 & --2.78 & 207 \\
Inner Cone & 4.5 &--10.0 & 3.9 & --3.05 & 227 \\
Outer edge & 4.5 &--14.8 & 11.4 & --3.4 & 253 \\
& & & & & \\
Core & 1.5 & --6.1 & 0.00 & --3.05 & 227 \\
Inner Cone & 1.5 &--10.9 & 4.5 & --3.2 & 238 \\
Outer edge & 1.5 &--16.8 & 11.6 & --2.6 & 193 \\
& & & & & \\
\hline
\end{tabular}
\label{tab1}
\end{table}
\section{Conal and Core Emission Characteristics}
\label{sec8}
Few opportunities exist to study the properties of the conal emission
in core-single {\bf S$_t$} pulsars because the conal outriders appear
only at high frequency where the overall emission already tends to be
weak relative to meter wavelengths. So it is interesting to carry out
some single-pulse polarimetric analyses on the conal portions of
B1933+16's 1.5-GHz profile.
In order to study the dynamics of the component structure we computed
a longitude-longitude correlation diagram at zero delay, and this is
shown in the top panel of Fig.~\ref{fig6}. Along the diagonal of
complete correlation, several regions are seen that correspond to the
leading conal, core and trailing conal parts of the profile. The core
feature shows as a rectangle of weak positive correlation (yellow)
bounded by weak negative (blue). The conal regions show similar
regions of correlation and interestingly the trailing conal region
seems to bifurcate into an inner and outer region as discussed above.
The lower display in Fig.~\ref{fig6} computes the correlation at
one-pulse delay, and note that the region above the diagonal is no
longer a mirror of that below because the upper portion gives the
correlations at a delay of +1 pulse and the lower at --1. The region
of large positive correlation at about --4\degr\ longitude falls on
the diagonal and the correlation here is perceptible to delays as long
as 5-10 pulses. The area of large negative correlation associates the
leading part of the core with the leading conal component, and this
changes greatly with increasing delay.
We also carried out fluctuation spectra on the leading and trailing
conal regions, but these show no strong periodicities. We see no hint
of intensity-dependent effects as in the core. The
longitude-longitude correlations do hint at subpulse motion, but not
in a very regular manner as in some slower pulsars. It is not very
surprising that we see no regular conal modulation, as such ``drift''
modulation seems almost never to occur in stars rotating as rapidly as
B1933+16.
Returning to core properties, Fig.~\ref{fig7} displays the
correlation between the 3-way separated primary and secondary modal
power. The two different modal profiles that earlier appeared in
Fig.~\ref{fig4} are shown in the side and bottom panels for reference.
First note the bright positive correlation in the early part of the
core feature around --4\degr\ that represents the modal fractions of
power in this region fluctuating together. Also important are the
weaker regions of positive and negative correlation showing how the
later primary-mode core radiation is coupled dynamically to the
earlier power in the secondary mode.
\section{Aberration/Retardation and Geometrical Emission Heights}
\label{sec9}
Pulsar B1933+16 exhibits a core-cone geometry. The prominent conal
peaks are seen clearly at 4.5 GHz adjacent to the central core
component. At 1.5 GHz the low level conal emission extends far on
both the leading and trailing sides of the profile, however no clear
outer conal peaks are discernible. Our analysis of the segregated
modes exhibits the largely RVM character of the PPA traverse, and this
prompts us to estimate the emission heights ($r$) for both the core
and conal radiation using the A/R method.
In this method the height $r$ can be determined according to the A/R
shift $\Delta t$=$4r/c$ [=$\Delta\phi P/360\degr$], where $P$ is the
pulsar period in seconds and $\Delta\phi$ is the longitude difference
in degrees between the center of the core or cone emission obtained
from the total intensity profiles and the steepest gradient point of
the PPA traverse ({\it i.e.,}\thinspace Blaskiewicz {\it et al.}\thinspace\ 1991; Mitra \& Li 2004; Dyks
2008). $\Delta\phi$ requires determination of the profile center in
terms of the longitude difference between a trailing feature $\phi_t$
and its leading counterpart $\phi_t$, such that $\Delta\phi = \phi_c
- \phi_0$ where $\phi_c = \phi_l + (\phi_t-\phi_l)/2$ and $\phi_0$ is
the ``fiducial'' longitude at the steepest gradient point. Several
assumptions and systematic errors are associated with $\phi_l$,
$\phi_t$ and $\phi_0$ measurements as discussed in Mitra \& Li
(2004). Also, the A/R method usually assumes an RVM PPA traverse from
a single emission height. Aggregate emission from several heights
could compromise interpretation of the SG point. Our purpose here,
however, is to estimate emission heights which are small compared to
the light-cylinder radius.
We apply the A/R method first to B1933+16's conal emission. At 1.5
GHz the peak location of the inner conal emission based on the average
profile was not easily discernible. Rather in some low S/N
intensity-segregated profiles the peaks of the cone were visible, and
we used these locations to identify the center of the cone. The
situation is relatively simple at 4.5 GHz where the conal peaks are
prominent. We use the profiles in Fig.~\ref{fig1} which are plotted
in a manner such that the longitude origin is taken at $\phi_0$ and
estimate $\phi_l$ and $\phi_t$ for the conal peaks of the inner cone.
Additionally we also measure $\phi_l$ and $\phi_t$ at the extreme
outer edges of the profile at an intensity level of five times that of
the noise level. These profile measures are given in
Table~\ref{tab1}, including the estimated emission heights. The
errors in the $\phi_l$ and $\phi_t$ measurements are much smaller than
the systematic errors in this analysis. Typically we find the conal
emission heights to be about 200 km above the neutron star surface.
Next we proceed to determine the core emission height using A/R. We
estimate $\phi_l$ and $\phi_t$ using the half power points on either
side of the core feature for both the frequencies. It is interesting
to see that the midpoint $\phi_c$ falls close to the zero-crossing
point of the sign changing circular. The resulting emission heights
that we obtain are again about 200 km (see Table~\ref{tab1})---very
similar to the conal emission heights. This is a remarkable result.
For the first time we have successfully been able to apply the A/R
tools to estimate the physical emission heights for both the core and
conal radio emission regions within the profile.
\section{Frequency-Independent OPM Emission}
\label{sec10}
In general PPA measurements reflect the orientation of the linear polarization electric
vector with respect to a reference direction. In our case we measure the PPAs at 1.5
and 4.5 GHz in an absolute sense (Fig.~\ref{fig1}), which essentially means that the
PPA values at these two frequencies represent the position angle of the linear
polarization with respect to celestial North as the emission detaches from the pulsar
magnetosphere. As a result, comparison of the PPAs across the profile allows us to
directly investigate the frequency evolution of the direction of the linear polarization
for a broad frequency range.
For PSR B1933+16 the PPA comparison is not straightforward due to the
complex nature of the polarization distributions (see \S\ref{sec3}).
In the central core region between --8\degr\ and 0\degr, the PPA
distributions appear to be wildly different at the two frequencies.
Particularly the PPAs near --5\degr\ longitude are concentrated around
--50\degr\ at 1.5 GHz and about --30\degr\ at 4.5 GHz. The rest of
the PPA distribution earlier than about --8\degr\ and later than
0\degr\ longitude correspond to the conal emission and shows
remarkable similarity between the two frequencies. The PPA ``patch''
at longitude --10\degr\ is about PPA --60\degr\ and that at longitude
+5\degr\ about --45\degr---and these are virtually invariant with
frequency. Also the RVM fit that was obtained using at 1.5-GHz PPA
traverse (as shown in Fig.~\ref{fig1}) fits the PPAs at 4.5 GHz in the
conal regions well.
We found an even lower frequency polarization measurement for PSR
B1933+16 at 774 MHz by Han {\it et al.}\thinspace\ (2009). Although the paper does not
explicitly state that their PPA measurements were calibrated in an
absolute sense, the lead author confirms that they were so (Han 2016).
Their PPAs below the leading conal component fall at about --65\degr\
and those of the trailing at about --50\degr, so they compare quite
closely with our values at 1.4 and 4.5 GHz. Of course, this situation
was constructed by measuring the Faraday rotation and derotating the
PPAs to infinite frequency in order to achieve the ``absolute''
calibration. However, it is important here to reverse the question in
order to ask what the broad band PPA alignment indicates about the
propagation of radiation through the pulsar magnetosphere.
\begin{table*}
\caption[]{Summary of A/R core and conal emission heights as in Table~\ref{tab1} for several pulsars in the literature.
}
\begin{tabular}{cccccccccc}
\hline
Pulsar & P & $\dot{P}$ & {\bf Component} & $\nu_{obs}$ & $\phi_l$ & $\phi_t$ & $\phi_c$ & $r$ & Ref \\
&(s)&(10$^{-15}$ s/s)& & (GHz)& (\degr) & (\degr) & (\degr) & (km) & \\
\hline
B0329+54 & 0.714 & 2.05 & Core & 0.3 & & & --1.5 $\pm$ 0.2 & 223$\pm$30 &MRG07 \\
& & & Inner peaks& 0.3 &--6.2$\pm$0.2 &5.0$\pm$0.2&--0.7$\pm$0.2 & 104$\pm$30 & \\
& & & Outer peaks& 0.3 &--13.8$\pm$0.2 &10.0$\pm$0.2&--1.9$\pm$0.2 & 282$\pm$30 & \\
& & & & & & & & & \\
B0355+54 & 0.156 & 4.4 & Core & 0.3 &--15.0 $\pm$ 0.5&--8.0 $\pm$0.5 & --11.5$\pm$0.5 & 373$\pm$15 &MR11 \\
& & & Outer peaks& &--39.7 $\pm$ 0.5& --9.3 $\pm$ 0.4& --15.2$\pm$0.6 & 494$\pm$19 & \\
& & & & & & & & & \\
B0450+55 & 0.340 & 2.37 & Core & 0.3 &--14.5 $\pm$ 0.5&--7.5 $\pm$0.5 & --11.0$\pm$0.5 & 779$\pm$35 &MR11 \\
& & & Outer peaks& 0.3 &−18.5 $\pm$ 0.2&10.8 $\pm$0.2 & --3.9 $\pm$0.2 & 274$\pm$12 & \\
& & & & & & & & & \\
B1237+25 & 1.382 & 9.06 & Core & 0.3 & & & --0.4$\pm$0.1 & 117$\pm$30 &ERM14 \\
& & & Inner peaks& & --4.0$\pm$0.2 &3.3$\pm$0.2 &--0.35$\pm$0.2 &101$\pm$50 & \\
& & & Outer peaks& & --6.4$\pm$0.2 &5.1$\pm$0.2 &--0.6$\pm$0.2 &172$\pm$50 & \\
\hline
\end{tabular}
\label{tab2}
References: MRG07, Mitra, Rankin \& Gupta (2007); MR11, Mitra \& Rankin (2011); ERM13, Smith, Rankin \& Mitra (2013).
\end{table*}
\begin{table*}
\caption[]{Summary of plasma properties for core and conal emission. The frequency $\nu_{B}$ is
the cyclotron frequency estimated for $\gamma_s = 200$, frequency
$\nu_p$ estimated for $\gamma_s = 200$ and two values of $\kappa =
100, 10^4$, the cyclotron frequency estimated for two values of
$\Gamma_s = 300, 600$. The radius of the light cylinder is given by
$R_{lc} = c P/2\pi$ and the height of the radio emission $r$ in terms
of light cylinder distance is given by $R = r/R_{lc}$. The quantities
$\Psi = PA_v- PA_{\circ}$, measured for the primary $\Psi_{PPM}$ and
secondary $\Psi_{SPM}$ polarization modes, are obtained from Rankin
(2015).} \begin{tabular}{cccccccccccc}
\hline
Pulsar & P & $\dot{P}$ & {\bf Component} & $\nu_{obs}$ & $\nu_B$ & $\nu_p$ & $\nu_{cr}$ & $R$ & R$_{lc}$ & $\Psi_{PPM}$ & $\Psi_{SPM}$ \\
&(s) &(10$^{-15}$ s/s)& & (GHz)& (GHz) & (GHz) & (GHz) & & (km) &($^{\circ}$) &($^{\circ}$) \\
\hline
B1933+16 & 0.358 & 6.0 & Core & 4.5 & 4676 &20$-$200 &0.5$-$4.3 & 0.01 & 17093 &+112(5)&+22(5) \\
& & & Inner Cone & 4.5 & 3545 &17$-$174 &0.4$-$4.2 & 0.01 & & & \\
& & & Outer edge & 4.5 & 2561 &15$-$148 &0.5$-$3.9 & 0.01 & & & \\
& & & & & & & & & & & \\
& & & Core & 1.5 & 3545 &17$-$174 &0.4$-$4.2 & 0.01 & & & \\
& & & Inner Cone & 1.5 & 3076 &16$-$162 &0.5$-$4.1 & 0.01 & & & \\
& & & Outer edge & 1.5 & 5769 &22$-$222 &0.5$-$4.5 & 0.01 & & & \\
& & & & & & & & & & & \\
B0355+54 & 0.156 & 4.4 & Core & 0.3 & 451 &9.4$-$94.4&0.6$-$4.9&0.05&7448 &+89(5) &--1(5) \\
& & & Outer peaks& & 195 &6.1$-$62 &0.5$-$4.2&0.07& & & \\
& & & & & & & & & & & \\
B0450+55 & 0.340 & 2.37 & Core & 0.3 & 54 & 2.2$-$22&0.3$-$2.3 & 0.05 & 16233&+86(16)&--4(16) \\
& & & Outer peaks& 0.3 & 1234 & 10$-$105&0.5$-$3.9 & 0.02 & & & \\
& & & & & & & & & & \\
B0329+54 & 0.714 & 2.05 & Core & 0.3 &3087 & 11$-$115& 0.4$-$3& 0.006 & 34090 &+99(4)&+9(4) \\
& & & Inner peaks& 0.3 &30438& 36$-$362 &0.5$-$4.3& 0.003& & & \\
& & & Outer peaks& 0.3 &1526 & 8$-$81 &0.3$-$2.6& 0.008& & & \\
& & & & & & & & & & \\
B1237+25 & 1.382 & 9.06 & Core & 0.3 &19910& 21$-$210& 0.4$-$3.0 &0.002&65890 &+143(4)&+53(4) \\
& & & Inner peaks& 0.3 &30612& 26$-$261& 0.4$-$3.1 &0.002& & & \\
& & & Outer peaks& 0.3 &6198 & 11$-$117& 0.3$-$2.5 &0.002& & & \\
\hline
\end{tabular}
\label{tab3}
\end{table*}
\section{On the Structure of Core Components}
\label{sec11}
The obvious composite structure of B1933+16's core component has long
been perplexing. An idea had developed that core components would be
Gaussian shaped, however those we have studied closely are not, and
perhaps the poor resolution of many older observations had contributed
to this incorrect view. In any case we must now attempt to interpret
the origins of this structure
The two parts of B1933+16's core component are clearly demarcated by
their differing circular and linear polarization in Figs.~\ref{fig1}
and \ref{fig2}. The trailing half of the feature has positive
circular and substantial linear, whereas the leading half is linearly
depolarized and shows negative circular. In earlier work we at some
points tried to understand this two part structure in terms of a
missing ``notch'', but we now see strong evidence that the leading and
trailing parts of the core have very different polarizations and
dynamics.
Paper ET VI interpreted the trailing part of the core as corresponding
to the full width of the polar cap. The core width there was measured
as twice the peak to trailing half power point or some 4.3\hbox{$^{\circ}$},
suggesting a nearly orthogonal value of $\alpha$ (77\degr). The full
core feature width suggested a less orthogonal geometry which appeared
incompatible quantitatively with the steep $R_{PA}$. The issue
remains here: we fitted Gaussians to the pulsar's 1.5-GHz profile,
finding that the central core region was well fitted by two Gaussians,
the trailing one with a halfwidth of about 4.5-5\degr, and a leading
one some 3\degr\ earlier of about 2\degr\ halfwidth and half the
amplitude. This analysis is then compatible with the earlier
assumption that the trailing region is associated with the full polar
cap emission and that the leading feature is ``something extra''.
This line of interpretation is also compatible with the more complete,
probably X-mode linear polarization of the trailing region---and seen
in many other core features (ET XI)---and the complex, depolarized
largely O-mode emission in the earlier part. The polarization mixing
around --5\degr\ longitude results in the strongly diverted average
PPA traverse and coincides with the peak of the leading fitted
``extra'' Gaussian above. Both polarization modes are clearly present
in this region and mix both to produce the linear depolarization and
the splay of different polarization states seen in individual samples.
Further, we have seen that higher resolution exhibits the diversion of
the average PPA in this region ever more strongly, so the ``extra''
emission in this narrow region is probably O mode such that its
``patch'' of PPAs in Fig.~\ref{fig1} (upper) would fall on the modal
RVM track if our resolution was even higher.
The distinction between the trailing core region and the ``extra''
leading emission is also seen in the contrasting forms of the
mode-segregated profiles. We can also see that something peculiar is
happening dynamically at --5\degr\ longitude in the persistent
correlation at 1-pulse delay (Fig.~\ref{fig6} lower), and this effect
is further clarified in Fig.~\ref{fig7} where strong positive
correlation is indicated between the power of the two modes in this
region. However, it is important to note that the peak falls well off
the diagonal of the diagram, indicating correlation between the modal
power at different longitudes.
Our earlier pulse-sequence polarimetric studies of core emission in
pulsars B0329+54 (Mitra {\it et al.}\thinspace\ 2007) and B1237+25 (Smith {\it et al.}\thinspace\ 2013)
also exhibited core features of two parts, marked by distinct leading
and trailing regions of modal emission and hands of circular
polarization. Our analyses of these two pulsars went in somewhat
different directions because their cores were modulated by
intensity-dependent A/R, whereas B1933+16's emission is much steadier.
A further difference was that the ``extra'' leading emission had
smaller relative intensity, so that the core width could be directly
interpreted geometrically.
\section{Summary}
\label{sec:summary}
Using high resolution polarimetric Arecibo observations of pulsar
B1933+16 at 1.5 and 4.6 GHz, we have conducted a comprehensive
analysis of this brightest core dominated pulsar in the Arecibo sky.
Here we summarize the main results W of our analysis:
\begin{itemize}
\item With the exception of one region near --5\degr\ longitude, the 1.5-GHz PPA
traverse can be well fitted by an RVM curve. This shows that most of the core
power is in the primary mode and most of the conal in the secondary.
\item The 4.6-GHz core polarization is much more difficult to interpret, whereas
the conal PPAs are nearly identical to their counterparts at 1.5 GHz. Imposing
the 1.5-GHz RVM fit on the 4.6-GHz PPA provides useful insight on how the
latter should be interpreted.
\item High resolution analysis reveals the apparently non-RVM region of emission
at about --5\degr\ longitude with unprecedented clarity.
\item The 1.5-GHz RVM fit gives $\alpha$ and $\beta$ values of 125\degr and --1.2\degr,
respectively, with large errors due to the nearly complete correlation between them.
The resulting PPA at the inflection point is --58\degr$\pm$5\degr, where we have taken
the longitude origin in our various plots. See Fig.~\ref{fig1}. Efforts were made to
interpret the several different core and conal modal ``patches'' in different ways, and
we believe the above fit properly represents the pulsar's actual emission geometry.
\item Such a ``fiducial'' PPA of --58\degr\ together with the measured $PA_v$ direction of
+176(0)\degr\ indicates polarization orientation +54\degr$\pm$5\degr. Despite this
measured cant from 90\degr, we believe that it probably still identifies the extraordinary
(X) propagation mode.
\item An intensity-fractionation analysis showed only weak differences in the several
populations of pulses in B1933+16, in large part because the pulsar emission tends
to be very steady from pulse to pulse. See Fig.~\ref{fig2}.
\item However, B1933+16's core structure is similar to that seen before in B0329+54
and B1237+25---that is, it shows a double modal structure with both opposite senses
of circular polarization and orthogonal linear polarization. In these other pulsars, the
two parts are strongly correlated dynamically, but the connection is less apparent
in B1933+16 due to its steady intensity from pulse to pulse. Fig.~\ref{fig7}, however,
shows the highly correlated mixed-mode power early and its bridge to the later X-mode
emission.
\item The 1.5-GHz modal emission was segregated using both the two- and three-way
techniques as seen in Figs.~\ref{fig3} and \ref{fig4}. These show that the later parts of
the core are dominated by X-mode emission, whereas the earlier parts represent the
O mode.
\item Analysis of broad microstructures in the emission of B1933+16's core component
shows that their timescales are nearly identical with the similar largely conal emissions
studied earlier----in this case some 500 $\mu$sec. See Fig.~\ref{fig5-1} \& \ref{fig5-2}.
\item Longitude-longitude correlations and fluctuation-spectral analyses of B1933+16's
1.5-GHz emission show distinct conal and core regions. The former again show a mix
of inner and outer conal radiation. Fluctuation spectra provide no signs of periodicity.
\item Aberration/retardation analyses provide physical emission height measurements
for B1933+16's conal emission and unusually here for its core radiation as well. These
computations in Table~\ref{tab1} suggest that both the conal and core radiation stems
from an average height of around 200 km above the neutron-star surface.
\item Conal linear polarization in B1933+16 exhibits very similar absolute PPAs
between 0.77 and 4.6 GHz after derotating them according to the small interstellar
Faraday rotation exhibited by the pulsar.
\end{itemize}
\section{Theoretical Interpretation and Discussion}
\label{sec:discussion}
The enormous brightness temperatures of pulsar radio emission demand a
coherent plasma radiation mechanism. Relating physical theories of
coherent radio emission to the observations requires above all else
one crucial datum: the location of the regions within the
magnetosphere where the core and conal radio emission are emitted. The
A/R method for determining radio emission heights is physically
grounded and little model dependent, but it has been successfully
applied more to the conal emission, where the RVM can more usually be
fitted to conal PPA traverses and the SG points accurately determined.
Cores have too often presented complex traverses for which it could be
doubted whether an RVM fit was appropriate or even possible. Some
earlier efforts were made to measure A/R heights for cores (Mitra \&
Li 2004), but the results have remained inconclusive and in some cases
seemed to show the wrong sense.
In an effort to resolve important questions about cores---and their
emission heights in particular---we have embarked on a systematic
study of pulsars with bright core components as well as prominent
conal emission. This paper is the third in a series that began with
pulsar B0329+54 (Mitra, Rankin \& Gupta 2007) and then studied
B1237+25 (Smith, Rankin \& Mitra 2013). Each of the pulsars exhibits
distinct conal and central core emission components along with a
complex PPA traverse. By using high quality single pulse polarimetry
and PPA mode segregation techniques, we were able to identify the
underlying RVM as we did earlier for B0329+54.
This in turn permitted us to estimate the core and conal A/R shifts
that are here summarized in Table~\ref{tab2} (refer to notes in the
Table for details about individual measurements). Amongst these three
pulsars PSR B1933+16 is the fastest and has the best determined A/R
shift. We searched the literature and found that for two more
pulsars, PSR B0355+54 and PSR B0450+55, the A/R-based core and conal
shifts could be determined with confidence---{\it i.e.,}\thinspace we could identify
their components clearly as core or conal and then fit the RVM to
their PPA traverses. As seen in Table~\ref{tab2} the conal and core
centers $\Delta \phi_c$ lead the SG point in each case, which in turn
suggests that the core and conal emission arises from a finite heights
above the polar cap. However, unlike for B1933+16, the core and conal
centers do not coincide, and hence the A/R shift method, using the
simple relation $r=c\Delta\phi P/4 \times 360^{\circ}$, cannot be used
to estimate these emission heights accurately. There can be a few
reasons for this, for example: the methods employed to measure the
emission centers are not sufficiently accurate (see Mitra \& Li 2004);
the emission patterns are not centered about the dipole axis: or
finally the core and conal emission arises from different heights.
Our aim here, however, is not to find exact emission heights, but to
illustrate that the radiation arises from deep within the
magnetosphere at altitudes of a few hundred kilometers. In this
spirit we simply estimate the core and conal emission heights $r$ in
Table~\ref{tab2} assuming their centers are shifted by A/R. However,
the excellent RVM fits to the PPA traverses in all these cases also
implies that the height differences between the core and conal
emission must be relatively small, as large differences can cause
distortions and kinks in the PPA traverse (see Mitra \& Seiradakis
2004).
These measurements, taken together, provide definitive evidence, we
believe, that the core and conal (hence all) radio emission arises in
regions at heights of typically no more than a few hundred kilometers
above the neutron-star surface. .
Although the PPA traverses of these pulsars are complex and heretofore
inscrutable in RVM terms, our techniques of high resolution
polarimetry, intensity fractionation and modal segregation have been
able to reliably identify their underlying RVM traverses. B1933+16
was the most complex in these terms we have studied, and we began our
work presuming that its traverse must entail a coherent combination of
polarized power to so distort its observed PPA traverse. However, as
we have seen, this is not the case: the non-RVM distortions in each of
these pulsars are primarily polarization-modal in origin. The two
orthogonal polarization (PPM and SPM) modes undergo incoherent mixing
of varying intensities and degrees of orthogonality across the pulse
profile---with the mixing occurring on such short timescales that our
high resolution observations can partially resolve them. These
effects are particularly acute in the core emission, and there also
complicated by intensity-dependent A/R.
Our ability to trace the PPM and the SPM across the profile and to
identify them with the absolute $PA_{\circ}$ together with the $PA_v$
direction further permits us to associate the OPMs with the X and O
propagation modes in the pulsar magnetosphere. This is then
summarized in Table~\ref{tab3}. For pulsars B0329+54, B0355+54 and
PSR B0450+55 the association of the PPM with the X and the SPM with
the O mode is very good. This is in general agreement with the
measured $\Psi$ distribution for core dominated pulsars (Rankin 2015)
where the PPM emission for core components is associated with the X
mode. For PSR B1933+16 the association is less precise and B1237+25's
canted orientation may stem from its position near the North Galactic
pole.\footnote{It remains surprising that $PA_v-PA_0$ for many pulsars
falls close to either 0 or 90\degr. Excellent polarimetry and careful
analysis has determined $PA_0$ values for these pulsars with
certainty, and the proper-motion directions measured by both VLBI and
timing are in good agreement with each other. To the extent that
pulsar velocities are due to natal supernova ``kicks'', the one
alignment can be understood, but binary disruption would tend to
produce the other. So we have much to learn both about why the
alignments tend to be close to 0 or 90\degr\ as well as why some show
significant misalignments.} In any case, the observations now
strongly suggest that the emission emerging from the pulsar
magnetosphere is comprised of the X and O modes.
The observational evidence locating the pulsar radio-emission region
and associating the radiation with the X and O plasma modes have
strengthened our ability to identify the curvature-radiation mechanism
as responsible for exciting coherent radio emission in pulsars ({\it e.g.,}\thinspace
Melikidze, Gil \& Pataraya 2000, hereafter MGP00; Gil, Luybarski \&
Meikidze 2004, hereafter GLM04; Mitra, Gil \& Melikidze 2009;
Melikidze, Mitra \& Gil 2014, hereafter MMG14). In most pulsar
emission models the the rotating neutron star is a unipolar inductor
due to the enormous magnetic fields, and a strong electric field is
generated around the star. In such strong electric and magnetic
fields charged particles can be pulled out from the neutron star
and/or be created by the process of magnetic pair creation. Finally
the region around the neutron star becomes a charge-separated
magnetosphere which is force free. The physics of how particles
populate and flow in the pulsar magnetosphere is a matter of intense
research; however, the consensus is now that, in the presence of
sufficient plasma, a relativistic plasma flow can be generated along
the global open dipolar magnetic field [see Spitkovsky (2011) for a
review]. In order to achieve a force-free condition and to maintain
corotation the magnetosphere needs a minimum charge density
(Goldreich \& Julian 1968) of $n_{GJ} = \Omega \cdot B/2 \pi c$
(rotational frequency $\Omega = 2 \pi / P$, where $P$ is the pulsar
period, $B$ is the magnetic field and $c$ is velocity of light). A
key aspect of the magnetosphere is that enormous electric fields will
be generated in regions depleted below the $n_{GJ}$ value, where the
remaining charges can be accelerated to high energies and more charges
produced in that region.
In order to explain the origin of pulsar radio emission arising
relatively close to the star, a charge-depleted acceleration region
just above the polar cap is found to be necessary. Such a prototype
region was envisaged as an inner vacuum gap (IVG) by Ruderman \&
Sutherland (1975, RS75). The gap has strong electric and magnetic
fields and is initially charge starved. Apparently, the IVG can
eventually discharge as electron-positron pairs in the form of
``sparks'' at specific locations, and corresponding non-stationary
spark-associated relativistic primary particles with Lorentz factors
of $\gamma_p$ can be generated. This discharge process continues
until the force-free charge density $n_{GJ}$ is reached. The
electron-positron pairs are separated due to the electric field in the
IVG, and one kind of charge streams outward towards the upper
magnetosphere further radiating in the strong magnetic fields and the
resulting photons thereby producing secondary electron-positron plasma
with Lorentz factor $\gamma_s$. Thus the charge density in the
outflowing secondary plasma $n_s$ gets multiplied by a factor
$\kappa \sim n_{s}/n_{GJ}$ (Sturrock 1971). The backflowing plasma
then heats the polar cap surface and can generate thermal x-ray
emission. For the above process to work, the RS75 model assumed a
magnetic curvature radius of 10$^6$ cm in the vacuum gap, which in
fact implies strongly non-dipolar fields (see {\it e.g.,}\thinspace Gil, Melikidze
\& Mitra 2004). As a consequence the range for $\gamma_p \sim 10^{6}$, $\nu_s \sim$ few
hundreds, and $\kappa \sim 100-10^{4}$. The RS75 model had some shortcomings, however,
as it could not explain the slower-than-expected subpulse-drifting effect (Deshpande \& Rankin
2001), and it also predicted higher temperatures for the hot polar caps than were observed.
A refinement of the model by GLM04 suggested that a partially screened IVG can be
realized that can explain these effects. In any case, the core and conal radio emission are
generated through the growth of plasma instabilities in the spark-associated relativistically
flowing secondary plasma.
Knowledge of the location of the pulsar emission and using the above
model, it is possible to compute the various frequency scales involved
in the pulsar emission problem. Following eqs. (10), (11) and (12) in
MMG14, the cyclotron frequency $\nu_{B}$, and plasma frequency
$\nu_{p}$ at a fractional light cylinder distance $R$ for a pulsar
with period $P$ and $\dot{P} = \dot{P}_{15} \times 10^{-15}$ can be
expressed as $\nu_{B} = 5.2
\times 10^{-2} (1/\gamma_s) \times (\dot{P}_{15}/P^5)^{0.5} R^{-3}$ GHz and $\nu_{p} = 2 \times
10^{-5} \kappa^{0.5} \sqrt{\gamma_s} (\dot{P}/P^7)^{0.25} R^{-1.5}$
GHz. In Table~\ref{tab3} we provide the values of $\nu_{B}$ and
$\nu_{p}$ for a reasonable value of $\gamma_s =200$ and two values of
$\kappa = 100 , 10^4$, respectively. For all cases we find $\nu_{p}
< \nu_{B}$, and further the the observed radio emission $\nu_{obs}
< \nu_{p}$, or in other words the observed frequency of radio emission
is lower than the plasma frequency of the emitting plasma.
A model of pulsar radio emission for $\nu_{obs} < \nu_{p} < \nu_{B}$
has been developed by MGP00. It is important to notice that the only
known instability that can operate at low radio emission heights
(typically below 10\% of the light cylinder) is the two-stream
instability. The non-stationary sparking IVG discharge leads to
generation of secondary plasma clouds with slight spreads in their
particle velocities. The faster and slower velocities of two
successive clouds overlap at the heights of radio emission to trigger
strong Langmuir turbulence in the secondary plasma, which can become
modulationally unstable. MGP00 demonstrated that nonlinear growth of
the modulational instability can lead to formation of charged solitons
which are capable of emitting curvature radiation in curved magnetic
fields. The soliton size must be larger than the linear Langmuir
wave, and to maintain coherence the emission should have wavelengths
larger than the soliton size, or in other words $\nu_{obs} < \nu_{p}$
as observed. The theory suggests that the maximum frequency of the
soliton coherent curvature radiaiton given by eq. (12) in MMG14 is
$\nu_{cr} = 0.8 \times 10^{9} (\Gamma^3/P) R^{-0.5}$ GHz, where
$\Gamma$ is the Lorentz factor of the emitting soliton. The value of
$\Gamma$ is slightly different from $\gamma_s$, and assuming a two
reasonable values of $\Gamma = 300,600$ we estimate values of
$\nu_{cr}$ as given in Table~\ref{tab1}. Clearly the values lie in
the observed frequency range---{\it i.e.,}\thinspace $\nu_{cr} \sim \nu_{obs}$.
Finally we turn our attention to the prediction of the coherent
curvature radiation theory for the X and O modes and their propagation
in the magnetosphere. We have emphasized in \S\ref{sec10} that in PSR
B1933+16 the linear polarization direction remains unchanged for conal
emission between 0.77 and 4.5 GHz. In turn, if we invoke the RVM
solution, we can conclude that with respect to the magnetic field line
planes there is no rotation or adiabatic walking of the linearly
polarized conal modal emission in the pulsar magnetosphere. We note
that several current studies suggest the invariance of the PPAs with
frequency. One class of investigation involves establishing the
rotation measure variation as a function of pulse phase
(Ramachandran {\it et al.}\thinspace\ 2004; Noutsos {\it et al.}\thinspace\ 2009), and these studies
indicate that the rotation measure across a pulsar's profile is
largely due to the interstellar medium. Another study by
Karastergiou \& Johnston (2006) compared absolute PPAs between 1.4 and
3.1 GHz and found very little change with frequency.
If the PPA modes are interpreted as the X and O modes, then the
absence of adiabatic walking, at least for the X mode, can be
understood in the framework of the propagation effects predicted by
the coherent curvature radiation theory as shown in a recent work by
MMG14. This theory demonstrated that for radio emission to be excited
and detach from the magnetosphere below 10\% of the light cylinder
(which is also the case for PSR B1933+16 as shown above), the
refractive index of the emitting plasma for the X-mode propagation is
unity, and hence the X mode can emerge from the plasma preserving its
emitted polarization direction. However the theory has no prediction
for the O mode nor for the existence of circular polarization. These
appear to be two fundamental problems that need to be resolved in
pulsar emission theories as emphasized in the latter paper.
\section{Conclusion}
In this paper we have established that the core and conal emission lie
at heights of no more than several hundred kilometers, core and conal
emission display similar short timescale microstructures, and their
emission can also be associated with the X and O modes. These
similarities strongly suggest, as expected, that the core and conal
emission processes have the same physical origin. However, we have
yet to understand the causes of their individual geometric,
polarization and modulation properties. We argue that, overall, the
pulsar radio emission mechanism is excited by coherent curvature
radiation.
\section*{Acknowledgments} We thank our referee Jarsalow Dyks for providing constructive comments
which helped in improving the paper. Much of the work was made
possible by support from the NASA Space Grants and from US National
Science Foundation grant 09-68296. One of us (JMR) also thanks the
Anton Pannekoek Astronomical Institute of the University of Amsterdam
for their support. Arecibo Observatory is operated by SRI
International under a cooperative agreement with the US National
Science Foundation, and in alliance with SRI, the Ana
G. M\'endez-Universidad Metropolitana, and the Universities Space
Research Association. We thank Prof. Jinlin Han for sharing their
774-MHz polarimetry of PSR B1933+16. This work made use of the NASA
ADS astronomical data system.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 1,111
|
{"url":"https:\/\/carmamaths.org\/eventprint.php?n=709","text":"# CARMA Discrete Mathematics Seminar\n\n## Thursday, 5th Mar 2015\n\nV106, Mathematics Building\n\n# Dr Yuqing Lin\n\n(School of Electrical Engineering and Computer Science, The University of Newcastle)\n\n# The number of maximal state circles of plane graphs\n\nIt is well known that there is a one-to-one correspondence between signed plane graphs and link diagrams via the medial construction. The relationship was once used in knot tabulations in the early time of study in knot theory. Indeed, it provides a method of studying links using graphs. Let $G$ be a plane graph. Let $D(G)$ be the alternating link diagram corresponding to the (positive) $G$ obtained from the medial construction. A state $S$ of $D$ is a labeling each crossing of $D$ by either $A$ or $B$. Making the corresponding split for each crossing gives a number of disjoint embedded closed circles, called state circles. We call a state which possesses maximum number of state circles a maximum state. The maximum state is closely related to the genus of the corresponding link, thus has been studied. In this talk, we will discuss some of the recent progress we have made on this topic.","date":"2023-01-30 17:50:23","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\": 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\": 0.5007322430610657, \"perplexity\": 329.3406929392378}, \"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-2023-06\/segments\/1674764499826.71\/warc\/CC-MAIN-20230130165437-20230130195437-00821.warc.gz\"}"}
| null | null |
Tall spires of hooded white flowers with smudges of pale purple in late summer. Attractive foliage spring to autumn. Good for shady spots. To 1.5m. Hardy.
Aconitum carmichaelii 'Cloudy' (PBR) (Monkshood 'Cloudy'): Tall spires of hooded white flowers with smudges of pale purple in late summer. Attractive foliage spring to autumn. To 1.5m. Hardy.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,517
|
\section{Introduction}
Observational investigations of coronal loop structure have been undertaken since the 1940s \citep{bray91}; however, due to insufficient spatial resolution of current and previous instrumentation, the definitive resolved widths of these fundamental structures have not been fully realised. Recent high-resolution data from NASA's Interface Region Imaging Spectrometer (IRIS; \citealp{depontieu14}) and the High-resolution Coronal imager (Hi-C; \citealp{kobayashi14}) have led to coronal loop width studies in unprecedented detail. For short loops whose lengths are of the scale of a granule, \citet{peter13} found widths $\lesssim200$\,km within the Hi-C data. Similarly, \citet{aschwanden17} sampled $10^5$ loop width measurements from the Hi-C field of view (FOV) with their analysis finding the most-likely width $\approx550$~km, arguing the possibility that Hi-C fully resolved the 193\,\AA\ loops/strands. This agrees with previous work \citep{peter13} where it is proposed that at least some of the wider loops with diameters $\approx$1\,Mm observed by NASA's Solar Dynamic Observatory Atmospheric Imaging Assembly (AIA; \citealp{lemen12}) do not appear to show what they consider to be obvious signs of substructure when compared to the coincident Hi-C data-set. Combining IRIS data with hydrodynamic simulations, \citet{brooks16} find transition region temperature loops with widths between $266-386$\,km, and showcase that these structures appear to be composed of singular magnetic threads.
\citet{klimchuk15} investigates the widths of four EUV loops as a function of position using Hi-C and AIA data, and obtain widths of $880-1410$\,km with Hi-C. They also find that, while the analysed loops have relatively constant cross-section along their lengths, those measured with Hi-C are typically less than 25\% narrower than their AIA counterparts. Therefore, they suggest that loops are not highly under-resolved by AIA and these results further supports previous findings of measured widths along both EUV \citep{fuentes06} and soft x-ray \citep{klimchuk92,klimchuk00} loop structures where no significant or observable expansion from the loop base to apex are determined.This work has been developed further by \citep{klimchuk20} where, for 20 loops from the first Hi-C flight data, intensity versus width measurements tended to be uncorrelated or have a direct dependence, implying that the loop flux tube cross sections themselves are approximately circular (assuming that there is non-negligible twist along the flux tube and that the plasma emission is nearly uniform along the magnetic field).
More recently, \citet{williams19} investigate loops from five regions within the FOV of the latest Hi-C flight but at 172\,\AA\ wavelengths \citep[termed Hi-C\,2.1;][]{instpaper}. As with \citet{aschwanden17}, coronal strand widths of $\approx513$~km were determined for four of the five regions analysed. The final region, which investigates low emission, low density loops, finds much narrower coronal strands of $\approx388$\,km, placing those structures beneath the width of a single AIA pixel. The fact that these strands are above the smallest spatial scale at which Hi-C\,2.1\ can resolve individual structures ($220-340$\,km; \citealp{instpaper}) suggests that Hi-C\,2.1\ may be beginning to resolve a key spatial scale of coronal loops.
Notably, and the focus for this work, \citet{williams19} also find example structures that may not be fully resolved within the Hi-C\,2.1\ data. These relate to smaller `bumps' or turning points in the intensity profiles that are larger than the observational error bars but do not constitute a full, completely isolated strand. Could these be the result of projection effects of overlapping structures along the integrated line of sight for this optically thin plasma, or are they the result of further structures beneath even the resolving abilities of Hi-C\,2.1 ?
Thus, this current paper outlines approaches to further investigate the possible spatial scale of Hi-C\,2.1\ coronal strands reported upon by \citet{williams19} but are not fully resolved as defined above. In \S\ref{sec:data} the Hi-C\,2.1\ data preparation is discussed including the Gaussian fit method employed to estimate the width of these partially resolved coronal features. The resulting distribution of fitted widths is described in \S\ref{sec:res} with conclusions reached on the analysis outlined in \S\ref{sec:conc}.
\section{Data Preparation and Analysis Method}\label{sec:data}
On 29\textsuperscript{th} May 2018 at 18:54 UT, Hi-C\,2.1\ was successfully relaunched from the White Sands Missile Range, NM, USA, capturing high-resolution data (2k$\times$2k pixels; $4.4^\prime\times4.4^\prime$ field of view) of target active region AR\,12712\ in EUV emission of wavelength 172\,\AA\ (dominated by Fe IX emission $\approx0.8$\,MK) with a plate scale of 0.129\ensuremath{^{\prime\prime}}. During the flight Hi-C\,2.1\ captured 78 images with a 2s exposure time and a 4.4s cadence between 18:56 and 19:02 UT. Full details on the Hi-C\,2.1\ instrument can be found in \citet{instpaper}.
\begin{figure*}
\centerline{\includegraphics[width=0.9\textwidth]{HiC_MGN_FOV.pdf}}
\caption{Reverse colour image showing the Hi-C\,2.1\ field of view which has been time-averaged for $\approx60$~s and then, for the purpose of this figure only, sharpened with Multi-Scale Gaussian Normalisation \citep[MGN]{morgandruckmuller14}. The locations of the cross section slices where multiple-peaked structures are observed are shown in \textit{blue}, whilst the \textit{green} slices are displayed more clearly in Figure\,\ref{fig:closeup}. The cross-sectional profiles for the slices numbered 1\,-\,4 (5 and 6) are shown in Figure\,\ref{fig:example1} (\ref{fig:fwhmcomp}).}
\label{fig:fov}
\end{figure*}
\begin{figure*}
\centerline{\includegraphics[width=0.9\textwidth]{BackgroundRemoval.pdf}}
\caption{Left: an example cross-section slice from the Hi-C\,2.1\ field-of-view (slice\,2 in Figure\,\ref{fig:fov}) is shown in black. The global background obtained by interpolating with a cubic spline through the inflection points is shown via the dashed blue line. Right: the isolated coronal strands that are obtained by subtracting the global background from the Hi-C\,2.1\ intensity. The error bars denote five times the Poisson error (\textit{grey}) due to the small magnitude (mean: $1.8\times10^{-3}$) making them difficult to see without magnification.}
\label{fig:background}
\end{figure*}
\subsection{Dataset Extraction and Background Subtraction}\label{sec:bkgd}
The basis of the sample data-set investigated here include a number of subsets from the ten higher-emission cross-section slices analysed by \citet{williams19} plus nine other additional slices from within the Hi-C\,2.1\ field of view (see Figure~\ref{fig:fov} where all dataset locations are indicated). In each case the resulting emission profile across the structures would indicate sub-structure strands that are not fully resolved i.e., a non-Gaussian shape.
Following the method outlined in \citep{williams19}, the Hi-C\,2.1\ dataset under consideration is time-averaged over a period $\approx60$\,s that is free from spacecraft jitter\footnote{A consequence of the instability experienced during the Hi-C\,2.1\ flight is that ghosting of the mesh could not be avoided \citep{instpaper}. This leads to the diamond patterns across the entire Hi-C\,2.1\ field-of-view, which are exaggerated when the data is enhanced with MGN (Figure\,\ref{fig:fov}).}. Each cross-section normal to each strand is taken to be 3-pixels deep and the background emission is then subtracted. As outlined in Figure~\ref{fig:background}, this background subtraction is performed by firstly finding all the local minima of a slice, and interpolating through these values using a cubic spline \citep{yi15} to obtain a global trend (dashed blue line). The global trend is then subtracted from the intensity profile along the slice, leaving behind the background subtracted coronal strands (similar to \citealp{aschwanden11,williams19}). Due to the large number of counts detected by Hi-C\,2.1\, the Poisson error associated with these isolated coronal strands is minimal (Figure\,\ref{fig:background}).
\subsection{Gaussian Fitting and FWHM Measurements}\label{sec:gaussfit}
The analysis method is based on the assumption that at rest, an isolated coronal strand element has an observed emission profile across its width and normal to the strand axis that is approximately Gaussian. It is important to note that as indicated by \citet{pontin17}, instantaneously coronal strands may not necessarily have a clear Gaussian cross-section. On the other hand, \citep{klimchuk20} have shown from Hi-C observations that coronal strands are likely to have circular cross-sections. To attempt to address this and as indicated previously, the data samples are time-averaged over $\approx60$\,s (the first 11 Hi-C\,2.1\ frames) to average out any short timescale changes. Whilst no obvious signs of motion within the structures analysed are noticed in this 60s window, the authors acknowledge that as indicated by \citet{morton13}, small amplitude oscillations could be present which would lead to the measured widths being broader than the structural width due to the time integration performed.
Previous width studies (such as \citealp{aschwanden17,williams19}) would have considered the features under examination here (e.g. those between $0\ensuremath{^{\prime\prime}}-6.5\ensuremath{^{\prime\prime}}$ in Figure~\ref{fig:example1}) as individual, whole structures in spite of their outline. However, due to their distinct non-Gaussian cross section that is itself well resolved by Hi-C\,2.1\,, in this study they are considered to be subsequently modelled as the combination of several Gaussian-shaped coronal strands.
The observed Hi-C\,2.1\ intensity profile of a cross-sectional slice is reproduced by simultaneously fitting Gaussian profiles, the number of which is determined by the Akaike Information Criterion \citep[AIC]{akaike74} along with a corrective term (AICc) for small sample sizes. This is fully described in Appendix \ref{app:aic}. Subsequently, the full width at half maximum (FWHM) of the Gaussian profile is measured to provide an estimate of the possible width of the sub-structures likely present within the Hi-C\,2.1\ data.
Thus, the method employed to fit Gaussian profiles to the observed Hi-C\,2.1\ intensity is as follows. Firstly, the following expression for a Gaussian function, $Y_G$ is used:
\begin{equation}\label{eq:gaussian}
Y_{G} = A \exp\left(\frac{-(x-x_{p})^2}{2W^2}\right),
\end{equation}
whereby $x$ is position along the cross-section slice, $A$ and $x_{p}$ are the amplitude and location of the peak, and $W$ is the Gaussian RMS width. This can be related to the FWHM by: $FWHM=2\,\sqrt{2\ln{2}}\,W\approx2.35\,W$.
An estimate is made on the number of structures, $N$ that could be present within the intensity profile along with their approximate location, width, and amplitude. Summing the $Y_G$ values for $N$ number of Gaussian curves at each pixel yields the model fit:
\begin{equation}\label{eq:fit}
f\left(x\right) = \sum_{i=1}^N Y_{G\left(i\right)}\left(x\right).
\end{equation}
The closeness of the fit at each pixel, $\chi^2\left(x\right)$ is then determined by measuring the deviation of the fit from the original intensity:
\begin{equation}\label{eq:acc}
\chi^2\left(x\right) = \left(\frac{f\left(x\right)-y\left(x\right)}{\sigma\left(x\right)}\right)^2,
\end{equation}
where $y\left(x\right)$ and $\sigma\left(x\right)$ are the observed Hi-C\,2.1\ intensity and Poisson error at each pixel. The overall closeness of fit is then taken as $\sum \chi^2\left(x\right)$, which is then reduced to its smallest value by simultaneously adjusting the free parameters $A$, $x_p$, and $W$ for the $N$ Gaussian curves in $f\left(x\right)$. The minimisation of $\sum \chi^2\left(x\right)$ is performed by using the non-linear least-squares curve fitting method, \textit{MPFIT}\footnote{\textit{MPFIT} is freely-available at: http://purl.com/net/mpfit} \citep{mpfit} which is based on the MINPACK-1 FORTRAN library \citep{minpack}. During the fitting process the one-$\sigma$ uncertainties are returned from \textit{MPFIT}. These error values are only accurate if the shape of the likelihood surface is well approximated by a parabolic function. Whether fitting multiple Gaussian profiles to each slice satisfies this condition or not would require analysis beyond the scope of this study, however, it is likely the one-$\sigma$ uncertainties do provide a lower-bound of the FWHM errors.
To determine the appropriate number of Gaussian profiles, $N$ within a given slice, the AIC model selection is employed. This is done by firstly generating several candidate models, where the number of Gaussian curves differs in each model. The non-linear least-squares curve fitting method is then employed for each candidate model and finally the AICc is then computed. The model with the smallest AICc value is then selected as the preferred model for that Hi-C\,2.1\ slice.
Once the number of Gaussian profiles contained within a Hi-C\,2.1\ slice is determined, the strand width(s) are taken as the Gaussian FWHM value(s). As with previous loop-width studies \citep{aschwanden17,brooks13,brooks16,peter13,williams19} the width measurements are then collated into statistical samples in order to deduce if key structural widths can be extracted from the data.
\section{Results \& Analysis}\label{sec:res}
\begin{figure*}
\centerline{\includegraphics[width=0.95\textwidth]{closeup.pdf}}
\caption{A close-up view of slices numbered 1 - 6 in Figure\,\ref{fig:fov}, which have been sharpened using MGN. The colour tables are normalised to each sub-region and are shown in reverse colour. The cross-sectional profiles of these slices are shown in Figures\,\ref{fig:fwhmcomp} and \ref{fig:example1}.}
\label{fig:closeup}
\end{figure*}
\begin{figure*}
\centerline{\includegraphics[width=0.95\textwidth]{FWHM_compare.pdf}}
\caption{A comparison between the FWHM measurement method of this paper with that of \citet{williams19} for slices 5 and 6 shown in Figure\,\ref{fig:fov}. The \textit{blue} line is the background-subtracted Hi-C\,2.1\ intensity, the Gaussian profiles obtained from using the non-linear least-squares fitting method are shown in \textit{red} and their FWHM are denoted by the \textit{green} asterisks. The solid \textit{grey} bands indicate the FWHM of the Hi-C\,2.1\ structures as determined by our previous method \citep{williams19}.}
\label{fig:fwhmcomp}
\end{figure*}
\begin{figure}[!htbp]
\includegraphics[width=0.97\columnwidth]{Example_1.pdf}
\caption{The intensity profiles of the Hi-C\,2.1\ cross-sectional slices (1\,-\,4 in Figure\,\ref{fig:fov}) are shown in blue. The Gaussian profiles (grey) generated by the non-linear least-squares fitting algorithm and the subsequent fit, $f\left(x\right)$ (red) are also plotted for the four example slices.}
\label{fig:example1}
\end{figure}
\begin{figure}
\includegraphics[width=\columnwidth]{FWHM3.pdf}
\caption{a) Occurrence frequency plot of the 183 FWHM measurements for the fitted Gaussian curves with a bin width of 125\,km ; b) a subset of the Gaussian profiles plotted at bin widths of 62.5\,km that correspond to the most populous widths in a). The dashed vertical navy (red) lines indicate the low (high) emission strand widths obtained in \citet{williams19}. Panel c) shows the $1-\sigma$ errors for the 183 widths obtained for the Gaussian profiles in a).}
\label{fig:fwhm}
\end{figure}
Employing the non-linear least-squares curve fitting method\ discussed in the previous section, a total of 183 Gaussian profiles are fitted to twenty four Hi-C\,2.1\ cross-sectional slices. As seen in the field-of-view plot (Figure\,\ref{fig:fov}) it is not easy to completely isolate a coronal strand. For example, to the north of slice 5 (Figure\,\ref{fig:closeup}) there is an increase in intensity due to a crossing of another emitting feature along the integrated line-of-sight. Care is taken to avoid contamination from such structures, though it is possible that some residual emission may remain in the slices selected. However, the relative intensity of the much brighter strand to the often weaker contaminating emission means that it is removed during background subtraction.
The following subsections compare the twenty four cross-sectional slices to those outlined in \citet{williams19} as well as examining the frequency distribution of the newly fitted Gaussian profiles.
\subsection{FWHM Method Comparison}\label{sec:paper1}
Here, comparison is made between the resulting strand widths obtained by fitting multiple Gaussian profiles to Hi-C\,2.1\ structures versus the widths obtained using the previous method \citep{williams19} now with the improved background subtraction discussed in \S\,\ref{sec:bkgd}. Two examples are outlined (Figure\,\ref{fig:fwhmcomp}) where a non-Gaussian distribution is seen. For slice 5 (6), the AICc determined that six (four) Gaussian profiles are supported by the Hi-C\,2.1\ data whereas measuring the widths of the non-Gaussian profiles provides two (one) structures.
The \citet{williams19} method provides widths of $\approx745$\,km and $980$\,km yielding a mean of $\approx860$\,km for slice 5. Comparatively, the non-linear least-squares curve fitting method\ yields minimum and maximum widths of $\approx320$\,km and $\approx590$\,km, and a mean of $\approx480$\,km. The width of the single structure in slice 6 measured with the \citet{williams19} method is $\approx1145$\,km. As with slice 5, the non-linear least-squares curve fitting method\ provides narrower widths with the minimum, maximum, and mean now being $\approx405$\,km, $\approx530$\,km, and $\approx450$\,km, respectively.
The width estimates of the structures centred at 1\ensuremath{^{\prime\prime}} (slice 5) and 10.4\ensuremath{^{\prime\prime}} (slice 6) may be artificially broadened by their shape using the method employed by \citet{williams19} due to the observable change in gradient that occurs in the vicinity of the half-maximum intensity value. The structure at 3.8\ensuremath{^{\prime\prime}} (slice 5) does not appear to be affected by this as the change in gradient (or `bump') occurs much lower along the structure than the half-maximum intensity value. However, the measured width of this structure is still $\approx300$\,km broader than the maximum AICc determined Gaussian width, which indicates previous analysis methods may have over-estimated the strand widths of structures that are potentially not completely isolated from the background and/or other structures along the integrated line-of-sight.
\subsection{Distribution of Fitted Widths}\label{sec:diswid}
In Figure \ref{fig:example1}, the cross-sectional profiles (\textit{blue}) are shown of the Hi-C\,2.1\ slices numbered 1\,-\,4 in Figure\,\ref{fig:fov} along with the best AICc-determined fits and Gaussian profiles generated from the non-linear least-squares curve fitting method, shown in \textit{red} and \textit{grey} respectively. From the four examples shown here, it is seen that there is good agreement between the observed Hi-C\,2.1\ intensity and the generated fit though some minor discrepancies may occasionally occur (e.g. slice 2 between 3.5\ensuremath{^{\prime\prime}} and 6.5\ensuremath{^{\prime\prime}}). These discrepancies could be eradicated by adding additional Gaussian profiles along the slices; however, the additional parameters introduced are not supported by the AICc model selection.
In Figure\,\ref{fig:fwhm}\,(a) the FWHM values of the 183 Gaussian profiles are collated into an occurrence frequency plot binned at $125$\,km intervals so as to be consistent with the previous study \citep{williams19} where $1\ensuremath{^{\prime\prime}} \approx 725$\,km. A sub-section of this data ($200-760$\,km) is shown in Figure\,\ref{fig:fwhm}\,(b) which is binned at half the spatial scale of Figure\,\ref{fig:fwhm}\,(a) ($62.5$\,km). Figure\,\ref{fig:fwhm}\,(c) shows the one-$\sigma$ errors for the 183 Gaussian widths indicating the majority of errors are $\lesssim50$\,km.
The distribution of all analysed widths in Figure\,\ref{fig:fwhm}\,(a) reveals that the most populous widths are between $450-575$\,km; this matches the high-emission region results from \citet{williams19}. The median width for this data is $645$\,km and is akin to that obtained by \citet{brooks13}; however, this value is due largely to the presence of a number of broader strands ($>1000$\,km). Furthermore, $\approx21\%$ of widths exceed $1000$\,km whilst $\approx32\%$ of the strands studied are at the SDO/AIA resolving scale of $600-1000$\,km. From this, $\approx47\%$ of the strands are beneath the resolving scale of AIA.
Figure\,\ref{fig:fwhm}\,(a) reveals the most populous strand widths in this study occur between $\approx200-760$\,km with the number of width samples above this spatial scale rapidly decreasing. Figure\,\ref{fig:fwhm}\,(b) shows a subset of the obtained widths having been re-binned to 62.5\,km intervals, which allows for further insight on the distribution of widths for the most populous occurrence frequency bins of Figure\,\ref{fig:fwhm}\,(a). The obtained Hi-C\,2.1\ strand widths reveal the presence of numerous strands ($\approx32\%$ of the 183 Gaussian widths) whose FWHM are beneath the most frequent high-emission strand widths seen previously \citep[$\approx513$\,km]{williams19}. Similarly, $\approx17\%$ reside beneath an AIA pixel width of 435\,km. Comparatively then, only $\approx6\%$ of the strands are actually at the the scale at which Hi-C\,2.1\ can resolve structures \citep[$\approx220-340$\,km]{instpaper}, indicating that current instrumentation may now be beginning to observe a prevalent spatial scale.
As with \citet{williams19}, this analysis reveals that the most-likely strand widths of $450-575$\,km are typically of the order of an AIA pixel width. This result coupled with the low-percentage ($\approx6\%$) of strands at the Hi-C\,2.1\ resolving scale suggests that the non-Gaussian structures observed are predominately the result of multiple, potentially resolvable strands overlapping along the integrated line-of-sight rather than the result of finer strands that even Hi-C\,2.1\ is unable to resolve into distinct features. It should be noted that these results are $50-250$\,km narrower than previous Hi-C findings that focused on 193\,\AA\ emission \citep{aschwanden17,brooks13}.
Nevertheless, this does not rule out the possibility of strands within or beneath the resolving power of Hi-C\,2.1. For example, using CRISP H-$\alpha$ data, \citet{scullion14} find that the most populous strand width is $\approx100$\,km. However, the temperature of those structures is 1-2 orders of magnitude lower than that observed with Hi-C\,2.1.
\section{Summary and Conclusions}\label{sec:conc}
This work outlines a follow-up analysis to \citet{williams19} where non-Gaussian shaped width profiles that are not fully resolved within the Hi-C\,2.1\ data are further investigated. To estimate the widths of possible strands, Gaussian functions are first fitted to approximate the Hi-C\,2.1\ intensity profiles using the method outlined in \S\ref{sec:gaussfit}. The non-linear least-squares curve fitting method employed automatically determines the Gaussian RMS width due to $W$ being a free parameter (Equation~\ref{eq:gaussian}) used to reduce $\sum \chi^2\left(x\right)$. The number of Gaussian profiles and subsequently the number of RMS widths measured are determined by the AICc, which are then converted to FWHM for our width analysis study.
The FWHM are collated into occurrence frequency plots (Figure~\ref{fig:fwhm}) revealing the most frequent strand width is $\approx450-575$\,km. The spatial scales obtained in this study largely agree with previous findings \citep{aschwanden17,williams19} where typical widths the size of an AIA pixel are seen. Additionally, the results reveal that only $\approx6\%$ of the strands analysed reside at the smallest spatial scales that Hi-C\,2.1\ can resolve into distinct structures \citep[$220-340$\,km]{instpaper}. Together, these findings strongly suggest that structures emitting at 172\,\AA\ that cannot be resolved into distinct features by Hi-C\,2.1\ are likely to comprise of multiple strands overlapping along the integrated line-of-sight rather than being an amalgamation of strands at/below the resolving scale of Hi-C\,2.1. For coronal loop modelling, the onus must now be on the determination of the spatial scale at which heating occurs that leads to the formation of individual magnetic strands that i) have widths $450-575$\,km and ii) are filled with plasma around 1\,MK.
Furthermore, recent work by \citet{klimchuk20} investigated the widths along the length of isolated coronal structures and found no correlation between width and intensity. However, as is noted in \citet{klimchuk20}, if a structure indicated signs of any possible substructure, then that particular example was not included in the study data. Thus, employing the methods adopted in this work on those rejected examples would allow for that type of analysis to be performed along the observable length of coronal structural sub-elements and not only the aforementioned monolithic features. This will be addressed in a follow-up study using the Hi-C\,2.1\ data set examined here.
The highly anticipated ESA mission Solar Orbiter (SolO) will provide close-up ($\approx0.28$\,AU), high-latitude (34$^{\circ}$) solar observations. During the mission there will be several observation windows where the spatial resolution of EUV Imager (EUI) HRI as well as the selected passband (174\,\AA) will be similar to that of Hi-C. However, it is likely SolO will have longer observation windows over which any target active region may be studied (Hi-C only captures 2.5 minutes of usable data per flight). This will allow for significantly improved strand width determination across many differing coronal structures.
\begin{appendix}
\section{Akaike Information Criterion}\label{app:aic}
To determine the number of strands that may be hidden within the Hi-C\,2.1\ data the Akaike Information Criterion \citep{akaike74} is employed to determine the optimal number of Gaussian profiles supported by the data. Each additional Gaussian that is added to the model introduces an additional three parameters that may be tweaked to better allow for the observed data to be replicated by equations \ref{eq:gaussian}\,-\,\ref{eq:acc}. Employing a model selection method such as AIC helps minimise the possibility of selecting a model with too many(few) Gaussian curves, and thus the danger of over(under)-fitting the Hi-C\,2.1\ data.
Often, the AIC is defined as $AIC = 2k -2\ln{\left(L_{max}\right)}$ where $k$ is the number of parameters in the model and $L_{max}$ is the maximum likelihood. In this study, a least-squares model fitting is employed and thus the maximum likelihood estimate for the variance of a model's distribution of residuals is $\hat{\sigma}^2 = RSS/n$, where $n$ is the sample size and $RSS$ is the residual sum of squares:
\begin{equation}\label{eq:rss}
RSS = \Sigma_{i=1}^{n}\left(y_i-f\left(x_i\right)\right)^2.
\end{equation}
Thus, the maximum value of a model's likelihood function can be expressed as:
\begin{equation}\label{eq:llh}
-\frac{n}{2}\ln{\left(2\pi\right)}-\frac{n}{2}\ln{\left(\hat{\sigma}^2\right)}-\frac{1}{2\hat{\sigma}^2}RSS = -\frac{n}{2}\ln{\left(\frac{RSS}{n}\right)}+C,
\end{equation}
where $C$ is an independent constant that does not change unless $y$ does. Following \citet[p.63]{burnham02}, this means the AIC for a least-squares model can be expressed as:
\begin{equation}
AIC = 2k + n\ln{\left(\frac{RSS}{n}\right)}-2C = 2k + n\ln{\left(RSS\right)}-\left(n\ln{\left(n\right)}+2C\right),
\end{equation}
which can be further simplified to
\begin{equation}\label{eq:AICRSS}
AIC = 2k + n\ln{\left(RSS\right)},
\end{equation}
as $\left(n\ln{\left(n\right)}+2C\right)$ is a constant (provided $y$ does not change) and only the differences in AIC are meaningful.
If $n$ is small, AIC may prefer models which have more parameters and lead to over-fitting of the data. As such, a correction for this is to use the AICc, which provides an additional term accounting for $n$ and $k$:
\begin{equation}
AICc = AIC + \frac{2k^2 + 2k}{n-k-1},
\end{equation}
that converges to 0 as $n\xrightarrow{}\infty$ meaning $AICc \equiv AIC$ for large values of $n$.
\subsection{AIC Test Cases}
\begin{figure*}[ht!]
\centering
\begin{minipage}[b]{.6\textwidth}
\centerline{\includegraphics[width=\textwidth]{TEST1.pdf}}
\end{minipage}
\caption{The first AIC model selection test case. The upper-left panel labelled synthetic data is generated using three Gaussian functions. Using the Gaussian fitting method employed in this paper, a number of Gaussian functions (shown in gray) are used to replicate the synthetic data and their AIC and AICc values are indicated.}
\label{fig:test1}
\end{figure*}
\begin{figure*}[ht!]
\centering
\begin{minipage}[b]{.47\textwidth}
\centerline{\includegraphics[width=\textwidth]{TEST2.pdf}}
\caption{The second AIC model selection test case. The upper-left panel labelled synthetic data is generated using four Gaussian functions. Using the Gaussian fitting method employed in this paper, a number of Gaussian functions (shown in gray) are used to replicate the synthetic data and their AIC and AICc values are indicated.}
\label{fig:test2}
\end{minipage}
\begin{minipage}[b]{.47\textwidth}
\centerline{\includegraphics[width=\textwidth]{TEST3.pdf}}
\caption{The thrid AIC model selection test case. The upper-left panel labelled synthetic data is generated using three Gaussian functions. Using the Gaussian fitting method employed in this paper, a number of Gaussian functions (shown in gray) are used to replicate the synthetic data and their AIC and AICc values are indicated.}
\label{fig:test3}
\end{minipage}
\end{figure*}
To validate the AICc model selection, three test cases are devised that are similar to what a Hi-C\,2.1\ cross-sectional slice may look like in this study. The test cases are generated by specifying a number of Gaussian profiles using equation \ref{eq:gaussian}, which are then combined using equation \ref{eq:fit} to generate the synthetic data for each test case. This process allows for the AICc model selection accuracy to be verified as the number of Gaussian profiles required to generate the three test cases are known.
The first test case is composed of three distinct Gaussian profiles which is shown by the blue plot in Figure\,\ref{fig:test1}. An initial guess on the number of Gaussian profiles (2, 3, 4, 5, and 6 Gaussian curves) and their associated free parameters ($A$, $x_p$, and $W$) are made, which are then passed through our non-linear least squares fitting method. During this fitting method, the AIC and AICc values are computed for each model shown in Figure\,\ref{fig:test1}. This reveals that the smallest AIC/AICc values are for the model consisting of three Gaussian functions, which matches the number used to generate the synthetic data. Test cases 2 and 3 (Figures\,\ref{fig:test2}\,and\,\ref{fig:test3}) are more complex than the first test case, and subsequently provide a closer representation of Hi-C\,2.1\ data. Again, the lowest AIC and AICc values correspond to the models consisting of four and three Gaussian profiles, which match the number of Gaussian profiles used in generating the test data.
Whilst both the AIC and AICc show agreement on the model selection for the three test cases analysed in this appendix, some of the Hi-C\,2.1\ cross-sectional slices selected may be of sufficiently small sample size that AIC would favour over-fitting the data, which would lead to artificially narrow widths of the strands in question. Therefore and as outlined above, the AICc is employed in this case for determining the number of strands within a given cross-sectional slice.
\end{appendix}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 6,686
|
{"url":"http:\/\/stem.mrozarka.com\/calculus-1\/lessons-2014-s1\/unit-5\/day-68","text":"### Day 68 - Bases Other Than e - 11.25.14\n\n UpdatesN\/ABell Ringer QuizDifferentiation\/Integration of Natural Log and Exponential FunctionsFind the derivative with respect to x of the following: .all of the abovenone of the aboveFind .both b and cnone of the abovenone of the aboveFind of the following: none of the aboveReviewAlgebraDifferentiation of the Natural Logarithmic FunctionIntegration with the Natural Logarithmic FunctionExponential Functions: Differentiation and IntegrationLessonBases Other Than eCheckpoints$\\int\\sin(x)dx=-\\cos(x)+C$$\\int\\cos(x)dx=\\sin(x)+C$$\\int\\tan(x)dx=-\\ln |cos(x)|+C$$\\int \\csc(x)dx=-\\ln|\\csc(x)+\\cot(x)|+C$$\\int \\sec(x)dx=\\ln|\\sec(x)+\\tan(x)|+C$$\\int \\cot(x)dx=\\ln|\\sin(x)|+C$Exit TicketPosted on the board at end of the block. Lesson Objective(s)How are derivatives found when bases are not e?Standard(s)APC.9Apply formulas to find derivatives.Includes:derivatives of algebraic, trigonometric, exponential, logarithmic, and inverse trigonometric functionsderivations of sums, products, quotients, inverses, and composites (chain rule) of elementary functionsderivatives of implicitly defined functionshigher order derivatives of algebraic, trigonometric, exponential, and logarithmic, functionsMathematical Practice(s)#1 - Make sense of problems and persevere in solving them#2 - Reason abstractly and quantitatively#5 - Use appropriate tools strategically#6 - Attend to precision#8 - Look for and express regularity in repeated reasoningPast CheckpointsDifferentiation of the Natural Logarithmic FunctionCheckpointsIntegration with the Natural Logarithmic FunctionCheckpointsExponential Functions: Differentiation and IntegrationCheckpoints","date":"2021-01-22 21:53:03","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\": 6, \"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\": 0.8450945019721985, \"perplexity\": 6701.710576731357}, \"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-2021-04\/segments\/1610703531429.49\/warc\/CC-MAIN-20210122210653-20210123000653-00111.warc.gz\"}"}
| null | null |
## Contents
I
1 THE GIRL WITH YIN EYES
2 FISHER OF MEN
3 THE DOG AND THE BOA
4 THE GHOST MERCHANT'S HOUSE
5 LAUNDRY DAY
II
6 FIREFLIES
7 THE HUNDRED SECRET SENSES
8 THE CATCHER OF GHOSTS
9 KWAN'S FIFTIETH
III
10 KWAN'S KITCHEN
11 NAME CHANGE
12 THE BEST TIME TO EAT DUCK EGGS
13 YOUNG GIRLS WISH
14 HELLO GOOD-B YE
15 THE SEVENTH DAY
16 BIG MA'S PORTRAIT
17 THE YEAR OF NO FLOOD
18 SIX-ROLL SPRING CHICKEN
19 THE ARCHWAY
20 THE VALLEY OF STATUES
21 WHEN HEAVEN BURNED
22 WHEN LIGHT BALANCES WITH DARK
IV
23 THE FUNERAL
24 ENDLESS SONGS
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.
The Hundred Secret Senses
A G.P. Putnam's Sons Book / published by arrangement with the author
All rights reserved.
Copyright © 1995 by Amy Tan
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: 978-1-101-20294-4
A G.P. PUTNAM'S SONS BOOK®
G.P. Putnam's Sons Books first published by The G.P. Putnam's Sons Publishing Group, a member of Penguin Putnam Inc.,
375 Hudson Street, New York, New York 10014.
G.P. PUTNAM'S SONS and the "P" design are trademarks belonging to Penguin Putnam Inc.
Version_2
Also by Amy Tan
THE JOY LUCK CLUB
THE KITCHEN GOD'S WIFE
for children
THE MOON LADY
THE CHINESE SIAMESE CAT
FOR FAITH
##
To write this story, I depended on the indulgence, advice, conversations, and sustenance of many: Babalu, Ronald Bass, Linden and Logan Berry, Dr. Thomas Brady, Sheri Byrne, Joan Chen, Mary Clemmey, Dr. Asa DeMatteo, Bram and Sandra Dijkstra, Terry Doxey, Tina Eng, Dr. Joseph Esherick, Audrey Ferber, Robert Foothorap, Laura Gaines, Ann and Gordon Getty, Molly Giles, Amy Hempel, Anna Jardine, Peter Lee Kenfield, Dr. Eric Kim, Gus Lee, Cora Miao, Susanne Pari, the residents of Pei Sa Bao village, Robin and Annie Renwick, Gregory Atsuro Riley, the Rock Bottom Remainders, Faith and Kirkpatrick Sale, Orville Schell, Gretchen Schields, the staff of Shelburne House Library, Kelly Simon, Dr. Michael Strong, Daisy Tan, John Tan, Dr. Steven Vandervort, Lijun Wang, Wayne Wang, Yuhang Wang, Russell Wong, the people of Yaddo, and Zo.
I thank them, but do not hold them accountable for the felicitous and sometimes unwitting ways in which they contributed to the truth of this fiction.
# I
## [1
THE GIRL WITH YIN EYES](TheHundredSecretSenses-toc.html#TOC-2)
My sister Kwan believes she has yin eyes. She sees those who have died and now dwell in the World of Yin, ghosts who leave the mists just to visit her kitchen on Balboa Street in San Francisco.
"Libby-ah," she'll say to me. "Guess who I see yesterday, you guess." And I don't have to guess that she's talking about someone dead.
Actually, Kwan is my half sister, but I'm not supposed to mention that publicly. That would be an insult, as if she deserved only fifty percent of the love from our family. But just to set the genetic record straight, Kwan and I share a father, only that. She was born in China. My brothers, Kevin and Tommy, and I were born in San Francisco after my father, Jack Yee, immigrated here and married our mother, Louise Kenfield.
Mom calls herself "American mixed grill, a bit of everything white, fatty, and fried." She was born in Moscow, Idaho, where she was a champion baton twirler and once won a county fair prize for growing a deformed potato that had the profile of Jimmy Durante. She told me she dreamed she'd one day grow up to be different—thin, exotic, and noble like Luise Rainer, who won an Oscar playing O-lan in _The GoodEarth._ When Mom moved to San Francisco and became a Kelly girl instead, she did the next-best thing. She married our father. Mom thinks that her marrying out of the Anglo race makes her a liberal. "When Jack and I met," she still tells people, "there were laws against mixed marriages. We broke the law for love." She neglects to mention that those laws didn't apply in California.
None of us, including my mom, met Kwan until she was eighteen. In fact, Mom didn't even know Kwan _existed_ until shortly before my father died of renal failure. I was not quite four when he passed away. But I still remember moments with him. Falling down a curly slide into his arms. Dredging the wading pool for pennies he had tossed in. And the last day I saw him in the hospital, hearing what he said that scared me for years.
Kevin, who was five, was there. Tommy was just a baby, so he was in the waiting room with my mom's cousin, Betty Dupree—we had to call her Aunt Betty—who had moved out from Idaho as well. I was sitting on a sticky vinyl chair, eating a bowl of strawberry Jell-O cubes that my father had given me from his lunch tray. He was propped up in bed, breathing hard. Mom would cry one minute, then act cheerful. I tried to figure out what was wrong. The next thing I remember, my father was whispering and Mom leaned in close to listen. Her mouth opened wider and wider. Then her head turned sharply toward me, all twisted with horror. And I was terror-struck. How did he know? How did Daddy find out I flushed my turtles, Slowpoke and Fastpoke, down the toilet that morning? I had wanted to see what they looked like without their coats on, and ended up pulling off their heads.
"Your daughter?" I heard my mom say. "Bring her back?" And I was sure that he had just told her to bring me to the pound, which is what he did to our dog Buttons after she chewed up the sofa. What I recall after that is a jumble: the bowl of Jell-O crashing to the floor, Mom staring at a photo, Kevin grabbing it and laughing, then me seeing this tiny black-and-white snapshot of a skinny baby with patchy hair. At some point, I heard my mother shouting: "Olivia, don't argue, you have to leave now." And I was crying, "But I'll be good."
Soon after that, my mother announced: "Daddy's left us." She also told us she was going to bring Daddy's other little girl from China to live in our house. She didn't say she was sending me to the pound, but I still cried, believing everything was vaguely connected—the headless turtles whirling down the toilet, my father abandoning us, the other girl who was coming soon to take my place. I was scared of Kwan before I ever met her.
When I was ten, I learned that my father's kidneys had killed him. Mom said he was born with four instead of the usual two, and all of them were defective. Aunt Betty had a theory about why this happened. She _always_ had a theory, usually obtained from a source like the _Weekly World News._ She said he was supposed to be a Siamese twin. But in the womb, my father, the stronger twin, gobbled up the weaker one and grafted on the two extra kidneys. "Maybe he also had two hearts, two stomachs, who knows." Aunt Betty came up with this scenario around the time that _Life_ magazine ran a pictorial about Siamese twins from Russia. I saw the same story: two girls, Tasha and Sasha, conjoined at the hip, too heartbreakingly beautiful to be freaks of nature. This must have been in the mid-sixties, around the time I learned fractions. I remember wishing we could exchange Kwan for those Siamese twins. Then I'd have two half sisters, which equaled a whole, and I figured all the kids on the block would try to be our friends, hoping we'd let them watch as we jumped rope or played hopscotch.
Aunt Betty also passed along the story of Kwan's birth, which was not heartbreaking, just embarrassing. During the war, she said, my father had been a university student in Guilin. He used to buy live frogs for his supper at the outdoor market from a young woman named Li Chen. He later married her, and in 1944 she gave birth to their daughter, the skinny baby in the picture, Kwan.
Aunt Betty had a theory about the marriage as well. "Your dad was good-looking, for a Chinese man. He was college-educated. And he spoke English like me and your mom. Now why would he marry a little peasant girl? Because he _had to,_ that's why." By then, I was old enough to know what _had to_ meant.
Whatever the case, in 1948, my father's first wife died of a lung disease, perhaps TB. My father went to Hong Kong to search for work. He left Kwan in the care of his wife's younger sister, Li Bin-bin, who lived in a small mountain village called Changmian. Of course, he sent money for their support—what father would not? But in 1949, the Communists took over China, and it was impossible for my father to return for his five-year-old daughter. So what else could he do? With a heavy heart, he left for America to start a new life and forget about the sadness he left behind. Eleven years later, while he was dying in the hospital, the ghost of his first wife appeared at the foot of his bed. "Claim back your daughter," she warned, "or suffer the consequences after death!" That's the story my father gave just before he died—that is, as told by Aunt Betty years later.
Looking back, I can imagine how my mom must have felt when she first heard this. Another wife? A daughter in China? We were a modern American family. We spoke English. Sure, we ate Chinese food, but take-out, like everyone else. And we lived in a ranch-style house in Daly City. My father worked for the Government Accounting Office. My mother went to PTA meetings. She had never heard my father talk about Chinese superstitions before; they attended church and bought life insurance instead.
After my father died, my mother kept telling everyone how he had treated her "just like a Chinese empress." She made all sorts of grief-stricken promises to God and my father's grave. According to Aunt Betty, at the funeral, my mother vowed never to remarry. She vowed to teach us children to do honor to the Yee family name. She vowed to find my father's firstborn child, Kwan, and bring her to the United States.
The last promise was the only one she kept.
MY MOTHER has always suffered from a kind heart, compounded by seasonal rashes of volunteerism. One summer, she was a foster mother for Yorkie Rescue; the house still stinks of dog pee. For two Christmases, she dished out food to the homeless at St. Anthony's Dining Room; now she goes away to Hawaii with whoever is her current boyfriend. She's circulated petitions, done fund-raising, served on boards of alternative-health groups. While her enthusiasm is genuine, eventually, always, it runs out and then she's on to something new. I suspect she thought of Kwan as a foreign exchange student she would host for a year, a Chinese Cinderella, who would become self-sufficient and go on to have a wonderful American life.
During the time before Kwan came, Mom was a cheerleader, rallying my brothers and me to welcome a big sister into our lives. Tommy was too little to do anything except nod whenever Mom said, "Aren't you excited about having another big sister?" Kevin just shrugged and acted bored. I was the only one who did jumping jacks like a gung-ho recruit, in part because I was ecstatic to learn Kwan would be _in addition_ to me, not _instead of._
Although I was a lonely kid, I would have preferred a new turtle or even a doll, not someone who would compete for my mother's already divided attention and force me to share the meager souvenirs of her love. In recalling this, I know that my mother loved me—but not absolutely. When I compared the amount of time she spent with others—even total strangers—I felt myself sliding further down the ranks of favorites, getting bumped and bruised. She always had plenty of room in her life for dates with men or lunch with her so-called gal pals. With me, she was unreliable. Promises to take me to the movies or the public pool were easily erased with excuses or forgetfulness, or worse, sneaky variations of what was said and what was meant: "I hate it when you pout, Olivia," she once told me. "I didn't guarantee I'd _go_ to the swim club with you. I said I would _like to._ " How could I argue my need against her intention?
I learned to make things not matter, to put a seal on my hopes and place them on a high shelf, out of reach. And by telling myself that there was nothing inside those hopes anyway, I avoided the wounds of deep disappointment. The pain was no worse than the quick sting of a booster shot. And yet thinking about this makes me ache again. How is it that as a child I knew I should have been loved more? Is everyone born with a bottomless emotional reservoir?
So of course, I didn't want Kwan as my sister. Just the opposite. Which is why I made great efforts in front of my mother to appear enthusiastic. It was a distorted form of inverse logic: If hopes never come true, then hope for what you don't want.
Mom had said that a big sister was a bigger version of myself, sweet and beautiful, only more Chinese, and able to help me do all kinds of fun things. So I imagined not a sister but another me, an older self who danced and wore slinky clothes, who had a sad but fascinating life, like a slant-eyed version of Natalie Wood in _West Side Story,_ which I saw when I was five. It occurs to me only now that my mother and I both modeled our hopes after actresses who spoke in accents that weren't their own.
One night, before my mother tucked me in bed, she asked me if I wanted to pray. I knew that praying meant saying the nice things that other people wanted to hear, which is what my mom did. So I prayed to God and Jesus to help me be good. And then I added that I hoped my big sister would come soon, since my mother had just been talking about that. When I said, "Amen," I saw she was crying and smiling proudly. Under my mother's eye I began to collect welcome presents for Kwan. The scarf my aunt Betty gave me for my birthday, the orange blossom cologne I received at Christmas, the gooey Halloween candy—I lovingly placed all these scratchy, stinky, stale items into a box my mother had marked "For Olivia's big sister." I convinced myself I had become so good that soon Mom would realize we didn't need another sister.
My mother later told my brothers and me how difficult it was to find Kwan. "In those days," she said, "you couldn't just write a letter, stick a stamp on it, and send it to Changmian. I had to cut through _mounds_ of red tape and fill out dozens of forms. And there weren't too many people who'd go out of their way to help someone from a communist country. Aunt Betty thought I was crazy! She said to me, 'How can you take in a nearly grown girl who can't speak a word of English? She won't know right from wrong or left from right.' "
Paperwork wasn't the only obstacle Kwan had to unknowingly surmount. Two years after my father died, Mom married Bob Laguni, whom Kevin today calls "the fluke in our mother's history of dating foreign imports—and that's only because she thought Laguni was Mexican instead of Italian." Mom took Bob's name, and that's how my brothers and I also ended up with Laguni, which I gladly changed to Bishop when I married Simon. The point is, Bob never wanted Kwan to come in the first place. And my mom usually put his wishes above everyone else's. After they divorced—I was in college by then—Mom told me how Bob pressured her, just before they were married, to cancel the paperwork for Kwan. I think she intended to and forgot. But this is what she told me: "I watched you pray. You looked so sweet and sad, asking God, 'Please send me my big sister from China.' "
I WAS NEARLY SIX by the time Kwan came to this country. We were waiting for her at the customs area of San Francisco Airport. Aunt Betty was also there. My mother was nervous and excited, talking nonstop: "Now listen, kids, she'll probably be shy, so don't jump all over her. . . . And she'll be skinny as a beanpole, so I don't want any of you making fun of her. . . ."
When the customs official finally escorted Kwan into the lobby where we were waiting, Aunt Betty pointed and said, "That's her. I'm telling you that's her." Mom was shaking her head. This person looked like a strange old lady, short and chubby, not exactly the starving waif Mom pictured or the glamorous teenage sister I had in mind. She was dressed in drab gray pajamas, and her broad brown face was flanked by two thick braids.
Kwan was anything but shy. She dropped her bag, fluttered her arms, and bellowed, "Hall-oo! Hall-oo!" Still hooting and laughing, she jumped and squealed the way our new dog did whenever we let him out of the garage. This total stranger tumbled into Mom's arms, then Daddy Bob's. She grabbed Kevin and Tommy by the shoulders and shook them. When she saw me, she grew quiet, squatted on the lobby floor, and held out her arms. I tugged on my mother's skirt. "Is _that_ my big sister?"
Mom said, "See, she has your father's same thick, black hair."
I still have the picture Aunt Betty took: curly-haired Mom in a mohair suit, flashing a quirky smile; our Italo-American stepfather, Bob, appearing stunned; Kevin and Tommy mugging in cowboy hats; a grinning Kwan with her hand on my shoulder; and me in a frothy party dress, my finger stuck in my bawling mouth.
I was crying because just moments before the photo was taken, Kwan had given me a present. It was a small cage of woven straw, which she pulled out of the wide sleeve of her coat and handed to me proudly. When I held it up to my eyes and peered between the webbing, I saw a six-legged monster, fresh-grass green, with saw-blade jaws, bulging eyes, and whips for eyebrows. I screamed and flung the cage away.
At home, in the bedroom we shared from then on, Kwan hung the cage with the grasshopper, now missing one leg. As soon as night fell, the grasshopper began to chirp as loudly as a bicycle bell warning people to get out of the road.
After that day, my life was never the same. To Mom, Kwan was a handy baby-sitter, willing, able, and free. Before my mother took off for an afternoon at the beauty parlor or a shopping trip with her gal pals, she'd tell me to stick to Kwan. "Be a good little sister and explain to her anything she doesn't understand. Promise?" So every day after school, Kwan would latch on to me and tag along wherever I went. By the first grade, I became an expert on public humiliation and shame. Kwan asked so many dumb questions that all the neighborhood kids thought she had come from Mars. She'd say: "What M&M?" "What ching gum?" "Who this Popeye Sailor Man? Why one eye gone? He bandit?" Even Kevin and Tommy laughed.
With Kwan around, my mother could float guiltlessly through her honeymoon phase with Bob. When my teacher called Mom to say I was running a fever, it was Kwan who showed up at the nurse's office to take me home. When I fell while roller-skating, Kwan bandaged my elbows. She braided my hair. She packed lunches for Kevin, Tommy, and me. She tried to teach me to sing Chinese nursery songs. She soothed me when I lost a tooth. She ran the washcloth over my neck while I took my bath.
I should have been grateful to Kwan. I could always depend on her. She liked nothing better than to be by my side. But instead, most of the time, I resented her for taking my mother's place.
I remember the day it first occurred to me to get rid of Kwan. It was summer, a few months after she had arrived. Kwan, Kevin, Tommy, and I were sitting on our front lawn, waiting for something to happen. A couple of Kevin's friends sneaked to the side of our house and turned on the sprinkler system. My brothers and I heard the telltale spit and gurgle of water running into the lines, and we ran off just before a dozen sprinkler heads burst into spray. Kwan, however, simply stood there, getting soaked, marveling that so many springs had erupted out of the earth all at once. Kevin and his friends were howling with laughter. I shouted, "That's not nice."
Then one of Kevin's friends, a swaggering second-grader whom all the little girls had a crush on, said to me, "Is that dumb Chink your sister? Hey, Olivia, does that mean you're a dumb Chink too?"
I was so flustered I yelled, "She's not my sister! I hate her! I wish she'd go back to China!" Tommy later told Daddy Bob what I had said, and Daddy Bob said, "Louise, you better do something about your daughter." My mother shook her head, looking sad. "Olivia," she said, "we don't ever hate anyone. 'Hate' is a _terrible_ word. It hurts you as much as it hurts others." Of course, this only made me hate Kwan even more.
The worst part was sharing my bedroom with her. At night, she liked to throw open the curtains so that the glare of the street lamp poured into our room, where we lay side by side in our matching twin beds. Under this "beautiful American moon," as she called it, Kwan would jabber away in Chinese. She kept on talking while I pretended to be asleep. She'd still be yakking when I woke up. That's how I became the only one in our family who learned Chinese. Kwan infected me with it. I absorbed her language through my pores while I was sleeping. She pushed her Chinese secrets into my brain and changed how I thought about the world. Soon I was even having nightmares in Chinese.
In exchange, Kwan learned her English from me—which, now that I think of it, may be the reason she has never spoken it all that well. I was not an enthusiastic teacher. One time, when I was seven, I played a mean trick on her. We were lying in our beds in the dark.
"Libby-ah," Kwan said. And then she asked in Chinese, "The delicious pear we ate this evening, what's its American name?"
"Barf," I said, then covered my mouth to keep her from hearing my snickers.
She stumbled over this new sound—"bar-a-fa, bar-a-fa"—before she said, "Wah! What a clumsy word for such a delicate taste. I never ate such good fruit. Libby-ah, you are a lucky girl. If only my mother did not die." She could segue from just about any topic to the tragedies of her former life, all of which she conveyed to me in our secret language of Chinese.
Another time, she watched me sort through Valentine's Day cards I had spilled onto my bed. She came over and picked up a card. "What's this shape?"
"It's a heart. It means love. See, all the cards have them. I have to give one to each kid in my class. But it doesn't really mean I love everyone."
She went back to her own bed and lay down. "Libby-ah," she said. "If only my mother didn't die of heartsickness." I sighed, but didn't look at her. This again. She was quiet for a few moments, then went on. "Do you know what heartsickness is?"
"What?"
"It's warming your body next to your family, then having the straw roof blow off and carry you away."
"Oh."
"You see, she didn't die of lung sickness, no such thing."
And then Kwan told me how our father caught a disease of too many good dreams. He could not stop thinking about riches and an easier life, so he became lost, floated out of their lives, and washed away his memories of the wife and baby he left behind.
"I'm not saying our father was a bad man," Kwan whispered hoarsely. "Not so. But his loyalty was not strong. Libby-ah, do you know what loyalty is?"
"What?"
"It's like this. If you ask someone to cut off his hand to save you from flying off with the roof, he immediately cuts off both hands to show he is more than glad to do so."
"Oh."
"But our father didn't do this. He left us when my mother was about to have another baby. I'm not telling you lies, Libby-ah, this is true. When this happened, I was four years old by my Chinese age. I can never forget lying against my mother, rubbing her swollen belly. Like a watermelon, she was this big."
She reached out her arms as far as she could. "Then all the water in her belly poured out as tears from her eyes, she was so sad." Kwan's arms fell suddenly to her sides. "That poor starving baby in her belly ate a hole in my mother's heart, and they both died."
I'm sure Kwan meant some of this figuratively. But as a child, I saw everything Kwan talked about as literal truth: chopped-off hands flying out of a roofless house, my father floating on the China Sea, the little baby sucking on his mother's heart. The images became phantoms. I was like a kid watching a horror movie, with my hands clapped to my eyes, peering anxiously through the cracks. I was Kwan's willing captive, and she was my protector.
At the end of her stories, Kwan would always say: "You're the only one who knows. Don't tell anyone. Never. Promise, Libby-ah?"
And I would always shake my head, then nod, drawn to allegiance through both privilege and fear.
One night, when my eyelids were already heavy with sleep, she started droning again in Chinese: "Libby-ah, I must tell you something, a forbidden secret. It's too much of a burden to keep inside me any longer."
I yawned, hoping she'd take the hint.
"I have yin eyes."
"What eyes?"
"It's true. I have yin eyes. I can see yin people."
"What do you mean?"
"Okay, I'll tell you. But first you must promise never to tell anyone. Never. Promise, ah?"
"Okay. Promise."
"Yin people, they are those who have already died."
My eyes popped open. "What? You see dead people? . . . You mean, _ghosts_?"
"Don't tell anyone. Never. Promise, Libby-ah?"
I stopped breathing. "Are there ghosts here now?" I whispered.
"Oh yes, many. Many, many good friends."
I threw the covers over my head. "Tell them to go away," I pleaded.
"Don't be afraid. Libby-ah, come out. They're your friends too. Oh see, now they're laughing at you for being so scared."
I began to cry. After a while, Kwan sighed and said in a disappointed voice, "All right, don't cry anymore. They're gone."
So that's how the business of ghosts got started. When I finally came out from under the covers, I saw Kwan sitting straight up, illuminated by the artificial glow of her American moon, staring out the window as if watching her visitors recede into the night.
The next morning, I went to my mother and did what I promised I'd never do: I told her about Kwan's yin eyes.
NOW THAT I'm an adult, I realize it wasn't my fault that Kwan went to the mental hospital. In a way, she brought it on herself. After all, I was just a little kid then, seven years old. I was scared out of my mind. I had to tell my mother what Kwan was saying. I thought Mom would just ask her to stop. Then Daddy Bob found out about Kwan's ghosts and blew his stack. Mom suggested taking her to Old St. Mary's for a talk with the priest. But Daddy Bob said no, confession wouldn't be enough. He booked Kwan into the psychiatric ward at Mary's Help instead.
When I visited her there the following week, Kwan whispered to me: "Libby-ah, listen, I have secret. Don't tell anyone, ah?" And then she switched to Chinese. "When the doctors and nurses ask me questions, I treat them like American ghosts—I don't see them, don't hear them, don't speak to them. Soon they'll know they can't change me, why they must let me go." I remember the way she looked, as immovable as a stone palace dog.
Unfortunately, her Chinese silent treatment backfired. The doctors thought Kwan had gone catatonic. Things being what they were back in the early 1960s, the doctors diagnosed Kwan's Chinese ghosts as a serious mental disorder. They gave her electroshock treatments, once, she said, then twice, she cried, then over and over again. Even today it hurts my teeth to think about that.
The next time I saw her at the hospital, she again confided in me. "All that electricity loosened my tongue so I could no longer stay silent as a fish. I became a country duck, crying _gwa-gwa-gwa!—_ bragging about the World of Yin. Then four bad ghosts shouted, 'How can you tell our secrets?' They gave me a _yin-yang tou—_ forced me to tear out half my hair. That's why the nurses shaved everything off. I couldn't stop pulling, until one side of my head was bald like a melon, the other side hairy like a coconut. The ghosts branded me for having two faces: one loyal, one traitor. But I'm not a traitor! Look at me, Libby-ah. Is my face loyal? What do you see?"
What I saw paralyzed me with fear. She looked as if she'd been given a crew cut with a hand-push lawn mower. It was as bad as seeing an animal run over on the street, wondering what it once had been. Except I knew how Kwan's hair used to be. Before, it flowed past her waist. Before, my fingers swam through its satin-black waves. Before, I'd grab her mane and yank it like the reins of a mule, shouting, "Giddyap, Kwan, say hee-haw!"
She took my hand and rubbed it across her sandpapery scalp, whispering about friends and enemies in China. On and on she went, as if the shock treatments had blown off the hinges of her jaw and she could not stop. I was terrified I'd catch her crazy talking disease.
To this day, I don't know why Kwan never blamed me for what happened. I'm sure she knew I was the one who got her in trouble. After she came back from Mary's Help, she gave me her plastic ID bracelet as a souvenir. She talked about the Sunday-school children who came to the hospital to sing "Silent Night," how they screamed when an old man yelled, "Shut up!" She reported that some patients there were possessed by ghosts, how they were not like the nice yin people she knew, and this was a real pity. Not once did she ever say, "Libby-ah, why did you tell my secret?"
Yet the way I remember it is the way I have always felt—that I betrayed her and that's what made her insane. The shock treatments, I believed, were my fault as well. They released all her ghosts.
THAT WAS more than thirty years ago, and Kwan still mourns, "My hair sooo bea-you-tiful, shiny-smooth like waterfall, slippery-cool like swimming eel. Now look. All that shock treatment, like got me bad home permanent, leave on cheap stuff too long. All my rich color—burnt out. All my softness—crinkle up. My hairs now just stiff wires, pierce message to my brain: No more yin-talking! They do this to me, hah, still I don't change. See? I stay strong."
Kwan was right. When her hair grew back, it was bristly, wiry as a terrier's. And when she brushed it, whole strands would crackle and rise with angry static, popping like the filaments of light bulbs burning out. Kwan explained, "All that electricity doctor force into my brain, now run through my body like horse go 'round racetrack." She claims that's the reason she now can't stand within three feet of a television set without its hissing back. She doesn't use the Walkman her husband, George, gave her; she has to ground the radio by placing it against her thigh, otherwise no matter what station she tunes it to, all she hears is "awful music, boom-pah-pah, boom-pah-pah." She can't wear any kind of watch. She received a digital one as a bingo prize, and after she strapped it on, the numbers started mutating like the fruits on a casino slot machine. Two hours later the watch stopped. "I gotta jackpot," she reported. "Eight-eight-eight-eight-eight. Lucky numbers, bad watch."
Although Kwan is not technically trained, she can pinpoint in a second the source of a fault in a circuit, whether it's in a wall outlet or a photo strobe. She's done that with some of my equipment. Here _I_ am, the commercial photographer, and _she_ can barely operate a point-and-shoot. Yet she's been able to find the specific part of the camera or cable or battery pack that was defective, and later, when I ship the camera to Cal Precision in Sacramento for troubleshooting, I'll find she was exactly right. I've also seen her temporarily activate a dead cordless phone just by pressing her fingers on the back recharger nodes. She can't explain any of this, and neither can I. All I can say is, I've seen her do these things.
The weirdest of her abilities, I think, has to do with diagnosing ailments. She can tell when she shakes hands with strangers whether they've ever suffered a broken bone, even if it healed many years before. She knows in an instant whether a person has arthritis, tendinitis, bursitis, sciatica—she's really good with all the musculoskeletal stuff— maladies that she calls "burning bones," "fever arms," "sour joints," "snaky leg," and all of which, she says, are caused by eating hot and cold things together, counting disappointments on your fingers, shaking your head too often with regret, or storing worries between your jaw and your fists. She can't cure anybody on the spot; she's no walking Grotto of Lourdes. But a lot of people say she has the healing touch. Like her customers at Spencer's, the drugstore in the Castro neighborhood where she works. Most of the people who pick up their prescriptions there are gay men—"bachelors," she calls them. And because she's worked there for more than twenty years, she's seen some of her longtime customers grow sick with AIDS. When they come in, she gives them quickie shoulder rubs, while offering medical advice: "You still drink beer, eat spicy food? Together, _same_ time? Wah! What I tell you? Tst! Tst! How you get well do this? Ah?"—as if they were little kids fussing to be spoiled. Some of her customers drop by every day, even though they can receive home delivery free. I know why. When she puts her hands on the place where you hurt, you feel a tingling sensation, a thousand fairies dancing up and down, and then it's like warm water rolling through your veins. You're not cured, but you feel released from worry, becalmed, floating on a tranquil sea.
Kwan once told me, "After they die, the yin bachelors still come visit me. They call me _Doctor_ Kwan. Joking, of course." And then she added shyly in English: "Maybe also for respect. What you think, Libby-ah?" She always asks me that: "What you think?"
No one in our family talks about Kwan's unusual abilities. That would call attention to what we already know, that Kwan is wacky, even by Chinese standards—even by San Francisco standards. A lot of the stuff she says and does would strain the credulity of most people who are not on antipsychotic drugs or living on cult farms.
But I no longer think my sister is crazy. Or if she is, she's fairly harmless, that is, if people don't take her seriously. She doesn't chant on the sidewalk like that guy on Market Street who screams that California is doomed to slide into the ocean like a plate of clams. And she's not into New Age profiteering; you don't have to pay her a hundred fifty an hour just to hear her reveal what's wrong with your past life. She'll tell you for free, even if you don't ask.
Most of the time, Kwan is like anyone else, standing in line, shopping for bargains, counting success in small change: "Libby-ah," she said during this morning's phone call, "yesterday, I buy two-for-one shoes on sale, Emporium Capwell. Guess how much I don't pay. You guess."
But Kwan is odd, no getting around that. Occasionally it amuses me. Sometimes it irritates me. More often I become upset, even angry—not with Kwan but with how things never turn out the way you hope. Why did I get Kwan for a sister? Why did she get me?
Every once in a while, I wonder how things might have been between Kwan and me if she'd been more normal. Then again, who's to say what's normal? Maybe in another country Kwan would be considered ordinary. Maybe in some parts of China, Hong Kong, or Taiwan she'd be revered. Maybe there's a place in the world where everyone has a sister with yin eyes.
KWAN'S NOW NEARLY FIFTY, whereas I'm a whole twelve years younger, a point she proudly mentions whenever anyone politely asks which of us is older. In front of other people, she likes to pinch my cheek and remind me that my skin is getting "wrinkle up" because I smoke cigarettes and drink too much wine and coffee—bad habits she does not have. "Don't hook on, don't need stop," she's fond of saying. Kwan is neither deep nor subtle; everything's right on the surface, for anybody to see. The point is, no one would ever guess we are sisters.
Kevin once joked that maybe the Communists sent us the wrong kid, figuring we Americans thought all Chinese people looked alike anyway. After hearing that, I fantasized that one day we'd get a letter from China saying, "Sorry, folks. We made a mistake." In so many ways, Kwan never fit into our family. Our annual Christmas photo looked like those children's puzzles, "What's Wrong with This Picture?" Each year, front and center, there was Kwan—wearing brightly colored summer clothes, plastic bow-tie barrettes on both sides of her head, and a loony grin big enough to burst her cheeks. Eventually, Mom found her a job as a bus-girl at a Chinese-American restaurant. It took Kwan a month to realize that the food they served there was supposed to be Chinese. Time did nothing to either Americanize her or bring out her resemblance to our father.
On the other hand, people tell me I'm the one who takes after him most, in both appearance and personality. "Look how much Olivia can eat without gaining an ounce," Aunt Betty is forever saying. "Just like Jack." My mother once said, "Olivia analyzes every single detail to death. She has her father's accountant mentality. No wonder she became a photographer." Those kinds of comments make me wonder what else has been passed along to me through my father's genes. Did I inherit from him my dark moods, my fondness for putting salt on my fruit, my phobia about germs?
Kwan, in contrast, is a tiny dynamo, barely five feet tall, a miniature bull in a china shop. Everything about her is loud and clashing. She'll wear a purple checked jacket over turquoise pants. She whispers loudly in a husky voice, sounding as if she had chronic laryngitis, when in fact she's never sick. She dispenses health warnings, herbal recommendations, and opinions on how to fix just about anything, from broken cups to broken marriages. She bounces from topic to topic, interspersing tips on where to find bargains. Tommy once said that Kwan believes in free speech, free association, free car-wash with fill-'er-up. The only change in Kwan's English over the last thirty years is in the speed with which she talks. Meanwhile, she thinks her English is great. She often corrects her husband. "Not _stealed_ ," she'll tell George. _"Stolened."_
In spite of all our obvious differences, Kwan thinks she and I are exactly alike. As she sees it, we're connected by a cosmic Chinese umbilical cord that's given us the same inborn traits, personal motives, fate, and luck. "Me and Libby-ah," she tells new acquaintances, "we same in here." And she'll tap the side of my head. "Both born Year the Monkey. Which one older? You guess. Which one?" And then she'll squash her cheek against mine.
Kwan has never been able to correctly pronounce my name, Olivia. To her, I will always be Libby-ah, not plain Libby, like the tomato juice, but Libby-ah, like the nation of Muammar Qaddafi. As a consequence, her husband, George Lew, his two sons from a first marriage, and that whole side of the family all call me Libby-ah too. The "ah" part especially annoys me. It's the Chinese equivalent of saying "hey," as in "Hey, Libby, come here." I asked Kwan once how she'd like it if I introduced her to everyone as "Hey, Kwan." She slapped my arm, went breathless with laughter, then said hoarsely, "I like, I like." So much for cultural parallels, Libby-ah it is, forever and ever.
I'm not saying I don't love Kwan. How can I not love my own sister? In many respects, she's been more like a mother to me than my real one. But I often feel bad that I don't want to be close to her. What I mean is, we're _close_ in a manner of speaking. We know things about each other, mostly through history, from sharing the same closet, the same toothpaste, the same cereal every morning for twelve years, all the routines and habits of being in the same family. I really think Kwan is sweet, also loyal, extremely loyal. She'd tear off the ear of anyone who said an unkind word about me. That counts for a lot. It's just that I wouldn't want to be closer to her, not the way some sisters are who consider themselves best friends. As it is, I don't share everything with her the way she does with me, telling me the most private details of her life— like what she told me last week about her husband:
"Libby-ah," she said, "I found mole, big as my nostril, found on— what you call this thing between man legs, in Chinese we say _yinnang,_ round and wrinkly like two walnut?"
"Scrotum."
"Yes-yes, found big mole on scrotum! Now every day–every day, must examine Georgie-ah, his scrotum, make sure this mole don't start grow."
To Kwan, there are no boundaries among family. Everything is open for gruesome and exhaustive dissection—how much you spent on your vacation, what's wrong with your complexion, the reason you look as doomed as a fish in a restaurant tank. And then she wonders why I don't make her a regular part of my social life. She, however, invites me to dinner once a week, as well as to every boring family gathering—last week, a party for George's aunt, celebrating the fact that she received her U.S. citizenship after fifty years, that sort of thing. Kwan thinks only a major catastrophe would keep me away. She'll worry aloud: "Why you don't come last night? Something the matter?"
"Nothing's the matter."
"Feel sick?"
"No."
"You want me come over, bring you orange? I have extra, good price, six for one dollar."
"Really, I'm _fine._ "
She's like an orphan cat, kneading on my heart. She's been this way all my life, peeling me oranges, buying me candy, admiring my report cards and telling me how smart I was, smarter than she could ever be. Yet I've done nothing to endear myself to her. As a child, I often refused to play with her. Over the years, I've yelled at her, told her she embarrassed me. I can't remember how many times I've lied to get out of seeing her.
Meanwhile, she has _always_ interpreted my outbursts as helpful advice, my feeble excuses as good intentions, my pallid gestures of affection as loyal sisterhood. And when I can't bear it any longer, I lash out and tell her she's crazy. Before I can retract the sharp words, she pats my arm, smiles and laughs. And the wound she bears heals itself instantly. Whereas I feel guilty forever.
IN RECENT MONTHS, Kwan has become even more troublesome. Usually after the third time I say no to something, she quits. Now it's as though her mind is stuck on automatic rewind. When I'm not irritated by her, I worry that maybe she's about to have a nervous breakdown again. Kevin said she's probably going through menopause. But I can tell it's more than that. She's more obsessed than usual. The ghost talk is becoming more frequent. She mentions China in almost every conversation with me, how she must go back before everything changes and it's too late. Too late for what? She doesn't know.
And then there's my marriage. She simply won't accept the fact that Simon and I have split up. In fact, she's purposely trying to sabotage the divorce. Last week, I gave a birthday party for Kevin and invited this guy I was seeing, Ben Apfelbaum. When he told Kwan he worked as a voice talent for radio commercials, she said, "Ah, Libby-ah and me too, both talent for get out of tricky situation, also big talent for get own way. Is true, Libby-ah?" Her eyebrows twitched. "You husband, Simon, I think he agree with me, ah?"
"My soon-to-be _ex-_ husband." I then had to explain to Ben: "Our divorce will be final five months from now, December fifteenth."
"Maybe not, maybe not," Kwan said, then laughed and pinched my arm. She turned to Ben: "You meet Simon?"
Ben shook his head and started to say, "Olivia and I met at the—"
"Oh, very handsome," Kwan chirped. She cupped her hand to the side of her mouth and confided: "Simon look like Olivia twin brother. Half Chinese."
"Half _Hawaiian_ ," I said. "And we don't look alike at all."
"What you mother father do?" Kwan scrutinized Ben's cashmere jacket.
"They're both retired and live in Missouri," said Ben.
"Misery! Tst! Tst!" She looked at me. "This too sad."
Every time Kwan mentions Simon, I think my brain is going to implode from my trying not to scream in exasperation. She thinks that because I initiated the divorce I can take it back.
"Why not forgive?" she said after the party. She was plucking at the dead blooms of an orchid plant. "Stubborn and anger together, very bad for you." When I didn't say anything, she tried another tack: "I think you still have strong feeling for him—mm-hm! Very, very strong. Ah— see!—look you face. So red! This love feeling rushing from you heart. I right? Answer. I right?"
And I kept flipping through the mail, scrawling MOVED across any envelope with Simon Bishop's name on it. I've never discussed with Kwan why Simon and I broke up. She wouldn't understand. It's too complex. There's no one event or fight I can put my finger on to say, "That was the reason." Our breakup was the result of many things: a wrong beginning, bad timing, years and years of thinking habit and silence were the same as intimacy. After seventeen years together, when I finally realized I needed more in my life, Simon seemed to want less. Sure, I loved him—too much. And he loved me, only not enough. I just want someone who thinks I'm number one in his life. I'm not willing to accept emotional scraps anymore.
But Kwan wouldn't understand that. She doesn't know how people can hurt you beyond repair. She believes people who say they're sorry. She's the naive, trusting type who believes everything said in television commercials is certifiable truth. Look at her house: it's packed to the gills with gadgets—Ginsu knives, slicers and dicers, juicers and french-fry makers, you name it, she's bought it, for "only nineteen ninety-five, order now, offer good until midnight."
"Libby-ah," Kwan said on the phone today. "I have something must tell you, very important news. This morning I talk to Lao Lu. We decide: You and Simon shouldn't get divorce."
"How nice," I said. "You decided." I was balancing my checkbook, adding and subtracting as I pretended to listen.
"Me and Lao Lu. You remember him."
"George's cousin." Kwan's husband seemed to be related to just about every Chinese person in San Francisco.
"No-no! Lao Lu not cousin. How you can forget? Lots times I already tell you about him. Old man, bald head. Strong arm, strong leg, strong temper. One time loose temper, loose head too! Chopped off. Lao Lu say—"
"Wait a minute. Someone without a head is now telling _me_ what to do about my marriage?"
"Tst! Chopped head off over one hundred year ago. Now look fine, no problem. Lao Lu think you, me, Simon, we three go China, everything okay. Okay, Libby-ah?"
I sighed. "Kwan, I really don't have time to talk about this now. I'm in the middle of something."
"Lao Lu say cannot just balance checkbook, see how much you got left. Must balance life too."
How the hell did Kwan know I was balancing my checkbook?
That's how it's been with Kwan and me. The minute I discount her, she tosses in a zinger that keeps me scared, makes me her captive once again. With her around, I'll never have a life of my own. She'll always claim a major interest.
Why do I remain her treasured little sister? Why does she feel that I'm the most important person in her life?—the most! Why does she say over and over again that even if we were not sisters, she would feel this way? "Libby-ah," she tells me, "I never leave you."
No! I want to shout, I've done nothing, don't say that anymore. Because each time she does, she turns all my betrayals into love that needs to be repaid. Forever we'll know: She's been loyal, someday I'll have to be.
But even if I cut off both my hands, it'd be no use. As Kwan has already said, she'll never release me. One day the wind will howl and she'll be clutching a tuft of the straw roof, about to fly off to the World of Yin.
"Let's go! Hurry come!" she'll be whispering above the storm. "But don't tell anyone. Promise me, Libby-ah."
## [2
FISHER OF MEN](TheHundredSecretSenses-toc.html#TOC-3)
Before seven in the morning, the phone rings. Kwan is the only one who would call at such an ungodly hour. I let the answering machine pick up.
"Libby-ah?" she whispers. "Libby-ah, you there? This you big sister . . . Kwan. I have something important tell you. . . . You want hear? . . . Last night I dream you and Simon. Strange dream. You gone to bank, check you savings. All a sudden, bank robber run through door. Quick! You hide you purse. So bank robber, he steal everybody money but yours. Later, you gone home, stick you hand in purse—ah!—where is?—gone! Not money but you heart. Stolened! Now you have no heart, how can live? No energy, no color in cheek, pale, sad, tired. Bank president where you got all you savings, he say, 'I loan you my heart. No interest. You pay back whenever.' You look up, see his face—you know who, Libby-ah? You guess. . . . Simon! Yes-yes, give you his heart. You see! Still love you. Libby-ah, do you believe? Not just dream . . . Libby-ah, you listening me?"
BECAUSE OF KWAN, I have a talent for remembering dreams. Even today, I can recall eight, ten, sometimes a dozen dreams. I learned how when Kwan came home from Mary's Help. As soon as I started to wake, she would ask: "Last night, Libby-ah, who you meet? What you see?"
With my half-awake mind, I'd grab on to the wisps of a fading world and pull myself back in. From there I would describe for her the details of the life I'd just left—the scuff marks on my shoes, the rock I had dislodged, the face of my true mother calling to me from underneath. When I stopped, Kwan would ask, "Where you go before that?" Prodded, I would trace my way back to the previous dream, then the one before that, a dozen lives, and sometimes their deaths. Those are the ones I never forget, the moments just before I died.
Through years of dream-life, I've tasted cold ash falling on a steamy night. I've seen a thousand spears flashing like flames on the crest of a hill. I've touched the tiny grains of a stone wall while waiting to be killed. I've smelled my own musky fear as the rope tightens around my neck. I've felt the heaviness of flying through weightless air. I've heard the sucking creak of my voice just before life snaps to an end.
"What you see after die?" Kwan would always ask.
I'd shake my head. "I don't know. My eyes were closed."
"Next time, open eyes."
For most of my childhood, I thought everyone remembered dreams as other lives, other selves. Kwan did. After she came home from the psychiatric ward, she told me bedtime stories about them, yin people: a woman named Banner, a man named Cape, a one-eyed bandit girl, a half-and-half man. She made it seem as if all these ghosts were our friends. I didn't tell my mother or Daddy Bob what Kwan was saying. Look what happened the last time I did that.
When I went to college and could finally escape from Kwan's world, it was already too late. She had planted her imagination into mine. Her ghosts refused to be evicted from my dreams.
"Libby-ah," I can still hear Kwan saying in Chinese, "did I ever tell you what Miss Banner promised before we died?"
I see myself pretending to be asleep.
And she would go on: "Of course, I can't say exactly how long ago this happened. Time is not the same between one lifetime and the next. But I think it was during the year 1864. Whether this was the Chinese lunar year or the date according to the Western calendar, I'm not sure. . . ."
Eventually I would fall asleep, at what point in her story I always forgot. So which part was her dream, which part was mine? Where did they intersect? Every night, she'd tell me these stories. And I would lie there silently, helplessly, wishing she'd shut up.
Yes, yes, I'm sure it was 1864. I remember now, because the year sounded very strange. Libby-ah, just listen to it: _Yi-ba-liu-si._ Miss Banner said it was like saying: Lose hope, slide into death. And I said, No, it means: Take hope, the dead remain. Chinese words are good and bad this way, so many meanings, depending on what you hold in your heart.
Anyway, that was the year I gave Miss Banner the tea. And she gave me the music box, the one I once stole from her, then later returned. I remember the night we held that box between us with all those things inside that we didn't want to forget. It was just the two of us, alone for the moment, in the Ghost Merchant's House, where we lived with the Jesus Worshippers for six years. We were standing next to the holy bush, the same bush that grew the special leaves, the same leaves I used to make the tea. Only now the bush was chopped down, and Miss Banner was saying she was sorry that she let General Cape kill that bush. Such a sad, hot night, water streaming down our faces, sweat and tears, the cicadas screaming louder and louder, then falling quiet. And later, we stood in this archway, scared to death. But we were also happy. We were happy to learn we were unhappy for the same reason. That was the year that both our heavens burned.
Six years before, that's when I first met her, when I was fourteen and she was twenty-six, maybe younger or older than that. I could never tell the ages of foreigners. I came from a small place in Thistle Mountain, just south of Changmian. We were not Punti, the Chinese who claimed they had more Yellow River Han blood running through their veins, so everything should belong to them. And we weren't one of the Zhuang tribes either, always fighting each other, village against village, clan against clan. We were Hakka, Guest People—hnh!—meaning, guests not invited to stay in any good place too long. So we lived in one of many Hakka roundhouses in a poor part of the mountains, where you must farm on cliffs and stand like a goat and unearth two wheelbarrows of rocks before you can grow one handful of rice.
All the women worked as hard as the men, no difference in who carried the rocks, who made the charcoal, who guarded the crops from bandits at night. All Hakka women were this way, strong. We didn't bind our feet like Han girls, the ones who hopped around on stumps as black and rotten as old bananas. We had to walk all over the mountain to do our work, no binding cloths, no shoes. Our naked feet walked right over those sharp thistles that gave our mountain its famous name.
A suitable Hakka bride from our mountains had thick calluses on her feet and a fine, high-boned face. There were other Hakka families living near the big cities of Yongan, in the mountains, and Jintian, by the river. And the mothers from poorer families liked to match their sons to hardworking pretty girls from Thistle Mountain. During marriage-matching festivals, these boys would climb up to our high villages and our girls would sing the old mountain songs that we had brought from the north a thousand years before. A boy had to sing back to the girl he wanted to marry, finding words to match her song. If his voice was soft, or his words were clumsy, too bad, no marriage. That's why Hakka people are not only fiercely strong, they have good voices, and clever minds for winning whatever they want.
We had a saying: When you marry a Thistle Mountain girl, you get three oxen for a wife: one that breeds, one that plows, one to carry your old mother around. That's how tough a Hakka girl was. She never complained, even if a rock tumbled down the side of the mountain and smashed out her eye.
That happened to me when I was seven. I was very proud of my wound, cried only a little. When my grandmother sewed shut the hole that was once my eye, I said the rock had been loosened by a ghost horse. And the horse was ridden by the famous ghost maiden Nunumu—the _nu_ that means "girl," the _numu_ that means "a stare as fierce as a dagger." Nunumu, Girl with the Dagger Eye. She too lost her eye when she was young. She had witnessed a Punti man stealing another man's salt, and before she could run away, he stabbed his dagger in her face. After that, she pulled one corner of her headscarf over her blind eye. And her other eye became bigger, darker, sharp as a cat-eagle's. She robbed only Punti people, and when they saw her dagger eye, oh, how they trembled.
All the Hakkas in Thistle Mountain admired her, and not just because she robbed Punti people. She was the first Hakka bandit to join the struggle for Great Peace when the Heavenly King came back to us for help. In the spring, she took an army of Hakka maidens to Guilin, and the Manchus captured her. After they cut off her head, her lips were still moving, cursing that she would return and ruin their families for one hundred generations. That was the summer I lost my eye. And when I told everyone about Nunumu galloping by on her ghost horse, people said this was a sign that Nunumu had chosen me to be her messenger, just as the Christian God had chosen a Hakka man to be the Heavenly King. They began to call me Nunumu. And sometimes, late at night, I thought I could truly see the Bandit Maiden, not too clearly, of course, because at that time I had only one yin eye.
Soon after that, I met my first foreigner. Whenever foreigners arrived in our province, everyone in the countryside—from Nanning to Guilin—talked about them. Many Westerners came to trade in foreign mud, the opium that gave foreigners mad dreams of China. And some came to sell weapons—cannons, gunpowder, rifles, not the fast, new ones, but the slow, old kind you light with a match, leftovers from foreign battles already lost. The missionaries came to our province because they heard that the Hakkas were God Worshippers. They wanted to help more of us go to their heaven. They didn't know that a God Worshipper was not the same as a Jesus Worshipper. Later we all realized our heavens were not the same.
But the foreigner I met was not a missionary. He was an American general. The Hakka people called him Cape because that's what he always wore, a large cape, also black gloves, black boots, no hat, and a short gray jacket with buttons—like shiny coins!—running from the waist to his chin. In his hand he carried a long walking stick, rattan, with a silver tip and an ivory handle carved in the shape of a naked woman.
When he came to Thistle Mountain, people from all the villages poured down the mountainsides and met in the wide green bowl. He arrived on a prancing horse, leading fifty Cantonese soldiers, former boatmen and beggars, now riding ponies and wearing colorful army uniforms, which we heard were not Chinese or Manchu but leftovers from wars in French Africa. The soldiers were shouting, "God Worshippers! We are God Worshippers too!"
Some of our people thought Cape was Jesus, or, like the Heavenly King, another one of his younger brothers. He was very tall, had a big mustache, a short beard, and wavy black hair that flowed to his shoulders. Hakka men also wore their long hair this way, no pigtail anymore, because the Heavenly King said our people should no longer obey the laws of the Manchus. I had never seen a foreigner before and had no way of knowing his true age. But to me, he looked old. He had skin the color of a turnip, eyes as murky as shallow water. His face had sunken spots and sharp points, the same as a person with a wasting disease. He seldom smiled, but laughed often. And he spoke harsh words in a donkey bray. A man always stood by his side, serving as his go-between, translating in an elegant voice what Cape said.
The first time I saw the go-between, I thought he looked Chinese. The next minute he seemed foreign, then neither. He was like those lizards that become the colors of sticks and leaves. I learned later this man had the mother blood of a Chinese woman, the father blood of an American trader. He was stained both ways. General Cape called him _yiban ren,_ the one-half man.
Yiban told us Cape had just come from Canton, where he became friends with the Heavenly King of the Great Peace Revolution. We were all astounded. The Heavenly King was a holy man who had been born a Hakka, then chosen by God to be his treasured younger son, little brother to Jesus. We listened carefully.
Cape, Yiban said, was an American military leader, a supreme general, the highest rank. People murmured. He had come across the sea to China, to help the God Worshippers, the followers of Great Peace. People shouted, "Good! Good!" He was a God Worshipper himself, and he admired us, our laws against opium, thievery, the pleasures of the dark parts of women's bodies. People nodded, and I stared with my one eye at the naked lady on the handle of Cape's walking stick. He said that he had come to help us win our battle against the Manchus, that this was God's plan, written more than a thousand years before in the Bible he was holding. People pushed forward to see. We knew that same plan. The Heavenly King had already told us that the Hakka people would inherit the earth and rule God's Chinese kingdom. Cape reported the Great Peace soldiers had already captured many cities, had gathered much money and land. And now, the struggle was ready to move north— if only the rest of the God Worshippers in Thistle Mountain would join him as soldiers. Those who fought, he added, would share in the bounty—warm clothes, plenty to eat, weapons, and later, land of their own, new status and ranks, schools and homes, men and women separate. The Heavenly King would send food to their families left behind. By now, everybody was shouting, "Great Peace! Great Peace!"
Then General Cape tapped his walking stick on the ground. Everyone grew quiet again. He called Yiban to show us the gifts the Heavenly King had asked him to bring. Barrels of gunpowder! Bushels of rifles! Baskets of French African uniforms, some torn and already stained with blood. But everyone agreed they were still very fine. Everybody was saying, "Hey, see these buttons, feel this cloth." That day, many, many people, men and women, joined the army of the Heavenly King. I could not. I was too young, only seven, so I was very unhappy inside. But then the Cantonese soldiers passed out uniforms—only to the men, none to the women. And when I saw that, I was not as unhappy as before.
The men put on their new clothes. The women examined their new rifles, the matches for lighting them. Then General Cape tapped his walking stick again and asked Yiban to bring out his gift to us. We all pressed forward, eager to see yet another surprise. Yiban brought back a wicker cage, and inside was a pair of white doves. General Cape announced in his curious Chinese that he had asked God for a sign that we would be an ever victorious army. God sent down the doves. The doves, General Cape said, meant we poor Hakkas would have the rewards of Great Peace we had hungered for over the last thousand years. He then opened the cage door and pulled out the birds. He threw them into the air, and the people roared. They ran and pushed, jumping to catch the creatures before they could fly away. One man fell forward onto a rock. His head cracked open and his brains started to pour out. But people jumped right over him and kept chasing those rare and precious birds. One dove was caught, the other flew away. So someone ate a meal that night.
My mother and father joined the struggle. My uncles, my aunts, my older brothers, nearly everyone over thirteen in Thistle Mountain and from the cities down below. Fifty or sixty thousand people. Peasants and landowners, soup peddlers and teachers, bandits and beggars, and not just Hakkas, but Yaos and Miaos, Zhuang tribes, and even the Puntis who were poor. It was a great moment for Chinese people, all of us coming together like that.
I was left behind in Thistle Mountain to live with my grandmother. We were a pitiful village of scraps, babies and children, the old and the lame, cowards and idiots. Yet we were happy, because just as he had promised, the Heavenly King sent his soldiers to bring us food, more kinds than we could have ever imagined in a hundred years. And the soldiers also brought us stories of great victories: How the Heavenly King had set up his new kingdom in Nanjing. How taels of silver were more plentiful than rice. What fine houses everyone lived in, men in one compound, women in another. What a peaceful life—church on Sunday, no work, only rest and happiness. We were glad to hear that we now lived in a time of Great Peace.
The following year, the soldiers came with rice and salt-cured fish. The next year, it was only rice. More years passed. One day, a man who had once lived in our village returned from Nanjing. He said he was sick to death of Great Peace. When there is great suffering, he said, everyone struggles the same. But when there is peace, no one wants to be the same. The rich no longer share. The less rich envy and steal. In Nanjing, he said, everyone was seeking luxuries, pleasures, the dark places of women. He said the Heavenly King now lived in a fine palace and had many concubines. He allowed his kingdom to be ruled by a man possessed with the Holy Ghost. And General Cape, the man who rallied all the Hakkas to fight, had joined the Manchus and was now a traitor, bound by a Chinese banker's gold and marriage to his daughter. Too much happiness, said the man who returned, always overflows into tears of sorrow.
We could feel in our stomachs the truth of what this man said. We were hungry. The Heavenly King had forgotten us. Our Western friends had betrayed us. We no longer received food or stories of victory. We were poor. We had no mothers, no fathers, no singing maidens and boys. We were bitter cold in the wintertime.
The next morning, I left my village and went down the mountain. I was fourteen, old enough to make my own way in life. My grandmother had died the year before, but her ghost didn't stop me. It was the ninth day of the ninth month, I remember this, a day when Chinese people were supposed to climb the heights, not descend from them, a day for honoring ancestors, a day that the God Worshippers ignored to prove they abided by a Western calendar of fifty-two Sundays and not the sacred days of the Chinese almanac. So I walked down the mountain, then through the valleys between the mountains. I no longer knew what I should believe, whom I could trust. I decided I would wait for a sign, see what happened.
I arrived at the city by the river, the one called Jintian. To those Hakka people I met, I said I was Nunumu. But they didn't know who the Bandit Maiden was. She was not famous in Jintian. The Hakkas there didn't admire my eye that a ghost horse had knocked out. They pitied me. They put an old rice ball into my palm and tried to make me a half-blind beggar. But I refused to become what people thought I should be.
So I wandered around the city again, thinking about what work I might do to earn my own food. I saw Cantonese people who cut the horns off toes, Yaos who pulled teeth, Puntis who pierced needles into swollen legs. I knew nothing about drawing money out of the rotten parts of other people's bodies. I continued walking until I was beside the low bank of a wide river. I saw Hakka fishermen tossing big nets into the water from little boats. But I had no nets, no little boat. I did not know how to think like a fast, sly fish.
Before I could decide what to do, I heard people along the riverbank shouting. Foreigners had arrived! I ran to the dock and watched two Chinese _kuli_ boatmen, one young, one old, walking down a narrow plank, carrying boxes and crates and trunks from a large boat. And then I saw the foreigners themselves, standing on the deck—three, four, five of them, all in dull black clothes, except for the smallest one, who had clothing and hair the shiny brown of a tree-eating beetle. That was Miss Banner, but of course I didn't know it at the time. My one eye watched them all. Their five pairs of foreign eyes were on the young and old boatmen balancing their way down the long, thin gangplank. On the shoulders of the boatmen were two poles, and in the saggy middle a large trunk hung from twisted ropes. Suddenly, the shiny brown foreigner ran down the plank—who knew why?—to warn the men, to ask them to be more careful. And just as suddenly, the plank began to bounce, the trunk began to swing, the men began to sway, and the five foreigners on the boat began to shout. Back and forth, up and down—our eyes leapt as we watched those boatmen clenching their muscles and the shiny foreigner flapping her arms like a baby bird. In the next moment, the older man, at the bottom of the plank, gave one sharp cry—I heard the crack, saw his shoulder bone sticking out. Then two _kuli_ s, one trunk, and a shiny-clothed foreigner fell with great splashes into the water below.
I ran to the river edge. The younger _kuli_ had already swum to shore. Two fishermen in a small boat were chasing the contents that had spilled out of the trunk, bright clothing that billowed like sails, feathered hats that floated like ducks, long gloves that raked the water like the fingers of a ghost. But nobody was trying to help the injured boatman or the shiny foreigner. The other foreigners would not; they were afraid to walk down the plank. The Punti people on the shore would not; if they interfered with fate, they would be responsible for those two people's undrowned lives. But I didn't think this way. I was a Hakka. The Hakkas were God Worshippers. And the God Worshippers were fishers of men. So I grabbed one of the bamboo poles that had fallen in the water. I ran along the bank and stuck this out, letting the ropes dangle downstream. The _kuli_ and the foreigner grabbed them with their eager hands. And with all my strength, I pulled them in.
Right after that the Punti people pushed me aside. They left the injured boatman on the ground, gasping and cursing. That was Lao Lu, who later became the gatekeeper, since with a broken shoulder he could no longer work as a _kuli._ As for Miss Banner, the Puntis dragged her higher onto the shore, where she vomited, then cried. When the foreigners finally came down from the boat, the Puntis crowded around them, shouting, "Give us money." One of the foreigners threw small coins on the ground, and the Puntis flocked like birds to devour them, then scattered away.
The foreigners loaded Miss Banner in one cart, the broken boatman in another. They loaded three more carts with their boxes and crates and trunks. And as they made their way to the mission house in Changmian, I ran behind. So that's how all three of us went to live in the same house. Our three different fates had flowed together in that river, and became as tangled and twisted as a drowned woman's hair.
It was like this: If Miss Banner had not bounced on the plank, Lao Lu never would have broken his shoulder. If his shoulder had not broken, Miss Banner never would have almost drowned. If I hadn't saved Miss Banner from drowning, she never would have been sorry for breaking Lao Lu's shoulder. If I hadn't saved Lao Lu, he never would have told Miss Banner what I had done. If Miss Banner hadn't known this, she never would have asked me to be her companion. If I hadn't become her companion, she wouldn't have lost the man she loved.
THE GHOST MERCHANT'S HOUSE was in Changmian, and Changmian was also in Thistle Mountain, but north of my village. From Jintian it was a half-day's journey. But with so many trunks and moaning people in carts, we took twice as long. I learned later that _Changmian_ means "never-ending songs." Behind the village, higher into the mountains, were many caves, hundreds. And when the wind blew, the mouths of the caves would sing _wu! wu!—_ just like the voices of sad ladies who have lost sons.
That's where I stayed for the last six years of my life—in that house. I lived with Miss Banner, Lao Lu, and the missionaries—two ladies, two gentlemen, Jesus Worshippers from England. I didn't know this at the time. Miss Banner told me many months later, when we could speak to each other in a common tongue. She said the missionaries had sailed to Macao, preached there a little while, then sailed to Canton, preached there another little while. That's also where they met Miss Banner. Around this time, a new treaty came out saying the foreigners could live anywhere in China they pleased. So the missionaries floated inland to Jintian, using West River. And Miss Banner was with them.
The mission was a large compound, with one big courtyard in the middle, then four smaller ones, one big fancy main house, then three smaller ones. In between were covered passageways to connect everything together. And all around was a high wall, cutting off the inside from the outside. No one had lived in that place for more than a hundred years. Only foreigners would stay in a house that was cursed. They said they didn't believe in Chinese ghosts.
Local people told Lao Lu, "Don't live there. It's haunted by fox-spirits." But Lao Lu said he was not afraid of anything. He was a Cantonese _kuli_ descended from ten generations of _kuli_ s! He was strong enough to work himself to death, smart enough to find the answer to whatever he wanted to know. For instance, if you asked him how many pieces of clothing did the foreign ladies own, he wouldn't guess and say maybe two dozen each. He would go into the ladies' rooms when they were eating, and he would count each piece, never stealing any, of course. Miss Banner, he told me, had two pairs of shoes, six pairs of gloves, five hats, three long costumes, two pairs of black stockings, two pairs of white stockings, two pairs of white undertrousers, one umbrella, and seven other things that may have been clothing, but he could not determine which parts of the body they were supposed to cover.
Through Lao Lu, I quickly learned many things about the foreigners. Only later did he tell me why local people thought the house was cursed. Many years before, it had been a summer mansion, owned by a merchant who died in a mysterious and awful way. Then his wives died, four of them, one by one, also in mysterious and awful ways, youngest first, oldest last, all of this happening from one full moon to the next.
Like Lao Lu, I was not easily scared. But I must tell you, Libby-ah, what happened there five years later made me believe the Ghost Merchant had come back.
## [3
THE DOG AND THE BOA](TheHundredSecretSenses-toc.html#TOC-4)
Ever since we separated, Simon and I have been having a custody spat over Bubba, _my_ dog. Simon wants visitation rights, weekend walks. I don't want to deny him the privilege of picking up Bubba's poop. But I hate his cavalier attitude about dogs. Simon likes to walk Bubba off leash. He lets him romp through the trails of the Presidio, along the sandy dog run by Crissy Field, where the jaws of a pit bull, a rottweiler, even a mad cocker spaniel could readily bite a three-pound Yorkie-chihuahua in half.
This evening, we were at Simon's apartment, sorting through a year's worth of receipts for the free-lance business we haven't yet divided. For the sake of tax deductions, we decided "married filing joint return" should still apply.
"Bubba's a dog," Simon said. "He has the right to run free once in a while."
"Yeah, and get himself killed. Remember what happened to Sarge?"
Simon rolled his eyes, his look of "Not that again." Sarge had been Kwan's dog, a scrappy Pekingese-Maltese that challenged any male dog on the street. About five years ago, Simon took him for a walk—off leash—and Sarge tore open the nose of a boxer. The owner of the boxer presented Kwan with an eight-hundred-dollar veterinary bill. I insisted Simon should pay. Simon said the boxer's owner should, since his dog had provoked the attack. Kwan squabbled with the animal hospital over each itemized charge.
"What if Bubba runs into a dog like Sarge?" I said.
"The boxer started it," Simon said flatly.
"Sarge was a vicious dog! You were the one who let him off leash, and Kwan ended up paying the vet bill!"
"What do you mean? The boxer's owner paid."
"Oh no, he didn't. Kwan just said that so you wouldn't feel bad. I told you that, remember?"
Simon twisted his mouth to the side, a grimace of his that always preceded a statement of doubt. "I don't remember that," he said.
"Of course you don't! You remember what you _want_ to remember."
Simon sneered. "Oh, and I suppose you don't?" Before I could respond, he held up his hand, palm out, to stop me. "I know, I know. _You_ have an indelible memory! _You_ can never forget a thing! Well, let me tell you, your recollection of every last detail has nothing to do with memory. It's called holding a goddamn grudge."
WHAT SIMON SAID has annoyed me all night long. Am I really the kind of person who hangs on to resentments? No, Simon was being defensive, throwing back barbs. Can I help it if I was born with a knack for remembering all sorts of things?
Aunt Betty was the first person to tell me I had a photographic memory; her comment made me believe I would grow up to be a photographer. She said this because I once corrected her in front of a bunch of people on her account of a movie we had all seen together. Now that I've been making my living behind the camera lens for the last fifteen years, I don't know what people mean by photographic memory. How I remember the past isn't like flipping through an indiscriminate pile of snapshots. It's more selective than that.
If someone asked me what my address was when I was seven years old, the numbers wouldn't flash before my eyes. I'd have to relive a specific moment: the heat of the day, the smell of the cut lawn, the slap-slap-slap of rubber thongs against my heels. Then once again I'd be walking up the two steps of the poured-concrete porch, reaching into the black mailbox, heart pounding, fingers grasping— Where is it? Where's that stupid letter from Art Linkletter, inviting me to be on his show? But I wouldn't give up hope. I'd think to myself, Maybe I'm at the wrong address. But no, there they are, the brass numbers above, 3-6-2-4, complete with tarnish and rust around the screws.
That's what I remember most, not addresses but pain—that old lump-in-the-throat conviction that the world had fingered me for abuse and neglect. Is that the same as a grudge? I wanted so much to be a guest on _Kids Say the Darndest Things._ It was the kiddie route to fame, and I wanted once again to prove to my mother that I was special, in spite of Kwan. I wanted to snub the neighborhood kids, to make them mad that I was having more fun than they would ever know. While riding my bicycle around and around the block, I'd plot what I'd say when I was finally invited to be on the show. I'd tell Mr. Linkletter about Kwan, just the funny stuff—like the time she said she loved the movie _Southern Pacific._ Mr. Linkletter would raise his eyebrows and round his mouth. "Olivia," he'd say, "doesn't your sister mean _South_ Pacific?" Then people in the audience would slap their knees and roar with laughter, and I'd glow with childish wonder and a cute expression.
Old Art always figured kids were so sweet and naive they didn't know they were saying embarrassing things. But all those kids on the show knew precisely what they were doing. Why else didn't they ever mention the _real_ secrets—how they played night-night nurse and dickie doctor, how they stole gum, gunpowder caps, and muscle magazines from the corner Mexican store. I knew kids who did those things. They were the same ones who once pinned down my arms and peed on me, laughing and shouting, "Olivia's sister is a retard." They sat on me until I started crying, hating Kwan, hating myself.
To soothe me, Kwan took me to the Sweet Dreams Shoppe. We were sitting outside, licking cones of rocky road ice cream. Captain, the latest mutt my mother had rescued from the pound, whom Kwan had named, was lying at our feet, vigilantly waiting for drips.
"Libby-ah," Kwan said, "what this word, lee-tahd?"
"Reee-tard," I corrected, lingering over the word. I was still angry with Kwan and the neighbor kids. I took another tongue stab of ice cream, thinking of retarded things Kwan had done. "Retard means _fan-tou_ ," I said. "You know, a stupid person who doesn't understand anything." She nodded. "Like saying the wrong things at the wrong time," I added. She nodded again. "When kids laugh at you and you don't know why."
Kwan was quiet for the longest time, and the inside of my chest began to feel tickly and uncomfortable. Finally she said in Chinese: "Libby-ah, you think this word is me, _retard_? Be honest."
I kept licking the drips running down the side of my cone, avoiding her stare. I noticed that Captain was also watching me attentively. The tickly feeling grew, until I let out a huge sigh and grumbled, "Not really." Kwan grinned and patted my arm, which just about drove me crazy. "Captain," I shouted. "Bad dog! Stop begging!" The dog cowered.
"Oh, he not begging," Kwan said in a happy voice. "Only hoping." She petted his rump, then held her cone above the dog's head. "Talk English!" Captain sneezed a couple of times, then let go with a low _wuff._ She allowed him a lick. _"Jang Zhongwen!_ Talk Chinese!" Two high-pitched yips followed. She gave him another lick, then another, sweet-talking him in Chinese. And it annoyed me to see this, how any dumb thing could make her and the dog instantly happy.
Later that same night Kwan asked me again about what those kids had said. She pestered me so much I thought she really was retarded.
Libby-ah, are you sleeping? Okay, sorry, sorry, go back to sleep, it's not important. . . . I only wanted to ask you again about this word, _retard._ Ah, but you're sleeping now, maybe tomorrow, after you come home from school. . . .
Funny, I was thinking how I once thought Miss Banner was this way, _retard._ She didn't understand anything. . . . Libby-ah, did you know I taught Miss Banner to talk? Libby-ah? Sorry, sorry, go back to sleep, then.
It's true, though. I was her teacher. When I first met her, her speech was like a baby's! Sometimes I laughed, I couldn't help it. But she did not mind. The two of us had a good time saying the wrong things all the time. We were like two actors at a temple fair, using our hands, our eyebrows, the fast twist of our feet to show each other what we meant. That's how she told me about her life before she came to China. What I thought she said was this:
She was born to a family who lived in a village far, far west of Thistle Mountain, across a tumbling sea. It was past the country where black people live, beyond the land of English soldiers and Portuguese sailors. Her family village was bigger than all these lands put together. Her father owned many ships that crossed this sea to other lands. In these lands, he gathered money that grew like flowers, and the smell of this money made many people happy.
When Miss Banner was five, her two little brothers chased a chicken into a dark hole. They fell all the way to the other side of the world. Naturally, the mother wanted to find them. Before the sun came up and after the sun came down, she puffed out her neck like a rooster and called for her lost sons. After many years, the mother found the same hole in the earth, climbed in, and then she too fell all the way to the other side of the world.
The father told Miss Banner, We must search for our lost family. So they sailed across the tumbling sea. First they stopped at a noisy island. Her father took her to live in a large palace ruled by tiny people who looked like Jesus. While her father was in the fields picking more flower-money, the little Jesuses threw stones at her and cut off her long hair. Two years later, when her father returned, he and Miss Banner sailed to another island, this one ruled by mad dogs. Again he put Miss Banner in a large palace and went off to pick more flower-money. While he was gone, the dogs chased Miss Banner and tore at her dress. She ran around the island, searching for her father. She met an uncle instead. She and this uncle sailed to a place in China where many foreigners lived. She did not find her family there. One day, as she and the uncle lay in bed, the uncle became hot and cold at the same time, rose up in the air, then fell into the sea. Lucky for her, Miss Banner met another uncle, a man with many guns. He took her to Canton, where foreigners also lived. Every night, the uncle laid his guns on the bed and made her polish them before she could sleep. One day, this man cut off a piece of China, one with many fine temples. He sailed home on this floating island, gave the temples to his wife, the island to his king. Miss Banner met a third uncle, a Yankee, also with many guns. But this one combed her hair. He fed her peaches. She loved this uncle very much. One night, many Hakka men burst into their room and took her uncle away. Miss Banner ran to the Jesus Worshippers for help. They said, Fall on your knees. So she fell on her knees. They said, Pray. So she prayed. Then they took her inland to Jintian, where she fell in the water and prayed to be saved. That's when I saved her.
Later on, as Miss Banner learned more Chinese words, she told me about her life again, and because what I heard was now different, what I saw in my mind was different too. She was born in America, a country beyond Africa, beyond England and Portugal. Her family village was near a big city called _Nu Ye,_ sounds like Cow Moon. Maybe this was New York. A company called Russia or Russo owned those ships, not her father. He was a clerk. The shipping company bought opium in India— those were the flowers—then sold it in China, spreading a dreaming sickness among Chinese people.
When Miss Banner was five, her little brothers did not chase chickens into a hole, they died of chicken pox and were buried in their backyard. And her mother did not puff her neck out like a rooster. Her throat swelled up and she died of a goiter disease and was buried next to her sons. After this tragedy, Miss Banner's father took her to India, which was not ruled by little Jesuses. She went to a school for Jesus-worshipping children from England, and they were not holy but naughty and wild. Later, her father took her to Malacca, which was not ruled by dogs. She was talking about another school, where the children were also English and even more disobedient than the ones in India. Her father sailed off to buy more opium in India but never returned—why, she did not know, so she grew many kinds of sadness in her heart. Now she had no father, no money, no home. When she was still a young maiden, she met a man who took her to Macao. Lots of mosquitoes in Macao; he died of malaria there and was buried at sea. Then she lived with another man, this one an English captain. He helped the Manchus, fought the God Worshippers, earned big money for each city he captured. Later, he sailed home, bearing many looted temple treasures for England and his wife. Miss Banner then went to live with another soldier, a Yankee. This one, she said, helped the God Worshippers, fought against the Manchus, also earned money by looting the cities he and the God Worshippers burned to the ground. These three men, Miss Banner told me, were not her uncles.
I said to her, "Miss Banner-ah, this is good news. Sleeping in the same bed with your uncles is not good for your aunts." She laughed. So you see, by this time, we could laugh together because we understood each other very well. By this time, the calluses on my feet had been exchanged for an old pair of Miss Banner's tight leather shoes. But before this happened, I had to teach her how to talk.
To begin, I told her my name was Nunumu. She called me Miss Moo. We used to sit in the courtyard and I would teach her the names of things, as if she were a small child. And just like a small child, she learned eagerly, quickly. Her mind wasn't rusted shut to new ideas. She wasn't like the Jesus Worshippers, whose tongues were creaky old wheels following the same grooves. She had an unusual memory, extraordinarily good. Whatever I said, it went in her ear then out her mouth.
I taught her to point to and call out the five elements that make up the physical world: metal, wood, water, fire, earth.
I taught her what makes the world a living place: sunrise and sunset, heat and cold, dust and heat, dust and wind, dust and rain.
I taught her what is worth listening to in this world: wind, thunder, horses galloping in the dust, pebbles falling in water. I taught her what is frightening to hear: fast footsteps at night, soft cloth slowly ripping, dogs barking, the silence of crickets.
I taught her how two things mixed together produce another: water and dirt make mud, heat and water make tea, foreigners and opium make trouble.
I taught her the five tastes that give us the memories of life: sweet, sour, bitter, pungent, and salty.
One day, Miss Banner touched her palm on the front of her body and asked me how to say this in Chinese. After I told her, she said to me in Chinese: "Miss Moo, I wish to know many words for talking about my breasts!" And only then did I realize she wanted to talk about the feelings in her heart. The next day, I took her wandering around the city. We saw people arguing. Anger, I said. We saw a woman placing food on an altar. Respect, I said. We saw a thief with his head locked in a wooden yoke. Shame, I said. We saw a young girl sitting by the river, throwing an old net with holes into the shallow part of the water. Hope, I said.
Later, Miss Banner pointed to a man trying to squeeze a barrel that was too large through a doorway that was too small. "Hope," Miss Banner said. But to me, this was not hope, this was stupidity, rice for brains. And I wondered what Miss Banner had been seeing when I was naming those other feelings for her. I wondered whether foreigners had feelings that were entirely different from those of Chinese people. Did they think all our hopes were stupid?
In time, however, I taught Miss Banner to see the world almost exactly like a Chinese person. Of cicadas, she would say they looked like dead leaves fluttering, felt like paper crackling, sounded like fire roaring, smelled like dust rising, and tasted like the devil frying in oil. She hated them, decided they had no purpose in this world. You see, in five ways she could sense the world like a Chinese person. But it was always this sixth way, her American sense of importance, that later caused troubles between us. Because her senses led to opinions, and her opinions led to conclusions, and sometimes they were different from mine.
For most of my childhood, I had to struggle _not_ to see the world the way Kwan described it. Like her talk about ghosts. After she had the shock treatments, I told her she had to pretend she didn't see ghosts, otherwise the doctors wouldn't let her out of the hospital.
"Ah, keep secret," she said, nodding. "Just you me know."
When she came home, I then had to pretend the ghosts _were_ there, as part of our secret of pretending they weren't. I tried so hard to hold these two contradictory views that soon I started to see what I wasn't supposed to. How could I not? Most kids, _without_ sisters like Kwan, imagine that ghosts are lurking beneath their beds, ready to grab their feet. Kwan's ghosts, on the other hand, sat _on_ the bed, propped against her headboard. I saw them.
I'm not talking about filmy white sheets that howled "Oooooohh." Her ghosts weren't invisible like the affable TV apparitions in _Topper_ who moved pens and cups through the air. Her ghosts looked alive. They chatted about the good old days. They worried and complained. I even saw one scratching our dog's neck, and Captain thumped his leg and wagged his tail. Apart from Kwan, I never told anyone what I saw. I thought I'd be sent to the hospital for shock treatments. What I saw seemed so real, not at all like dreaming. It was as though someone _else's_ feelings had escaped, and my eyes had become the movie projector beaming them into life.
I remember a particular day—I must have been eight—when I was sitting alone on my bed, dressing my Barbie doll in her best clothes. I heard a girl's voice say: _"Gei wo kan."_ I looked up, and there on Kwan's bed was a somber Chinese girl around my age, demanding to see my doll. I wasn't scared. That was the other thing about seeing ghosts: I always felt perfectly calm, as if my whole body had been soaked in a mild tranquilizer. I politely asked this little girl in Chinese who she was. And she said, _"Lili-lili, lili-lili,"_ in a high squeal.
When I threw my Barbie doll onto Kwan's bed, this _lili-lili_ girl picked it up. She took off Barbie's pink feather boa, peered under the matching satin sheath dress. She violently twisted the arms and legs. "Don't break her," I warned. The whole time I could feel her curiosity, her wonder, her fear that the doll was dead. Yet I never questioned why we had this emotional symbiosis. I was too worried that she'd take Barbie home with her. I said, "That's enough. Give her back." And this little girl pretended she didn't hear me. So I went over and yanked the doll out of her hands, then returned to my bed.
Right away I noticed the feather boa was missing. "Give it back!" I shouted. But the girl was gone, which alarmed me, because only then did my normal senses return, and I knew she was a ghost. I searched for the feather boa—under the covers, between the mattress and the wall, beneath both twin beds. I couldn't believe that a ghost could take something real and make it disappear. I hunted all week for that feather boa, combing through every drawer, pocket, and corner. I never found it. I decided that the girl ghost really had stolen it.
Now I can think of more logical explanations. Maybe Captain took it and buried it in the backyard. Or my mom sucked it up into the vacuum cleaner. It was probably something like that. But when I was a kid, I didn't have strong enough boundaries between imagination and reality. Kwan saw what she believed. I saw what I _didn't_ want to believe.
When I was a little older, Kwan's ghosts went the way of other childish beliefs, like Santa Claus, the Tooth Fairy, the Easter Bunny. I did not tell Kwan that. What if she went over the edge again? Privately I replaced her notions of ghosts and the World of Yin with Vatican-endorsed saints and a hereafter that ran on the merit system. I gladly subscribed to the concept of collecting goody points, like those S&H green stamps that could be pasted into booklets and redeemed for toasters and scales. Only instead of getting appliances, you received a one-way ticket to heaven, hell, or purgatory, depending on how many good and bad deeds you'd done and what other people said about you. Once you made it to heaven, though, you didn't come back to earth as a ghost, unless you were a saint. This would probably not be the case with me.
I once asked my mom what heaven was, and she said it was a permanent vacation spot, where all humans were now equal—kings, queens, hoboes, teachers, little kids. "Movie stars?" I asked. Mom said I could meet all kinds of people, as long as they had been nice enough to get into heaven. At night, while Kwan rattled on with her Chinese ghosts, I would list on my fingers the people I wanted to meet, trying to put them in some sort of order of preference, if I was limited to meeting, say, five a week. There was God, Jesus, and Mary—I knew I was supposed to mention them first. And then I'd ask for my father and any other close family members who might have passed on—although not Daddy Bob. I'd wait a hundred years before I put him on my dance card. So that took care of the first week, sort of boring but necessary. The next week was when the good stuff would begin. I'd meet famous people, if they were already dead—the Beatles, Hayley Mills, Shirley Temple, Dwayne Hickman—and maybe Art Linkletter, the creep, who'd finally realize why he should have had me on his dumb show.
By junior high, my version of the afterlife was a bit more somber. I pictured it as a place of infinite knowledge, where all things would be revealed—sort of like our downtown library, only bigger, where pious voices enumerating what thou shall and shall not do echoed through loudspeakers. Also, if you were slightly but not hopelessly bad, you didn't go to hell, but you had to pay a huge fine. Or maybe if you did something worse, you went to a place similar to continuation school, which was where all the bad kids ended up, the ones who smoked, ran away from home, shoplifted, or had babies out of wedlock. But if you had followed the rules, and didn't wind up a burden on society, you could advance right away to heaven. And there you'd learn the answers to all the stuff your catechism teachers kept asking you, like:
What should we learn as human beings?
Why should we help others less fortunate than ourselves?
How can we prevent wars?
I also figured I'd learn what happened to certain things that were lost, such as Barbie's feather boa and, more recently, my rhinestone necklace, which I suspected my brother Tommy had filched, even though he said, "I didn't take it, swear to God." What's more, I wanted to look up the answers to a few unsolved mysteries, like: Did Lizzie Borden kill her parents? Who was the Man in the Iron Mask? What really happened to Amelia Earhart? And out of all the people on death row who had been executed, who was actually guilty and who was innocent? For that matter, which felt worst, being hanged, gassed, or electrocuted? In between all these questions, I'd find the proof that it was my father who told the truth about how Kwan's mother died, not Kwan.
By the time I went to college, I didn't believe in heaven and hell anymore, none of those metaphors for reward and punishment based on absolute good and evil. I had met Simon by then. He and I would get stoned with our friends and talk about the afterlife: "It just doesn't make sense, man—I mean, you live for less than a hundred years, then everything's added up and, boom, you go on for billions of years after that, either lying on the proverbial beach or roasting on a spit like a hot dog." And we couldn't buy the logic that Jesus was the only way. That meant that Buddhists and Hindus and Jews and Africans who had never even heard of Christ Almighty were doomed to hell, while Ku Klux Klan members were not. Between tokes, we'd speak while trying not to exhale: "Wow, what's the point in that kind of justice? Like, what does the universe learn after that?"
Most of our friends believed there was nothing after death—lights out, no pain, no reward, no punishment. One guy, Dave, said immortality lasted only as long as people remembered you. Plato, Confucius, Buddha, Jesus—they were immortal, he said. He said this after Simon and I attended a memorial service for a friend, Eric, whose number came up in the draft and who was killed in Vietnam.
"Even if they weren't really the way they're now remembered?" Simon asked.
Dave paused, then said, "Yeah."
"What about Eric?" I asked. "If people remember Hitler longer than Eric, does that mean Hitler is immortal but Eric isn't?"
Dave paused again. But before he could answer, Simon said firmly, "Eric was great. Nobody will ever forget Eric. And if there's a paradise, that's where he is right now." I remember I loved Simon for saying that. Because that's what I felt too.
How did those feelings disappear? Did they vanish like the feather boa, disappear when I wasn't looking? Should I have tried harder to find them again?
It's not just grudges that I hang on to. I remember a girl on my bed. I remember Eric. I remember the power of inviolable love. In my memory, I still have a place where I keep all those ghosts.
## [4
THE GHOST MERCHANT'S HOUSE](TheHundredSecretSenses-toc.html#TOC-5)
My mother has another new boyfriend, Jaime Jofré. I don't have to meet him to know he'll have charm, dark hair, and a green card. He'll speak with an accent and my mother will later ask me, "Isn't he passionate?" To her, words are more ardent if a man must struggle to find them, if he says _"amor"_ with a trill rather than ordinary "love."
Romantic though she is, my mother is a practical woman. She wants proof of love: Give and you should receive. A bouquet, ballroom dancing lessons, a promise of eternal fidelity—it must be up to the man to decide. And there's also Louise's corollary of sacrificial love: Give up smoking for him and receive a week at a health spa. She prefers the Calistoga Mud Baths or the Sonoma Mission Inn. She thinks men who understand this kind of exchange are from emerging nations—she would never say "the third world." A colony under foreign dictatorship is excellent. When emerging nation isn't available, she'll settle for Ireland, India, Iran. She firmly believes that men who have suffered from oppression and a black-market economy know there's more at stake. They try harder to win you over. They're willing to deal. Through these guiding thoughts, my mother has found true love as many times as she's quit smoking for good.
Hell yes, I'm furious with my mother. This morning she asked if she could drop by to cheer me up. And then she spent two hours comparing my failed marriage with hers to Bob. A lack of commitment, an unwillingness to make sacrifices, no give, all take—those are the common faults she's noticed in Simon and Bob. And she and I both "gave, gave, gave from the bottom of our hearts." She bummed a cigarette from me, then a match.
"I saw it coming," she said, and inhaled deeply. "Ten years ago. Remember that time Simon went to Hawaii and left you home when you had the flu?"
"I told him to go. We had nonrefundable airline tickets and he could sell only one." Why was I defending him?
"You were sick. He should have been giving you chicken soup rather than cavorting on the beach."
"He was cavorting with his grandmother. She'd had a stroke." I was starting to sound as whiny as a kid.
She gave me a sympathetic smile. "Sweetie, you don't have to be in denial anymore. I know what you're feeling. I'm your mother, remember?" She stubbed out her cigarette before assuming her matter-of-fact, social worker manner: "Simon didn't love you enough, because _he_ was lacking, not you. You are abundantly lovable. There is _nothing wrong with you._ "
I gave a stiff nod. "Mom, I really should get to work now."
"You go right ahead. I'll just have another cup of coffee." She looked at her watch and said, "The exterminators flea-bombed my apartment at ten. Just to be safe, I'd like to wait another hour before I go back."
And now I'm sitting at my desk, unable to work, completely drained. What the hell does she know about my capacity for love? Does she have any idea how many times she's hurt me without knowing it? She complains that all that time she spent with Bob was a big waste. What about me? What about the time she didn't spend with me? Wasn't that a waste too? And why am I now devoting any energy to thinking about this? I've been reduced to a snivelly little kid again. There I am, twelve years old, facedown on my twin bed, a corner of the pillow stuffed into my mouth so that Kwan can't hear my mangled sobs.
"Libby-ah," Kwan whispers, "something matter? You sick? Eat too much Christmas cookie? Next time I don't make so sweet. . . . Libby-ah, you like my present? You don't like, tell me, okay? I make you another sweater. You tell me what color. Knit it take me only one week. I finish, wrap up, like surprise all over again. . . . Libby-ah? I think Daddy Mommy come back from Yosemite Park bring you beautiful present, pictures too. Pretty snow, mountaintop . . . Don't cry! No! No! You not mean this. How you can _hate_ you own mother? . . . Oh? Daddy Bob too? _Ah, zemma zaogao. . . ."_
Libby-ah, Libby-ah? Can I turn on the light? I want to show you something. . . .
Okay, okay! Don't get mad! I'm sorry. I'm turning it off. See? It's dark again. Go back to sleep. . . . I was going to show you the pen that fell out of Daddy Bob's trouser pocket. . . . You tilt it one way, you see a lady in a blue dress. You tilt it the other way, wah!—the dress falls down. I'm not lying. See for yourself. I'll turn on the light. Are you ready? . . . Oh, Libby-ah, your eyes are swollen big as plums! Put the wet towel back over them. Tomorrow they won't itch as much. . . . The pen? I saw it sneaking out of his pocket when we were at Sunday mass. He didn't notice because he was pretending to pray. I know it was just pretend, mm-hmm, because his head went this way _—booomp!—_ and he was snoring. _Nnnnnnnhhh!_ It's true! I gave him a little push. He didn't wake up, but his nose stopped making those sounds. Ah, you think that's funny? Then why are you laughing?
So anyway, after a while I looked at the Christmas flowers, the candles, the colored glass. I watched the priest waving the smoky lantern. Suddenly I saw Jesus walking through the smoke! Yes, Jesus! I thought he had come to blow out his birthday candles. I told myself, Finally I can see him—now I am a Catholic! Oh, I was so excited. That's why Daddy Bob woke up and pushed me down.
I kept smiling at Jesus, but then I realized—ah?—that man was not Jesus but my old friend Lao Lu! He was pointing and laughing at me. "Fooled you," he said. "I'm not Jesus! Hey, you think he has a bald head like mine?" Lao Lu walked over to me. He waved his hand in front of Daddy Bob. Nothing happened. He touched his little finger light as a fly on Daddy Bob's forehead. Daddy Bob slapped himself. He slowly pulled the nasty pen from Daddy Bob's pocket and rolled it into a fold of my skirt.
"Hey," Lao Lu said. "Why are you still going to a foreigners' church? You think a callus on your butt will help you see Jesus?"
Don't laugh, Libby-ah. What Lao Lu said was not polite. I think he was remembering our last lifetime together, when he and I had to sit on the hard bench for two hours every Sunday. Every Sunday! Miss Banner too. We went to church for so many years and never saw God or Jesus, not Mary either, although back then it was not so important to see her. In those days, she was also mother to baby Jesus but only concubine to his father. Now everything is Mary this and that! _—Old St. Mary's, Mary's Help, Mary Mother of God, forgiving me my sins._ I'm glad she got a promotion. But as I said, in those days, the Jesus Worshippers did not talk about her so much. So I had to worry only about seeing God and Jesus. Every Sunday, the Jesus Worshippers asked me, "Do you believe?" I had to say not yet. I wanted to say yes to be polite. But then I would have been lying, and when I died maybe they would come after me and make me pay two kinds of penalty to the foreign devil, one for not believing, another for pretending that I did. I thought I couldn't see Jesus because I had Chinese eyes. Later I found out that Miss Banner never saw God or Jesus either. She told me she wasn't a religious kind of person.
I said, "Why is that, Miss Banner?"
And she said, "I prayed to God to save my brothers. I prayed for him to spare my mother. I prayed that my father would come back to me. Religion teaches you that faith takes care of hope. All my hopes are gone, so why do I need faith anymore?"
"Ai!" I said. "This is too sad! You have no hopes?"
"Very few," she answered. "And none that are worth a prayer."
"What about your sweetheart?"
She sighed. "I've decided he's not worth a prayer either. He deserted me, you know. I wrote letters to an American navy officer in Shanghai. My sweetheart's been there. He's been in Canton. He's even been in Guilin. He knows where I am. So why hasn't he come?"
I was sad to hear that. At the time, I didn't know her sweetheart was General Cape. "I still have many hopes of finding my family again," I said. "Maybe I should become a Jesus Worshipper."
"To be a true worshipper," she said, "you must give your whole body to Jesus."
"How much do you give?"
She held up her thumb. I was astonished, because every Sunday she preached the sermon. I thought this should be worth two legs at least. Of course, she had no choice about preaching. No one understood the other foreigners, and they couldn't understand us. Their Chinese was so bad it sounded just like their English. Miss Banner had to serve as Pastor Amen's go-between. Pastor Amen didn't ask. He said she must do this, otherwise no room for her in the Ghost Merchant's House.
So every Sunday morning, she and Pastor stood by the doorway to the church. He would cry in English, "Welcome, welcome!" Miss Banner would translate into Chinese: "Hurry-come into God's House! Eat rice after the meeting!" God's House was actually the Ghost Merchant's family temple. It belonged to his dead ancestors and their gods. Lao Lu thought the foreigners showed very bad manners picking this place for God's House. "Like a slap in the face," he said. "The God of War will drop horse manure from the sky, you wait and see." Lao Lu was that way—you make him mad, he'll pay you back.
The missionaries always walked in first, Miss Banner second, then Lao Lu and I, as well as the other Chinese people who worked in the Ghost Merchant's House—the cook, the two maids, the stableman, the carpenter, I forget who else. The visitors entered God's House last. They were mostly beggars, a few Hakka God Worshippers, also an old woman who pressed her hands together and bowed three times to the altar, even though she was told over and over again not to do that anymore. The newcomers sat on the back benches—I'm guessing this was in case the Ghost Merchant came back and they needed to run away. Lao Lu and I had to sit up front with the missionaries, shouting "Amen!" whenever the pastor raised his eyebrows. That's why we called him Pastor Amen—also because his name sounded like "Amen," Hammond or Halliman, something like that.
As soon as we flattened our bottoms on those benches, we were not supposed to move. Mrs. Amen often jumped up, but only to wag her finger at those who made too much noise. That's how we learned what was forbidden. No scratching your head for lice. No blowing your nose into your palm. No saying "Shit" when clouds of mosquitoes sang in your ear—Lao Lu said that whenever anything disturbed his sleep.
That was another rule: No sleeping except when Pastor Amen prayed to God, long, boring prayers that made Lao Lu very happy. Because when the Jesus Worshippers closed their eyes, he could do the same and take a long nap. I kept my eye open. I would stare at Pastor Amen to see if God or Jesus was coming down from the heavens. I had seen this happen to a God Worshipper at a temple fair. God entered an ordinary man's body and threw him to the ground. When he stood up again, he had great powers. Swords thrust against his stomach bent in half. But no such thing ever happened to Pastor Amen. Although one time when Pastor was praying, I saw a beggar standing at the door. I remembered that the Chinese gods sometimes did this, came disguised as beggars to see what was going on, who was being loyal, who was paying them respect. I wondered if the beggar was a god, now angry to see foreigners standing at the altar where he used to be. When I looked back a few minutes later, the beggar had disappeared. So who knows if he was the reason for the disasters that came five years later.
At the end of the prayer time, the sermon would begin. The first Sunday, Pastor Amen spoke for five minutes—talk, talk, talk!—a lot of sounds that only the other missionaries could understand. Then Miss Banner translated for five minutes. Warnings about the devil. Amen! Rules for going to heaven. Amen! Bring your friends with you. Amen! Back and forth they went, as if they were arguing. So boring! For two hours, we had to sit still, letting our bottoms and our brains grow numb.
At the end of the sermon, there was a little show, using the music box that belonged to Miss Banner. Everyone liked this part very much. The singing was not so good, but when the music started, we knew our suffering was almost at an end. Pastor Amen lifted both hands and told us to rise. Mrs. Amen walked to the front of the room. So did the nervous missionary named Lasher, like _laoshu,_ "mouse," so that was what we called her, Miss Mouse. There was also a foreign doctor named Swan, which sounded like _suan-le,_ "too late"—no wonder sick people were scared to see him. Dr. Too Late was in charge of opening Miss Banner's music box and winding it with a key. When the music started, the three of them sang. Mrs. Amen had tears pouring from her eyes. Some of the old country people asked out loud if the box contained tiny foreigners.
Miss Banner once told me the music box was a gift from her father, the only memory of her family that she had left. Inside, she kept a little album for writing down her thoughts. The music, she said, was actually a German song about drinking beer, dancing, and kissing pretty girls. But Mrs. Amen had written new words, which I heard a hundred times but only as sounds: "We're marching with Jesus on two willing feet, when Death turns the corner, our Lord we shall meet." Something like that. You see, I remember that old song, but this time the words have new meaning. Anyway, that was the song we heard every week, telling everyone to go outside to eat a bowl of rice, a gift from Jesus. We had many beggars who thought Jesus was a landlord with many rice fields.
The second Sunday, Pastor Amen spoke for five minutes, Miss Banner for three. Then Pastor for another five minutes, Miss Banner for one. Everything became shorter and shorter on the Chinese side, and the flies drank from our sweat for only one and a half hours that Sunday. The week after that it was only one hour. Later, Pastor Amen had a long talk with Miss Banner. The following week, Pastor Amen spoke for five minutes, Miss Banner spoke the same amount. Again Pastor spoke for five minutes, Miss Banner the same amount. But now she didn't talk about rules for going to heaven. She was saying, "Once upon a time, in a kingdom far away, there lived a giant and the filial daughter of a poor carpenter who was really a king. . . ." At the end of each five minutes, she would stop at a very exciting part and say something like: "Now I must let Pastor speak for five minutes. But while you wait, ask yourself, Did the tiny princess die, or did she save the giant?" After the sermon and story were over, she told people to shout "Amen" if they were ready to eat their free bowl of rice. Ah, big shouts!
Those Sunday sermons became very popular. Many beggars came to hear Miss Banner's stories from her childhood. The Jesus Worshippers were happy. The rice-eaters were happy. Miss Banner was happy. I was the only one who worried. What if Pastor Amen learned what she was doing? Would he beat her? Would the God Worshippers pour coals over my body for teaching a foreigner to have a disobedient Chinese tongue? Would Pastor Amen lose face and have to hang himself? Would the people who came for rice and stories and not Jesus go to a foreigners' hell?
When I told Miss Banner my worries, she laughed and said no such thing would happen. I asked her how she knew this. She said, "If everyone is happy, what harm can follow?" I remembered what the man who returned to Thistle Mountain had said: "Too much happiness always overflows into tears of sorrow."
WE HAD five years of happiness. Miss Banner and I became great and loyal friends. The other missionaries remained strangers to me. But from seeing little changes every day, I knew their secrets very well. Lao Lu told me about shameful things he saw from outside their windows, also strange things he saw when he was inside their rooms. How Miss Mouse cried over a locket holding a dead person's hair. How Dr. Too Late ate opium pills for his stomachache. How Mrs. Amen hid pieces of Communion bread in her drawer, never eating it, just saving it for the end of the world. How Pastor Amen reported to America that he had made one hundred converts when really it was only one.
In return, I told Lao Lu some of the secrets I had seen myself. That Miss Mouse had feelings for Dr. Too Late, but he didn't notice. That Dr. Too Late had strong feelings for Miss Banner, and she pretended not to notice. But I did not tell him that Miss Banner still had great feelings for her number-three sweetheart, a man named Waren. Only I knew this.
For five years, everything was the same, except for these small changes. That was our life back then, a little hope, a little change, a little secret.
And yes, I had my secrets too. My first secret was this. One night, I dreamed I saw Jesus, a foreign man with long hair, long beard, many followers. I told Miss Banner, except I forgot to mention the part about the dream. So she told Pastor Amen, and he put me down for a hundred converts—that's why I knew it was only one. I didn't tell Miss Banner to correct him. Then he would have been more ashamed that his hundred converts was not even one.
My second secret was much worse.
This happened soon after Miss Banner told me she had lost her family and her hopes. I said I had so much hope I could use my leftovers to wish her sweetheart would change his mind and return. This pleased her very much. So that's what I prayed for, for at least one hundred days.
One evening, I was sitting on a stool in Miss Banner's room. We were talking, talking, talking. When we ran out of the usual complaints, I asked if we could play the music box. Yes, yes, she said. I opened the box. No key. It's in the drawer, she said. Ah! What's this? I picked up an ivory carving and held it to my eye. It was in the shape of a naked lady. Very unusual. I remembered seeing something like it once. I asked her where the little statue had come from.
"It belonged to my sweetheart," she said. "The handle of his walking stick. When it broke off, he gave it to me as a remembrance."
Wah! That's when I knew Miss Banner's sweetheart was the traitor, General Cape. All this time, I had been praying for him to come back. Just thinking about it shriveled my scalp.
So that was my second secret: that I knew who he was. And the third was this: I started praying he would stay away.
Let me tell you, Libby-ah, I didn't know how much she hungered for love, any kind. Sweet love didn't last, and it was too hard to find. But rotten love!—there was plenty to fill the hollow. So that's what she grew accustomed to, that's what she took as soon as it came back.
## [5
LAUNDRY DAY](TheHundredSecretSenses-toc.html#TOC-6)
Just like clockwork, the phone rings at eight. That makes it the third morning in a row Kwan's called at the exact moment that I'm buttering toast. Before I can say hello, she blurts out: "Libby-ah, ask Simon— name of stereo fix-it store, what is?"
"What's wrong with your stereo?"
"Wrong? Ahhhh . . . too much noise. Yes-yes, I play radio, it go _ccccchhhhhhhssss._ "
"Did you try adjusting the frequency?"
"Yes-yes! I often adjust."
"How about standing back from the stereo? Maybe you're conducting a lot of static today. It's supposed to rain."
"Okay-okay, maybe try that first. But just case, you call Simon, ask him store name."
I'm in a good mood. I want to see how far she'll carry her ruse. "I know the store," I say, and search for a likely-sounding name. "Yeah, it's Bogus Boomboxes. On Market Street." I can practically hear Kwan's mind whirring and clicking into alternate mode.
Finally she laughs and says, "Hey, you bad girl—lie! No such name."
"And no such stereo problem," I add.
"Okay-okay. You call Simon, tell him Kwan say Happy Birthday."
"Actually, I was going to call him for the same reason."
"Oh, you so bad! Why you torture me, embarrass this way!" She lets out a wheezy laugh, then gasps and says, "Oh, and Libby-ah, after call Simon, call Ma."
"Why? Is her stereo broken too?"
"Don't joke. Her heart feel bad."
I'm alarmed. "What's wrong? Is it serious?"
"Mm-hmm. So sad. You remember new boyfriend she have, I May Hopfree?"
_"High-_ may ho _-fray,"_ I pronounce slowly. "Jaime Jofré."
"I always remember, I May Hopfree. And that's what he do! Turn out he married already. Chile lady. She show up, pinch his ear, take home."
"No!" A ripple of glee flows into my cheeks, and I mentally slap myself.
"Yes-yes, Ma so mad! Last week she buy two loveboat cruise ticket. Hopfree say use your Visa, I pay you back. Now no pay, no cruise, no refund. Ah! Poor Ma, always find wrong man. . . . Hey, maybe I do matchmake for her. I choose better for her than she choose herself. I make good match, bring me luck."
"What if it's not so good?"
"Then I must fix, make better. My duty."
After we hang up, I think about Kwan's duty. No wonder she sees my impending divorce as a personal and professional failure on her part. She still believes she was our spiritual _mei-po,_ our cosmic matchmaker. And I'm hardly in the position to tell her that she wasn't. I was the one who asked her to convince Simon we were destined to be together, linked by the necessity of fate.
SIMON BISHOP AND I met more than seventeen years ago. At that moment in our lives, we were willing to place all our hopes on the ridiculous—pyramid power, Brazilian _figa_ charms, even the advice of Kwan and her ghosts. We both were terribly in love, I with Simon, he with someone else. The someone else happened to have died before I ever met Simon, although I didn't know that until three months later.
I spotted Simon in a linguistics class at UC Berkeley, spring quarter 1976. I noticed him right away because like me he had a name that didn't fit with his Asian features. Eurasian students weren't as common then as they are now, and as I stared at him, I had the sense I was seeing my male doppelgänger. I started wondering how genes interact, why one set of racial characteristics dominates in one person and not in another with the same background. I once met a girl whose last name was Chan. She was blond-haired and blue-eyed, and no, she wearily explained, she was not adopted. Her father was Chinese. I figured that her father's ancestors had engaged in secret dalliances with the British or Portuguese in Hong Kong. I was like that girl, always having to explain about my last name, why I didn't look like a Laguni. My brothers look almost as Italian as their last name implies. Their faces are more angular than mine. Their hair has a slight curl and is a lighter shade of brown.
Simon didn't look like any particular race. He was a perfectly balanced blend, half Hawaiian-Chinese, half Anglo, a fusion of different racial genes and not a dilution. When our linguistics class formed study groups, Simon and I drifted toward the same one. We didn't mention what we so obviously shared.
I remember the first time he brought up his girlfriend, because I had been hoping he didn't have one. Five of us were cramming for a midterm. I was listing the attributes of Etruscan: a dead language, as well as an isolate, unrelated to other languages . . . In the middle of my summary, Simon blurted: "My girlfriend, Elza, she went on a study tour of Italy and saw these incredible Etruscan tombs."
We looked at him—like, So? Mind you, Simon didn't say, "My girlfriend, who, by the way, is as dead as this language." He talked about her in passing, as if she were alive and well, traveling on Eurail and sending postcards from Tuscany. After a few seconds of awkward silence, he looked sheepish and mumbled the way people do when they're caught arguing with themselves while walking down the sidewalk. Poor guy, I thought, and at that moment my heartstrings went _twing._
After class, Simon and I would often take turns buying each other coffee at the Bear's Lair. There we added to the drone of hundreds of other life-changing conversations and epiphanies. We discussed primitivism as a Western-biased concept. Mongrelization as the only long-term answer to racism. Irony, satire, and parody as the deepest forms of truth. He told me he wanted to create his own philosophy, one that would guide his life's work, that would enable him to make _substantive_ changes in the world. I looked up the word _substantive_ in the dictionary that night, then realized I wanted a substantive life too. When I was with him, I felt as if a secret and better part of myself had finally been unleashed. I had dated other guys to whom I felt attracted, but those relationships seldom went beyond the usual good times induced by all-night parties, stoned conversations, and sometimes sex, all of which soon grew as stale as morning breath. With Simon, I laughed harder, thought more deeply, felt more passionately about life beyond my own cubbyhole. We could volley ideas back and forth like tennis pros. We wrestled with each other's minds. We unearthed each other's past with psychoanalytic gusto.
I thought it was eerie how much we had in common. Both of us had lost a parent before the age of five, he a mother, I a father. We both had owned pet turtles; his died after he accidentally dropped them into a chlorinated swimming pool. We both had been loners as kids, abandoned to caretakers—he to two unmarried sisters of his mother's, I to Kwan.
"My mom left me in the hands of someone who talked to ghosts!" I once told him.
"God! I'm amazed you aren't crazier than you already are." We laughed, and I felt giddy about our making fun of what had once caused me so much pain.
"Good ol' Mom," I added. "She's the quintessential social worker, totally obsessed with helping strangers and ignoring the homefront. She'd rather keep an appointment with her manicurist than lift a finger to help her kids. Talk about phony! It wasn't that she was pathological, but, you know—"
And Simon jumped in: "Yeah, even benign neglect can hurt for a lifetime." Which was _exactly_ what I was feeling but couldn't put into words. And then he clinched my heart: "Maybe her lack of attention is what made you as strong as you are today." I nodded eagerly as he went on: "I was thinking that, because my girlfriend—you know, Elza—well, she lost both parents when she was a baby. Talk about strong-willed— whew!"
That's how we were together, intimate in every way—up to a point. I sensed we were attracted to each other. From my end it was a strong sexual charge. From his it was more like static cling—which he easily shook off: "Hey, Laguni," he'd say, and put his hand firmly on my shoulder. "I'm bushed, gotta run. But if you want to go over notes this weekend, give me a call." With this breezy sendoff, I'd trudge back to my apartment, nothing to do on a Friday night, because I had turned down a date hoping that Simon would ask me out. By then I was stupid-in-love with Simon—goo-goo-eyed, giggly-voiced, floaty-headed, infatuated in the worst way. There were so many times when I lay in bed, disgusted that I was twitching with unspent desire. I wondered: Am I crazy? Am I the only one who's turned on? Sure, he has a girlfriend. So what? As everyone knows, when you're in college and changing your mind about a million things, a current girlfriend can turn into a former one overnight.
But Simon didn't seem to know that I was flirting with him. "You know what I like about you?" he asked me. "You treat me like a good buddy. We can talk about anything and we don't let the other thing get in the way."
"The other what?"
"The fact that we are . . . Well, you know, the opposite-sex thing."
"Really?" I said, faking astonishment. "You mean, I'm a girl but you're a— I had no idea!" And then we both broke into hearty guffaws.
At night I'd cry angrily, telling myself that I was a fool. I vowed many times to give up any hope of romance with Simon—as if it were possible to will myself _not_ to be in love! But at least I knew how to put on a good front. I continued to play the jovial good buddy, listening with a smile on my face and a cramp in my heart. I expected the worst. And sure enough, sooner or later, he would bring up Elza, as though he knew she was on my mind as well.
Through three months of masochistic listening I came to know the minutiae of her life: That she lived in Salt Lake City, where she and Simon had grown up, tussling with each other since the fifth grade. That she had a two-inch scar on the back of her left knee the shape and color of an earthworm, a mysterious legacy from infancy. That she was athletic; she kayaked, backpacked, and was an expert cross-country skier. That she was musically gifted, a budding composer, who had studied with Artur Balsam at a famous summer music camp in Blue Hill, Maine. She'd even written her own thematic variation on the Goldberg Variations. "Really?" I said to each praiseworthy thing he said about her. "That's amazing."
The strange thing is, he kept speaking about her in the present tense. Naturally I thought she was alive in the present time. Once, Simon pointed out I had smeared lipstick on my teeth, and as I hurriedly rubbed it off, he added, "Elza doesn't wear makeup, not even lipstick. She doesn't believe in it." I wanted to scream, What's there to believe!? You either wear makeup or you don't! By then I wanted to smack her, a girl so morally upright she had to be the most odious hominid ever to walk planet Earth, in her non–animal hide shoes. Even if Elza had been sweet and insipid, it wouldn't have mattered, I still would have despised her. To me, Elza didn't _deserve_ Simon. Why should she have him as one of her perks of life? She deserved an Olympic gold medal for Amazon discus-throwing. She deserved a Nobel Peace Prize for saving retarded baby whales. She deserved to play organ for the Mormon Tabernacle Choir.
Simon, on the other hand, deserved me, someone who could help him discover the recesses of his soul, the secret passageways that Elza had barricaded with constant criticism and disapproval. If I complimented Simon—told him what he had said was profound, for example— he'd say, "You think so? Elza says one of my biggest faults is going along with whatever's nice and easy, that I don't think things through hard enough."
"You can't believe everything that Elza says."
"Yeah, that's what she says too. She hates it when I just go along with what's been handed to me as truth. She believes in trusting your own intuitions, sort of like that guy who wrote _Walden,_ what's his name, Thoreau. Anyway, she thinks it's important for us to argue, to get to the marrow of what we believe and why."
"I hate to argue."
"I don't mean argue in the sense of a fight. More of a debate, like what you and I do."
I hated being compared and falling short. I tried to sound playful. "Oh? So what do _you two_ debate?"
"Like whether celebrities have a responsibility as symbols and not just as people. Remember when Muhammad Ali refused to be drafted?"
"Sure," I lied.
"Elza and I both thought he was great, taking a personal stand like that against the war. But then he wins back the heavyweight title and later President Ford invites him to the White House. Elza said, 'Can you believe it?' I said, 'Hell, if I were invited, I'd go to the White House too.' And she said, 'By a _Republican_ president? During an election year?' She wrote him a letter."
"The president?"
"No, Muhammad Ali."
"Oh, right. Of course."
"Elza says you can't just talk politics or watch it happen on television. You have to do something, otherwise you're part of it."
"Part of what?"
"You know, hypocrisy. It's the same as corruption."
I imagined Elza looking like Patty Hearst, wearing a beret and combat fatigues, an automatic rifle perched on her hip.
"She believes all people should take an active moral position on life. Otherwise the world's going to end in thirty years or less. A lot of our friends say she's a pessimist. But she thinks she's the real optimist, because she wants to _do_ something to change the world in a positive way. If you think about it, she's right."
While Simon grew more expansive about Elza's ridiculous opinions, I'd be dreamily analyzing his features, how chameleonlike they were. His face would change—from Hawaiian to Aztecan, Persian to Sioux, Bengali to Balinese.
"What kind of name is Bishop?" I asked one day.
"On my father's side, missionary eccentrics. I'm descended from _the_ Bishops—you know?—the family of Oahu Island fame. They went to Hawaii in the eighteen hundreds to convert lepers and heathens, then ended up marrying royalty and owning half the island."
"You're kidding."
"Unfortunately, I'm also from the side of the family that didn't inherit any of the wealth, not a single pineapple orchard or golf course. On my mother's side, we're Hawaiian-Chinese, with a couple of royal princesses swimming in the gene pool. But again, no direct access to beachfront property." And then he laughed. "Elza once said I inherited from the missionary side of my family the laziness of blind faith, and from my royal Hawaiian side a tendency to use others to take care of my needs rather than working to fulfill them myself."
"I don't think that's true, that stuff about inherited nature, as if we're destined to develop into a certain kind of person without choice. I mean, hasn't Elza ever heard of _determinism_?"
Simon looked stumped. "Hmmm," he said, thinking. For a moment, I felt the satisfaction of having vanquished a competitor with a subtle and deft move.
But then he remarked: "Doesn't the doctrine of determinism say that all events and even human choices follow natural laws, meaning it kind of goes along with what Elza was saying?"
"What I mean is," and I began to stammer as I tried to recall what I'd skimmed over in philosophy class. "I mean, how do we define _natural_? Who's to say what's natural and what's not?" I was flailing, trying to keep my pathetic self above water. "Besides, what's her background?"
"Her folks are Mormon, but they adopted her when she was a year old and named her Elsie, Elsie Marie Vandervort. She doesn't know who her biological parents were. But ever since she was six, before she knew how to read music, she could hear a song just once, then play it exactly, note for note. And she especially loved music by Chopin, Paderewski, Mendelssohn, Gershwin, Copland—I forget the others. Later she discovered every single one of them was either Polish or Jewish. Isn't that weird? So that made her think she was probably a Polish Jew. She started calling herself Elza instead of Elsie."
"I like Bach, Beethoven, and Schumann," I said smartly, "but that doesn't make me a German."
"It wasn't just that. When she was ten, something happened which will sound really bizarre, but I swear it's true, because I saw part of it. She was in the school library, flipping through an encyclopedia, and she saw a photo of some crying kid and his family being rounded up by soldiers. The caption said they were Jews being taken to Auschwitz. She didn't know where Auschwitz was or even that it was a concentration camp. But she literally smelled something horrible that made her shake and gag. And then she fell to her knees and started chanting: 'Osh-vee-en-shim, osh-vee-en-shim,' something like that. The librarian shook her, but Elza wouldn't stop—she couldn't. So the librarian dragged her to the school nurse, Mrs. Schneebaum. And Mrs. Schneebaum, who was Polish, heard Elza chanting 'Osh-vee-en-shim' and freaked. She thought Elza was saying this to make fun of her. Well, get this: It turned out 'Oswie¸cim' is the way you say 'Auschwitz' in Polish. After Elza came out of her trance, she knew her parents were Polish Jews who had survived Auschwitz."
"What do you mean, she knew?"
"She just _knew—_ like the way hawks know to hover on a stream of air, the way rabbits freeze with fear. It's knowledge that can't be taught. She said her mother's memories passed from heart to womb, and they're now indelibly printed on the walls of her brain."
"Come on!" I said dismissively. "She sounds like my sister Kwan."
"How so?"
"Oh, she just makes up any old theory to suit whatever she believes. Anyway, biological instinct and emotional memories aren't the same thing. Maybe Elza read or heard about Auschwitz before and didn't remember. You know how people see old photos or movies and later think they were personal memories. Or they have a déjà vu experience—and it's just a bad synapse feeding immediate sensory perception into long-term memory. I mean, does she even _look_ Polish or Jewish?" And right after I said that I had a dangerous thought. "You have a picture of her?" I asked as casually as possible.
While Simon dug out his wallet, I could feel my heart revving like a race car, about to confront my competition. I feared she would look devastatingly beautiful—a cross between Ingrid Bergman illuminated by airport runway lights and Lauren Bacall sulking in a smoke-filled bar.
The photo showed an outdoorsy girl, backlit by a dusk-hour glow, frizzy hair haloing a sullen face. Her nose was long, her chin childishly small, her lower lip curled out in mid-utterance, so that she looked like a bulldog. She was standing next to a camping tent, arms akimbo, hands perched on chunky hips. Her cutoff jeans were too tight, sharply creased at the crotch. There was also her ridiculous T-shirt, with its "Question Authority" in lumpy letters stretched over the mounds of her fatty breasts.
I thought to myself, Why, she isn't gorgeous. She isn't even button-nose cute. She's as plain as a Polish dog without mustard. I was trying to restrain a smile, but I could have danced the polka I was so happy. I knew that comparing myself with her that way was superficial and irrelevant. But I couldn't help feeling happily superior, believing I was prettier, taller, slimmer, more stylish. You didn't have to like Chopin or Paderewski to recognize that Elza was descended from Slavic peasant stock. The more I looked, the more I rejoiced. To finally see the demons of my insecurity, and they were no more threatening than her cherub-faced kneecaps.
What the hell did Simon see in her? I tried to be objective, look at her from a male point of view. She was athletic, there was that. And she certainly gave the impression of being smart, but in an intimidating, obnoxious way. Her breasts were far bigger than mine; they might be in her favor—if Simon was stupid enough to like fleshy globules that would someday sag to her navel. You might say that her eyes were interesting, slanted and catlike. Although on second glance, they were disturbing, smudged with dark hollows. She stared straight into the camera and her look was both penetrating and vacant. Her expression suggested that she knew the secrets of the past and future and they were all sad.
I concluded Simon had confused loyalty with love. After all, he had known Elza since childhood. In a way, you had to admire him for that. I handed the picture back to him, trying not to appear smug. "She seems _awfully_ serious. Is that something you inherit being a Polish Jew?"
Simon studied the photo. "She can be funny when she wants. She can do impersonations of anyone—gestures, speech patterns, foreign accents. She's hilarious. She can be. Sometimes. But." He paused, struggling. "But you're right. She broods a lot about how things can be better, why they should be, until she goes into a funk. She's always been that way, moody, serious, I guess you might even say depressed. I don't know where that comes from. Sometimes she can be so, you know, unreasonable," and he trailed off, seemingly troubled, as if he were now viewing her from a new light and her features were glaringly unattractive.
I hoarded these observational tidbits as weapons to use in the future. Unlike Elza, I would become a _true_ optimist. I would take action. In contrast to her lugubriousness, I would be buoyant. Instead of being a critical mirror, I would admire Simon's insights. I too would take active political stands. But I'd laugh often and show Simon that life with a spiritual soul mate didn't have to be all doom and gloom. I was determined to do whatever was necessary to unseat her from Simon's heart.
After seeing Elza's picture, I thought she would be easy to displace. Foolish me, I didn't know I would have to pry Simon from the clutches of a ghost. But that day, I was so happy I even accepted an invitation from Kwan to come to dinner. I brought my laundry, and just to be pleasant, I pretended to listen to her advice.
Libby-ah, let me do this. You don't know how to use my washing machine. Not too much soap, not too much hot, always turn the pockets inside out. . . .
Libby-ah, ai-ya, why do you have so many black clothes? You should wear pretty colors! Little flowers, polka dots, purple is a good color for you. White, I don't like. Not because of superstition. Some people think that white means death. No such thing. In the World of Yin, there are many, many colors you don't even know, because you can't see them with your eyes. You have to use your secret senses, imagine them when you are full of genuine feelings and memories, both happy and sad. Happy and sad sometimes come from the same thing, did you know this?
Anyway, white I don't like because it's too easy to get dirty, too hard to clean. It's not practical. I know, because in my last lifetime, I had to wash lots of white laundry—lots, lots, lots. That was one of the ways I earned my room in the Ghost Merchant's House.
On the First Day of each week I had to wash. On the Second Day, I ironed what I had washed. The Third Day was for polishing shoes and mending clothes. The Fourth Day was for sweeping the courtyard and passageways, the Fifth Day for mopping the floors and wiping the furniture in God's House. The Sixth Day was for important business.
I liked the Sixth Day the most. Together Miss Banner and I walked around the village, handing out pamphlets called "The Good News." Even though the paper contained English words turned into Chinese, I couldn't read them. Since I couldn't read, I couldn't teach Miss Banner to read. And in the poor parts of the village that we walked through, nobody knew how to read either. But people were glad to take those pamphlets. They used them to stuff inside their winter clothes. They put them over rice bowls to keep out flies. They pasted them over cracks in walls. Every few months, a boat from Canton came and brought more boxes of these pamphlets. So every week, on the Sixth Day, we had plenty to hand out. We didn't know that what we really were giving those people was plenty of future trouble.
When we returned to the Ghost Merchant's House, happy and empty-handed, Lao Lu would put on a little show for us. He would climb up a column, then walk quickly along the edge of the roof, while we gasped and cried, "Don't fall!" Then he would turn around and pick up a brick and place this on his head, then a teacup on top of that, then a bowl, a plate, all sorts of things of different sizes and weights. Again he would walk along that skinny edge, while we screamed and laughed. I think he was always trying to recover face from that time he fell into the water with Miss Banner and her trunk.
The Seventh Day, of course, was for going to God's House, then resting in the afternoon, talking in the courtyard, watching the sunset, the stars, or a lightning storm. Sometimes I plucked leaves from a bush that grew in the courtyard. Lao Lu always corrected me: "That's not a bush. It's a holy tree. See here." He would stand with his arms straight out, like a ghost walking in the night, claiming that the spirit of nature now flowed from the tree's limbs into his. "You eat the leaves," he said, "and you find peace, balance in yourself, piss on everyone else." So every Sunday, I used those leaves to make a tea, like a thank-you gift to Lao Lu for his show. Miss Banner always drank some too. Each week, I would say, "Hey, Lao Lu, you are right, the tea from this bush makes a person feel peaceful." Then he would say, "That's not just any dog-pissing bush, it's a holy tree." So you see, those leaves did nothing to cure him of cursing, too bad.
After the Seventh Day, it was the First Day all over again, the one I'm now going to talk about. And as I said, I had to wash the dirty clothes.
I did my washing in the large walled passageway just outside the kitchen. The passageway had a stone-paved floor and was open to the sky but shaded by a big tree. All morning long, I kept big pots of water and lime boiling, two pots because the missionaries didn't allow me to have men and ladies swimming together in the same hot water. One pot I scented with camphor, the other with cassia bark, which smells like cinnamon. Both were good for keeping away cloth-eating moths. In the camphor water, I boiled white shirts and the secret underclothes of Pastor Amen and Dr. Too Late. I boiled their bedding, the cloths they used to wipe their noses and brows. In the pot with cassia bark, I boiled the blouses, the secret underclothes of the ladies, their bedding, the cloths they used to wipe their lady noses.
I laid the wet clothes on the wheel of an old stone mill, then rolled the stone to squeeze out the water. I put the squeezed clothes into two baskets, men and ladies still separate. I poured the leftover cassia water over the kitchen floor. I poured the leftover camphor water over the passageway floor. And then I carried the baskets through the gateway, into the back area, where there were two sheds along the wall, one for a mule, one for a buffalo cow. Between these two sheds was a rope stretched very tight. And this is where I hung the laundry to dry.
On my left side was another wall, and a gateway that led into a large strolling garden, bounded by high stone walls. It was a beautiful place, once tamed by the hands of many gardeners, now neglected and wild. The stone bridges and ornamental rocks still stood, but the ponds underneath were dried up, no fish only weeds. Everything was tangled together—the flowering bushes, the branches of trees, weeds and vines. The pathways were thick with the leaves and blossoms of twenty seasons, so soft and cool on my feet. The paths rolled up and down in surprising ways, letting me dream I was climbing back up Thistle Mountain. The top of one of these hills was just big enough for a small pavilion. Inside the pavilion were stone benches covered with moss. In the middle of the stone floor was a burnt spot. From this pavilion, I could look over the wall, see the village, the limestone peaks, the archway going into the next mountain valley. Every week, after I washed the clothes, I soaked duck eggs in leftover lime and buried them in the garden to let them cure. And when I was done with that, I stood in the pavilion, pretending the world I saw beyond the wall was mine. I did this for several years, until one day Lao Lu saw me standing there. He said, "Ai, Nunumu, don't go up there anymore, that's where the Punti merchant died, in the pavilion."
Lao Lu said the merchant was standing there one evening, with his four wives down below. He gazed at the sky and saw a cloud of black birds. The merchant cursed them, then burst into flames. Wah! The fire roared, the merchant's fat hissed and spattered. Below, his terrified wives yowled, smelling the pungent odor of fried chili and garlic. All at once, the fire went out, and smoke in the shape of the merchant rose and blew away. When his wives crept up to the pavilion, they found no ash, only his feet and shoes remained. Also, the smell, terrible and delicious at the same time.
After Lao Lu told me that, I worried about that smell every time I hung the laundry, every time I went into the garden to bury my eggs. I smelled camphor, cassia, dead leaves, and flowering bushes. But the day that I'm now talking about, I thought I smelled the Ghost Merchant, his fear of death, very strong, chili and garlic, maybe a little vinegar too. It was a day of great heat, during the month when the cicadas unbury themselves after lying four years in the ground. They were singing, the males shrieking for females, each one trying to be the loudest. I kept my one eye aimed toward the gateway, just in case the Ghost Merchant was in there, looking for his feet. I heard a rustling sound, dry leaves crackling, twigs snapping, and black birds rushed up out of the bushes and scattered in the sky. The cicadas fell silent.
My bones were trembling. I wanted to run away. But I heard the Ghost Bandit Maiden inside me say, "Scared? How can you be scared of a Punti merchant with no feet? Go inside and see where he is." I was now both scared and ashamed to be scared. I carefully went to the gateway, peeked in. When the cicadas began to buzz, I ran into the garden, my feet crunching dead leaves. I darted onto the stone bridge, past the dry pond, over the hills rolling up and down. When the buzzing turned into clacks, I stopped, knowing the cicadas would soon exhaust themselves and fall quiet. Using their song, I ran and stopped, ran and stopped, until I was standing at the bottom of the hill big enough for a small pavilion. I circled its bottom when the clacking stopped, and stared at a man sitting on a stone bench, eating a tiny banana. I had never heard of a ghost eating a banana. Of course, since then, other ghosts have told me that they sometimes pretend to eat bananas, although never ones with lots of black bruises, which is what this man's banana had.
When the man saw me, he leapt to his feet. He had a peculiar but elegant face, not Chinese, not foreign. He wore gentleman's clothes. I had seen this man before, I was sure of it. Then I heard sounds coming from the other side of the hill, a loud stream of water splashing on rocks, a man sighing, feet crunching twenty seasons of leaves. I saw the flash of a silver-tipped walking stick, the hollowed face of the man who owned it. His hands were busy closing the many buttons of his trousers. This was General Cape, and the elegant man with the banana was the one-half man called Yiban.
Wah! Here was the man I had prayed would return to Miss Banner. I later prayed that he would stay away, but I must not have asked God that as many times.
Cape barked to Yiban, and then Yiban said to me, "Little Miss, this gentleman is a famous Yankee general. Is this the house where the God-worshipping foreigners live?"
I didn't answer. I was remembering what the man who came back to Thistle Mountain had said: that General Cape had turned traitor against the Hakkas. I saw General Cape looking at my shoes. He spoke again, and Yiban translated: "The lady who gave you those leather shoes is a great friend of the general. She is anxious to see him."
So the shoes with my feet inside led the two men to Miss Banner. And Yiban was right. She was anxious to see General Cape. She threw her arms around him and let him lift her in the air. She did this in front of Pastor Amen and Mrs. Amen, who although they were husband and wife never touched each other, not even in their own room—that's what Lao Lu told me. Late at night, when everyone was supposed to be asleep but was not, Miss Banner opened her door and General Cape quickly walked from his room into hers. Everyone heard this; we had no windows, only wooden screens.
I knew Miss Banner would call the general into her room. Earlier that evening, I had told her Cape was a traitor to Hakka people, that he would be a traitor to her as well. She became very angry with me, as if I were saying these things to curse her. She said General Cape was a hero, that he had left her in Canton only to help the God Worshippers. So then I told her what the man who returned to Thistle Mountain had said: that General Cape had married a Chinese banker's daughter for gold. She said my heart was rotten meat and my words were maggots feeding on gossip. She said if I believed these things about General Cape, then I would no longer be her loyal friend.
I said to her, "When you already believe something, how can you suddenly stop? When you are a loyal friend, how can you no longer be one?" She didn't answer.
Late at night, I heard the music box play, the one her father had given her when she was a young girl. I heard the music that made tears pour from Mrs. Amen's eyes, but now the music was making a man kiss a girl. I heard Miss Banner sigh, again and again. And her happiness was so great it spilled over, leaked into my room, and turned into tears of sorrow.
I've started doing my laundry at Kwan's house again. Simon used to take care of the wash—that was one of the nice things about being married to him. He liked to tidy up the house, snap fresh sheets and smooth them onto the bed. Since he left, I've had to wash my own clothes. The coin-op machines are in the basement of my building, and the mustiness and dim light give me the willies. The atmosphere preys on my imagination. But then, so does Kwan.
I always wait until I run out of clean underwear. And then I throw three bagfuls of laundry into the car and head for Balboa Street. Even now, as I stuff my clothes into Kwan's dryer, I think about that story she told me the day I was so hopeful with love. When she got to the part about joy turning into sorrow, I said, "Kwan, I don't want to hear this anymore."
"Ah? Why?"
"It bums me out. And right now, I want to stay in a good mood."
"Maybe I tell you more, don't become bum. You see mistake Miss Banner do—"
"Kwan," I said, "I don't want to hear about Miss Banner. _Ever._ "
What power! What relief! I was amazed how strong Simon made me feel. I could stand up to Kwan. I could decide whom I should listen to and why. I could be with someone like Simon, who was down-to-earth, logical, and sane.
I never thought that he too would fill my life with ghosts.
# II
## [6
FIREFLIES](TheHundredSecretSenses-toc.html#TOC-8)
The night Simon kissed me for the first time was when I finally learned the truth about Elza. The spring quarter had ended and we were walking in the hills behind the Berkeley campus, smoking a joint. It was a warm June night, and we came upon an area where tiny white lights were twinkling in the oak trees as if it were Christmas.
"Am I hallucinating?" I asked.
"Fireflies," Simon answered. "Aren't they amazing?"
"Are you sure? I don't think they exist in California. I've never seen them before."
"Maybe some student bred them for a work-study experiment and let them go."
We sat on the scabby trunk of a fallen tree. Two flickering bugs were zigzagging their way toward each other, their attraction looking haphazard yet predestined. They flashed on and off like airplanes headed for the same runway, closer and closer, until they sparked for an instant as one, then extinguished themselves and flitted darkly away.
"That's romance for you," I said.
Simon smiled and looked right at me. He awkwardly put his arm around my waist. Ten seconds passed, twenty seconds, and we hadn't moved. My face grew hot, my heart was beating fast, as I realized we were crossing the confines of friendship, about to leap over the fence and run for the wilds. And sure enough, our mouths, like those fireflies, bobbed and weaved toward each other. I closed my eyes when his lips reached mine, both of us trembly and tentative. Just as I pressed in closer to let him take me into a more passionate embrace, he released me, practically pushed me. He started talking in an apologetic tone.
"Oh God, I'm sorry. I really like you, Olivia. A lot. It's complicated, though, because of—well, you know."
I flicked a bug off the trunk, stared at it dumbly as it twirled on its back.
"You see, the last time I saw her, we had a terrible fight. She got very angry with me, and I haven't seen her since. That was six months ago. The thing is, I still love her. But—"
"Simon, you don't have to explain." I stood up on shaky legs. "Let's just forget it, okay?"
"Olivia, sit down. Please. I have to tell you. I want you to understand. This is important."
"Let go of me. Forget it, okay? Oh, shit! Just pretend it never happened!"
"Wait. Come back. Sit down, please sit down. Olivia, I have to tell you this."
"What the hell for?"
"Because I think I love you too."
I caught my breath. Of course, I would have preferred if he hadn't qualified his declaration with "I think" and "too," as if I could be part of an emotional harem. But infatuated as I was, "love" was enough to act as both balm and bait. I sat down.
"If you hear what happened," he said, "maybe you'll understand why it's taken me so long to tell you how I feel about you."
My heart was still pounding wildly with a strange mixture of anger and hope. We sat in nervous silence for a few minutes. When I was ready, I said in a cool voice, "Go ahead."
Simon cleared his throat. "This fight Elza and I had, it was in December, during the quarter break. I was back in Utah. We had planned to go cross-country skiing in Little Cottonwood Canyon. The week before, we'd been praying for new snow, and then it finally came in truckloads, three feet of fresh powder."
"She didn't want to go," I guessed, trying to hurry up the story.
"No, we went. So we were driving up the canyon, and I remember we were talking about the SLA and whether giving food to the poor made extortion and bank robbery less reprehensible. Out of the blue, Elza asked me, 'What do you feel about abortion?' And I thought I heard wrong. 'Extortion?' I said. And she said, 'No, abortion.' So I said, 'You know, like what we said before, about _Roe versus Wade,_ that the decision didn't go far enough.' She cut me off and said, 'But what do you _really_ feel about abortion?' "
"What did she mean, really feel?"
"That's what I asked. And she said slowly, enunciating every syllable: 'I mean emotionally, what do you feel?' And I said, 'Emotionally, I think it's fine.' Then she blew up: 'You didn't even think about the question! I'm not asking you about the weather. I'm asking you about the lives of human beings! I'm talking about the real life of a woman versus the potential life in her womb!' "
"She was hysterical." I was eager to emphasize Elza's volatile and unreasonable nature.
He nodded. "At the trailhead, she jumped out of the car, really pissed, threw on her skis. Just before she took off, she screamed, 'I'm pregnant, you idiot. And there's no way I'm having this baby and ruining my life. But it tears me up to abort it and you're just sitting there, smiling, saying it's fine.' "
"Omigod. Simon. How were you supposed to know?" So that was it, I thought: Elza had wanted to get married, and confronted with the prospect, Simon had refused. Good for him.
"I was stunned," Simon continued. "I had no idea. We were always careful about birth control."
"You think she slipped up on purpose?"
He frowned. "She's not that kind of person." He seemed defensive.
"What did you do?"
"I put on my skis, followed her tracks. I kept shouting for her to wait, but she went over a crest and I couldn't see her anymore. God, I remember how beautiful it was that day, sunny, peaceful. You know, you never think terrible things can happen when the weather's nice." He laughed bitterly.
I thought he was through—since that day he and Elza hadn't seen each other, end of story, time for the sequel, me. "Well," I said, trying to sound sympathetic, "the least she could have done was given you a chance to discuss the situation before jumping all over you."
Simon leaned forward and buried his face in his hands. "Oh God!" he said in an anguished voice.
"Simon, I understand, but it wasn't your fault, and now it's over."
"No, wait," he said hoarsely. "Let me finish." He stared at his knees, took a few deep breaths. "I got to this steep fire road, and there was an out-of-bounds sign. Just beyond that, she was sitting at the top of a ledge, hugging herself, crying. I called to her and she looked up, really pissed. She pushed off and headed down this steep wide-open bowl. I can still see it: The snow, it was incredible, pristine and bottomless. And she was gliding down, taking the fall line. But about halfway down, she hit some heavier snow, her skis sank, and she sagged to a stop."
I looked at Simon's eyes. They were fixed on something faraway and lost, and I became scared.
"I yelled her name as loud as I could. She was mashing her poles against the snow, trying to kick up the tips of her skis. I yelled again— 'Goddamnit, Elza!'—and I heard this sound, like a muffled gunshot, and then it was perfectly quiet again. She turned around. She was squinting—she must have been blinded by the sun. I don't think she saw it— the slope, two hundred yards above her. It was slowly tearing, no sound, like a giant zipper opening up. The seam became a crack, an icy blue shadow. And then it was snaking fast, straight across. The crack slipped down a little, and it was huge, glassy as an ice rink. Then everything began to rumble, the ground, my feet, my chest, my head. And Elza— I could tell she knew. She was struggling to get out of her skis."
Like Elza, I knew what was coming. "Simon, I don't think I want to hear any more of this—"
"She threw off her skis and her backpack. She was jumping through the snow, sinking to her hips. I started yelling, 'Go to the side!' And then the mountain collapsed and all I could hear was this train roar, trees snapping, whole stands of them, popping like toothpicks."
"Oh God," I whispered.
"She was swimming on top of the crud—that's what you're supposed to do, swim, swim, keep swimming. And then . . . she was swallowed up . . . gone. Everything creaked and settled, then grew absolutely still. I could smell pine pitch from the broken trees. My mind was going a million miles a minute. Don't panic, I told myself, if you panic it's all over. I skied down the side, between the trees where the snow was intact. I kept telling myself, Remember where she went under. Look for skis sticking up. Use one of your skis as a marker. Dig with your pole. Spread out in a widening circle.
"But when I got to the bottom, nothing looked the way it did from the top. The point I'd marked in my head, shit, it wasn't there, just this huge field of rubbly snow, heavy as wet cement. I was stumbling around, feeling like I was in one of those nightmares where your legs are paralyzed."
"Simon," I said, "you don't have to—"
But he kept talking: "All of a sudden this strange calm hit me, the eye of the hurricane. I could see Elza in my mind, where she was. We were so connected. She was guiding me with her thoughts. I shoved my way across to where I thought she was. I started to dig with one of my skis, telling her I'd have her out soon. And then I heard a helicopter. Thank God! I waved like mad, then two ski patrol guys were jumping out with a rescue dog and avalanche probes. I was so nuts I was saying how aerobically fit she was, what her heart rate was, how many miles she ran every week, where they should dig. But the ski patrol guys and the dog started going down the slope in a zigzag pattern. So I kept digging in the same area where I was sure she was. Pretty soon, I heard the dog yelping, and the guys shouting down below that they'd found her. That surprised me, that she wasn't where I thought she was. When I got down to where the ski patrol guys were, I saw they had her top half uncovered. I was shoving my way through, sweaty and out of breath, thanking them, telling them how great they were, because I could see she was okay. She was there, right there, all along she'd been only two feet under the surface. I was so damn happy to see she was alive."
"Oh, thank God," I whispered. "Simon, until you said that, you know what? I actually thought—"
"Her eyes were already open. But she was stuck, crouched on her side with her hands cupped together in front of her mouth, like this, which was what I'd taught her to do, to push out an air pocket so you can breathe longer. I was laughing and saying, 'God, Elza, I can't believe you were calm enough to remember that part about the air pocket.' Only the rescue guys were now pushing me back, saying, 'We're sorry, man, but she's gone.' And I said, 'What the fuck are you talking about? She's still there, I can see her, get her out.' And one of the guys put his arm on my shoulder and said, 'Hey, bud, we've been digging for an hour and the avalanche was reported an hour before that. The most she ever had was twenty, twenty-five minutes tops.'
"And I yelled back, 'It's been _ten minutes_!' I was so crazed—you know what I thought? That Elza told them to say that because she was still mad at me. I pushed past them. You see, I was going to tell her that I knew—I knew in my heart and my gut—how special life is, how hard it is to give it up, yours or anyone else's."
I put my hand on Simon's shoulder. He was sucking in air like an asthmatic. "When I reached her," he said, "I scraped out the snow stuck in her mouth. And, and, and—and that's when I realized she wasn't breathing, you know, she wasn't really breathing into that little air space I had taught her to make. And, and, and I saw how dark her face was, the tears frozen in her open eyes, you know, and I said, 'Elza, please, come on, please don't do this, please don't be scared.' I grabbed her hands like this—oh God, oh shit, they were so cold—but she wouldn't stop, she wouldn't . . . She was—"
"I know," I said softly.
Simon shook his head. "She was praying, you see, hands cupped together like this, the way I had taught her. And even though I already knew, oh shit, oh Jesus, even though I knew she wasn't really saying anything, I could hear her, she was crying, 'Please, God, please, please, please don't let me die.' "
I turned away. My throat was making stupid noises as I tried not to cry. I didn't know what to say, how to console him. And I know I should have felt horrible sadness, great sympathy for Simon, which I did feel. But to be completely honest, what I felt most of all was gut-wrenching fear. I had hated her, wished her dead, and now it was as if I'd killed her. I would have to pay for this. It would all come back to me, the full karmic circle, like Kwan and the mental hospital. I looked at Simon. He was gazing clear-eyed at the silhouettes of oak trees, the sparks of fireflies.
"You know, most of the time I know that she's gone," he said with an eerie calm. "But sometimes, when I think about her, our favorite song will come on the radio. Or a friend of hers from Utah will call right at that moment. And I don't think it's just coincidence. I sense her. She's there. Because, you see, we were connected, really connected, in every way. It wasn't just physical, that was the least of it. It was like well . . . can I read you something she wrote?"
I nodded blankly. Simon took out his wallet and unfolded a sheet of paper taped at the seams. "She sent this to me about a month before the accident, as part of my birthday present." I listened with a sickened heart.
" 'Love is tricky,' " he read in a quavery voice. " 'It is never mundane or daily. You can never get used to it. You have to walk with it, then let it walk with you. You can never balk. It moves you like the tide. It takes you out to sea, then lays you on the beach again. Today's struggling pain is the foundation for a certain stride through the heavens. You can run from it but you can never say no. It includes everyone.' " Simon folded the letter back up. "I still believe that," he said.
I was desperately trying to figure out what the words meant. But my mind was churning everything I had heard into frothy gibberish. Did he read the letter to say that's what he wanted from me?
"That was beautiful." I was ashamed I couldn't think of anything else to say.
"God! You don't know how relieved I am, I mean, being able to talk to you about her." His eyes were shiny, speedy, his words spilled out with abandon. "It's like she's the only one who _knows_ me, really knows me. It hits me all the time, and I know I need to let her go. But I'll be walking around campus thinking, No, she can't be gone. And then I see her, her same wavy hair, only when she turns around, it's someone else. But no matter how many times I'm wrong, I can't stop looking for her. It's like being addicted, going through the worst kind of withdrawal. I find her in everything, everyone." His eyes locked crazily on mine. "Like your voice. When I first met you, I thought it sounded a lot like hers."
I must have jumped a couple of inches, because Simon quickly added, "You have to understand I was sort of whacked-out when I met you. It was only three months after she, you know, had the accident. I wanted to believe she was still alive, living in Utah, that she was mad at me, that's why I hadn't seen her in a while. . . . Actually, now that I think about it, you two don't sound that much alike, not really." He traced a finger around my knuckles. "I never wanted to love anyone else. I figured it was enough, what Elza and I had. I mean, I figure most people never have that kind of love in a whole lifetime—you know what I mean?"
"You were lucky."
He kept stroking my knuckles. "And then I remembered what she wrote about not running away from love, not saying no. That you can't." He glanced up at me. "Anyway, that's why I had to tell you everything, so I can be open with you from here on out. And so you'd understand that I also have these other feelings, besides what I have for you, and if I'm not always there . . . well, you know."
I could barely breathe. I whispered in the softest possible voice, "I understand. I do. Really, I do." And then we both stood up and, without another word between us, walked out of the hills and back to my apartment.
What should have been one of the most romantic nights of my life was a nightmare for me. The whole time we made love I had the sense that Elza was watching us. I felt as if I were having sex during a funeral. I was afraid to make a sound. Yet Simon didn't act bereaved or guilty at all. You wouldn't have known he had just told me the saddest story I'd ever heard. He was like other lovers on first nights, eager to show me how versatile and experienced he was, worried he wasn't pleasing me, soon ready for a second round.
Afterward, I lay in bed, sleepless, thinking about music by Chopin and Gershwin, what they could possibly have in common. I could picture Elza's cherub-faced kneecaps, one of them with a beatific smile, and I wondered how a little baby could get a scar the shape and color of an earthworm. I thought about her eyes—what memories of hope and pain and violence had she inherited? Love moves you like the tide, she had said. I saw her floating on the wave of an avalanche.
By dawn, I could see Elza as Simon had. Her head was surrounded by a halo of light, her skin was as soft as a cherub's wings. And her icy blue eyes could see everything, the past into the future. She would always be dangerously beautiful, as pristine and alluring as a slope of fresh, bottomless snow.
LOOKING BACK, I can see I was an idiot to continue with Simon. But I was young, I was stupid-in-love. I confused a pathetic situation with a romantic one, sympathy with a mandate to save Simon from sorrow. And I've always been a magnet for guilt. My father, then Kwan, now Elza. I felt guilty about every bad thought I'd ever had about Elza. As penance, I sought her approval. I became her conspirator. I helped resuscitate her.
I remember the time I suggested to Simon that we go backpacking in Yosemite. "You told me how much Elza loved nature," I said. "I was thinking, if we went, well, she'd be there too." Simon looked grateful that I understood, and for me that was enough, that this was the way our love would grow. I just had to wait a bit. That's what I reminded myself later, when we were camped at a place called Rancheria Falls. Above us was a magnificent canopy of stars. It was so vast, so vivid, and my hope was the same. I struggled in my heart, then my brain to tell Simon this, but it all came out sounding trite. "Simon, look," I said. "Do you realize they're the same stars that the first lovers on earth saw?"
And Simon breathed in and exhaled deeply. I could tell he did this not with wonder but with informed sadness. So I was quiet, I understood, the way I said I would. I knew he was thinking about Elza again. Maybe he was thinking that she used to see these same stars. Or that she once expressed a similar thought, only more elegantly. Or that in the dark my voice was hers, with the same overly passionate tone, the one I used to express ordinary ideas, the one she would have used to save the whole damn world.
And then I felt myself becoming smaller yet denser, about to be crushed by the weight of my own heart, as if the laws of gravity and balance had changed and I was now violating them. I stared once again at those sharp little stars, twinkling like fireflies. Only now they were splotched and melting, and the night heaven was tilting and whirling, too immense to hold itself up any longer.
## [7
THE HUNDRED SECRET SENSES](TheHundredSecretSenses-toc.html#TOC-9)
The way I embraced Elza's former life, you'd have thought she was once my dearest, closest friend. When Simon and I had to pick recipes for Thanksgiving, we chose Elza's oyster-and-chestnut stuffing over my Chinese sticky-rice-and-sausage. We drank our coffee out of two-handled ceramic mugs Elza had made at a summer camp for musically gifted children. In the evenings and on weekends, we played Elza's favorite tapes: songs by the Blues Project, Randy Newman, Carole King, as well as a rather bathetic symphony that Elza herself had composed, which her college orchestra had recently performed and recorded as a memorial to her. To Simon, I said the music was living proof of her beliefs. But secretly, I thought it sounded like alley cats yowling on garbage night, with a finale of cans crashing as a well-aimed shoe flew out the window.
Then December rolled around, and Simon asked me what special gift I wanted for Christmas. The radio was playing holiday songs, and I was trying to think what Simon would want for Elza—a donation in her name to the Sierra Club? a collection of Gershwin records? That's when I heard Yogi Yorgesson singing "Yingle Bells."
The last time I had heard that song I was twelve, when I thought sarcasm was the height of cool. That year, I gave Kwan a Ouija board for Christmas. While she stared in bafflement at the old-fashioned letters and numbers, I told her she could use the Ouija to ask American ghosts how to spell English words. She patted the board and said, "Wunnerful, so useful." My stepfather threw a fit.
"Why do you feel you have to make fun of her?" Daddy Bob said to me sternly. Kwan examined the Ouija board, more puzzled than before.
"It was just a joke, okay?"
"Then it's a mean joke and you have a mean heart to do it." He grabbed my hand and jerked me up, saying, "Young lady, Christmas is over for you."
Alone, in the bedroom, I turned on the radio. That's when I heard "Yingle Bells" playing. The song was supposed to be a "yoke," like Kwan's present. I was crying bitterly: How was I being mean to Kwan if she didn't even _know_ it? Besides, I reasoned, if I was being mean, which I wasn't, she deserved it, she was so wacko. She invited people to play funny jokes on her. And what was so wrong about having fun on Christmas? It was the holier-than-thou people who were mean. Well, if everyone thought I was bad, I'd show them what bad was.
I turned the radio way up. I imagined the volume knob was Daddy Bob's big Italian nose, and I twisted it so hard it broke off, and now Yogi Yorgesson was singing "laughing all the way—ha-ha-ha!" at the top of his lungs while Daddy Bob was swearing, "Olivia, turn off that goddamn radio," which was not a Christian thing to say, especially on Christmas. I pulled the plug with a vengeance. Later Kwan came into the bedroom and told me she liked my spelling gift "oh very-very much."
"Stop acting so retarded," I growled. And I kept my face looking as mean as I could, but it scared me to see how much I had hurt her.
Now here was Simon asking me what I wanted for Christmas. Once again I was listening to "Yingle Bells" on the radio. And I wanted to cry out that being understanding gets you nowhere. At that moment, I knew what I really wanted for Christmas. I wanted to pull the plug. I wanted Elza dead.
But after six months of acting like the noble runner-up, how could I suddenly tell Simon I wanted to kick Elza's ghostly butt out of our bed? I imagined packing her photos, her records, her irritating kitsch into a box. "For safekeeping," I'd tell Simon, "while I do some spring cleaning." Then I'd load the box into the trunk of my car and late at night I'd drive to Lake Temescal. I'd weigh the box down with bleach bottles filled with sand, dump the whole mess into the dark moonless water, and watch the bubbles surface as my nemesis sank into liquid oblivion.
And later, what would I say to Simon, what explanation would I give him? "God, it's terrible, but the box with all of Elza's things?—it was stolen. I can't believe it either. The burglars must have thought it was valuable. I mean, it is, but just to you and me. God, you're right, I don't know why they didn't take the stereo."
He'd notice my evasive eyes, the corners of my mouth turned up in an irrepressible smile. I'd have to confess what I'd done, what I really felt about Elza and her two-handled coffee mugs. He'd be pissed, and that would be the end of Simon and me. If that was the case, to hell with him. But after I had exhausted my imaginary self with variations of this pyrrhic victory, I was lost. I couldn't let go of Simon any more than he could Elza.
It was in this wretched and murderous frame of mind that I sought an accomplice to do the dirty deed. I called Kwan.
I DISCREETLY OUTLINED the situation to my sister. I didn't say I was in love with Simon. To Kwan? And suffer her sisterly chuckles, endless teasing, and crackpot advice? I said Simon was a friend.
"Ah! Boyfriend," she guessed, all excited.
"No. Just a friend."
"Close friend."
" _Just_ a friend."
"Okay-okay, now I understand you meaning."
I told her that one of Simon's friends had died in an accident. I said that Simon was sad, that he couldn't let go of this friend who was dead. He was obsessed, it was unhealthy. I said it might help him if he heard from this friend as a yin person. Knowing how suggestible Kwan was, also how eager she was to help me in any way, I made the requirements as clear as possible.
"Maybe," I hinted, "Simon's dead friend can tell him they must both start a new life. He must forget about her, never mention her name."
"Ah! She was girlfriend."
"No, just a friend."
"Ah, like you, _just_ friend." She smiled, then asked, "Chinese too?"
"Polish, I think. Maybe also Jewish."
"Tst! Tst!" Kwan shook her head. "Polish-Jewish, very hard to find, so many dead Polish-Jewish. Many dead Chinese people too, but I have many connection for Chinese—this yin person know that yin person, easier for me to find if Chinese. But Polish-Jewish—ah!—maybe she don't even go to Yin World, maybe go someplace else."
"The next world is _segregated_? You can go to the World of Yin only if you're Chinese?"
"No-no! Miss Banner, she not Chinese, she go to Yin World. All depend what you love, what you believe. You love Jesus, go Jesus House. You love Allah, go Allah Land. You love sleep, go sleep."
"What if you don't believe in anything for sure before you die?"
"Then you go big place, like Disneyland, many places can go try— you like, you decide. No charge, of course."
As Kwan continued to ramble, I imagined an amusement park filled with ex–insurance agents dressed in hand-me-down angel costumes, waving fake lightning bolts, exhorting passersby to take an introductory tour of Limbo, Purgatory, the Small World of Unbaptized Infants. Meanwhile, there'd be hordes of former Moonies and est followers signed up for rides called the Pandemonium, the Fire and Brimstone, the Eternal Torture Rack.
"So who goes to the World of Yin?"
"Lots people. Not just Chinese, also people have big regret. Or people think they miss big chance, or miss wife, miss husband, miss children, miss sister." Kwan paused and smiled at me. "Also, people miss Chinese food, they go Yin World, wait there. Later can be born into other person."
"Oh, you mean yin people are those who believe in reincarnation."
"What mean recarnation?"
"Reincarnation. You know, after you die, your spirit or soul or whatever can be reborn as another human being."
"Yes, maybe this same thing, something like that. You not too picky, can come back fast, forty-nine days. You want something special—born to this person, marry that person—sometimes must wait long time. Like big airport, can go many-many places. But you want first class, window seat, nonstop, or discount, maybe have long delay. Hundred year at least. Now I tell you something, secret, don't tell anyone, ah. Many yin people, next life, guess who they want be. You guess."
"President of the United States."
"No."
"The Who."
"Who?"
"Never mind. Who do they want to be?"
"Chinese! I telling you true! Not French, not Japanese, not Swedish. Why? I think because Chinese food best, fresh and cheap, many-many flavors, every day different taste. Also, Chinese family very close, friends very loyal. You have Chinese friend or family one lifetime, stay with you ten thousand lifetime, good deal. That's why so many Chinese people live in world now. Same with people from India. Very crowded there. India people believe many lifetime too. Also, I hear India food not too bad, lots spicy dishes, curry flavors too. Course, Chinese curry still best. What you think, Libby-ah? You like my curry dish? You like, I can make for you tonight, okay?"
I steered Kwan back to the matter of Elza. "So what's the best way to go about finding Simon's friend? Where do Polish Jews usually go?"
Kwan started muttering: "Polish-Jewish, Polish-Jewish. So many places can go. Some believe nothing after die. Some say go in-between place, like doctor waiting room. Others go Zion, like fancy resort, no one ever complaining, no need tip, good service anyway." She shook her head, then asked, "How this person die?"
"A skiing accident in Utah. Avalanche. It's like drowning."
"Ah!—waterski affa lunch! Stomach too full, no wonder drown."
"I didn't say after lunch. I said—"
"No lunch? Then why she drown? Cannot swim?"
"She didn't drown! She was buried in the snow."
"Snow!" Kwan frowned. "Then why you say she drown?"
I sighed, about to go insane.
"She very young?"
"Twenty-one."
"Tst! This too sad. Happen when?"
"About a year ago."
Kwan clapped her hands. "How could I forgotten! My bachelor friend! Toby Lipski. Lipski, sound like 'ski.' Jewish too. Oh!—very funny yin person. Last year he die, liver cancer. He tell me, 'Kwan, you right, too much drinking at disco club, bad for me, very-very bad. When I come back, _no more drinking._ Then I can have long life, long love, _long_ penis.' Last part, course he just joking. . . ." Kwan looked at me to make sure she'd made her point about the evils of alcohol. "Toby Lipski also tell me, 'Kwan, you need yin favor, you ask for Toby Lipski.' Okay. Maybe I ask Toby Lipski find this girl. What's name?"
"Elza."
"Yes-yes, Elza. First I must send Toby message, like write letter with my mind." She squeezed her eyes shut and tapped the side of her head. Her eyes flew back open. "Send to Yin World. Everything with mind and heart together, use hundred secret sense."
"What do you mean, secret sense?"
"Ah! I already tell you so many time! You don't listen? Secret sense not really secret. We just call secret because everyone has, only forgotten. Same kind of sense like ant feet, elephant trunk, dog nose, cat whisker, whale ear, bat wing, clam shell, snake tongue, little hair on flower. Many things, but mix up together."
"You mean instinct."
"Stink? Maybe sometimes stinky—"
"Not stink, _in_ stinc _t._ It's a kind of knowledge you're born with. Like . . . well, like Bubba, the way he digs in the dirt."
"Yes! Why you let dog do that! This not sense, just nonsense, mess up you flower pot!"
"I was just making a—ah, forget it. What's a secret sense?"
"How I can say? Memory, seeing, hearing, feeling, all come together, then you know something true in you heart. Like one sense, I don't know how say, maybe sense of tingle. You know this: Tingly bones mean rain coming, refreshen mind. Tingly skin on arms, something scaring you, close you up, still pop out lots a goose bump. Tingly skin top a you brain, oh-oh, now you know something true, leak into you heart, still you don't want believe it. Then you also have tingly hair in you nose. Tingly skin under you arm. Tingly spot in back of you brain—that one, you don't watch out, you got a big disaster come, mm-hm. You use you secret sense, sometimes can get message back and forth fast between two people, living, dead, doesn't matter, same sense."
"Well, whatever you need to do," I said. "But put a rush on it."
"Wah!" Kwan snorted. "You think I work post office—shop late, mail Christmas Eve, deliver Christmas Day, everything rush-rush-rush? No such thing here, no such thing there either! Anyway, in Yin World, no need save time. Everything already too late! You want reach someone, must sense that person feeling, that person sense youself feeling. Then— _pung!—_ like happy accident when two self run into each other."
"Well, whatever. Just be sure to tell this Toby guy that the woman's name is Elza Vandervort. That's her adopted name. She doesn't know who her real parents were. She thinks they were Polish Jews who had been in Auschwitz. And she may be thinking about Chopin, musical stuff."
"Wah! You talking too fast."
"Here, I'll write it down for you."
Only afterward did I consider the irony of the whole matter: that I was helping Kwan with her illusions so that she could help Simon let go of his.
Two weeks later, Kwan told me Toby had hit a jackpot. He had set a date with Elza for the night of the next full moon. Kwan said that people in the World of Yin were very bad about making appointments, because nobody used a calendar or a clock anymore. The best method was to watch the moon. That was why so many strange things happened when the moon was at its brightest, Kwan said: "Like porch light, telling you guests welcome-welcome, come inside."
I STILL FEEL GUILTY over how easy it was to fool Simon. It went like this.
I mentioned to him we'd been invited to Kwan's for dinner. He accepted. The moment we walked in her house, Kwan said, "Ohhhh, so good-looking." As if he'd been prompted, Simon said to Kwan, "You're kidding. You don't look twelve years older than Olivia." Then Kwan beamed and said, "Ohhhh, good manner too."
The curry wasn't bad, the conversation wasn't too painful. Kwan's husband and her stepsons chatted excitedly about a fistfight they'd witnessed in the parking lot of Safeway. Throughout the dinner, Kwan did not act that weird, although she did ask Simon nosy questions about his parents. "Which one Chinese? Mother side. But not Chinese?. . . Ah, Hawaii-ah, I know, Chinese already premix. She do hula-hula dance? . . . Ah. Dead? So young? Ai, so sad. I see hula-hula once on TV, hip go 'round like wash machine, wavy hands like flying bird. . . ."
When Simon went to the bathroom, she gave me a wink and whispered loudly: "Hey! Why you say he just friend? This look on you face, his face, hah, not just friend! I right?" And then she broke into gales of belly laughter.
After dinner, on cue, George and the boys went to the family room to watch _Star Trek._ Kwan told Simon and me to come to the living room; she had something important to say. We sat down on the couch, Kwan in her easy chair. She pointed to the fake fireplace with its gas heater insert.
"Too cold?" she asked.
We shook our heads.
Kwan folded her hands in her lap. "Simon," she said, smiling like a genie, "tell me—you like my little sister, ah?"
"Kwan," I warned, but Simon was already answering her: "Very much."
"Hmm-mm." She looked as pleased as a cat after a tongue bath. "Even you don't tell me, I already see this. Mm-hmm . . . You know why?"
"I guess it's apparent," Simon said with a sheepish grin.
"No-no, you parent not telling me. I know—in here," and she tapped her forehead. "I have yin eyes, mm-hm, yin eyes."
Simon gave me a searching look, as if to ask, Help me out, Olivia— what's going on? I shrugged.
"Look there." Kwan was pointing to the fireplace. "Simon, what you see?"
He leaned forward, then made a stab at what he must have thought was a Chinese game. "You mean those red candles?"
"No-no, you see _fireplace._ I right?"
"Oh, yeah. Over there, a fireplace."
"You see fireplace. I see something else. A yin person standing there, somebody already dead."
Simon laughed. "Dead? You mean like a ghost?"
"Mm-hm. She say her name—Elsie." Good old Kwan, she accidentally said Elza's name wrong in exactly the right way. "Simon-ah, maybe you know this girl Elsie? She saying she know you, mm-hm."
His smile gone, Simon now sat forward. "Elza?"
"Oh, now she sooo happy you remember her." Kwan poised her ear toward the imaginary Elza, listening attentively. "Ah? . . . Ah. Okay-okay." She turned back to us. "She saying you won't believe, she already meet many famous music people, all dead too." She consulted the fireplace. "Oh? . . . Oh . . . Oh! . . . Ah, ah. No-no, stop, Elsie, too many name! You saying so many famous people name I can't repeat! Okay, one . . . Showman? No? I not pronouncing right?"
"Chopin?" I hinted.
"Yes-yes. Chopin too. But this one she say name like Showman. . . . Oh! Now understand— _Schumann_!"
Simon was mesmerized. I was impressed. I didn't know before that Kwan knew anything about classical music. Her favorite songs were country-western tunes about heartbroken women.
"She also saying so happy now meet her mother, father, big brother. She mean other family, not adopt-her one. Her real name she say sound like Wawaski, Wakowski, I think Japanese name. . . . Oh? Not Japanese? . . . Mm. She say Polish. Polish-Jewish. What? . . . Oh, okay. She saying her family die long time ago, because auto in ditch."
"Auschwitz," I said.
"No-no. Auto in ditch. Yes-yes, I right, auto go in ditch, turn over, crash!" Kwan cupped her right ear. "Lots time, beginning very hard understand what yin person saying. Too excited, talk too fast. Ah? . . ." She cocked her head slightly. "Now she saying, grandparents, they die this place, Auschwitz, wartime Poland." Kwan looked at me and gave me a wink, then quickly turned back to the fireplace with a surprised and concerned expression. "Ai-ya! Tst! Tst! Elsie, you suffer too much. So sad. Oh." Kwan touched her knee. "She saying, auto accident, this how she got scar on her little baby leg."
I didn't think I had written down that detail about Elza's scar. But I must have, and was glad I had. It added a nice authentic touch.
Simon blurted out a question: "Elza, the baby. What about the baby you were going to have? Is it with you?"
Kwan looked at the fireplace, puzzled, and I held my breath. Shit! I forgot to mention the damn baby. Kwan concentrated on the fireplace. "Okay-okay." She turned to us and brushed the air with one nonchalant hand. "Elsie say no problem, don't worry. She met this person, very nice person suppose be her baby. He not born yet, so didn't die. He have only small waiting time, now already born someone else."
I exhaled in relief. But then I saw that Kwan was staring at the fireplace with a worried face. She was frowning, shaking her head. And just as she did this, the top of my head began to tingle and I saw sparks fly around the fireplace.
"Ah," Kwan said quietly, more hesitantly. "Now Elsie saying you, Simon, you must no longer think about her. . . . Ah? Mm-hm. This wrong, yes-yes—too much waste you life think about her. . . . Ah? Hm. You must forget her, she say, yes, _forget!—_ never say her name. She have new life now. Chopin, Schumann, her mommy, daddy. You have new life too. . . ."
And then Kwan told Simon that he should grab me before it was too late, that I was his true-love girl, that he'd be forever sorry if he missed this good chance of many lifetimes. She went on and on about how honest and sincere I was, how kind, how loyal, how smart. "Oh, maybe she not so good cook, not yet, but you patient, wait and see. If not, I teach her."
Simon was nodding, taking it all in, looking sad and grateful at the same time. I should have been ecstatic right then, but I was nauseated. Because I also had seen Elza. I had heard her.
She wasn't like the ghosts I saw in my childhood. She was a billion sparks containing every thought and emotion she'd ever had. She was a cyclone of static, dancing around the room, pleading with Simon to hear her. I knew all this with my one hundred secret senses. With a snake's tongue, I felt the heat of her desire to be seen. With the wing of a bat, I knew where she fluttered, hovering near Simon, avoiding me. With my tingly skin, I felt every tear she wept as a lightning bolt against my heart. With the single hair of a flower, I felt her tremble, as she waited for Simon to hear her. Except I was the one who heard her—not with my ears but with the tingly spot on top of my brain, where you know something is true but still you don't want to believe it. And her feelings were not what came out of Kwan's well-meaning mouth. She was pleading, crying, saying over and over again: "Simon, don't forget me. Wait for me. I'm coming back."
I NEVER TOLD KWAN what I saw or heard. For one thing, I didn't want to believe it was anything but a hallucination. Yet over these last seventeen years, I've come to know that the heart has a will of its own, no matter what you wish, no matter how often you pull out the roots of your worst fears. Like ivy, they creep back, latching on to the chambers in your heart, leeching out the safety of your soul, then slithering through your veins and out your pores. On countless nights, I've awakened in the dark with a recurring fever, my mind whirling, scared about the truth. Did Kwan hear what I heard? Did she lie for my sake? If Simon found out we'd tricked him, what would he do? Would he realize he didn't love me?
On and on the questions came, and I let them pile up, until I was certain our marriage was doomed, that Elza would pull it down. It was an avalanche waiting to happen, balanced on one dangerous and slippery question: Why are we together?
And then the sun would climb above the sill. Morning light would make me squint. I'd look at the clock. I'd rise and touch the faucets of the shower. I'd adjust the hot and the cold, then awaken my mind with water sprayed hard against my skin. And I'd be grateful to return to what was real and routine, confined to the ordinary senses I could trust.
## [8
THE CATCHER OF GHOSTS](TheHundredSecretSenses-toc.html#TOC-10)
I had the Internal Revenue Service to thank for leading us to the altar.
We had been living together for three years, two of those post-college. In keeping with our shared dream to "make a substantive difference," we worked in the human services field. Simon was a counselor for Clean Break, which helped troubled teens with criminal records. I was an outreach worker for Another Chance, a program for pregnant drug addicts. We didn't earn much, but after we saw how much withholding tax the IRS took out of our monthly paychecks, we calculated how much we would save if we filed a joint return: a whole three hundred forty-six dollars a year!
With this sum dangling before our impoverished eyes, we debated whether it was right for the government to favor married couples. We both agreed that taxes were an insidious form of government coercion. But why give the government three hundred forty-six dollars to buy more weapons? We could use that money to buy new stereo speakers. It was definitely Simon who suggested we get married, I remember. "What do you think?" he said. "Should we be co-opted and file jointly?"
The wedding took place near the Rhododendron Gardens of Golden Gate Park, a site we figured was both free and romantically al fresco. But that June day, the fog rolled in on an arctic breeze, whipping our clothes and hair around, so that in the wedding photos we and our guests look deranged. While the Universal Life Church minister was intoning the blessing of the marriage, a park official announced loudly, "Excuse me, folks, but you need a permit to hold a gathering like this." So we rushed through the exchange of vows, packed up the wedding picnic and gifts, and hauled them back to our cramped apartment on Stanyan Street.
As icing to our ruined cake, the wedding gifts included none of the practical things we desperately needed to replace our ragtag assortment of sheets, towels, and kitchenware. Most of our friends provided joke gifts of the marital-aids variety. My former stepfather, Bob, gave us a crystal vase. Simon's parents presented us with an engraved sterling tray.
The rest of my family tried to outdo one another in finding that "special something" that our future grandchildren would inherit as heirlooms. From my mother, it was an original metal sculpture of a man and woman embracing, a work of art that Bharat Singh, her boyfriend of the moment, had made. My brother Tommy supplied us with a vintage pachinko machine, which he played every time he visited. Kevin gave us a case of red wine, which we were supposed to let age fifty years. But after a few impromptu weekend parties with friends, we had a fine collection of empty bottles.
Kwan's gift was actually quite beautiful, surprisingly so. It was a Chinese rosewood box with a carved lid. When I lifted the lid, the music of "The Way We Were" broke out in a stiff and mindless rhythm. In the jewelry compartment was a package of tea. "Make good feeling last long time," Kwan explained, and gave me a knowing look.
FOR THE FIRST SEVEN YEARS of our marriage, Simon and I went out of our way to agree on almost everything. For the next seven, we seemed to do the opposite. We didn't debate as he and Elza had on important issues, such as due process, affirmative action, and welfare reform. We argued over petty matters: Does food taste better if you heat the pan before putting olive oil in it? Simon yes, me no. We didn't have blowouts. But we squabbled often, as if from habit. And this made us ill-tempered with each other, less than loving.
As to our hopes, our dreams, our secret desires, we couldn't talk about those. They were too vague, too frightening, too important. And so they stayed inside us, growing like a cancer, a body eating away at itself.
In retrospect, I'm amazed how long our marriage lasted. I wonder about the marriages of other people, our friends, if they continue out of habit, or lethargy, or some strange combination of fear feeding into hope, then hope unleashing fear. I never thought our marriage was worse than anyone else's. In some respects, I felt ours was better than most. We made a nice-looking couple at dinner parties. We kept our bodies in shape, had a decent sex life. And we had one very big thing in common, our own business, public relations, mostly for nonprofit and medical groups.
Over the years, we developed a steady roster of clients—the National Kidney Foundation, the Brain Tumor Research Foundation, Paws for a Cause, a couple of hospitals, and one lucrative account, a sleaze-ball clinic that insisted on print ads using a lot of before-and-after photos of liposucted female buttocks. Simon and I worked out of a room in our apartment. I was the photographer, designer, desktop typographer, and pasteup artist. Simon was the copywriter, client manager, print buyer, and accounts receivable department. In matters of aesthetics, we treated each other with careful respect. We sought agreement in brochure layouts, type sizes, and headlines. We were extremely professional.
Our friends used to say, "You two are so lucky." And for years I wanted to believe we were as lucky as they enviously thought we were. I reasoned that the quarrels we had were only minor irritations, like slivers under the skin, dents on the car, easy enough to remove once we got around to it.
And then, almost three years ago now, Dudley, my godfather, a retired accountant whom I hadn't seen since babyhood, died and left me stocks in a small gene-splicing company. They weren't worth much when he died. But by the time the executor passed them along to me, the gene company had gone public, the stocks had split a couple of times, and thanks to the commercial miracles of DNA, Simon and I had enough money to buy, even with inflated San Francisco prices, a decent house in a terrific neighborhood. We did, that is, until my mother suggested that I share my luck with my brothers and Kwan. After all, she pointed out, Dudley was Daddy's friend and not anyone I had been especially close to. She was right, but I was hoping Kevin, Tommy, and Kwan would say, "Keep it, thanks for the thought." So much for hope. The one who surprised me the most was Kwan. She shrieked and danced like a contestant on _Wheel of Fortune._ After we cut up the inheritance pie and removed a hefty wedge for taxes, Simon and I had enough for a down payment on a modest house in an iffy neighborhood.
As a result, our search for a home took more than a year. Simon had suggested a 1950s fixer-upper in the fog-ridden Sunset district, which he thought we could sell in a few years for double our investment. What I had in mind was more of a shabby Victorian in up-and-coming Bernal Heights, a place we could remodel as home sweet home and not an investment. "You mean hovel sweet hovel," Simon said, after viewing one property.
We didn't see eye to eye on what we called "future potential." The potential, of course, had more to do with us. We both knew that living in a small dump required the kind of fresh, exuberant love in which nothing mattered except happily snuggling for warmth in the same cramped double bed. Simon and I had long before progressed to a king-size bed and an electric blanket with dual controls.
One foggy summer Sunday, we spotted an Open House sign for a co-op in a six-unit building on the fringe of Pacific Heights. By fringe, I mean that it was hanging on to neighborhood chic by a tattered thread. The building's backside rested in the Western Addition, where windows and doors were covered with saw-proof steel bars. And it was fully three blocks and two tax brackets away from the better streets of Pacific Heights, populated by families who could afford dog-walkers, au pairs, and two second homes.
In the common hallway, Simon picked up a sales leaflet riddled with hyphenated disclaimers. " 'A semi-luxurious, bi-level, lower Pacific Heights co-op,' " he read out loud, " 'located in a prestigious, once-grandiose Victorian mansion constructed in 1893 by the quite-famous architect Archibald Meyhew.' " Amazingly, the leaflet also boasted of ten rooms and a parking space, all for an asking price only a tad out of our budget. Everything else we had seen that was affordable had no more than five rooms, six if it lacked a garage.
I rang the bell for unit five. "It's a good price for the neighborhood," I remarked.
"It's not even a condo," said Simon. "With co-ops, I hear you have to go by loony rules to even change the wattage of your light bulbs."
"Look at that banister. I wonder if it's the original woodwork. Wouldn't that be great?"
"It's faux. You can tell by the lighter swirls. They're too regular."
Since Simon seemed to be quashing all interest in the place, I was going to suggest we leave. But then we heard rapid footsteps on the staircase and a man calling: "I'll be with you in a sec." Simon casually clasped his hand around mine. I couldn't remember the last time he had done that. Despite his criticisms, he must have liked the building's possibilities, enough anyway to want us to have the appearance of a happily married couple, sturdy in our finances, sufficiently stable to last through the escrow period.
The real estate agent and, as it turned out, creator of the sales sheet was a nattily dressed, balding young guy named Lester Roland or Roland Lester. He had the annoying habit of frequently clearing his throat, thus giving the impression he was either lying or on the verge of making an embarrassing confession.
He handed us a business card. "Have you bought in this neighborhood before, Mr. and Mrs. uh—?"
"Bishop. Simon and Olivia," Simon answered. "We live in the Marina district now."
"Then you know this is one of _the_ best residential areas of the city."
Simon acted blasé. "Pacific Heights, you mean, not the Western Addition."
"Well! You must be old pros at this. Want to see the basement first, I suppose."
"Yep. Let's get that over with."
Lester dutifully showed us the separate meters and hot water tanks, the common boiler and the copper pipes, while we both made experienced, noncommittal grunts. "As you'll notice"—Lester cleared his throat—"the foundation is the original brick."
"Nice." Simon nodded approvingly.
Lester frowned and gave us a moment of profound silence. "I mention this because"—he coughed—"as you may already know, most banks won't finance a building with a brick foundation. Earthquake fears, you know. But the owner is willing to carry a second mortgage, and at comparable market rates, if you qualify, of course."
Here it is, I thought, the reason why the place is for sale so cheap. "Has there been a problem with the building?"
"Oh no, not at all. Of course, it's gone through the usual settling, cosmetic cracks and such. All classic buildings get a few wrinkles—that's the privilege of age. Hell, we should all look so good at a hundred! And you also have to bear in mind this old painted lady has already survived the 'eighty-nine quake, not to mention the big one of 'aught-six. You can't say that about the newer buildings, can you now?"
Lester sounded all too eager, and I started smelling the unpleasant mustiness of a dump. In dark corners, I saw piles of beaten suitcases, their mouse-gnawed leather and cracked vinyl ashy with dust. In another storage area was an assortment of rusted heavy things—automobile parts, barbells, a metal toolbox—a monument to some prior tenant's overproduction of testosterone. Simon let go of my hand.
"The unit comes with only one garage space," said Lester. "But luckily, the man in unit two is blind and you can rent his space for a second car."
"How much?" Simon asked, just as I announced, "We don't have a second car."
Like a cat, Lester looked serenely at both of us, then said to me, "Well, that saves a lot of trouble, doesn't it." We started climbing a narrow stairwell. "I'm taking you up the back entry, what was once the servants' staircase, leading to the available unit. Oh, and by the way, a couple of blocks down—walking distance, you know?—there's a terrific private school, absolutely top-notch. By the third grade, those little monsters know how to tear apart a 386 computer and upgrade it to a 486. Incredible what they can teach your kids these days!"
And this time, Simon and I said in the same two beats, "No kids." We looked at each other, startled. Lester smiled, then said, "Sometimes that's very wise."
EARLIER IN OUR MARRIAGE, having children was the one big dream we shared. Simon and I were infatuated with the possibilities of our genetic merger. He wanted a girl who looked like me, I wanted a boy who looked like him. After six years of taking my temperature daily, of abstaining from alcohol between periods, of having sex by clockwork, we went to a fertility specialist, Dr. Brady, who told us Simon was sterile.
"You mean Olivia is sterile," Simon said.
"No, the tests indicate it's you," Dr. Brady answered. "Your medical records also report that your testicles didn't descend until you were three."
"What? I don't remember that. Besides, they're descended now. What does that have to do with anything?"
That day we learned a lot about the fragility of sperm, how sperm has to be kept cooler than body temperature; that's why the testicles hang outside, natural air-conditioning. Dr. Brady said that Simon's sterility wasn't simply a matter of low sperm count or low motility, that he had been sterile probably since pubescence, meaning since his first ejaculation.
"But that's impossible," Simon said. "I already know I can—well, it can't be. The tests are wrong."
Dr. Brady said in a voice practiced at consoling a thousand disbelieving men: "I assure you, sterility has nothing whatsoever to do with masculinity, virility, sexual drive, erection, ejaculation, or your ability to satisfy a partner." I noticed the doctor said "a partner" and not "your wife," as if to include many possibilities, past, present, and future. He then went on to discuss the contents of ejaculate, the physics of an erection, and other trivia that had nothing to do with the tiny duck rain boots that sat on our dresser, the Beatrix Potter books my mother had already collected for her future grandchild, the memory of a pregnant Elza screaming at Simon from the top of an avalanche-prone slope.
I knew Simon was thinking about Elza, wondering whether she had been wrong about the pregnancy. If so, that made her death all the more tragic, based on one stupid mistake after another. I also knew Simon had to be considering that Elza might have lied, that she hadn't been pregnant at all. But why? And if she had been pregnant, who had been her other lover? Why, then, did she lash out at Simon? None of the possible answers made any sense.
Ever since our yin-talk session with Kwan years before, Simon and I had avoided bringing up Elza's name. Now we found ourselves doubly tongue-tied, unable to discuss Simon's sterility, the questions it raised about Elza or, for that matter, our feelings about artificial insemination and adoption. Year after year, we avoided talking about babies, real, imagined, or hoped for, until there we were, on this third-floor landing, both of us informing this odious stranger named Lester, "No kids," as if we'd made our decision years earlier and it was as final then as it was now.
LESTER WAS SEARCHING through dozens of keys strung on a wire. "It's here somewhere," he muttered. "Probably the last one. Yep, wouldn't you know it—voilà!" He swung open the door and tapped his hand against the wall until he found the light switch. The apartment felt familiar at first—as if I had secretly visited this place a thousand times before, the rendezvous house of nightly dreams. There they were: the heavy wooden double doors with panes of wavy old glass, the wide hallway with its wainscoting of dark oak, the transom window throwing a shaft of light glittery with ancient dust. It was like coming back to a former home, and I couldn't decide whether my sense of familiarity was comforting or oppressive. And then Lester cheerily announced that we should start by looking at the "reception parlor," and the feeling evaporated.
"The architecture is what we call Eastlake and gothic revival," Lester was telling us. He went on to explain how the place had become a boardinghouse for itinerant salesmen and war widows in the twenties. In the forties, "gothic revival" evolved into "handyman special" when the building was converted into twenty-four dinky studios, cheap wartime housing. In the sixties, it became student apartments, and during the real estate boom of the early eighties, the building was again reincarnated, this time into the present six "semi-luxurious" co-ops.
I figured "semi-luxurious" referred to the cheap-glass chandelier in the foyer. "Semi-funky" would have been a more honest way to describe the apartment, which embodied an incongruous mix of its former incarnations. The kitchen with its Spanish-red tiles and wood-laminate cupboards had lost all traces of its Victorian lineage, whereas the other rooms were still generously decorated with useless gingerbread spandrels and plaster friezes in the corners of the ceilings. The radiator pipes no longer connected to radiators. The brick fireplaces had had their jowls bricked over. Hollow-core doors made do for recently improvised closets. And through Lester's grandiloquent real estate parlance, useless Victorian spaces had sprung important new purposes. A former stair landing backlit by a panel of amber glass became "the music hall"— perfect, I imagined, for a string quartet of midgets. What were once the suffocating quarters of a bottom-of-the-rung laundry maid now became at Lester's suggestion "the children's library," _not_ that there was an adult library. And half of a once commodious dressing room with a built-in cedar wardrobe—the other half was in the adjoining apartment—was now "the scriptorium." We listened patiently to Lester, words skittering out of his mouth like cartoon dogs on fresh-waxed linoleum, frantically going nowhere.
He must have noticed our dwindling interest; he toned down the bluster, changed tack, and aimed us toward "the excellent economics of classic lines and a little bit of elbow grease." We made a perfunctory inspection of the remaining rooms, a maze of cubbyholes, similarly inflated with pseudobaronial terms: the nursery, the breakfast parlor, the water closet, the last being an actual closet big enough only for one toilet and its seated occupant, knees pressed against the door. In a modern apartment, the whole floor space would have amounted to no more than four average-size rooms at best.
Only one room, on the top floor, remained to be seen. Lester invited us to climb the narrow staircase to the former attic, now "the grand boudoir." There, the jaws on our cynical faces dropped. We gazed about slowly like people awestruck from sudden religious conversion. Before us was an enormous room with ceilings that sloped into walls. It was equivalent in floor space to the entire nine rooms below. And in contrast to the musty darkness of the third floor, the attic was light and airy, painted clean white. Eight dormer windows jutted out of the sloped ceiling, leading our eyes into the cloud-spotted sky. Below our feet, wide-plank floors gleamed, shiny as an ice rink. Simon took my hand again and squeezed it. I squeezed back.
This had potential. Together, I thought, Simon and I could dream up ways to fill the emptiness.
THE DAY WE MOVED IN, I began stripping layers from the walls of the former nursery, soon to be dubbed my "inner sanctum." Lester had said that the original walls were mahogany with inlays of burl, and I was eager to uncover this architectural treasure. Aided by the dizzying fumes of paint thinner, I imagined myself an archaeologist digging through the strata of former lives whose histories could be reconstructed by their choice of wall coverings. First to peel off was a yuppie skin of Chardonnay-colored latex, stippled to look like the walls of a Florentine monastery. This was followed by flaky crusts of the preceding decades—eighties money green, seventies psychedelic orange, sixties hippie black, fifties baby pastels. And beneath those rolled off sheaves of wallpaper in patterns of gold-flocked butterflies, cupids carrying baskets of primroses, the repetitious flora and fauna of past generations who stared at these same walls during sleepless nights soothing a colicky baby, a feverish toddler, a tubercular aunt.
A week later, with raw fingertips, I reached a final layer of plaster and then the bare wood, which was not mahogany, as Lester had said, but cheap fir. Where it was not charred it was blackened with mildew, the probable result of an overzealous turn-of-the-century firehose. While I'm not someone prone to violence, this time I kicked at the wall so hard one of the boards caved in and exposed masses of coarse gray hair. I let out a tremendous scream, grade-B horror movie in pitch, and Simon bounded into the room, waving a trowel—as if _that_ would have been an effective weapon against a mass murderer. I pointed an accusing finger toward the hairy remains of what I believed was an age-old unsolved crime.
After an hour, Simon and I had torn off nearly all of the damaged and rotting wood. On the floor lay piles of hair resembling giant rats' nests. It was not until we called in a contractor to install drywall that we discovered we had removed bushels of horsehair, a form of Victorian insulation. The contractor also said that horsehair made for effective soundproofing. Well-to-do Victorians, we learned, constructed their homes so that one would not have to listen to anything as indelicate as a trill of sexual ecstasy or the trumpet blasts of indigestion emanating from adjoining rooms.
I mention this because Simon and I didn't bother to put back the horsehair, and at first I believed that had something to do with the strange acoustics we began to experience in the first month. The space between our wall and the adjoining apartment's had become a hollow shaft about a foot in width. And this shaft, I thought, served as a sounding board, capable of transmitting noises from the entire building, then converting them into thumps, hisses, and what sometimes sounded like lambada lessons being conducted upstairs in our bedroom.
Whenever we tried to describe our noise problem, I would imitate what I had heard: _Tink-tink-tink, whumpa-whumpa-whumpa, chh-chh-shhh._ Simon would compare the noise to a possible source: the tapping of an out-of-tune piano key, the flitter of mourning doves, the scraping of ice. We perceived the world so differently—that's how far apart we had grown.
There was another strange aspect to all this: Simon never seemed to be at home when the creepiest sounds occurred—like the time I was in the shower and heard the theme to _Jeopardy_ being whistled, a melody I found especially haunting since I couldn't get the annoying tune out of my mind the rest of the day. I had the feeling I was being stalked.
A structural engineer suggested that the racket might be coming from the useless radiator pipes. A seismic safety consultant told me that the problem might be simply the natural settling of a wood-frame building. With a little imagination, he explained, you might think creaks and groans were all sorts of things, doors slamming, people running up and down the stairs—although he had never known of anyone else who complained of the sound of glass breaking followed by snickers of laughter. My mother said it was rats, possibly even raccoons. She'd had that problem once herself. A chimneysweep diagnosed pigeons nesting in our defunct flues. Kevin said that dental fillings can sometimes transmit radio waves and I should see Tommy, who was my dentist. The problems persisted.
Strangely enough, our neighbors said they weren't bothered by any unusual sounds, although the blind man below us acidly mentioned that he could hear us playing our stereo too loud, especially in the mornings. That's when he did his daily Zen meditation, he said.
When my sister heard the thumps and hisses, she came up with her own diagnosis: "The problem not some _thing_ but some _body._ Mm-hm." As I continued to unpack books, Kwan walked around my office, her nose upraised, scenting like a dog in search of its favorite bush. "Sometime ghost, they get lost," she said. "You want, I try catch for you." She held out one hand like a divining rod.
I thought of Elza. Long ago, she had vanished from conversation, but she managed to dwell in the back of my brain, frozen in time, like a tenant under rent control who was impossible to evict. Now, with Kwan's ghosts, she had wriggled her way out.
"It's not ghosts," I said firmly. "We took out the insulation. The room's like an echo chamber."
Kwan dismissed my explanation with an authoritative sniff. She placed her hand over a spot on the floor. She wandered about the room, her hand quivering, tracking like a bloodhound. She emitted a series of "hmmms," each growing more conclusive: "HHhhmm! HhhmmMM!" Finally she stood in the doorway, absolutely still.
"Very strange," she said. "Someone here, I feel this. But not ghost. Living person, full of electricity, stuck in wall, also under floor."
"Well!" I joked. "Maybe we should charge this person rent."
"Living people always more trouble than ghost," Kwan continued. "Living people bother you because angry. Ghost make trouble only because sad, lost, confused."
I thought of Elza, pleading for Simon to hear her.
"Ghost, I know how catch," said Kwan. "My third auntie teach me how. I call ghost—'Listen me, ghost!'—one heart speaking each other." She gazed upward, looking sincere. "If she old woman, show her old slippers, leather bottoms already soft, very comfortable wear. If she young girl, show comb belong her mother. Little girl always love own mother hair. I put this treasure ghost love so much in big oil jar. When she go in—quick!—I put lid on tight. Now she ready listen. I tell her, 'Ghost! Ghost! Time you go Yin World.' "
Kwan looked at my frowning face and added: "I know–I know! In America don't have big oil jar, maybe don't even know what kind I mean. For American ghost, must use something else—like maybe big Tupperware. Or travel suitcase, Samsonite kind. Or box from very fancy store, not discount place. Yes-yes, this better idea, I think. Libby-ah, what's a name that fancy store, everybody know everything cost so much? Last year Simon bought you hundred-dollar pen there."
"Tiffany."
"Yes-yes, Tiffany! They give you blue box, same color like heaven. American ghost love heaven, pretty clouds. . . . Oh, I know. Where music box I give you wedding time? Ghost love music. Think little people inside play song. Go inside see. My last lifetime, Miss Banner have music box like this—"
"Kwan, I have work to do—"
"I know–I know! Anyway, you don't have ghost, you got living person sneak in you house. Maybe he did some sort a bad thing, now hiding, don't want get caught. Too bad I don't know how catch loose person. Better you call FBI. Ah—I know! Call that man on TV show, _American Most Wanted._ You call. I telling you, every week, they catch someone." So much for Kwan's advice.
And then something else occurred, which I tried to pass off as coincidence. Elza came back into our lives in rather dramatic fashion. One of her college classmates, who had gone on to become a producer of New Age music, revived a number of pieces Elza had composed called _Higher Consciousness._ The music later became the sound track to a television series on angels, which was ironic, as Simon pointed out, since Elza was not fond of Christian mythology. But then, overnight it seemed, everyone was wild for anything having to do with angels. The series received huge ratings, a CD of the sound track sold moderately well, and Simon started finding new self-worth in Elza's small fame. I never thought I'd hate angels so much. And Simon, who once loathed New Age music, would play her album whenever friends came over. He would casually remark that the composer had dedicated the music to him. Why's that, they would ask. Well, they had been lovers, best friends. Naturally, this caused some friends to smile at me in a consoling way, which I found maddening. I would then explain matter-of-factly that Elza had died before I met Simon. Yet somehow it sounded more like a confession, as if I'd said I had killed her myself. And then silence would permeate the room.
So along with all the sound effects in our house, I tried to pretend I wasn't bothered by Elza's music. I tried to ignore the increasing distance between Simon and me. I tried to believe that in matters of marriage, as with earthquakes, cancer, and acts of war, people such as myself were immune to unexpected disaster. But to pretend that all was right with the world, I first had to know what was wrong.
## [9
KWAN'S FIFTIETH](TheHundredSecretSenses-toc.html#TOC-11)
Simon and I never replaced the cheap-glass chandelier. When we first moved in, we found it offensive, a glaring insult to good taste. Later, the fixture became a joke. And soon it was merely a source of light we took for granted. It was there but not noticed, except when one of the bulbs burned out. We even tried to rid ourselves of this reminder by buying a dozen light bulbs from a blind-veterans' organization, sixty watts each, guaranteed to last fifty thousand hours, forever in foyer-light years. But then five out of six bulbs burned out within the year. We never got around to putting up the ladder to change them. With one bulb burning, the chandelier was practically invisible.
One night, this was about six months ago now, that last bulb gave out with a small pop, leaving us in darkness. Simon and I were about to go to our usual neighborhood restaurant for an after-work supper. "I'll buy some real bulbs tomorrow," said Simon.
"Why not a whole new light fixture?"
"What for? This one's not so bad. Come on, let's go. I'm hungry."
As we walked to the restaurant, I was wondering about what he had said, or rather, how he had said it, as if he didn't care about our life together anymore. Tacky was now good enough for us.
The restaurant was half empty. Soft, soporific music was playing in the background, white noise, the kind no one really listens to. While glancing at a menu I knew by heart, I noticed a couple in their fifties seated across from us. The woman wore a sour expression. The man seemed bored. I watched them awhile longer. They chewed, buttered bread, sipped water, never making eye contact, never saying a word. They didn't seem to be having a fight. They just acted resigned, disconnected from both happiness and discomfort. Simon was studying the wine list. Did we ever order anything except the house white?
"You want to share a bottle of red this time?" I said.
He didn't look up. "Red has all that tannin. I don't want to wake up at two in the morning."
"Well, let's get something different. A fumé blanc maybe."
He handed me the wine list. "I'm just going to have the house Chablis. But you go ahead."
As I stared at the list, I began to panic. Suddenly, everything about our life seemed predictable yet meaningless. It was like fitting all the pieces of a jigsaw puzzle only to find the completed result was a reproduction of corny art, great effort leading to trivial disappointment. Sure, in some ways we were compatible—sexually, intellectually, professionally. But we weren't _special,_ not like people who truly belonged to each other. We were partners, not soul mates, two separate people who happened to be sharing a menu and a life. Our whole wasn't greater than the sum of our parts. Our love wasn't destined. It was the result of a tragic accident and a dumb ghost trick. That's why he had no great passion for me. That's why a cheap chandelier fit our life.
When we arrived home, Simon flopped on the bed. "You've been awfully quiet," he said. "Anything wrong?"
"No," I lied. Then: "Well, I don't know, exactly." I sat on my side of the bed and started to page through a shopping catalogue, waiting for him to ask me again.
Simon was now using the television remote to change channels every five seconds: a news flash about a kidnapped little girl, a _teleno-vela_ in Spanish, a beefy man selling exercise equipment. As pieces of televised life blipped past me, I tried to gather my emotions into coherent logic Simon could understand. But whatever I'd been stifling hit me in a jumble and ached in my throat. There was the fact that we couldn't talk about Simon's sterility—not that I wanted to have children at this point in our lives. And the spooky sounds in the house, how we pretended they were normal. And Elza, how we couldn't talk about her, yet she was everywhere, in the memory of lies Kwan had told during her yin-talk session, in the damn music Simon played. I was going to suffocate if I didn't make drastic changes in my life. Meanwhile, Simon was still bouncing from one channel to the next.
"Do you know how irritating that is?" I said tersely.
Simon turned off the TV. He rolled over to face me, propped on one arm. "What's wrong?" He looked tenderly concerned.
My stomach clenched. "I just wonder sometimes, Is this all there is? Is this how we're going to be ten, twenty years from now?"
"What do you mean, is this all?"
"You know, living here in this funky house, putting up with the noise, the tacky chandelier. Everything feels stale. We go to the same restaurant. We say the same things. It's the same old shit over and over again."
He looked puzzled.
"I want to _love_ what we do as a couple. I want us to be closer."
"We're together twenty-four hours a day as it is."
"I'm not talking about work!" I felt like a small child, hungry and hot, itchy and tired, frustrated that I couldn't say what I really wanted to. "I'm talking about us, what's important. I feel like we're stagnating and mold is growing around the edges."
"I don't feel that way."
"Admit it, our life together won't be any better next year than it is today. It'll be worse. Look at us. What do we share now besides doing the same work, seeing the same movies, lying in the same bed?"
"Come on. You're just depressed."
"Of course I'm depressed! Because I can see where we're headed. I don't want to become like those people we saw in the restaurant tonight—staring at their pasta, nothing to say to each other except, 'How's the linguini?' As it is, we never talk, not really."
"We talked tonight."
"Yeah, sure. How the new client is a neo-Nazi. How we should put more in our SEP account. How the co-op board wants to raise the dues. That's not real talk! That's not real life. That's not what's important in _my_ life."
Simon playfully rubbed my knee. "Don't tell me you're having a mid-life crisis? People had those only in the seventies. Besides, today there's Prozac."
I brushed his hand away. "Stop being so condescending."
He put his hand back. "Come on, I'm joking."
"Then why do you always joke about important things?"
"Hey, you're not the only one. I wonder about my life too, you know, how long I have to do the things that really matter."
"Yeah? Like what?" I sneered. "What matters to you?"
He paused. I imagined what he was going to say: the business, the house, having enough money to retire early.
"Go on. Tell me."
"Writing," he finally said.
"You already write."
"I don't mean what I write now. Do you really think that's all I'm about—writing brochures on cholesterol and sucking fat out of flabby thighs? Gimme a break."
"What, then?"
"Stories." He looked at me, waiting for a reaction.
"What kind?" I wondered if he was making this up on the spot.
"Stories about real life, people here, or in other countries, Madagascar, Micronesia, one of those islands in Indonesia where no tourists have ever been."
"Journalism?"
"Essays, fiction, whatever allows me to write about the way I see the world, where I fit in, questions I have. . . . It's hard to explain."
He started to remove the catalogue from my hand. I grabbed it back. "Don't." We were on the defensive again.
"All right, stay in your goddamn funk!" he shouted. "So we're not perfect. We slip up. We don't talk enough. Does that make us miserable failures? I mean, we're not homeless or sick or working in mindless jobs."
"What, I'm supposed to be happy thinking, 'Gee, someone else has it worse than I do'? Who do you think I am—Pollyanna?"
"Shit! What do you want?" he snapped. "What could possibly make you happy?"
I felt stuck in the bottom of a wishing well. I was desperate to shout what I wanted, but I didn't know what that was. I knew only what it wasn't.
Simon lay back on the pillow, his hands locked over his chest. "Life's always a big fucking compromise," he said. He sounded like a stranger. "You don't always get what you want, no matter how smart you are, how hard you work, how good you are. That's a myth. We're all hanging in the best we can." He exhaled a cynical laugh.
And then I spit out what I had been afraid to say: "Yeah, well, I'm sick of hanging in as Elza's lousy replacement."
Simon sat up. "What the hell does Elza have to do with this?" he asked.
"Nothing." I was being stupid and childish, but I couldn't stop. A few tense minutes went by before I said: "Why do you have to play that goddamn CD all the time and tell everyone she was your girlfriend?"
Simon stared at the ceiling. He sighed sharply, a signal he was just about to give up. "What's going on?"
"I just want us, you know, to have a better life," I stammered. "Together." I couldn't meet his eyes. "I want to be important to you. I want you to be important to me. . . . I want us to have dreams together."
"Yeah, what kind of dreams?" he said hesitantly.
"That's the point—I don't know! That's what I want us to talk about. It's been so long since we had dreams together, we don't even know what that means anymore."
We were at a standstill. I pretended to read my magazine. Simon went to the bathroom. When he returned, he sat on the bed and put his arm around me. I hated myself for crying, but I couldn't stop. "I don't know, I don't know," I kept sobbing. He patted my eyes with a tissue, wiped my nose, then eased me down on the bed.
"It's all right," he consoled. "You'll see, tomorrow, it'll be all right."
But his niceness made me despair even more. He wrapped himself around me, and I tried to choke back my sobs, pretending to be calmed, because I didn't know what else to do. And then Simon did what he always did when we didn't know what else to do—he started to make love. I stroked his hair, to let him think this was what I wanted too. But I was thinking, Doesn't he worry about what's going to happen to us? Why doesn't he worry? We're doomed. It's just a matter of time.
The next morning, Simon surprised me. He brought me coffee in bed and brightly announced, "I've been thinking about what you said last night—about having a dream together. Well, I have a plan."
Simon's idea was to draw up a wish list: something we could do together, which would allow us to define what he called the creative parameters of our life. We talked openly, excitedly. We agreed the dream should be risky but fun, include exotic travel, good food, and most important, a chance to create something that was emotionally satisfying. We did not mention romance. "That takes care of the dream part," he said. "Now we have to figure out how to make it happen."
At the end of our three-hour discussion, we had conjured up a proposal that we would mail to a half-dozen travel and food magazines. We would offer to write and photograph a story on village cuisine of China; this would involve a junket to serve as the model for future food and folk-culture articles, possibly a book, a lecture tour, maybe even a cable TV series.
It was the best talk Simon and I had had in years. I still didn't think he completely understood my fears and despair, but he had responded in the best way he could. I wanted a dream. He made a plan. And when I thought about it, wasn't that enough to give us hope?
I realized we had about a one-zillionth-percent chance of getting even a nibble on our proposal. But once the letters were out in the universe, I felt better, as if I'd hauled off my old life to Goodwill. Whatever came next had to be better.
A FEW DAYS after Simon and I had our tête-à-tête, my mother called with a reminder to bring my camera to Kwan's house that evening. I looked at the calendar. Shit, I had completely forgotten we were supposed to go to Kwan's for her birthday. I ran upstairs to the bedroom, where Simon was watching Super Bowl highlights, his lanky body stretched across the rug in front of the TV. Bubba was lying next to him, chewing on a squeaky toy.
"We have to be at Kwan's in an hour. It's her birthday."
Simon groaned. Bubba jumped up into sitting position, front paws paddling, whining for his leash.
"No, Bubba, you have to stay." He slumped to the floor, head on paws, gazing at me with woeful eyes.
"We'll stay long enough to be social," I offered, "then skip out early."
"Oh, sure," said Simon, eyes still on the screen. "You know how Kwan is. She'll never let us leave early."
"Well, we have to go. It's her fiftieth."
I scanned the bookshelves to find something that might pass for a birthday gift. An art book? No, I decided, Kwan wouldn't appreciate it, she has no aesthetic sense. I looked in my jewelry box. How about this silver-and-turquoise necklace I hardly ever wear? No, my sister-in-law gave me that, and she'd be at the party. I went downstairs to my office, and that's where I spotted it: a mock-tortoiseshell box slightly larger than a pack of cards, perfect accompaniment for Kwan's kitschy junk. I had bought the box while Christmas shopping two months before. At the time, it looked like one of those all-purpose gifts, compact enough to tuck in my purse, just in case someone, for instance a client, surprised me with a Christmas present. But this year, no one did.
I went to Simon's study and rummaged around in his desk for wrapping paper and ribbon. In the bottom left-hand drawer, tucked in the back, I found a misplaced diskette. I was about to file it in Simon's storage box, when I noticed the index name he had written on the label: "Novel. Opened: 2/20/90." So he _was_ trying to write something important to him after all. He'd been working on it for a long time. I felt wounded that he hadn't shared this with me.
At that point, I should have respected Simon's privacy and filed the diskette away. But how could I not look? There was his heart, his soul, what mattered to him. I turned on the computer with shaky hands, slipped in the diskette. I called up the file named "Chap. 1." A screenful of words flashed on a blue background, and then the first sentence: _From the time she was six, Elise could hear a song once, then play it from memory, a memory inherited from her dead grandparents._
I scrolled through the first page, then the second. This is schlock, this is drivel, I kept telling myself. I read page after page, gorging myself on poison. And I imagined her, Elza, stroked by his fingertips, gazing back at him on the screen. I could see her smirking at me: "I came back. That's why you've never been happy. I've been here all along."
CALENDARS DON'T MEASURE time for me anymore. Kwan's birthday was six months ago, a lifetime ago. After I came home from her party, Simon and I fought viciously for another month. The pain seemed to last forever, but love disintegrated in a second. He camped out in his study, then moved out at the end of February, which now feels so long ago I can't even remember what I did the first few weeks alone.
But I'm getting used to the change. No routines, no patterns, no old habits, that's the norm for me now. It suits me. As Kevin told me last week at his birthday party, "You look good, Olivia, you really do."
"It's the new me," I said in a flip tone. "I'm using a new face cream, fruit acids."
I've surprised everyone by how well I've been doing—not just coping, but actually carving out a new life. Kwan is the only one who thinks otherwise.
Last night on the phone, she had this to say: "You voice, so tired-sounding! Tired live alone, I think. Simon same way. Tonight two you come my house eat dinner, like old time, just be friend—"
"Kwan, I don't have time for this."
"Ah, so busy! Okay, not tonight. Tomorrow, too busy again? You come tomorrow, ah."
"Not if Simon's there."
"Okay-okay. Just you come tonight. I make you potsticker, your favorite. Also give you wonton take home for you freezer."
"No talking about Simon, right?"
"No talk, just eat. Promise."
I'M INTO my second helping of potstickers. I keep waiting for Kwan to slip in some mention of my marriage. She and George are talking heatedly about Virginia, a cousin of George's dead wife, in Vancouver, whose nephew in China wants to immigrate to Canada.
George is chewing a mouthful. "His girlfriend wanted to catch a ride to Canada too. Forced him to marry her. My cousin, she had to start the paperwork all over again. Everything was almost approved, now— Hey! Go back to the end of the line. Wait eighteen more months."
"Two hundred dollar, new paperwork." Kwan reaches for a green bean with her chopsticks. "Many-many hour wasted, going this office, that office. Then what? Surprise—baby pop out."
George nods. "My cousin said, 'Hey, why didn't you wait? Now we have to add the baby, start the application process over again.' The nephew said, 'Don't tell the officials we got a baby. The two of us, we come first, go to college, find high-paying jobs, buy a house, car. Later, we find a way to bring the baby, one, two years from now.' "
Kwan puts her rice bowl down. "Leave baby behind! What sort a thinking this?" She glares at me, as if I were the one with child abandonment ideas. "College, money, house, job—where you think find such things? Who pay for college, big down payment?"
I shake my head. George grumbles, and Kwan makes a disgusted face. "Bean not tender, too old, no taste."
"So? What happened?" I ask. "Are they bringing the baby?"
"No." Kwan puts down her chopsticks. "No baby, no nephew, no wife. Virgie move San Francisco soon. America don't have immigration for nephew, Auntie Virgie can't sponsor. Now the nephew mother in China, sister to Virgie, she blame us lose her son's good chance!"
I wait for further explanation. Kwan pokes the air with her chopsticks. "Wah! Why you think you son so important? Own sister don't consider how much trouble! You son spoil. I already smell from here. _Hwai dan._ Bad egg."
"You told her this?"
"Never meet her."
"Then why is she blaming you?"
"Blame in letter because Virgie tell her we invite her stay with us."
"Did you?"
"Before not. Now letter say this, we invite. Otherwise she loose face. Next week, she come."
Even with constant exposure to Kwan, I don't think I will ever understand the dynamics of a Chinese family, all the subterranean intricacies of who's connected to whom, who's responsible, who's to blame, all that crap about losing face. I'm glad my life isn't as complicated.
At the end of the night, Kwan hands me a video. It's of her birthday party, the same day Simon and I had our major blowout, the one that led to our end.
I remember that I had raced upstairs, where Simon was getting dressed. I opened a dormer window and held up his diskette and shouted, "Here's your fucking novel! Here's what's important to you!" And then I let go of the diskette.
We shouted at each other for an hour, and then I said in a calm and detached voice the words that were more terrible than any curse: "I want a divorce." Simon shocked me by saying, "Fine," then stomped down the stairs, banged the door, and was gone. Not five minutes later, the phone rang. I made myself as unemotional as I could. No hurt, no anger, no forgiveness. Let him beg. On the fifth ring, I picked it up.
"Libby-ah?" It was Kwan, her voice shy and girlish. "Ma call you? You coming? Everyone already here. Lots food . . ."
I mumbled some sort of excuse.
"Simon sick? Just now? . . . Oh, food poison. Okay, you take care him. No-no. He more important than birthday." And when she said that, I was determined that Simon would no longer be more important than anything in my life, not even Kwan. I went to the party alone.
"Very funny video," Kwan is now saying, as she sees me to the door. "Maybe no time watch. Take anyway." And so the evening ends, without one mention of Simon.
Once home, I am forlorn. I try watching television. I try to read. I look at the clock. It's too late to call anyone. For the first time in six months, my life seems hollow and I'm desperately lonely. I see Kwan's video lying on the dresser. Why not? Let's go to a party.
I've always thought home videos are boring, because they're never edited. You see moments from your life that never should be replayed. You see the past happening as the present, yet you already know what's coming.
This one opens with blinking holiday lights, then pans out to show we are at the Mediterranean doorway of Kwan and George's house on Balboa Street. With the blurry sweep of the camera, we enter. Even though it's the end of January, Kwan always keeps the holiday decorations up until after her birthday. The video captures it all: plastic wreaths hung over aluminum-frame windows, the indoor-outdoor green-and-blue carpeting, the fake-wood-grain paneling, and a mish-mash of furniture bought at warehouse discount centers and Saturday tag sales.
The back of Kwan's permed hair looms into the frame. She calls out in her too loud voice: "Ma! Mr. Shirazi! Welcome-welcome, come inside." My mother and her boyfriend of the moment bounce into view. She's wearing a leopard-print blouse, leggings, and a black velour jacket with gold braid trim. Her bifocals have a purple gradient tint. Ever since her facelift, my mother has been wearing progressively more outrageous clothes. She met Sharam Shirazi at an advanced salsa dance class. She told me she liked him better than her last beau, a Samoan, because he knows how to hold a lady's hand, "not like a drumstick." Also, by my mother's estimation, Mr. Shirazi is quite the lover boy. She once whispered to me: "He does things maybe even you young people don't do." I didn't ask what she meant.
Kwan stares back at the camera to make sure George has recorded our mother's arrival properly. And then more people arrive. The video swerves toward them: Kwan's two stepsons, my brothers, their wives, their cumulative four children. Kwan greets them all, shouting out each child's name—"Melissa! Patty! Eric! Jena!"—then motions to George to get a shot of the kids grouped together.
Finally, there's my arrival. "Why so late!" Kwan complains happily. She grabs my arm and escorts me to the camera so that our faces fill the screen. I look tired, embarrassed, red-eyed. It's obvious I want to escape.
"This my sister, Libby-ah," Kwan is saying to the camera. "My _favorite_ best sister. Which one older? You guess. Which?"
In the next few scenes, Kwan acts as if she were on amphetamines, bouncing off the walls. There she is, standing next to her fake Christmas tree. She points to ornaments, gestures like the gracious hostess of a game show. There she is, picking up her presents. She exaggerates their heaviness, then shakes, tilts, smells each one before reading the name tag of the lucky recipient. Her mouth rounds in fake astonishment: "For _me_?" And then she laughs gruffly and holds up all ten fingers, closing and opening them like a flashing signal: "Fifty years!" she shouts. "Can you believe? No? How 'bout forty?" She comes closer to the camera and nods. "Okay-okay, forty."
The camera ricochets from one ten-second scene to the next. There they are, my mother sitting on Mr. Shirazi's lap: someone yells for them to kiss and they gladly oblige. Next are my brothers in the bedroom, watching ESPN; they wave to the camera with sloshing cans of beer. Now my sisters-in-law, Tabby and Barbara, are helping Kwan in the kitchen; Kwan holds up a coin-shaped slice of pork and cries, "Taste! Come close, taste!" In another bedroom, kids are huddled around a computer game; they cheer each time a monster has been killed. And now the whole family and I are standing in the buffet line, finding our way to a dining table which has been enlarged by the addition of a mah jong table at one end and a card table at the other.
I see a close-up of myself: I wave, toast Kwan, then go back to stabbing at my plate with a plastic fork, all the usual party behavior. But the camera is heartlessly objective. Anyone can see it in my face: my expressions are bland, my words are listless. It's so obvious how depressed I am, entirely resistant to what life has to offer. My sister-in-law Tabby is talking to me, but I'm staring vacantly at my plate. The cake arrives and everyone breaks into the Happy Birthday song. The camera pans the room and finds me on the sofa, setting into motion a tabletop toy of steel balls that make a perpetual and annoying _clack-clack_ sound. I look like a zombie.
Kwan opens her presents. The Hummel knockoff of skating children is from her coworkers at the drugstore. "Oh, so cute-cute," she croons, putting it next to her other figurines. The coffee maker is from my mother. "Ah, Ma! How you know my other coffee machine broken?" The silk blouse in her favorite color, red, is from her younger stepson, Teddy. "Too good to wear," Kwan laments with joy. The silver-plated candlestick holders are from her other stepson, Timmy; she puts candles in them, then sets them on the table he helped her refinish last year. "Just like First Lady in White House!" she gloats. The clay-blob sculpture of a sleeping unicorn is from our niece Patty; Kwan puts this carefully on the mantel, promising: "I never sell it, even when Patty become famous artist and this worth one million dollar." The daisy-patterned bathrobe is from her husband. She looks at the designer-soundalike label: "Ohhhh. Giorgio Laurentis. Too expensive. Why you spend so much?" She shakes her finger at her husband, who smiles, bashfully proud.
Another pile is set in front of Kwan. I fast-forward through the unveiling of placemats, a clothes steamer, a monogrammed tote bag. Finally I see her picking up my present. I press the Stop button, then hit Play.
". . . Always save best for last," she's proclaiming. "Must be very-very special, because Libby-ah my favorite sister." She unties the ribbon, puts it aside for safekeeping. The wrapping paper falls away. She purses her lips, staring at the tortoiseshell box. She turns it slowly from top to bottom, then lifts off the top and looks inside. She touches her hand to one cheek and says, "Beautiful, so useful too." She holds up the box for video-recorded history: "See?" she says, grinning. "Travel soap dish!"
In the background, you can hear my strained voice. "Actually, it's not for soap. It's for, you know, jewels and stuff."
Kwan looks at the box again. "Not for soap? For jews? Ohhh!" She holds up the box again, giving it more respect. Suddenly she brightens. "George, you hear? My sister Libby-ah say I deserve good jews. Buy me diamond, big diamond put in travel soap dish!"
George grunts and the camera swings wildly as he calls out: "The two sisters, stand by the fireplace." I'm protesting, explaining that I have to go home, that I have work to do. But Kwan is pulling me up from the sofa, laughing and calling to me, "Come-come, lazy girl. Never too busy for big sister."
The video camera whirs. Kwan's face freezes into a grin, as if she's waiting for a flash to go off. She squeezes me tight, forcing me to be even closer to her, then murmurs in a voice full of wonder, "Libby-ah, my sister, so special, so good to me."
And I'm on the verge of tears, both in the video and now watching my life happen over again. Because I can't deny it any longer. Any second, my heart is going to break.
# III
## [10
KWAN'S KITCHEN](TheHundredSecretSenses-toc.html#TOC-13)
Kwan says to come over at six-thirty, which is what time she always says to come over, only usually we don't start eating until closer to eight. So I ask if dinner will _really_ be ready at six-thirty, otherwise I'll come later, because I am _really_ busy. Six-thirty for sure, she says.
At six-thirty, George answers the door, bleary-eyed. He isn't wearing his glasses, and his thin patch of hair looks like an advertisement for anti–static cling products. He's just been promoted to manager at a Food-4-Less store in the East Bay. When he first started working there, Kwan didn't notice the 4 in the store's name, and even with reminders she still calls it Foodless.
I find her in the kitchen, cutting off the stems of black mushrooms. The rice hasn't been washed, the prawns haven't been deveined. Dinner is two hours away. I thump my purse down on the table, but Kwan is oblivious to my irritation. She pats a chair.
"Libby-ah, sit down, I have something must tell you." She slices mushrooms for fully half a minute more before dropping her bombshell. "I was talking to a yin person." She's now speaking in Chinese.
I sigh deeply, letting her know I am not in the mood for this line of conversation.
"Lao Lu, you also know him, but not in this lifetime. Lao Lu said that you must stay together with Simon. This is your _yinyuan,_ the fate that brings lovers together."
"And why is this my fate?" I say unpleasantly.
"Because in your last lifetime together, you loved someone else before Simon. Later, Simon trusted you with his whole life that you loved him too."
I almost fall off my chair. I have never told Kwan or anyone else the real reason we are getting divorced. I've said simply that we've grown apart. And now here's Kwan talking about it—as if the whole damn universe, dead and alive, knows.
"Libby-ah, you must believe," she says in English. "This yin friend, he say Simon telling you true. You think he love you less, she more— no!—why you think like this, always compare love? Love not like money. . . ."
I am livid to hear her defending him. "Come on, Kwan! Do you realize how crazy-stupid you sound? If anyone else heard you talking like this, they'd think you were nuts! If there really are ghosts, why don't I ever see them? Tell me that, huh."
She's now slicing open the backs of the prawns, pulling out their black intestines, leaving on their shells. "One time you can see," she says calmly. "Little-girl time."
"I was pretending. Ghosts come from the imagination, not the World of Yin."
"Don't say 'ghost.' To them this like racist word. Only bad yin person you call ghost."
"Oh, right. I forgot. Even dead people have become politically correct. Okay, so what do these _yin_ people look like? Tell me. How many of them are here tonight? Who's sitting in this chair? Mao Tse-tung? Chou En-lai? How about the Dowager Empress?"
"No-no, they not here."
"Well, tell them to drop by! Tell them I want to see them. I want to ask them if they have degrees in marriage counseling."
Kwan spreads newspapers on the floor to catch the grease from the stove. She slides the prawns into a hot pan and instantly the roar of blistering oil fills the kitchen. "Yin people want come, that's when come," she says above the din. "They never saying when, because treat me just like close family—come by no invitation, 'Surprise, we here.' But most times, come for dinner, when maybe one two dish not cook right. They say, 'Ah! This sea bass, too firm, not flaky, maybe cook one minute too much. And these pickle-turnips, not crunchy enough, should make sound like walk in snow, crunch-crunch, then you know ready eat. And this sauce—tst!—too much sugar, only foreigner want eat it.' "
Blah, blah, blah. It's so ludicrous! She's describing precisely what she, George, and his side of the family do all the time, the kind of talk I find boring as hell. It makes me want to laugh and scream at the same time, hearing her version of the pleasures of the afterlife described as amateur restaurant reviewing.
Kwan dumps the glistening prawns into a bowl. "Most yin people very busy, working hard. They want relax, come to me, for good conversation, also because say I'm excellent cook." She looks smug.
I try to trap Kwan in her own faulty logic: "If you're such an excellent cook, why do they come so often and criticize your cooking?"
Kwan frowns and sticks out her lower lip—as if I could be stupid enough to ask such a question! "Not real criticize, just friendly way open-talk, be frank like close friend. And don't really come eat. How can eat? They already dead! Only pretend eat. Anyway, most times they praise my cooking, yes, saying they never so lucky enough eat such good dish. Ai-ya, if only can eat my green-onion pancake, then can die oh so happy. But—too late—already dead."
"Maybe they should try take-out," I grumble.
Kwan pauses for a moment. "Ah-ha-ha, funny! You make joke." She pokes my arm. "Naughty girl. Anyway, yin people like come visit me, talk about life already gone, like banquet, many-many flavors. 'Oh,' they say, 'now I remember. This part I enjoy, this I not enjoy enough. This I eat up too fast. Why I don't taste that one? Why I let this piece my life gone spoiled, complete wasted?' "
Kwan pops a prawn into her mouth, slides it around from cheek to cheek, until she has extricated the shell intact minus every bit of flesh. I'm always amazed that she can do this. To me, it looks like a circus stunt. She smacks her lips approvingly. "Libby-ah," she says, holding up a small plate of golden shreds, "you like dried scallop?" I nod. "Georgie cousin Virgie send me from Vancouver. Sixty dollar one pound. Some people think too good for everyday. Should save best for later on." She throws the scallops into a pan of sliced celery. "To me, best time now. You wait, everything change. Yin people know this. Always ask me, 'Kwan, where best part my life gone? Why best part slip through my fingers like fast little fish? Why I save for last, find out later last already come before?' . . . Libby-ah, here, taste. Tell me, too salty, not salty enough?"
"It's fine."
She continues: " 'Kwan,' they telling me, 'you still alive. You can still make memory. You can make good one. Teach us how make good one so next time we remember what not suppose forget.' "
"Remember what?" I ask.
"Why they want come back, of course."
"And you help them remember."
"I already help many yin people this way," she brags.
"Just like Dear Abby."
She considers this. "Yes-yes, like Dear Abby." She's visibly pleased with the comparison. "Many-many yin people in China. America too, plenty." And then she starts to tally them on her fingers: "That young police officer—come my house time my car get stolened?—last lifetime he missionary in China, always saying 'Amen, Amen.' That pretty girl, work at bank now watch over my money so good, she another—bandit girl, long time ago rob greedy people. And Sarge, Hoover, Kirby, now Bubba, doggies, all them so loyal. Last lifetime they same one person. You guess who."
I shrug. I hate this game, the way she always inveigles me into her delusions.
"You guess."
"I don't know."
"Guess."
I throw my hands up. "Miss Banner."
"Ha! You guess wrong!"
"All right, tell me. Who, then?"
"General Cape!"
I slap my forehead. "Of course." I have to admit the whole idea of my dog's being General Cape is rather amusing.
"Now you know reason why first dog name Captain," Kwan adds.
"I named him that."
She wags her finger. "Demote him lower rank. You smart, teach him lesson."
"Teach him! Pff. That dog was so dumb. He wouldn't sit, wouldn't come, all he could do was beg for food. And then he ran away."
Kwan shakes her head. "Not run away. Run over."
"What?"
"Mm-hm. I see, didn't want tell you, you so little. So I say, Oh, Libby-ah, doggie gone, run away. I not lying. He run away to street before run over. Also, my English then not so good. Run away, run over, sound same to me. . . ." As Kwan belatedly speaks of Captain's death, I feel a twinge of childlike sadness, of wanting things back, of believing I can change the fact that I was less than kind to Captain if only I could see him one more time.
"General Cape, last lifetime no loyalty. That's why come back doggie so many times. He choose himself do this. Good choice. Last lifetime he so bad—so bad! I know because his one-half man told me. Also I can see . . . Here, Libby-ah, _huang do-zi,_ big bean sprout, see how yellow? Bought fresh today. Break off tails. You see any rotten spot, throw away. . . ."
General Cape, he was rotten too. He threw away other people. Nunumu, I told myself, you pretend General Cape is not here. I had to pretend for a long time. For two months, General Cape lived in the Ghost Merchant's House. For two months, Miss Banner opened her door every night to let him in. For those same two months, she didn't speak to me, not as her loyal friend. She treated me as if I were her servant. She pointed to spots on the bosom part of her white clothes, spots she claimed I had not washed out, spots I knew were the dirty fingerprints of General Cape. On Sundays, she preached exactly what Pastor Amen said, no more good stories. And there were other great changes during that time.
At meals, the missionaries, Miss Banner, and General Cape sat at the table for foreigners. And where Pastor Amen used to sit, that's where General Cape put himself. He talked in his loud, barking voice. The others, they just nodded and listened. If he raised his soup spoon to his lips, they raised their spoons. If he put his spoon down to say one more boast, they put their spoons down to listen to one more boast.
Lao Lu, the other servants, and I sat at the table for Chinese. The man who translated for Cape, his name, he told us, was Yiban Johnson, One-half Johnson. Even though he was half-and-half, the foreigners decided he was more Chinese than Johnson. That's why he had to sit at our table as well. At first, I didn't like this Yiban Johnson, what he said—how important Cape was, how he was a hero to both Americans and Chinese. But then I realized: What he spoke was what General Cape put in his mouth. When he sat at our dinner table, he used his own words. He talked to us openly, like common people to common people. He was genuinely polite, not pretending. He joked and laughed. He praised the food, he did not take more than his share.
In time, I too thought he was more Chinese than Johnson. In time, I didn't even think he looked strange. His father, he told us, was American-born, a friend of General Cape's from when they were little boys. They went to the same military school together. They were kicked out together. Johnson sailed to China with an American company doing the cloth trade, Nankeen silk. In Shanghai, he bought the daughter of a poor servant as his mistress. Just before she was about to have his child, Johnson told her, "I'm going back to America, sorry, can't take you with me." She accepted her fate. Now she was the leftover mistress of a foreign devil. The next morning, when Johnson awoke, guess who he saw hanging from the tree outside his bedroom window?
The other servants cut her down, wrapped a cloth around the red neck gash where the rope had twisted life out of her body. Because she had killed herself, they held no ceremonies. They put her in a plain wood coffin and closed it up. That night, Johnson heard a crying sound. He rose and went into the room where the coffin lay. The crying grew louder. He opened the box, and inside he found a baby boy, lying between the legs of the dead mistress. Around the baby's neck, just under his tiny chin, was a red mark, thick as a finger, the same half-moon shape of the rope burn on his mother.
Johnson took that baby who was one-half his blood to America. He put the baby in a circus, told people the hanging story, showed them the mysterious rope-burn scar. When the boy was five, his neck was bigger, his scar looked smaller, and nobody paid to see if it was mysterious anymore. So Johnson went back to China with the circus money and his half-blood son. This time, Johnson took up the opium trade. He went from one treaty-port city to another. He made a fortune in each city, then gambled each fortune away. He found a mistress in each city, then left each mistress behind. Only the little Yiban cried to lose so many mothers. That was who taught him to speak so many Chinese dialects— Cantonese, Shanghainese, Hakka, Fukien, Mandarin—those mistress-mothers. English he learned from Johnson.
One day, Johnson ran into his old schoolmate Cape, who now worked for any kind of military army—the British, the Manchus, the Hakkas, it didn't matter which—whoever would pay him. Johnson said to Cape, "Hey, I have a big debt, lots of trouble, can you loan your old friend some money?" As proof that he would repay him, Johnson said, "Borrow my son. Fifteen years old and he speaks many languages. He can help you work for any army you choose."
Since that day, for the next fifteen years, young Yiban Johnson belonged to General Cape. He was his father's never paid debt.
I asked Yiban: Who does General Cape fight for now—the British, the Manchus, the Hakkas? Yiban said Cape had fought for all three, had made money from all three, had made enemies among all three. Now he was hiding from all three. I asked Yiban if it was true that General Cape had married a Chinese banker's daughter for gold. Yiban said Cape married the banker's daughter not just for gold, but for the banker's younger wives as well. Now the banker was looking for him too. Cape, he said, was addicted to golden-millet dreams, riches that could be harvested in one season, then plowed under, gone.
I was happy to hear that I was right about General Cape, that Miss Banner was wrong. But in the next instant, I became sick with sadness. I was her loyal friend. How could I be glad, watching this terrible man devour her heart?
Then Lao Lu spoke up: "Yiban, how can you work for such a man? No loyalty, not to country, not to family!"
Yiban said, "Look at me. I was born to a dead mother, so I was born to no one. I have been both Chinese and foreign, this makes me neither. I belonged to everyone, so I belong to no one. I had a father to whom I am not even one-half his son. Now I have a master who considers me a debt. Tell me, whom do I belong to? What country? What people? What family?"
We looked at his face. In all my life, I had never seen a person so intelligent, so wistful, so deserving to belong. We had no answer for him.
That night, I lay on my mat, thinking about those questions. What country? What people? What family? To the first two questions, I knew the answers right away. I belonged to China. I belonged to the Hakkas. But to the last question, I was like Yiban. I belonged to no one else, only myself.
Look at me, Libby-ah. Now I belong to lots of people. I have family, I have you. . . . Ah! Lao Lu says no more talking! Eat, eat before everything gets cold.
## [11
NAME CHANGE](TheHundredSecretSenses-toc.html#TOC-14)
As it turns out, Kwan was right about the sounds in the house. There _was_ someone in the walls, under the floors, and he was full of anger and electricity.
I found out after our downstairs neighbor, Paul Dawson, was arrested for making crank phone calls to thousands of women in the Bay Area. My automatic response was sympathy; after all, the poor man was blind, he was lonely for companionship. But then I learned the nature of his calls: he had claimed to be a member of a cult that kidnapped "morally reprehensible" women and turned them into "sacrificial village dolls," destined to be penetrated by male cult members during a bonding rite, then eviscerated alive by their female worker bees. To those who laughed at his phone threats, he said, "Would you like to hear the voice of a woman who also thought this was a joke?" And then he played a recording of a woman screaming bloody murder.
When the police searched Dawson's apartment, they found an odd assortment of electronic equipment: tape recorders hooked up to his telephone, redialers, voice changers, sound-effects tapes, and more. He hadn't limited his terrorist activities to the telephone. Apparently, he felt the prior owners of our apartment also had been too noisy, inconsiderate of his morning Zen meditations. When they temporarily moved out during a remodeling phase, he punched holes in his ceiling and installed speakers and bugging devices underneath the upstairs floor, enabling him to monitor the doings of his third-floor neighbors and spook them with sound effects.
My sympathy immediately turned into rage. I wanted Dawson to rot in jail. For all this time, I had been driven nearly crazy with thoughts of ghosts—one in particular, even though I would have been the last to admit so.
But I'm relieved to know what caused the sounds. Living alone edges my imagination toward danger. Simon and I see each other only for business reasons. As soon as we're fiscally independent, we'll divorce ourselves from our clients as well. In fact, he's coming over later to deliver copy for a dermatologist's brochure.
But now Kwan has dropped by, uninvited, while I'm in the middle of a phone call to the printer's. I let her in, then return to my office. She's brought some homemade wontons, which she is storing in my freezer, commenting loudly on the lack of provisions in my fridge and cupboards: "Why mustard, pickles, but no bread, no meat? How you can live like this way? And beer! Why beer, no milk?"
After a few minutes, she comes into my office, a huge grin on her face. In her hands is a letter I had left on the kitchen counter. It's from a travel magazine, _Lands Unknown,_ which has accepted Simon's and my proposal for a photo essay on village cuisines of China.
When the letter arrived the day before, I felt as though I had won the lottery only to remember I'd thrown my ticket away. It's a cruel joke played on me by the gods of chance, coincidence, and bad luck. I've spent the better part of the day and night gnawing on this turn of events, playing out scenarios with Simon.
I pictured him scanning the letter, saying, "God! This is unbelievable! So when are we going?"
"We're not," I'd say. "I'm turning it down." No hint of regret in my voice.
Then he'd say something like: "What do you mean, turning it down?"
And I'd say, "How could you even _think_ we'd go together?"
Then maybe—and this really got my blood boiling—maybe he'd suggest that _he_ would still go, and take along another photographer.
So I'd say, "No, you're not, because _I'm_ going and I'm bringing along another writer, a _better_ writer." And then the whole thing would escalate into a volley of insults about morals, business ethics, and comparative talent, variations of which kept me awake most of the night.
"Ohhhh!" Kwan is now cooing, waving the letter with joy. "You and Simon, going to China! You want, I go with you, be tour guide, do translation, help you find lots bargains. Of course, pay my own way. For long time I want go back anyway, see my auntie, my village—"
I cut her off: "I'm not going."
"Ah? Not going? Why not?"
"You know."
"I know?"
I turn around and look at her. "Simon and I are getting divorced. Remember?"
Kwan ponders this for two seconds, before answering: "Can go like friends! Why not just friends?"
"Drop it, Kwan, _please._ "
She looks at me with a tragic face. "So sad, so sad," she moans, then walks out of my office. "Like two starving people, argue-argue, both throw out rice. Why do this, why?"
When I show Simon the letter, he is stunned. Are those actually tears? In all the years I've known him, I've never seen him cry, not at sad movies, not even when he told me about Elza's death. He swipes at the wetness on his cheeks. I pretend not to notice. "God," he says, "the thing we wished for so much came through. But we didn't."
We're both quiet, as if to remember our marriage with a few moments of respectful silence. And then, in a bid for strength, I take a deep breath and say, "You know, painful as it's been, I think the breakup has been good for us. I mean, it forces us to examine our lives separately, you know, without assuming our goals are the same." I feel my tone has been pragmatic, but not overly conciliatory.
Simon nods and softly says, "Yeah, I agree."
I want to shout, What do you mean, _you_ agree! All these years we never agreed on anything, and now you agree? But I say nothing, and even congratulate myself for being able to keep my ill feelings in check, to not show how much I hurt. A second later, I am overwhelmed with sadness. Being able to restrain my emotions isn't a great victory—it's the pitiful proof of lost love.
Every word, every gesture is now loaded with ambiguity, nothing can be taken at face value. We speak to each other from a safe distance, pretending all the years we soaped each other's backs and pissed in front of each other never happened. We don't use any of the baby talk, code words, or shorthand gestures that had been our language of intimacy, the proof that we belonged to each other.
Simon looks at his watch. "I better go. I'm supposed to meet someone by seven."
Is he meeting a woman? This soon? I hear myself say, "Yeah, I have to get ready for a date too." His eyes barely flicker, and I blush, certain that he knows I've told a pathetic lie. As we walk to the door, he glances up.
"I see you finally got rid of that stupid chandelier." He gazes back at the apartment. "The place looks different—nicer, I think, and more quiet."
"Speaking of quiet," I say, and tell him about Paul Dawson, the house terrorist. Simon's the only one I know who can fully appreciate the outcome.
"Dawson?" Simon shakes his head, incredulous. "What a bastard. Why would he do something like that?"
"Loneliness," I say. "Anger. Revenge." And I sense the irony of what I've just said, a poker stabbing the ashes of my heart.
After Simon leaves, the apartment does feel awfully quiet. I lie on the rug in the bedroom and stare at the night sky through a dormer window. I think about our marriage. The weft of our seventeen years together was so easily torn apart. Our love was as ordinary as the identical welcome mats found in the suburbs we grew up in. The fact that our bodies, our thoughts, our hearts had once moved in rhythm with each other had only fooled us into thinking we were special.
And all that talk about the breakup being good for us—who am I trying to fool? I'm cut loose, untethered, not belonging to anything or anybody.
And then I think about Kwan, how misplaced her love for me is. I never go out of my way to do anything for her unless it's motivated by emotional coercion on her part and guilt on mine. I never call her out of the blue to say, "Kwan, how about going to dinner or a movie, just the two of us?" I never take any pleasure in simply being nice to her. Yet there she is, always hinting about our going together to Disneyland or Reno or China. I bat away her suggestions as though they were annoying little flies, saying I hate gambling, or that southern California is definitely not on my list of places to visit in the near future. I ignore the fact that Kwan merely wants to spend more time with me, that I am her greatest joy. Oh God, does she hurt the way I do now? I'm no better than my mother!—careless about love. I can't believe how oblivious I've been to my own cruelty.
I decide to call Kwan and invite her to spend a day, maybe even a weekend, with me. Lake Tahoe, that would be nice. She'll go berserk. I can't wait to hear what she says. She won't believe it.
But when Kwan answers the phone, she doesn't wait for me to explain why I've called. "Libby-ah, this afternoon I talking to my friend Lao Lu. He agree, you _must_ go China—you, Simon, me together. This year Dog Year, next year Pig, too late. How you cannot go? This you fate waiting to happen!"
She rambles on, countering my silences with her own irrefutable logic. "You half-Chinese, so must see China someday. What you think? We don't go now, maybe never get another chance! Some mistake you can change, this one cannot. Then what you do? What you think, Libby-ah?"
In hopes that she'll cease and desist, I say, "All right, I'll think about it."
"Oh, I know you change mind!"
"Wait a minute. I didn't say I'd go. I said I'd think about it."
She's off and running. "You and Simon _love_ China, guarantee one hundred percent, specially my village. Changmian so beautiful you can't believe. Mountain, water, sky, like heaven and earth come together. I have things I leave there, always want give you. . . ." She goes on for another five minutes, extolling the virtues of her village before announcing, "Oh-oh, doorbell ringing. I call you again later, okay?"
"Actually, I called you."
"Oh?" The doorbell sounds once more. "Georgie!" she cries. "Georgie! Answer door!" Then she shouts, "Virgie! Virgie!" Is George's cousin from Vancouver already living with them? Kwan comes back on the line. "Wait minute. I go answer door." I hear her welcoming someone, and then she's on the line again, slightly breathless. "Okay. Why you call?"
"Well, I wanted to ask you something." I immediately regret what I haven't said yet. What am I getting myself into? I think about Lake Tahoe, being marooned with Kwan in a dinky motel room. "This is sort of last-minute, so I understand if you're too busy—"
"No-no, never too busy. You need something, ask. My answer always yes."
"Well, I was wondering, well"—and then I say, all in a rush—"what are you doing tomorrow for lunch? I have to take care of some business near where you work. But if you're busy, we could do it another day, no big deal."
"Lunch?" Kwan says brightly. "Oh! Lunch!" Her voice sounds heartbreakingly happy. I curse myself for being so stingy with my token gift. And then I listen, flabbergasted, as she turns away from the receiver to announce, "Simon, Simon—Libby-ah call me have lunch tomorrow!" I hear Simon in the background: "Make sure she takes you somewhere expensive."
"Kwan? Kwan, what's Simon doing there?"
"Come over eat dinner. Yesterday I already ask you. You say busy. Not too late, you want come now, I have extra."
I look at my watch. Seven o'clock. So this is his date. I nearly jump for joy. "Thanks," I tell her. "But I'm busy tonight." My same excuse.
"Always too busy," she answers. Her same lament.
Tonight, I make sure my excuse isn't a lie. As penance, I busy myself making a to-do list of unpleasant tasks I've been putting off, one of which is changing my name. That necessitates changing my driver's license, credit cards, voter registration, bank account, passport, magazine subscriptions, not to mention informing our friends and clients. It also means deciding what last name I will use. Laguni? Yee?
Mom suggested I keep the name Bishop. "Why go back to Yee?" she reasoned. "There aren't any other Yees you're related to in this country. So who's going to care?" I didn't remind Mom about her pledge to do honor to the Yee family name.
As I think more about my name, I realize I've never had any sort of identity that suited me, not since I was five at least, when my mother changed our last name to Laguni. She didn't bother with Kwan's. Kwan's name remained Li. When Kwan came to America, Mom said that it was a Chinese tradition for girls to keep their mother's last name. Later, she admitted that our stepfather didn't want to adopt Kwan since she was nearly an adult. Also, he didn't want to be legally liable for any trouble she might cause as a Communist.
Olivia Yee. I say the name aloud several times. It sounds alien, as though I'd become totally Chinese, just like Kwan. That bothers me a little. Being forced to grow up with Kwan was probably one of the reasons I never knew who I was or wanted to become. She was a role model for multiple personalities.
I call Kevin for his opinion on my new name. "I never liked the name Yee," he confesses. "Kids used to yell, 'Hey, Yee! Yeah, you, yee-eye-yee-eye-oh.' "
"The world's changed," I say. "It's hip to be ethnic."
"But wearing a Chinese badge doesn't really get you any bonus points," Kevin says. "Man, they're cutting Asians out, not making more room for them. You're better off with Laguni." He laughs. "Hell, some people think Laguni's Mexican. Mom did."
"Laguni doesn't feel right to me. We don't really belong to the Laguni lineage."
"Nobody does," says Kevin. "It's an orphan's name."
"What are you talking about?"
"When I was in Italy a couple of years ago, I tried to look up some Lagunis. I found out it's just a made-up name that nuns gave to orphans. Laguni—like 'lagoon,' isolated from the rest of the world. Bob's grandfather was an orphan. We're related to a bunch of orphans in Italy."
"Why didn't you ever tell us this before?"
"I told Tommy and Mom. I guess I forgot to tell you because—well, I figured you weren't a Laguni anymore. Anyway, you and Bob didn't get along that much. To me, Bob's the only dad I ever knew. I don't remember anything about our real father. Do you?"
I do have memories of him: flying into his arms, watching him crack open the claws of a crab, riding on his shoulders as he walked through a crowd. Aren't they enough that I should pay tribute to his name? Isn't it time to feel connected to somebody's name?
AT NOON, I go to the drugstore to pick up Kwan. We first spend twenty minutes while she introduces me to everyone in the store—the pharmacist, the other clerk, her customers, all of whom happen to be her "favoritests." I choose a Thai restaurant on Castro, where I can watch the street traffic from a window table while Kwan carries on a one-sided conversation. Today I'm taking it like a good sport; she can talk about China, the divorce, my smoking too much, whatever she wants. Today is my gift to Kwan.
I put on my reading glasses and scan the menu. Kwan scrutinizes the restaurant surroundings, the posters of Bangkok, the purple-and-gold fans on the walls. "Nice. Pretty," she says, as if I had taken her to the finest place in town. She pours us tea. "So!" she proclaims. "Today you not too busy."
"Just taking care of personal stuff."
"What kind personal?"
"You know, renewing my residential parking permit, getting my name changed, that sort of thing."
"Name change? What's name change?" She unfolds her napkin in her lap.
"I have to do all this junk to change my last name to Yee. It's a hassle, going to the DMV, the bank, City Hall. . . . What's the matter?"
Kwan is vigorously shaking her head. Her face is scrunched up. Is she choking?
"Are you all right?"
She flaps her hands, unable to speak, looking frantic.
"Omigod!" I try to recall how to do the Heimlich maneuver.
But now Kwan motions that I should sit down. She swallows her tea, then moans, "Ai-ya, ai-ya. Libby-ah, now I sorry must tell you something. Change name to Yee, don't do this."
I steel myself. No doubt she's going to argue once again that Simon and I shouldn't divorce.
She leans forward like a spy. "Yee," she whispers, "that not really Ba's name."
I sit back, heart pounding. "What are you talking about?"
"Ladies," the waiter says. "Have we decided?"
Kwan points to an item on the menu, asking first how to pronounce it. "Fresh?" she asks. The waiter nods, but not with the enthusiasm that Kwan requires. She points to another item. "Tender?"
The waiter nods.
"Which one better?"
He shrugs. "Everything is good," he says. Kwan looks at him suspiciously, then orders the pad thai noodles.
When the waiter leaves, I ask, "What were you saying?"
"Sometimes menu say fresh _—not_ fresh!" she complains. "You don't ask, maybe they serve you yesterday leftovers."
"No, no, not the food. What were you saying about Daddy's name?"
"Oh! Yes-yes." She hunches her shoulders, and drops once again to her spy pose. "Ba's name. Yee not his name, no. This true, Libby-ah! I only telling you so you don't go through life with wrong name. Why make ancestors happy not our own?"
"What are you talking about? How could Yee not be his name?"
Kwan looks from side to side, as if she were about to reveal the identities of drug lords. "Now I going to tell you something, ah. Don't tell anyone, promise, Libby-ah?"
I nod, reluctant but already caught. And then Kwan begins to talk in Chinese, the language of our childhood ghosts.
I'm telling you the truth, Libby-ah. Ba took another person's name. He stole the fate of a lucky man.
During the war, that's when it happened, when Ba was at National Guangxi University, studying physics. This was in Liangfeng, near Guilin. Ba was from a poor family, but his father sent him to a missionary boarding school when he was a young boy. You didn't have to pay anything, just promise to love Jesus. That's why Ba's English was so good.
I don't remember any of this. I'm just telling you what Li Bin-bin, my auntie, said. Back then, my mother, Ba, and I lived in a small room in Liangfeng, near the university. In the mornings, Ba went to his classes. In the afternoons, he worked in a factory, putting together radio parts. The factory paid him by the number of pieces he finished, so he didn't make very much money. My auntie said Ba was more nimble with his mind than with his fingers. At night, Ba and his classmates threw their money together to buy kerosene for a shared lamp. On full-moon nights they didn't need a lamp. They could sit outside and study until dawn. That's what I also did when I was growing up. Did you know that? Can you see how in China the full moon is both beautiful and a bargain?
One night, when Ba was on his way home from his studies, a drunkard stepped out of an alley and blocked his way. He waved a suit coat in his hands. "This coat," he said, "has been in my family for many generations. But now I must sell it. Look at my face, I'm just a common man from the hundred family names. What use do I have for such fancy clothes?"
Ba looked at the suit coat. It was made of excellent cloth, lined and tailored in a modern style. You have to remember, Libby-ah, this was 1948, when the Nationalists and Communists were fighting over China. Who could afford a coat like this? Someone important, a big official, a dangerous man who got all his money taking bribes from scared people. Our Ba didn't have cotton batting for brains. Hnh! He knew the drunkard had stolen the coat and they both could lose their heads trading in such goods. But once Ba put his fingers on that coat, he was like a small fly caught in a big spider's web. He could not let go. A new feeling ran through him. Ah! To feel the seams of a rich man's coat—to think this was the closest he had ever come to a better life. And then this dangerous feeling led to a dangerous desire, and this desire led to a dangerous idea.
He shouted at the drunkard: "I know this coat is stolen, because I know its owner. Quick! Tell me where you got it or I'll call the police!" The guilty thief dropped the coat and ran off.
Back in our little room, Ba showed my mother the coat. She told me later how he slipped his arms into the sleeves, imagining that the power of its former owner now streamed through his own body. In one pocket he found a pair of thick eyeglasses. He put these on and swung out one arm, and in his mind a hundred people leapt to attention and bowed. He clapped his hands lightly, and a dozen servants rushed from his dreams to bring him food. He patted his stomach, full from his make-believe meal. And that's when Ba felt something else.
Eh, what's this? Something stiff was caught in the lining of the suit coat. My mother used her fine scissors to cut away the threads along the seam. Libby-ah, what they found must have caused their minds to whirl like clouds in a storm. From within the lining fell out a stack of papers— official documents for immigrating to America! On the first page there was a name written in Chinese: Yee Jun. Below that, it was in English: Jack Yee.
You have to imagine, Libby-ah, during civil wartime, papers like these were worth many men's lives and fortunes. In our Ba's trembling hands were certified academic records, a quarantine health certificate, a student visa, and a letter of enrollment to Lincoln University in San Francisco, one year's tuition already paid. He looked inside an envelope: it contained a one-way ticket on American President Lines and two hundred U.S. dollars. And there was also this: a study sheet for passing the immigration examination upon landing.
Oh, Libby-ah, this was very bad business. Don't you see what I'm saying? In those days, Chinese money was worthless. It must be that this man Yee had bought the papers for a lot of gold and bad favors. Did he betray secrets to the Nationalists? Did he sell the names of leaders in the People's Liberation Army?
My mother was scared. She told Ba to throw the coat into the Li River. But Ba had a wild-dog look in his eyes. He said, "I can change my fate. I can become a rich man." He told my mother to go live with her sister in Changmian and wait. "Once I'm in America, I will send for you and our daughter, I promise."
My mother stared at the visa photo of the man Ba would soon become, Yee Jun, Jack Yee. He was an unsmiling thin man, only two years older than Ba. He was not handsome, not like Ba. This man Yee had short hair, a mean face, and he wore thick glasses in front of his cold eyes. You can see a person's heart through his eyes, and my mother said this man Yee looked like the sort of person who would say, "Roll out of the way, you worthless maggot!"
That night my mother watched Ba turn himself into this man Yee by putting on his clothes, cutting his hair. She watched him put on the thick eyeglasses. And when he faced her, she saw his tiny eyes, so cold-looking. He had no warm feelings anymore for my mother. She said it was as though he had become this man Yee, the man in the photo, a man who was arrogant and powerful—eager to be rid of his past, in a hurry to start his new fate.
So that's how Ba stole his name. As to Ba's real name, I don't know what it is. I was so young, and then, as you already know, my mother died. You are lucky no such tragedy like this has happened to you. Later, my auntie refused to tell me Ba's real name because he left her sister. That was my auntie's revenge. And my mother wouldn't tell me either, even after she died. But I've often wondered what his name was. A few times I invited Ba to visit me from the World of Yin. But other yin friends tell me he is stuck somewhere else, a foggy place where people believe their lies are true. Isn't this sad, Libby-ah? If I could learn his real name, I would tell him. Then he could go to the World of Yin, say sorry to my mother, so sorry, and live in peace with our ancestors.
That's why you must go to China, Libby-ah. When I saw that letter yesterday I said to myself, This is your fate waiting to happen! People in Changmian might still remember his name, my auntie for one, I'm sure of it. The man who became Yee, that's what Big Ma, my auntie, always called him. You ask my Big Ma when you go. Ask her what our Ba's real name is.
Ah! What am I saying! You won't know how to ask. She doesn't speak Mandarin. She's so old she never went to school to learn the people's common language. She speaks the Changmian dialect, not Hakka, not Mandarin, something in between, and only people from the village speak that. Also, you have to be very clever how you ask her questions about the past, otherwise she'll chase you away like a mad duck plucking at your feet. I know her ways. What a temper she has!
Don't worry, though, I'm going with you. I already promised. I never forget my promises. You and me, the two of us, we can change our father's name back to its true one. Together we can send him at last to the World of Yin.
And Simon! He must come along too. That way, you can still do the magazine article, get some money to go. Also, we need him to carry the suitcases. I have to bring lots of gifts. I can't go home with empty hands. Virgie can cook for Georgie, her dishes aren't so bad. And Georgie can take care of your dog, no need to pay anyone.
Yes, yes, the three of us together, Simon, you, me. I think this is the most practical, the best way to change your name.
Hey, Libby-ah, what do you think?
## [12
THE BEST TIME TO EAT DUCK EGGS](TheHundredSecretSenses-toc.html#TOC-15)
Kwan doesn't argue to get her own way. She uses more effective methods, a combination of the old Chinese water torture approach and American bait-and-switch.
"Libby-ah," she says. "What month we go China, see my village?"
"I'm not going, remember?"
"Oh, right-right. Okay, what month you think I should go? September, probably still too hot. October, too many tourist. November, not too hot, not too cold, maybe this best time."
"Whatever."
The next day, Kwan says, "Libby-ah, Georgie can't go, not enough vacation time earn up yet. You think Virgie and Ma come with me?"
"Sure, why not? Ask them."
A week later, Kwan says, "Ai-ya! Libby-ah! I already buy three ticket. Now Virgie got new job, Ma got new boyfriend. Both say, Sorry, can't go. And travel agent, she say sorry too, no refund." She gives me a look of agony. "Ai-ya, Libby-ah, what I do?"
I think about it. I could pretend to fall for her routine. But I can't bring myself to do it. "I'll see if I can find someone to go with you," I say instead.
In the evening, Simon calls me. "I was thinking about the China trip. I don't want our breakup to be the reason you miss out on this. Take another writer—Chesnick or Kelly, they're both great on travel pieces. I'll call them for you, if you want."
I'm stunned. He keeps persuading me to go with Kwan, to use her homecoming as a personal angle to the story. I turn over in my head all the permutations of meanings in what he's saying. Maybe there's a chance we can become friends, the kind of buddies we were when we first met. As we continue on the phone, I recall what initially attracted us to each other—the way our ideas grew in logic or hilarity or passion the more we talked. And that's when I feel the grief for what we've lost over the years: the excitement and wonder of being in the world at the same time and in the same place.
"Simon," I say at the end of our two-hour conversation, "I really appreciate this. . . . I think it'd be nice to one day be friends."
"I never stopped being yours," he says.
And at that moment I let go of all restraint. "Well, then, why don't you come to China too?"
ON THE PLANE, I begin to look for omens. That's because Kwan said, when we checked in at the airport: "You, me, Simon—going China! This our fate join together at last."
And I think, Fate as in "the mysterious fate of Amelia Earhart." Fate as in the Latin root of "fatal." It doesn't help matters that the Chinese airline Kwan chose for its discount fare has suffered three crashes in the past six months, two of them while landing in Guilin, where we're headed, after a four-hour stopover in Hong Kong. My confidence in the airline takes another nosedive when we board. The Chinese flight attendants greet us wearing tam-o'-shanters and kilts, an inexplicable fashion choice that makes me question our caretakers' ability to deal with hijackers, loss of engine parts, and unscheduled ocean landings.
As Kwan, Simon, and I struggle down the narrow aisle, I notice there isn't a single white person in coach, unless you count Simon and me. Does this mean something?
Like many of the Chinese people on board, Kwan is gripping a tote bag of gifts in each hand. These are in addition to the suitcase full of presents that has already gone into checked baggage. I imagine tomorrow's television newscast: "An air-pump thermos, plastic food-savers, packets of Wisconsin ginseng—these were among the debris that littered the runway after a tragic crash killed Horatio Tewksbury III of Atherton, who was seated in first class, and four hundred Chinese who dreamed of returning as success stories to their ancestral homeland."
When we see where our assigned seats are, I groan. Center row, smack in the middle, with people on both sides. An old woman sitting at the other end of the aisle stares at us glumly, then coughs. She prays aloud to an unspecified deity that no one will take the three seats next to her, and cites that she has a very bad disease and needs to lie down and sleep. Her coughing becomes more violent. Unfortunately for her, the deity must be out to lunch, because we sit down.
When the drink trolley finally arrives, I ask for relief in the form of a gin and tonic. The flight attendant doesn't understand.
"Gin and tonic," I repeat, and then say in Chinese: "A slice of lemon, if you have it."
She consults her comrade, who likewise shrugs in puzzlement.
_"Ni you_ scotch _meiyou?"_ I try. "Do you have scotch?"
They laugh at this joke.
Surely you have scotch, I want to shout. Look at the ridiculous costumes you have on!
But "scotch" is not a word I've learned to say in Chinese, and Kwan isn't about to assist me. In fact, she looks rather pleased with my frustration and the flight attendants' confusion. I settle for a Diet Coke.
Meanwhile, Simon sits on my other side, playing Flight Simulator on his laptop. "Whoa-whoa- _whoa!_ Shit." This is followed by the sounds of a crash-and-burn. He turns to me. "Captain Bishop here says drinks are on the house."
Throughout the trip, Kwan acts tipsy with happiness. She repeatedly squeezes my arm and grins. For the first time in more than thirty years, she'll be on Chinese soil, in Changmian, the village where she lived until she was eighteen. She'll see her aunt, the woman she calls Big Ma, who brought her up and, according to Kwan, horribly abused her, pinching her cheeks so hard she left her with crescent-shaped scars.
She'll also be reunited with old schoolmates, at least those who survived the Cultural Revolution, which started after she left. She's looking forward to impressing her friends with her English, her driver's license, the snapshots of her pet cat sitting on the floral-patterned sofa she bought recently at a warehouse sale—"fifty percent off for small hole, maybe no one even see it."
She talks of visiting her mother's grave, how she'll make sure it's swept clean. She'll take me to a small valley where she once buried a box filled with treasures. And because I am her darling sister, she wants to show me her childhood hiding place, a limestone cave that contains a magic spring.
The trip presents a number of firsts for me as well. The first time I've gone to China. The first time since I was a child that Kwan will be my constant companion for two weeks. The first time that Simon and I will travel together and sleep in separate rooms.
Now squished into my seat between Simon and Kwan, I realize how crazy it is that I am going—the physical torture of being in planes and airports for almost twenty-four hours, the emotional havoc of going with the very two people who are the source of my greatest heartaches and fears. And yet for the sake of my heart, that's what I have to do. Of course, I have pragmatic reasons for going—the magazine article, finding my father's name. But my main motivation is fear of regret. I worry that if I didn't go, one day I'd look back and wonder, What if I had?
Perhaps Kwan is right. Fate is the reason I'm going. Fate has no logic, you can't argue with it any more than you can argue with a tornado, an earthquake, a terrorist. Fate is another name for Kwan.
WE'RE TEN HOURS from China. My body is already confused as to whether it's day or night. Simon is snoozing, I haven't slept a wink, and Kwan is waking up.
She yawns. In an instant, she's alert and restless. She fidgets with her pillows. "Libby-ah, what you thinking?"
"Oh, you know, business." Before the trip, I made an itinerary and a checklist. I've taken into account jet lag, orientation, location scouting, the possibility that the only lighting available will be fluorescent blue. I've penciled in reminders to get shots of small grocery stores and big supermarkets, fruit stands and vegetable gardens, stoves and cooking utensils, spices and oils. I've also fretted many nights over logistics and budget. The distance to Changmian is a major problem, a three- or four-hour ride from Guilin, according to Kwan. The travel agent wasn't even able to find Changmian on a map. He has us booked into a hotel in Guilin, two rooms at sixty dollars each a night. There might be places that are cheaper and closer, but we'll have to find them after we arrive.
"Libby-ah," Kwan says. "In Changmian, things maybe not too fancy."
"That's fine." Kwan has already told me that the dishes are simple, similar to what she cooks, not like those in an expensive Chinese restaurant. "Actually," I reassure her, "I don't want to take pictures of fancy stuff. Believe me, I'm not expecting champagne and caviar."
"Cavi-ah, what that?"
"You know, fish eggs."
"Oh! Have, _have._ " She looks relieved. "Cavi-egg, crab egg, shrimp egg, chicken egg—all have! Also, thousand-year duck egg. Course, not really thousand year, only one, two, three year most . . . Wah! What I thinking! I know where find you duck egg older than that. Long time ago, I hide some."
"Really?" This sounds promising, a nice detail for the article. "You hid them when you were a girl?"
"Until I twenty."
"Twenty? . . . You were already in the States then."
Kwan smiles secretively. "Not _this_ time twenty. _Last_ time." She leans her head against the seat. "Duck egg—ahh, so good . . . Miss Banner, she don't like too much. Later, starving time come, eat anything, rat, grasshopper, cicada. She think thousand-year _yadan_ taste better than eat those. . . . When we in Changmian, Libby-ah, I show you where hide them. Maybe some still there. You and me go find, ah?"
I nod. She looks so damn happy. For once, her imaginary past does not bother me. In fact, the idea of searching for make-believe eggs in China sounds charming. I check my watch. Another twelve hours and we'll be in Guilin.
"Mmm," Kwan murmurs. _"Yadan . . ."_
I can tell that Kwan is already there, in her illusory world of days gone by.
Duck eggs, I loved them so much I became a thief. Before breakfast, every day except Sunday, that's when I stole them. I wasn't a terrible thief, not like General Cape. I took only what people wouldn't miss, one or two eggs, that sort of thing. Anyway, the Jesus Worshippers didn't want them. They liked the eggs of chickens better. They didn't know duck eggs were a great luxury—very expensive if you bought them in Jintian. If they knew how much duck eggs cost, they'd want to eat them all the time. And then what? Too bad for me!
To make thousand-year duck eggs, you have to start with eggs that are very, very fresh, otherwise, well, let me think . . . otherwise . . . I don't know, since I used only fresh ones. Maybe the old ones have bones and beaks already growing inside. Anyway, I put these very fresh eggs into a jar of lime and salt. The lime powder I saved from washing clothes. The salt was another matter, not cheap like today. Lucky for me, the foreigners had lots. They wanted their food to taste as if it were dipped in the sea. I liked salty things too, but not _everything_ salty. When they sat down to eat, they took turns saying "Please pass the salt," and added even more.
I stole the salt from the cook. Her name was Ermei, Second Sister, one daughter too many of a family with no sons. Her family gave her to the missionaries so they wouldn't have to marry her off and pay a dowry. Ermei and I had a little back-door business. The first week, I gave her one egg. She then poured salt into my empty palms. The next week, she wanted two eggs for the same amount of salt! That girl knew how to bargain.
One day, Dr. Too Late saw our exchange. I walked to the alleyway where I did the wash. When I turned around, there he was, pointing to the little white mound lying in the nest of my palms. I had to think fast. "Ah, this," I said. "For stains." I was not lying. I needed to stain the eggshells. Dr. Too Late frowned, not understanding my Chinese. What could I do? I dumped all that precious salt into a bucket of cold water. He was still watching. So I pulled something from the basket of ladies' private things, threw that in the bucket, and began to scrub. "See?" I said, and held up a salty piece of clothing. Wah! I was holding Miss Mouse's panties, stained at the bottom with her monthly blood! Dr. Too Late—ha, you should have seen his face! Redder than those stains. After he left, I wanted to cry for having spoiled my salt. But when I fished out Miss Mouse's panties—ah?—I saw I'd been telling the truth! That bloodstain, it was gone! It was a Jesus miracle! Because from that day on, I could help myself to as much salt as I needed, one handful for stains, one handful for eggs. I didn't have to go to Ermei through the back door. But every now and then, I still gave her an egg.
I put the lime, salt, and eggs into earthen jars. The jars I got from a one-eared peddler named Zeng in the public lane just outside the alleyway. One egg traded for a jar that was too leaky for oil. He always had plenty of cracked jars. This made me think that man was either very clumsy, or crazy about duck eggs. Later, I learned he was crazy for me! It's true! His one ear, my one eye, his leaky jars, my tasty eggs—maybe that's why he thought we were a good match. He didn't say he wanted me to become his wife, not in so many words. But I knew he was thinking this, because one time he gave me a jar that was not even cracked. And when I pointed this out to him, he picked up a stone, knocked a little chip off the mouth of the jar, and gave it back to me. Anyway, that's how I got jars and a little courtship.
After many weeks, the lime and the salt soaked through the eggshells. The whites of the eggs became firm green, the yellow yolks hard black. I knew this because I sometimes ate one to be certain the others were ready to go into their mud coats. Mud, I didn't have to steal that. In the Ghost Merchant's garden I could mix plenty. While the mud-coated eggs were still wet, I wrapped them in paper, pages torn from those pamphlets called "The Good News." I stuck the eggs into a small drying oven I had made out of bricks. I didn't steal the bricks. They had fallen out of the wall and were cracked. Along every crack, I smeared glue squeezed from a sticky poisonous plant. That way the sun could flow through the cracks, the bugs got stuck and couldn't eat my eggs. The next week, when the clay coats were hard, I put my eggs once more in the curing jar. I buried them in the northwest corner of the Ghost Merchant's garden. Before the end of my life, I had ten rows of jars, ten paces long. That's where they still might be. I'm sure we didn't eat them all. I saved so many.
To me, a duck egg was too good to eat. That egg could have become a duckling. That duckling could have become a duck. That duck could have fed twenty people in Thistle Mountain. And in Thistle Mountain, we rarely ate a duck. If I ate an egg—and sometimes I did—I could see twenty hungry people. So how could I feel full? If I hungered to eat one, but saved it instead, this satisfied me, a girl who once had nothing. I was thrifty, not greedy. As I said, every now and then I gave an egg to Ermei, to Lao Lu as well.
Lao Lu saved his eggs too. He buried them under the bed in the gatehouse, where he slept. That way, he said, he could dream about tasting them one day. He was like me, waiting for the best time to eat those eggs. We didn't know the best time would later be the worst.
ON SUNDAYS, the Jesus Worshippers always ate a big morning meal. This was the custom: long prayer, then chicken eggs, thick slices of salty pork, corn cakes, watermelon, cold water from the well, then another long prayer. The foreigners liked to eat cold and hot things together, very unhealthy. The day that I'm now talking about, General Cape ate plenty. Then he stood up from the table, made an ugly face, and announced he had a sour stomach, too bad he couldn't visit God's House that morning. That's what Yiban told us.
So we went to the Jesus meeting, and while I was sitting on the bench, I noticed Miss Banner could not stop tapping her foot. She seemed anxious and happy. As soon as the service was over, she picked up her music box and went to her room.
During the noonday meal of cold leftovers, General Cape didn't come to the dining room. Neither did Miss Banner. The foreigners looked at his empty chair, then hers. They said nothing, but I knew what they were thinking, mm-hmm. Then the foreigners went to their rooms for the midday nap. Lying on my straw mat, I heard the music box playing that song I had grown to hate so much. I heard Miss Banner's door open, then close. I put my hands over my ears. But in my mind I could see her rubbing Cape's sour stomach. Finally, the song stopped.
I awoke hearing the stableman shout as he ran along the passageway: "The mule, the buffalo, the cart! They're gone." We all came out of our rooms. Then Ermei ran from the kitchen and cried: "A smoked pork leg and a sack of rice." The Jesus Worshippers were confused, shouting for Miss Banner to come and change the Chinese words into English. But her door stayed closed. So Yiban told the foreigners what the stableman and the cook had said. Then all the Jesus Worshippers flew to their rooms. Miss Mouse came out, crying and pulling at her neck; she had lost her locket with the hair of her dead sweetheart. Dr. Too Late couldn't find his medicine bag. For Pastor and Mrs. Amen, it was a silver comb, a golden cross, and all the mission money for the next six months. Who had done such a thing? The foreigners stood like statues, unable to speak or move. Maybe they were wondering why God let this happen on the day they worshipped him.
By this time, Lao Lu was banging on General Cape's door. No answer. He opened the door, looked in, then said one word: Gone! He knocked on Miss Banner's door. Same thing, gone.
Everyone began to talk all at once. I think the foreigners were trying to decide what to do, where to look for those two thieves. But now they had no mule, no buffalo cow, no cart. Even if they did have them, how would they know where to look? Which way did Cape and Miss Banner go? To the south into Annam? To the east along the river to Canton? To Guizhou Province, where wild people lived? The nearest _yamen_ for reporting big crimes was in Jintian, many hours' walking distance from Changmian. And what would the _yamen_ official do when he heard that the foreigners had been robbed by their own kind? Laugh ha-ha-ha.
That evening, during the hour of insects, I sat in the courtyard, watching the bats as they chased after mosquitoes. I refused to let Miss Banner float into my mind. I was saying to myself, "Nunumu, why should you waste one thought on Miss Banner, a woman who favors a traitor over a loyal friend? Nunumu, you remember from now on, foreigners cannot be trusted." Later I lay in my room, still not thinking about Miss Banner, refusing to give her one piece of my worry or anger or sadness. Yet something leaked out anyway, I don't know how. I felt a twist in my stomach, a burning in my chest, an ache in my bones, feelings that ran up and down my body, trying to escape.
The next morning was the first day of the week, time to wash clothes. While the Jesus Worshippers were having a special meeting in God's House, I went into their rooms to gather their dirty clothes. Of course, I didn't bother with Miss Banner's room. I walked right past it. But then my feet started walking backward and I opened her door. The first thing I saw was the music box. I was surprised. Must be she thought it was too heavy for her to carry. Lazy girl. I saw her dirty clothes lying in the basket. I looked in her wardrobe closet. Her Sunday dress and shoes were gone, also her prettiest hat, two pairs of gloves, the necklace with a woman's face carved on orange stone. Her stockings with the hole in one heel, they were still there.
And then I had a bad thought and a good plan. I wrapped a dirty blouse around the music box and put it in the basket of clothes. I carried this down the passageway, through the kitchen, then along the hall to the open alleyway. I walked through the gate into the Ghost Merchant's garden. Along the northwest wall, where I kept my duck eggs, that's where I dug another hole and buried the box and all memories of Miss Banner.
As I was patting dirt over this musical grave, I heard a low sound, like a frog: "Wa- _ren_! Wa- _ren_!" I walked along the path, and above the crunch-crunch of leaves I heard the sound again, only now I knew it was Miss Banner's voice. I hid behind a bush and looked up at the pavilion. Wah! There was Miss Banner's ghost! Her hair, that's what made me think this, it was wild-looking, flowing to her waist. I was so scared I fell against the bush, and she heard my noise.
"Wa- _ren_? Wa- _ren_?" she called, as she ran down the pathway with a wild, lost look on her face. I was crawling away as fast as I could. But then I saw her Sunday shoes in front of me. I looked up. I knew right away she wasn't a ghost. She had many mosquito bites on her face, her neck, her hands. If there had also been _ghost_ mosquitoes out there, they might have done that. But I didn't think of that until just now. Anyway, she was carrying her leather bag for running away. She scratched at her itchy face, asked me in a hopeful voice: "The general—has he come back for me?"
So then I knew what had happened. She had been waiting in the pavilion since the day before, listening for every small sound. I shook my head. And I felt both glad and guilty to see misery crawl over her face. She collapsed to the ground, then laughed and cried. I stared at the back of her neck, the bumpy leftovers where mosquitoes had feasted, the proof that her hope had lasted all night long. I felt sorry for her, but also I was angry.
"Where did he go?" I asked. "Did he tell you?"
"He said Canton. . . . I don't know. Maybe he lied about that as well." Her voice was dull, like a bell that is struck but doesn't ring.
"You know he stole food, money, lots of treasures?"
She nodded.
"And still you wanted to go with him?"
She moaned to herself in English. I didn't know what she was saying, but it sounded as if she was pitying herself, sorry she was not with that terrible man. She looked up at me. "Miss Moo, whatever should I do?"
"You didn't respect my opinion before. Why ask me now?"
"The others, they must think I'm a fool."
I nodded. "Also a thief."
She was quiet for a long time. Then she said, "Perhaps I should hang myself—Miss Moo, what do you think?" She began laughing like a crazy person. Then she picked up a rock and placed this in my lap. "Miss Moo, please do me the favor of smashing my head in. Tell the Jesus Worshippers that the devil Cape killed me. Let me be pitied instead of despised." She threw herself on the dirt, crying, "Kill me, please kill me. They'll wish me dead anyway."
"Miss Banner," I said, "you are asking me to be a murderer?"
And she answered: "If you are my loyal friend, you would do me this favor."
Loyal friend! Like a slap in the face! I said to myself, "Who is she to talk about being a loyal friend?" Kill me, Miss Moo! Hnh! I knew what she really wanted—for me to soothe her, tell her how the Jesus Worshippers would not be angry, how they would understand that she'd been fooled by a bad man.
"Miss Banner," I said, choosing my words very carefully, "don't be an even bigger fool. You don't really want me to smash your head. You're pretending."
She answered: "Yes, yes, kill me! I want to die!" She beat her fist on the ground.
I was supposed to persuade her against this idea at least one or two more times, arguing until she agreed, with much reluctance, to live. But instead I said, "Hm. The others will hate you, this is true. Maybe they will even kick you out. Then where will you go?"
She stared at me. Kick her out? I could see this idea running through her mind.
"Let me think about this," I said. After a few moments, I announced in a firm voice: "Miss Banner, I've decided to be your loyal friend."
Her eyes were two dark holes swimming with confusion.
"Sit with your back against this tree," I told her. She didn't move. So I grabbed her arm and dragged her to the tree and pushed her down. "Come, Miss Banner, I'm only trying to help you." I put the hem of her Sunday dress between my teeth and ripped it off.
"What are you doing!" she cried.
"What does it matter?" I said. "Soon you'll be dead anyway." I tore her hem into three pieces of cloth. I used one strip to tie her hands behind the skinny tree trunk. By now she was trembling a lot.
"Miss Moo, please let me explain—" she started to say, but then I tied another strip around her mouth. "Now, even if you must scream," I said, "no one will hear you." She was mumbling uh-uh-uh. I wrapped the other strip over her eyes. "Now you can't see the terrible thing I must do." She began to kick her feet. I warned her: "Ah, Miss Banner, if you struggle like this, I may miss, and smash only your eye or nose. Then I would have to do it again. . . ."
She was making muffled cries, wagging her head and bouncing on her bottom.
"Are you ready, Miss Banner?"
She was making the uh-uh-uh sounds and shaking her head, her whole body, the tree, shaking so hard the leaves started to fall as if it were autumn. "Farewell," I said, then touched her head lightly with my fist. Just as I thought she would, she fainted right away.
What I had done was mean but not terrible. What I did next was kind but a lie. I walked over to a flowering bush. I broke off a thorn and pricked my thumb. I squeezed and dribbled blood onto the front of her dress, along her brow and nose. And then I ran to get the Jesus Worshippers. Oh, how they praised and comforted her. Brave Miss Banner!—tried to stop the General from stealing the mule. Poor Miss Banner!—beaten, then left to die. Dr. Too Late apologized that he had no medicine to put on the bumps on her face. Miss Mouse said it was so sad that Miss Banner had lost her music box. Mrs. Amen made her invalid soup.
When she and I were alone in her room, Miss Banner said, "Thank you, Miss Moo. I don't deserve such a loyal friend." Those were her words, I remember, because I was very proud. She also said, "From now on, I'll always believe you." Just then Yiban entered the room without knocking. He threw a leather bag on the floor. Miss Banner gasped. It was her bag of clothes for running away. Now her secret had been discovered. All my meanness and kindness were for nothing.
"I found this in the pavilion," he said. "I believe it belongs to you. It contains your hat, also some gloves, a necklace, a lady's hairbrush." Yiban and Miss Banner stared at each other a long time. Finally he said, "Lucky for you, the General forgot to take it with him." That's how he let her know that he too would keep her pitiful secret.
All that week as I did my work, I asked myself, Why did Yiban save Miss Banner from disgrace? She had never been his friend, not like me. I thought about that time I pulled Miss Banner from the river. When you save a person's life, that person becomes a part of you. Why is that? And then I remembered that Yiban and I had the same kind of lonely heart. We both wanted someone to belong to us.
Soon Yiban and Miss Banner were spending many long hours together. Mostly they spoke in English, so I had to ask Miss Banner what they said. Oh, she told me, nothing very important: their life in America, their life in China, what was different, what was better. I felt jealous, knowing she and I had never talked about these not very important things.
"What is better?" I asked.
She frowned and searched her mind. I guessed she was trying to decide which of the many Chinese things she loved should be mentioned first. "Chinese people are more polite," she said, then thought some more. "Not so greedy."
I waited for her to continue. I was sure she would say that China was more beautiful, that our thinking was better, our people more refined. But she did not say these things. "Is there anything better in America?" I asked.
She thought a little bit. "Oh . . . comfort and cleanliness, stores and schools, walkways and roadways, houses and beds, candies and cakes, games and toys, tea parties and birthdays, oh, and big loud parades, lovely picnics on the grass, rowing a boat, putting a flower in your hat, wearing pretty dresses, reading books, and writing letters to friends . . ." On and on she went, until I felt myself growing small and dirty, ugly, dumb, and poor. Often I have not liked my situation. But this was the first time I had this feeling of not liking myself. I was sick with envy— not for the American things she mentioned, but that she could tell Yiban what she missed and he could understand her old desires. He belonged to her in ways that I could not.
"Miss Banner," I asked her, "you feel something for Yiban Johnson, ah?"
"Feel? Yes, perhaps. But just as a friend, though not as good a friend as you. Oh! And not with the feeling between a man and woman—no, no, _no!_ After all, he's Chinese, well, not completely, but half, which is almost worse. . . . Well, in our country, an American woman can't possibly . . . What I mean is, such romantic friendships would _never_ be allowed."
I smiled, all my worries put to rest.
Then, for no reason, she began to criticize Yiban Johnson. "I must tell you, though, he's awfully serious! No sense of humor! So gloomy about the future. China is in trouble, he says, soon even Changmian will not be safe. And when I try to cheer him up, tease him a little, he won't laugh. . . ." For the rest of the afternoon, she criticized him, mentioning all his tiny faults and the ways she could change them. She had so many complaints about him that I knew she liked him better than she said. _Not_ just a friend.
The next week, I watched them sitting in the courtyard. I saw how he learned to laugh. I heard the excited voices of boy-girl teasing. I knew something was growing in Miss Banner's heart, because I had to ask many questions to find out what it was.
I'll tell you something, Libby-ah. What Miss Banner and Yiban had between them was love as great and constant as the sky. She told me this. She said, "I have known many kinds of love before, never this. With my mother and brothers, it was tragic love, the kind that leaves you aching with wonder over what you might have received but did not. With my father, I had uncertain love. I loved him, but I don't know if he loved me. With my former sweethearts, I had selfish love. They gave me only enough to take back what they wanted from me.
"Now I am content," Miss Banner said. "With Yiban, I love and am loved, fully and freely, nothing expected, more than enough received. I am like a falling star who has finally found her place next to another in a lovely constellation, where we will sparkle in the heavens forever."
I was happy for Miss Banner, sad for myself. Here she was, speaking of her greatest joy, and I did not understand what her words meant. I wondered if this kind of love came from her American sense of importance and had led to conclusions that were different from mine. Or maybe this love was like an illness—many foreigners became sick at the slightest heat or cold. Her skin was now often flushed, her eyes shiny and big. She was forgetful of time passing. "Oh, is it that late already?" she often said. She was also clumsy and needed Yiban to steady her as she walked. Her voice changed too, became high and childlike. And at night she moaned. Many long hours she moaned. I worried that she had caught malaria fever. But in the morning, she was always fine.
Don't laugh, Libby-ah. I had never seen this kind of love in the open before. Pastor and Mrs. Amen were not like this. The boys and girls of my old village never acted like this, not in front of other people, at least. That would have been shameful—showing you care more for your sweetheart than for all your family, living and dead.
I thought that her love was another one of her American luxuries, something Chinese people could not afford. For many hours each day, she and Yiban talked, their heads bent together like two flowers reaching for the same sun. Even though they spoke in English, I could see that she would start a thought and he would finish it. Then he would speak, stare at her, and misplace his mind, and she would find the words that he had lost. At times, their voices became low and soft, then lower and softer, and they would touch hands. They needed the heat of their skin to match the warmth of their hearts. They looked at the world in the courtyard—the holy bush, a leaf on the bush, a moth on the leaf, the moth he put in her palm. They wondered over this moth as though it were a new creature on earth, an immortal sage in disguise. And I could see that this life she carefully held was like the love she would always protect, never let come to harm.
By watching all these things, I learned about romance. And soon, I too had my own little courtship—you remember Zeng, the one-eared peddler? He was a nice man, not bad-looking, even with one ear. Not too old. But I ask you: How much exciting romance can you have talking about cracked jars and duck eggs?
Well, one day Zeng came to me as usual with another jar. I told him, "No more jars. I have no eggs to cure, none to give you."
"Take the jar anyway," he said. "Give me an egg next week."
"Next week I still won't have any to give you. That fake American general stole the Jesus Worshippers' money. We have only enough food to last until the next boat from Canton comes with Western money."
The next week Zeng returned and brought me the same jar. Only this time, it was filled with rice. So heavy with feelings! Was this love? Is love rice in a jar, no need to give back an egg?
I took the jar. I didn't say, Thank you, what a kind man you are, someday I'll pay you back. I was like—how do you say it?—a _diplomat._ "Zeng-ah," I called as he started to leave. "Why are your clothes always so dirty? Look at all those grease spots on your elbows! Tomorrow you bring your clothes here, I'll wash them for you. If you're going to court me, at least you should look clean."
You see? I knew how to do romance too.
WHEN WINTER CAME, Ermei was still cursing General Cape for stealing the pork leg. That's because all the cured meat was gone, and so was the fresh. One by one, she had killed the pigs, the chickens, the ducks. Every week, Dr. Too Late, Pastor Amen, and Yiban walked many hours down to Jintian to see if the boat from Canton had come, bearing them money. And every week, they walked home with the same long faces.
One time, they returned with blood running down their long faces. The ladies went running toward them, screaming and crying: Mrs. Amen to Pastor Amen, Miss Mouse to Dr. Too Late, Miss Banner to Yiban. Lao Lu and I ran to the well. While the ladies fussed and washed off the blood, Pastor Amen explained what happened and Yiban translated for us.
"They called us devils, enemies of China!"
"Who? Who?" the ladies cried.
"The Taiping! I won't call them God Worshippers anymore. They're madmen, those Taiping. When I said, 'We're your friends,' they threw rocks at me, tried to kill me!"
"Why? Why?"
"Their eyes, because of their eyes!" Pastor shouted more things, then fell to his knees and prayed. We looked at Yiban and he shook his head. Pastor began punching the air with his fists, then prayed again. He pointed to the mission and wailed, prayed more. He pointed to Miss Mouse, who started to cry, patting Dr. Too Late's face, even though there was no more blood to wipe away. He pointed to Mrs. Amen, spit more words out. She stood up, then walked away. Lao Lu and I were like deaf-mutes, still innocent of what he had said.
At night, we went to the Ghost Merchant's garden to find Yiban and Miss Banner. I saw their shadows in the pavilion on top of the little hill, her head on his shoulder. Lao Lu would not go up there, because of the ghost. So I hissed until they heard me. They walked down, holding hands, letting go after they saw me. By the light of a melon slice of moon, Yiban told us the news.
He had talked to a fisherman when he went with Pastor and Dr. Too Late to the river to learn about the arrival of boats. The fisherman told him, "No boats, not now, not soon, maybe never. The British boats choked off the rivers. No coming in, no going out. Yesterday the foreigners fought for God, today for the Manchus. Maybe tomorrow China will break into little pieces and the foreigners will pick them up, sell them along with their opium." Yiban said there was fighting from Suzhou to Canton. The Manchus and foreigners were attacking all the cities ruled by the Heavenly King. Ten-ten thousand Taiping killed, babies and children too. In some places, all a man could see were rotting Taiping bodies; in other cities, only white bones. Soon the Manchus would come to Jintian.
Yiban let us think about this news. "When I told Pastor what the fisherman said, he went to his knees and prayed, just as you saw him do this afternoon. The God Worshippers threw stones at us. Dr. Too Late and I began to run, calling Pastor, but he wouldn't come. Stones hit his back, his arm, a leg, then his forehead. When he fell to the ground, blood and patience ran out of his head. That's when he lost his faith. He cried, 'God, why did you betray me? Why? Why did you send us the fake general, let him steal our hopes?' "
Yiban stopped talking. Miss Banner said something to him in English. He shook his head. So Miss Banner continued. "This afternoon, when you saw him fall to his knees, he again let the bad thoughts spill out of his brain. Only now he had lost not just his faith, but also his mind. He was shouting, 'I hate China! I hate Chinese people! I hate their crooked eyes, their crooked hearts. They have no souls to save.' He said, 'Kill the Chinese, kill them all, just don't let me die with them.' He pointed to the other missionaries and cried, 'Take her, take him, take her.' "
After that day, many things changed, just like my eggs. Pastor Amen acted like a little boy, complaining and crying often, acting stubborn, forgetting who he was. But Mrs. Amen was not angry with him. Sometimes she scolded him, most times she tried to comfort him. Lao Lu said that night she let Pastor curl against her. Now they were like husband and wife. Dr. Too Late let Miss Mouse nurse his wounds long after there was nothing more to heal. And late at night, when everyone was supposed to be asleep but was not, a door would open, then close. I heard footsteps, then Yiban's whispers, then Miss Banner's sighs. I was so embarrassed to hear them that soon after that I dug up her music box and gave it back. I told her, "Look what else General Cape forgot to take."
One by one the servants left. By the time the air was too cold for mosquitoes to come out at night, the only Chinese who remained at the Ghost Merchant's House were Lao Lu and I. I'm not counting Yiban, because I no longer thought he was more Chinese than Johnson. Yiban stayed because of Miss Banner. Lao Lu and I stayed because we still had our duck-egg fortunes buried in the Ghost Merchant's garden. But we also knew that if we left, none of those foreigners would know how to stay alive.
Every day Lao Lu and I searched for food. Since I had once been a poor girl in the mountains, I knew where to look. We poked in the places beneath tree trunks where cicadas slept. We sat in the kitchen at night, waiting for insects and rats to come out for crumbs we couldn't see. We climbed up the mountains and picked wild tea and bamboo. Sometimes we caught a bird that was too old or too stupid to fly away fast enough. In the springtime, we plucked locusts and grasshoppers hatching in the fields. We found frogs and grubs and bats. Bats you have to chase into a small place and keep them flying until they fall from exhaustion. We fried what we caught in oil. The oil I got from Zeng. Now he and I had more to talk about than just cracked jars and eggs—funny things, like the first time I served Miss Banner a new kind of food.
"What's this?" she asked. She put her nose to the bowl, looked and sniffed. So suspicious. "Mouse," I said. She closed her eyes, stood up, and left the room. When the rest of the foreigners demanded to know what I had said, Yiban explained in their language. They all shook their heads, then ate with good appetites. I later asked Yiban what he told them. "Rabbit," he said. "I said Miss Banner once had a rabbit for a pet." After that, whenever the foreigners asked what Lao Lu and I had cooked, I had Yiban tell them, "Another kind of rabbit." They knew not to ask whether we were telling the truth.
I'm not saying we had plenty to eat. You need many kinds of rabbits to feed eight people two or three times a day. Even Mrs. Amen grew thin. Zeng said the fighting was getting worse. We kept hoping one side would win, one side would lose, so we could return to life being better. Only Pastor Amen was happy, babbling like a baby.
One day Lao Lu and I both decided everything had become worse and worse, until now it was the worst. We agreed that this was the best time to eat duck eggs. We argued a little over how many eggs to give each person. This depended on how long Lao Lu and I thought the worst time would last and how many eggs we had to make things better. Then we had to decide whether to give the eggs to people in the morning or at night. Lao Lu said the morning was best, because we could have dreams of eating eggs and have them come true. This, he said, would make us glad if we woke up and discovered we were still alive. So every morning, we gave each person one egg. Miss Banner, oh, she loved those green-skinned eggs—salty, creamy, better than rabbits, she said.
Help me count, Libby-ah. Eight eggs, every day for almost one month, that's what?—two hundred and forty duck eggs. Wah! I made that many! If I sold those today in San Francisco, ah, what a fortune! Actually, I made even more than that. By the middle of summer, the end of my life, I had at least two jars left. The day we died, Miss Banner and I were laughing and crying, saying we should have eaten more eggs.
But how can a person know when she's going to die? If you knew, what would you change? Can you crack open more eggs and avoid regrets? Maybe you'd die with a stomachache.
Anyway, Libby-ah, now that I think of this, I don't have regrets. I'm glad I didn't eat all those eggs. Now I have something to show you. Soon we can dig them up. You and I, we can taste what's left.
## [13
YOUNG GIRLS WISH](TheHundredSecretSenses-toc.html#TOC-16)
My first morning in China, I awake in a dark hotel room in Guilin and see a figure leaning over my bed, staring at me with the concentrated look of a killer. I'm about to scream, when I hear Kwan saying in Chinese, "Sleeping on your side—so _this_ is the reason your posture is so bad. From now on, you must sleep on your back. Also do exercises."
She snaps on the light and proceeds to demonstrate, hands on hips, twisting at the waist like a sixties PE teacher. I wonder how long she's stood by my bed, waiting for me to waken so she can present her latest bit of unsolicited advice. Her bed is already made.
I look at my watch and say in a grumpy voice, "Kwan, it's only five in the morning."
"This is China. Everyone else is up. Only you're asleep."
"Not anymore."
We've been in China less than eight hours, and already she's taking control of my life. We're on her terrain, we have to go by her rules, speak her language. She's in Chinese heaven.
Snatching my blankets, she laughs. "Libby-ah, hurry and get up. I want to go see my village and surprise everyone. I want to watch Big Ma's mouth fall open and hear her words of surprise: 'Hey, I thought I chased you away. Why are you back?' "
Kwan pushes open the window. We're staying at the Guilin Sheraton, which faces the Li River. Outside it's still dark. I can hear the _trnnng! trnnng!_ of what sounds like a noisy pachinko parlor. I go to the window and look down. Peddlers on tricycle carts are ringing their bells, greeting one another as they haul their baskets of grain, melons, and turnips to market. The boulevard is bristling with the shadows of bicycles and cars, workers and schoolchildren—the whole world chirping and honking, shouting and laughing, as though it were the middle of the day. On the handlebar of a bicycle dangle the gigantic heads of four pigs, roped through the nostrils, their white snouts curled in death grins.
"Look." Kwan points down the street to a set of stalls lit by low-watt bulbs. "We can buy breakfast there, cheap and good. Better than paying nine dollars each for hotel food—and for what? Doughnut, orange juice, bacon, who wants it?"
I recall the admonition in our guidebooks to steer clear of food sold by street vendors. "Nine dollars, that's not much," I reason.
"Wah! You can't think this way anymore. Now you're in China. Nine dollars is lots of money here, one week's salary."
"Yeah, but cheap food might come with food poisoning."
Kwan gestures to the street. "You look. All those people there, do they have food poisoning? If you want to take pictures of Chinese food, you have to taste real Chinese food. The flavors soak into your tongue, go into your stomach. The stomach is where your true feelings are. And if you take photos, these true feelings from your stomach can come out, so that everyone can taste the food just by looking at your pictures."
Kwan is right. Who am I to begrudge carrying home a few parasites? I slip some warm clothes on and go into the hallway to knock on Simon's door. He answers immediately, fully dressed. "I couldn't sleep," he admits.
In five minutes, the three of us are on the sidewalk. We pass dozens of food stalls, some equipped with portable propane burners, others with makeshift cooking grills. In front of the stalls, customers squat in semicircles, dining on noodles and dumplings. My body is jittery with exhaustion and excitement. Kwan chooses a vendor who is slapping what look like floury pancakes onto the sides of a blazing-hot oil drum. "Give me three," she says in Chinese. The vendor pries the cooked pancakes off with his blackened bare fingers, and Simon and I yelp as we toss the hot pancakes up and down like circus jugglers.
"How much?" Kwan opens her change purse.
"Six yuan," the pancake vendor tells her.
I calculate the cost is a little more than a dollar, dirt cheap. By Kwan's estimation, this is tantamount to extortion. "Wah!" She points to another customer. "You charged him only fifty fen a pancake."
"Of course! He's a local worker. You three are tourists."
"What are you saying! I'm also local."
"You?" The vendor snorts and gives her a cynical once-over. "From where, then?"
"Changmian."
His eyebrows rise in suspicion. "Really, now! Who do you know in Changmian?"
Kwan rattles off some names.
The vendor slaps his thigh. "Wu Ze-min? You know Wu Ze-min?"
"Of course. As children, we lived across the lane from each other. How is he? I haven't seen him in over thirty years."
"His daughter married my son."
"Nonsense!"
The man laughs. "It's true. Two years ago. My wife and mother opposed the match—just because the girl was from Changmian. But they have old countryside ideas, they still believe Changmian is cursed. Not me, I'm not superstitious, not anymore. And now a baby's been born, last spring, a girl, but I don't mind."
"Hard to believe Wu Ze-min's a grandfather. How is he?"
"Lost his wife, this was maybe twenty years ago, when they were sent to the cowsheds for counterrevolutionary thinking. They smashed his hands, but not his mind. Later he married another woman, Yang Ling-fang."
"That's not possible! She was the little sister of an old schoolmate of mine. I can't believe it! I still see her in my mind as a tender young girl."
"Not so tender anymore. She's got _jiaoban_ skin, tough as leather, been through plenty of hardships, let me tell you."
Kwan and the vendor continue to gossip while Simon and I eat our pancakes, which are steaming in the morning chill. They taste like a cross between focaccia and a green-onion omelet. At the end of our meal, Kwan and the vendor act like old friends, she promising to send greetings to family and comrades, he advising her on how to hire a driver at a good price.
"All right, older brother," Kwan says, "how much do I owe you?"
"Six yuan."
"Wah! Still six yuan? Too much, too much. I'll give you two, no more than that."
"Make it three, then."
Kwan grunts, settles up, and we leave. When we're half a block away, I whisper to Simon, "That man said Changmian is cursed."
Kwan overhears me. "Tst! That's just a story, a thousand years old. Only stupid people still think Changmian is a bad-luck place to live."
I translate for Simon, then ask, "What kind of bad luck?"
"You don't want to know."
I am about to insist she tell me, when Simon points to my first photo opportunity—an open-air market overflowing with wicker baskets of thick-skinned pomelos, dried beans, cassia tea, chilies. I pull out my Nikon and am soon busy shooting, while Simon jots down notes.
"Plumes of acrid breakfast smoke mingled with the morning mist," he says aloud. "Hey, Olivia, can you do a shot from this direction? Get the turtles, the turtles would be great."
I inhale deeply and imagine that I'm filling my lungs with the very air that inspired my ancestors, whoever they might have been. Because we arrived late the night before, we haven't yet seen the Guilin landscape, its fabled karst peaks, its magical limestone caves, and all the other sites listed in our guidebook as the reasons this is known in China as "the most beautiful place on earth." I have discounted much of the hype and am prepared to focus my lens on the more prosaic and monochromatic aspects of communist life.
No matter which way we go, the streets are chock-full of brightly dressed locals and bloated Westerners in jogging suits, as many people as one might see in San Francisco after a 49ers Super Bowl victory. And all around us is the hubbub of a free-market economy. There they are, in abundance: the barterers of knickknacks; the hawkers of lucky lottery tickets, stock market coupons, T-shirts, watches, and purses with bootlegged designer logos. And there are the requisite souvenirs for tourists—Mao buttons, the Eighteen Lohan carved on a walnut, plastic Buddhas in both Tibetan-thin and roly-poly models. It's as though China has traded its culture and traditions for the worst attributes of capitalism: rip-offs, disposable goods, and the mass-market frenzy to buy what everyone in the world has and doesn't need.
Simon sidles up to me. "It's fascinating and depressing at the same time." And then he adds, "But I'm really glad to be here." I wonder if he's also referring to being with me.
Looking up toward cloud level, we can still see the amazing peaks, which resemble prehistoric shark's teeth, the clichéd subject of every Chinese calendar and scroll painting. But tucked in the gums of these ancient stone formations is the blight of high-rises, their stucco exteriors grimy with industrial pollution, their signboards splashed with garish red and gilt characters. Between these are lower buildings from an earlier era, all of them painted a proletarian toothpaste-green. And here and there is the rubble of prewar houses and impromptu garbage dumps. The whole scene gives Guilin the look and stench of a pretty face marred by tawdry lipstick, gapped teeth, and an advanced case of periodontal disease.
"Boy, oh boy," whispers Simon. "If Guilin is China's most beautiful city, I can't wait to see what the cursed village of Changmian looks like."
We catch up with Kwan. "Everything is entirely different, no longer the same." Her voice seems tinged with nostalgia. She must be sad to see how horribly Guilin has changed over the past thirty years. But then Kwan says in a proud and marveling voice: "So much progress, everything is so much better."
A couple of blocks farther on, we come upon a part of town that screams with more photo opportunities: the bird market. Hanging from tree limbs are hundreds of decorative cages containing singing finches, and exotic birds with gorgeous plumage, punk crests, and fanlike tails. On the ground are cages of huge birds, perhaps eagles or hawks, magnificent, with menacing talons and beaks. There are also the ordinary fowl, chickens and ducks, destined for the stew pot. A picture of them, set against a background of beautiful and better-fated birds, might make a nice visual for the magazine article.
I've shot only half another roll at the bird market, when I see a man hissing at me. "Ssssss!" He sternly motions me to come over. What is he, the secret police? Is it illegal to take pictures here? If he threatens to take my camera away, how much should I offer as a bribe?
The man solemnly reaches underneath a table and brings out a cage. "You like," he says in English. Facing me is a snowy-white owl with milk-chocolate highlights. It looks like a fat Siamese cat with wings. The owl blinks its golden eyes and I fall in love.
"Hey, Simon, Kwan, come here. Look at this."
"One hundred dollar, U.S.," the man says. "Very cheap."
Simon shakes his head and says in a weird combination of pantomime and broken English: "Take bird on plane, not possible, customs official will say stop, not allowed, must pay big fine—"
"How much?" the man asks brusquely. "You say. I give you morning price, best price."
"There's no use bargaining," Kwan tells the man in Chinese. "We're tourists, we can't bring birds back to the United States, no matter how cheap."
"Aaah, who's talking about bringing it back?" the man replies in rapid Chinese. "Buy it today, then take it to that restaurant across the street, over there. For a small price, they can cook it tonight for your dinner."
"Omigod!" I turn to Simon. "He's selling this owl as food!"
"That's disgusting. Tell him he's a fucking goon."
"You tell him!"
"I can't speak Chinese."
The man must think I am urging my husband to buy me an owl for dinner. He zeroes in on me for a closing sales pitch. "You're very lucky I even have _one._ The cat-eagle is rare, very rare," he brags. "Took me three weeks to catch it."
"I don't believe this," I tell Simon. "I'm going to be sick."
Then I hear Kwan saying, "A cat-eagle is not that rare, just hard to catch. Besides, I hear the flavor is ordinary."
"To be honest," says the man, "it's not as pungent as, say, a pangolin. But you eat a cat-eagle to give you strength and ambition, not to be fussy over taste. Also, it's good for improving your eyesight. One of my customers was nearly blind. After he ate a cat-eagle, he could see his wife for the first time in nearly twenty years. The customer came back and cursed me: 'Shit! She's ugly enough to scare a monkey. Fuck your mother for letting me eat that cat-eagle!' "
Kwan laughs heartily. "Yes, yes, I've heard this about cat-eagles. It's a good story." She pulls out her change purse and holds up a hundred-yuan note.
"Kwan, what are you doing?" I cry. "We are _not_ going to eat this owl!"
The man waves away the hundred yuan. "Only American money," he says firmly. "One hundred _American_ dollars."
Kwan pulls out an American ten-dollar bill.
"Kwan!" I shout.
The man shakes his head, refusing the ten. Kwan shrugs, then starts to walk away. The man shouts to her to give him fifty, then. She comes back and holds out a ten and a five, and says, "That's my last offer."
"This is insane!" Simon mutters.
The man sighs, then relinquishes the cage with the sad-eyed owl, complaining the whole time: "What a shame, so little money for so much work. Look at my hands, three weeks of climbing and cutting down bushes to catch this bird."
As we walk away, I grab Kwan's free arm and say heatedly: "There's no way I'm going to let you eat this owl. I don't care if we are in China."
"Shh! Shh! You'll scare him!" Kwan pulls the cage out of my reach. She gives me a maddening smile, then walks over to a concrete wall overlooking the river and sets the cage on top. She meows to the owl. "Oh, little friend, you want to go to Changmian? You want to climb with me to the top of the mountain, let my little sister watch you fly away?" The owl twists his head and blinks.
I almost cry with joy and guilt. Why do I think such bad things about Kwan? I sheepishly tell Simon about my mistake and Kwan's generosity. Kwan brushes off my attempt to apologize.
"I'm going back to the bird market," says Simon, "to take some notes on the more exotic ones they're selling for food. Want to come?"
I shake my head, content to admire the owl Kwan has saved.
"I'll be back in ten or fifteen minutes."
Simon strides off, and I notice how American his swagger looks, especially here on foreign soil. He walks in his own rhythm; he doesn't conform to the crowd.
"See that?" I hear Kwan say. "Over there." She's pointing to a cone-shaped peak off in the distance. "Just outside my village stands a sharp-headed mountain, taller than that one even. We call it Young Girl's Wish, after a slave girl who ran away to the top of it, then flew off with a phoenix who was her lover. Later, she turned into a phoenix, and together, she and her lover went to live in an immortal white pine forest."
Kwan looks at me. "It's a story, just superstition."
I'm amused that she thinks she has to explain.
Kwan continues: "Yet all the girls in our village believed in that tale, not because they were stupid but because they wanted to hope for a better life. We thought that if we climbed to the top and made a wish, it might come true. So we raised little hatchlings and put them in cages we had woven ourselves. When the birds were ready to fly, we climbed to the top of Young Girl's Wish and let them go. The birds would then fly to where the phoenixes lived and tell them our wishes."
Kwan sniffs. "Big Ma told me the peak was named Young Girl's Wish because a crazy girl climbed to the top. But when she tried to fly, she fell all the way down and lodged herself so firmly into the earth she became a boulder. Big Ma said that's why you can see so many boulders at the bottom of that peak—they're all the stupid girls who followed her kind of crazy thinking, wishing for hopeless things."
I laugh. Kwan stares at me fiercely, as if I were Big Ma. "You can't stop young girls from wishing. No! Everyone must dream. We dream to give ourselves hope. To stop dreaming—well, that's like saying you can never change your fate. Isn't that true?"
"I suppose."
"So now you guess what I wished for."
"I don't know. What?"
"Come on, you guess."
"A handsome husband."
"No."
"A car."
She shakes her head.
"A jackpot."
Kwan laughs and slaps my arm. "You guessed wrong! Okay, I'll tell you." She looks toward the mountain peaks. "Before I left for America, I raised three birds, not just one, so I could make three wishes at the top of the peak. I told myself, If these three wishes come true, my life is complete, I can die happy. My first wish: to have a sister I could love with all my heart, only that, and I would ask for nothing more from her. My second wish: to return to China with my sister. My third wish"—Kwan's voice now quavers—"for Big Ma to see this and say she was sorry she sent me away."
This is the first time Kwan's ever shown me how deeply she can resent someone who's treated her wrong. "I opened the cage," she continues, "and let my three birds go free." She flings out her hand in demonstration. "But one of them beat its wings uselessly, drifting in half-circles, before it fell like a stone all the way to the bottom. Now you see, two of my wishes have already happened: I have you, and together we are in China. Last night I realized my third wish would never come true. Big Ma will never tell me she is sorry."
She holds up the cage with the owl. "But now I have a beautiful cat-eagle that can carry with him my new wish. When he flies away, all my old sadnesses will go with him. Then both of us will be free."
Simon comes bounding back. "Olivia, you won't believe the things people here consider food."
We head to the hotel, in search of a car that will take one local, two tourists, and a cat-eagle to Changmian village.
## [14
HELLO GOOD-B YE](TheHundredSecretSenses-toc.html#TOC-17)
By nine, we've procured the services of a driver, an amiable young man who knows how to do the capitalist hustle. "Clean, cheap, fast," he declares in Chinese. And then he makes an aside for Simon's benefit.
"What'd he say?" Simon asks.
"He's letting you know he speaks English."
Our driver reminds me of the slick Hong Kong youths who hang out in the trendy pool halls of San Francisco, the same pomaded hair, his inch-long pinkie nail, perfectly manicured, symbolizing that his lucky life is one without back-breaking work. He flashes us a smile, revealing a set of nicotine-stained teeth. "You call me Rocky," he says in heavily accented English. "Like famous movie star." He holds up a tattered magazine picture of Sylvester Stallone that he has pulled out of his Chinese–English dictionary.
We stash a suitcase of gifts and my extra camera gear into the trunk of his car. The rest of the luggage is at our hotel. Rocky will have to take us back there tonight, unless Kwan's aunt insists that we stay at her place—always a possibility with Chinese families. With this in mind, I've tucked an overnight kit into my camera bag. Rocky opens the door with a flourish, and we climb into a black Nissan, a late-model sedan that curiously lacks seat belts and safety headrests. Do the Japanese think Chinese lives aren't worth saving? "China has either better drivers or no liability lawyers," Simon concludes.
Having learned that we're Americans, Rocky happily assumes we like loud music. He slips in a Eurythmics tape, which was a gift from one of his "excellent American customers." And so with Kwan in the front seat, and Simon, the owl, and me in back, we start our journey to Changmian, blasted by the beat of "Sisters Are Doing It for Themselves."
Rocky's excellent American customers have also taught him select phrases for putting tourists at ease. As we traverse the crowded streets of Guilin, he recites them to us like a mantra: "Where you go? I know it. Jump in, let's go." "Go faster? Too fast? No way, José." "How far? Not far. Too far." "Park car? Wait a sec. Back in flash." "Not lost. No problem. Chill out." Rocky explains that he is teaching himself English so he can one day fulfill his dream and go to America.
"My idea," he says in Chinese, "is to become a famous movie actor, specializing in martial arts. For two years I've practiced tai chi chuan. Of course, I don't expect a big success from the start. Maybe when I first arrive I'll have to take a job as a taxi driver. But I'm hardworking. In America, people don't know how to be as hardworking as we Chinese. We also know how to suffer. What's unbearable to Americans would be ordinary conditions for me. Don't you think that's true, older sister?"
Kwan gives an ambiguous "Hm." I wonder whether she is thinking of her brother-in-law, a former chemist, who immigrated to the States and now works as a dishwasher because he's too scared to speak English lest people think he is stupid. Just then Simon's eyes grow round, and I shout, "Holy shit," as the car nearly sideswipes two schoolgirls holding hands. Rocky blithely goes on about his dream:
"I hear you can make five dollars an hour in America. For that kind of money, I'd work ten hours a day, every day of the year. That's fifty dollars a day! I don't make that much in a month, even with tips." He looks at us in the rearview mirror to see if we caught this hint. Our guidebook says tipping in China is considered insulting; I figure the book must be out-of-date.
"When I live in America," Rocky continues, "I'll save most of my money, spend only a little on food, cigarettes, maybe the movies every now and then, and of course a car for my taxi business. My needs are simple. After five years, I'll have almost one hundred thousand American dollars. Here that's a half-million yuan, more if I exchange it on the streets. Even if I don't become a movie star in five years, I can still come back to China and live like a rich man." He's grinning with happiness at the prospect. I translate for Simon what Rocky has said.
"What about expenses?" says Simon. "There's rent, gas, utilities, car insurance."
"Don't forget income taxes," I say.
And Simon adds: "Not to mention parking tickets and getting mugged. Tell him most people would probably starve in America on fifty dollars a day."
I'm about to translate for Rocky, when I remember Kwan's story about Young Girl's Wish. You can't stop people from hoping for a better life.
"He'll probably never make it to America," I reason to Simon. "Why spoil his dreams with warnings he'll never need?"
Rocky looks at us through the rearview mirror and gives us a thumbs-up. A second later, Simon grips the front seat, and I shout, "Holy Jesus shit!" We are about to hit a young woman on a bicycle with her baby perched on the handlebar. At the last possible moment, the bicyclist wobbles to the right and out of our way.
Rocky laughs. "Chill out," he says in English. And then he explains in Chinese why we shouldn't worry. Kwan turns around and translates for Simon: "He said in China if driver run over somebody, driver always at fault, no matter how careless other person."
Simon looks at me. "This is supposed to reassure us? Did something get lost in the translation?"
"It doesn't make any sense," I tell Kwan, as Rocky veers in and out of traffic. "A dead pedestrian is a dead pedestrian, no matter whose fault it is."
"Tst! This American thinking," Kwan replies. The owl swings his head and stares at me, as if to say, Wise up, gringa, this is China, your American ideas don't work here. "In China," Kwan goes on, "you always responsible for someone else, no matter what. You get run over, this my fault, you my little sister. Now you understand?"
"Yeah," Simon says under his breath. "Don't ask questions." The owl pecks at the cage.
We drive by a strip of shops selling rattan furniture and straw hats. And then we're in the outskirts of town, both sides of the road lined with mile after mile of identical one-room restaurants. Some are in the stages of being built, their walls layers of brick, mud plaster, and whitewash. Judging from the garish billboard paintings on the front, I guess that all the shops employ the same artist. They advertise the same specialties: orange soda pop and steamy-hot noodle soup. This is competitive capitalism taken to a depressing extreme. Idle waitresses squat outside, watching our car whiz by. What an existence. Their brains must be atrophied from boredom. Do they ever rail against the sheer randomness of their lot in life? It's like getting the free space on the bingo card and nothing else. Simon is furiously jotting down notes. Has he observed the same despair?
"What are you writing?"
"Billions and billions unserved," he answers.
A few miles farther on, the restaurants give way to simple wooden stalls with thatched roofs, and even farther, peddlers without any shelter from the damp chill. They stand by the side of the road, yelling at the top of their lungs, waving their string bags of pomelos, their bottles of homemade hot sauce. We are moving backward in the evolution of marketing and advertising.
As we drive through one village, we see a dozen or so men and women dressed in identical white cotton jackets. Next to them are stools, buckets of water, wooden tool chests, and hand-painted signboards. Being illiterate in Chinese, I have to ask Kwan what the signs say. " 'Expert haircut,' " she reads. "Also each one can drain boil, clip off corns, remove earwax. Two ears same price as one."
Simon is taking more notes. "Whew! How'd you like to be the tenth person offering to remove earwax, when no one is stopping for the first? That's my definition of futility."
I remember an argument we once had, in which I said you couldn't compare your happiness with someone else's unhappiness and Simon said why not. Perhaps we were both wrong. Now, as I watch these people waving at us to stop, I feel lucky I'm not in earwax removal. Yet I'm also afraid that the core of my being, stripped of its mail-order trappings, is no different from that of the tenth person who stands on the road wishing for someone to stop and single her out. I nudge Simon. "I wonder what they hope for, if anything."
He answers mock-cheery: "Hey, the sky's the limit—as long as it doesn't rain."
I imagine a hundred Chinese Icaruses, molding wings out of earwax. You can't stop people from wishing. They can't help trying. As long as they can see sky, they'll always want to go as high as they can.
The stretches between villages and roadside bargains grow longer. Kwan is falling asleep, her head bobbing lower and lower. She half awakens with a snort every time we hit a pothole. After a while, she emits long rhythmical snores, blissfully unaware that Rocky is driving faster and faster down the two-lane road. He routinely passes slower vehicles, clicking his fingers to the music. Each time he accelerates, the owl opens his wings slightly, then settles down again in the cramped cage. I'm gripping my knees, then sucking air between clenched teeth whenever Rocky swings into the left lane to pass. Simon's face is tense, but when he catches me looking at him, he smiles.
"Don't you think we should tell him to slow down?" I say.
"We're fine, don't worry." I take Simon's "don't worry" to be patronizing. But I resist the urge to argue with him. We are now tailgating a truck filled with soldiers in green uniforms. They wave to us. Rocky honks his horn, then swerves sharply to pass. As we go by the truck, I can see an oncoming bus bearing down on us, the urgent blare of its horn growing louder and louder. "Oh my God, oh my God," I whimper. I close my eyes and feel Simon grab my hand. The car jerks back into the right lane. I hear a _whoosh,_ then the blare of the bus horn receding.
"That's it," I say in a tense whisper. "I'm going to tell him to slow down."
"I don't know, Olivia. He might be offended."
I glare at Simon. "What? You'd rather die than be rude?"
He affects an attitude of nonchalance. "They all drive like that."
"So mass suicide makes it okay? What kind of logic is that?"
"Well, we haven't seen any accidents."
The knot of irritation in my throat bursts. "Why do you always think it's best not to say anything? Tell me, who gets to pick up the pieces after the damage is done?"
Simon stares at me, and I can't tell whether he is angry or apologetic. At that moment, Rocky brakes abruptly. Kwan and the owl awake with a flutter of arms and wings. Perhaps Rocky has picked up the gist of our argument—but no, we are now almost at a standstill in bumper-to-bumper traffic. Rocky rolls down the window and sticks out his head. He curses under his breath, then starts punching the car horn with the heel of his hand.
After a few minutes, we can see the source of our delay: an accident, a bad one, to judge from the spray of glass, metal, and personal belongings that litter the road. The smells of spilled gasoline and scorched rubber hang in the air. I am about to say to Simon, "See?" But now our car is inching by a black minivan, belly up, its doors splayed like the broken wings of a squashed insect. The front passenger section is obliterated. There is no hope for anyone inside. A tire lies in a nearby vegetable field. Seconds later, we go by the other half of the impact: a red-and-white public bus. The large front window is smashed, the hound-nosed hood is twisted and smeared with a hideous swath of blood, and the driver's seat is empty, a bad sign. About fifty gawkers, farm tools still in hand, mill around, staring and pointing at various parts of the crumpled bus as if it were a science exhibit. And then we are passing the other side of the bus and I can see a dozen or so injured people, some clutching themselves and bellowing in pain, others lying quietly in shock. Or perhaps they are already dead.
"Shit, I can't believe this," says Simon. "There's no ambulance, no doctors."
"Stop the car," I order Rocky in Chinese. "We should help them." Why did I say that? What can I possibly do? I can barely look at the victims, let alone touch them.
"Ai-ya." Kwan stares at the field. "So many yin people." Yin people? Kwan is saying there are dead people out there? The owl coos mournfully and my hands turn slippery-cold.
Rocky keeps his eyes on the road ahead, driving forward, leaving the tragedy behind us. "We'd be of no use," he says in Chinese. "We have no medicine, no bandages. Besides, it's not good to interfere, especially since you're foreigners. Don't worry, the police will be along soon."
I'm secretly relieved he isn't heeding my instructions.
"You're Americans," he continues, his voice deep with Chinese authority. "You're not used to seeing tragedies. You pity us, yes, because you can later go home to a comfortable life and forget what you've seen. For us, this type of disaster is commonplace. We have so many people. This is our life, always a crowded bus, everyone trying to squeeze in for himself, no air to breathe, no room left for pity."
"Would someone please tell me what's going on!" Simon exclaims. "Why aren't we stopping?"
"Don't ask questions," I snap. "Remember?" Now I'm glad Rocky's dreams of American success will never come true. I want to tell him about illegal Chinese immigrants who are duped by gangs, who languish in prison and then are deported back to China. I'll fill his ear with stories about homeless people, about the crime rate, about people with college degrees who are standing in unemployment lines. Who is he to think his chances of succeeding are any better than theirs? Who is he to assume we know nothing about misery? I'll rip up his Chinese–English dictionary and stuff it in his mouth.
And then I feel literally sick with disgust at myself. Rocky is right. I can't help anyone, not even myself. I weakly ask him to pull over so I can throw up. As I lean out of the car, Simon pats my back. "You're okay, you're going to be fine. I feel queasy too."
When we get back on the open road, Kwan gives Rocky some advice. He solemnly nods, then slows down.
"What'd she say?" Simon asks.
"Chinese logic. If we're killed, no payment. And in the next lifetime, he'll owe us big time."
ANOTHER THREE HOURS PASS. I know we have to be getting close to Changmian. Kwan is pointing out landmarks. "There! There!" she cries huskily, bouncing up and down like a little child. "Those two peaks. The village they surround is called Wife Waiting for Husband's Return. But where is the tree? What happened to the tree? Right there, next to that house, there was a very big tree, maybe a thousand years old."
She scans ahead. "That place there! We used to hold a big market there. But now look, it's just an empty field. And there—that mountain up ahead! That's the one we called Young Girl's Wish. I once climbed all the way to the top."
Kwan laughs, but the next second she seems puzzled. "Funny, now that mountain looks so small. Why is that? Did it shrink, washed down by the rain? Or maybe the peak was worn down by too many girls running up there to make a wish. Or maybe it's because I've become too American and now I see things with different eyes, everything looking smaller, poorer, not as good."
All at once, Kwan shouts to Rocky to turn down a small dirt road we just passed. He makes an abrupt U-turn, knocking Simon and me into each other, and causing the owl to shriek with indignity. Now we are rumbling along a rutted lane, past fields with pillows of moist red dirt. "Turn left, turn left!" Kwan orders. She has her hands clasped in her lap. "Too many years, too many years," she says, as if chanting.
We approach a stand of trees, and then, as soon as Kwan announces, "Changmian," I see it: a village nestled between two jagged peaks, their hillsides a velvety moss-green with folds deepening into emerald. More comes into view: crooked rows of buildings whitewashed with lime, their pitched tile roofs laid in the traditional pattern of dragon coils. Surrounding the village are well-tended fields and mirrorlike ponds neatly divided by stone walls and irrigation trenches. We jump out of the car. Miraculously, Changmian has avoided the detritus of modernization. I see no tin roofs or electrical power lines. In contrast to other villages we passed, the outlying lands here haven't become dumping grounds for garbage, the alleys aren't lined with crumpled cigarette packs or pink plastic bags. Clean stone pathways crisscross the village, then thread up a cleft between the two peaks and disappear through a stone archway. In the distance is another pair of tall peaks, dark jade in color, and beyond those, the purple shadows of two more. Simon and I stare at each other, wide-eyed.
"Can you fucking believe this?" he whispers, and squeezes my hand. I remember other times he has said those same words: the day we went to City Hall to be married, the day we moved into our co-op. And then I think to myself: Happy moments that became something else.
I reach into my bag for my camera. As I look through the viewfinder, I feel as though we've stumbled on a fabled misty land, half memory, half illusion. Are we in Chinese Nirvana? Changmian looks like the carefully cropped photos found in travel brochures advertising "a charmed world of the distant past, where visitors can step back in time." It conveys all the sentimental quaintness that tourists crave but never actually see. There must be something wrong, I keep warning myself. Around the corner we'll stumble on reality: the fast-food market, the tire junkyard, the signs indicating this village is really a Chinese fantasyland for tourists: Buy your tickets here! See the China of your dreams! Unspoiled by progress, mired in the past!
"I feel like I've seen this place before," I whisper to Simon, afraid to break the spell.
"Me too. It's so perfect. Maybe it was in a documentary." He laughs. "Or a car commercial."
I gaze at the mountains and realize why Changmian seems so familiar. It's the setting for Kwan's stories, the ones that filter into my dreams. There they are: the archways, the cassia trees, the high walls of the Ghost Merchant's House, the hills leading to Thistle Mountain. And being here, I feel as if the membrane separating the two halves of my life has finally been shed.
From out of nowhere we hear the din of squeals and cheers. Fifty tiny schoolchildren race toward the perimeter of a fenced-in yard, hailing our arrival. As we draw closer, the children shriek, turn on their heels, and run back to the school building laughing. After a few seconds, they come screaming toward us like a flock of birds, followed by their smiling teacher. They stand at attention, and then, through some invisible signal, shout all together in English, "A-B-C! One-two-three! How are you! Hello good-bye!" Did someone tell them American guests were coming? Did the children practice this for us?
The children wave and we wave back. "Hello good-bye! Hello good-bye!" We continue along the path past the school. Two young men on bicycles slow down and stop to stare at us. We keep walking and round a corner. Kwan gasps. Farther up the path, in front of an arched gateway, stand a dozen smiling people. Kwan puts her hand to her mouth, then runs toward them. When she reaches the group, she grabs each person's hand between her two palms, then hails a stout woman and slaps her on the back. Simon and I catch up with Kwan and her friends. They are exchanging friendly insults.
"Fat! You've grown unbelievably fat!"
"Hey, look at you—what happened to your hair? Did you ruin it on purpose?"
"This is the style! What, have you been in the countryside so long you don't recognize good style?"
"Oh, listen to her, she's still bossy, I can tell."
"You were always the bossy one, not—"
Kwan stops in mid-sentence, transfixed by a stone wall. You would think it's the most fascinating sight she's ever seen.
"Big Ma," she murmurs. "What's happened? How can this be?"
A man in the crowd guffaws. "Ha! She was so anxious to see you she got up early this morning, then jumped on a bus to meet you in Guilin. And now look—you're here, she's there. Won't she be mad!"
Everyone laughs, except Kwan. She walks closer to the wall, calling hoarsely, "Big Ma, Big Ma." Several people whisper, and everyone draws back, frightened.
"Uh-oh," I say.
"Why is Kwan crying?" Simon whispers.
"Big Ma, oh, Big Ma." Tears are streaming down Kwan's cheeks. "You must believe me, this is not what I wished. How unlucky that you died on the day that I've come home." A few women gasp and cover their mouths.
I walk over to Kwan. "What are you saying? Why do you think she's dead?"
"Why is everyone so freaked?" Simon glances about.
I hold up my hand. "I'm not sure." I turn back to her. "Kwan?" I say gently. "Kwan?" But she does not seem to hear me. She is looking tenderly at the wall, laughing and crying.
"Yes, I knew this," she is saying. "Of course, I knew. In my heart, I knew all the time."
IN THE AFTERNOON, the villagers hold an uneasy homecoming party for Kwan in the community hall. The news has spread through Changmian that Kwan has seen Big Ma's ghost. Yet she has not announced this to the village, and since there is no proof that Big Ma has died, there is no reason to call off a food-laden celebration that evidently took her friends days to prepare. During the festivities, Kwan does not brag about her car, her sofa, her English. She listens quietly as her former childhood playmates recount major events of their lives: the birth of twin sons, a railway trip to a big city, and the time a group of student intellectuals was sent to Changmian for reeducation during the Cultural Revolution.
"They thought they were smarter than us," recounts one woman whose hands are gnarled by arthritis. "They wanted us to raise a fast-growing rice, three crops a year instead of two. They gave us special seeds. They brought us insect poison. Then the little frogs that swam in the rice fields and ate the insects, they all died. And the ducks that ate the frogs, they all died too. Then the rice died."
A man with bushy hair shouts: "So we said, 'What good is it to plant three crops of rice that fail rather than two that are successful?' "
The woman with arthritic hands continues: "These same intellectuals tried to breed our mules! Ha! Can you believe it? For two years, every week, one of us would ask them, 'Any luck?' And they'd say, 'Not yet, not yet.' And we'd try to keep our faces serious but encouraging. 'Try harder, comrade,' we'd say. 'Don't give up.' "
We are still laughing when a young boy runs into the hall, shouting that an official from Guilin has arrived in a fancy black car. Silence. The official comes into the hall, and everyone stands. He solemnly holds up the identity card of Li Bin-bin and asks if she belonged to the village. Several people glance nervously at Kwan. She walks slowly toward the official, looks at the identity card, and nods. The official makes an announcement, and a ripple of moans and then wails fills the room.
Simon leans toward me. "What's wrong?"
"Big Ma's dead. She was killed in that bus accident we saw this morning."
Simon and I walk over and each put a hand on one of Kwan's shoulders. She feels so small.
"I'm sorry," Simon stammers. "I . . . I'm sorry you didn't get to see her again. I'm sorry we didn't meet her."
Kwan gives him a teary smile. As Li Bin-bin's closest relative, she has volunteered to perform the necessary bureaucratic ritual of bringing the body back to the village the next day. The three of us are returning to Guilin.
As soon as Rocky sees us, he stubs out his cigarette and turns off the car radio. He must have heard the news. "What a tragedy," he says. "I'm sorry, big sister, I should have stopped. I'm to blame—"
Kwan waves off his apologies. "No one's to blame. Anyway, regrets are useless, always too late."
When Rocky opens the car door, we see that the owl is still in his cage on the backseat. Kwan lifts the cage gently and stares at the bird. "No need to climb the mountain anymore," she says. She sets the cage on the ground, then opens its door. The owl sticks out his head, hops to the edge of the doorway and onto the ground. He twists his head and, with a great flap of wings, takes off toward the peaks. Kwan watches him until he disappears. "No more regrets," she says. And then she slips into the car.
As Rocky warms the engine, I ask Kwan, "When we passed the bus accident this morning, did you see someone who looked like Big Ma? Is that how you knew she'd died?"
"What are you saying? I didn't know she was dead until I saw her yin self standing by the wall."
"Then why did you tell her that you knew?"
Kwan frowns, puzzled. "I knew what?"
"You were telling her you knew, in your heart you knew it was true. Weren't you talking about the accident?"
"Ah," she says, understanding at last. "No, not the accident." She sighs. "I told Big Ma that what _she_ was saying was true."
"What did she say?"
Kwan turns to the window, and I can see the reflection of her stricken face. "She said she was wrong about the story of Young Girl's Wish. She said all my wishes had already come true. She was always sorry she sent me away. But she could never tell me this. Otherwise I wouldn't have left her for a chance at a better life."
I search for some way to console Kwan. "At least you can still see her," I say.
"Ah?"
"I mean as a yin person. She can visit you."
Kwan stares out the car window. "But it's not the same. We can no longer make new memories together. We can't change the past. Not until the next lifetime." She exhales heavily, releasing all her unsaid words.
As our car rumbles along the road, the children in the playground run toward us and press their faces against the slat fence. "Hello good-bye!" they cry. "Hello good-bye!"
## [15
THE SEVENTH DAY](TheHundredSecretSenses-toc.html#TOC-18)
Kwan is devastated, I can tell. She isn't crying, but when I suggested earlier that we simply order hotel room service rather than troop outside for a bargain, she readily agreed.
Simon offers her a few more awkward condolences. He kisses her on the cheek, then leaves the two of us alone in our room. We're dining on lasagne, twelve dollars a plate, wildly extravagant by Chinese standards. Kwan stares at her dinner, her face blank, a windswept plain before the storm. For me, lasagne is comfort food. I'm hoping it will fortify me enough to comfort Kwan.
What should I say? "Big Ma—she was a great lady. We're all going to miss her"? That would be insincere, since Simon and I never met her. And Kwan's stories of Big Ma's mistreating her always sounded to me like material for an _Auntie Dearest_ memoir. Yet here is Kwan, grieving over this vile woman who literally left her with scars. Why do we love the mothers of our lives even if they were lousy caretakers? Are we born with blank hearts, waiting to be imprinted with any imitation of love?
I think about my own mother. Would I be desolate if she died? I feel terrified and guilty even pondering the question. Think about it, though: Would I revisit my childhood to cull happy memories and find they are as rare as ripe blackberries on a well-picked bush? Would I stumble into thorns, stir up the queen wasp surrounded by her adoring drones? Upon my mother's death, would I forgive her, then breathe a sigh of relief? Or would I go to an imaginary dell where my mother is now perfect, attentive and loving, where she embraces me and says, "I'm sorry, Olivia. I was a terrible mother, really shitty. I wouldn't blame you if you never forgave me." That's what I want to hear. I wonder what she would in fact tell me.
"Lasagne," Kwan says out of the blue.
"What?"
"Big Ma ask what we eating. Now she say big regret don't have time try American food."
"Lasagne is Italian."
"Shh! Shh! I know, but you tell her that, then regret don't have time see Italy. Already too many regret."
I lean toward Kwan and say sotto voce, "Big Ma doesn't understand English?"
"Just Changmian dialect and a little heart-talk. After she dead longer, learn more heart-talk, maybe even some English. . . ."
Kwan keeps talking, and I'm glad she's not drowning in grief, because I wouldn't know how to save her.
". . . Yin people, after while, just speak heart-talk. Easier, faster that way. No misconfusion like with words."
"What does heart-talk sound like?"
"I already tell you."
"You did?"
"Many time. Don't just use tongue, lip, teeth for speaking. Use hundred secret sense."
"Oh, right, right." I recall snatches of conversations we've had about this: the senses that are related to primitive instincts, what humans had before their brains developed language and the higher functions—the ability to equivocate, make excuses, and lie. Spine chills and musky scents, goose bumps and blushing cheeks—those are the vocabulary of the secret senses. I think.
"The secret senses," I say to Kwan. "Is it like your hair standing on end means you're afraid?"
"Mean someone you _love_ now afraid."
"Someone you love?"
"Yes, secret sense always between two people. How you can have secret just you know, ah? You hair raise up, you know someone secret."
"I thought you meant that they were secret because people have forgotten they have these senses."
"Ah, yes. People often forget until die."
"So it's a language of ghosts."
"Language of love. Not just honey-sweetheart kind love. Any kind love, mother-baby, auntie-niece, friend-friend, sister-sister, stranger-stranger."
"Stranger? How can you love a stranger?"
Kwan grins. "When you first meet Simon, he stranger, right? First time I meet you, you stranger too. And Georgie! When I first see Georgie, I say myself, 'Kwan, where you know this man from?' You know what? Georgie my sweetheart from last lifetime!"
"Really? Yiban?"
"No, Zeng!"
Zeng? I draw a blank.
And she answers in Chinese: "You know—the man who brought me oil jars."
"Yes, I remember now."
"Wait, Big Ma, I'm telling Libby-ah about my husband." Kwan looks past me. "Yes, you know him—no, not in this lifetime, the last, when you were Ermei, and I gave you duck eggs, and you gave me salt."
As I poke my fork into the lasagne, Kwan is happily chatting away, distracted from grief by her memories of a make-believe past.
The last time I saw Zeng before he became Georgie, that was . . . ah, yes, the day before I died.
Zeng brought me a small bag of dried barley, and some bad news. When I handed him his clean clothes, he gave me back nothing to wash. I was standing near my steaming pots, boiling the clothes.
"No need to worry about what's clean or dirty anymore," he told me. He was looking at the mountains, not at me. Ah, I thought, he's saying our courtship is over. But then he announced: "The Heavenly King is dead."
Wah! This was like hearing a thunderclap when the sky is blue. "How could this happen? The Heavenly King can't die, he's immortal!"
"No longer," Zeng said.
"Who killed him?"
"Died by his own hand, that's what people are saying."
This news was even more shocking than the first. The Heavenly King didn't allow suicide. Now he had killed himself? Now he was admitting he wasn't baby brother to Jesus? How could a Hakka man disgrace his own people that way? I looked at Zeng, his gloomy face. I guessed he had my same feelings. He was Hakka too.
I thought about this as I drew the heavy, wet clothes out of the water. "At least the battles will now end," I said. "The rivers will flow with boats again."
That's when Zeng gave me the third piece of news, even worse than the other two. "The rivers are already flowing, not with boats but with blood." When someone says "not with boats but with blood," you don't just listen and say, "Oh, I see." I had to nudge him to get every piece of information, like begging for a bowl of rice one grain at a time. He was so stingy with words. Little bit by little bit, this is what I learned.
Ten years before, the Heavenly King had sent a tide of death from the mountains to the coast. Blood flowed, millions died. Now the tide was returning. In the port cities, the Manchus had slaughtered all the God Worshippers. They were moving inland, burning down houses, digging up graves, destroying heaven and earth at the same time.
"All dead," Zeng told me. "No one is spared. Not even babies."
When he said that, I saw so many crying babies. "When will they come to our province?" I whispered. "Next month?"
"Oh no. The messenger reached our village only a few paces ahead of death."
"Ai-ya! Two weeks? One? How long?"
"Tomorrow the soldiers will destroy Jintian," he said. "The day after that—Changmian."
All feeling drained from my body. I leaned against the grinding mill. In my head, I could already see the soldiers marching along the road. As I was imagining swords dripping with blood, Zeng asked me to marry him. Actually, he didn't use this word "marry." He said in a rough voice: "Hey, tonight I'm going to the mountains to hide in the caves. You want to come with me or not?"
To you, this may sound clumsy, not so romantic. But if someone offers to save your life, isn't that as good as going to church in a white dress and saying "I do"? If my situation had been different, that's what I would have said: I do, let's go. But I had no room in my mind to think about marriage. I was wondering what would happen to Miss Banner, Lao Lu, Yiban—yes, even the Jesus Worshippers, those white faces of Pastor and Mrs. Amen, Miss Mouse and Dr. Too Late. How strange, I thought. Why should I care what happens to them? We have nothing in common—no language, no ideas, no same feelings about the earth and sky. Yet I could say this about them: Their intentions are sincere. Maybe some of their intentions are not so good to begin with, they lead to bad results. Still, they try very hard. When you know this about a person, how can you not have something in common?
Zeng broke into my thoughts. "You coming or not?"
"Let me consider this awhile," I answered. "My mind is not as quick as yours."
"What's left to consider?" Zeng said. "You want to live, or you want to die? Don't think too much. That makes you believe you have more choices than you do. Then your mind becomes confused." He went over to the bench next to the passageway wall, put his hands behind his head, and lay down.
I threw the wet clothes on the grinding mill. I rolled the stone to squeeze out the water. Zeng was right; I was confused. In one corner of my mind, I was thinking, Zeng is a good man. For the remainder of my life, I may never get another chance like this, especially if I die soon.
Then I went to another corner of my mind: If I go with him, then I'll have no more questions or answers of my own. I can no longer ask myself, Am I a loyal friend? Should I help Miss Banner? What about the Jesus Worshippers? These questions would no longer exist. Zeng would decide what should concern me, what should not. That's how things are between a man and a woman.
My mind was going back and forth, this way, that way. A new life with Zeng? Old loyalty to friends? If I hide in the mountains, will I suffer from fear, then die no matter what? If I stay, will my death be quick? Which life, which death, which way? It was like chasing a chicken, then becoming the chicken being chased. I had only one minute to decide which feeling was strongest. And that was the one I followed.
I looked at Zeng lying on the bench. His eyes were closed. He was a kind person, not too clever, but always honest. I decided to end our courtship the same way I started it. I would be a diplomat and make him think it was his idea to stop.
"Zeng-ah," I called.
He opened his eyes, sat up.
I started hanging the wet clothes. "Why should we run away?" I said. "We're not Taiping followers."
He put his hands on his knees. "Listen to your friend, ah," he said, very patient with me. "The Manchus need only a hint that you're friends with God Worshippers. Look where you live. That's enough for a death sentence."
I knew this, but instead of agreeing, I argued: "What are you saying? The foreigners don't worship the Heavenly King. Many times, I heard them say, 'Jesus has no Chinese baby brother.' "
Zeng huffed at me, as if he had never realized what a stupid girl I was. "Say that to a Manchu soldier, your head will already be rolling on the ground." He sprang to his feet. "Don't waste time talking anymore. Tonight I'm leaving. Are you coming?"
I continued my foolish talk: "Why not wait a little longer? Let's see what really happens. The situation can't be as bad as you think. The Manchus will kill some people here and there, but only a few, to set an example. As for the foreigners, the Manchus won't bother them. They have a treaty. Now that I think of this, maybe it's safer to stay here. Zeng-ah, you come live with us. We have room."
"Live here?" he shouted. "Wah! Maybe I should cut my own throat right now!" Zeng squatted and I could see his mind was bubbling like my laundry pots. He was saying all kinds of impolite things, loud enough for me to hear: "She's an idiot. Only one eye—no wonder she can't see what's the right thing to do!"
"Hey! Who are you to criticize me?" I said. "Maybe a fly has flown into your one ear and filled your head with locust fever." I held up the tip of my little finger and made zigzags in the air. "You hear _zzz-zzz,_ you think clouds of disaster are coming from behind. Scared for no reason."
"No reason!" Zeng shouted. "What's happened to your mind? Have you been living in foreign holy clouds so long you think you're immortal?" He stood up, looked at me with disgust a few moments, then said, "Pah!" He turned around and walked away. Instantly, my heart was hurting. As his voice drifted off, I heard him say, "What a crazy girl! Lost her mind, now she's going to lose her head. . . ."
I kept hanging the laundry, but now my fingers were trembling. How quickly good feelings turn to bad. How easily he was fooled. A tear was burning my eye. I pushed it back. No self-pity. Crying was a weak person's luxury. I started to sing one of the old mountain songs, which one I don't remember now. But my voice was strong and clear, young and sad.
"All right, all right. No more arguing." When I turned around, there Zeng was, a tired look on his face. "We can take the foreigners to the mountains too," he said.
Take them with us! I nodded. As I watched him walk away, he started to sing the boy's answer to my song. This man was smarter than I had thought. What a clever husband he would make. With a good voice too. He stopped and called to me: "Nunumu?"
"Ah!"
"Two hours after the sun goes down, that's when I'll come. Tell everyone to be ready in the main courtyard. Do you understand?"
"Understood!" I shouted.
He walked a few more steps, then stopped again. "Nunumu?"
"Ah!"
"Don't wash any more clothes. The only one left to wear them will be a corpse."
So you see? He was already being bossy, making decisions for me. That's how I knew we were married. That's how he told me _I do._
AFTER ZENG LEFT, I went into the garden and climbed up to the pavilion where the Ghost Merchant died. I looked over the wall and saw the rooftops of many houses, the little pathway leading to the mountains. If you had just come to Changmian for the first time, you might think, Ah, what a beautiful place. So quiet. So peaceful. Maybe I should do my honeymoon here.
But I knew this stillness meant the season of danger was now ending, and disaster would soon begin. The air was thick and damp, hard to breathe. I saw no birds, I saw no clouds. The sky was stained orange and red, as if the bloodshed had already reached into the heavens. I was nervous. I had a feeling that something was crawling on my skin. And when I looked, creeping on my arm was one of the five evils, a centipede, its legs marching in a wave! Wah! I whipped out my arm to shake it off, then crushed that centipede flat as a leaf. Even though it was dead, my foot kept stomping on it, up and down, until it was a dark smear on the stone floor. And still I couldn't get rid of this feeling that something was crawling on my skin.
After a while, I heard Lao Lu ringing the dinner bell. Only then did I come to my senses. In the dining room, I took a seat next to Miss Banner. We didn't have separate tables for Chinese and foreigners anymore, not since I started sharing my duck eggs. Same as usual, Mrs. Amen said the mealtime prayer. Same as usual, Lao Lu brought out a dish of fried grasshoppers, which he said was chopped-up rabbit. I was going to wait until we had finished eating, but my thoughts burst from my mouth: "How can I eat when tomorrow we might die!"
When Miss Banner finished translating my bad news, everyone was quiet for a moment. Pastor Amen leapt out of his chair, held up his arms, and cried to God in a happy voice. Mrs. Amen led her husband back to the table and made him sit down. Then she spoke, and Miss Banner translated her words: "Pastor can't go. You see how he is, still feverish. Out there he would call attention to himself, bring danger to others. We'll stay here. I'm sure the Manchus won't harm us, since we're foreigners."
Was this bravery or stupidity? Maybe she was right and the Manchus would not kill the foreigners. But who could be sure?
Miss Mouse spoke next. "Where is this cave? Do you know how to find it? We might become lost! Who is this man Zeng? Why should we trust him?" She couldn't stop worrying. "It's so dark! We should stay here. The Manchus can't kill us. It's not allowed. We are subjects of the Queen. . . ."
Dr. Too Late ran to Miss Mouse and took her pulse. Miss Banner whispered to me what he said: "Her heart beats too fast. . . . A journey into the mountains would kill her. . . . Pastor and Miss Mouse are his patients. . . . He will stay with them. . . . Now Miss Mouse is crying and Dr. Too Late is holding her hand. . . ." Miss Banner was translating things I could see for myself. That's how dazed she was.
Then Lao Lu spoke up: "I'm not staying. Look at me. Where's my long nose, my pale eyes? I can't hide behind this old face. At least in the mountains there are a thousand caves, a thousand chances. Here I have none."
Miss Banner stared at Yiban. So much fright in her eyes. I knew what she was thinking: That this man she loved looked more Chinese than Johnson. Now that I think about this, Yiban's face was similar to Simon's, sometimes Chinese, sometimes foreign, sometimes mixed up. But that night, to Miss Banner he looked very Chinese. I know this, because she turned to me and said: "What time will Zeng come for us?"
We didn't have wristwatches back then, so I said something like, "When the moon has risen halfway into the night sky," which meant around ten o'clock. Miss Banner nodded, then went to her room. When she came out, she was wearing all her best things: her Sunday dress with the torn hem, the necklace with a woman's face carved into orange stone, gloves of very thin leather, her favorite hairpins. They were tortoiseshell, just like that soap dish you gave me for my birthday. Now you know why I loved it so much. Those were the things she decided to wear in case she died. Me, I didn't care about my clothes, even though that night was supposed to be like my honeymoon. Also, my other trousers and blouse were still wet, hanging in the garden. And they were no better than what I had on.
THE SUN WENT DOWN. The half-moon rose, then climbed even higher. We grew feverish, waiting in the dark courtyard for Zeng to come. To be honest, we didn't have to wait for him. I knew the way into the mountains as well as he did, maybe even better. But I didn't tell the others that.
Finally, we heard a fist pounding on the gate. _Bom! Bom! Bom!_ Zeng was here! Before Lao Lu reached the gate, the sound came again: _Bom! Bom!_ So Lao Lu shouted: "You made us wait, now you can do the same while I take a piss!" He opened one side of the gate, and instantly two Manchu soldiers with swords sprang inside, knocking him to the ground. Miss Mouse gave out one long scream— _"Aaaaahhhhh!"_ —followed by many fast ones, _"Aahh-aahh-aahh!"_ Dr. Too Late put his hand over her noisy mouth. Miss Banner pushed Yiban away, and he crept behind a bush. I did nothing. But in my heart, I was crying, What's happened to Zeng? Where's my new husband?
Just then, someone else walked into the courtyard. Another soldier. This one was of high rank, a foreigner. His hair was short. He had no beard, no cape. But when he spoke—shouting "Nelly!" while tapping his stick—we knew who this traitor thief was. There he stood, General Cape, searching the courtyard for Miss Banner. Did he look sorry for what he had done? Did the Jesus Worshippers run over and beat him with their fists? He held out his arm to Miss Banner. "Nelly," he said once again. She didn't move.
And then everything wrong happened all at once. Yiban came out from behind the bush and walked angrily toward Cape. Miss Banner rushed past Yiban and threw herself into Cape's arms, murmuring, "Wa-ren." Pastor Amen began to laugh. Lao Lu shouted: "The bitch can't wait to fuck the dog!" A sword flashed down— _crack!_ —then again— _whuck!_ And before any of us could think to cry stop, a head came rolling toward me, its lips still formed in a yell. I stared at Lao Lu's head, waiting to hear his usual curse. Why didn't he speak? Behind me, I heard the foreigners, their whimpers and moans. And then a howl rose out of my chest and I threw myself on the ground, trying to bring these two pieces together as Lao Lu again. Useless! I jumped to my feet, glared at Cape, ready to kill and be killed. I took only one step and my legs went soft, as if they had no bones inside. The night grew darker, the air even heavier, as the ground rose up and smashed against my face.
When I opened my eye, I saw my hands and brought them to my neck. My head was there, so was a big bump on the side. Had someone struck me down? Or had I fainted? I looked around. Lao Lu was gone, but the dirt nearby was still wet with his blood. In the next moment, I heard shouts coming from around the other end of the house. I scurried over and hid behind a tree. From there, I could see into the open windows and doors of the dining room. It was like watching a strange and terrible dream. Lamps were burning. Where did the foreigners find the oil? At the small table where the Chinese used to eat sat the two Manchu soldiers and Yiban. In the middle of the foreigner table lay a large shank bone, its blackened meat still smoking from the fire. Who brought this food? General Cape held a pistol in each hand. He lifted one and aimed it at Pastor Amen, who was sitting next to him. The pistol made a loud click but no explosion. Everybody laughed. Pastor Amen then began tearing off pieces of meat with his bare hands.
Soon, Cape barked to his soldiers. They picked up their swords, walked briskly across the courtyard, opened the gate, and went outside. Cape then stood up and bowed to the Jesus Worshippers as if to thank them for being his honored guests. He held out his hand to Miss Banner, and just like emperor and empress they strolled down the corridor until they reached her room. Soon enough, I heard the awful sounds of her music box.
My eye flew back to the dining room. The foreigners were no longer laughing. Miss Mouse had her face pressed against the palms of her hands. Dr. Too Late was comforting her. Only Pastor Amen was smiling as he examined the shank bone. Yiban was already gone.
So many bad thoughts whirled through my head. No wonder the foreigners are called white devils! They had no morals. They couldn't be trusted. When they say to turn the other cheek, they really mean they have two faces, one tricky, the other false. How could I be so stupid as to think they were my friends! And now where was Zeng? How could I have risked his life for theirs?
A door swung open and Miss Banner stepped out, holding a lantern in her hand. She called back to Cape in a teasing voice, then closed the door and walked toward the courtyard. _"Nuli!"_ she cried sharply in Chinese. " _Nuli,_ come! Don't keep me waiting!" Oh, I was mad. Who does she think she's calling slave girl? She was searching for me, turning in circles. My hand scraped along the ground for a stone. But all I found was a pebble, and clutching this tiny weapon, I told myself, This time I'll smash her head in for sure.
I came out from behind the tree. _"Nuwu!"_ I answered back.
As soon as I called her witch, she spun around, the lantern glowing on her face. She still couldn't see me. "So, witch," I said, "you know your name." One of the soldiers opened the gate and asked if anything was the matter. Cut off her head, I expected Miss Banner to say. But instead she answered in a calm voice: "I was calling for my servant."
"You want us to look for her?"
"Ah! No need, I've already found her. See, she's over there." She was pointing to a dark spot at the opposite end of the courtyard. _"Nuli!"_ she shouted to the empty corner. "Quick now, bring me the key to my music box!"
What was she saying? I wasn't there. The soldier stepped back outside, banging the gate closed. Miss Banner turned around and hurried toward me. In a moment, her face was near mine. By lantern light I could see the anguish in her eyes. "Are you still my loyal friend?" she asked in a soft, sad voice. She held up the music box key. Before I could think what she meant, she whispered, "You and Yiban must leave tonight. Let him despise me, otherwise he won't leave. Make sure he is safe. Promise me this." She squeezed my hand. "Promise," she said again. I nodded. Then she opened my fist and saw the pebble in my palm. She took this and replaced it with the key. "What?" she shouted. "You left the key in the pavilion? Stupid girl! Now take this lantern, go to the garden. Don't you dare come back until you find it."
I was so happy to hear her say these meaningless things. "Miss Banner," I whispered. "You come with us—now."
She shook her head. "Then he'll kill us all. After he's gone, we'll find each other." She let go of my hand and walked in the dark, back to her room.
I FOUND YIBAN in the Ghost Merchant's garden, burying Lao Lu.
"You are a good person, Yiban." I covered the dirt with old leaves so the soldiers would not find his grave.
When I finished, Yiban said, "Lao Lu knew how to keep the gate closed to everything except his own mouth."
I nodded, then remembered my promise. So I said in an angry voice: "It's Miss Banner's fault he's dead. Throwing herself in the traitor's arms!" Yiban was staring at his fists. I nudged his arm. "Hey, Yiban, we should leave, run away. Why die for the sins of these foreigners? None of them are any good."
"You're mistaken," Yiban said. "Miss Banner only pretends to give Cape her heart, to save us all." You see how well he knew her? Then you also know how strongly I had to lie.
"Hnh! Pretend!" I said. "I'm sorry to have to tell you the truth. Many times she's told me she wished he would come back for her. Of course she was fond of you, but only half as much as she is fond of Cape. And you know why? You are only half a foreigner! That's how these Americans are. She loves Cape because he is her same kind. You can't easily change ruts that have been carved in mud."
Yiban's fists were still clenched, and his face grew sad, very sad. Lucky for me, I didn't have to tell too many more lies about Miss Banner. He agreed we should leave. But before we did that, I went to the northwest corner and reached into an open jar that had two duck eggs left. There was no time to dig up more. "We'll go to Hundred Caves Mountain," I said. "I know how to find it." I blew out the lantern Miss Banner had given me and handed it to Yiban. Then the two of us slipped through the alleyway gate.
We did not take the path through the village. We crept along the foot of the mountain, where prickly bushes grew. As we began to climb toward the first ridge and its wall, my heart was pounding hard with worry that the soldiers would see us. Even though I was a girl and Yiban a man, I climbed faster. I had my mountain legs. When I reached the archway, I waited for him to catch up. From there my eye searched for the Ghost Merchant's House. It was too dark. I imagined Miss Banner gazing into the night, wondering if Yiban and I were safe. And then I thought of Zeng. Had he seen Cape and his soldiers? Did he flee to the mountains by himself? Just as I thought this, I heard his voice calling me from behind.
"Nunumu?"
"Ah!" I turned around. I saw his shadow at the end of the archway tunnel. How happy I was. "Zeng, there you are! I've been worried to pieces about you. We waited, and then the soldiers came—"
He cut me off. "Nunumu, hurry. Don't waste time talking. Come this way." He was still bossy, no time to say, "Oh, my little treasure, at last I have found you." As I walked through the arch, I let him know I was glad to see him, by complaining in a teasing way: "Hey, when you didn't come, I thought you had changed your mind, taken another woman, one with two eyes." I stepped out of the archway tunnel. Zeng was walking along the ridge beside the wall. He waved for me to follow.
"Don't travel through the valley," he said. "Stay high along the mountain."
"Wait!" I said. "Another person is coming." He stopped. I turned around to see if Yiban was following. And then I heard my new husband say, "Nunumu, tonight the soldiers killed me. Now I will wait for you forever."
"Ai-ya!" I grumbled. "Don't joke like that. Tonight the soldiers killed Lao Lu. I've never seen a more terrible sight—"
At last, Yiban came out of the archway. "Who are you speaking to?" he asked.
"Zeng," I said. "He's here. See?" I turned around. "Zeng? I can't see you. Wave your hand. . . . Hey, where are you? Wait!"
"I will wait for you forever," I heard him whisper in my ear. Ai-ya! That's when I knew Zeng wasn't joking. He was dead.
Yiban came next to me. "What's wrong? Where is he?"
I bit my lips to keep from crying out. "I was mistaken. I saw a shadow, that's all." My eye was burning and I was grateful for the dark. What did it matter if I died now rather than later? If I had not made a promise to Miss Banner, I would have gone back to the Ghost Merchant's House. But now here was Yiban, waiting for me to decide which way to go.
"High along the mountain," I said.
As Yiban and I pushed past bushes and stumbled over rocks, we said nothing to each other. I think he was like me, hurting over people he had lost. He and Miss Banner might one day be together again. No such hope for Zeng and me. But then I heard Zeng say, "Nunumu, how can you decide the future? What about the next lifetime? Can't we marry then?" Wah! Hearing this, I almost tumbled down the mountain. Marry! He used the word "marry"!
"Nunumu," he continued, "before I leave, I'll take you to a cave where you should hide. Use my eyes in the dark."
Right away, I could see through the patch of my blind eye. And before me was a small path, cast with dusk light. The land everywhere else was hidden by night. I turned to Yiban. "Quick now," I said, and I marched ahead as bravely as any soldier.
After several hours, we were standing in front of a bush. As I pulled back the branches, I saw a hole just big enough for one person to squeeze through. Yiban climbed in first. He called back: "It's too shallow. It ends a few paces in."
I was surprised. Why would Zeng take us to such a poor cave? My doubt insulted him. "It's not too shallow," he said. "On the left side are two boulders. Reach down between them." I climbed in and found a cool opening that slanted down.
"This is the right cave," I said to Yiban. "You just didn't look carefully enough. Light the lantern and climb down after me."
That hole was the beginning of a long, twisty passageway with a small stream running along one side. Sometimes the tunnel split into two directions. "Where one way goes up, the other down," Zeng said, "always go lower. Where one has a stream, and the other is dry," Zeng said, "follow the water. Where one is narrow, the other wide," Zeng said, "squeeze in." The farther we went, the cooler the air became, very refreshing.
We turned one corner after another, until we saw a celestial light. What was this? We were in a place, like a palace room, that could have held a thousand people. It was very bright. In the middle of the floor was a lake—with water that glowed. It was a greenish-golden color, and not like light that comes from a candle, a lamp, or the sun. I thought it was the beams of the moon shining through a hole in the world.
Yiban thought it might be a volcano bubbling underneath. Or ancient sea creatures with shining eyes. Or perhaps a star that broke in two, fell to earth, and splashed into the lake.
I heard Zeng say, "Now you can find the rest of the way yourself. You won't get lost."
Zeng was leaving me. "Don't go!" I shouted.
But only Yiban answered back: "I haven't moved."
And then I could no longer see out of my blind eye. I waited for Zeng to speak again. Nothing. Gone like that. No "Good-bye, my little heart-liver. Soon I will meet you in the next world." That's the trouble with yin people. Unreliable! They come when they want, go when they please. After I died, Zeng and I had a big argument about that.
And then I told him what I am telling you now, Big Ma: That with your death, I know too late what I have truly lost.
## [16
BIG MA'S PORTRAIT](TheHundredSecretSenses-toc.html#TOC-19)
I listened to Kwan talk to Big Ma for half the night. Now I'm bleary-eyed. She's as perky as can be.
Rocky is driving us to Changmian in a trouble-prone van. Big Ma's shrouded body is lying across the bench in the back. At every intersection, the van coughs to a stop, belches, and dies. Rocky then jumps out, flings open the hood, and bangs on various metallic parts, while bellowing in Chinese, "Fuck your ancestors, you lazy worm." Miraculously, this incantation works, much to our relief and that of the honking drivers behind us. Inside, the van feels like an icebox; out of consideration to Big Ma and her sad condition, Rocky has kept the heater turned off. Staring out the windows, I can see fog rising off the banks of irrigation ditches. The peaks have blended into the thick mist. This does not look like the beginning of a good day.
Kwan is sitting in the back, chatting loudly to Big Ma's body as though they were girls on their way to school. I am one bench up, and Simon is sitting on the seat behind Rocky, maintaining proletarian camaraderie and, I suspect, keeping an eye on dangerous driving maneuvers. Earlier this morning, after we checked out of the Sheraton and loaded the luggage into the van, I told Simon, "Thank God this will be the last ride we have to make with Rocky." Kwan gave me a horrified look: "Wah! Don't say 'last.' Bad luck to say such a thing." Bad luck or not, at least we won't have to make the daily commute to and from Changmian. We are going to live in the village for the next two weeks, rent-free, courtesy of Big Ma, who, according to Kwan, "invite us to stay her place, even before she die."
Above the metallic rattles of the van, I can hear Kwan bragging to the dead woman: "This sweater, see, it looks like wool, doesn't it? But it's _crylic-ah,_ mm-hmm, _machine wash_." She says "acrylic" and "machine washable" in her version of English, then explains how washing machines and dryers figure into the American judicial system: "In California, you can't hang laundry from your balcony or window, oh no. Your neighbors will call the police for shaming them. America doesn't have as much freedom as you think. So many things are forbidden, you would not believe it. I think some rules are good, though. You can't smoke except in jail. You can't throw an orange peel on the road. You can't let your baby poop on the sidewalk. But some rules are ridiculous. You can't talk in a movie theater. You can't eat too many fatty foods. . . ."
Rocky revs the engine and speeds down the bumpy road. Now I'm concerned not only for Kwan's state of mind but also at the possibility that Big Ma's body will soon go hurtling onto the floor.
"Also, you can't make your children work," Kwan is saying with absolute authority. "I'm telling the truth! Remember how you made me gather twigs and sticks for fuel? Oh yes, I remember. I had to run all over the place in the wintertime, up and down, back and forth, here and there! My poor little fingers, swollen stiff with cold. And then you sold my bundles to other households and kept the money yourself. No, I'm not blaming you, not now. Of course, I know, in those days everyone had to work hard. But in America, they would have put you in jail for treating me this way. Yes, and for slapping my face so many times and pinching my cheeks with your sharp fingernails. You don't remember? See the scars, here on my cheek, two of them, like a rat bite. And now that I'm remembering this, I'm telling you again, I didn't give those moldy rice cakes to the pigs. Why should I lie now? Just as I told you then, the one who stole them was Third Cousin Wu. I know because I watched her cutting off the green mold, one little rice cake at a time. Ask her yourself. She must be dead by now. Ask her why she lied and said I threw them away!"
Kwan is unusually quiet for the next ten minutes, and I figure that she and Big Ma are giving each other the Chinese silent treatment. But then I hear Kwan shout to me in English: "Libby-ah! Big Ma ask me can you take her picture? She say there no good picture of her when she still alive." Before I can answer, Kwan does more yin-speak translation: "This afternoon, she say best time take picture. After I put on best clothes, best shoes." Kwan smiles broadly at Big Ma, then turns to me. "Big Ma say she proud beyond words have such famous photographer in family."
"I'm not famous."
"Don't argue with Big Ma. To her, you famous. That's what matter."
Simon wobbles toward the back and sits next to me, whispering, "You're not actually going to take a picture of a corpse, are you?"
"What am I supposed to say?—'I'm sorry, I don't do dead people, but I can refer you to someone who does'?"
"She might not be very photogenic."
"No kidding."
"You realize this is Kwan's wish for a photo, not Big Ma's."
"Why are you saying things that are completely unnecessary?"
"Just checking, now that we're in China. A lot of weird stuff has already happened, and it's only the second day."
WHEN WE ARRIVE in Changmian, four elderly women snatch our luggage and wave off our protests with laughter and assertions that each is stronger than the three of us combined. Unencumbered, we wend our way through a maze of stone-paved lanes and narrow alleyways to reach Big Ma's house. It is identical to every other house in the village: a one-story walled hut made out of mud bricks. Kwan opens the wooden gate and Simon and I step over the threshold. In the middle of an open-air courtyard, I see a tiny old woman draining water from a hand pump into a bucket. She looks up, first with surprise, then with delight, in seeing Kwan. "Haaaa!" she cries, her open mouth releasing clouds of moist breath. One of her eyes is pinched shut, the other turned outward like that of a frog on the alert for flies. Kwan and the woman grab each other by the arms. They poke at each other's waistlines and then break into rapid Changmian dialect. The old woman gestures toward a crumbling wall, throws a deprecating scowl at her untended fire. She seems to be apologizing for the poor condition of the house and her failure to have ready a banquet and forty-piece orchestra to hail our arrival.
"This Du Lili, my old family friend," Kwan tells Simon and me in English. "Yesterday she gone to mountainside, pick mushrooms. Come back, find out I already come and gone."
Du Lili crinkles her face into an expression of agony, as if she understood this translation of her disappointment. We nod in sympathy.
Kwan continues: "Long time ago we live together, this same house. You speak Mandarin to her. She understand." Kwan turns back to her friend and explains on our behalf: "My little sister, Libby-ah, she speaks a strange kind of Mandarin, American style, her thoughts and sentences running backward. You'll see. And this one here, her husband, Simon, he's like a deaf-mute. English, that's all he speaks. Of course, they're only half Chinese."
"Ahhhhh!" Du Lili's tone suggests either shock or disgust. "Only half! What do they speak to each other?"
"American language," Kwan answers.
"Ahhhhh." Another note of apparent revulsion. Du Lili inspects me as if the Chinese part of my face were going to peel off any second.
"You can understand a little?" she asks me slowly in Mandarin. And when I nod, she complains in more rapid speech: "So skinny! Why are you so skinny? Tst! Tst! I thought people in America ate lots. Is your health poor? Kwan! Why don't you feed your little sister?"
"I try," Kwan protests. "But she won't eat! American girls, they all want to be skinny."
Next, Du Lili gives Simon the once-over: "Oh, like a movie star, this one." She stands on tiptoe for a better view.
Simon looks at me with raised eyebrows. "Translation, please."
"She says you'd make a good husband for her daughter." I wink at Kwan and try to keep a straight face.
Simon's eyes widen. This is a game he and I used to play in the early days of living together. I'd give him bogus translations, and both of us would play out the lie until one of us broke down.
Du Lili takes Simon's hand and leads him inside, saying, "Come, I want to show you something."
Kwan and I follow. "She needs to check your teeth first," I tell Simon. "It's a custom before the betrothal." We find ourselves in an area about twenty feet by twenty feet, which Du Lili calls the central room. It is dark, and sparsely furnished with a couple of benches, a wooden table, and a scattering of jars, baskets, and boxes. The ceiling is peaked. From the rafters hang dried meat and peppers, baskets, and no light fixtures. The floor is made of tamped earth. Du Lili points toward a plain wooden altar table pushed against a back wall. She asks Simon to stand next to her.
"She wants to see if the gods approve of you," I say. Kwan rounds her mouth, and I wink at her.
Tacked above the table are pink paper banners with faded inscriptions. In the middle is a picture of Mao with yellowed tape across his torn forehead. On the left is a cracked gilt frame containing a portrait of Jesus, hands raised to a golden ray of light. And on the right is what Du Lili wants Simon to see: an old calendar photo featuring a Bruce Lee look-alike in ancient warrior costume, guzzling a green-colored soda pop. "See this movie star?" Du Lili says. "I think you look like him— thick hair, fierce eyes, strong mouth, the same, oh, very handsome."
I peer at the photo and then at Simon, who is waiting for my translation. "She says you resemble this criminal who's on China's most-wanted list. Forget the marriage. She's going to collect a thousand yuan for turning you in."
He points to the calendar photo, then to himself, mouthing, "Me?" He shakes his head vigorously and protests in pidgin English: "No, no. Wrong person. Me American, nice guy. This man bad, someone else."
I can't hold on to the façade any longer. I burst out laughing.
"I win," Simon gloats. Kwan translates our silliness to Du Lili. For a few seconds, Simon and I smile at each other. It's the first warm moment we've shared in a long time. At what point in our marriage did our teasing drift into sarcasm?
"Actually, what Du Lili said was, you're as handsome as this movie star."
Simon clasps his hands together and bows, thanking Du Lili. She bows back, glad that he finally understands her compliment.
"You know," I tell him, "for some reason, in this light, you do look, well, different."
"Hmm. How so?" His eyebrows dance flirtatiously.
I feel awkward. "Oh, I don't know," I mumble, my face growing warm. "Maybe you look more Chinese or something." I turn away and pretend to be absorbed in the picture of Mao.
"Well, you know what they say about people who are married, how we become more and more alike over the years."
I keep staring at the wall, wondering what Simon is really thinking. "Look at this," I say, "Jesus right next to Mao. Isn't this illegal in China?"
"Maybe Du Lili doesn't know who Jesus is. Maybe she thinks he's a movie star selling light bulbs."
I am about to ask Du Lili about the Jesus picture, when Kwan spins around and calls to some dark figures standing in the bright doorway. "Come in! Come in!" She becomes all bustle and business. "Simon, Libby-ah, quick! Help the aunties with our luggage." Our elderly bellhops push us aside and with mighty huffs finish dragging in our suitcases and duffel bags, the bottoms of which are spattered with mud.
"Open your purse," Kwan says, and before I can comply she is rifling through my bag. She must be looking for money for a tip. But instead she pulls out my Marlboro Lights and gives the women the whole damn pack. One of the women gleefully passes the pack around, then pockets the rest. The old ladies start puffing away. And then, in a cloud of smoke, they leave.
Kwan drags her suitcase into a dark room on the right. "We sleep here." She motions me to follow. I expect a grim communist bedroom, decor that will match the minimalist look of the rest of the house. But when Kwan opens a window to let in the late-morning sun, I spot an ornately carved marriage bed, enclosed with a shredded canopy of grayed mosquito netting. It is a wonderful antique, almost exactly like one I coveted in a shop on Union Street. The bed is made up in the same way Kwan does hers at home: a sheet pulled taut over the mattress, the pillow and folded quilt stacked neatly at the foot end. "Where did Big Ma get this?" I marvel.
"And this." Simon brushes his hand along a marble-topped dresser, its mirror showing more silver than reflection. "I thought they got rid of all this imperialist furniture during the revolution."
"Oh, those old things." Kwan gives them a dismissive wave, full of pride. "Been in our family long time. During Cultural Revolution time, Big Ma hid them under lots a straw, in shed. That's how everything saved."
"Saved?" I ask. "Then where did our family get them originally?"
"Original, missionary lady give our mother's grandfather, payment for big debt."
"What big debt?"
"Very long story. This happen, oh, one hundred year—"
Simon cuts in. "Could we talk about this later? I'd like to get settled in the other bedroom."
Kwan lets out a derisive snort.
"Oh." Simon's face goes blank. "I take it there's no other bedroom."
"Other bedroom belong Du Lili, only one small bed."
"Well, where are we all going to sleep?" I search the room for an extra mattress, a cushion.
Kwan gestures nonchalantly toward the marriage bed. Simon smiles at me and shrugs in an apologetic manner that is clearly insincere.
"That bed's barely big enough for two people," I say to Kwan. "You and I can sleep there, but we're going to have to find a spare bed for Simon."
"Where you can find spare bed?" She stares at the ceiling, palms up, as if beds might materialize out of thin air.
Panic grows in my throat. "Well, somebody must have an extra mattress pad or _something._ "
She translates this to Du Lili, who also turns her palms up. "See?" Kwan says. "Nothing."
"It's okay, I can sleep on the floor," Simon offers.
Kwan translates this to Du Lili as well, and it elicits chuckles. "You want sleep with bugs?" says Kwan. "Biting spiders? Big rats? Oh yes, many rats here, chew off you finger." She makes chomping-teeth sounds. "How you like that, ah? No. Only way, we three sleep same bed. Anyway, only for two weeks."
"That's not a solution," I reply.
Du Lili looks concerned and whispers to Kwan. Kwan whispers back, tilts her head toward me, then Simon. _"Bu-bu-bu-bu-bu!"_ Du Lili cries, the nos punctuated by rapid shakes of her head. She grabs my arm, then Simon's, and pushes us together as though we were two squabbling toddlers. "Listen, you two hotheads," she scolds in Mandarin. "We don't have enough luxuries for your American kind of foolishness. Listen to your auntie, ah. Sleep in one bed, and in the morning you'll both be warm and happy as before."
"You don't understand," I say.
_"Bu-bu-bu!"_ Du Lili waves off any lame American excuses.
Simon blows out a sigh of exasperation. "I think I'll go take a short walk while you three figure this out. Me, I'm amenable to three's a crowd or rats on the floor. Really, whatever you decide."
Is he angry with me for protesting so much? This isn't my fault, I want to shout. As Simon walks out, Du Lili follows him, scolding him in Chinese: "If you have troubles, you must fix them. You're the husband. She'll listen to your words, but you must be sincere and ask forgiveness. A husband and wife refusing to sleep together! This is not natural."
When we are alone, I glower at Kwan: "You planned this whole thing, didn't you?"
Kwan acts offended. "This not plan. This China."
After a few moments of silence, I grumpily excuse myself. "I have to use the bathroom. Where is it?"
"Down lane, turn left, you see small shed, big pile black ash—"
"You mean there's _no_ bathroom in the house?"
"What I tell you?" Kwan answers, now grinning victoriously. "This China."
WE EAT a proletarian lunch of rice and yellow soybeans. Kwan insisted that Du Lili throw together some simple leftovers. After lunch, Kwan returns to the community hall so she can prepare Big Ma for her portrait session. Simon and I take off in different directions to explore the village. The route I choose leads to an elevated narrow stone lane that cuts through soggy fields. In the distance, I see ducks waddling in a line parallel to the horizon. Are Chinese ducks more orderly than American ones? Do their quacks differ? I click off a half-dozen shots with my camera, so I can remind myself later of what I was thinking at the moment.
When I return to the house, Du Lili announces that Big Ma has been waiting for her picture to be taken for more than half an hour. As we walk toward the hall, Du Lili takes my hand and speaks to me in Mandarin: "Your big sister and I once splashed together in those rice paddies. See, over there."
I imagine Du Lili as a younger woman caring for a child-size version of Kwan.
"Sometimes we caught tadpoles," she says, acting girlish, "using our headscarves as nets, like this." She makes scooping motions, then pretends she is wading in mud. "In those days, our village leaders told married women that swallowing a lot of tadpoles was good for birth control. Birth control! We didn't even know what that was. But your sister said, 'Du Lili, we must be good little Communists.' She ordered me to eat the black creatures."
"You didn't!"
"How could I not obey? She was older than I was by two months!"
_Older?_ My mouth drops. How can Kwan be older than Du Lili? Du Lili looks ancient, at least a hundred. Her hands are rough and callused. Her face is heavily furrowed and several of her teeth are missing. I guess that's what happens when you don't use Oil of Olay after a long, hard day in the rice paddies.
Du Lili smacks her lips. "I swallowed a dozen, maybe more. I could feel them wiggling down my throat, swimming in my stomach, then sliding up and down my veins. They squirmed all around my body, until one day I fell down with a fever and a doctor from the big city said, 'Hey, Comrade Du Lili, have you been eating tadpoles? You have blood flukes!' "
She laughs, and a second later turns somber. "Sometimes I wonder if that's why no one wanted to marry me. Yes, I think that's the reason. Everybody heard I ate tadpoles and could never bear a son."
I glance at Du Lili's wandering eye, her sun-coarsened skin. How unfair life has been to her. "Don't worry." She pats my hand. "I'm not blaming your sister. Many times I'm glad I never married. Yes, yes— what a lot of trouble, taking care of a man. I heard that half a man's brain lies between his legs—hah!" She grabs her crotch and ambles forward in a drunken manner. Then she grows serious again. "Some days, though, I say to myself, Du Lili, you would have been a good mother, yes, watchful and strict about morals."
"Sometimes children are a lot of trouble too," I say quietly.
She agrees. "A lot of heartaches."
We walk in silence. Unlike Kwan, Du Lili seems sensible, down-to-earth, someone you can confide in. She doesn't commune with the World of Yin, or at least she doesn't talk about it. Or does she?
"Du Lili," I say. "Can you see ghosts?"
"Ah, you mean like Kwan? No, I don't have yin eyes."
"Does anyone else in Changmian see ghosts?"
She shakes her head. "Only your big sister."
"And when Kwan says she sees a ghost, does everyone believe her?" Du Lili looks away, uncomfortable. I urge her to be open with me: "Myself, I don't believe in ghosts. I think people see what they wish in their hearts. The ghosts come from their imaginations and longings. What do you think?"
"Ah! What does it matter what I think?" She won't meet my eye. She bends down and wipes the toe of her muddied shoe. "It's like this. For so many years, others have been telling us what to believe. Believe in gods! Believe in ancestors! Believe in Mao Tse-tung, our party leaders, dead heroes! As for me, I believe whatever is practical, the least trouble. Most people here are the same way."
"So you don't really believe Big Ma's ghost is here in Changmian." I want to pin her down.
Du Lili touches my arm. "Big Ma is my friend. Your sister is also my friend. I would never damage either friendship. Maybe Big Ma's ghost is here, maybe not. What does it matter? Now do you understand? Ah?"
"Hmm." We keep walking. Will Chinese thinking ever take root in my brain? As if she heard me, Du Lili chuckles. I know what she's thinking. I'm like those intellectuals who came to Changmian, so smart, so sure of their own ideas. They tried to breed mules and ended up making asses out of themselves.
WE ARRIVE at the gateway to the community hall just as a heavy rain begins pelting the ground so violently my chest thumps, urgent and panicky. We dash across the open yard, through double doors leading into a large room that is bone-chillingly cold. The air holds an old, musty dampness, which I imagine to be the byproduct of hundreds of years of moldering bones. The balmy autumn weather, for which Guilin is supposedly known, has taken an early exodus, and although I have on as many layers of clothes as I can fit under my Gore-Tex parka, my teeth are chattering, my fingers are numb. How am I going to shoot any pictures this afternoon?
A dozen people are in the hall, painting white funeral banners and decorating the walls and tables with white curtains and candles. Their loud voices rise above the rain, echo in the room. Kwan is standing next to the coffin. As I approach, I find myself loath to see my photo subject. I imagine she'll look fairly battered. I nod to Kwan when she sees me.
When I look in the coffin, I'm relieved to see that Big Ma's face is covered with a white paper sheet. I try to keep my voice respectful. "I guess the accident damaged her face."
Kwan seems puzzled. "Oh, you mean this paper," she says in Chinese. "No, no, it's customary to cover the face."
"Why?"
"Ah?" She cocks her head, as if the answer would descend from heaven and fall into her ear. "If the paper moves," she says, "then the person is still breathing, and it's too soon to bury the body. But Big Ma is dead for sure, she just told me." Before I can prepare myself, Kwan reaches over and removes the sheet.
Big Ma certainly looks lifeless, although not horribly so. Her brow is bunched into a worried expression and her mouth is twisted into an eternal grimace. I've always thought that when people die, their facial muscles relax, giving them a look of grateful tranquility.
"Her mouth," I say in awkward Chinese. "The way it's crooked. It looks like her dying time was very painful."
Kwan and Du Lili lean forward in unison to stare at Big Ma. "That might be," Du Lili says, "but right now she looks very much as she did when she was alive. The turn in her mouth, that's what she always wore."
Kwan agrees. "Even before I left China, her face was this way, worried and dissatisfied at the same time."
"She was very heavy," I note.
"No, no," Kwan says. "You only think so because now she's dressed for her journey to the next world. Seven layers for her top half, five for the bottom."
I point to the ski jacket Kwan selected as the seventh layer. It's iridescent purple with flashy southwestern details, one of the gifts she bought on sale at Macy's, hoping to impress Big Ma. The price tag is still attached, to prove the jacket isn't a hand-me-down. "Very nice," I say, wishing I were wearing it right now.
Kwan looks proud. "Practical too. All waterproof."
"You mean it rains in the next world?"
"Tst! Of course not. The weather is always the same. Not too hot, not too cold."
"Then why did you say the jacket is waterproof?"
She stares at me blankly. "Because it is."
I cup my numb fingers near my mouth and blow on them. "If the weather's so nice in the other world, why so many clothes, seven and five layers?"
Kwan turns to Big Ma and repeats my question in Chinese. She nods as if listening on a telephone. "Ah. Ah. Ah. Ah-ha-ha-ha!" then translates the answer for my mortal ears: "Big Ma says she doesn't know. Ghosts and yin people were forbidden by the government for so long, now even she's forgotten all the customs and their meanings."
"And now the government _allows_ ghosts?"
"No, no, they just don't fine people anymore for letting them come back. But this is the correct custom, seven and five, always two more on top than on the bottom. Big Ma thinks seven is related to the seven days of the week, one layer for each day. In the old days, people were supposed to mourn their relatives seven weeks, seven times seven, forty-nine days. But nowadays, we're just as bad as foreigners, a few days is enough."
"But why only five layers on her lower half?"
Du Lili cracks a smile. "It means that two days a week Big Ma must wander about with her bottom naked in the underworld."
She and Kwan laugh so hard that people in the hall turn and stare. "Stop! Stop!" Kwan cries, trying to stifle her giggles. "Big Ma is scolding us. She says she hasn't been dead long enough for us to make such jokes." When she regains her composure, Kwan goes on: "Big Ma isn't sure, but she thinks five is for all the common things that attach mortals to the living world—the five colors, the five flavors, the five senses, the five elements, the five emotions—"
And then Kwan stops. "Big Ma, there are seven emotions, ah, not just five." She counts them on her hand, beginning with her thumb: "Joy, anger, fear, love, hate, desire . . . One more—what is it? Ah, yes, yes! Sorrow! No, no, Big Ma, I didn't forget. How could I forget? Of course I have sorrow now that you are leaving this world. How can you say such a thing? Last night I cried, and not just to show off. You saw me. My sorrow was genuine, not fake. Why do you always believe the worst about me?"
"Ai-ya!" Du Lili cries to Big Ma's body. "Don't fight anymore now that you're dead." She looks at me and winks.
"No, I won't forget," Kwan is telling Big Ma. "A rooster, a dancing rooster, not a hen or a duck. I already know this."
"What's she saying?" I ask.
"She wants a rooster tied to the lid of her coffin."
"Why?"
"Libby-ah wants to know why." Kwan listens for a minute, then explains. "Big Ma can't remember exactly, but she thinks her ghost body is supposed to enter the rooster and fly away."
"And you believe that?"
Kwan smirks. "Of course not! Even Big Ma doesn't believe it. This is just superstition."
"Well, if she doesn't believe it, why do it?"
"Tst! For tradition! Also, to give something scary for children to believe. Americans do the same thing."
"No we don't."
Kwan gives me a superior big-sister look. "You don't remember? When I first arrived in the United States, you told me rabbits laid eggs once a year and dead people came out of caves to look for them."
"I did not."
"Yes, and you also said if I did not listen to you, Santa Claus would come down the chimney and put me in a bag, then take me to a very cold place, colder than a freezer."
"I never said that." And even as I protest, I vaguely recall a Christmas joke I once played on Kwan. "Maybe you just misunderstood what I meant."
Kwan sticks out her lower lip. "Hey, I'm your big sister. You think I don't understand your meaning? Hnh! Ah, never mind. Big Ma says no more chitter-chatter. Time to take the picture."
I try to clear my head by doing a routine reading with the light meter. Definitely, time for the tripod. Aside from the illumination of a few white candles near a spirit table, the available light is northern, glary, and coming through dirty windows. There are no ceiling fixtures, no lamps, no wall outlets for strobes. If I use a flash unit, I won't be able to control the amount of light I want, and Big Ma might come out looking even more ghoulish. A chiaroscuro effect is what I prefer anyway, a combination of airiness and murkiness. A full second at f/8 will produce nice detail on one half of Big Ma's face, the shadow of death on the other.
I take out the tripod, set up my Hasselblad, and attach a color Polaroid back to do a test. "Okay, Big Ma," I say, "don't move." Am I losing my mind? I'm talking to Big Ma as if I too believe she can hear me. And why am I making such a big deal over a photo of a dead woman? I'm not going to be able to use it in the article. Then again, everything matters, or should. Every frame should be the best I can shoot. Or is this another one of those myths in life, passed along by high achievers so that everyone else will feel like a perpetual failure?
Before I can ponder this further, a dozen people gather around me, clamoring to see what's come out of the camera. No doubt many of them have seen tourist photo stands, the ones offering instant pictures at extortionist prices.
"Hold on, hold on," I say, as they crowd closer. I place the print against my chest to speed up the developing. The villagers fall silent; they must think noise will disturb the process. I peel off the top and look at the test. The contrast is too sharp for my taste, but I show them the photo anyway.
"Very realistic!" one person exclaims.
"Fine quality!" another says. "See how Big Ma looks—like she's about to wake up and feed her pigs."
Someone jokes: " 'Wah!' she'll say. 'Why are so many people around my bed?' "
Du Lili steps up. "Libby-ah, now take my picture." She is flattening a bristly cowlick with her palm, tugging on her jacket sleeve to smooth out the wrinkles. I look through the viewfinder. She has assumed the stiff posture of a soldier on guard, her face turned toward me, her wandering eye pointed upward. The camera whirs. As soon as I pull the Polaroid test off, she snatches it from my hands and hugs it to her chest, tapping her foot and grinning madly.
"The last time I saw a photograph of myself was many years ago," she says with excitement. "I was very young." When I give her the okay signal, she yanks off the top layer and whips the picture up to her eager face. She squints with her turned-out eye and blinks several times. "So this is what I look like." Her expression speaks of reverence for the miracle of photography. I'm touched, proud in an "aw, shucks" way.
Du Lili carefully hands Kwan the photo as if it were a newly hatched chick. "A good likeness," Kwan says. "What did I tell you? My little sister is very skillful." She passes the Polaroid around for the others to see.
"Very true to life," a man says enthusiastically. The others chime in:
"Exceptionally clear."
"Extraordinarily realistic."
The Polaroid comes around to Du Lili again. She cradles it in her palms. "Then I don't look so good," she says in a wan voice. "I'm so old. I never thought I was so old, so ugly. Am I really that ugly, that stupid-looking?"
A few people laugh, thinking Du Lili is making a joke. But Kwan and I can see that she is genuinely shocked. She wears the face of someone betrayed, and I am the one who has wounded her. Certainly she must have seen herself recently in a mirror? But then the way we see our reflections from changing angles allows us to edit out what we don't like. The camera is a different sort of eye, one that sees a million present particles of silver on black, not the old memories of a person's heart.
Du Lili wanders off, and I want to say something to comfort her, to tell her that I am a bad photographer, that she has wonderful qualities a camera can never see. I start to follow her, but Kwan takes my arm and shakes her head. "Later I will talk to her," she says, and before she can say anything else, I am surrounded by a dozen people, all of them pleading that I take their photos as well. "Me first!" "Take one of me with my grandson!"
"Wah!" Kwan scolds. "My sister is not in the business of giving away free portraits." The people keep insisting: "Just one!" "Give me a photo too!" Kwan holds up her hands and yells sternly: "Quiet! Big Ma just told me everyone must leave right away." The shouts dwindle. "Big Ma says she needs to rest before her journey to the next world. Otherwise she might go crazy with grief and stay here in Changmian." Her comrades quietly absorb this proclamation, then file out, grousing good-naturedly.
When we are alone, I grin my thanks to Kwan. "Did Big Ma really say that?" Kwan gives me a sidelong glance and bursts into laughter. I join her, grateful for her quick thinking.
Then she adds: "The truth is, Big Ma said to take more photos of her, but this time from another angle. She said the last one you took makes her look almost as old as Du Lili."
I'm taken aback. "That's a mean thing to say."
Kwan acts clueless. "What?"
"Saying Du Lili looks older than Big Ma."
"But she is older, at least five or six years."
"How can you say that! She's younger than you are."
Kwan tilts her head, attentive. "Why do you think this?"
"Du Lili told me."
Kwan is now reasoning with Big Ma's lifeless face: "I know, I know. But because Du Lili has mentioned this, we must tell her the truth." Kwan walks up to me. "Libby-ah, now I must tell you a secret."
A stone drops in my stomach.
"Almost fifty years ago, Du Lili adopted a little girl she found on the road during civil war time. Later, that daughter died and Du Lili was so crazy with sorrow she believed she became her daughter. I remember this, because the little girl was my friend, and yes, if she had lived, she would have been two months younger than I am now and not the seventy-eight-year-old woman that Du Lili is today. And now that I am telling you this—" Kwan breaks off and begins to argue with Big Ma again. "No, no, I can't tell her that, it's too much."
I stare at Kwan. I stare at Big Ma. I think about what Du Lili has said. Who and what am I supposed to believe? All the possibilities whirl through my brain, and I feel I am in one of those dreams where the threads of logic between sentences keep disintegrating. Maybe Du Lili is younger than Kwan. Maybe she's seventy-eight. Maybe Big Ma's ghost is here. Maybe she isn't. All these things are true and false, yin and yang. What does it matter?
Be practical, I tell myself. If the frogs eat the insects and the ducks eat the frogs and the rice thrives twice a year, why question the world in which they live?
## [17
THE YEAR OF NO FLOOD](TheHundredSecretSenses-toc.html#TOC-20)
Why question the world, indeed? Because I'm not Chinese like Kwan. To me, yin isn't yang, and yang isn't yin. I can't accept two contradictory stories as the whole truth. When Kwan and I walk back to Big Ma's house, I quietly ask, "How did Du Lili's daughter die?"
"Oh, it's a very sad story," Kwan answers in Chinese. "Maybe you don't want to know."
We continue in silence. I know she is waiting for me to ask again, so finally I say, "Go ahead."
Kwan stops and looks at me. "You won't be scared?"
I shake my head while thinking, How the hell would I know if I'm not going to be scared? As Kwan begins to speak, I shiver, and it is not due to the cold.
Her name was Buncake and we were five years old when she drowned. She was as tall as I was, eye to eye, her quiet mouth to my noisy one. That's what my auntie complained, that I talked too much. "If you let go with one more word," Big Ma would warn, "I'll send you away. I never promised your mother I'd keep you."
Back then, I was skinny, nicknamed Pancake, _bao-bing—_ "a flimsy piece of fritter," Big Ma called me—always with four scabby bites on my knees and elbows. And Buncake, she was plump, her arms and legs creased like a steamed stuffed _bao-zi._ Du Yun was the one who found her on the road—that was Du Lili's name back then, Du Yun. Big Ma was the one who named Buncake Lili, because when she first came to our village, _lili-lili-lili_ was the only sound she could make, the warbling of an oriole. _Lili-lili-lili,_ that's what came out of the pucker of her small red mouth, as if she had just bitten into a raw persimmon, expecting sweet, tasting bitter. She watched the world like a baby bird, her eyes round and black, looking for danger. Why she was like this nobody knew except me, because she never talked, not with words at least. But in the evening, when lamplight danced on the ceiling and walls, her little white hands spoke. They would glide and swoop with the shadows, soar and float, those pale birds through clouds. Big Ma would watch and shake her head: Ai-ya, how strange, how strange. And Du Yun would laugh like an idiot watching a play. Only I understood Buncake's shadow talk. I knew her hands were not of this world. You see, I also was a child, still close to the time before this life. And so I too remembered I had once been a spirit who left this earth in the body of a bird.
To Du Yun's face, everyone in the village would smile and tease, "That little Buncake of yours, she's peculiar, ah?" But outside our courtyard, they would whisper rotten words, and these sounds would drift over our wall and into my ears.
"That girl is so spoiled she's turned crazy," I heard neighbor Wu say. "Her family must have been bourgeois-thinking. Du Yun should beat her often, at least three times a day."
"She's possessed," another said. "A dead Japanese pilot fell from the sky and lodged in her body. That's why she can't speak Chinese, only grunt and twirl her hands like a suicide plane."
"She's stupid," yet another neighbor said. "Her head is as hollow as a gourd."
But to Du Yun's way of thinking, Buncake didn't speak because Du Yun could speak for her. A mother always knows what's best for her daughter, she'd say, isn't that true?—what she should eat, what she should think and feel. As to Buncake's shadow-dancing hands, this was proof, Du Yun once said—genuine proof!—that her ancestors had been ladies of the court. And Big Ma replied, "Wah! Then she has counterrevolutionary hands, hands that will be chopped off one day. Better that she learn how to press one finger to her nostril and blow snot into her palm."
Only one thing about Buncake made Du Yun sad. Frogs. Buncake didn't like springtime frogs, green-skinned frogs as small as her fist. At dusk, that's when you could hear them, groaning like ghost gates: _Ahh-wah, ahh-wah, ahh-wah._ Big Ma and Du Yun would grab buckets and nets, then wade into the watery fields. And all those frogs held their breath, trying to disappear with their silence. But soon they could not hold in their wishes any longer: _Ahh-wah, ahh-wah, ahh-wah,_ even louder than before, wailing for love to find them.
"Who could love such a creature?" Du Yun used to joke. And Big Ma always answered, "I can—once it is cooked." How easily they caught those lovesick creatures. They put them in buckets, shiny as oil in the rising moonlight. By morning, Big Ma and Du Yun were standing by the road, calling, "Frogs! Juicy frogs! Ten for one yuan!" And there we were, Buncake and I, seated on upturned buckets, chins resting on palms, nothing to do but feel the sun rise, warming one cheek, one arm, one leg.
No matter how good the business, Big Ma and Du Yun always saved at least a dozen frogs for our noon meal. By mid-morning, we trudged home, seven buckets empty, one half-full. In the courtyard kitchen, Big Ma would make a big hot fire. Du Yun would reach into the bucket and grab a frog, and Buncake would hurry behind me, hiding. I could feel her chest heave up and down, fast and hard, just like that frog squirming in Du Yun's hand, puffing its throat in and out.
"Watch closely, ah," Du Yun would say to Buncake and me. "This is the best way to cook a frog." She would turn the frog on its back and—quick!—stick the sharp end of a pair of scissors up its anus _— szzzzzzzz!—_ slicing all the way to its throat. Her thumb would dip under the slit, and with one fast tug, out slid a belly full of mosquitoes and silver-blue flies. With another tug, at the frog's throat, off slipped the skin, snout to tail, and it hung from Du Yun's fingers like the shriveled costume of an ancient warrior. Then chop, chop, chop, and the frog lay in pieces, body and legs, the head thrown away.
While Du Yun peeled those frogs, one after another, Buncake kept her fist wedged hard between her teeth, like a sandbag stopping a leak in a riverbank. So no scream came out. And when Du Yun saw the anguish on Buncake's face, she would croon in a mother's sweet voice: "Baby-ah, wait a little longer. Ma will feed you soon."
Only I knew what words were stuck in Buncake's screamless mouth. In her eyes I could see what she had once seen, as clearly as if her memories were now mine. That this tearing of skin from flesh was how her mother and father had died. That she had watched this happen from a leafy limb, hidden high in a tree where her father had put her. That in the tree an oriole called, warning Buncake away from her nest. But Buncake would not make a sound, not a cry or even a whimper, because she had promised her mother to be quiet. That's why Buncake never talked. She had promised her mother.
In twelve minutes, twelve frogs and skins flew into the pan and crackled in oil, so fresh some of the legs would leap from the pan— wah!—and Du Yun would catch them with one hand, her other hand still stirring. That's how good Du Yun was at cooking frogs.
But Buncake didn't have the stomach to appreciate this. By the dim lamplight, she watched as we greedy ones ate those delicious creatures, our teeth busy searching for shreds of meat on bones as tiny as embroidery needles. The skins were the best, soft and full of flavor. Second best I liked the crunchy small bones, the springy ones just above the feet.
Often Du Yun would look up and urge her new daughter, "Don't play now, eat, my treasure, eat." But Buncake's hands would be flapping and flying, soaring with her shadows. Then Du Yun would become sad that her daughter wouldn't eat the dish she cooked best. You should have seen Du Yun's face—so much love for a leftover girl she had found on the road. And I know Buncake tried to love Du Yun with the leftover scraps of her heart. She followed Du Yun's footsteps around the village, she raised one arm so her new mother could hold her hand. But on those nights when the frogs sang, when Du Yun picked up her swinging buckets, Buncake would run to a corner, press in tight, and begin to sing: _Lili-lili-lili._
That's how I remember Buncake. She and I were good little friends. We lived in the same house. We slept in the same bed. We were like sisters. Without talking, we each knew what the other was feeling. At such a young age, we knew about sadness, and not just our own. We both knew about the sadness of the world. I lost my family. She lost hers.
The year Du Yun found Buncake on the road, that was a strange year, the year of no flood. In the past, our village always had too much rain, at least one flood in the springtime. Sudden rivers that swept through our homes, washing the floors of insects and rats, slippers and stools, then retching all this into the fields. But the year Buncake came— no flood, just rain, enough for crops and frogs, enough for people in our village to say, "No flood, why are we so lucky? Maybe it was the girl Du Yun found on the road. Yes, she must be the reason."
The following year, there was no rain. In all the villages surrounding ours, rain fell as usual, big rain, small rain, long rain, short rain. Yet in our village, none. No rain for the spring tilling. No rain for the summer harvest. No rain for the fall planting. No rain, no crops. No water to cook the rice that no longer grew, no chaff to feed the pigs. The rice fields cooked hard as porridge crust, and the frogs lay on top, dry as twigs. The insects climbed out of the cracked ground, waving their feelers toward the sky. The ducks withered and we ate them, skin on bones. When we stared too long at the mountain peaks, our hungry eyes saw roasted sweet potatoes with their skins broken open. Such a terrible year, so terrible that people in our village said Buncake, that crazy girl, she must be the reason.
One hot day, Buncake and I were sitting in the hull of a dusty ditch that ran alongside our house. We were waiting for our imaginary boat to take us to the land of fairy queens. Suddenly we heard a groan from the sky, then another groan, then a big crack _—kwahhh!—_ and hard rain fell like pellets of rice. I was so happy and scared! More lightning came, more thunder. At last our boat is going, I shouted. And Buncake laughed. For the first time, I heard her laugh. I saw her reach her hands toward the flashes in the sky.
The rain kept gurgling _—gugu-gugu-gugu—_ rolling down the mountains, filling their wrinkles and veins. And the hollows couldn't gulp fast enough, there was that much water. Soon, so quick, this friendly ditch-boat became a brown river, pushing against our legs. White water-tails grabbed our little wrists and ankles. Faster and faster we tumbled, arms first, then feet first, until the water spat us out into a field.
Through whispered talk, I later learned what happened. When Big Ma and Du Yun took us out of the water, we both were pale and still, wrapped in weeds, two soggy cocoons with no breath bursting through. They dug the mud out of our nostrils and mouths, they pulled the weeds out of our hair. My thin body was broken to pieces, her sturdy one was not. They dressed us in farewell clothes. Then they went to the courtyard, washed two pig troughs that were no longer needed, broke off the seats of two benches for lids. They put us in these poor little coffins, then sat down and cried.
For two days, we lay in those coffins. Big Ma and Du Yun were waiting for the rain to stop so they could bury us in the rocky soil where nothing ever grew. On the third morning, a big wind came and blew the clouds away. The sun rose, and Du Yun and Big Ma opened the coffins to see our faces one last time.
I felt fingers brushing my cheek. I opened my eyes and saw Du Yun's face, her mouth stretched big with joy. "Alive!" she cried. "She's alive!" She grabbed my hands and rubbed them against her face. And then Big Ma's face was looking down at me too, searching. I was confused, my head as thick as morning fog.
"I want to get up." That's what I first said. Big Ma jumped back. Du Yun dropped my hands. I heard them howl: "How can this be! It can't be!"
I sat up. "Big Ma," I said, "what's the matter?" They began to scream, loud screams, so terrible I thought my head would burst from fright. I saw Big Ma running to the other coffin. She flung open the lid. I saw myself. My poor broken body! And then my head whirled, my body fell, and I saw nothing more until evening came.
When I awoke, I was lying on the cot I once shared with Buncake. Big Ma and Du Yun were standing across the room, in the doorway. "Big Ma," I said, yawning. "I had a nightmare."
Big Ma said, "Ai-ya, look, she speaks." I sat up and slid off the cot, and Big Ma cried, "Ai-ya, look, she moves." I stood up, complained I was hungry and wanted to pee. She and Du Yun backed off from the doorway. "Go away or I'll beat you with peach twigs!" Big Ma cried.
And I said, "Big Ma, we have no peach trees." She clapped her hand to her mouth. At the time, I didn't know that ghosts were supposed to be scared of peach twigs. Later, of course, I learned that this was just superstition. I've since asked many ghosts, and they all laugh and say, "Scared of peach twigs? No such thing!"
Anyway, as I was saying, my bladder was about to burst. I was anxious to pieces, hopping and holding myself in. "Big Ma," I said, more politely this time, "I want to visit the pigs." Next to the pen we had a small pit, a plank of wood on each side for balancing yourself while doing your business, both kinds. That was before our village went through reeducation on collective waste. And after that, it was no longer enough to give your mind, body, and blood to the common good—you had to donate your shit too, just like American taxes!
But Big Ma did not say I could visit the pigs. She walked up to me and spit in my face. This was another superstition about ghosts: Spit on them and they'll disappear. But I did not disappear. I wet my pants, a warm stream dribbling down my legs, a puddle darkening the floor. I was sure that Big Ma would beat me, but instead she said, "Look, she's pissing."
And Du Yun said, "How can that be? A ghost can't piss."
"Well, use your own eyes, you fool. She's pissing."
"Is she a ghost or isn't she?"
They went on and on, arguing about the color, the stink, the size of my puddle. Finally they decided to offer me a little something to eat. This was their thinking: If I was a ghost, I would take this bribe and leave. If I was a little girl, I would stop complaining and go back to sleep, which is what I did after eating a little piece of stale rice ball. I slept and dreamed that all that had happened was part of the same long dream.
When I awoke the next morning, I again told Big Ma that I had suffered a nightmare. "You're still sleeping," she said. "Now get up. We're taking you to someone who will wake you out of this dream."
We walked to a village called Duck's Return, six _li_ south of Changmian. In this village lived a blind woman named Third Auntie. She was not really my auntie. She was auntie to no one. It was just a name, Third Auntie, what you call a woman when you should not say the word "ghost-talker." In her youth, she had become famous all around the countryside as a ghost-talker. When she was middle-aged, a Christian missionary redeemed her and she gave up talking to ghosts, all except the Holy Ghost. When she was old, the People's Liberation Army reformed her, and she gave up the Holy Ghost. And when she grew very old, she no longer remembered whether she was redeemed or reformed. She was finally old enough to forget all she had been told to be.
When we entered her room, Third Auntie was sitting on a stool in the middle of the floor. Big Ma pushed me forward. "What's wrong with her?" Du Yun asked in a pitiful voice. Third Auntie took my hands in her rough ones. She had eyes the color of sky and clouds. The room was quiet except for my breathing. At last, Third Auntie announced: "There's a ghost inside this girl." Big Ma and Du Yun gasped. And I jumped and kicked, trying to rid myself of the demon.
"What can we do?" Du Yun cried.
And Third Auntie said, "Nothing. The girl who lived in this body before doesn't want to come back. And the girl who lives in it now can't leave until she finds her." That's when I saw her, Buncake, staring at me from a window across the room. I pointed to her and shouted, "Look! There she is!" And when I saw her pointing back at me, her puckered mouth saying my words, I realized I was looking at my own reflection.
On the way home, Big Ma and Du Yun argued, saying things a little girl should never hear.
"We should bury her, put her in the ground where she belongs." This was Big Ma talking.
"No, no," Du Yun moaned. "She'll come back, still a ghost, and angry enough to take you and me with her."
Then Big Ma said, "Don't say she's a ghost! We can't bring home a ghost. Even if she is—wah, what trouble!—we'll have to be reformed."
"But when people see this girl, when they hear the other girl's voice . . ."
By the time we reached Changmian, Big Ma and Du Yun had decided they would pretend nothing was the matter with me. This was the attitude people had to take with many things in life. What was wrong was now right. What was right was now left. So if someone said, "Wah, this girl must be a ghost," Big Ma would answer, "Comrade, you are mistaken. Only reactionaries believe in ghosts."
At Buncake's funeral, I stared at my body in the coffin. I cried for my friend, I cried for myself. The other mourners were still confused over who was dead. They wept and called my name. And when Big Ma corrected them, they again wept and called Buncake's name. Then Du Yun would begin to wail.
For many weeks, I scared everyone who heard my voice come out of that puckered mouth. No one talked to me. No one touched me. No one played with me. They watched me eat. They watched me walk down the lane. They watched me cry. One night, I woke up in the dark to find Du Yun sitting by my bed, pleading in a lilting voice. "Buncake, treasure, come back home to your ma." She lifted my hands, moved them near the candlelight. When I yanked them back, she churned the air with her arms—oh, so clumsy, so desperate, so sad, a bird with broken wings. I think that's when she started believing she was her daughter. That's how it is when you have a stone in your heart and you can't cry out and you can't let it go. Many people in our village had swallowed stones like this, and they understood. They pretended I was not a ghost. They pretended I had always been the plump girl, Buncake the skinny one. They pretended nothing was the matter with a woman who now called herself Du Lili.
In time, the rains came again, then the floods, then the new leaders who said we must work harder to wash away the Four Olds, build the Four News. The crops grew, the frogs creaked, the seasons went by, one ordinary day after another, until everything changed and was the same again.
One day, a woman from another village asked Big Ma: "Hey, why do you call that fat girl Pancake?" And Big Ma looked at me, trying to remember. "Once she was skinny," she said, "because she wouldn't eat frogs. Now she can't stop."
You see, everyone decided not to remember. And later, they really did forget. They forgot there was a year of no flood. They forgot that Du Lili was once called Du Yun. They forgot which little girl drowned. Big Ma still beat me, only now I had a body with more fat, so her fists did not hurt me as much as before.
Look at these fingers and hands. Sometimes even I believe they have always been mine. The body I thought I once had, maybe that was a dream I confused with waking life. But then I remember another dream.
In this dream, I went to the World of Yin. I saw so many things. Flocks of birds, some arriving, some leaving. Buncake soaring with her mother and father. All the singing frogs I had ever eaten, now with their skin coats back on. I knew I was dead, and I was anxious to see my mother. But before I could find her, I saw someone running to me, anger and worry all over her face.
"You must go back," she cried. "In seven years, I'll be born. It's all arranged. You promised to wait. Did you forget?" And she shook me, shook me until I remembered.
I flew back to the mortal world. I tried to return to my body. I pushed and shoved. But it was broken, my poor thin corpse. And then the rain stopped. The sun was coming out. Du Yun and Big Ma were opening the coffin lids. Hurry, hurry, what should I do?
So tell me, Libby-ah, did I do wrong? I had no choice. How else could I keep my promise to you?
## [18
SIX-ROLL SPRING CHICKEN](TheHundredSecretSenses-toc.html#TOC-21)
"Now you remember?" Kwan asks.
I am transfixed by her plump cheeks, the crease of her small mouth. Looking at her is like viewing a hologram: locked beneath the shiny surface is the three-dimensional image of a girl who drowned.
"No," I say.
Is Kwan—that is, this woman who claims to be my sister—actually a demented person who _believed_ she was Kwan? Did the flesh-and-blood Kwan drown as a little girl? That would account for the disparity between the photo of the skinny baby our father showed us and the chubby girl we met at the airport. It would also explain why Kwan doesn't resemble my father or my brothers and me in any way.
Maybe my wish from childhood came true: The real Kwan died, and the village sent us this other girl, thinking we wouldn't know the difference between a ghost and someone who thought she was a ghost. Then again, how can Kwan not be my sister? Did a terrible trauma in childhood cause her to believe she had switched bodies with someone else? Even if we aren't genetically related, isn't she still my sister? Yes, of course. Yet I want to know what parts of her story might be true.
Kwan smiles at me, squeezing my hand. She points to birds flying overhead. If only she said they were elephants. Then, at least, her madness would be consistent. Who can tell me the truth? Du Lili? She isn't any more reliable than Kwan. Big Ma is dead. And no one else in the village who would be old enough to remember speaks anything other than Changmian. Even if they did speak Mandarin, how can I ask? "Hey, tell me, is my sister really my sister? Is she a ghost or just insane?" But I have no time to decide what to do. Kwan and I are now walking through the gate to Big Ma's house.
In the central room, we find Simon and Du Lili carrying on a spirited conversation in the universal language of charades. Simon rolls down an imaginary car window and shouts, "So I stuck my head out and said, 'Come on, move your ass!' " He leans on an invisible horn, and then _—"Bbbbrr-ta-ta! Bbbbbrrr-ta-ta!"—_ imitates an Uzi-wielding thug blowing up his tires.
Du Lili says in Changmian what sounds like the equivalent of "Hnh! That's nothing." She acts out a pedestrian lugging bags of groceries— heavy bags, we are asked to observe, that stretch her arms like noodle dough. Suddenly she glances up, leaps back nearly onto Simon's toes, and launches her heavy bags just as a car zigzagging like a snake flies past the tip of her nose and plows into a crowd of people. Or maybe she means a stand of trees. In any case, some sort of limbs are flying this way and that through the air. To end her little drama, she walks over to the driver and spits in his face, which in this reenactment is the bucket next to Simon's shoe.
Kwan bursts into hoots and cheers, I clap. Simon pouts like the second runner-up in _Queen for a Day._ He accuses Du Lili of exaggerating— maybe the car was _not_ going fast like a snake, no indeed, but slow as a lame cow. _"Bu-bu-bu!"_ she cries, giggling and stamping her foot. Yes, and maybe she was walking with her head in the clouds and _she_ caused the accident. _"Bu-bu-bu!"_ As she pummels his back, Simon cowers: "All right, you win! Your drivers are worse!"
Except for their age difference, they look like new lovers who flirt with each other, bantering, provoking, finding excuses to touch. I feel a twitch in my heart, although it can't be jealousy, because who could ever think that those two— Well, whether or not Kwan's story about Du Lili and her dead daughter is true, one thing is certain: Du Lili is way past old.
The charades now over, she and Kwan drift toward the courtyard, discussing what they are going to make for dinner. When they are out of earshot, I pull Simon aside.
"How did you and Du Lili get on the topic of bad drivers, of all things?"
"I started off trying to tell her about yesterday's trip here with Rocky, about the accident."
Makes sense. I then recount to him what Kwan told me. "So what do you think?"
"Well, number one, Du Lili doesn't seem crazy to me, nor does Kwan. And number two, they're the same old stories you've heard all your life."
"But this one's different. Don't you see? Maybe Kwan really isn't my sister."
He frowns. "How can she _not_ be your sister? Even if she isn't blood-related, she's still your sister."
"Yeah, but that means there was another girl who was _also_ my sister."
"Even so, what would you do? Disown Kwan?"
"Of course not! It's just that—well, I need to know for sure what really happened."
He shrugs. "Why? What difference would it make? All I know is what I see. To me, Du Lili seems like a nice lady. Kwan is Kwan. The village is great. And I'm glad to be here."
"So what about Du Lili? Do you believe her when she says she's fifty? Or do you believe Kwan, who says Du Lili is—"
Simon breaks in: "Maybe you didn't understand what Du Lili was saying. You said it yourself, your Chinese isn't all that great."
I'm annoyed. "I just said I couldn't _speak_ it as well as Kwan."
"Maybe Du Lili used an expression like—well, 'young as a spring chicken.' " His voice has the assured tone of masculine reason. "Maybe you took her literally to mean she thought she was a chicken."
"She didn't say she was a chicken." My temples are pounding.
"See, now you're taking even me too literally. I was only giving an example of—"
I lose it. "Why do you always have to prove you're so goddamn right?"
"Hey, what is this? I thought we were just talking here. I'm not trying to—"
And then we hear Kwan shout from the courtyard: "Libby-ah! Simon! Hurry come! We cook now. You want take picture, yes?"
Still irritated, I run into Big Ma's room to retrieve my photo gear. There it is again: the marriage bed. Don't even think about it, I tell myself. I look out the window, then at my watch: almost dusk, the golden half-hour. If ever there were a time and a place to allow gut passion into my work, this is it, in China, where I have no control, where everything is unpredictable, totally insane. I pick up the Leica, then stuff ten rolls of high-speed film into my jacket pocket.
In the courtyard, I take out a numbered roll and load it. After the heavy rain, the sky has drained itself to a soft gouache blue, splotches of powder-puff clouds swimming behind the peaks. I inhale deeply and smell the woody kitchen smoke of Changmian's fifty-three households. And beneath that bouquet wafts the ripe odor of manure.
I scope out the elements of the scene. The mud-brick walls of the courtyard will serve well as overall backdrop. I like the orange tinge, the rough texture. The tree in the middle has anemic-looking leaves— avoid that. The pigpen has definite foreground potential—nicely positioned on the right side of the courtyard under an eave of thatched twigs. It's rustically simple, like a manger in a kids' Christmas play. But instead of Jesus, Mary, and Joseph, there are three pigs rooting in muck. And a half-dozen chickens, one lacking a foot, another with part of its beak missing. I dance in an expanding then shrinking arc around my subject. Out of the corner of my eye, I see a slop bucket full of grayish rice gruel and flies. And a pit with a terrible stench, a dark watery hole. I lean over and glimpse a gray-furred creature, along with clumps of puffy rice that writhe—maggots, that's what they are.
Life in Changmian now seems futile. I should be "previsualizing" the moment I want, willing spontaneity to coincide with what's given. But all I see in my head are well-heeled readers flipping through a chic travel magazine that specializes in bucolic images of third world countries. I know what people want to see. That's why my work usually feels unsatisfying, pre-edited into safe dullness. It isn't that I want to take photos that are deliberately unflattering. What's the point in that? There's no market for them, and even if there were, hard realism would give people the wrong impression, that _all_ of China is this way, backward, unsanitary, miserably poor. I hate myself for being American enough to make these judgments. Why do I always edit the real world? For whose sake?
Screw the magazine. To hell with right and wrong impressions. I check the light, the f-stop. I'll just do my best to capture a moment, the sense of it as it happens. And that's when I spot Du Lili crouched next to the hand pump, draining water into a pan. I circle her, focus, and begin to shoot. But upon seeing my camera, she jumps up to pose and tugs at the bottom of her old green jacket. So much for spontaneity.
"You don't have to stand still," I call to her. "Move around. Ignore me. Do what you want."
She nods, then walks around the courtyard. And in her diligence to forget the camera's presence, she admires a stool, gestures toward baskets hanging from a tree, marvels over an ax covered with mud, as if she were displaying priceless national treasures. "One, two, three," I count stiffly in Chinese, then take a few posed shots to satisfy her. "Good, very good," I say. "Thank you."
She looks puzzled. "I did it wrong?" she asks in a plaintive child's voice. Ah—she's been waiting for a flash, the click of the shutter, neither of which the Leica will produce. I decide to tell a small lie.
"I'm not really taking pictures," I explain. "I'm only looking—just practice."
She gives me a relieved smile and walks back to the pigpen. As she opens the gate, the pigs snort and run toward her, snouts raised, sniffing for handouts. A few hens warily circle her for the same reason. "A nice fat one," Du Lili says, considering her choices. I skulk around the courtyard like a thief, trying to remain unobtrusive, while I search for the best combination of subject, light, background, and framing. The sun sinks another degree and sends filtered rays through the twig roof, throwing warm light on Du Lili's gentle face. With this bit of serendipity, my instincts take over. I can feel the shift, the power that comes from abandoning control. I'm shooting now between breaths. Unlike other cameras, which leave me blind while the shutter's open, the Leica lets me see the moment I'm capturing: the blur of Du Lili's hand grabbing a chicken, the flurry of the other chickens, the pigs turning in unison like a marching band. And Simon—I shoot a few pictures of him taking notes for possible captions. This is like the old days, the way we used to work in rhythm with each other. Only now he isn't in his business-as-usual mode. His eyes are wonderfully intense. He glances at me and smiles.
I turn my camera back to Du Lili. She is walking toward the hand pump, the squawking chicken in her grip. She holds it over a white enamel bowl that sits on a bench. Her left hand firmly clutches the chicken's neck. In her other hand is a small knife. How the hell is she going to lop off the chicken's head with that? Through the viewfinder, I see her pressing the blade against the bird's neck. She slowly saws. A thin ribbon of blood springs up. I'm as stunned as the chicken. She dangles the bird so that its neck is extended downward, and blood starts to trickle into the white bowl.
In the background, I can hear the pigs screaming. They are actually _screaming,_ like people in terror. Someone once told me that pigs can go into a deadly fever when they're being led to the slaughterhouse, that they are smart enough to know what awaits them. And now I wonder if they also could have sympathy for the pain of a dying chicken. Is that evidence of intelligence or a soul? In spite of all the open-heart surgeries and kidney transplants I've photographed, I feel queasy. Yet I keep shooting. I notice Simon is no longer writing captions.
When the bowl is half filled with blood, Du Lili lets the chicken fall to the ground. For several agonizing minutes, we watch as it stumbles and gurgles. At last, with dazed eyes, it slumps over. Well, if Du Lili believes she is Buncake, she certainly must have lost her compassion for birds.
Simon comes over to me. "That was fucking barbaric. I don't know how you could keep shooting."
His remark irritates me. "Stop being so ethnocentric. You think killing chickens in the States is more humane? Anyway, she probably did it that way so the meat will be free of toxins. It's like a tradition, a kosher process or something."
"Kosher my ass! Kosher is killing the animal quickly so it _doesn't_ suffer. And the blood's drained after the animal's dead, then thrown away."
"Well, I still think she did it that way for health reasons." I turn to Du Lili and ask her in Chinese.
_"Bu-bu,"_ she replies, shaking her head with a laugh. "After I have enough blood, I usually cut off the head right away. But this time I let the chicken dance a bit."
"Why?"
"For you!" she says happily. "For your photos! More exciting that way, don't you agree?" Her eyebrows flick up as she waits for my thanks. I fake a smile.
"Well?" says Simon.
"It's . . . Well, you're right, it's not kosher." And then I can't help it, seeing that smug look on his face. "Not kosher in a Jewish sense," I add. "It's more of an ancient Chinese ritual, a spiritual cleansing . . . for the chicken." I return my attention to the viewfinder.
Du Lili dumps the chicken into a pot of boiling water. And then, with her bare hands, she begins dipping the bird as though she were washing a sweater. She has so many calluses they cover her palms like asbestos gloves. At first she seems to be petting the dead chicken, consoling it. But with each stroke, a handful of feathers comes away, until the bird emerges from its bath pimply-skinned pink.
Simon and I follow Du Lili as she carries the carcass across the courtyard to the kitchen. The roof is so low we have to crouch to avoid brushing our heads against the thatching. From a dark corner in the back, Kwan pulls out a bundle of twigs, then feeds them into the mouth of a blazing mud-brick hearth. Over the fire sits a wok big enough to cook a boar. She grins at me. "Good picture?"
How could I have had any doubts she is my sister? They're all stories, I tell myself. She just has a crazy imagination.
Kwan guts the chicken in one motion, then chops it up, head, feet, and tail, and dumps the pieces into a bubbling broth. Into this mixture, she tosses several handfuls of greens, what looks like chard. "Fresh," she notes in English to Simon. "Everything always fresh."
"You went to the market today?"
"What market? No market. Only backyard, pick youself." Simon writes this down.
Du Lili is now bringing in the bowl of chicken blood. It has congealed to the color and consistency of strawberry gelatin. She cuts the blood into cubes, then stirs them into the stew. As I watch the red swirls, I think of the witches of _Macbeth,_ their faces lit by fire, with steam rising from the caldron. How did the line go? "For a charm of powerful trouble," I recite, "like a hell-broth boil and bubble."
Simon looks up. "Hey, that's what I was just thinking." He leans over to smell the stew. "This is great material."
"Don't forget, we have to eat this great material."
When the fire dies down, so does my available light. I slip the Leica into a jacket pocket. God! I'm hungry! If I don't eat the chicken and its bloody broth, what are my other choices? There's no ham and cheese in the fridge—there's no fridge. And if I wanted ham, I'd have to slaughter the screaming pigs first. But there isn't time to consider alternatives. Kwan is now at a half-crouch, grabbing the handles of the giant wok. She does a dead lift. "Eat," she announces.
In the center of the courtyard, Du Lili has made a small twig fire within an iron ring. Kwan sets the wok on top and Du Lili passes out bowls, chopsticks, and cups for tea. Following her lead, we squat around our improvised dinner table. "Eat, eat," Du Lili says, motioning to Simon and me with her chopsticks. I eye the pot, looking for something that resembles my supermarket version of packaged meat. But before I can find it, Du Lili plucks a chicken's foot from the stew and plops it in my bowl.
"No, no. You take that," I protest in Chinese. "I can help myself."
"Don't be polite," she argues. "Eat before it gets cold."
Simon smirks. I transfer the foot to his bowl. "Eat, eat," I say with a gracious smile, then help myself to a thigh. Simon stares morosely at the once dancing foot. He takes a tentative bite and chews with a thoughtful expression. After a while, he nods politely to Du Lili and says, "Hmm-mmm. Good, very good." The way she beams, you'd think she's just won a cook-off.
"That was nice of you to say that."
"It _is_ good," he says. "I wasn't just being polite."
I ease my teeth on the edge of the thigh and take a puppy nip. I chew, let it roll onto my tongue. No taste of blood. The meat is amazingly flavorful, velvety! I eat more, down to the bone. I sip the broth, so clean-tasting yet buttery rich. I reach into the pot and fish out a wing. I chew, and conclude that Chinese courtyard chickens taste better than American free-range. Does the tastiness come from what they eat? Or is it the blood in the broth?
"How many rolls did you shoot?" Simon asks.
"Six," I say.
"Then we'll call this six-roll spring chicken."
"But it's autumn."
"I'm naming it in honor of Du Lili, who ain't no spring chicken, as you pointed out." Simon quivers and pleads, Quasimodo style: "Please, Mistress, don't beat me."
I make the sign of the cross over his head. "All right. You're forgiven, you jerk."
Du Lili holds up a bottle of colorless liquor. "When the Cultural Revolution ended, I bought this wine," she proclaims. "But for the last twenty years, I've had no reason to celebrate. Tonight, I have three." She tips the bottle toward my cup, sighs a protracted "ahhhh," as if she were relieving her bladder, not pouring us wine. When all of our cups are filled, she lifts hers _—"Gan-bei!"—_ and sips noisily, her head tilting back slowly until she has emptied the contents.
"You see?" Kwan says in English. "Must keep cup going back, back, back, until all finish." She demonstrates by chugging hers. "Ahhhh!" Du Lili pours herself and Kwan another round.
Well, if Kwan, the queen of the teetotalers, can drink this, it must not be very strong. Simon and I clink glasses, then toss the liquor down our throats, only to gasp immediately like city slickers in a cowboy saloon. Kwan and Du Lili slap their knees and chortle. They point to our cups, still half full.
"What is this?" Simon gasps. "I think it just removed my tonsils."
"Good, ah?" Kwan tops his cup before he can refuse.
"It tastes like sweat socks," he says.
"Sweet suck?" Kwan takes another sip, smacks her lips, and nods in agreement.
Three rounds and twenty minutes later, my head feels clear, but my feet have gone to sleep. I stand up and shake my legs, wincing as they tingle. Simon does the same.
"That tasted like shit." He stretches his arms. "But you know, I feel _great._ "
Kwan translates for Du Lili: "He says, Not bad."
"So what do you call this drink, anyway?" Simon asks. "Maybe we should take some with us when we go back to the States."
"This drink," says Kwan, and she pauses to look at her cup with great respect, "this drink we call pickle-mouse wine, something like that. Very famous in Guilin. Taste good, also good for health. Take long time make. Ten, maybe twenty year." She motions to Du Lili and asks her to show the bottle. Du Lili holds up the bottle and taps the red-and-white label. She passes it to Simon and me. It's nearly empty.
"What's this at the bottom?" Simon asks.
"Mouse," says Kwan. "That's why call pickle-mouse wine."
"What is it really?"
"You look." Kwan points to the bottom of the bottle. "Mouse."
We look. We see something gray. With a tail. Somewhere in my brain I know I should retch. But instead, Simon and I look at each other and we both start laughing. And then we can't stop. We laugh until we choke, clutching our aching stomachs.
"Why are we laughing?" Simon is panting.
"We must be drunk."
"You know, I don't feel drunk. I feel, well, happy to be alive."
"Me too. Hey, look at those stars. Don't they seem bigger? Not just brighter, but bigger? I feel like I'm shrinking and everything else is getting bigger."
"You see like tiny mouse," Kwan says.
Simon points to the shadows of mountains jutting above the courtyard wall. "And those," he says. "The peaks. They're huge."
We stare in silence at the mountains, and then Kwan nudges me. "Now maybe you see dragon," she says. "Two side-by-side dragon. Yes?"
I squint hard. Kwan grabs my shoulders and repositions me. "Squeeze-close eyes," she orders. "Sweep from mind American ideas. Think Chinese. Make you mind like dreaming. Two dragon, one male, one female."
I open my eyes. It's as though I'm viewing the past as the foreground, the present as a faraway dream. "The peaks going up and down," I say, tracing in the air, "those are their two spines, right? And the way the two front peaks taper into those mounds, those are their two heads, with the valley tucked between their two snouts."
Kwan pats my arm, as if I were a student who has recited her geography lessons well. "Some people think, 'Oh, village sit right next to dragon mouth—what bad _feng shui,_ no harmony.' But to my way thinking, all depend what type dragon. These two dragon very loyal, good _chi—_ how you say in English, good _chi_?"
"Good vibes," I say.
"Yes-yes, good vibe." She translates for Du Lili what we are talking about.
Du Lili breaks into a huge grin. She chatters something in Changmian and starts humming: "Daaa, dee-da-da."
Kwan sings back: "Dee, da-da-da." Then to us she says, "Okay-okay. Simon, Libby-ah, sit back down. Du Lili say I should tell you dragon love-story." We're like kindergartners around a campfire. Even Du Lili is leaning forward.
"This the story," Kwan begins, and Du Lili smiles, as if she understood English. "Long time ago two black dragon, husband and wife, live below ground near Changmian. Every springtime, wake up, rise from earth like mountain. Outside, these two dragon look like human person, only black skin, also very strong. In one day, two together can dig ditch all around village. Water run down mountain, caught in ditch. That way, no rain come, doesn't matter, plenty for grow plants. Libby-ah, what you call this kind watering, flow by itself?"
"Irrigation."
"Yes-yes. What Libby-ah say, irritation—"
"Irrigation."
"Yes-yes, irrigationing, they make this for whole village. So everybody love these two black dragon people. Every year throw big feast, celebrate them. But one day, Water God, real low-level type, he get mad—'Hey, somebody took water from my river, not asking permission.' "
"Darn." Simon snaps his fingers. "Water rights. It's always water rights."
"Yes-yes. So big fight, back and forth. Later Water God hire some wild people from other tribe, not our village, somewhere else, far away. Maybe Hawaii." She elbows Simon. "Hey. Joke, I just joke! Not Hawaii. I don't know where from. Okay, so people use arrow, kill dragon man and woman, pierce their body all over every place. Before die, crawl back inside earth, turn into dragon. See! Those two backs now look-alike six peaks. And where arrows gone in, make ten thousand cave, all twist together, lead one heart. Now when rain come, water flow through mountain, pour through holes, just like tears, can't stop running down. Reach bottom—flood! Every year do this."
Simon frowns. "I don't get it. If there's a flood every year, what's the good _chi_?"
"Tst! Flood not big flood. Only little flood. Just enough wash floor clean. In my lifetime, only one bad flood, one long drought. So pretty good luck."
I could remind her that she lived in Changmian for only eighteen years before moving to America. But why ruin her story and our good time? "What about the Water God?" I ask.
"Oh, that river—no longer. Flood wash him away!"
Simon claps and whistles, startling Du Lili out of her doze. "The happy ending. All right!" Du Lili stands up and stretches, then begins clearing away the remnants of our chicken feast. When I try to help, she pushes me down.
"So who told you that story?" I ask Kwan.
She's placing more twigs on top of the fire. "All Changmian people know. For five thousand year, every mother singing this story to little children, song call 'Two Dragon.' "
"Five thousand years? How do you know that? It couldn't be written down anywhere."
"I know, because—well, I tell you something, secret. Between two dragon, in small valley after this one, locate small cave. And this little cave lead to other cave, so big you can't believe almost. And inside that big cave—lake, big enough for boat ride! Water so beautiful you never seen, turquoise and gold. Deep, glowing too! You forget bring lamp, you still see entire ancient village by lakeside—"
"Village?" Simon comes over. "You mean a real village?"
I want to tell him it's another one of Kwan's stories, but I can't catch his eye.
Kwan is pleased by his excitement. "Yes-yes, ancient village. How old, don't know exact. But stone house still standing. No roof, but wall, little doorway to crawl inside. And inside—"
"Wait a second," Simon interrupts. "You've been in this cave, you've seen this village?"
Kwan goes on rather cockily: "Course. And inside stone house, many thing, stone chair, stone table, stone bucket with handle, two dragon carve on top. You see—two dragon! That story same age stone village. Maybe older, maybe five thousand year not right. Maybe more like ten thousand. Who know how old for sure."
A prickle of goose bumps rises along my back. Maybe she's talking about a different cave. "How many people have been to this village?" I ask.
"How many? Oh, don't know exact amount. House very small. Not too many people can live there one time."
"No, what I mean is, do people go there now?"
"Now? No, don't think so. Too scared."
"Because—"
"Oh, you don't want know."
"Come on, Kwan."
"Okay-okay! But you get scared, not my fault."
Simon leans against the water pump. "Go ahead."
Kwan takes a deep breath. "Some people say, you go inside, not just this cave, any cave this valley, never come back." She hesitates, then adds: "Except as ghost." She checks us for a reaction. I smile. Simon is transfixed.
"Oh, I get it." I try again to catch Simon's attention. "This is the Changmian curse that man mentioned yesterday."
Simon is pacing. "God! If this is true . . ."
Kwan smiles. "You think true, I'm ghost?"
"Ghost?" Simon laughs. "No, no! I meant the part about the cave itself—if _that's_ true."
"Course, true. I already telling you, I see so myself."
"I'm only asking because I read somewhere, what was it? . . . I remember now. It was in the guidebook, something about a cave with Stone Age dwellings inside. Olivia, did you read about that?"
I shake my head. And now I'm wondering if I've taken Kwan's story about Nunumu and Yiban too skeptically. "You think this is the cave?"
"No, that's some big tourist attraction closer to Guilin. But the book said that this mountain area is so riddled with caves there are probably thousands no one's ever seen before."
"And the cave Kwan's talking about might be another—"
"Wouldn't that be incredible?" Simon turns to Kwan. "So you think no one else has been there before?"
Kwan frowns. "No-no. I not saying this. Lots people been."
Simon's face falls. He rolls his eyes. Oh well.
"But now all dead," Kwan adds.
"Whoa." Simon holds up his hand, stop sign–like. "Let's see if we can get this straight." He starts pacing again. "What you're telling us is, no one _living_ knows about the cave. Except you, of course." He waits for Kwan to confirm what he's said so far.
"No-no. Changmian people know. Just don't know where locate."
"Ah!" He slowly walks around us. "No one knows where the cave is located. But they know _about_ the cave."
"Course. Many Changmian stories concerning this. Many."
"For example." He motions for Kwan to take the floor.
She furrows her brow and crinkles her nose, as if searching through her extensive repository of ghost stories, all of them secrets we would be sworn never to reveal. "Most famous ones," she says after this pause, "always concerning foreigner. When they die cause so much trouble!"
Simon nods sympathetically.
"Okay, one story go like this. This happen maybe one hundred year ago. So I didn't see, only hear Changmian people talk. Concerning four missionaries, come from England, riding in little wagons, big umbrella on top, just two mule in front pulling those fat people. Hot day too. Jump out two Bible ladies, one young and nervous, one old and bossy, also two men, one has beard, other one, oh, so fat no one from our village can believe. And these foreigners, they wearing Chinese clothes—yes!— but still look strange. Fat man, he speak Chinese, little bit, but very hard understanding him. He say something like, 'Can we do picnic here?' Everybody say, 'Welcome-welcome.' So they eat, eat, eat, eat, eat, so much food."
I interrupt Kwan. "You're talking about Pastor Amen?"
"No-no. Entirely different people. I already told you, didn't see, only hear. Anyway, after they done eating, fat man ask, 'Hey, we hear you have famous cave, ancient city inside. You show us?' Everybody make excuse: 'Oh, too far. Too busy. Nothing see.' So old Bible lady, she hold up pencil—'Whoever want it, take me see cave, you can have!' Those days, long time ago, our people never yet seen pencil—writing brush, course, but pencil, no. Course, probably Chinese people invent pencil, we invent so many things—gunpowder but not for killing, noodle too. Italian people always say they invent noodle—not true, only copy Chinese from Marco Polo time. Also, Chinese people invent zero for number. Before zero, people don't know have nothing. Now everybody have zero." Kwan laughs at her own joke. " . . . What I saying before?"
"You were talking about the Bible lady with the pencil."
"Ah, yes. In our poor village, no one seen pencil. Bible lady, she show them can make mark just like that, no need mixing ink. One young man, family name Hong—he always dreaming he better than you—he took that pencil. Today, his family still have, on altar table, same pencil cost his life." Kwan crosses her arms, as if to suggest pencil greediness deserved death.
Simon picks up a twig. "Wait a minute. We're missing something here. What happened to the missionaries?"
"Never come back."
"Maybe they went home," I reason. "Nobody saw them leave."
"That young man also don't come back."
"Maybe he became a Christian and joined the missionaries."
Kwan gives me a doubtful look. "Why someone do that? Also, why those missionaries don't take their wagons, their mules? Why Bible church later send all kinds foreign soldiers searching for them? Causing so much trouble, knock on this door, that door—'What happen? You don't tell us what, burn you down.' Pretty soon, everybody got same idea, they say, 'Oh, so sad, bandits, that's what.' And now, today, everyone still know this story. If someone acting like better than you, you say, 'Huh! You don't watch, maybe you later turn into pencil man.' "
"Hear that?" I poke Simon.
Kwan sits up straight and cocks her ear toward the mountains. "Ah, you hear?"
"What?" Simon and I say at the same time.
"Singing. Yin people singing."
We fall quiet. After a few moments, I hear a slight whishing sound. "Sounds like wind to me."
"Yes! To most people, just wind _—wu! wu!—_ blow through cave. But you have big regret, then hear yin people calling you, 'Come here, come here.' You grow more sad, they sing more louder: 'Hurry! Hurry!' You go see inside, oh, they so happy. Now you take someone place, they can leave. Then they fly to Yin World, peace at last."
"Sort of a tag-you're-it kind of place," Simon adds.
I pretend to laugh, but I'm bothered. Why does Kwan have so many stories about switching places with dead people?
Kwan turns to me. "So now you know why village name become Changmian. _Chang_ mean 'sing,' _mian_ mean 'silk,' something soft but go on forever like thread. Soft song, never ending. But some people pronounce 'Changmian' other way, rising tone change to falling, like this: _Chang._ This way _chang_ mean 'long,' _mian_ mean 'sleep.' Long Sleep. Now you understand?"
"You mean songs that put you to sleep," says Simon.
"No-no-no-no-no. Long Sleep—this another name for _death._ That's why everybody say, 'Changmian cave, don't go there. Doorway to World of Yin.' "
My head tingles. "And you believe that?"
"What believe? I already there. I know. Lots yin people stuck there, waiting, waiting."
"So why is it you were able to come back?" I catch myself before she can answer. "I know, you don't have to tell me." I don't want Kwan to go into the whole story of Buncake or Zeng now. It's late. I need sleep, and I don't want to feel I'm lying next to someone who's possessed a dead girl's body.
Simon crouches next to me. "I think we should go see this cave."
"You're kidding."
"Why not?"
"Why not! Are you nuts? People die in there!"
"You believe that stuff about ghosts?"
"Of course not! But there must be something bad in there. Gas fumes, cave-ins, who knows what else."
"Drowning," adds Kwan. "Lots sad people drown themself, fall to bottom, down, down, down."
"Hear that, Simon? Drowning, down, down, down!"
"Olivia, don't you realize? This could be an incredible find. A prehistoric cave. Stone Age houses. Pottery—"
"And bone," Kwan offers, looking helpful.
"And bones!" reiterates Simon. "What bones?"
"Mostly foreigner bone. They lose way, then lose mind. But don't want die. So lie by lakeside long time, until body become stone."
Simon stands up, facing the peaks.
I say to him: "People lose their minds in there. They turn to stone."
But Simon isn't listening anymore. I know he's mentally wending his way into the cave and into the world of fame and fortune. "Can you imagine what the magazine editors will say when they see our story? Shit! From chicken soup to major archaeological find! Or maybe we should call _National Geographic_ or something. I mean, it's not like we owe _Lands Unknown_ the rights to this story. And we should also take some of the pottery back with us as proof, definitely."
"I'm not going in there," I say firmly.
"Fine. I'll go by myself."
I want to shout, I forbid you. But how can I? I don't have an exclusive claim anymore on his body, mind, or soul. Kwan is looking at me, and I want to shout at her as well: This is your fault! You and your damn stories! She gives me that annoying sisterly look, pats my arm, trying to calm me down. I yank my arm away.
She turns to Simon. "No, Simon. Can't go youself."
He spins around. "What do you mean?"
"You don't know where cave locate."
"Yeah, but you'll show me." He states this like a fact.
"No-no, Libby-ah right, too dangerous."
Simon scratches his neck. I figure he is gathering his arguments to beat us both down, but instead he shrugs. "Well, maybe. But why don't we all sleep on it?"
I LIE in the middle of the crowded marriage bed, as stiff as Big Ma in her coffin. My limbs ache in my effort not to touch Simon. We are in the same bed for the first time in nearly ten months. He's wearing silk thermal underwear. Every now and then I feel the sharp ridge of his shins or the cleft of his butt against my thigh, and I carefully ease away, only to be rebuffed by Kwan's knees, her jabbing toes. I have the sneaking suspicion she's pushing me toward Simon.
Strange groaning sounds erupt. "What was that?" I whisper.
"I didn't hear anything," Simon answers. So he's still awake too.
Kwan yawns. "Singing from cave. I already tell you this."
"It sounds different now, like someone complaining."
She rolls over to her side. In a few minutes, she's snoring, and after a while, Simon's breathing deeply. So there I am, crammed between two people, yet alone, wide awake, staring at the dark, seeing the moments of the past twenty-four hours: The refrigerated van ride and Big Ma's ski parka. Buncake and Kwan in their coffins. The poor chicken and its death dance. The dead mouse in the wine, the dead missionaries in the cave. And Simon's face, his excitement when we looked at the dragon peaks together. That was nice, special. Was it the old feeling we once had? Maybe we could become friends. Or maybe it meant nothing. Maybe it was only the pickle-mouse wine.
I flip onto my side and Simon follows suit. I make myself as straight as a chopstick to avoid touching him. The body, however, isn't meant to be stiff and still, except in death. I long to bend my body into his, to allow myself this comfort. But if I do, maybe he'll assume too much, think that I'm forgiving him. Or admitting that I need him. He smacks his lips and snuffles—the sounds he always makes as he enters into deep sleep. And soon I can feel his breath rolling in waves on my neck.
I've always envied the way he can sleep solidly through the night, undisturbed by car alarms, earthquakes, and now, those persistent scratching sounds beneath the bed. Or is it more like sawing? Yes, that's the teeth of a saw, the sawteeth of a rat, chewing on a bedpost, sharpening its fangs before climbing up here. "Simon," I whisper, "do you hear that? Simon!" And then, as in the old days, he loops one arm over my hip and nuzzles his face against my shoulder. I instantly stiffen. Is he asleep? Did he do this by instinct? I wiggle my hip slightly to see if he'll rouse and remove his hand. He groans. Maybe he's testing me.
I lift his hand from my hip. He stirs and says in a groggy voice, "Mmm, sorry," then extricates himself, snorts, and turns over. So his embrace was an accident of sleep. He meant nothing by it. My throat tightens, my chest hurts.
I remember how he always wanted to cuddle and make love after an argument, as if connecting our bodies that way mended whatever rift was between us. I resented the easy supposition of all's well that ends well. And yet I'd resist only slightly when he'd raise my chin. I'd hold in my anger and my breath as he nibbled my lips, my nose, my brow. The more upset I was, the more places he'd woo: my neck, my nipples, my knees. And I'd let him—not because I weakened and wanted sex, but because it would have been spiteful, beyond redemption, not to allow us this hope.
I planned to talk about the problems later. How he saw avoidance as normal and I saw it as a warning. How we didn't know how to talk to each other anymore, how in protecting our own territory we were losing common ground. Before it was too late, I wanted to say that whatever love had brought us together had dwindled and now needed to be restocked. At times I feared that our love had never overflowed into plenty, that it had been enough for a few years but was never meant to last a lifetime. We mistook a snack for a recurring harvest. We were two people starved for abundant love but too tired to say so, leg-ironed together until time passed us by and we left this world, two vague hopes without dreams, just another random combination of sperm and egg, male and female, once here now gone.
I used to think these things while he undressed me, resenting the fact that he saw nakedness as intimacy. I'd let him stroke me where he knew me best, which was my body and not my heart. He'd be seeking my rhythm, saying, "Relax, relax, relax, relax." And I'd slip, let go of all that was wrong. I'd yield to my rhythm, his rhythm, our rhythm, love by practice, habit, and reflex.
In the past, after we made love, I would feel better, no longer quite as upset. I'd try to remember the worries once again—about harvests and abundance, fruitless love and hopeless death—and they were no longer feelings but notions, silly, even laughable.
Now that our marriage is over, I know what love is. It's a trick on the brain, the adrenal glands releasing endorphins. It floods the cells that transmit worry and better sense, drowns them with biochemical bliss. You can know all these things about love, yet it remains irresistible, as beguiling as the floating arms of long sleep.
## [19
THE ARCHWAY](TheHundredSecretSenses-toc.html#TOC-22)
I'm jolted out of sleep by screams—young girls being raped or killed or both! Then Du Lili's voice cries out, "Wait, wait, you greedy things." And the pigs shriek even louder as she coos: "Eat, eat. Eat and grow fat."
Before I can relax, I sense another unpleasant eye-opener. During the night, my body must have gravitated toward the nearest heat source, that being Simon. More precisely, my butt is now snuggled against the springy nest of his groin, which, I notice, is sprouting a morning erection, what we once fondly referred to as "the alarm cock." Kwan's third of the bed is empty, her indentation already cool to the touch. When did she leave? Oh yes, I know what she's up to, the sneak. And Simon, is he really asleep? Is he secretly laughing?
The awful truth is, I'm aroused. In spite of everything I thought the night before, my lower body has a pulsing, heat-seeking, rub-craving itch. And the rest of me longs for comfort. I curse myself: You have a fucking brainless burrow! The IQ of a pop-bead! I slide away from danger and hop out of Kwan's side of the bed. Simon stirs. Shivering in my nightshirt, I hurry to the foot of the bed, where I dumped my luggage yesterday. The air temperature must be forty-five degrees. My hands go trawling for warm clothes.
Simon yawns, sits up and stretches, then peels back the mosquito netting. "I slept well," he says ambiguously. "How 'bout you?"
I pull out my parka and drape it over my shoulders. It's so stiff with cold it crackles. My teeth are chattering as I speak: "So how does one take a shower or bath around here?" Simon has an amused look on his face. Does he suspect anything?
"There's a public bathhouse next to the toilet shack," he says. "I checked it out yesterday while you were shooting. It has an Esalen-spa charm to it. Gender neutral. One trough, no waiting. But I don't think anyone's used it for ages. The water's kind of scummy. And if you want a warm bath, bring a pail of hot water."
I was prepared for bad, but not incredibly bad. "They use the same bathwater—all day?"
"All week, it looks like. God, I know, we're so _wasteful_ in the States."
"What are you grinning at?" I ask.
"You. I know how obsessed you are about cleanliness."
"No I'm not."
"Oh? Then why is it that when you stay in a hotel, you pull down the bedspread first thing?"
"Because they don't get changed that often."
"So?"
"So I'm not fond of lying on top of someone else's skin flakes and dried-up bodily fluids."
"Aha! I rest my case. Now go to the bathhouse. I dare you."
For a moment, I weigh which is worse, bathing in the common broth or going funky for the next two weeks.
"Of course, you could fill a basin and take a sponge bath right here. I could be your water boy."
I pretend not to hear him. My cheek muscles are nearly spastic from trying not to smile. I pull out two pairs of leggings. I reject the thin cotton, choose the Polarfleece, regretting I didn't bring more. Simon's suggestion is a good one, the part about the sponge bath, that is. Water boy, yeah, fat chance. I can just picture it, Simon as Egyptian slave, wearing one of those twisted-cloth jockstraps, a look of excruciating desire on his face as he silently ladles warm water over my breasts, my stomach, my legs. And heartless me, I'd treat him like a faucet: More hot! More cold! Hurry up!
"By the way," he says, interrupting my thoughts, "you were talking in your sleep again."
I avoid meeting his eye. Some people snore. I sleep-talk, not in mumbles but in complete, well-articulated sentences. Nightly. Loudly. Sometimes I even wake myself up. Simon's heard me tell knock-knock jokes, order a three-course meal of desserts, shout for Kwan to keep her ghosts from me.
Simon lifts one eyebrow. "Last night, what you said was certainly revealing."
Shit. What the hell did I dream? I always remember my dreams. Why can't I now? Was Simon in the dream? Did we have sex? "Dreams don't mean anything," I say. I take out a thermal undershirt and bottle-green velour top. "They're just flotsam and jetsam."
"Don't you want to know what you said?"
"Not really."
"It relates to something you _love_ to do."
I throw down the clothes and snap, "I don't love it as much as you think!"
Simon blinks twice, then starts laughing. "Oh yes you do! Because you said, 'Simon, wait. I haven't paid for this yet!' " He allows five seconds for this to sink in. "You were shopping. What did you think I was referring to?"
"Shut up." My face is burning. I thrust my hand into the suitcase and angrily grab some woolen socks. "Turn around. I want to get dressed."
"I've already seen you naked a thousand times."
"Well, it's not going to be a thousand and one. Turn around."
With my back to him, I whip off the parka and my nightshirt, still berating myself for being taken in by him. He baited me! And what an idiot I was, going for it. I should have known he'd trick me. And then I sense something else. I spin around.
"You don't have to suck your tummy in." He's holding up the gauzy curtain. "You look great. You always have. I never get tired of looking at you."
"You shithead!"
"What! We're still married!"
I wad up a sock and throw it at him. He ducks, letting go of the mosquito netting, which must be a hundred years old, because when the sock hits, the mesh blows apart—poof!—and wispy tufts are lofted high into the air.
We both stare at the damage. I feel like a kid who's broken a neighbor's window with a baseball, wickedly thrilled.
"Uh-oh." I cover my mouth and snicker.
Simon shakes his head. "Bad girl."
"It's your fault."
"What do you mean! You threw the sock."
"You were looking!"
"I still am."
And there I am, standing stark naked, freezing my ass off.
I throw the other sock at him, then my leggings, the velour top, my nightshirt. Clutching a slipper, I fly toward Simon and strike him on the back. He grabs my hand and we both fall onto the bed, where we tussle and roll, slap and shove, so grateful finally to have this excuse to touch each other. And when we exhaust ourselves with this playful joust, we look at each other, silently, eye to eye, no smiles, nothing more to be said. All at once, we both leap, like wolf mates reunited, searching for that which identifies us as belonging to each other: the scent of our skin, the taste of our tongues, the smoothness of our hair, the saltiness of our necks, the ridges of our spines, the slopes and creases we know so well yet feel so new. He is tender and I am wild, nuzzling and nipping, both of us tumbling until we lose all memory of who we were before this moment, because at this moment we are the same.
WHEN I WALK out to the courtyard, Kwan gives me one of her innocent but knowing grins. "Libby-ah, why you smiling?"
I look at Simon. "No rain," I answer. No matter who Kwan really is, sister or not, I'm glad that she suggested we come to China.
In front of her, on the ground, is an open suitcase, stuffed with various gizmos and gadgets. According to Kwan, Big Ma bequeathed these gifts to Du Lili, everything except a wooden music box that plays a tinkly version of "Home on the Range." I pull out my camera and begin shooting.
Kwan picks up the first item. Simon and I lean forward to see. It's a Roach Motel. "In America," she explains to Du Lili with a serious face, "they call this a guesthouse for roaches." She points to the label.
"Wah!" cries Du Lili. "Americans are so rich they make toy houses for insects! Tst! Tst!" She shakes her head, her mouth turned down in proletarian disgust. I tell Simon what she said.
"Yes, and Americans feed them delicious food." Kwan peers into the motel's door. "And the food is so good, the bugs never want to leave. They stay forever."
Du Lili slaps Kwan's arm and pretends to be angry. "You're so bad! You think I don't know what this is?" She then says to me in an excited voice, "Chinese people have the same thing. We use pieces of bamboo, cut open like this and filled with sweet sap. Your big sister and I used to make them together. Our village held contests to see who could catch the most pests—the most flies, the most rats, the most roaches. Your big sister was often praised for catching the most roaches. Now she's trying to catch me with a prank."
Kwan unveils more treasures, and it's obvious that many of them have come from a sporting goods store. First, there's a day pack. "Strong enough to carry bricks, with many pockets, on the sides, underneath, here, there, see. Unzip them like this— Wah, what's in here?" She pulls out a portable water purifier, a tiny backpacking stove, a small medical kit, an inflatable cushion, resealable baggies, heavy-duty trash liners, a space blanket, and—"Wah! Unbelievable!"—even more things: a waterproof match holder, a flashlight, and a Swiss Army knife with a built-in toothpick, "very practical." Like an Avon Lady, Kwan explains the particular use of each item.
Simon examines the stash. "Amazing. How'd you think of all this?"
"Newspaper," Kwan answers. "They have article on earthquake, if big one come, this what you need for survive. In Changmian, you see, no need wait for earthquake. Already no electricity, no running water, no heat."
Next Kwan lifts from the suitcase a plastic sweater box, the kind used for storing junk under the bed, and out come gardening gloves, gel-filled insoles, leggings, towels, T-shirts. Du Lili exclaims and sighs and laments that Big Ma did not live long enough to enjoy such luxuries. I take a picture of Du Lili surrounded by her inheritance. She is wearing wraparound sunglasses and a 49ers Super Bowl cap, the word "Champs" studded in rhinestones.
After a simple breakfast of rice porridge and pickled vegetables, Kwan brings out stacks of photos that document her thirty-two years of American life. She and Du Lili sit on a bench, poring over them. "Look here," Kwan says. "This is Libby-ah, only six years old. Isn't she cute? See the sweater she's wearing? I knit it myself before I left China."
"These little foreigner girls"—Du Lili points—"who are they?"
"Her schoolmates."
"Why are they being punished?"
"Punished? They're not being punished."
"Then why are they wearing the tall dunce hats?"
"Ah-ha-ha-ha! Yes, yes, tall hats for punishing counterrevolutionaries, that's what they look like! In America, foreigners wear tall hats to celebrate birthdays, also New Year's. This is a party for Libby-ah's birthday. It's a common American custom. The schoolmates offer gifts, nothing useful, just pretty things. And the mother makes a sweet cake and puts flaming candles on top. The child plants a wish in her head, and if she can blow out the candles all at once, the wish will grow true. Then the children feast on sweet cake, guzzle sweet drinks, eat sweet candy, so much sweetness their tongues roll back and they can't swallow any more."
Du Lili rounds her mouth in disbelief. "Tst! Tst! A party for every birthday. A simple charm for a birthday wish. Why do Americans still wish so much, when they already have too much? For me, I don't even need a party. A wish once every twenty years would be enough. . . ."
Simon pulls me aside. "Let's go for a walk."
"Where to?"
He leads me out of the courtyard, then points to the archway between the mountains, the entrance to the next valley.
I wag my finger at him like a nursery school teacher. "Simon, you aren't still thinking about that cave, are you?"
He returns a phony look of offense. " _Moi?_ Of course not. I just thought it would be nice to go for a walk. We have things to talk about."
"Oh? Like what?" I say coyly.
"You know." He takes my hand, and I call out over the wall: "Kwan! Simon and I are going for a walk."
"Where?" she shouts back.
"Around."
"When you return?"
"You know, whenever."
"How I know what time worry?"
"Don't worry." And then I have second thoughts about where we might be headed. So I add, "If we're not back in two hours, call the police."
I hear her happily grumbling to Du Lili in Chinese: "She says if they're lost, telephone the police. What telephone? We have no telephone. . . ."
We walk quietly, holding hands. I'm thinking of what I should say. I'm sure Simon is doing the same. I'm not going to settle for patching things up automatically. I want a commitment to become closer, to be intimate with our minds and not just our bodies. And so with our own yet-to-be-spoken thoughts, we head in the general direction of the stone wall that separates Changmian from the next valley.
Our meandering takes us through private alleyways that interlace related compounds, and we apologize to those families who stare at us with curiosity, then apologize again when they run to their doorways, showing us coins for sale, green tarnished disks they claim are at least five hundred years old. I shoot a couple of frames and imagine a caption that would suit them: "Changmian residents staring at intruders." We peek into the open gates of courtyards and see old men coughing, smoking the stubs of cigarettes, young women holding babies, their fat cheeks bright pink from the pinching cold. We pass an old woman with a huge bundle of kindling balanced on her shoulders. We smile at children, several of whom have cleft palates or clubfeet, and I wonder if this is the result of inbreeding. We see this together, two aliens in the same world. Yet what we see is also different, because I wince at such hardship, the life that Kwan once had, that I could have had. And Simon remarks, "You know, they're sort of lucky."
"What do you mean?"
"You know, the small community, family histories linked for generations, focused on the basics. You need a house, you get your friends to help you slap a few bricks together, no bullshit about qualifying for a loan. Birth and death, love and kids, food and sleep, a home with a view—I mean, what more do you need?"
"Central heating."
"I'm serious, Olivia. This is . . . well, this is _life._ "
"You're being sentimental. This is the pits, this is basic survival."
"I still think they're lucky."
"Even if they don't think so?"
He pauses, then raises his lower lip like a bulldog. "Yeah." His smart-ass tone is begging for an argument. And then I think, What's the matter with me? Why do I have to escalate everything into a moral battle of right and wrong? The people here don't care what we think. Let it go, I tell myself.
"I guess I can see your point," I say. And when Simon smiles, the embers of my irritation are fanned once again.
The path leads up the hill. As we round the top, we spot two little girls and a boy, all around five or six years old, playing in the dirt. About ten yards beyond them is the high stone wall and its archway, blocking our view of what lies beyond. The children look up, cautious and alert, their faces and clothes covered with mud.
_"Ni hau?"_ Simon says in a flat American accent—"How are you?" being one of the few Chinese phrases he knows. Before the children notice, I bring out the Leica and fire off five shots. The children giggle, then return to their play. The boy is patting out the finishing touches on a mud fortress, his thumbprints still visible on the walls and gate. One girl is using her fingers to tweeze small blades of grass. The other girl delicately transfers the green slivers to the thatched roof of a miniature hut. And crawling near the hut are several brown grasshoppers, the captive residents of this elaborate compound. "Aren't those kids smart?" I say. "They made toys out of nothing."
"Smart and messy," Simon answers. "I'm kidding. They're cute." He points to the smaller girl. "That one sort of looks like you at age six, you know, the birthday party snapshot."
As we walk toward the archway, the children jump up. "Where are you going?" the boy says gruffly in his childish Mandarin.
"To see what's over there." I point to the tunnel. "Do you want to come?" They run ahead of us. But when they arrive at the entrance, they turn around and look at us. "Go ahead," I tell them. "You go first." They don't budge, only shake their heads solemnly. "We'll go together." I extend my hand to the smaller girl. She backs away and stands behind the boy, who says, "We can't." The bigger girl adds, "We're scared." The three of them huddle together even more closely, their huge eyes fixed on the archway.
After I translate this for Simon, he says, "Well, I'm going through now. If they don't want to come, fine." The moment he steps into the archway, the children scream, turn on their heels, and flee at top speed. "What was that about?" Simon's voice echoes in the rounded entrance.
"I don't know." My eyes follow the children until they drop behind the hill. "Maybe they've been warned not to talk to strangers."
"Come on," he calls. "What are you waiting for?"
I'm looking at the walls along the ridge. Unlike the mud-brick ones in the village, these are made out of huge blocks of cut stone. I imagine the laborers from long ago hauling them into place. How many died of exhaustion? Were their bodies used as mortar, the way workers' bodies were when the Great Wall was built? In fact, this looks like a miniature version of the Great Wall. But why is it here? Was it also built as a barrier during the days of warlords and Mongolian invaders? As I step through the archway, the pulse in my neck pounds. My head begins to float. I stop in the middle of the tunnel and put my hand on the wall. The tunnel is about five feet long and five feet high, tomblike. I imagine ghostly warring troops waiting for us on the other side.
What I see instead is a small, flat valley, a rain-soaked pasture on one side, a sectioned field on the other, with the pathway we are on continuing straight down the middle like a flat brown ribbon. Flanking both sides of the valley are dozens of loaf-shaped mountains much smaller than the two peaks ahead of us. It would be the perfect setting for a pastoral romance, if not for the fact that I can't get the faces of those scared children out of my mind. Simon is already walking down the hill.
"Do you think we're trespassing?" I say. "I mean, maybe this is private property."
He looks back at me. "In China? Are you kidding? They don't call it communist for nothing, you know. It's all public land."
"I don't think that's true anymore. People can own houses and even their own businesses now."
"Hey, don't worry. If we're trespassing, they're not going to shoot us. They'll just tell us to get out and we'll get out. Come on. I want to see what's in the valley after this one."
I keep expecting an angry farmer to come charging at us with an upraised hoe. But the lush pasture is empty, the fields are quiet. Isn't this a workday? Why isn't anyone out here? And those high stone walls, why are they there, if not to keep someone out? Why is it so deathly quiet? No sign of life, not even a peeping bird. "Simon," I start to say, "doesn't it seem, well—"
"I know, isn't it amazing, more like the fields of an English country manor, a scene out of _Howards End._ "
In an hour, we've traveled the length of the valley. We begin walking up another hill, this one steeper and rockier than the last. The path narrows to a rough trail of switchbacks. I can see the wall and the second archway above, the limestone peaks looking like sharp coral thrust up from an ancient ocean floor. Dark clouds whirl in front of the sun, and the air turns chilly. "Maybe we should head back," I suggest. "It looks like rain."
"Let's see what's at the top first." Without waiting for me to agree, Simon climbs the path. As we wind our way up, I think about Kwan's story of the missionaries, how the villagers said they were killed by bandits. Maybe there's truth in the lie. Just before we left the hotel in Guilin—when was that? only yesterday?—I picked up _The China Daily,_ the English-language newspaper. On the front page was a report that violent crime, once unheard of in China, is now on the increase, especially in tourist locales such as Guilin. In one village of only two hundred seventy-three people, five men were executed by firing squad a couple of days ago, one for rape, two for robbery, two for murder, crimes committed all in the last year. Five violent crimes, five executions—and from one tiny village! That's swift justice for you: accused, found guilty, kaboom. The newspaper further reported that the crime wave stemmed from "Western pollution and degenerative thinking." Before being executed, one of the hooligans confessed that his mind had rotted after he watched a bootlegged American movie called _Naked Gun 33 1⁄3._ He swore, however, that he was innocent of murder, that hillside bandits had killed the Japanese tourist and his crime lay only in buying the dead woman's stolen Seiko watch. Remembering this account, I do an appraisal of our robbery potential. My watch is a cheap plastic Casio. Although who knows, maybe hillside bandits crave digital watches with built-in calculators the size of thumbnails. I left my passport at Big Ma's house, thank God. I heard passports are worth about five thousand U.S. dollars each on the black market. Thieves would kill for those.
"Where's your passport?" I ask Simon.
"Right here." He pats his fanny pack. "What, you think we'll run into the border patrol or something?"
"Shit, Simon! You shouldn't carry your passport with you!"
"Why not?"
Before I can answer, we hear rustling in the bushes, followed by a clop-clop sound. I picture bandits on horseback. Simon keeps walking ahead. "Simon! Come back here."
"In a sec." He rounds the bend, out of sight.
And then I hear him yelp: "Hey, there. Whoa! Wait . . . hey, wait!" He comes scrambling down the path, yelling, "Olivia, get out—" then flies into me so hard he knocks the wind out of me. As I lie in the dirt, my mind detaches itself from my body. Strange, I'm so lucid and calm. My senses seem sharper. I examine the knot on my lower shin, the popped-out vein on my kneecap. No pain. No pain! I _know_ without any doubt or fear that this is a sign that death is around the bend. I've read this in how-to books on dying, that somehow you _know,_ although you can't explain why. The moments slow. This is the one-second flashback that dying people have, and I'm surprised how long the second is lasting. I seem to have an infinite amount of time to sum up what's been important in my life—laughter, unanticipated joy, Simon . . . even Simon. And yes, love, forgiveness, a healing inner peace, knowing I'm not leaving behind any big rifts or major regrets. I laugh: Thank God I have on clean underwear, although who in China would care? Thank God Simon is with me, that I'm not alone in this terrible yet wonderful moment. Thank God he'll be by my side later—that is, if there is a heaven or World of Yin, whatever. And if there is indeed a whatever, what if . . . what if Elza is there? Into whose angelic arms will Simon fly? My thoughts aren't quite as lucid or healing anymore, the seconds tick at their usual workaday pace, and I jump to my feet, saying to myself, Fuck this shit.
That's when they appear, our would-be assassins, a cow and her calf, so startled by my scream they skid to a mud-flying stop. "What's the matter?" Simon asks. The cow gives me a big-gummed moo. If self-humiliation were fatal, that's what I would have died of. My big spiritual epiphany is a joke on me. And I can't even laugh about it. How stupid I feel. I can no longer trust my perceptions, my judgments. I know how schizophrenics must feel, trying to find order in chaos, inventing bootstrap logic that would hold together what might otherwise further unravel.
The cow and her calf lope off. But just as we step back onto the path, a young man strides down, stick in hand. He wears a gray sweater over a white shirt, new blue jeans, and clean-white sneakers. "He must be the cow herder," says Simon.
I'm wary of making any assumptions now. "He could be a bandit, for all we know."
We stand off to the side to let him pass. But when the young man is in front of us, he stops. I keep expecting him to ask us a question, yet he says nothing. His expression is bland, his eyes intense, observant, almost critical.
_"Ni hau!"_ Simon waves, even though the guy is standing right in front of us.
The young man remains silent. His eyes flick up and down. I begin to babble in Chinese. "Are those your cows? They scared me to death. Maybe you heard me scream. . . . My husband and I, we're Americans, from San Francisco. Do you know the place? Yes? No? . . . Well, now we're visiting my sister's aunt in Changmian. Li Bin-bin."
Still no answer.
"Do you know her? Actually, she's dead. Yesterday, that's when she died, before we could meet her, a real pity. So now we want to make a, a . . ." I'm so flustered I can't think of the Chinese word for "funeral," so I say, "Make a party for her, a sad party." I laugh nervously, ashamed of my Chinese, my American accent.
He stares me straight in the eye. And I say to him in my mind, Okay, buster, if you want to play this game, I'll stare too. But after ten seconds I look down.
"What's with this guy?" Simon asks. I shrug. The cow herder does not look like other men we have seen in Changmian, the ones with cold-coarsened hands and home-chopped hair. He's groomed, his fingernails are clean. And he looks arrogantly smart. In San Francisco, he could pass for a doctoral student, a university lecturer, a depressed poet-activist. Here he's a cow herder, a cow herder who disapproves of us for reasons I can't fathom. And because of that, I want to win him over, make him smile, assure myself I'm not as ridiculous as I feel.
"We're taking a walk," I continue in Mandarin. "Having a look around. It's very pretty here. We want to see what lies between those mountains." I point toward the archway, just in case he doesn't understand.
He looks up, then turns back to us with a scowl. Simon smiles at him, then leans toward me. "He obviously doesn't know what you're talking about. Come on, let's go."
I persist. "Is that all right?" I say to the cow herder. "Do we need to receive permission from someone? Is it safe? Can you advise us?" I wonder how it would feel to be smart but to have your prospects go no further than a pasture in Changmian. Maybe he envies us.
As if he heard my thoughts, he smirks. "Assholes," he says in perfectly enunciated English, then turns and walks down the path. For a few seconds, we're too stunned to say anything.
Simon starts walking. "That was weird. What did you say to him?"
"I didn't say anything!"
"I'm not accusing you of saying anything wrong. But what did you say?"
"I said we were going for a walk. Okay? I asked if we needed permission to be here."
We trudge up the hill again, no longer holding hands. The two strange encounters, first with the kids and now with the cow herder, have put a pall on any sort of romantic talk. I try to dismiss them, but unable to make any sense of them, I worry. This is a warning. It's as clear as smelling a bad odor, knowing it leads to something rotten, dead, decayed.
Simon puts his hand on the small of my back. "What is it?"
"Nothing." Yet I long to confide in him, to have our fears if not our hopes be in synch. I stop walking. "This is going to sound silly, but actually I was wondering—maybe these things are like omens."
"What things?"
"The kids telling us not to go in here—"
"They said _they_ couldn't go in. There's a difference."
"And that guy. His evil chuckle, like he knows we shouldn't go into the next valley, but he's not going to say."
"His laugh wasn't evil. It was a laugh. You're acting like Kwan, linking two coincidences and coming up with a superstition."
I explode: "You asked me what I was thinking, and I told you! You don't have to contradict everything I say and make fun of me."
"Hey, hey, easy. I'm sorry. . . . I was only trying to put your mind at ease. Do you want to go back now? Are you really that nervous?"
"God, I hate it when you say that!"
"What! Now what'd I do?"
" ' _That_ nervous,' " I say testily. "You only say it about women and yappy little dogs. It's condescending."
"I don't mean it that way."
"You never describe men as nervous."
"All right, all right! Guilty as charged. You're not nervous, you're . . . you're hysterical! How's that?" He grins. "Come on, Olivia, lighten up. What's the matter?"
"I'm just . . . well, I'm concerned. I'm concerned we might be trespassing, and I don't want to come across as ugly Americans, you know, presuming we can do anything we want."
He puts his arm around me. "Tell you what. We're almost at the top. We'll have a quick look, then head back. If we see anyone, we'll apologize and leave. Of course, if you really are nervous, I mean, _concerned—_ "
"Would you stop!" I give him a shove. "Go on. I'll catch up."
He shrugs, then climbs the path with big strides. I stand there for a moment, mentally lashing myself for not saying what I feel. But I'm irked that Simon can't sense what I really want. I shouldn't have to spell it out like a demand, making me the bitch of the day and him the all-suffering nice guy.
When I reach the top, he is in the second archway, which is nearly identical to the first, except that it seems older, or perhaps battered. Part of the wall has caved in, and it looks as though it happened not through gradual decay but rather with the sudden force of a cannon or a ramrod.
"Olivia!" Simon shouts from the other side. "Come here. You won't believe this!"
I hurry over, and when I emerge from the archway and look down, I see a landscape that both chills and mesmerizes me, a fairy-tale place I've seen in nightmares. It is completely unlike the smooth, sunlit valley we just crossed. This is a deep and narrow ravine shaped by violent upheavals, as lumpy as an unmade bed, a scratchy blanket of moss with patches of light and pockets of shadows, the faded hues of a perpetual dusk.
Simon's eyes are glazed with excitement. "Isn't this great?"
Sprouting here and there are mounds of rocks, stacked high as men. They look like monuments, cairns, an army of petrified soldiers. Or perhaps they are the Chinese version of Lot's salty wife, pillars of human weakness, the fossilized remains of those who entered this forbidden place and dared to look back.
Simon points. "Look at those caves! There must be hundreds."
Along the walls, from the bottom of the ravine to the tops of the peaks, are cracks and fissures, pockmarks and caves. They look like the shelves and storage bins of a huge prehistoric mortuary.
"It's incredible!" Simon exclaims. I know he's thinking about Kwan's cave. He goes down a trail of sorts, more gully than path, with rocky footholds that give under his weight.
"Simon, I'm tired. My feet are starting to hurt."
He turns around. "Just wait there. I'll look around for about five minutes. Then we'll go back together. Okay?"
"No more than five minutes!" I yell. "And don't go in any caves." He is clambering down the trail. What is it that makes him so oblivious to danger? Probably one of those biological differences between men and women. Women's brains use higher and more evolved functions, which account for their sensitivity, their humaneness, their worry, whereas men rely on more primitive functions. See rock, chase. Danger, sniff, go find. Smoke cigar later. I resent Simon's carelessness. And yet, I have to admit, I find it seductive, his boyish disregard of peril, his pursuit of fun without consequences. I think about the type of men I consider sexy: they are always the ones who climb the Himalayas, who paddle through alligator-infested jungle rivers. It isn't that I consider them brave. They are reckless, unpredictable, maddeningly unreliable. But like rogue waves and shooting stars, they also lend thrills to a life that otherwise would be as regular as the tide, as routine as day passing into night.
I look at my watch. Five minutes have dragged by. Then it becomes ten, fifteen, twenty. Where the hell is Simon? The last time I saw him he was making his way toward a cluster of those cairns, or whatever they are. He walked behind a bush, and then I couldn't follow where he went. A raindrop hits my cheek. Another splatters on my jacket. In an instant, it's pouring. "Simon!" I yell. "Simon!" I expect my voice to echo loudly, but it is muffled, absorbed by the rushing rain. I duck under the archway. The rain is falling so fast and hard it makes a hazy curtain. The air smells metallic, minerals flushed from rocks. The peaks and hillsides are turning dark, glistening. Rain streams along their sides, and small rivulets send loose rocks down. Flash flood, what if there's a flash flood? I curse Simon for making me worry about him. But within another minute, my worry spills into panic. I have to leave the shelter and search for him. I pull the parka hood over my head, step into the downpour, and march toward the slope.
I'm counting on altruistic courage to seize hold of me and guide me down. Yet when I lean toward the dark ravine, fear slips into my veins, paralyzes my limbs. My throat tightens and I plead out loud: "Please, dear God, or Buddha, whoever's listening. Make him return right now. I can't stand it anymore. Make him return, and I promise—"
Simon's face pops into view. His hair, his down jacket, his jeans are soaking wet, and there he is, panting like a dog who wants to play more fetch. My one second of gratitude dissolves into anger.
We run for the archway. Simon slips off his jacket and twists the sleeves, releasing a small flood onto the ground. "Now what are we going to do?" I grouse.
"Keep each other warm." His teeth are chattering. He leans against the tunnel wall, then pulls me toward him, my back against his chest, his arms wrapped around me. His hands are icy. "Come on, relax." He rocks me slightly. "There, that's better."
I try to recall the morning's lovemaking, the unexpected joy, the welling up of emotion we both had. But throughout my body, bundles of muscles tighten and cramp—my jaw, my chest, my forehead. I feel restrained, stifled. I ask myself, How can I relax? How can I let go of everything that's happened? You need complete trust to do that.
And then a bad thought lands on my brain: Has Simon slept with other women since our separation? Of course he has! The guy can't go without sex for more than a couple of days. One time—this was a few years ago—we came across a magazine questionnaire, "Your Lover's Inner Sex Life," something stupid like that. I read the first question out loud to Simon: "How often does your lover masturbate?" And I was mentally checking "never or seldom" when Simon said, "Three or four times a week. It depends."
"It depends?" I blurted. "On what? Sunny weather?"
"Boredom, as much as anything," he answered. And I thought, Twice a week with me is _boring_?
And now I'm wondering, How many women has he been _bored_ by since we broke up?
Simon massages my neck. "God, you're so tight in here. Can you feel that?"
"Simon, this morning, you know?"
"Hmm, that was nice."
"But don't you think we should have used a condom?" I'm hoping he'll say, "What for? I'm shooting blanks. You know that." But instead he catches his breath. His fingers stop kneading. And then he rubs one of my arms briskly.
"Hmm. Yeah. I guess I forgot."
My eyes lock. I try to breathe calmly. I'm going to ask him. But whatever he says next, I can take it. Besides, I'm not so holy myself. I slept with that creepy marketing director, Rick—or rather, we groped, never needed the condom lying on the nightstand, because "the big bruiser," as my date called his flaccid penis, went on strike, something it had never done, he assured me, ever. And of course, I felt sexually humiliated, especially after pretending with well-timed sighs and shivers that I was aroused.
Simon's mouth is near my ear. His breathing reminds me of the rushing sea sounds you can hear in a nautilus, a memory now endlessly trapped in a spiral.
"Simon, about the condom—are you saying you slept with someone else?"
His breathing stops. He lifts his head away from my ear. "Well . . . well, if I did, I don't remember anymore." He squeezes me. "Anyway, they don't matter. Only you." He strokes my hair.
"They? How many is _they_?"
"Uh . . . Hell, I don't know."
"Ten? Twelve?"
He laughs. "Gimme a break."
"Three? Four?"
He's quiet. I'm quiet. He exhales, shifts his posture slightly. "Hmm. Maybe something like that."
"Which is it? Three or four?"
"Olivia, let's not talk about this. It'll only get you upset."
I pull away from him. "I'm already upset. You slept with four other women and you didn't even bother to use a goddamn condom this morning!" I walk to the opposite side of the archway and glare at him with my shit-detector eyes.
"It was three." He's looking at his feet. "And I was careful. I didn't catch anything. I used a condom every time."
" _Every_ time! Boxes and boxes of condoms! How thoughtful of you to be thinking of me."
"Come on, Olivia. Stop it."
"Who were they? Someone I know? Tell me." And then I think of a woman I despise, Verona, a free-lance art director we hired last year for one project. Everything about her was fake, her name, her eye-lashes, her fingernails, her breasts. I once told Simon her breasts were too symmetrical to be real. He laughed and said, "Well they sure squished like the real thing." And when I asked how he knew that, he said that whenever they looked at the layouts together, she leaned over his shoulder and pressed her tits into his back. "Why didn't you say something?" I asked. He said that that would only have called more attention to the fact she was flirting, that it was better to just ignore it, since he wasn't going to do anything about it anyway.
"Was one of them Verona?" I hold my arms tightly across my chest in an effort to stop trembling. He opens his mouth slightly, then closes it, resigned. "You did, didn't you? You fucked that bitch."
"I'm not saying that, you are."
I'm crazed. "So tell me, were they real? Did she have squishy tits?"
"Come on, Olivia. Quit. Why on earth is this so important to you? It shouldn't mean a damn thing."
"It means you never intended to get back together with me! It means I can't trust you. I've never been able to trust you." I'm raging, drowning, needing to take Simon down with me. "I've never been important to you! I only tricked myself into thinking I was. And Kwan tricked you with her stupid ghost tricks, that séance. Yeah, remember that? What Elza said? You were supposed to forget her, move on with your life. And you know what? Kwan made that up. She lied! I told her to."
Simon gives a small laugh. "Olivia, you're acting crazy. Do you actually think I _believed_ that séance shit? I thought we were both just humoring Kwan."
I'm sobbing: "Right, barrels of laughs . . . Only it wasn't a joke, Simon, she was there! I swear, she was, I saw her. And you know what she was saying? Forget her? Fuck no! She was begging you to forget _me._ She said to wait—"
Simon claps his hand to his forehead. "You just never give up, do you?"
" _Me_ give up? You've never given _her_ up!"
Simon's eyes narrow. "You want to know what the real problem is? You use Elza as a scapegoat for all your insecurities. You've made her a bigger deal in your life than she ever was in mine. You never even knew her, but you project every doubt about yourself onto her. . . ."
I put my hands over my ears. And as he continues to heap his pseudo-analytic garbage on me, I rack my brain for another weapon, a final, fatal bullet to the heart. And that's when I recall secretly reading some letters Elza had written Simon, their pet names, their youthful promises. I turn to him. "You think I'm crazy? Well, maybe I am, because I can see her right now! Yeah, Elza! She's standing right in front of you. She just said, 'Angel Buns, what do you mean I was no big deal?' " Simon's face freezes. "You were supposed to wait, we were supposed to plant those trees together, one for every year."
Simon tries to cover my mouth with his hand. I jump back.
"Don't you see?" I wail. "She's here! She's in your head. She's in your heart! She's always here, right now, in this stupid fucking place, with her stupid fucking omens, telling us we're doomed, Simon, we're doomed!"
At last, Simon has a stricken look on his face I've never seen before. It scares me. He's shaking. Drops are running down his cheeks—are they rain or tears?
"Why are you doing this?" he howls.
I turn and race out of the archway, into the rain. I run across the valley, gasping, pushing my heart to burst. By the time I reach Big Ma's house, the rain has stopped. I walk through the courtyard, and Kwan gives me one of her knowing looks.
"Libby-ah, oh, Libby-ah," she moans. "Why you crying?"
## [20
THE VALLEY OF STATUES](TheHundredSecretSenses-toc.html#TOC-23)
Simon still isn't back. I look at my watch. An hour has gone by. I figure he's fuming by himself. Fine, let him freeze his ass out there. It's not quite noon. I pull out a paperback and climb into the bed. The trip to China is now a debacle. Simon will have to leave. That makes the most sense. After all, he doesn't speak any Chinese. And this is Kwan's village, and she's _my_ sister. As for the magazine piece, I'll just have to take notes from here on out and find someone back home to polish it into an article.
Kwan calls out that it's time for lunch. I muster my composure, ready to face the Chinese inquisition. "Where Simon?" she'll ask. "Ai-ya, why you fight too much?" Kwan is in the central room, putting a steaming bowl on the table. "See? Tofu, tree ear, pickled green. You want take picture?" I have no desire to eat, or shoot photos. Du Lili bustles into the room with a pot of rice and three bowls. We begin to eat, or rather, they do, eagerly, critically.
"First not salty enough," Kwan complains. "Now too salty." Is this some sort of veiled message about Simon and me? A few minutes later she says to me, "Early this morning big sun, now look, rain come back." Is she making a sneaky analogy to my fight with Simon? But throughout the rest of the meal, she and Du Lili don't even mention his name. Instead, they gossip animatedly about people in the village, thirty years' worth of marriages and diseases, unexpected tragedies and hilarious outcomes, all of which I have no interest in. My ear is attuned to the gate, waiting for the creak and slam of Simon's return. I hear only the meaningless spatter of rain.
After lunch, Kwan says she and Du Lili are going to the community hall to visit Big Ma. Do I want to come? I imagine Simon returning to the house, searching for me, growing uneasy, worried, maybe even frantic. Shit, he wouldn't worry, that's what _I_ do. "I think I'll stay here," I tell Kwan. "I need to reorganize my camera gear and enter some notes on what I've taken so far."
"Okay. You get done later, come visit Big Ma. Last chance. Tomorrow we do funeral."
When I'm finally alone, I sort through my baggies of film, checking for moisture. Damn this weather! It's so damp and chilly that even with four layers of clothing, my skin feels clammy, my toes practically numb. Why did I let pride take precedence over warm clothes?
Before we left for China, Simon and I discussed what we should bring. I packed a large suitcase, a duffel, and my camera bag. Simon said he had two carry-ons, and then he goaded me: "By the way, don't count on me to lug your extra junk." I shot back: "Who asked you to?" And he retaliated with another taunt: "You never _ask,_ you _expect._ " After that remark, I decided I wouldn't let Simon help me—even if he _insisted._ Like a pioneer faced with a dead team of oxen and a desert to cross, I took a long, hard look at my travel inventory. I was determined to reduce my luggage to complete self-sufficiency: one wheeled carry-on and my camera bag. I pitched everything that was not absolutely essential. Out went the portable CD player and CDs, the exfoliant, skin toner, and rejuvenating cream, the hair dryer and conditioner, two pairs of leggings and matching tunic tops, half my stash of underwear and socks, a couple of novels I'd been meaning to read for the last ten years, a bag of prunes, two out of three rolls of toilet paper, a pair of fleece-lined boots, and the saddest omission, a purple down vest. In deciding what should go in my allotted space, I bet on tropical weather, hoped for the occasional night of Chinese opera, and didn't even question whether there would be electricity.
And so, among the things I packed and now resent seeing in my tiny suitcase are two silk tanktops, a pair of canvas shorts, a clothes steamer, a pair of sandals, a swimsuit, and a neon-pink silk blazer. The only opera I'll wear that to is the soap going on in my own little courtyard. At least I have the waterproof shell. Small consolation, big regret. I long for the down vest the way a person adrift in the sea deliriously dreams of water. Warmth—what I would give for it! Damn this weather! Damn Simon for being nice and toasty in his own down jacket. . . .
His down jacket—it's drenched, sopping and useless. Just before I left him, he was shaking, with anger, I thought at the time. Now I wonder. Oh God! What are the signs of hypothermia? A vague memory about cold and anger strays across my mind. When was that, five or six years ago?
I was shooting photos in an emergency room, the usual dramatic stock for a hospital's annual report. A team of paramedics wheeled in a shabbily dressed woman who reeked of urine. Her speech was slurred, she complained she was burning up and had to take off a mink coat she didn't have. I assumed she was drunk or high on drugs. And then she started convulsing. "Defibrillator!" someone shouted. I later asked one of the nurses what I should put in the caption—heart attack? alcoholism? "Put down that she died of January," the nurse said angrily. And because I didn't understand, he said, "It's January. It's cold. She died of hypothermia, just like six other homeless people this month."
That won't happen to Simon. He's healthy. He's always too warm. He rolls down the car window when other people are freezing, and he doesn't even ask. He's inconsiderate that way. He keeps people waiting. He doesn't even think they might be worried. He'll be here any minute. He'll arrive with that irritating grin of his, and I'll be pissed for worrying without reason.
After five minutes of trying to convince myself of these things, I run to the community hall to find Kwan.
IN THE TUNNEL of the second archway, Kwan and I find Simon's jacket crumpled on the ground like a broken corpse. Stop whimpering, I tell myself. Crying means you expect the worst.
I stand at the top of the ledge that leads into the ravine and look down, searching for movement. Various scenarios play in my head. Simon, now delirious, wandering half clothed in the ravine. Rocks tumbling down from the peaks. The young man, who is not a cow herder at all but a modern-day bandit, stealing Simon's passport. I blurt out to Kwan: "We ran into some kids, they screamed at us. And later this guy with some cows, he called us assholes. . . . I was tense. I went a little crazy, and Simon . . . he was trying to be nice, but then he got mad. And what I said, well, I didn't mean it." In the coved tunnel, my words sound confessional and hollow at the same time.
Kwan listens quietly, sadly. She doesn't say anything to take away my guilt. She doesn't respond with false optimism that all will be well. She unzips the day pack that Du Lili insisted we bring along. She spreads the space blanket on the ground, inflates the cushion, lays out the little camping stove and an extra canister of fuel.
"If Simon returns to Big Ma's house," she reasons in Chinese, "Du Lili will send someone to let us know. If he comes to this place, you will be here to help him get warm." She opens her umbrella.
"Where are you going?"
"A short look around, that's all."
"What if you get lost too?"
_"Meiyou wenti."_ Don't worry, she's telling me. "This is my childhood home. Every rock, every twist and turn in the hills, I know them all like old friends." She steps outside, into the drizzle.
I call to her: "How long will you be gone?"
"Not long. Maybe one hour, no more."
I look at my watch. It's almost four-thirty. At five-thirty, the golden half-hour will come, but now dusk scares me. By six, it will be too dark to walk.
After she leaves, I pace between the archway's two openings. I look out on one side, see nothing, then repeat the process on the opposite side. You're not going to die, Simon. That's fatalistic bullshit. I think of people who beat the odds. The lost skier at Squaw Valley who dug a snow cave and was saved three days later. And that explorer who was trapped on an ice floe—was it John Muir?—who did jumping jacks all night long to stay alive. And of course, there was that Jack London tale about a man caught in a blizzard who manages to build a fire out of wet twigs. But then I remember the ending: A clump of snow slides off the branch above and extinguishes his hope below. And then other endings come to mind: The snowboarder who fell into a tree well and was found dead the next morning. The hunter who sat down to rest one day on the Italian–Austrian border and wasn't discovered until spring thaw thousands of years later.
I try meditating to block out these negative thoughts: palms open, mind open. But all I can think about is how cold my fingers are. Is that how cold Simon is?
I imagine myself as Simon, standing in the same archway, overheated from our argument, muscles tight, wanting to bolt in any dangerous direction. I've seen that happen before. When he learned that our friend Eric had been killed in Vietnam, he went meandering alone and wound up lost in the eucalyptus groves of the Presidio. The same thing happened when we visited some friends of friends in the country, and one man started telling racist jokes. Simon stood up and announced that the guy had his head screwed on wrong. At the time, I was angry that he had created a scene and left me to deal with the aftermath. But now, recalling this moment, I feel a mournful admiration for him.
The rain has stopped. That's what he too must be seeing. "Hey," I imagine him saying, "let's check out those rocks again." I walk outside to the ledge, look down. He wouldn't see stomach-churning steepness the way I do. He wouldn't see a hundred ways to crack your skull open. He'd just walk down the trail. So I do. Did Simon go this way? About halfway down, I look back, then around. There isn't any other way into this place, unless he threw himself over the ledge and dropped seventy feet to the bottom. Simon isn't suicidal, I tell myself. Besides, suicidal people talk about killing themselves before they do it. And then I remember reading a story in the _Chronicle_ about a man who parked his new Range Rover on the Golden Gate Bridge during rush hour, then threw himself over the railing. His friends expressed the usual shock and disbelief. "I saw him at the health club just last week," one was reported as saying. "He told me he held two thousand shares of Intel stock at twelve that were now trading at seventy-eight. Man, he was talking about the _future._ "
Toward the bottom of the ravine, I check the sky for the amount of daylight still left. I see dark birds fluttering like moths; they fall suddenly, then flap upward again. They're making shrill, high-pitched noises, the sounds of frightened creatures. Bats—that's what they are! They must have escaped from a cave, now out for a flight at dusk, the hour of insects. I saw bats in Mexico once _—mariposas,_ the waiters called them, butterflies, so as not to scare the tourists. I wasn't afraid of them then, nor am I now. They are harbingers of hope, as welcome as the dove that brought a leafy twig to Noah. Salvation is nearby. Simon is nearby too. Perhaps the bats soared out because he entered their lair and disturbed their upside-down slumber.
I follow the twisty, uneven path, trying to see where the bats are coming from, where they return. My foot slides, and I wrench my ankle. I hobble over to a rock and sit down. "Simon!" I expect my yelling to carry as if in an amphitheater. But my cries are sucked into the hollows of the ravine.
At least I'm not cold anymore. There's hardly any wind down here. The air is still, heavy, almost oppressive. That's strange. Isn't the wind supposed to blow _faster_? What was in that brochure Simon and I did on Measure J, the one against Manhattanization—the Bernoulli effect, how forests of skyscrapers create wind tunnels, because the smaller volume through which air passes decreases pressure and increases velocity—or does it increase pressure?
I look at the clouds. They're streaming along. The wind is definitely blowing up there. And the more I watch, the more unsteady the ground feels, like the bottom of a salad spinner. And now the peaks, the trees, the boulders grow enormous, ten times larger than they were a minute ago. I stand and walk again, this time careful of my footing. Although the ground appears level, it's as if I were climbing a steep incline. A force seems to push me back. Is this one of those places on earth where the normal properties of gravity and density, volume and velocity have gone haywire? I grab on to the cracks of a rocky mound and strain so hard to pull myself up I'm sure a blood vessel in my brain will burst.
And then I gasp. I'm standing on a crest. Below is an abrupt drop of twenty feet or so, as if the earth collapsed like a soufflé, creating a giant sinkhole. Stretching out to the mountains at the other end of the ravine is a bumpy wasteland pincushioned with those things I saw earlier—cairns, monuments, whatever they are. It alternately resembles a petrified forest of burnt trees and a subterranean garden of stalagmites from a former cave. Did a meteor fall here? The Valley of the Shadow of Death, that's what this is.
I go up to one of the formations and circle it like a dog, then circle it again, attempting to make sense of it. Whatever this is, it sure does not grow that way naturally. Someone deliberately stacked these rocks—and at angles that don't look balanced. Why don't the rocks fall? Large boulders perch on the points of small spires. Other rocks tilt on dime-size spots, as if they were iron filings latched on to a magnet. They could pass for modern art, sculptures of lamps and hat racks, precisely planned to give them a haphazard look. On one pile the topmost rock looks like a misshapen bowling ball, its holes suggesting vacant eyes and a screaming mouth, like the person in that Edvard Munch painting. I see other formations with the same features. When were these made? By whom and why? No wonder Simon wanted to come down here. He came back to investigate further. As I continue walking, this strange gathering of rocks resembles more and more the blackened victims of Pompeii, Hiroshima, the Apocalypse. I'm surrounded by an army of these limestone statues, bodies risen from the calcified remains of ancient sea creatures.
A dank, fusty odor hits my nose and panic rises in my throat. I look around for signs of decay. I've smelled this same stench before. But where? When? It feels overwhelmingly familiar, an olfactory version of déjà vu—déjà senti. Or perhaps it's instinctive, like the way animals know that smoke comes from fire and fire leads to danger. The odor is trapped in my brain as visceral memory, emotional residue of stomach-cramping fear and sadness, but without the reason that caused it.
I hurry past another stack of rocks. But my shoulder catches a jutting edge, and I scream as the entire load collapses. I stare at the rubble. Whose magic did I just destroy? I have the uneasy feeling I have broken a spell and these metamorphisms will soon begin to sway and march. Where is the archway? Now there seem to be more rock mounds—have they multiplied?—and I must weave through this maze, my legs going one direction, my mind arguing that I should go another. What would Simon do? Whenever I've become uncertain about accomplishing a physical feat, he's been the reasoning voice, assuring me I can run another half-mile, or hike to the next hill, or swim to the dock. And there were times in the past when I believed him, and was grateful that he believed in me.
I imagine Simon urging me on now. "Come on, Girl Scout, move your ass." I look for the stone wall and archway to orient myself. But nothing is distinct. I see only gradations of flat-light shadows. Then I remember those times I became angry with Simon because I listened to him and failed. When I yelled at him after I tried rollerblading and fell on my butt. When I cried because my backpack was too heavy.
I sit on the ground in exasperation, whimpering. Fuck this, I'm calling a taxi. Look how dopey I've become. Do I really believe I can stick my hand up, hail a taxi, and get out of this mess? Is that all that I've managed to store in the emergency section of my inner resources—my willingness to pay cab fare? Why not a limo? I must be losing my mind!
"Simon! Kwan!" Hearing the panic in my voice, I grow even more panicked. I try to move more quickly but my body feels heavy, pulled to the earth's core. I bump into one of the statues. A rock topples, grazes my shoulder. And just like that, all the terror I've been holding in bubbles out of my mouth and I begin to cry like an infant. I can't walk. I can't think. I sink to the ground and clutch myself. I'm lost! They're lost! All three of us are trapped in this terrible land. We'll die here, rot and slough, then petrify and become other faceless statues! Shrill voices accompany my screams. The caves are singing, songs of sorrow, songs of regret.
I cover my ears, close my eyes, to shut out the craziness of the world, my mind, both. You can make it stop, I tell myself. I'm straining hard to believe this; I can feel a cord in my brain stretching taut, and then it rips and I'm soaring, free of my body and its mortal fears, growing light and giddy. So this is how people become psychotic, they simply let go. I can see myself in a boring Swedish movie, slow to react to painfully obvious ironies. I howl like a madwoman at how ridiculous I look, how stupid it is to die in a place like this. And Simon will never know how _nervous_ I became. He's right. I'm hysterical!
A pair of hands grabs my shoulders, and I yell.
It's Kwan, her face full of worry. "What happen? Who you talking to?"
"Oh God!" I jump up. "I'm lost. I thought you were too." I'm sniffling and babbling between staccato breaths. "I mean, are we? Are we lost?"
"No-no-no," she says. I notice then that she has a wooden box tucked under one arm and balanced on her hip. It looks like an old chest for silverware.
"What's that?"
"Box." With her free hand, Kwan helps me onto my rubbery legs.
"I _know_ it's a box."
"This way." She guides me by the elbow. She says nothing about Simon. She is strangely solemn, unusually taciturn. And fearing that she might have bad news to tell me, I feel my chest tighten.
"Did you see—" She cuts me off by shaking her head. I'm relieved, then disappointed. I no longer know how I should feel from moment to moment. We're edging our way past the strange statues. "Where'd you get that box?"
"Found it."
I am beyond being frustrated. "Really!" I snap. "I thought you bought it at Macy's."
"This my box I hide long time ago. Already tell you this. This box always want show you."
"Sorry. I'm just frazzled. What's inside?"
"We go up there, open and see."
We walk quietly. As my fear ebbs, the landscape begins to look more benign. The wind pushes against my face. I was perspiring earlier, and now I'm growing chilled. The path is still uneven and tricky, but I no longer sense any strange gravitational pull. I berate myself: Girl, the only thing haywire in this place is your mind. I went through nothing more dangerous than a panic attack. Rocks, I was scared of rocks.
"Kwan, what are these things?"
She stops and turns around. "What thing?"
I gesture to one of the piles.
"Rocks." She starts walking again.
"I _know_ they're rocks. I mean, how did they get here, what are they supposed to be? Do they mean something?"
She stops once more, casts her eye over the gully. "This secret."
The hair rises on the back of my neck. I put some casual bluff in my voice. "Come on, Kwan. Are they like gravestones? Are we walking across a cemetery or something? You can tell me."
She opens her mouth, about to answer. But then a stubborn look crosses her face. "I tell you later. Not now."
"Kwan!"
"After we back." She points to the sky. "Dark soon. See? Don't waste time talk." And then she adds softly: "Maybe Simon already come back."
My chest swoops with hope. She knows something I don't, I'm sure of it. I concentrate on this belief as we thread our way up and around several boulders, down a gully, then past a high-walled crevasse. Soon we are at the small trail leading to the top. I can see the wall and the archway.
I scramble ahead of Kwan, heart pounding. I'm convinced Simon is there. I believe that the forces of chaos and uncertainty will allow me another chance to make amends. At the top, my lungs are nearly bursting. I'm dizzy with joy, crying with relief, because I feel the clarity of peace, the simplicity of trust, the purity of love.
And there!—the day pack, the stove, the damp jacket, everything as we left it, nothing more or less. Fear nicks my heart, but I cling to the absolute strength of faith and love. I walk to the other end of the tunnel, knowing that Simon will be there, he _has_ to be.
The ledge is empty, nothing out there but the wind, its slap. I lean against the tunnel wall and slump to the ground, hugging my knees. I look up. Kwan's there. "I'm not leaving," I tell her. "Not until I find him."
"I know this." She sits on top of the wooden chest, opens the day pack, and takes out a glass jar of cold tea and two tins. One contains roasted peanuts, the other fried fava beans. She cracks open a peanut and offers it to me.
I shake my head. "You don't need to stay. I know you have to get ready for Big Ma's funeral tomorrow. I'll be okay. He'll probably show up soon."
"I stay with you. Big Ma already tell me, delay funeral two three day, still okay. Anyway, more time cook food."
An idea hits me. "Kwan! Let's ask Big Ma where Simon is." As soon as I say this, I realize how desperate I've become. This is how parents of dying children react, turning to psychics and New Age healers, anything as long as there is a thread of possibility somewhere in this universe or the next.
Kwan gives me a look so tender I know I have hoped for too much. "Big Ma doesn't know," she says quietly in Chinese. She pulls off the cup that covers the camping stove, and lights the burner. Blue flames shoot through tiny slots with a steady hiss. "Yin people," she now says in English, "not know everything, not like you thinking. Sometimes they lost themself, don't know where should go. That's why some yin people come back so often. Always looking, asking, 'Where I lose myself? Where I go?' "
I'm glad Kwan can't see how dejected I feel. The camp stove throws only enough light to outline us as shadows. "You want," she says softly, "I ask Big Ma help us go look. We make like FBI search party. Okay, Libby-ah?"
I'm touched by her eagerness to help me. That's all that makes sense out here.
"Anyway, no funeral tomorrow. Big Ma have nothing else can do." Kwan pours cold tea into the metal stove cup and sets this on the burner. "Of course, I can't ask her tonight," she says in Chinese. "It's already dark—ghosts, they scare her to death, even though she's a ghost herself. . . ."
I absently watch blue and orange flame tips licking at the bottom of the metal cup.
Kwan holds her palms toward the stove. "Once a person has the bad habit of being scared of ghosts, it's hard to break. Me, I'm lucky, I never started this habit. When I see them, we just talk like friends. . . ."
At that moment, a dreadful possibility grips me. "Kwan, if you saw Simon, I mean, as a yin person, you would tell me, wouldn't you? You wouldn't pretend—"
"I don't see him," she answers right away. She strokes my arm. "Really, I'm telling the truth."
I allow myself to believe her, to believe that she wouldn't lie and he isn't dead. I bury my head in the nest of my arms. What should we do next, what rational, efficient plan should we use in the morning? And later, say by noon, if we still haven't found him, then what? Should one of us call the police? But then I remember there are no phones, no car. Maybe I could hitchhike and go directly to the American consulate. Is there a branch in Guilin? How about an American Express office? If there is, I'll lie and tell them I'm a Platinum Card member, charge me for whatever is needed, search and rescue, emergency airlift.
I hear scraping sounds and raise my head. Kwan is poking the Swiss Army knife into the keyhole on the front of the chest. "What are you doing?"
"Lost key." She holds up the knife, searching among its tools for a better implement. She chooses the plastic toothpick. "Long time ago, I put many thing inside here." She inserts the toothpick into the hole. "Libby-ah, flashlight in bag, you get for me, okay?"
With the light, I can see that the box is made of a dark reddish wood and trimmed in tarnished brass. On the lid is a bas-relief carving of thick trees, a Bavarian-looking hunter, a small dead deer slung over his shoulders, and a dog leaping in front of him.
"What's in there?"
There's a click and Kwan sits up. She smiles, gestures toward the box. "You open, see youself."
I grasp a small brass latch and slowly pull up the lid. Tinkling sounds burst out. Startled, I let the lid drop. Silence. It's a music box.
Kwan titters. "Hnh, what you thinking—ghost in there?"
I lift the lid again, and the plucked sounds of a silvery tune fill our small tunnel, sounding jarringly cheerful, a jaunty military march for prancing horses and people in bright costumes. Kwan hums along, obviously familiar with the melody. I aim the light toward the interior of the box. In one corner, under a panel of glass, is the apparatus that makes the music, a metal comb brushing against the pins of a rolling cylinder. "It doesn't sound very Chinese," I tell Kwan.
"Not Chinese. German-made. You like music?"
"Very pleasant." So this is the source of her music box story. I'm relieved to know there's at least some basis for her delusions. I too hum along with the tune.
"Ah, you know song?"
I shake my head.
"I once give you music box, wedding present. Remember?"
Abruptly, the music stops; the tune hangs in the air a few seconds before it fades away. There is only the awful hissing of the stove, a reminder of rain and cold, of Simon's being in danger. Kwan slides open a wooden panel in the box. She takes out a key, inserts it into a slot, and begins cranking. The music resumes, and I'm grateful for its artificial comfort. I glance at the section of the box that is now exposed. It's a knickknack drawer, a catchall for loose buttons, a frayed ribbon, an empty vial—things once treasured but eventually forgotten, things meant to be repaired, then put aside for too long.
When the music stops again, I wind the box myself. Kwan is examining a kidskin glove, its fingers permanently squeezed into a brittle bunch. She puts it to her nose and sniffs.
I pick up a small book with deckled edges. _A Visit to India, China, and Japan_ by Bayard Taylor. Inserted between two pages is a bookmark of sorts, the torn flap of an envelope. A phrase on one of the pages is underlined: "Their crooked eyes are typical of their crooked moral vision." What bigot owned this book? I turn over the envelope flap. Written in brown ink is a return address: Russell and Company, Acropolis Road, Route 2, Cold Spring, New York. "Did this box belong to someone named Russell?" I ask.
"Ah!" Kwan's eyes grow big. "Russo. You remember!"
"No." I point the flashlight at the envelope flap. "It says here 'Russell and Company.' See?"
Kwan seems disappointed. "At that time, I didn't know English," she says in Chinese. "I couldn't read it."
"So this box belonged to Mr. Russell?"
_"Bu-bu."_ She takes the envelope flap and stares at it. "Ah! Russell. I thought it was 'Russo' or 'Russia.' The father worked for a company named Russell. His name was . . ." Kwan looks me in the eye. "Banner," she says.
I laugh. "Oh, right. Like Miss Banner. Of course. Her father was a merchant seaman or something."
"Opium boat."
"Yeah, I remember now. . . ." And then the oddity of this strikes me, that we're no longer talking about a bedtime story of ghosts. Here is the music box, here are the things that supposedly belonged to them. I can barely speak. "This was Miss Banner's music box?"
Kwan nods. "Her first name—ai-ya!—now it's run out of my head." She reaches into the knickknack drawer and removes a small tin. "Tst! Her name," she keeps saying to herself, "how can I forget her name?" From the tin, she removes a small black brick. I think it is an inkstone, until she pinches off a piece and adds it to the tea, now boiling on the stove.
"What's that?"
"Herb." She switches to English. "From special tree, new leaf only, very sticky. I make for Miss Banner myself. Good for drinking, also just for smelling. Loosen you mind. Make you feel peace. Maybe give me memory back."
"Is this from the holy tree?"
"Ah! You remember!"
"No. I remember the story you told." My hands are shaking. I have a terrible craving for a cigarette. What the hell is going on? Maybe I have become as crazy as Kwan. Maybe the water in Changmian is contaminated with a hallucinogen. Or maybe I've been bitten by a Chinese mosquito that infects the brain with insanity. Maybe Simon isn't missing. And I don't have things in my lap that belong to a woman from a childhood dream.
The mist and sharp scent of the tea waft upward. I hover over the cup, the steam dampening my face, close my eyes, and inhale the fragrance. It has a calming effect. Maybe I am actually asleep. This is a dream. And if it is, I can pull myself out. . . .
"Libby-ah, look."
Kwan gives me a hand-stitched book. The cover is made of soft, floppy suede, sepia-colored. OUR SUSTENANCE, it says in embossed gothic lettering. Traces of gold flake rim the bottoms of the letters. As I turn the cover, bits of endpaper crumble off and I see from the exposed leather underneath that the now faded covers were once a somber purple, a color that reminds me of a Bible picture from childhood: a wild-looking Moses, standing on a boulder against a purple sky, breaking the tablets in front of a crowd of turban-headed heathens.
I open the book. On the left side of one page is a message typeset in cramped, uneven lines: "Trust in the Lord delivers us from temptations of the Devil. If you are overflowing with the Spirit, you cannot be fuller." On the opposite page are the typeset words "The Amen Corner." And beneath that, in a scrawl full of ink smears and sputters, is a quirky list: "Rancid beans, putrid radishes, opium leaf, pigweed, shepherd's purse, artemisia, foul cabbage, dried seeds, stringy pods, and woody bamboo. Much served cold or adrift in a grim sea of castor oil. God have mercy." The pages that follow contain similar juxtapositions, Christian inspiration related to thirst and salvation, hunger and fulfillment, answered by an Amen Corner listing foods that the owner of this journal obviously found offensive yet useful for heretic humor. Simon would love seeing this. He can use it in our article.
"Listen." I read aloud to Kwan: " 'Canine cutlets, bird fricassee, stewed holothuria, worms, and snakes. A feast for honourable guests. In the future, I shall strive to be less than honourable!' " I put the journal down. "I wonder what holothuria is."
"Nelly."
I look up. "Holothuria means Nelly?"
She laughs, spanks my hand lightly. "No-no-no! Miss Banner, her first name Nelly. But I always call her Miss Banner. That's why almost don't remember whole name. Ha. What bad memory! Nelly Banner." She chuckles to herself.
I grip the journal. My ears are ringing. "When did you know Miss Banner?"
Kwan shakes her head. "Exact date, let me see—"
_"Yi ba liu si."_ I recall the Chinese words from one of Kwan's bedtime stories. "Lose hope, slide into death. One eight six four."
"Yes-yes. You good memory. Same time Heavenly King lose Great Peace Revolution."
The Heavenly King. I remember that part as well. Was there actually someone called the Heavenly King? I wish I knew more about Chinese history. I rub my palm on the soft cover of the journal. Why can't they make books like this today?—books that feel warm and friendly in your hands. I turn to another page and read the entry: " 'Biting off the heads of lucifer matches (agonizing). Swallowing gold leaf (extravagant). Swallowing chloride of magnesium (foul). Eating opium (painless). Drinking unboiled water (my suggestion). Further to the topic of suicide, Miss Moo informed me that it is strictly forbidden among Taiping followers, unless they are sacrificing themselves in the battle for God.' "
_Taiping. Tai_ means "great." _Ping_ means "peace." _Taiping,_ Great Peace. That took place—when?—sometime in the mid–nineteenth century. My mind is being pulled and I'm resisting, but barely hanging on. In the past, I've always maintained enough skepticism to use as an antidote to Kwan's stories when necessary. But now I'm staring at sepia ink on yellowed paper, a tarnished locket, the bunched glove, the cramped letters: OUR SUSTENANCE. I'm listening to the music, its lively, old-fashioned melody. I examine the box to see if there is any indication of a date. And then I remember the journal. On the back of the title page, there it is: Glad Tidings Publishers, MDCCCLIX. In Latin, damn it! I corral the letters into numbers: 1859. I flip open the Bayard Taylor book: G. P. Putnam, 1855. So what do these dates prove? That doesn't mean Kwan knew someone named Miss Banner during the Taiping Rebellion. It's just coincidence, the story, the box, the dates on the book.
But in spite of all my logic and doubt, I can't dismiss something larger I know about Kwan: that it isn't in her nature to lie. Whatever she says, she believes is true. Like what she said about Simon, that she hadn't seen him as a ghost, which means he's alive. I believe her. I have to. Then again, if I believe what she says, does that mean I now believe she has yin eyes? Do I believe that she talks to Big Ma, that there actually is a cave with a Stone Age village inside? That Miss Banner, General Cape, and One-half Johnson were real people? That she was Nunumu? And if that's all true, the stories she told throughout these years . . . well, she must have told me for a reason.
I know the reason. I've known since I was a child, really I have. Long ago I buried that reason in a safe place, just as she had done with her music box. Out of guilt, I listened to her stories, all the while holding on to my doubts, my sanity. Time after time, I refused to give her what she wanted most. She'd say, "Libby-ah, you remember?" And I'd always shake my head, knowing she hoped I would say, "Yes, Kwan, of course I remember. I was Miss Banner. . . ."
"Libby-ah," I hear Kwan say now, "what you thinking?"
My lips are numb. "Oh. You know. Simon. I keep thinking, and everything I think about gets worse and worse."
She scoots over so that we are sitting side by side. She massages my cold fingers, and instant warmth flows through my veins.
"How 'bout we talk? Nothing to talk about, that's what we talk. Okay? Talk movie we seen. Talk book you reading. Or talk weather— no-no, not this, then you worry again. Okay, talk political things, what I vote, what you vote, maybe argue. Then you don't think too much."
I'm confused. I return a half-smile.
"Ah! Okay. Don't talk. I talk. Yes, you just listen. Let see, what I talk about? . . . Ah! I know. I tell you story of Miss Banner, how she decide give me music box."
I catch my breath. "Okay. Sure."
Kwan switches to Chinese: "I have to tell you this story in Mandarin. It's easier for me to remember that way. Because when this happened, I couldn't speak any English. Of course, I didn't speak Mandarin then, only Hakka, and a little bit of Cantonese. But Mandarin lets me think like a Chinese person. Of course, if you don't understand a word here and there, you ask me, I'll try to think of the English word. Let me see, where should I start? . . .
"Ah, well, you already know this about Miss Banner, how she was not like other foreigners I knew. She could open her mind to different opinions. But I think sometimes this made her confused. Maybe you know how that is. You believe one thing. The next day, you believe the opposite. You argue with other people, then you argue with yourself. Libby-ah, do you ever do that?"
Kwan stops and searches my eyes for an answer. I shrug and this satisfies her. "Maybe too many opinions is an American custom. I think Chinese people don't like to have different opinions at the same time. We believe one thing, we stick to it for one hundred years, five hundred years. Less confusion that way. Of course, I'm not saying Chinese people never change their minds, not so. We can change if there's a good reason. I'm just saying we don't change back and forth, right and left, whenever we like, just to be interesting. Actually, maybe today, Chinese people are changing too much, whichever way the money is blowing, that's the direction they'll chase."
She nudges me. "Libby-ah, don't you think that's true? In China today people grow more capitalist ideas than pigs. They don't remember when capitalism was the number-one enemy. Short memory, big profits."
I respond with a polite laugh.
"Americans have short memories too, I think. No respect for history, only what's popular. But Miss Banner, she had a good memory, very unusual. That's why she learned to speak our language so quickly. She could hear something just once, then repeat it the next day. Libby-ah, you have a memory like this—don't you?—only it's with your eyes not your ears. What do you call this kind of memory in English? . . . Libby-ah, are you asleep? Did you hear what I asked?"
"Photographic memory," I answer. She's pressing all the buttons now. She's not going to let me hide this time.
"Photographic, yes. Miss Banner didn't have a camera, so she was not photographic, but she did have the memory part. She always could remember what people said, just like a tape recorder. Sometimes this was good, sometimes bad. She could remember what people said at dinnertime, how they said something completely different the next week. She remembered things that bothered her, could not let them out of her mind. She remembered what people prayed for, what they got instead. Also, she was very good at remembering promises. If you made her a promise, oh, she would never let you forget. This was like her memory specialty. And she also remembered promises she made to other people. To some people, making a promise is not the same thing as doing the promise. Not Miss Banner. To her a promise was forever, not just one lifetime. Like the vow she made to me, after she gave me the music box, when death marched toward us. . . . Libby-ah, where you going?"
"Fresh air." I walk to the archway, trying to push out of my mind what Kwan is telling me. My hands are trembling, and I know it isn't because of the cold. This is the promise Kwan always talked about, the one I never wanted to hear, because I was afraid. Of all times, why does she have to tell me this now? . . .
And then I think: What am I afraid of? That I might believe the story is true—that I made a promise and kept it, that life repeats itself, that our hopes endure, that we get another chance? What's so terrible about that?
I survey the night sky, now clear of rain clouds. I remember another night long ago with Simon, when I said something stupid about the night sky, how the stars were the same that the first lovers on earth had seen. I had been hoping with all my soul that someday he would love me above all others, above all else. But it was for just a brief moment, because my hope felt too vast, like the heavens, and it was easier to be afraid and keep myself from flying out there. Now I'm looking at the heavens again. This is the same sky that Simon is now seeing, that we have seen all our lives, together and apart. The same sky that Kwan sees, that all her ghosts saw, Miss Banner. Only now I no longer feel it is a vacuum for hopes or a backdrop for fears. I see what is so simple, so obvious. It holds up the stars, the planets, the moons, all of life, for eternity. I can always find it, it will always find me. It is continuous, light within dark, dark within light. It promises nothing but to be constant and mysterious, frightening and miraculous. And if only I can remember to look at the sky and wonder about this, I can use this as my compass. I can find my way through chaos no matter what happens. I can hope with all my soul, and the sky will always be there, to pull me up. . . .
"Libby-ah, you thinking too much again? Should I talk more?"
"I was just wondering."
"Wondering what?"
I keep my back to her, still searching the sky, finding my way from star to star. Their shimmer and glow have traveled a million light-years. And what I now see is a distant memory, yet as vibrant as life can ever be.
"You and Miss Banner. Did you ever look at the sky together on a night like this?"
"Oh yes, many times." Kwan stands up and walks over to me. "Back then, we had no TV of course, so at night the only thing to do was watch the stars."
"What I mean is, did you and Miss Banner ever have a night like this, when you were both scared and you didn't have any idea what would happen?"
"Ah . . . yes, this is true. She was scared to die, scared also because she had lost someone, a man she loved."
"Yiban."
Kwan nods. "I was scared too. . . ." She pauses before saying in a hoarse whisper: "I was the reason he wasn't there."
"What do you mean? What happened?"
"What happened was—ah, maybe you don't want to know."
"Is it . . . is it sad?"
"Sad, yes, happy too. Depends on how you remember."
"Then I want to remember."
Kwan's eyes are wet. "Oh, Libby-ah, I knew someday you would remember with me. I always wanted to show you I really was your loyal friend." She turns away, gathers herself, then squeezes my hand and smiles. "Okay, okay. Now this is a secret. Don't tell anyone. Promise, Libby-ah. . . . Ah yes, I remember the sky was dark, hiding us. Between those two mountains over there, it was growing brighter and brighter. A big orange fire was burning. . . ."
And I listen, no longer afraid of Kwan's secrets. She's offered me her hand. I'm taking it freely. Together we're flying to the World of Yin.
## [21
WHEN HEAVEN BURNED](TheHundredSecretSenses-toc.html#TOC-24)
Earlier I was with Yiban, in the cave—the one with the glowing lake, a stone village by its shore. And when I was there, Libby-ah, I did a terrible thing, that led to another. I made my last day on earth a day of lies.
First, I broke my promise to Miss Banner. I did so to be kind. I told Yiban the truth: "Miss Banner was pretending with Cape. She wanted to protect you, make sure you were safe. And see, now here you are."
You should have seen his face! Relief, joy, rage, then alarm—like the turning of leaves, all the seasons happening at once. "What good is it to have me alive when she's not with me?" he cried. "I'll kill that bastard Cape." He jumped up.
"Wah! Where are you going?"
"To find her, bring her here."
"No, no, you must not." And then I told my first lie of the day: "She knows how to come here. She and I have been here many times." Inside, I was worried about Miss Banner, because of course this was not true. So then I told my second lie. I excused myself, saying I needed a girl's privacy, meaning I had to find a dark place to pee. I picked up the lantern, because I knew that if I took this, Yiban would not be able to find his way out of the cave. And then I hurried through the twists and turns of the tunnel, vowing I would bring Miss Banner back.
As I climbed out of the mountain's womb, I felt I was being born into the world again. It was day, but the sky was white not blue, drained of color. Around the sun was a ring of many pale colors. Had the world already changed? What lay beyond these mountains—life or death?
When I reached the archway just above Changmian, I saw the village was there, the crowded marketplace, everything looking the same as before. Alive! Everyone was alive! This gave me hope about Miss Banner, and I cried. As I hurried down the path, I bumped into a man leading his buffalo ox. I stopped him, told him the news, and asked him to warn his family and others: "Remove all signs of 'The Good News,' God, and Jesus. Speak quietly and do not cause alarm. Otherwise the soldiers will see what we are doing. Then disaster will visit today instead of tomorrow."
I ran toward other people and said the same thing. I banged on the gate to the roundhouses where Hakkas lived, ten families under one roof. I went quickly from household to household. Hah! I thought I was so clever, warning the village in such a calm and orderly way. But then I heard a man shouting, "Death is coming for you, you shit-eating worm!" And his neighbor cried back, "Accuse me, will you? I'll tell the Manchus you're bastard brother to the Heavenly King."
At that instant _—ki-kak!—_ we all heard it, like the cracking of dry wood. Everyone fell quiet. Then came another crack, this one like the splitting of a tall tree at its thick trunk. Nearby, a man howled, "Guns! The soldiers are already here!" And in an instant people began spilling out of their houses, grabbing on to the sleeves of those fleeing down the street.
"Who's coming?"
"What! A death warrant for all Hakkas?"
"Go! Go! Find your brothers. We're running away!"
The warnings turned into shouts, the shouts into screams, and above that, I could hear the high-pitched wails of mothers calling for their children. I stood in the middle of the lane, bumped by people running this way and that. Look what I had done! Now the entire village would be killed with a single volley of gunfire. People were climbing into the mountains, spreading across like stars in the sky.
I raced down the lane, toward the Ghost Merchant's House. Then came another gunshot, and I knew it had come from within those walls. When I reached the back alleyway gate, there was another explosion, this one echoing through the lanes. I darted inside the back courtyard, then stood still. I was breathing and listening, then listening to my breathing. I scampered to the kitchen, pressed my ear against the door that led into the dining room. No sounds. I pushed the door open, ran to the window facing the courtyard. From there I could see the soldiers by the gate. What luck!—they were sleeping. But then I looked again. One soldier's arm was twisted, the other's leg was bent. Ai! They were dead! Who did this? Had they angered Cape? Was he now killing everybody? And where was Miss Banner?
When I turned down the corridor toward her room, I saw a man's naked body, smashed facedown on the ground. Flies were feasting in the fresh gourd of his brains. Ai-ya! Who was this unlucky person? Dr. Too Late? Pastor Amen? I crept past the body, as if he might awake. A few steps later, I saw last night's dinner, the shank bone now brown with hair and blood. General Cape must have done this. Who else had he killed? Before I could wonder too much longer, I heard sounds coming from God's House. The music box was playing, and Pastor was singing, as if this Seventh Day were like any other. As I hurried across the courtyard toward God's House, Pastor's singing turned to sobs, then the bellow of an animal. And above this, I heard Miss Banner— still alive!—scolding as if she were talking to a naughty child. But a moment later, she began to wail, "No, no, no, no!" before a big explosion cut her off. I raced into the room, and what I saw made my body turn to stone, then sand. By the altar, lying bent and crooked—Miss Banner in her yellow dress, the Jesus Worshippers in shiny Sunday black—like a butterfly and four beetles squashed dead on the stone floor. Wah! Gone so fast—I could still hear their cries echoing in the room. I listened more carefully. These were not echoes but— "Miss Banner?" I called. She lifted her head. Her hair was unbound, her mouth a silent dark hole. Blood was spattered on her bosom. Ai, maybe she really was dead.
"Miss Banner, are you a ghost?"
She moaned like one, then shook her head. She held out her arm. "Come help me, Miss Moo. My leg is broken."
As I walked toward the altar, I thought the other foreigners would rise too. But they remained still, holding hands, forever sleeping in pools of bright blood. I squatted beside her. "Miss Banner," I whispered, searching the corners of the room. "Where is Cape?"
"Dead," she answered.
"Dead! Then who killed—"
"I can't bear to talk about that now." Her voice was shaky, nervous, which of course made me wonder if she— But no, I couldn't imagine Miss Banner killing anyone. And then I heard her ask with a scared face: "Tell me, quick. Yiban—where is Yiban?"
When I said he was safe in a cave, her face sagged with relief. She sobbed, unable to stop. I tried to soothe her. "Soon you'll be reunited with him. The cave is not so far away."
"I can't walk even one step. My leg." She lifted her skirt, and I saw her right leg was swollen, a piece of bone sticking out. Now I told my third lie: "This is not so bad. Where I grew up, a person with a leg like this could still walk all over the mountain, no problem. Of course, being a foreigner, you are not as strong. But as soon as I find a way to bind your leg, we'll escape from here."
She smiled, and I was grateful to know that a person in love will believe anything as long as it gives hope. "Wait here," I said. I ran to her room and searched through the drawer containing her private ladies' things. I found the stiff garment she used for pinching in her waist and pushing up her bosom, also her stockings with the holes at the heels. I ran back and used these clothes to splint her leg. And when I was done, I helped her stand and limp to the bench at the back of God's House. Only then, away from those who were alive just a short while before, was she able to say how and why each person was killed.
She began by telling me what happened after Lao Lu lost his head and I fell senseless to the ground. The Jesus Worshippers, she said, joined hands and sang the music box song: "When Death turns the corner, our Lord we shall meet."
"Stop singing!" Cape then ordered. And Miss Mouse—you know how she was always so nervous—she shouted at Cape, "I don't fear you or death, only God. Because when I die, I'm going to heaven like this poor man you killed. And you, bastard of the devil, you'll roast in hell." Yes! Can you imagine Miss Mouse saying that? If I had been there, I would have cheered.
But her words did not frighten Cape. "Roast?" he said. "I'll show you what the devil likes to roast." He called his soldiers: "Cut off this dead man's leg and cook it over a fire." The soldiers laughed, thinking this was a joke. Cape barked out the order again, and the soldiers leapt forward to obey. The foreigners cried and tried to leave. How could they watch such an evil sight? Cape growled that if they didn't watch and laugh, each of their right hands would be next to go over the fire. So the foreigners stayed and watched. They laughed and vomited at the same time. Everyone was scared to death of Cape, everyone except Lao Lu, since he was already dead. And when he saw his leg turning on a spit— well, how much can a ghost stand before he turns to revenge?
Early in the morning, before the sun came up, Miss Banner heard a knock on her door. She rose and left Cape sleeping soundly in her bed. From outside, she heard an angry voice. It sounded familiar yet not. It was a man, shouting in the Cantonese of rough workingmen. "General Fake! General Fake! Get up, you lazy dog! Come and see! Brother Jesus has arrived. He's come to drag your carcass to hell." Wah! Who could this be? Certainly not one of the soldiers. But who else sounded like a coarse-talking _kuli_?
Cape then cursed: "Damn you, man, I'll kill you for ruining my sleep."
The Chinese voice yelled back: "Too late, you son of a bastard dog. I'm already dead."
Cape jumped out of bed, grabbed his pistol. But when he threw open the door, he began to laugh. There was Pastor Amen, the crazy man. He was cursing like a fifth-generation _kuli,_ the shank bone from last night's dinner balanced on his shoulder. Miss Banner thought to herself, How strange that Pastor can now speak the native tongue so well. Then she rushed to the door to warn the madman to go away. When Cape turned to push her back, Pastor swung the shank bone and cracked open the fake general's skull. He struck him again and again, his swings so wild that one of them caught Miss Banner on the shin. Finally Pastor threw the bone down and shouted at his enemy, long past dead: "I'll kick you with my good leg when we meet in the other world."
That's when Miss Banner suspected Lao Lu's ghost had jumped into Pastor's empty mind. She watched this man who was both living and dead. He picked up Cape's pistol and ran across the courtyard, and called to the soldiers guarding the gate. From where Miss Banner lay, she heard one explosion. Soon another came. And then she heard Pastor cry in his foreigner's tongue: "Dear God! What have I done?" All that noise had wakened him from his cloudy dreams.
Miss Banner said that when she saw Pastor next, he had the face of a living ghost. He staggered toward his room, but came across Cape's body first, then Miss Banner with her broken leg. She cowered as if he would strike her again.
For many hours, Pastor and the other Jesus Worshippers discussed what had happened, what they must do. Miss Banner listened to their talk of doom. If the Manchus saw what Pastor had done, Miss Mouse pointed out, he and the rest of them would be tortured alive. Which of them had the strength to lift the bodies and bury them? None. Should they run away? To where? There was no place they knew of where they could hide. Then Dr. Too Late suggested they end their suffering by killing themselves. But Mrs. Amen argued, "Taking our own lives would be a great sin, the same as murdering someone else."
"I'll put us all to rest," said Pastor. "I'm already condemned to hell for killing those three. At least let me be the one to deliver you to peace."
Only Miss Banner tried to persuade them against this idea. "There's always hope," she said. And they told her that any hope now lay beyond the grave. So she watched as they prayed in God's House, as they ate Mrs. Amen's stale Communion bread, as they drank water, pretending it was wine. And then they swallowed Dr. Too Late's pills to forget all their pains.
What happened after that you already know.
Miss Banner and I had no strength to bury the Jesus Worshippers. Yet we could not leave them as an easy meal for hungry flies. I went to the garden. I pulled down the white clothes I had washed the day before. I thought about all the terrible things that had happened during the time the laundry had changed from wet to dry. As I wrapped our friends in those hurry-up funeral shrouds, Miss Banner went to their rooms and tried to find a remembrance of each to put in her music box. Since Cape had already stolen their treasures, all that was left were pitiful scraps. For Dr. Too Late, it was a little bottle that once contained his opium pills. For Miss Mouse, a leather glove she always clutched when in fear. For Mrs. Amen, the buttons she popped off her blouses when she sang out loud. For Pastor Amen, a travel book. And for Lao Lu, the tin with leaves from the holy tree. She placed these things in the box, along with the album where she wrote her thoughts. Then we lighted the altar candles that had melted to stubs. I took from my pocket the key Miss Banner had given me the night before. I wound the box, we played the song. And Miss Banner sang the words the foreigners loved so much.
When the song was over, we prayed to their God. This time I was sincere. I bowed my head. I closed my eyes. I said out loud, "I lived with them for six years. They were like my family, although I didn't know them very well. But I can honestly say they were loyal friends of your son, also to us. Please welcome them to your home. Pastor too."
HOW MUCH TIME did we have before the Manchus would come? I did not know then, but I can tell you now. It was not enough.
Before we escaped, I tore off the skirt of Miss Banner's everyday dress and made a sling for the music box. I threw this over my left shoulder, Miss Banner leaned on my right, and we two hobbled out as one. But when we reached the door to leave God's House, a sudden wind blew past us. I turned around and saw the clothes of the Jesus Worshippers billowing as if their bodies were renewed with breath. Stacks of "The Good News" scattered, and those papers that flew on top of the burning candles burst into flame. Soon I could smell the Ghost Merchant, chili and garlic, very strong, as if a welcome-home banquet were being prepared. And maybe this was imagination that springs from too much fear. But I saw him—Miss Banner did not—his long robes, and beneath that, his two new feet in thick-soled shoes. He was walking and nodding, finally back in his unlucky house.
Hop by hop, Miss Banner and I climbed into those mountains. Sometimes she would stumble and land on her bad leg, then cry, "Leave me here. I can't go on."
"Stop this nonsense," I would scold each time. "Yiban is waiting, and you've already made us late." That was always enough to make Miss Banner try again.
At the top of the first archway, I looked back at the empty village. Half of the Ghost Merchant's House was on fire. A great black cloud was growing above it, like a message to the Manchus to hurry to Changmian.
By the time we reached the second archway, we heard the explosions. There was no way to hurry ourselves except in that place where our stomachs churned. It was growing dark, the wind had stopped. Our clothes were sweat-drenched from our struggle to come this far. Now we had to climb along the rocky side of the mountain, where one misstep could send us tumbling into the ravine. "Come, Miss Banner," I urged. "We're almost there." She was looking at her bad leg, now swollen to the size of two.
I had an idea. "Wait here," I told her. "I'll hurry to the cave where Yiban is. Then the two of us can carry you in." She grasped my hands, and I could see in her eyes that she was frightened of being left alone.
"Take the music box," she said. "Put it in a safe place."
"I'm coming back," I answered. "You know this, don't you?"
"Yes, yes, of course. I only meant you should take it now so that there is less to carry later on." I took her box of memories and staggered forward.
At each cave or crevice I passed, a voice would cry out, "This one's already taken! No room!" That's where people in the village had gone. The caves were plugged up with fear, a hundred mouths holding their breath. I climbed up, then down, searching for the cave hidden by a rock. More explosions! I began to curse like Lao Lu, regretting every wasted moment going by. And then—at last!—I found the rock, then the opening, and lowered myself in. The lamp was still there, a good sign that other people had not come in and that Yiban had not gone out. I put down the music box and lighted the lamp, and groped slowly through the twisty bowels of the cave, hoping with each step that my exhausted mind would not take me the wrong way. And then I saw the glow ahead, like dawn light in a trouble-free world. I burst into the room with the shining lake, crying, "Yiban! Yiban! I'm back. Hurry, come and help Miss Banner! She's standing outside, between safety and death."
No answer. So I called again, this time louder. I walked around the lake. A dozen fears pinched my heart. Had Yiban tried to make his way out and gotten lost? Had he fallen in the lake and drowned? I searched near the stone village. What was this? A wall had been knocked over. And along another part of the ledge, blocks of stone had been piled high. My eye traveled upward, and I could see where a person could grab here, step up there, all the way to a crack in the roof, an opening wide enough for a man to squeeze through. And I could see that through that hole all our hopes had flown out.
When I returned, Miss Banner was sticking her head out of the archway, calling, "Yiban, are you there?" When she saw I was alone, she cried, "Ai-ya! Has he been killed?"
I shook my head, then told her how I had broken my promise. "He's gone to find you," I said in a sorry voice. "This is my fault." She did not say what I was thinking: that if Yiban had still been in that cave, all three of us could have been saved. Instead, she turned, limped to the other side of the archway, and searched for him in the night. I stood behind her, my heart in shreds. The sky was orange, the wind tasted of ash. And now we could see small dots of light moving through the valley below, the lanterns of soldiers, bobbing like fireflies. Death was coming, we knew this, and it was terrible to wait. But Miss Banner did not cry. She said, "Miss Moo, where will you go? Which place after death? Your heaven or mine?"
What a peculiar question. As if I could decide. Didn't the gods choose for us? But I did not want to argue, on this, our last day. So I simply said, "Wherever Zeng and Lao Lu have gone, that's where I'll go too."
"That would be your heaven, then." We were quiet for a few moments. "Where you are going, Miss Moo, do you have to be Chinese? Would they allow me in?"
This question was even stranger than the last! "I don't know. I have never talked to anyone who has been there and back. But I think if you speak Chinese, maybe this is enough. Yes, I'm sure of it."
"And Yiban, since he is half-and-half, where would he go? If we choose the opposite—"
Ah, now I understood all her questions. I wanted to comfort her. So I told her the last lie: "Come, Miss Banner. Come with me. Yiban already told me. If he dies, he will meet you again, in the World of Yin."
She believed me, because I was her loyal friend. "Please take my hand, Miss Moo," she said. "Don't let go until we arrive."
And together we waited, both happy and sad, scared to death until we died.
## [22
WHEN LIGHT BALANCES WITH DARK](TheHundredSecretSenses-toc.html#TOC-25)
By the time Kwan finishes talking, the pinpricks of stars have dulled against the lightening sky. I stand at the ridge, searching among the bushy shadows for movement of any kind.
"You remember how we die?" Kwan asks from behind.
I shake my head, but then recall what I always thought was a dream: spears flashing by firelight, the grains of the stone wall. Once again, I can see it, feel it, the chest-tightening dread. I can hear the snorting of horses, their hooves stamping impatiently as a rough rope falls upon my shoulder blades, then scratches around my neck. I'm gulping air, the vein in my neck is pumping hard. Someone is squeezing my hand—Kwan, but I am surprised to see she is younger and has a patch over one eye. I'm about to say don't let go, when the words are jerked from my mouth, and I soar into the sky. I feel a snap, and my fears fall back to earth as I continue to rush through the air. No pain! How wonderful to be released! And yet I'm not, not entirely. For there is Kwan, still holding tight on to my hand.
She squeezes it again. "You remember, ah?"
"I think we were hanged." My lips are sluggish in the morning chill.
Kwan frowns. "Hang? Hmm. Don't think so. Back then, Manchu soldier don't hang people. Take too much trouble. Also, no tree."
I'm strangely disappointed to be told I'm wrong. "Okay, so how'd it happen?"
She shrugs. "Don't know. That's why ask you."
"What! _You_ don't remember how we died?"
"Happen so fast! One minute stand here, next minute wake up there. Long time already gone by. By time I realize, I already dead. Same like when I gone hospital, got electric shock. I wake up, Hey, where am I? Who knows, last lifetime maybe lightning come down, send you me fast to other world. Ghost Merchant think he die same way. _Pao!_ Gone! Only two feet leave behind."
I laugh. "Shit! I can't believe you told me this whole story, and you don't know the ending?"
Kwan blinks. "Ending? You die, that's not end story. That only mean story not finish. . . . Hey, look! Sun almost come up." She stretches her legs and arms. "We go find Simon now. Bring flashlight, blanket too." She plows ahead, sure of the way back. I know where we are headed: the cave, where Yiban promised he would remain, where I hope Simon might have gone.
We hike along the crumbly ridge, gingerly testing each foothold before bearing down with our full weight. My cheeks sting as warmth floods them. At last I'll see this damn cave that is both curse and hope. And what will we find? Simon, shivering but alive? Or Yiban, forever waiting for Miss Banner? While thinking this, I trip on a shifting pile of scree and land on my rear.
"Careful!" Kwan cries.
"Why do people say careful after it's too late?" I pick myself up.
"Not too late. Next time, maybe you don't fall. Here, take hand."
"I'm fine." I flex my leg. "See. No broken bone." We continue climbing, Kwan looking back at me every few seconds. Soon I come upon a cave. I peer in, search for signs of former life, prehistoric or more recently departed. "Say, Kwan, what became of Yiban and the people from Changmian?"
"I was already dead," she answers in Chinese, "so I don't know for sure. What I know comes from gossip I heard during this lifetime. So who knows what's true? People from other villages always added a bit of their own exaggeration and let the rumors trickle down the mountain like a roof leak. At the bottom, everybody's hearsay turned into one ghost story, and from there it spread throughout the province that Changmian was cursed."
"And—what's the story?"
"Ah, wait a little, let me catch my breath!" She sits on top of a flat boulder, huffing. "The story is this. People say that when the Manchu soldiers came, they heard people crying in the caves. 'Come out!' they ordered. No one did—would you? So the soldiers gathered dried twigs and dead bushes, then placed them near the mouths of those caves. When the fires started, the voices in the caves began to scream. All at once, the caves breathed a huge groan, then vomited a black river of bats. The sky was thick with the flying creatures, so many it was as though the ravine had been darkened by an umbrella. They fanned the fire, and then the whole valley burst into flame. The archway, the ridge— everywhere was surrounded by a burning wall. Two or three soldiers on horses got away, but the rest could not. One week later, when another troop came to Changmian, they found no one, either dead or alive. The village was empty, so was the Ghost Merchant's House, no bodies. And in the ravine, where the soldiers had gone, there was nothing but ash and the stone pilings of hundreds of graves." Kwan stands up. "Let's keep walking." And she's off.
I hurry after her. "The villagers died?"
"Maybe, maybe not. One month later, when a traveler from Jintian passed through Changmian, he found the village full of life on a busy market day. Dogs were lying in the gutter, people were arguing, children were toddling behind their mothers, as if life there had passed from one day to the next without any interruption. 'Hey,' the traveler said to the village elder, 'what happened when the soldiers marched into Changmian?' And the elder wrinkled his face and said, 'Soldiers? I have no memory of soldiers coming here.' So the traveler said, 'What about that mansion there? It's been blackened by fire.' And the villagers said, 'Oh, that. Last month, the Ghost Merchant came back and threw a banquet for us. One of the ghost chickens roasting on the stove flew up to the roof and set the eaves on fire.' By the time the traveler returned to Jintian, there was a pathway of people, from the top of the mountain to the bottom, all saying that Changmian was a village of ghosts. . . . What? Why are you laughing?"
"I think Changmian became a village of liars. They _let_ people think they were ghosts. Less trouble than going to the caves during future wars."
Kwan slaps her hands together. "What a smart girl. You're right. Big Ma told me a story once about an outsider who asked a young man from our village, 'Hey, are you a ghost?' The man frowned and swept out his arm toward an ungroomed field of rocks: 'Tell me, could a ghost have grown such a fine crop of rice?' The outsider should have realized the man was fooling him. A real ghost wouldn't brag about rice. He'd lie and say peaches instead! Ah?"
Kwan waits for me to acknowledge the logic of this. "Makes sense to me," I lie in the best Changmian tradition.
She goes on: "After a while, I think the village grew tired of everyone thinking they were ghosts. No one wanted to do business in Changmian. No one wanted their sons or daughters to marry into Changmian families. So later they told people, 'No, we're not ghosts, of course not. But there's a hermit who lives in a cave two mountain ridges over. He might be a ghost, or perhaps an immortal. He has long hair and a beard like one. I've never seen him myself. But I heard he appears only at dawn and dusk, when light balances with dark. He walks among the graves, looking for a woman who died. And not knowing which grave is hers, he tends them all."
"Were they talking about . . . Yiban?" I hold my breath.
Kwan nods. "Maybe this story started when Yiban was still alive, waiting for Miss Banner. But when I was six years old—this was shortly after I drowned—I saw him with my yin eyes, among the graves. By then, he really was a ghost. I was in this same ravine, gathering dried twigs for fuel. At the half-hour when the sun goes down, I heard two men arguing. I wandered among the graves and found them stacking rocks. 'Old uncles,' I said to be polite, 'what are you doing?'
"The bald one was very bad-tempered. 'Shit!' he said. 'Use your eyes, now that you have two. What do you think we're doing?' The longhaired man was more polite. 'See here, little girl,' he said. He held up a stone shaped like the head of an ax. 'Between life and death, there is a place where one can balance the impossible. We're searching for that point.' He placed the stone carefully on top of another one. But they both fell and struck the bald man on the foot.
" 'Fuck!' cursed the bald man. 'You nearly chopped off my leg. Take your time. The right place isn't in your hands, you fool. Use your whole body and mind to find it.' "
"That was Lao Lu?"
She grins. "Dead over a hundred years and still cursing! I found out that Lao Lu and Yiban were stuck, unable to go to the next world, because they had too many future regrets."
"How can you have a _future_ regret? That doesn't make any sense."
"No? You think to yourself, If I do this, then this will happen, then I will feel this way, so I shouldn't do it. You're stuck. Like Lao Lu. He was sorry he made Pastor believe he had killed Cape and the soldiers. To teach himself a lesson, he decided he would become Pastor's wife in the next lifetime. But whenever he thought about his future—that he would have to listen to Amen this and Amen that, every Sunday—he would start cursing again. How can he become a pastor's wife when his foul temper is still foul? That's why he's stuck."
"And Yiban?"
"When he couldn't find Miss Banner, he thought she had died. He grew sad. Then he wondered if she had gone back to Cape. He grew even sadder. When Yiban died, he flew to heaven to find Miss Banner, and because she was not there, he believed she was with Cape in hell."
"He never considered that she'd gone to the World of Yin?"
"See! That's what happens when you become stuck. Do good things enter your mind? Mm-mm. Bad things? Plenty."
"So he's still stuck?"
"Oh no-no-no-no-no! I told him about you."
"Told him what?"
"Where you were. When you would be born. And now he's waiting for you again. Somewhere here."
"Simon?"
Kwan flashes a huge smile and gestures toward a large rock. Behind it, barely visible, is a narrow opening.
"This is the cave with the lake?"
"The same."
I poke my head in and yell: "Simon! Simon! Are you in there? Are you all right?"
Kwan grabs my shoulders and gently pulls me back. "I go in, get him," she says in English. "Where flashlight?"
I fish it out of the day pack and click it on. "Shit, it must have been left on all night. The battery's dead."
"Let me see." She takes the flashlight and it immediately brightens. "See? Not dead. Okay!" She presses herself into the cave and I follow.
"No-no, Libby-ah! You stay outside."
"Why?"
"In case . . ."
"In case what?"
"Just in case! Don't argue." She clutches my hand so hard it hurts. "Promise, ah?"
"All right. Promise."
She smiles. The next moment, her face bunches up into an expression of pain, and tears spill down her round cheeks.
"Kwan? What is it?"
She squeezes my hand again and blubbers in English, "Oh, Libby-ah, I so happy can finally pay you back. Now you know all my secret. Give me peace." She throws her arms around me.
I'm flustered. I've always felt awkward with Kwan's emotional outpourings. "Pay me back—for what? Come on, Kwan, you don't owe me anything."
"Yes-yes! You my loyal friend." She's sniffling. "For me, you go Yin World, because I tell you, Sure-sure, Yiban follow you there. But no, he go heaven, you not there. . . . So you see, because me, you lost each other. That's why I so happy first time I meet Simon. Then I know, ah, at last!—"
I back away. My head is buzzing. "Kwan, that night you met Simon, do you remember talking to his friend Elza?"
She wipes her eyes with her sleeve. "Elza? . . . Ah! Yes-yes! Elsie. I remember. Nice girl. Polish-Jewish. Drown after lunch."
"What she said, that Simon should forget about her—did you make that up? Didn't she say something else?"
Kwan frowns. "Forget her? She say that?"
"You said she did."
"Ah! I remember now. Not 'forget.' _Forgive._ She want him forgive her. She done something make him feel guilty. He think his fault she die. She say no, her fault, no problem, don't worry. Something like that."
"But didn't she tell him to wait for her? That she was coming back?"
"Why you think this?"
"Because I saw her! I saw her with those secret senses you always talk about. She was begging Simon to see her, to know what she was feeling. I saw—"
"Tst! Tst!" Kwan puts her hand on my shoulder. "Libby-ah, Libby-ah! This not secret sense. This you own sense doubt. Sense worry. This nonsense! You see you own ghost self begging Simon, Please hear me, see me, love me. . . . Elsie not saying that. Two lifetime ago, you her daughter. Why she want you have misery life? No! She _help_ you. . . ."
I listen, stunned. Elza was my mother? Whether that was true or not, I feel lighthearted, giddy, a needless load of resentment removed, and with it a garbage pile of fears and doubts.
"All this time you think she chasing you? Mm-mm. You chase youself! Simon know this too." She kisses my cheek. "I go find him now, let him tell you hisself." I watch her squeeze into the cave.
"Kwan?"
She turns around. "Ah!"
"Promise you won't get lost. You'll come back."
"Yes, promise-promise! Course." She is lowering herself into the next chamber. "Don't worry." Her voice returns, deep and resonant. "I find Simon, be back soon. You wait for us. . . ." She trails off.
I wrap the space blanket around my shoulders and sit, leaning against the boulder that hid the entrance to the cave. Hope, nothing wrong with that. I scan the sky. Still gray. Is it going to rain again? And with that single unhappy possibility, bleakness and common sense take over. Did I become hypnotized while listening to Kwan's story? Am I as delusional as she is? How could I let my sister go into the cave by herself? I scramble to my feet and stick my head into the entrance. "Kwan! Come back!" I crawl into the mouth of the dark chamber. "Kwan! Kwan! Goddamnit, Kwan, answer me!" I venture forward, bump my head against the low ceiling, curse, then holler again. A few steps later, the light tapers, and at the next turn disappears. It's as if a thick blanket had been thrown over my eyes. I don't panic. I've worked in darkrooms half my life. But in here, I don't know the boundaries of the dark. The blackness is like a magnet drawing me in. I backtrack toward the cave opening, but I'm disoriented, with no sense of direction, not in or out, nor up or down. I shout for Kwan. My voice is growing hoarse, and I'm gasping for breath. Has all the air been sucked out of the cave?
"Olivia?"
I choke a yelp.
"Are you okay?"
"Oh God! Simon! Is that really you?. . ." I start to sob. "Are you alive?"
"Would I be talking to you if I weren't?"
I laugh and cry at the same time. "You never know."
"Here, reach out your hand."
I swat at the air until I touch flesh, his familiar hands. He pulls me toward him and I wrap my arms around his neck, leaning against his chest, rubbing his spine, assuring myself he is real. "God, Simon, what happened yesterday—I was insane. And later, when you didn't come back—did Kwan tell you what I've been through?"
"No, I haven't been back to the house yet."
I stiffen. "Oh God!"
"What's wrong?"
"Where's Kwan? Isn't she behind you?"
"I don't know where she is."
"But . . . she went in to find you. She went into the cave! And I've been calling for her! Oh God! This can't be happening. She promised she wouldn't get lost. She promised to come back. . . ." I keep babbling as Simon leads me out.
We stumble into the open, where the light is so hard I can't see. I blindly pat Simon's face, half expecting that when I can see the world again, he'll be Yiban and I'll be wearing a yellow dress stained with blood.
# IV
## [23
THE FUNERAL](TheHundredSecretSenses-toc.html#TOC-27)
Kwan disappeared two months ago. I don't say "died," because I haven't yet allowed myself to think that's what happened.
I sit in my kitchen, eating granola, staring at the pictures of missing kids on the back of the milk carton. "Reward for any information," it reads. I know what the mothers of those children feel. Until proven otherwise, you have to believe they're somewhere. You have to see them once more before it's time to say good-bye. You can't let those you love leave you behind in this world without making them promise they'll wait. And I have to believe it's not too late to tell Kwan, I was Miss Banner and you were Nunumu, and forever you'll be loyal and so will I.
Two months ago, the last time I saw her, I waited by the cave, certain that if I believed her story, she'd come back. I sat on the music box, with Simon beside me. He tried to appear optimistic, yet never cracked a joke, which was how I knew he was worried. "She'll turn up," he assured me. "I just wish you didn't have to go through this agony, first with me, now with Kwan."
As it turned out, he had been safe all along. After our fight, he too had left the archway. He was trudging back to Big Ma's house when he ran into the cow herder who had called us assholes. Only the guy was not a cow herder but a graduate student from Boston named Andy, the American nephew of a woman living in a village farther down the mountain. The two of them went to his aunt's house, where they chugged shots of _maotai_ until their tongues and brains went numb. Even if he hadn't passed out, Simon would have been all right, and it pained him to admit this had been so. In his fanny pack, he had a wool cap, which he put on after I ran away. And then he had raged, pitching rocks into the ravine until he worked up a sweat.
"You worried for nothing," he said, devastated.
And I said, "Better than finding out I worried for _something._ " I reasoned that if I was grateful that Simon had never been in any real danger, similar luck would be granted me any minute with regard to Kwan. "Sorry-sorry, Libby-ah," I imagined she'd say. "I took wrong turn in cave, lost. Took me whole morning come back! You worry for nothing." And later, I made adjustments in my hopes to account for the passing of time. "Libby-ah, where my head go? I see lake, can't stop dreaming. I think only one hour pass by. Ha!—don't know was ten!"
Simon and I stayed by the cave all night. Du Lili brought us food, blankets, and a tarp. We pushed away the boulder blocking the entrance of the cave, then climbed in and huddled in the shallow cove. I stared at the sky, a sieve punctured with stars, and considered telling him Kwan's story of Miss Banner, Nunumu, and Yiban. But then I became afraid. I saw the story as a talisman of hope. And if Simon or anyone else discounted even part of it, then some possibility out in the universe, the one I needed, might be removed.
On the second morning of Kwan's disappearance, Du Lili and Andy organized a search party. The older folks were too scared to go into the cave. So those who showed up were mostly young. They brought oil lamps and bundles of ropes. I tried to recall the directions to the cavern with the lake. What had Zeng said? Follow the water, stay low, choose the shallow route over the wide. Or was it narrow instead of deep? I didn't have to ask Simon not to go into the cave. He stayed close to me, and together we watched grimly as one man tied the rope around his waist and burrowed into the cave, while another man stood outside, holding the other end of the rope taut.
By the third day, the searchers had navigated a labyrinth that led them to dozens of other caves. But no trace yet was seen of Kwan. Du Lili went to Guilin to notify the authorities. She also sent a telegram that I had carefully written to George. In the afternoon, four vans arrived bearing green-uniformed soldiers and black-suited officials. The following morning, a familiar sedan rolled up the road, and out stepped Rocky and a somber old scholar. Rocky confided to me that Professor Po was the right-hand man to the paleontologist who discovered the Peking Man. The professor entered the intricate maze of caves, now explored more easily with the addition of guide ropes and lamps. When he emerged many hours later, he announced that many dynasties ago, the inhabitants of this area had excavated through dozens of caves, creating intentional dead ends, as well as elaborate interconnecting tunnels. It was likely, he theorized, the people of Changmian invented this maze to escape from Mongolians and other warring tribes. Those invaders who entered the labyrinth became lost, and scurried about like rats in a death trap.
A team of geologists was brought in. In the excitement that ensued, nearly everyone forgot about Kwan. Instead, the geologists found jars for grain and jugs for water. They barged into bat lairs and sent thousands of the frightened creatures shrieking into the blinding sun. They made an important scientific discovery of human shit that was at least three thousand years old!
On the fifth day, George and Virgie arrived from San Francisco. They had received my various telegrams with their progressively dire messages. George was confident that Kwan wasn't actually missing; it was only my poor Mandarin that had caused us to become temporarily separated. By evening, however, George was a wretched mess. He picked up a sweater that belonged to Kwan, buried his face in it, not caring who watched him cry.
On the seventh day, the search teams located the shining lake and the ancient village beside its shore. Still no Kwan. But now the village was crawling with officials of every rank, as well as a dozen more scientific teams, all of them trying to determine what caused the cave water to glow.
On each of those seven days, I had to give a report to yet another bureaucrat, detailing what had happened to Kwan. What was her birth-date? When did she become an overseas Chinese? Why did she come back here? Was she sick? You had a fight? Not with her but with your husband? Was your husband angry with her too? Is that why she ran away? Do you have a photo of her? You took this picture? What kind of camera do you use? You are a professional photographer? Really? How much money can a person make taking a picture like this? Is that so? That much? Can you take a picture of me?
At night, Simon and I hugged each other tightly in the marriage bed. We made love, but not out of lust. When we were together this way, we could hope, we could believe that love wouldn't allow us to be separated again. With each passing day, I didn't lose hope. I fought to have more. I recalled Kwan's stories. I remembered those times when she bandaged my wounds, taught me to ride a bike, placed her hands on my feverish six-year-old forehead and whispered, "Sleep, Libby-ah, sleep." And I did.
Meanwhile, Changmian had become a circus. The entrepreneurial guy who had tried to sell Simon and me so-called ancient coins was charging curiosity-seekers ten yuan for admission through the first archway. His brother charged twenty to go through the second. Many of the tourists trampled the ravine, and Changmian residents hawked rocks from the graves as souvenirs. An argument broke out between the village leaders and officials over who owned the caves and who could take what they contained. At that point, two weeks had gone by, and Simon and I couldn't stomach any more. We decided to take the plane home on the scheduled day.
BEFORE WE LEFT, Big Ma finally had her funeral. There were only eleven people in attendance that drizzly morning—two hired hands to transport the casket to the grave, a few of the older villagers, and George, Virgie, Du Lili, Simon, and I. I wondered if Big Ma was miffed at being upstaged by Kwan. The hired hands loaded the casket onto the back of a mule cart. Du Lili tied the requisite screaming rooster to the coffin lid. When we reached the bridge over the first irrigation pond, we found a television news crew blocking our way.
"Move your butts!" Du Lili shouted. "Can't you see? We have a funeral procession going through!" The crew came over and asked her to respect the rights of citizens to hear about the wonderful discovery in Changmian.
"Wonderful bullshit!" said Du Lili. "You're ruining our village. Now get out of our way." A stylish woman in snazzy jeans took Du Lili aside. I saw her offer money, which Du Lili angrily refused. My heart soared with admiration. The woman flashed more money. Du Lili pointed to the crew, then the coffin, complaining loudly again. A bigger wad of bills came out. And Du Lili shrugged. "All right," I heard her say as she pocketed the money. "At least the deceased can use this to buy a better life in the next world." My spirits went into a tailspin. Simon looked grim. We took a long detour, squeezing through alleyways until we reached the communal graveyard, a slope leading up the mountains facing west.
At the grave site, Du Lili cried while caressing Big Ma's parched face. I thought the body was amazingly well preserved after a two-week hiatus between departure and sendoff. "Ai, Li Bin-bin," Du Lili crooned, "you're too young to die. I should have gone before you." I translated this to Simon.
He glanced at Du Lili. "Is she saying she's older than Big Ma?"
"I don't know. I don't want to know what anything means anymore."
As the hired hands closed the coffin lid, I felt that the answers to so many questions were being sealed up forever: where Kwan was, what my father's real name had been, whether Kwan and a girl named Buncake had indeed drowned.
"Wait!" I heard Du Lili cry to the workers. "I almost forgot." She reached into her pocket and pulled out the wad of banknotes. As she wrapped Big Ma's stiff hand around the television crew's bribery money, I cried, my faith restored. And then Du Lili reached into the front of her padded jacket and extracted something else. It was a preserved duck egg. She put this in Big Ma's other hand. "Your favorite," she said. "In case you get hungry on your way there."
_Duck eggs!_ "I made so many," I could hear Kwan saying. "Maybe some still left."
I turned to Simon. "I have to go." I clutched my stomach and grimaced, feigning illness.
"You want me to help you?"
I shook my head and went over to Du Lili. "Bad stomach," I said. She gave me an understanding look. As soon as I was sure I was out of their view, I started running. I didn't give a damn about keeping my expectations in check. I was surrendering myself completely to hope. I was elated. I knew that what I believed was what I'd find.
I stopped at Big Ma's house and snatched a rusted hoe. And then I sped over to the community hall. When I reached the gate, I walked through slowly, searching for familiar signs. There!—the bottom bricks of the foundation—they were pockmarked black, and I was sure these were the burnt ruins of the Ghost Merchant's House. I raced through the empty building, glad that everyone was at the ravine gawking at three-thousand-year-old shit. In the back, I saw no garden, no rolling paths or pavilion. Everything had been leveled for the exercise yard. But just as I expected, the stones of the boundary walls were also blackened and blistered. I went to the northwest corner and calculated: Ten jars across, ten paces long. I began to hack at the mud with the hoe. I laughed out loud. If anyone saw me, they'd think I was as crazy as Kwan.
I unearthed a trench of mud five feet long and two feet deep, nearly big enough for a corpse. And then I felt the hoe hit something that was neither rock nor soil. I fell to my knees and frantically scooped out the dark moist dirt with my bare hands. And then I saw it, the lighter clay, firm and smooth as a shoulder. In my impatience, I used the handle of the hoe to break open the jar.
I pulled out a blackened egg, then another and another. I hugged them against my chest, where they crumbled, all these relics of our past disintegrating into gray chalk. But I was beyond worry. I knew I had already tasted what was left.
## [24
ENDLESS SONGS](TheHundredSecretSenses-toc.html#TOC-28)
George and Virgie have just come back from their honeymoon in Changmian. They claim I wouldn't recognize the place. "Tourist trap everywhere!" George says. "The whole village is now rich, selling plastic sea creatures, the glow-in-the-dark kind. That's why the lake was so bright. Ancient fish and plants living deep in the water. But now no more. Too many people made a wish, threw lucky coins in the lake. And all those sea creatures? Poisoned, went belly-up dead. So the village leaders installed underwater lights, green and yellow, very pretty, I saw them myself. Good show."
I think George and Virgie chose to go to Changmian as an apology to Kwan. To get married, George had to have Kwan declared legally dead. I still have mixed feelings about that. The marriage, I figure, is what Kwan had intended all along. At some level, she must have known she wasn't coming home. She never would have let George go through life without having enough to eat. I think she also would have laughed and said, "Too bad Virgie not better cook."
I've had almost two years to think about Kwan, why she came into my life, why she left. What she said about fate waiting to happen, what she might have meant. Two years is enough time, I know, to layer memories of what was with what might have been. And that's fine, because I now believe truth lies not in logic but in hope, both past and future. I believe hope can surprise you. It can survive the odds against it, all sorts of contradictions, and certainly any skeptic's rationale of relying on proof through fact.
How else can I explain why I have a fourteen-month-old baby girl? Like everyone else, I was astounded when I went to the doctor and he said I was three months along. I gave birth nine months after Simon and I made love in the marriage bed, nine months after Kwan disappeared. I'm sure there were those who suspected the father was some fly-by-night date, that I was careless, pregnant by accident. But Simon and I both know: That baby is ours. Sure, there was a reasonable explanation. We went back to the fertility specialist and he did more tests. Well, what do you know? The earlier tests were wrong. The lab must have made a mistake, switched the charts, because sterility, the doctor said, is not a reversible condition. Simon, he announced, had in fact not been sterile. I asked the doctor, "So how do you explain why I never became pregnant before?"
"You were probably trying too hard," he said. "Look how many women become pregnant after they adopt."
All I know is what I want to believe. I have a gift from Kwan, a baby girl with dimples in her fat cheeks. And no, I didn't name her Kwan or Nelly. I'm not that morbidly sentimental. I call her Samantha, sometimes Sammy. Samantha Li. She and I took Kwan's last name. Why not? What's a family name if not a claim to being connected in the future to someone from the past?
Sammy calls me "Mama." Her favorite toy is "ba," the music box Kwan gave me for my wedding. Sammy's other word is "Da," which is what she calls Simon, "Da" for "Daddy," even though he doesn't live with us all the time. We're still working things out, deciding what's important, what matters, how to be together for more than eight hours at a stretch without disagreeing about which radio station we should put on. On Fridays, he comes over and stays the weekend. We snuggle in bed, Simon and I, Sammy and Bubba. We are practicing being a family, and we're grateful for every moment together. The petty arguments, snipes and gripes, they still crop up. But it's easier to remember how unimportant they are, how they shrink the heart and make life small.
I think Kwan intended to show me the world is not a place but the vastness of the soul. And the soul is nothing more than love, limitless, endless, all that moves us toward knowing what is true. I once thought love was supposed to be nothing but bliss. I now know it is also worry and grief, hope and trust. And believing in ghosts—that's believing that love never dies. If people we love die, then they are lost only to our ordinary senses. If we remember, we can find them anytime with our hundred secret senses. "This a secret," I can still hear Kwan whispering. "Don't tell anyone. Promise, Libby-ah."
I hear my baby calling me. She gurgles and thrusts her hand toward the fireplace, at what I don't know. She insists. "What is it, Sammy? What do you see?" My heart races as I sense it might be Kwan.
"Ba," Sammy coos, her hand still reaching up. Now I see what she wants. I go to the mantel and take down the music box. I wind the key. I lift my baby into my arms. And we dance, joy spilling from sorrow.
**Looking for more?**
Visit Penguin.com for more about this author and a complete list of their books.
**Discover your next great read!**
|
{
"redpajama_set_name": "RedPajamaBook"
}
| 4,505
|
Multi-Platinum Alt Rock Geniuses, the Goo Goo Dolls, to Hit the Island Showroom
Home » Blog » Entertainment » Multi-Platinum Alt Rock Geniuses, the Goo Goo Dolls, to Hit the Island Showroom
The Goo Goo Dolls—John Rzeznik and Robby Takac—have spent the last 30 years topping the charts and garnering love from fans around the globe. John and Robby are set to take the stage at the Island Resort & Casino for two live shows on Friday, September 14th, and Saturday, September 15th. The Goo Goo Dolls are known for their 10 complete studio albums, as well as a staggering 19 top-10 singles on various charts.
Chart-Topping Hits by the Goo Goo Dolls
Chances are, you've heard of the Goo Goo Doll's top hits: "Iris," "Name," and "Slide." These songs and more have helped the band sell well over 10 million albums worldwide. Whether you've been a fan for the past 30 years or you're just discovering the greatness that is the Goo Goo Dolls, you will enjoy these samplings from their ever-evolving set list:
What Can You Expect From a Goo Goo Dolls Show?
The Goo Goo Dolls illicit emotion from the audience using popular, sing-along worthy jams and late 80s early 90s nostalgia. Those who have found themselves in that audience before have this to say about the show:
"This was my first time seeing The Goo Goo Dolls live and I came away a fan with them as well. I give my highest recommendations to go check out this very cool band and hear their incredible new music."
"I can't wait to see them again. And if you listen to the lyrics, they are timeless. That is what keeps them around in our crazy world."
– Lydia
About The Goo Goo Dolls
The band's whirlwind of a career began in Buffalo, New York, in 1985, when guitarist and vocalist John Rzeznik and bassist and vocalist Robby Takac joined musical forces. They started as a cover band but quickly began penning their own songs. Their debut album released in 1987; the second followed in 1989 and third in 1990.
Their album, "Superstar Car Wash," released in 1993 is credited to be their breakthrough as an artist. In 1995, A Boy Named Goo was released, where the hit song "Name" originated. The album quickly sold one million copies in the very first year. In 1998, Rzeznik wrote "Iris" for the movie City of Angels and that song alone spent nearly a year on the Billboard charts, with 18 weeks at number one. It also received 3 Grammy nominations.
The Goo Goo Dolls went on to release several more albums, touring with the likes of Bon Jovi, Lifehouse, Switchfoot and more, and performed at some of the world's best venues and events. It is still onward and upward for the Dolls, with a first-ever EP release last year titled, "You Should Be Happy."
Visit the Goo Goo Dolls website to learn more.
Location: Island Showroom
Find seats online or call 877-475-7469 to order tickets by phone.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 2,922
|
Monarchia (, ze , monarchía "jedynowładztwo", od μόναρχος "jedyny władca", od μόνος monos "jeden" i ἀρχός archós "początek") – ustrój polityczny lub forma rządów, gdzie głową państwa jest jedna osoba, nazywana monarchą, niewybierana w sposób republikański. Zazwyczaj:
monarcha sprawuje władzę dożywotnio (z wyjątkiem Andory, Malezji, Samoa oraz Zjednoczonych Emiratów Arabskich);
jego funkcja jest często dziedziczna (z wyjątkiem monarchii elekcyjnych);
jego stanowisko jest nieusuwalne.
Obecnie w wielu państwach z takowym ustrojem politycznym monarcha pełni jedynie funkcje reprezentacyjne. Kierowanie przez niego polityką jest znikome lub żadne.
Historia ustroju
W pierwszych wspólnotach plemiennych nie istniała hierarchia, która (wśród ludów europejskich) pojawiła się pod koniec demokracji wojennej, a na Bliskim Wschodzie przed powstaniem pierwszych cywilizacji.
W początkach starożytności monarcha zyskał praktycznie nieograniczoną władzę (w ustroju despotycznym), w niektórych krajach był uznawany za boga lub półboga (faraon, august), niekiedy miał pozostawać w łączności z bóstwami.
W średniowieczu powstały nowe, feudalne formy monarchii: patrymonialna oraz stanowa. Uzasadnieniem władzy monarchów było jej pochodzenie od Boga, stąd ich pozycja była teoretycznie niepodważalna. Byli suzerenami, jednak nie zawsze mogli wyegzekwować swą władzę i ich wpływ został ograniczony. Powstawały pierwsze parlamenty (tym samym monarchie parlamentarne).
Od XV wieku w Europie zachodniej ponownie wzrosło znaczenie monarchy, do władzy prawie nieograniczonej lub nieograniczonej, czyli absolutnej. Monarchia absolutna Ludwika XIV była wzorem dla innych monarchów Europy, szczególnie Prus, Rosji i Szwecji.
Rozwój przemysłu i handlu, wzrost pozycji burżuazji i klasy średniej, rewolucja francuska (1789-1799) doprowadziły do upadku absolutyzmu. Narodził się pogląd, że suwerenem jest społeczeństwo, które ma prawo powoływać i odwoływać sprawujących władzę.
Ta sytuacja, wsparta przez powstanie Stanów Zjednoczonych (4 lipca 1776), wymusiła ograniczenia praw monarchy. Doprowadziło to do powstania pierwszych monarchii konstytucyjnych (Francja, Rzeczpospolita Obojga Narodów), gdzie król był zwierzchnikiem administracji.
Nieco inna sytuacja miała miejsce w Anglii, a następnie w Wielkiej Brytanii. Ustrój panujący w tym kraju określa się mianem monarchii parlamentarnej.
Upadek wielu monarchii nastąpił po I wojnie światowej. Nowo powstałe państwa (Czechosłowacja, Polska itd.) wybrały ustrój republikański, w wielu krajach obalano monarchie, zastępując je republikami (Niemcy, Austro-Węgry, Turcja).
Nazewnictwo
Od kręgu kulturowego oraz religii zależy nazwa monarchii:
chrześcijańskie monarchie były m.in. cesarstwami, królestwami, księstwami (nazwy tworzono od tytułu monarsze, w tym od tytułów szlacheckich) czy carstwami;
islamskie monarchie są sułtanatami lub zapożyczają nazwy europejskie, np. Królestwo Maroka;
w Azji funkcjonowały także nazwy chanatu i kaganatu;
na Dalekim Wschodzie najważniejszymi monarchiami były cesarstwa w Indiach, Chinach i istniejące do dziś Cesarstwo Japonii.
Sukcesja tronu
Zazwyczaj monarchie są dziedziczne, tzn. po śmierci władcy tron obejmuje jedno z jego dzieci. W Europie początkowo był to najstarszy syn monarchy, później dopuszczano również córki (por. Jadwiga Andegaweńska, sankcja pragmatyczna). Czasem jednak zdarzało się, że monarcha umierał bezdzietnie, wówczas władzę mogła objąć osoba spokrewniona, niebędąca zstępnym. Czasem następcę monarcha mógł wyznaczyć na podstawie czynności prawnej: testamentem lub zawierając układ o przeżycie.
Zdarzało się, że sprawa sukcesji była niejasna, gdy wiele osób jednocześnie dowodziło swoich praw do tronu. Z tego powodu wybuchały wojny, np. wojna o sukcesję austriacką, wojna o sukcesję hiszpańską oraz wojna o sukcesję polską.
Niektóre państwa były monarchiami elekcyjnymi, tzn. takimi, gdzie monarchą zostawała osoba wybrana przez elektorów. Przykładem historycznych monarchii elekcyjnych mogą być takie państwa, jak Państwo Kościelne, Rzeczpospolita Obojga Narodów, Święte Cesarstwo Rzymskie Narodu Niemieckiego, a obecnie Watykan.
Obecne monarchie (stan na rok 2022)
Byłe monarchie
Poniżej podano wykaz monarchii, które istniały w okresie nowożytnym, ale już nimi nie są.
Oprócz tego władca brytyjski, który panuje obecnie łącznie w 15 monarchiach w przeszłości pełnił funkcję monarchy w następujących państwach, które obecnie mają ustrój republikański: Wolne Państwo Irlandzkie (1931–1949), Związek Południowej Afryki (1931–1961), Indie (1947–1950), Pakistan (1947–1956), Cejlon (1948–1972), Ghana (1957–1960), Nigeria (1960–1963), Sierra Leone (1961–1971), Tanganika (1961–1962), Trynidad i Tobago (1962–1976), Uganda (1962–1963), Kenia (1963–1964), Malawi (1964–1966), Malta (1964–1974), Gambia (1965–1970), Gujana (1966–1970), Barbados (1966–2021), Mauritius (1968–1992), Fidżi (1970–1987).
Żyjący byli monarchowie
Zobacz też
diarchia
republika
Uwagi
Przypisy
Bibliografia
Dynastie Europy, wyd. Ossolineum, Wrocław Warszawa Kraków 2003.
Morby John, Dynastie świata. Przewodnik chronologiczny i genealogiczny, Kraków 1995.
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 5,406
|
The Beginnings of Humankind
Free «The Beginnings of Humankind» Essay
Apes and human beings are two different creatures which form a part of the ecosystem. Apes are the primates that belong to the family pongidae and they are characterized by long arms, broad chest and an absence of the tail. They include the chimpanzee, gorilla, gibbon and orangutan. While human being is a man of the species of Homo sapiens, it is distinguished from other animals by its large brain capacity, intelligence and the power to articulate speech. Even though these two creatures are genetically different, they have the common features that can be compared; the dentition, jaws, skull and so on as observed by Johanson and Edey in their book; Lucy: the beginnings of human kind.
In terms of their dentition, the scholars noted that apes have larger developed jaws with the lower jaw being fairly deep and the two halves fused together in the front. Their dental formula resembles that of the old world monkeys; I2/2; C1/1; P2/2; M3/3 = 32. Their teeth is composed of the shovel shaped incisors and the prominent canines; with the rear edge of the upper canines honing against the edge of the first lower premolars. The molars are composed of 5 cusps with a complex arrangement of ridges on the grinding surfaces. On the other hand, human beings' teeth are arranged in upper arch (maxillary) and lower arch (mandible); forming a shape which is known as the Centenary Arch. Maxilla is firmly attached to the skull while mandible just forms a joint with the skull.
The dentition of a human being is also composed of the primary and permanent teeth. However, the primary teeth are replaced by all permanent teeth at the age of 12-13, whereby the permanent dentition is composed of thirty two teeth. These teeth are also classified as incisors, canines, premolars and molars; each with its specific adaptation for a specified function. They also have a special consistent arrangement that distinguishes them to the apes. The human beings' teeth are arranged in a manner in which each category lies directly opposite to the ones in the other jaw. For instance, the canines of the upper jaw rest on the ones of the lower jaws and so on, when the mouth is shut.
Concerning the skull, the ones for the apes have flattened faces with complete and forward pointing eye sockets which are very bony and with a plate of bone separating the orbit from the jaw muscle. Their skull also has no chin and the nose bone is flat, while the foramen magnum has crests and is ventrally placed to underneath the skull. On the other hand, the human skull is very different in that it has a smooth-domed shape and is without many crests, as is the case with the apes. The skull is also wider at the top and narrower at the base with a larger brain cavity than that of the apes. It is also made up of two pairs of bones; the cranium and facial bones. Finally, it is noted that the human cerebrospinal system is highly developed for manifestation of self-conscious intlligence.
Hadar and Laetoli are one of the best known archeological sites where a lot of fossils claimed to be for the ancestors of man have been discovered over the years. Hadar is located in 300 kilometers north east of Addis Ababa, Ethiopia and it derived its name from the Kada Hadar tributary; which is a tributary to river Awash. Here, many discoveries about the Australopithecus have been made, including that of the crude stone tools which suggest that the hominid might have lived during the Stone Age. On the other hand, Laetoli is located in Tanzania, about 50 kilometers south of Olduvai Gorge.
Want an expert write a paper for you?
Talk to an operator now!
According to the analysis carried out by Donald Johanson, Tim White and their colleagues on the hominids' fossils found in Hadar and Laetoli, these fossils resemble more human beings than the apes. For instance, the fossils found in Hadar by Donald Johanson which were that of the Australopithecus Afarensis (Lucky), had many similarities to the human being anatomy. These fossils were discovered in 1974 and are claimed to have existed between 3.9 and 3 million years ago. The dentition of Australopithecus is said to be more manlike than apelike. It had got large teeth with the back ones being larger than the front ones; just like in case of human beings. Its canines were also larger and pointed just like that of the humans. This is said to have been an adaptation to tearing and eating of flesh. Its molars were also in an ascending size and with serrated roots, just like that of human beings'. Therefore, the shape and arrangement of the entire teeth were closely in resemblance to that of human beings than the apes.
When it comes to the jaws, this analysis indicated that this hominid had slightly protruding jaws which almost resembled the ones for the human beings. While the apes had very larger and much protruded jaws; which were also stronger as compared to that of the modern man and Australopithecus. The jaws were also shaped in a way that they were in between the rectangular shape of the apes and the parabolic shape of the modern man's; parallel post canine tooth rows. The analysis of the skull of this hominid by the team also revealed a closer resemblance to human beings than the apes. It is said to have been big with a higher capacity (375 to 550 cc.) to that of apes. The skull was also big, but with a round shape, just like that of the human being. Finally, the attachment of the skull to the jaws is almost similar to that of man than to the apes.
Having done this analysis, a conclusion was made that the Laetoli hominid was also an ancestor to the modern man. That is because this hominid resembles the Australopithecus. The Laetoli fossils included the canine tooth, which was found in by Louis Leakey in 1935 and the upper jaw borne with a couple of premolars in it by Kohl-Larsen in 1938/39. They also include the lately discovered artifacts, such as the forty two teeth with slight attachment to jjaw bones and the mandible with nine teeth in 1974 by Mary Leakey and a Kenyan, Karnoya Kimeu. However, the most unique discovery in Laetoli was that of hominid footprints, which were discovered by Mary Leakey and her colleagues.
The team claimed that these footprints had been formed and preserved under a combination of events; volcanic eruption, rainstorm and another ash fall. All these artifacts are claimed to have existed between 3.76-3.56 million years ago. Even though there were so many footprints which could be of some other animals, such as the elephants, baboons, and birds; these ones were very unique, as they resembled the man's feet. These footprints were later studied by Tim White in the same year and his conclusion was that they were that of an erect ape and that they closely resembled the feet of the Lucky, which had been discovered in Hadar by Johanson and his colleges.
Based on the similarities, Johanson and Edey noted that the hominids could be of the same species. This was backed up by the fact that the earlier found fossils also resembled the ones which were discovered in Hadar. They both demonstrated the close resemblance to human beings. For instance, the feet have an arch-bending of the sole of the foot, just like in case of the human beings unlike the apes which did not have it. Its toe was equally not as big as that of the apes, which had a mobile big toe. It also had a lateral transmission of force from the heel then to the base of the lateral metatarsal like the human beings. These Laetoli fossils have also demonstrated resemblance to those from the other Australopithecus artifact sites, such as Turkwel and Maka, which existed 3.4 and 3.5 million years ago respectively.
Having carried out a thorough analysis of the fossils from Hadar and Laetoli, Donald Johanson, Tim White and their colleagues, one compared them to the features of the apes and the human beings. Their interpretation was that Australopithecus is a direct ancestor of man, and that the apes have not any genetically relationship to human beings. This is because the body features of Australopithecus and that of human beings closely resemble each other. While the body features of the apes neither resemble human beings or this hominid. These facts were very influencing towards their interpretation.
It can, thus, be concluded from the book Lucy: the beginnings of human kind that, Australopithecus afarensis is the direct ancestor to human beings, since it had a closely similar body features to that of human beings. However, the apes, which are normally thought to be the part of the human ancestors, are not. In addition, it is also true that the human ancestors must have lived in Africa and that they were free to walk and change residence. This is because the Australopithecus' fossils were discovered in different geographical regions within the African continent.
Related Informative essays
The Power of Images
Intersexuality
The Leadership of Bill Belichick
Safety and Security of Airport Ground
Heroin Chic and Contemporary Fashion
Fire Frequency and Severity
Problems with Overcrowded Prisons
Home Take Exam
The Call of the Wild by Jack London
Radio Frequency Identification in Supply Chain
Anthropology: Argonauts of the Western Pacific
Portfolio Project: Business Research
THOMAS JEFFERSON AND ALEXANDER HAMILTON
Fraud and Abuse in Managed Care
IBM Company and Marketing Mix
The Roles of Grapes and Grape Products in Health
Female Genital Mutilation in Egypt
Star Academy
PrimeWriting.net Testimonials
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,382
|
Who We Are Who We Serve Contact
Press Release Distribution Online Newsroom Platform id.
FAQ Sample Press Release Press Release Template Blog
Customer Reviews Login Contact Us
ADNOC Awards Cementing Services To NESR
Thursday, March 17, 2022 11:30 AM
National Energy Services Reunited Corp
Visit Newsroom
Company Update
ABU DHABI, UNITED ARAB EMIRATES / ACCESSWIRE / March 17, 2022 / National Energy Services Reunited Corp. ("NESR" or "the Company") (NASDAQ:NESR) (NASDAQ:NESRW), an international, industry-leading provider of integrated energy services in the Middle East and North Africa ("MENA") region, has announced that the Company has been awarded one of the major contracts for Cementing Services ("Cementing") in Abu Dhabi from ADNOC. In total, the announced award of framework agreements to five oilfield service providers is valued at $658 million USD, a strong portion of which includes NESR's scope for Cementing services up to seven years. The awards build on ADNOC's recent drilling-related investments to support its crude oil production capacity expansion, and through ADNOC's flagship in-country value ("ICV") program, has the potential to see over 65% of the total investment directly bolster the UAE economy.
NESR CEO & Chairman Sherif Foda commented, "We are proud to have again been selected among a short list of service suppliers, to support ADNOC's clear and unmistakable ambition for sustainable oil production capacity growth. Following on our recent announced award for Testing services with ADNOC onshore, this pivotal Cementing contract underscores our alignment with ADNOC's ICV vision and affords us the opportunity to deploy next-generation sustainable technologies across multiple service segments."
Mr. Foda continued, "I want to sincerely thank ADNOC leadership for their trust in us as a marquee service platform and partner. As the National Champion of the MENA region, NESR will continue to invest locally in its people, equipment and supply chain as its scope and share of services expand. ADNOC is a foremost leader in ICV empowerment, for which NESR's growth strategy is uniquely aligned."
About National Energy Services Reunited Corp.
Founded in 2017, NESR is one of the largest national oilfield services providers in the MENA and Asia Pacific regions. With over 5,000 employees, representing more than 60 nationalities in over 15 countries, the Company helps its customers unlock the full potential of their reservoirs by providing Production Services such as Hydraulic Fracturing, Cementing, Coiled Tubing, Filtration, Completions, Stimulation, Pumping and Nitrogen Services. The Company also helps its customers to access their reservoirs in a smarter and faster manner by providing Drilling and Evaluation Services such as Drilling Downhole Tools, Directional Drilling, Fishing Tools, Testing Services, Wireline, Slickline, Drilling Fluids and Rig Services.
This communication contains forward-looking statements (as such term is defined in Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended). Any and all statements contained in this communication that are not statements of historical fact may be deemed forward-looking statements. Terms such as "may," "might," "would," "should," "could," "project," "estimate," "predict," "potential," "strategy," "anticipate," "attempt," "develop," "plan," "help," "believe," "continue," "intend," "expect," "future," and terms of similar import (including the negative of any of these terms) may identify forward-looking statements. However, not all forward-looking statements may contain one or more of these identifying terms. Forward-looking statements in this communication may include, without limitation, statements regarding the potential scope and timing of the financial restatement, plans and objectives of management for future operations, projections of income or loss, earnings or loss per share, capital expenditures, dividends, capital structure or other financial items, the Company's future financial performance, expansion plans and opportunities, and the assumptions underlying or relating to any such statement.
The forward-looking statements are not meant to predict or guarantee actual results, performance, events or circumstances and may not be realized because they are based upon the Company's current projections, plans, objectives, beliefs, expectations, estimates and assumptions and are subject to a number of risks and uncertainties and other influences, many of which the Company has no control over. Actual results and the timing of certain events and circumstances may differ materially from those described by the forward-looking statements as a result of these risks and uncertainties. Factors that may influence or contribute to the accuracy of the forward-looking statements or cause actual results to differ materially from expected or desired results may include, without limitation: the amount, scope and timing of any financial restatement that may be required, information that may be discovered in the course of the Company's completion of the reconciliations of its financial results and related analysis; the ability to recognize the anticipated benefits of the Company's recent business combination transaction, which may be affected by, among other things, the price of oil, natural gas, natural gas liquids, competition, the Company's ability to integrate the businesses acquired and the ability of the combined business to grow and manage growth profitably; integration costs related to the Company's recent business combination; estimates of the Company's future revenue, expenses, capital requirements and the Company's need for financing; the risk of legal complaints and proceedings and government investigations; the Company's financial performance; success in retaining or recruiting, or changes required in, the Company's officers, key employees or directors; current and future government regulations; developments relating to the Company's competitors; changes in applicable laws or regulations; the possibility that the Company may be adversely affected by other economic and market conditions, political disturbances, war, terrorist acts, international currency fluctuations, business and/or competitive factors; and other risks and uncertainties set forth in the Company's most recent Annual Report on Form 20-F filed with the Securities and Exchange Commission (the "SEC").
You are cautioned not to place undue reliance on forward-looking statements because of the risks and uncertainties related to them and to the risk factors. The Company disclaims any obligation to update the forward-looking statements contained in this communication to reflect any new information or future events or circumstances or otherwise, except as required by law. You should read this communication in conjunction with other documents which the Company may file or furnish from time to time with the SEC.
For inquiries regarding NESR, please contact:
Blake Gendron - VP Investor Relations & Business Development
National Energy Services Reunited Corp.
SOURCE: National Energy Services Reunited Corp.
Drop us a line:
Copyright 2023 © ACCESSWIRE. All rights reserved. Privacy Policy | Terms of Service | Responsible Disclosure Guidelines | Status
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,593
|
{"url":"https:\/\/studysoup.com\/tsg\/17177\/university-physics-13-edition-chapter-3-problem-34e","text":"\u00d7\nGet Full Access to University Physics - 13 Edition - Chapter 3 - Problem 34e\nGet Full Access to University Physics - 13 Edition - Chapter 3 - Problem 34e\n\n\u00d7\n\n# Two piers, A and B, are located on a river; B is 1500 m\n\nISBN: 9780321675460 31\n\n## Solution for problem 34E Chapter 3\n\nUniversity Physics | 13th Edition\n\n\u2022 Textbook Solutions\n\u2022 2901 Step-by-step solutions solved by professors and subject experts\n\u2022 Get 24\/7 help from StudySoup virtual teaching assistants\n\nUniversity Physics | 13th Edition\n\n4 5 1 423 Reviews\n14\n4\nProblem 34E\n\nTwo piers, A and B, are located on a river; B is 1500 m downstream from A (?Fig. E3.32?). Two friends must make round trips from pier A to pier B and return. One rows a boat at a constant speed of 4.00 km\/h relative to the water; the other walks on the shore at a constant speed of 4.00 km\/h. The velocity of the river is 2.80 km\/h in the direction from A to B. How much time does it take each person to make the round trip?\n\nStep-by-Step Solution:\nStep 1 of 3\n\n3\/15\/15 Loanable funds market \uf0b7 Market that coordinates the borrowing and lending decisions of business firms and households o Price of loanable funds is the real interest rate \u201cr\u201d \uf0a7 \u201cReal interest rate\u201d means adjusted for inflation o Quantity of loanable funds is the amount saved or invested \u201cQs,I\u201d o Similar to goods\/services graph \uf0b7 Demand for loanable funds 1. Firms demand loanable funds (investment) 2. Downward sloping because as the interest rate decreases the firm will want to borrow more money o Increase in investment: demand cur Supply of loanable funds \uf0b7\uf020\uf020\uf020\uf020Individuals supply loanable funds (through savings) o Savings: after-tax income not spent on consumption \uf0a7 Income\uf0e0taxes\uf0e0disposable income\uf0e0 consumption OR savings \uf0b7\uf020\uf020\uf020\uf020Upward sloping because as the interest rate increases people will want to save more o Increase in savings: supply curve will shift right o Decrease in savings: supply curve will shift left The interest rate \uf0b7 Nominal interest rate: the percentage of the amount borrowed that must be paid to the lender in addition to the repayment of the principle \uf0b7 Real interest rate: the interest rate adjusted for inflation (real cost of borrowing and lending money) o r = i \u2013 pi Interest rate and inflation \uf0b7 When the actual rate of inflation is greater than anticipated: borrowers gain, lenders lose \uf0b7 When the actual rate of inflation is less than anticipated\u201d lenders gain, borrowers lose \uf0b7 Inflation does not help borrowers or lenders in a s systematic manner The interest rate and bond prices \uf0b7 Interest rate and bond prices are inversely rated \uf0b7 When the interest rate rises (falls), the market value of previously issued bonds will fall (rise) The foreign exchange market \uf0a7 Market in which the currencies of different countries are bought and sold o Price is price of foreign currency o Quantity is amount of foreign currency Changes in exchange rate \uf0b7 Appreciation: increase in value of currency relative to foreign currencies o Ex: dollar can buy more euros \uf0b7 Depreciation: reduction in value of currency relative to foreign currencies o Ex: dollar can buy less euros Demand for foreign currency \uf0b7 Demand for foreign currency is : imports + capital outflows o Capital outflows: domestic money invested abroad \uf0b7 Downward sloping because as dollar appreciates (foreign currency depreciates), people can import more and invest more in other countries Supple of foreign currency \uf0b7 Exports + capital inflows o Capital inflow: foreign money invested domestically \uf0b7 Upward sloping because as the dollar depreciates (foreign currency appreciates) foreign countries will demand more domestic exports and will invest more domestically o Demand increases: quantity increases, price of foreign currency increases o Demand decrease: quantity decreases, price of foreign currency decreases o Supply increases: quantity increases, price of foreign currency decreases o Supply decreases: quantity decreases, price of foreign currency increases 3\/17\/16 The foreign exchange market in equilibrium \uf0b7 Equilibrium occurs when supply of foreign currency equals demand for foreign currency o Imports + capital output = exports + capital input \uf0b7 Trade deficit: imports>exports o Capital output \u2013 capital input = exports \u2013 imports \uf0a7 Ex: 3 \u2013 4 = 2 - 3 \uf0b7 Trade surplus: exports> imports o Capital output \u2013 capital input = exports \u2013 imports \uf0a7 Ex: 5-4 = 6-5gn Aggregate goods\/services market \uf0b7 A market that includes all final goods and services (counts all items that enter into GDP) o Example: we would not be looking at the supply and demand of pizza or haircuts or highways etc. it would be the supply and demand of pizza and haircuts and highways etc. o Price of all things (and on graphs): the price index (PI) o Quantity of all things (and on graphs): the real GDP (Y) o Includes 2 supply curves The aggregate demand curve \uf0b7\uf020\uf020\uf020\uf020Aggregate demand (AD) curve: the relationship between the price level and the quantity of domestically produced goods and services all households, business firms, governments and foreigners are willing to purchase o Downward sloping because as price level goes down, quantity demanded of all goods will increase \uf0b7\uf020\uf020\uf020\uf0203 reasons why a decrease in price level will increase the quantity demanded of all goods 1. Increase the purchasing power of money \uf0a7 Graph: as PI decreases, Y will increase and shift to the right 2. Lead to a lower real interest rate, which increases consumption and investment \uf0a7 Graph: supply curve will shift to the right when you have more money in savings and the real interest rate will decrease. When real interest rate goes down, you will be less eager to put money in banks, thus you will start saving less and spend more. \uf0a7 r decreases \uf0e0 c increases, c increases \uf0e0 I increases \uf0a7 PI decreases \uf0e0 save more \uf0e0 r decreases \uf0e0 c increases, I increases \uf0e0 Y increase o WHEN PI DECREASES, Y INCREASES 3. Make domestically produced goods less expensive relative to foreign goods \uf0a7 Exports increase \u2013 imports The aggregate supple curve \uf0b7\uf020\uf020\uf020\uf020Relationship between a nations price level and the quantity of goods supplied by its producers o 2 relationships 1. Short-run aggregate supply curve 2. Long- run aggregate supply curve \uf0a7 Short run is time period in which something cannot be fixed, long run everything can be changed The short-run aggregate supply curve \uf0b7 Short-run aggregates supply curve (SRAS): upwards sloping because an increase in the price level will improve the profitability of the firms and cause them to increase output o Profit = revenue \u2013 cost o Ex: PI=1 \uf0e0 revenue ($1x100) - cost ($100)= $0 o PI=2\uf0e0 revenue ($2x100) \u2013 cost ($100)=$100 \uf0b7 Many of producers costa are still fixed The long-run aggregate supply curve \uf0b7 Long-run aggregate supply curve (LRAS): vertical because in the long-run people have had the time to adjust and so a higher price level will increase costs as much as it increases revenues o Profit = revenue \u2013 cost BUT ALL THINGS ARE UP FOR CHANGE o Ex: PI=1\uf0e0 revenue ($1x100)- cost ($100)= $0 o PI=2\uf0e0 revenue ($2x100)-cost ($200)=$0 o VERTICAL LINE ON GRAPH AT Y \uf0b7 Firms have no incentive to change production at any price level because in the long-run everything will balance out and profits wont change \uf0b7 Indicates potential output (Yf) of the economy \uf0b7 Where SRAS intersects LRAS: actual price level = expected price level Short-run equilibrium \uf0b7\uf020\uf020\uf020\uf020Occurs at the intersection of the AD and the SRAS o AD downward sloping, SRAS upward sloping o Intersection= e Long-run equilibrium \uf0b7 Occurs where AD, SRAS and LRAS all intersect at a single point o AD downward sloping, SRAS upward sloping, LRAS vertical line through intersection of AD and SRAS at point e o LRAS indicates wherever Yf is o Intersection is Y* \uf0a7 Yf = Y* \uf0e0 not in a recession, not in an expansion \uf0b7 Occurs when 1. We correctly anticipate price level 2. No expansion or recession (Y = Yf) 3. actual rate of unemployment = natural rate of unemployment ( U = U*) \uf0b7 If Yf is further left than Y* \uf0e0 expansion \uf0b7 If Yf is further right than Y*\uf0e0 contraction ***think of the AD and SRAS as changing in the same way as all other graphs and ignore LRAS because all it is showing is the potential output, it is not actually there. Like yellow first-down line when watching football on tv, it is just there to help the viewer*** 3\/15\/15 Loanable funds market \uf0b7 Market that coordinates the borrowing and lending decisions of business firms and households o Price of loanable funds is the real interest rate \u201cr\u201d \uf0a7 \u201cReal interest rate\u201d means adjusted for inflation o Quantity of loanable funds is the amount saved or invested \u201cQs,I\u201d o Similar to goods\/services graph \uf0b7 Demand for loanable funds 1. Firms demand loanable funds (investment) 2. Downward sloping because as the interest rate decreases the firm will want to borrow more money o Increase in investment: demand cur Supply of loanable funds \uf0b7\uf020\uf020\uf020\uf020Individuals supply loanable funds (through savings) o Savings: after-tax income not spent on consumption \uf0a7 Income\uf0e0taxes\uf0e0disposable income\uf0e0 consumption OR savings \uf0b7\uf020\uf020\uf020\uf020Upward sloping because as the interest rate increases people will want to save more o Increase in savings: supply curve will shift right o Decrease in savings: supply curve will shift left The interest rate \uf0b7 Nominal interest rate: the percentage of the amount borrowed that must be paid to the lender in addition to the repayment of the principle \uf0b7 Real interest rate: the interest rate adjusted for inflation (real cost of borrowing and lending money) o r = i \u2013 pi Interest rate and inflation \uf0b7 When the actual rate of inflation is greater than anticipated: borrowers gain, lenders lose \uf0b7 When the actual rate of inflation is less than anticipated\u201d lenders gain, borrowers lose \uf0b7 Inflation does not help borrowers or lenders in a s systematic manner The interest rate and bond prices \uf0b7 Interest rate and bond prices are inversely rated \uf0b7 When the interest rate rises (falls), the market value of previously issued bonds will fall (rise) The foreign exchange market \uf0a7 Market in which the currencies of different countries are bought and sold o Price is price of foreign currency o Quantity is amount of foreign currency Changes in exchange rate \uf0b7 Appreciation: increase in value of currency relative to foreign currencies o Ex: dollar can buy more euros \uf0b7 Depreciation: reduction in value of currency relative to foreign currencies o Ex: dollar can buy less euros Demand for foreign currency \uf0b7 Demand for foreign currency is : imports + capital outflows o Capital outflows: domestic money invested abroad \uf0b7 Downward sloping because as dollar appreciates (foreign currency depreciates), people can import more and invest more in other countries Supple of foreign currency \uf0b7 Exports + capital inflows o Capital inflow: foreign money invested domestically \uf0b7 Upward sloping because as the dollar depreciates (foreign currency appreciates) foreign countries will demand more domestic exports and will invest more domestically o Demand increases: quantity increases, price of foreign currency increases o Demand decrease: quantity decreases, price of foreign currency decreases o Supply increases: quantity increases, price of foreign currency decreases o Supply decreases: quantity decreases, price of foreign currency increases 3\/17\/16 The foreign exchange market in equilibrium \uf0b7 Equilibrium occurs when supply of foreign currency equals demand for foreign currency o Imports + capital output = exports + capital input \uf0b7 Trade deficit: imports>exports o Capital output \u2013 capital input = exports \u2013 imports \uf0a7 Ex: 3 \u2013 4 = 2 - 3 \uf0b7 Trade surplus: exports> imports o Capital output \u2013 capital input = exports \u2013 imports \uf0a7 Ex: 5-4 = 6-5gn Aggregate goods\/services market \uf0b7 A market that includes all final goods and services (counts all items that enter into GDP) o Example: we would not be looking at the supply and demand of pizza or haircuts or highways etc. it would be the supply and demand of pizza and haircuts and highways etc. o Price of all things (and on graphs): the price index (PI) o Quantity of all things (and on graphs): the real GDP (Y) o Includes 2 supply curves The aggregate demand curve \uf0b7\uf020\uf020\uf020\uf020Aggregate demand (AD) curve: the relationship between the price level and the quantity of domestically produced goods and services all households, business firms, governments and foreigners are willing to purchase o Downward sloping because as price level goes down, quantity demanded of all goods will increase \uf0b7\uf020\uf020\uf020\uf0203 reasons why a decrease in price level will increase the quantity demanded of all goods 1. Increase the purchasing power of money \uf0a7 Graph: as PI decreases, Y will increase and shift to the right 2. Lead to a lower real interest rate, which increases consumption and investment \uf0a7 Graph: supply curve will shift to the right when you have more money in savings and the real interest rate will decrease. When real interest rate goes down, you will be less eager to put money in banks, thus you will start saving less and spend more. \uf0a7 r decreases \uf0e0 c increases, c increases \uf0e0 I increases \uf0a7 PI decreases \uf0e0 save more \uf0e0 r decreases \uf0e0 c increases, I increases \uf0e0 Y increase o WHEN PI DECREASES, Y INCREASES 3. Make domestically produced goods less expensive relative to foreign goods \uf0a7 Exports increase \u2013 imports The aggregate supple curve \uf0b7\uf020\uf020\uf020\uf020Relationship between a nations price level and the quantity of goods supplied by its producers o 2 relationships 1. Short-run aggregate supply curve 2. Long- run aggregate supply curve \uf0a7 Short run is time period in which something cannot be fixed, long run everything can be changed The short-run aggregate supply curve \uf0b7 Short-run aggregates supply curve (SRAS): upwards sloping because an increase in the price level will improve the profitability of the firms and cause them to increase output o Profit = revenue \u2013 cost o Ex: PI=1 \uf0e0 revenue ($1x100) - cost ($100)= $0 o PI=2\uf0e0 revenue ($2x100) \u2013 cost ($100)=$100 \uf0b7 Many of producers costa are still fixed The long-run aggregate supply curve \uf0b7 Long-run aggregate supply curve (LRAS): vertical because in the long-run people have had the time to adjust and so a higher price level will increase costs as much as it increases revenues o Profit = revenue \u2013 cost BUT ALL THINGS ARE UP FOR CHANGE o Ex: PI=1\uf0e0 revenue ($1x100)- cost ($100)= $0 o PI=2\uf0e0 revenue ($2x100)-cost ($200)=$0 o VERTICAL LINE ON GRAPH AT Y \uf0b7 Firms have no incentive to change production at any price level because in the long-run everything will balance out and profits wont change \uf0b7 Indicates potential output (Yf) of the economy \uf0b7 Where SRAS intersects LRAS: actual price level = expected price level Short-run equilibrium \uf0b7\uf020\uf020\uf020\uf020Occurs at the intersection of the AD and the SRAS o AD downward sloping, SRAS upward sloping o Intersection= e Long-run equilibrium \uf0b7 Occurs where AD, SRAS and LRAS all intersect at a single point o AD downward sloping, SRAS upward sloping, LRAS vertical line through intersection of AD and SRAS at point e o LRAS indicates wherever Yf is o Intersection is Y* \uf0a7 Yf = Y* \uf0e0 not in a recession, not in an expansion \uf0b7 Occurs when 1. We correctly anticipate price level 2. No expansion or recession (Y = Yf) 3. actual rate of unemployment = natural rate of unemployment ( U = U*) \uf0b7 If Yf is further left than Y* \uf0e0 expansion \uf0b7 If Yf is further right than Y*\uf0e0 contraction ***think of the AD and SRAS as changing in the same way as all other graphs and ignore LRAS because all it is showing is the potential output, it is not actually there. Like yellow first-down line when watching football on tv, it is just there to help the viewer***\n\nStep 2 of 3\n\nStep 3 of 3\n\n#### Related chapters\n\nUnlock Textbook Solution","date":"2021-12-05 13:38:06","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\": 0.31558623909950256, \"perplexity\": 8108.972559473301}, \"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-2021-49\/segments\/1637964363189.92\/warc\/CC-MAIN-20211205130619-20211205160619-00272.warc.gz\"}"}
| null | null |
{"url":"https:\/\/www.r-bloggers.com\/2020\/05\/lying-with-statistics-one-beer-a-day-will-kill-you\/","text":"Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.\n\nAbout two years ago the renowned medical journal \u201cThe Lancet\u201d came out with the rather sensational conclusion that there is no safe level of alcohol consumption, so every little hurts! For example, drinking a bottle of beer per day (half a litre) would increase your risk of developing a serious health problem within one year by a whopping 7%! When I read that I had to calm my nerves by having a drink!\n\nOk, kidding aside: in this post, you will learn how to lie with statistics by deviously mixing up relative and absolute changes in risks, so read on!\n\nThe meta-study \u201cRisk thresholds for alcohol consumption\u201d adheres to the highest scientific standards, that is not the problem. The problem is how they chose to communicate the associated changes in risks for consuming alcohol.\n\nFor example, they tell you that by drinking a bottle of beer a day (half a litre) your risk of developing a serious health problem (like cardiovascular disease, cancer, cirrhosis of the liver, inflammation of the pancreas or diabetes) within one year would increase by 7%, i.e. 63 people on top of 914 people who would get a serious health problem anyway:\n\n63 \/ 914 * 100 # shock horror: nearly 7% more with health problems when drinking half a litre of beer per day!\n## [1] 6.892779\n\n\nSo, what does that mean? That about one in fourteen beer drinkers are going to bite the dust (no pun intended) next year? Fortunately not!\n\nThe problem is that this is a relative change in risk! It does not really help to assess the real danger. Only absolute changes in risk can do that!\n\nTo illustrate we use the personograph package (on CRAN) to show you what is really going on. Taking 2000 people about 18 would develop a serious health issue within one year anyway\u2026\n\nlibrary(personograph)\n\nn <- 2000\nprobl_wo_alc <- 18 \/ n\n\ndata <- list(first = probl_wo_alc, second = 1-probl_wo_alc)\npersonograph(data, colors = list(first = \"black\", second = \"#efefef\"),\nfig.title = \"18 of 2000 people with health problems\",\ndraw.legend = FALSE, n.icons = n, dimensions = c(20, 100),\nplot.width = 0.97)\n## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x,\n\n\n\u2026by consuming about 20 grams of alcohol per day (i.e. about 25 mL) a little more than one person would become seriously sick on top of that:\n\nprobl_w_alc <- 1 \/ n\n\ndata_2 <- list(first = probl_wo_alc, second = probl_w_alc, third = 1-(probl_wo_alc+probl_w_alc))\npersonograph(data_2, colors = list(first = \"black\", second = \"red\", third = \"#efefef\"),\nfig.title = \"About 1 additional case with half a litre of beer per day\",\ndraw.legend = FALSE, n.icons = n, dimensions = c(20, 100),\nplot.width = 0.97)\n\n\nAs you can see, this doesn\u2019t look spectacular at all! Yet, this would have been a good way to communicate the results so that everybody could get a feeling for what they really mean (but as I said, this doesn\u2019t look spectacular at all, go figure!).\n\nDoing the numbers also gives an absolute change in risk by only 0.063%! It is about 50% more probable to die in a house fire (and how many people do you personally know who actually died in a house fire? I don\u2019t know anybody\u2026)!\n\n63 \/ 100000 * 100 # only 0.063% in absolute numbers!\n## [1] 0.063\n\n\nPlease note: I do not say that it is safe to drink alcohol! But you have to put the numbers in perspective and the risk doesn\u2019t seem to be overly high (to put it mildly) when you drink responsibly!\n\nYou see that you can use statistics not only to \u201clie\u201d but to clarify things and communicate facts transparently. So, the problem lies not so much in statistics but in dishonesty and manipulation per se, which is the idea of one of my favorite cartoons (found here: CrossValidated):","date":"2021-06-23 09:40:57","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\": 0.2928362190723419, \"perplexity\": 2531.196187179396}, \"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-2021-25\/segments\/1623488536512.90\/warc\/CC-MAIN-20210623073050-20210623103050-00236.warc.gz\"}"}
| null | null |
Nike Zoom Kobe 8 EXT - "Year of the Snake"
The Year Of the Snake Nike Collection was unveiled this morning and amongst the footwear lineup shown we got a look at the Nike Zoom Kobe EXT. The off the court version of the Kobe 8 swaps out EM for Vac-Tech in order to provide a more comfort driven ride. The upper consists of a red suede one piece design that features laser etching to provide a sleek look. The outsole and liner are fitted with a royal purple and the lace tips, Swoosh and heel logo all get a gold finish to add to this celebratory piece. Keep checking back at Modern Notoriety for more info on these, but in the meantime check out the photos and let us know what you think.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,048
|
{"url":"https:\/\/commons.wikimedia.org\/wiki\/Commons:Village_pump\/Archive\/2015\/03","text":"# Commons:Village pump\/Archive\/2015\/03\n\n This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.\n\n## Pictures of pigeons that might be copyvio\n\none example\n\nHallo, is there someone able to read these pages (from April 2007)?\n\nHere on Commons are some pictures uploaded 2007-06-19 that might be copyvios (same user uploaded other pics as \"own\" from different sources, like Alex Sell[1] and a french(?) page[2]). My question is: is there a licens given for these pictures? Or might they be copyvios by themselves (there are also pics from featherside). It would be a loss to delete them all, they are also used to illustrate articles.\n\nThanks for your help, --PigeonIP (talk) 16:19, 28 February 2015 (UTC)\n\nEight years ago, too late for simple TinEye researchs. The uploader sticked to two camera models and some images without Exif, all about your Pigeon topic. And you already invested time for the Konigsberg Morehead. I'd assume good faith and vote keep for INUSE. \u2013Be..anyone (talk) 04:58, 1 March 2015 (UTC)\n\n# March 01\n\n## \"Female humans\"\n\nThere are no pictures directly in Category:Male humans. There are (at this moment) 556 photos directly in Category:Female humans. Recently, someone added the latter category to some of my photos, which is how I became aware of it.\n\nI really dislike this. It seems like a reduction of women to their gender rather than seeing them as fully human. I particularly dislike this when it is done to photos I took, because it feels like reduction of the subjects of my photos to their gender.\n\nI think that the correct solution to this is that neither Category:Male humans nor Category:Female humans should directly contain individual images. - Jmabel\u00a0! talk 16:25, 21 February 2015 (UTC)\n\nI agree with above. I warned this user, but he continue editing nevertheless, so I blocked him for 2 hours. And I removed this category with VFC. Regards, Yann (talk) 16:36, 21 February 2015 (UTC)\nThanks, Yann. Even though I'm an admin, I always hesitate to do something like this without having some sort of indication it's not just my solo view. I've suggested to him that if this is (as he now says on his talk page) part of a process of classifying more deeply that he use a hidden category as his temporary holder. - Jmabel\u00a0! talk 18:07, 21 February 2015 (UTC)\nBeing 'bold' about this (since it seems obvious) I added {{Categorize}} to both, which might help in the future. Revent (talk) 18:33, 21 February 2015 (UTC)\nSorry about the mess - I agree that there is no need for a lot of pictures in this category. It was intended to be only a temporary stage. Jmabel suggested a great idea of Hidden categories under my username. If this is acceptable to you would be happy if you pass to Category:Temporary categories for User:Chenspec Cat-a-lot - Female humans all the pictures that were in Category:Female humans. What do you think? Chenspec (talk) 18:56, 21 February 2015 (UTC)\n@Chenspec: Use {{user category}} instead of {{hiddencat}} for this, please. Other than that, don't know why anyone would complain about it as a temporary manrker. Revent (talk) 19:12, 21 February 2015 (UTC)\n@Revent: Excellent - done. Now, how do I pass the pictures to the new category? Need to restore ... Chenspec (talk) 19:25, 21 February 2015 (UTC)\nYes, a user category is more appropriate, sorry I didn't think of that. And I'm not sure what you mean by \"passing\" pictures to a category. You can place them in a user category or hidden category exactly like any other category. - Jmabel\u00a0! talk 01:25, 22 February 2015 (UTC)\n@Jmabel: All the pictures were removed from the category \"Female humans\" and it took me a long time to collect them. Do I have to find them again manually So I can categorize them in \"Category:Temporary categories for User:Chenspec Cat-a-lot - Female human\", or there is a way to restore them more effectively? Chenspec (talk) 07:36, 22 February 2015 (UTC)\nIf you know who removed them, they are probably all more or less in a row in his\/her contributions list. Failing that, your own contributions list would probably have them reasonably close together. And, yes, that categorization seems appropriate to me. Sorry you got blocked on this, I didn't mean to take it to that level but I guess that happened before we had time to discuss it calmly. - Jmabel\u00a0! talk 17:51, 22 February 2015 (UTC)\n@Jmabel: It's OK - The main thing is that everything worked out for the better. I will try to restore the pictures. Thanks for the help and guidance for the new category! Chenspec (talk) 20:30, 24 February 2015 (UTC)\nJust noting here for the record, just added (per a request on Yann's talk page) 583 images previously added by Chenspec to 'Female humans' to his user tracking category. More useful than would be apparent at first glance, nearly all are images of Wikipedians in various contexts with no categorization other than 'this is a Wikipedian'. Revent (talk) 22:28, 24 February 2015 (UTC)\n\n@Chenspec: I am sorry, but what is the befenit of copying pictures from Category:Females with birds. Where do you want to pass this pictures? other than \"with birds\"? --PigeonIP (talk) 11:03, 1 March 2015 (UTC)\n\nI understood that it does not matter what pictures I put my user category. Regarding your question - there are pictures that fit more than one category, according to what you see in them. If there are more categories they will be eligible to I'll put them, if not then do not. I hope this answers your question Chenspec (talk) 17:19, 1 March 2015 (UTC)\n\n# February 22\n\nI came across this image of a wartime radar system in Europe. This is clearly taken by Army personal, the only people who would have seen one in the field. Yet the Smithsonian claims copyright on it. Do we, as in the case of other examples, ignore their claim for these cases? Maury Markowitz (talk) 12:30, 24 February 2015 (UTC)\n\n\u2022 I probably would, while noting the dubious claim of copyright. - Jmabel\u00a0! talk 17:03, 24 February 2015 (UTC)\nFrom what I see, they don't seem to be explicitly claiming a copyright in this particular image (they credit it to the National Archives). The Smithsonian's Terms of Use make it clear the 'general' copyright claim on their website is to \"the compilation of content that is posted on the SI Websites, which consists of text, images, audio, video, databases, design, codes and software\" and that \"the Smithsonian does not necessarily own each component of the compilation.\" They just seem to engage in the (highly questionable, but common) practice of telling users they must license anything obtained from their website regardless of if they actually own the copyright in the particular work. Revent (talk) 18:34, 24 February 2015 (UTC)\n\u2022 The institution is free to make whatever claims they wish, even if they are not legally enforceable. If they have added value in the metadata, such as writing descriptions or adding lots of structured detail, then they can make a valid claim of creative ownership for that metadata. Automatically created metadata, such as what the source was, details copied from elsewhere, or basic facts like a date or the original author\/photographer, are not creative enough to make a claim for.\nIf anyone wished to scrape information and images from the si.edu, it would be a smart move to first write to the website contact and explain what the plan was, and give them a chance to object to it and explain if they have a legally valid claim that you may be unaware of. Putting aside this specific case, as I expect the Smithsonian to encourage open knowledge, if an institution were to issue take-down notices or legal challenges against a Commons uploader, having a previous good faith correspondence on record would be a great way of dismissing such actions. -- (talk) 19:10, 24 February 2015 (UTC)\nAnyway, the Smithsonian is part of the U.S. Federal Government, so it's questionable whether they can have copyright. {{PDUSGov}}. Adam Cuerden (talk) 00:26, 25 February 2015 (UTC)\nAdam, I believe the Smithsonian is actually something of a special case (though I'm not sure of the details and it wouldn't apply to this photo). As I understand it, a lot of their work is created by contractors who are not technically federal government employees, and so it can be copyrighted. - Jmabel\u00a0! talk 00:37, 25 February 2015 (UTC)\nThe US Government can not (in most cases) claim copyright in it's own works, but can own copyrights transferred to it by others (this is explicitly stated in 17 USC \u00a7 105). Not that this applies here, but to state that the US Government cannot own a copyright is mistaken. Revent (talk) 04:30, 25 February 2015 (UTC)\ni do not see a copyright claim here. you are making a guess about who took the photo. having had conversations with smithsonian people: they have a legal department, and practice of trying to pay for digitzation with fees. they tend to put NC on PD images. however, they do not issue takedown notices for PD claims over their NC. the smithsonian institution is a hybrid with federal support and private money. they are a repository of government and private collections, i.e. you cannot know what the copyright of an item is, but with research of the metadata on a case by case basis. we can go to the National Archives, who is the repository of the Naval Photographic Center, and find this item, or related film [3] but it does not appear to be digitized there yet. see also [4] Slowking4Farmbrough's revenge 01:31, 26 February 2015 (UTC)\n\nThe Smithsonian Institution is not the US Government. It publishes copyrighted materials on a regular basis, and actually pays photographers itself for its publications. Trying to say \"you are the government so you do not own what bears your copyright notice\" is unwise. [5] is clear. If they assert copyright, then you must abide by their terms of use. Collect (talk) 21:29, 26 February 2015 (UTC)\n\n@Collect: I do not think anyone is trying to say that copyright is not a 'real and valid concern', as you put it in your edit summary. It's simply that, with experience, Commons editors have learned that assertions of copyright from certain sources, including the Smithsonian, need to be evaluated critically (hence the discussion). This image, for example, is on the same website, with exactly as much of a 'copyright assertion' made. Are you going to claim that it's not usable on Commons because the Smithsonian asserts a copyright? Revent (talk) 22:17, 26 February 2015 (UTC)\nI am saying that folks who conflate the Smithsonian and the US Government are making an exceedingly grave error. The proper procedure is to contact the Smithsonian as they ask, and ask whether that particular image is covered under any copyright, not to assert \"the Smithsonian can not assert copyright on anything\" because that position will fail the second a lawyer sees it. Cheers. Collect (talk) 00:37, 27 February 2015 (UTC)\nnot very grave at all: nobody died from wishful thinking. and since they have not issued a DMCA, you would have a greater chance of federal court with a FOP german statue photo, where we have cases of a DMCA. in this case the metadata is pretty clear PD-USGov, but lets go scanning at Archives II where the original is. Slowking4Farmbrough's revenge 00:21, 28 February 2015 (UTC)\nOkay; if I assert copyright on that picture, then must you abide by my terms of use? A bullshit assertion is bullshit coming from me or them. Wikipedia says \"More than two-thirds of the Smithsonian's workforce of some 6,300 persons are employees of the federal government\" and by law, anything those employees create as part of their employment is not covered by copyright. It's true that the Smithsonian is the effective origin of a lot of copyrighted work, and it's rather unfortunate that it's a waste of time to try and contact them to get the correct legal copyright status of much of their work.--Prosfilaes (talk) 00:37, 28 February 2015 (UTC)\nwelcome to the internet. this is a common institutional attitude, that the researchers will come to them. the institutions tend not to have drunken the \"free\" kool-aid of license purity. the world does not exist to give you clear licenses. beware, you cannot separate the employees into fed and non-fed. they work on both fed and non-fed funded projects. and they are not going to share their time sheets for your convenience. you also have institutions sending nasty notices asserting \"sweat of the brow\"; and institutions not partnering with commons because admins don't like their name. it will take a long time hand holding, to change institutions. Slowking4Farmbrough's revenge 14:38, 1 March 2015 (UTC)\n\n## How to perform complex file searches in Commons\n\nHello, I am wondering if there is a way in Commons to carry out advanced search of files such as the following (I list them separately just in case one is feasible while another one is not):\n\n1.-\"Files with extension X (say .png)\" & \"Linked from Wikipedia Y (say, French) more than Z times\"\n\n2.-\"Files with extension X (say .svg)\" & \"uploaded in the last Y days\"\n\n3.-\"Files with extension X (say .svg)\" & \"uploaded in the last Y days\" & \"Uploaded by user Z\"\n\n4.-\"Files with extension X (say .svg)\" & \"Belonging to category Y\"\n\nAre the above and similar complex searches feasible in Commons? If so, how?\n\nThank you!--Rowanwindwhistler (talk) 06:26, 26 February 2015 (UTC)\n\n4 can be achieved to a large extend by simply searching for \"svg incategory:Y\" (link) this will also catch some extra stuff you don't want but in the results you can do \"ctrl-f\" and then search for \".svg\" another way (for categories with up to 200 files) is to go to the category you want to search (link) and above the media files (below the header) you can select the filetype, however I've found this to only work for small categories as it only filters the 200 results on the current page. For point 3 you can simply use the user uploads page and search again with ctrl-f for .svg (only works for up to 500 files). There are likely other and better ways to achieve this (for example using API-queries), these are however some quick and easy ones to start with. Basvb (talk) 16:16, 26 February 2015 (UTC)\nThank you for the suggestions. For the first one, I found the category has to be quoted if its name contains more than 1 word (svg incategory:\"Historical SVG maps in Greek\"). For the second, I am afraid I have not been able to see where to select the filetype yet... I have tried searching for it in a small category] but I did not find where to filter by filetype within it... Maybe I need to enable some configuration option to see it?--Rowanwindwhistler (talk) 20:43, 26 February 2015 (UTC)\nOw I forgot you can also use catscan, that scales up better. I've looked it up and you indeed have to enable a configuration, namely the \"GalleryFilterExtension\". Mvg, Basvb (talk) 00:46, 27 February 2015 (UTC)\nAha, I can see the extension menu now! Good indeed for quick filtering... Thank you for the tip on catscan. It seems to be more complex than I expected or else I am doing something wrong, though. If I try something simple like \"Categories=Maps of the Battle of the Nile+Last change Max age=24\" I should be getting a couple of maps I have uploaded a moment ago but I get nothing... I guess I need to check somthing else in the form to make it work...--Rowanwindwhistler (talk) 11:08, 27 February 2015 (UTC)\nWhen using the search on commons, I find doing intitle:svg can be a good (not perfect) way for finding things with a specific file type. You may also be interested in fr:Sp\u00e9cial:Fichiers_les_plus_li\u00e9s. All four of these queries can be done via sql access (For example: #1 [6], #2 [7], the last two can also be pretty easily done as well), but that's probably two complex to be usable by average commons user. Bawolff (talk) 16:57, 27 February 2015 (UTC)\nI will keep that in mind too. About the URL, I did not know the tool but, though relatively complex for me I think I could manage to create the queries I need by comparing with others and it does look very flexible... However, I wonder where we can find the different fields and variables available to build the query. Is there a list somewhere or does it have to be deduced from the output of a query to a file? Thank you again for all your suggestions.--Rowanwindwhistler (talk) 09:08, 28 February 2015 (UTC)\nJust one quick question related to the SQL syntax to use: would it be possible to filter also by category name text? For instance, would this be feasible:\n1. SVG files with \"map\" in at least one of its categories, linked 5k-10k times from the French Wikipedia & uploaded during the last month.\nPlaying around with the tool, I got as far as this (searching for \"map\" in the file name but not in the category as I do not know how to search the categories or whether this is possible at all). I think I would be able to add the time condition using the above examples but I have no clue how to check the categories...--Rowanwindwhistler (talk) 10:50, 28 February 2015 (UTC)\nthe schema (list of available variables) is available at https:\/\/tools.wmflabs.org\/tools-info\/schemas.php?schema=commonswiki and specificly the public parts you are allowed to use at https:\/\/tools.wmflabs.org\/tools-info\/schemas.php?schema=views (note its even possible to do crosswiki queries which combine fields from separate wikis. See for example the collapsible section on the bottom of [8]) the mediawiki wiki also has some information, e.g. mw:manual:categorylinks table. For your question about categories try something like (untested): use commonswiki_p;Select img_name, cl_to, count(*) from image inner join globalimagelinks on gil_to = img_name\n\ninner join page on page_namespace=6 and page_title = img_name inner join categorylinks on cl_from = page_id where img_media_type = 'DRAWING' and img_major_mime=\"image\" and img_minor_mime = 'svg+xml' and cl_to like '%map%' and gil_wiki = 'frwiki' and img_timestamp > 20150200000000 group by img_name , cl_to, having count(*) < 10000 and count(*)>5000;. Bawolff (talk) 19:56, 1 March 2015 (UTC)\n\n## File:Visualisation_1_Trillion.svg\n\nTo me the first four powers of 10 in this SVG appear as identical blobs in any browser window I can create. Is it possible to fix this file? Otherwise I doubt it's utility for illustrating the Powers of ten article on en:WP and elsewhere. Rich Farmbrough, 02:26 1 March 2015 (GMT).\n\nThe utility is up to editors at \"en:WP and elsewhere\", as they are the one deciding which graphics to use. But I agree that the creators of this graphics failed at show first 4 powers of 10 and I doubt that anybody will be able to show objects differing at 12 orders of magnitude on one graph. --Jarekt (talk) 04:09, 1 March 2015 (UTC)\nIt can be done as a video (a surf on Youtube will find a few that go from sub-atomic particles and zoom out to the entire known universe with a powers-of-ten countdown) however unless you show a logarithmic scale, it cannot be done in a static graph as the resolution would have to be 1012 pixels wide. -- (talk) 11:37, 1 March 2015 (UTC)\nWouldn't that be the cube route of 10^12, or 10,000 pixels? But still way too big to fit within a 2000 pixel image. Delphi234 (talk) 07:18, 3 March 2015 (UTC)\n\n# March 02\n\n## Crowdfunding campaign for a macro lens for Jee\n\nHi all,\nthis is just a short note to let you know that a small group of Commons contributors have started a crowdfunding campaign at Indiegogo to fund a new macro lens for our very own @Jee.\n\nThe campaign was coordinated at User talk:Jkadavoor\/campaign and will end on March 24, 2015. Please have a look at the campaign page to see if this is something that you're willing to support. Thanks, odder (talk) 14:08, 22 February 2015 (UTC)\n\nThanks for the announcement, odder. We in the campaign team were not sure if it was appripriate to announce a campaign run on a commercial website for a specific user here. -- Slaunger (talk) 20:41, 22 February 2015 (UTC)\nCool!\u00a0:) Rehman 14:30, 22 February 2015 (UTC)\nNice campaign. It is better done than previous ones. Maybe too much figures, I would have emphasise the description of the volunteer and his work (with a quote or an example of a photo report). Some remarks for the next campaigns: i) for a commonist, we should see his work on the main page, ii) avoid specific terms commonly used on Wikimedia (FP, QI) or explain them, iii) don't forget to create an hashtag to make a viral campaign. Pyb (talk) 15:14, 22 February 2015 (UTC)\n@Pyb:: Thank you for your feedback.\n\u2022 You may be right about the balance between facts about Jee and the campaign and slides with pretty pics. In a previous version of the promotional video for the campaign, there were many more pretty pics, so at least that aspect has improved, and I think the balance depends on the target audience - which was actually a bit hard for us to establish. Should we target Wikimedians or a completely different audience? We tried to do a bit of both, but it appears, so far that the the vast majority of donations is from Wikimedians with a high concentration of active Commons users. Maybe, as the campaign progress, it will attract a wider audience. Anyway, it seems like we are not doing too bad as 93% of the pledged amount have been sponsored already here on the launch day and there is still a month to go . (This should not keep people from donating though, as there is plenty of other useful gear, which could be of use for Jee (macro flash, bag, tripod, remote control, wildlife lens, spare battery, ...)). --Slaunger (talk) 20:41, 22 February 2015 (UTC)\n\u2022 i) I am not sure I understand what you mean about seeing his work on the main page? Do you mean the Main Page of Commons? Do you mean today? I am sure several of Jees pics have been picture of the day previously. I do not think it would be appropriate to try and coordinate a campaign done on a private web page for a single user with the Commons Main page.\n\u2022 ii) I am not sure I understand this thing about avoiding specific terms either. We do not mention the acronyms FP or QI anywhere. We mention featured pictures in the campain text explaining they are among the finest and linking to the actual Commons page. In the campaign video featured pictures is mentioned, but I do not think it necessarily needs further explanation at this stage. I think most people would understand that featured is something that somehow stands out as being especially good (which is sufficient).\n\u2022 iii) A hashtag is probably a good idea. I have no experience when it comes to hashtags and how that can aid the campaign. You mean something like #MacroLensForJee\u00a0?\nAgain, thanks for your comments. -- Slaunger (talk) 20:41, 22 February 2015 (UTC)\nI wonder if Pyb's comments refer to the Commons page where we planned the campaign, rather than the Indiegogo campaign itself? I can't really match the comments up with either the video or campaign page. I (and I suspect Slaunger) are too old for this hashtag stuff, but if anyone here is more social-media-aware and wants to help make this viral, please do so or offer suggestions on the User talk:Jkadavoor\/campaign page. We set a modest target for the campaign but there's plenty very useful equipment that could be purchased if the goal is exceeded. -- Colin (talk) 21:21, 22 February 2015 (UTC)\nI refer to the previous campaigns which didn't succeed or didn't succeed very well (Poco a poco, Tony the Tiger and Ryan Hodnett). I've nothing to say about Jkadavoor campaign because I like it\u00a0;) Pyb (talk) 21:57, 22 February 2015 (UTC)\nSo, we do have clients, like a for-profit outfit, yet we fund expenses on goodwill, like a non-profit. Sweet. What could ever go wrong\u2026? -- Tuv\u00e1lkin 19:16, 22 February 2015 (UTC)\n\u2022 Hi Tuv\u00e1lkin, as I don't speak very well english, can you precise please, is it a question or is there an issue for Commons maybe? -- Christian Ferrer 21:36, 22 February 2015 (UTC)\nPeople who send eMails to Wikimedia eMail addresses are called \"customers\" (English\u2192Malayalam\u2192English translation might have made this \"clients\") in OTRS, by the software. Regarding the lens, I think it's WMIN's job to fund its purchase (and lend it the WMUK Mac mini way). \u00a0\u00a0\u00a0FDMS\u00a0\u00a04\u00a0\u00a0\u00a0 20:34, 22 February 2015 (UTC)\nFDMS4 It would be nice if WMIN sponsored such projects, but from browsing their site it does not appear to me that they have any kind of grant program. They have a lot of information about how to donate to WMIN, not the other way around AFAICT. -- Slaunger (talk) 20:59, 22 February 2015 (UTC)\nIt would be nice if WMF + regional groups did more grant making for things like this. But it doesn't seem very high priority to fund individuals or they want to attach all sorts of strings (a loan rather than gift). -- Colin (talk) 21:33, 22 February 2015 (UTC)\nIn regards to grants, WM-AU has a camera equpment program. I got the large equipment support grant ($1000) which covered half the cost of my camera but I did reinvest the$1000 for a 50mm lens, more SD cards and a flash unit a few years ago. Bidgee (talk) 22:11, 22 February 2015 (UTC)\n@Slaunger: There is a Grants page, which redirects to a page called \"Microgrants\". However, no matter matter whether they would, I just think they should fund such projects. \u00a0\u00a0\u00a0FDMS\u00a0\u00a04\u00a0\u00a0\u00a0 22:13, 22 February 2015 (UTC)\nThat grants page seems a bit dead, with nothing listed as \"approved\" for ages. There are differences between Australia and other developed nations, and India and other developing nations -- camera equipment costs about the same yet wages and labour and local costs are hugely different. This may influence whether it is more cost effective to locally-fund activities such as training or hiring rooms vs purchasing equipment. And anyway, the money comes from donations whether via WMF or our own efforts. But I would like WMF to consider funding such grants, which are cheap compared to the cost of organising a conference or paying US salaries. -- Colin (talk) 10:01, 23 February 2015 (UTC)\n\nBtw, in case anyone where Jee is, he's had to go away for a short while for family reasons, so doesn't have wiki access. I'm sure he's very touched, as I am, by the generosity and goodwill shown. -- Colin (talk) 21:33, 22 February 2015 (UTC)\n\nI'm very happy to report that the modest target of $750 has been met in one day. Clearly we underestimated the generosity of the Commons community. We were encouraged to set a low target since failure to meet the target incurs hefty penalty fees from Indiegogo. But there is more equipment that will be very useful for Jee, from the essential components of every serious photographer's kit (good camera bag, tripod) to the specialist equipment to take the best macro pictures in poor light (a macro flash). So further donations are very very welcome and will be wisely used. Of the 1000-odd photographs Jee has uploaded to Commons so far, more than half are illustrating Wikipedia articles, which is a strong measure of high quality educational photography. -- Colin (talk) 23:28, 22 February 2015 (UTC)$1.635 now\u00a0:). --Steinsplitter (talk) 12:21, 24 February 2015 (UTC)\n2.000$now :). Awesome! -- Slaunger (talk) 20:44, 25 February 2015 (UTC) Thanks all for your helps and supports. I was away for a few days to to some unexpected personal matters. Back now and catching up. Jee 08:25, 24 February 2015 (UTC) Fantastic work. Congratulations to everyone involved. --99of9 (talk) 00:03, 3 March 2015 (UTC) Hi, this is Ravi from Wikimedia India. I am happy to know about Jeevan Jose's excellent work and the support he has been receiving from the community both on-wiki and off-wiki. Wikimedia India is well aware of needs like these and has put a Infrastructure Scholarship program in place for providing equipments and services that enable more and better contributions from already active community members. Wikimedia India is working under a limited budget. But, we will try our best to meet such needs. This is done under a FDC grant from WMF. So, in principle, the larger movement recognizes and supports needs like these. As Bidgee points out, there are also precedents in the Wikimedia World. Wikimedia Australia once had give a scholarship for the purchase of camera to User:99of9. WMAU has documented their learning for the wider movement here. <quote> The most recent innovation that is beneficial to the broader Wikimedia movement is our wmau:Proposal:Camera equipment program, which supports volunteers and improves Wikimedia Commons. Before the WMAU program was approved in January 2012, the committee had approved one small grant for camera equipment in July 2011 (wmau:Resolution:Toby Hudson's Small Grant). With the program in place, we have approved reimbursement for camera equipment purchases of$1600, and are currently reviewing another application for reimbursement of $500. <\/quote> Having said these, I would also like to highlight that grants for equipments like these have to be properly accounted by each organization according to the laws of the countries they have been registered. Some times, it can be impossible or may involve lot of paper work (especiallywhen foreign grant money is involved) and risk in case of damage or loss of the equipment. So, it is up to each organization to figure out how to support these needs. But, there is no second opinion that movement funds should be used for meeting such needs. Thanks.--Ravidreams (WMIN) (talk) 07:08, 2 March 2015 (UTC) The precedents aren't particularly impressive or indicative of a healthy regular grant support for Wikimedians. The one for 99of9 (Toby Hudson) wasn't (from what I read) a \"scholarship for the purchase of camera\" but a grant of$200 towards a $900 macro lens he was buying. In return he promised a certain number of usable images, which has been achieved. The Australian program page looks fairly dusty (is there another page where applications are discussed\/approved?) and seems to have rather stiff requirements (such as 1000 images in 1000 categories). It seems more concerned with quantity than quality (size requirements of 1000px are ridiculously meager and indicative of a history by some of only donating small size images to Commons while retaining full-size images for commercial sale). The India program has offered$79 towards the loan of a scanner. I understand the limited budget, which is why I think this is something WMF should be looking at, for whom $79 is small change. -- Colin (talk) 08:20, 2 March 2015 (UTC) From what I know of WM-AU, it is almost a zombie chapter, slowly eating its way through funding and offering very little back to the community or Wiki projects, arguably with the exception of those photography grants I suppose. I have a feeling that it's largely down to the lack of interest of its members, and partly due to the large geographical distances making it hard for members to meet and organise... I recently asked why WM-AU hasn't been involved in Wiki Loves Monuments, and the response was that nobody was interested in organising it. It's a shame. Australia is a relatively 'new' country but it has plenty of interesting 19th century monuments, and certainly not devoid of talented Wikipedian photographers (although history shows that the wildlife and landscapes are more of an interest than buildings!). Diliff (talk) 12:11, 2 March 2015 (UTC) @Diliff: I don't think that's a fair characterization. I'd say it's simply a small chapter in comparison to some of the behemoths we know and love. I'm not on the committee but am a happy member. There is very significant support from WMAU for activities in the GLAM sector (see the GLAMWIKI newsletter for month-by-month details, including stacks of library training and engagement). It's not always financial (because as you say we haven't applied for major central funding, apart from the country-based fundraiser). The committee networked with Wikibomb + and Wikimedia in Higher Education organizers to provide volunteer support. The photography equipment grant scheme discussed above is still in operation, and although the numbers sound low, there are not that many of us contributing high volumes on Commons. Commons also benefits from the GLAM relationships: I now have ~monthly contact with State Librarians in my state who are now established Wikipedians in their own right, and have managed to convince their institution to properly acknowledge the permission status of out-of-copyright works, enabling mass uploads of 12,000 items of historic media so far. And that's just the stuff I've benefitted from... There's also been a lot of work on wikitowns, and I believe there is a standing offer to support meetups (which is sometimes taken up in Melbourne). @Gnangarra: and @Kerry Raymond: are both Commons contributors on the current committee and may have more to add (sorry if I left out others). To find out more, please consider attending Wikiconference Australia 2015, another event they are organizing. --99of9 (talk) 23:40, 2 March 2015 (UTC) I'm happy to stand corrected then! I'd like to see the chapters more involved, but I guess we can't magic interested participants out of thin air, and it has to happen organically. Australia used to contribute a lot more active photographers in the past than seems to be the case now. I know that the images we see on QI and FPC are just the tip of the iceberg, but there used to be more active participation from Aussies than seems to be the case now. For what it's worth, I will one day soon (this year or perhaps next) be returning to Australia, so I suppose I shouldn't upset my future local chapter. :-) Diliff (talk) 00:16, 3 March 2015 (UTC) This probably isn't the place to have a conversation about WMAU and WLM, but the problem is not the organisation of it. The problem is that that we must upload datasets of monuments to the WLMdatabase. In most cases the datasets (which are maintained by a number of govt agencies plus some non-govt ones) are not avaialable to us. For example, we could do it for Queensland because I have negotiated access to that dataset, but we could not do it for most other states. To collect and enter all that data manually is a massive task, which understandably nobody is very keen to do. Also we wanted to include war memorials, which are very important culturally in Australia and particularly significant with the Gallipoli cententary approaching, yet we were told our war memorials were not acceptable for inclusion in WLM. If WLM would be more flexible about its requirements for participation, then we would probably be taking part. Also, our chapter is very active in outreach with programs of edit training and public talks (for any group or individual that asks us) at no cost (see our Past events page for details). We are rolling out hundreds of new articles using content that we have negotiated CC-BY access, we have the two WikiTowns projects running, etc. We have the camera scheme as previously mentioned. Where we have not been so successful is in organising local meet-ups, as turn-out has been pretty discouraging and we really don't know what we can do to improve that. So I am really not sure why people might think us inactive; perhaps we are just too busy doing things to have time to blow our own trumpet. Kerry Raymond (talk) @Colin: I agree that the WMF should (centrally or via chapters) expand something like the WMAU Photography equipment scheme more broadly. Contributing photos to Wiki*edia can be much more costly than contributing words to Wikipedia. IMO it was good to start with requirements on the stiff side, and small co-contributions ($200 of $900) to ensure the system is not gamed, and goes to genuine contributors rather than those in it for the$. Maybe other schemes could set a lower bar but achieve this with a strong oversight and screening process instead. Like you, I argued for quality (specifically the QI process) as a metric when the scheme was being set up, but it was considered complex as it is, and Wikipedia often values images that fill an important niche more than super high quality that we snobs on Commons look for. --99of9 (talk) 23:50, 2 March 2015 (UTC)\n. Although it's great that we're a repository for the world's media, we have to keep in mind that most of it gets very little traffic. One great image that is used in multiple articles on multiple language Wikipedia articles is (IMO) much more valuable than ten or one hundred sub-par images of some obscure object or building, only a couple of which could ever conceivably be used on Wikipedia. Both ends of the spectrum have their place and I'm certainly not saying we should sacrifice one for the other, but the potential utility of an image should be a factor in valuing it and I think those super high quality 'trophy images' of the sort that feature in POTY should be a strategic goal of Wikimedia chapters just as much as bulk dumps of images from GLAMs. Diliff (talk) 00:16, 3 March 2015 (UTC)\nOn the other hand, I took a series of photos at the Harvard Natural History Museum, e.g. File:Epomophorus labiatus Harvard.jpg. As you can see, that's not going to win any POTY competition, but it happens to be our best only photo of the Ethiopian epauletted fruit bat. OTOH, File:Swallow flying drinking.jpg is a striking photo, but the descriptions don't agree on what species it is, and the mainspace pages it's actually in use on have a good selection of alternate images for whatever species it is.--Prosfilaes (talk) 00:36, 3 March 2015 (UTC)\nThe way I see it: ultimately we want quality reader experience, which is roughly: eyeballs * quality. Eyeballs is roughly: project_usage * coreness. Project usage depends either on identified nicheness\/rareness (eg Prosfilaes) or best-in-class quality (eg Diliff). So both are obviously good targets. GLAM is mainly useful because it's a fast way to get a *lot* of diverse\/rare images (and it's also a place where the officialness of chapter backing reaps credibility rewards). It's interesting to compare two straightforward sets: modern quality mostly identified CSIRO images, 603 used + low quality but historic QSA images, 204 used which are roughly on par with my total QI contributions, 1246 used but about 500 of those are Jesus. Obviously I spent a lot more time getting the photographs! --99of9 (talk) 01:19, 3 March 2015 (UTC)\nThat's interesting analysis, I haven't seen Glamtools before. OK, but lets break the numbers down:\nQSA media. 4572 images, 204 total image uses and 156 distinct images used across the Wiki projects, meaning 3.7% of them are used at least once.\nCSIRO media. 3527 images, 603 total image uses and 314 distinct images used across the Wiki projects, with 8.9% of them used at least once.\n99of9's QI media. 143 images, 1246 total image uses and 139 distinct images used across the Wiki projects, with 97.2% of them used at least once.\nDiliff's FP media. 156 images, 11685 total image uses and 154 distinct images used across the Wiki projects, with 98.72% of them used at least once.\nThis is not an attempt to toot my own horn (but toot toot!)... It just goes to show that user-generated content, particularly the 'best-in-class' images are orders of magnitude more useful to the Wiki projects. Whether this is because they are genuinely more useful images, or whether they are used more simply because the users who created them have more of an incentive to find appropriate homes for them, I don't know. I suppose there is the relative obscurity of QSA and CSIRO's images compared to the more commonly referenced subjects that most of us tend to photograph. Either way, it seems like a strong case for the Wikimedia chapters valuing user-generated content. Diliff (talk) 02:10, 3 March 2015 (UTC)\nI think that someone who is going out and photographing content that they know is needed in existing articles is always the most valuable contribution, because those people both upload to Commmons and then use the photo in articles. But we have plenty of people just uploading photos for which we don't yet have articles, just as we have people writing articles for which we don't have photos, but slowly the two do converge. I write a lot of new articles and I am often amazed at how often I find a photo on Commons uploaded many years earlier, so a photo unused today isn't a never-used photo, just a not-yet used photo. And collections like QSA (which I know intimately as I categorised most of it) which are bulk uploaded will also slowly start to get used more as time passes, but of course I would never expect them to be as heavily used for two reasons: being out-of-copyright means they are mostly old low-res black-and-white images - of course we'd prefer high-res colour images where we have them. Secondly, bulk uploads tend to have a particular collection focus and possibly have too much on niche topics. For example, the QSA has hundreds of images of the construction of the Story Bridge. OK, we will never need all of them for Wikipedia articles, but we can and do include the Commons Categories in the article for anyone wanting more images that don't get included in the GLAM tools reporting. As I mostly write historical articles, I do draw on those collections. Can any Commons contributor take a photo for me of a 19th century politican? For historical people and historical events, we do depend on GLAM uploads to a large extent as our only source. Wikipedia's coverage is currently strongly skewed to the present day. We need both kinds of contributions. Kerry Raymond (talk) 02:34, 3 March 2015 (UTC)\n@Diliff: GLAMorous is a super-useful resource. To make the case you are suggesting, you first should tick \"Main namespace only\", whereupon my percentage drops to 70.63%... but more importantly it's worth notionally dividing the usage by the effort+expenses. For me the effort required to upload the entire QSA database was roughly equivalent to obtaining one Featured Picture! or about 20 QIs. Anyway, I don't think we're really arguing - I totally agree that Chapters should continue to value and support user-generated content. But they should also continue to support GLAM content. --99of9 (talk) 06:16, 3 March 2015 (UTC)\n@Ravidreams (WMIN): What do you mean by \"there is no second opinion that movement funds should be used for meeting such needs\"? Are you looking for a second chapter to set up a similar scheme (cf Wikimedia CH have a lending scheme that seems quite productive)? Do you take the absence of a second scheme as an indication that the chapters\/communities disapprove of using movement funds this way? --99of9 (talk) 23:58, 2 March 2015 (UTC)\nI agree with you. Local or on-wiki knowledge plus AGF is more important that 1000 images in 1000 categories (I'm struggling to think who might qualify for that and who I also regard as a great photographer?). And yes, local travel expenses might be one way to help a photographer rather than equipment, particularly if the local expenses are cheap vs imported electronics. Please do away with low-res thresholds. There hasn't been a camera made in the last 10 years that can't do a decent 6MP image. -- Colin (talk) 08:46, 3 March 2015 (UTC)\nSeveral chapters have some kind of Equipment lending scheme \u2212 Wikimedia CH was mentionned, but also Wikimedia \u00d6sterreich, Wikim\u00e9dia France, Wikimedia Sverige or Wikimedia UK (and maybe others). Jean-Fred (talk) 13:19, 3 March 2015 (UTC)\nand also the biggest one, WMDE: Wikipedia:Technikpool, Wikipedia:Festivalsommer\/Technik. See also the specific projets that intensively use lending programs to cover sports events (fr:Projet:Sport\/Photo) or music events (de:Wikipedia:Festivalsommer. Pyb (talk) 14:06, 3 March 2015 (UTC)\n\n# February 23\n\n## Video quality fixed incorrectly\n\nFile:Diamond Trust of London - Kickstarter.webm defaults to a shit quality transcode in the player. There is no setting to change the version being played back in the default viewer. The only way to get the original quality is to open up the file manually in the browser, bypassing TimedMediaHandler. Fairly sure this is a bug. - hahnchen 21:39, 2 March 2015 (UTC)\n\nHmm, it really should be choosing the original file as the source on browsers supporting webm. (The reason it only does crappy transcodes is because the height of the video is less than 314px high, and the smallest webm transcode we do is 360 px high. But in these cases its supposed to use original webm). Problem appears to be in player javascript. It works correctly with js disabled. Bawolff (talk) 18:51, 3 March 2015 (UTC)\nSilly TimedMediaHandler. It detects the video as video\/webm; codecs=\"vorbis, vp8\", but refuses to play it because it thinks it only supports video\/webm; codecs=\"vp8, vorbis\". This is definitely a bug. Bawolff (talk) 19:11, 3 March 2015 (UTC)\nFiled phab:T91431. Bawolff (talk) 19:11, 3 March 2015 (UTC)\nFFmpeg log on the talk page JFTR, but obviously encoding one MP4 bit in four WebM bits did not really help.Be..anyone (talk) 20:15, 3 March 2015 (UTC)\n\n# March 03\n\n## Licencing query\n\nNot sure what to do with this file - File:SoutheastAustralia MapLocator.png. It is good for using for ranges for southeastern Australian organisms but is the licencing a problem...also...does Tasmania look a little big on this? Do we think it is fixable? Casliber (talk) 04:08, 3 March 2015 (UTC)\n\nNo. It looks just right. --Dschwen (talk) 18:40, 3 March 2015 (UTC)\n\n## 25k PD-old photos by Pedro II of Brazil\n\n\u2026where are they in Commons? Here the tantalizing catalogue in Brazil\u2019s National Library, but I could not find a single entry among Common\u2019s photos credited to Pedro II of Brazil (1825-1891), let alone the mentioned 25 thousand. Any ideas? -- Tuv\u00e1lkin 00:24, 2 March 2015 (UTC)\n\n\u2022 Any better link on that Brazil National Library thing? Because that one comes up empty for me. - Jmabel\u00a0! talk 23:34, 2 March 2015 (UTC)\nIt is a quaint weird thing: I didn\u2019t noticed it before, but you need to click on the button with a big bold \">\" on it, next to the search box where there\u2019s already pre-filled the search term \"colecao|d.|teresa|cristina|maria\". The search lists items which are mostly photos or photo collections and links to their descriptions (the book icon), but is shows no photos, not even a tiny thumbnail. It is like they don\u2019t want to have it online at all (maybe because they know it would be impossible to enforce an exclusive copyright?). -- Tuv\u00e1lkin 02:07, 3 March 2015 (UTC)\nIts the online version of a en:library catalog, not an online copy of the content registered in that catalog. --Martin H. (talk) 20:21, 4 March 2015 (UTC)\n\n## Pages with no revisions\n\nThe following pages seem to exist but have no revisions:\n\nInstead of seeing any revisions, I see an error message:\n\nThe revision #0 of the page named \"Commons:Village pump\/Archive\/2015\/03\" does not exist.\n\nThis is usually caused by following an outdated history link to a page that has been deleted. Details can be found in the deletion log.\n\nWhat is this caused by? --Stefan4 (talk) 17:46, 5 March 2015 (UTC)\n\nStrange, must be some broken database, page from first link is at Commons:Deletion requests\/File:Pjy 2014-02-21 14-32.jpg a --Denniss (talk) 18:50, 5 March 2015 (UTC)\nI deleted Commons:Deletion requests\/File:Sltung-FB.png (sometimes deleting & restoring is crating page id in database). But now all revisions seems completely lost\u00a0:\/ --> https:\/\/commons.wikimedia.org\/w\/index.php?title=Special:Undelete&target=Commons%3ADeletion_requests%2FFile%3ASltung-FB.png --Steinsplitter (talk) 19:11, 5 March 2015 (UTC)\nCreated a bugreport: phabricator:T91679 --Steinsplitter (talk) 19:13, 5 March 2015 (UTC)\n\n## Category:James B. Weaver\n\nThis is a strange one: Category:James B. Weaver is an uncategorized and recent category that is a duplicate to the older Category:James Weaver. All files but one can be moved to the other category. But the one picture (File:James Weaver - Brady-Handy.jpg) leads to a warning message and seems to generate an automatic subcategorization of Category:James B. Weaver under Category:James Weaver. Is there any template sorcery involved or am I just to dumb to do it the right way? --Rudolph Buch (talk) 14:28, 11 March 2015 (UTC)\n\nlike that? Mvg, Basvb (talk) 14:41, 11 March 2015 (UTC)\nErrr - yes, exactly like that. When I tried this (and I did try several times) I got strange error messages about transcluded blocked templates and such. But thanks... --Rudolph Buch (talk) 15:14, 11 March 2015 (UTC)\nThis section was archived on a request by: --Rudolph Buch (talk) 15:17, 11 March 2015 (UTC)\n\n## Non-Latin-script page titles\n\nBrowsing around pages like \u6771\u4eac and \u0423\u043b\u0430\u0430\u043d\u0431\u0430\u0430\u0442\u0430\u0440, I take it there must be some kind of policy that city\/country\/similar pages' names be in the language that the place that they are describing uses. This is fair enough, and I understand that pages like Tokyo do redirect, but, if you came upon a page like \u120b\u120a\u1260\u120b by some means other than typing in the English\/Spanish\/... redirect, then you may have no idea what the page is about.\n\nAs a proposal, for logged-in users, at least, could the Commons software: (a) take your user language (say, Spanish), (b) if it isn't the same as the language the page's title is in, check Wikidata whether that wiki (i.e., es.wp) has a name for the Commons page you are on, (c), display the page title for you as something like \"\u6771\u4eac [Tokio]\" or \"\u120b\u120a\u1260\u120b [Lalibela]\", so that you had a better chance of understanding it? Note that not all articles have translation-boxes (e.g. \u6771\u4eac does, but \u120b\u120a\u1260\u120b currently doesn't), so these aren't always availalble for users. It Is Me Here t \/ c 20:34, 3 March 2015 (UTC)\n\nI don't know if there's any policy which requires galleries to be in native script, but some of the users who are speakers of such languages prefer it (especially since categories are currently required to be in Latin alphabet). AnonMoos (talk) 16:40, 6 March 2015 (UTC)\n\n## New York City help sought\n\nLast October there were over 4000 images in Category:New York City. I've been able to assign a more specific location category to the vast bulk of them (usually with a precise location such as a street address, particular building, etc., except for those which are, for example, a general Lower Manhattan skyline). I haven't lived in New York since the 1970s, but in some ways that was an advantage because many of these were historical photos of now-demolished buildings.\n\nSomewhere under 200 images remain in Category:Unidentified locations in New York City and Category:Unidentified locations in Manhattan. Perhaps half of these are hopeless (simply not enough visible to place them) but I suspect that someone who knows the city -- especially the present-day city -- better than I do could pin down another 50-100. - Jmabel\u00a0! talk 17:12, 4 March 2015 (UTC)\n\nNice work Jmabel! I think that for some of the images, such as portrait images or buses without any background, the location is not a relevant property of the image (and they are near impossible to find). I would suggest removing such images from the unidentified location cats. Other files such as File:Verkoopakte Manhattan.jpg are relevant for Manhattan, but don't have a location (It's the sale document of Manhattan from the Dutch). Mvg, Basvb (talk) 22:18, 4 March 2015 (UTC)\nFor the most part, I agree. User:Epicgenius placed that particular Verkoopakte file in that \"unidentified location\" category, not me. As for bus and taxi photos, in general I'm not the one who put them in \"unidentified location\" categories (although in some cases I may have moved files from Category:Unidentified locations in New York City to Category:Unidentified locations in Manhattan because I could tell exactly that much), but it's remarkable what little details of locations are sometimes enough to tell the tale. I got some out of photos like that (e.g. File:Academy MCI D4500 8950.jpg, which even shows one major building that has since been demolished) and I bet there are still a few that can be pinned down by someone else. For example, [:File:Orion hybrid bus in New York city-3.JPG]] and File:Orion hybrid bus in New York city-4.JPG are clearly the same location as each other, and I'd be almost certain they are on the Upper West Side of Manhattan, but I don't know quite where. - Jmabel\u00a0! talk 03:28, 5 March 2015 (UTC)\nThere are definitely many images that I can easily identify, since I live in NYC, so I'll be cleaning out Category:Unidentified locations in New York City and Category:Unidentified locations in Manhattan in the next few weeks or so. Epic Genius (talk) 03:32, 5 March 2015 (UTC)\n@Epicgenius: Wonderful! - Jmabel\u00a0! talk 19:04, 6 March 2015 (UTC)\n\n## Inspire Campaign: Improving diversity, improving content\n\nThis March, we\u2019re organizing an Inspire Campaign to encourage and support new ideas for improving gender diversity on Wikimedia projects. Less than 20% of Wikimedia contributors are women, and many important topics are still missing in our content. We invite all Wikimedians to participate. If you have an idea that could help address this problem, please get involved today! The campaign runs until March 31.\n\nAll proposals are welcome - research projects, technical solutions, community organizing and outreach initiatives, or something completely new! Funding is available from the Wikimedia Foundation for projects that need financial support. Constructive, positive feedback on ideas is appreciated, and collaboration is encouraged - your skills and experience may help bring someone else\u2019s project to life. Join us at the Inspire Campaign and help this project better represent the world\u2019s knowledge!\n\n(Sorry for the English - please translate this message!) MediaWiki message delivery (talk) 20:01, 4 March 2015 (UTC)\n\nThis is a very important and very needed outreach and I wish it full success. I have however a question about a minor point \u2014 one that is stated above and often repeated elsewhere: How can it be said that \u00abless than 20% of Wikimedia contributors are women\u00bb, or any other such exact value, when so many of us chose not to disclose any information about their gender?\u2026 -- Tuv\u00e1lkin 00:22, 6 March 2015 (UTC)\nAll I can see is the Wikimedia Foundation\/UNU-MERIT survey and a couple of papers listed at https:\/\/meta.wikimedia.org\/wiki\/Research:Gender_gap. It looks like it's all based on opt-in data from 2011 or earlier. --ghouston (talk) 08:40, 6 March 2015 (UTC)\nBesides that, even perfect data about the number of contributors wouldn't be the whole story. A wiki with one male contributor with 10,000 edits and one female contributor with 1 edit would still be male-dominated. --ghouston (talk) 09:41, 6 March 2015 (UTC)\nIt is based on a survey they did a while ago. Really there is no way to know exactly what the percentage is for certain. I would pose this though. For any of you who have ever seen the pictures from the meetups or from Wikimania, just scan those photos for women. There sure seems to be an awful lot in those pictures to me. Not as many as men I admit, but still a lot. For example, I just looked at 3 different Wikimania group photos and there are approximately 63 women in each one. Using File:Wikimania 2012 Group Photograph-0001.jpg as an example, it appears that women make up at least 20% of the crowd. Reguyla (talk) 20:14, 6 March 2015 (UTC)\n\n## Some Alaskans here? We need to photograph the Iditarod\n\nHello, I hope to contact some Alaskan editors here. I work on sled dog racing on several wikiprojects (Wikidata, Commons and several Wikipedias) and I need photographs of, well, about anyone, actually. So if some of you are in Anchorage on Saturday, in Fairbanks on Monday or on any of the checkpoints of the 2015 Iditarod, could you please go and photograph them? We have actually photos of several mushers but most of the participants this year doesn't have any free pictures (and when we have free pictures they are several years old most of the time). So any new photo, at all, would be pretty good. We need photos of every musher, even rookies, if possible. There are several categories on Commons, like Category:Mushers and subcategories, but you can just import on Category:2015 Iditarod (it doesn't exist right now but I'll create it once we have some photos) and I'll clean them up. Thank you very much. --Harmonia Amanda (talk) 23:39, 5 March 2015 (UTC)\n\n@Harmonia Amanda: Not that I am a Twitter person at all, but this seems like something where there is probably a relevant hashtag that could be poked at. Revent (talk) 10:27, 6 March 2015 (UTC)\nActually, I also did this announcement on twitter, and on Wikipedia. No one has yet responded to me but I still hope! --Harmonia Amanda (talk) 13:14, 6 March 2015 (UTC)\n\nGetting repeated message An unknown error occurred in storage backend \"local-swift-eqiad\". Will try again... AnonMoos (talk) 16:36, 6 March 2015 (UTC)\n\nSame here, glad it's not just me. mr.choppers\u00a0(talk)-en- 16:38, 6 March 2015 (UTC)\nUploading broken, reported to techs: phabricator:T91761 --Steinsplitter (talk) 16:44, 6 March 2015 (UTC)\nPeople are looking into the issue. Current theory is a recent config change broke things. They're going to try and revert it to see if that fixes it. Bawolff (talk) 16:49, 6 March 2015 (UTC)\nShould be fixed now (Thanks to Reedy). Bawolff (talk) 17:02, 6 March 2015 (UTC)\nWorked for me, thanks!!! mr.choppers\u00a0(talk)-en- 17:19, 6 March 2015 (UTC)\n\n## Proposal for Main Page migration into new translation system\n\nHi! I propose migration of Main Page into new translation system. This will simplify the centralized maintenance of all language versions, it will be easier to add new translations and all versions will be generally synchronous. Current scheme with localized page names will be saved. Any objections? --Kaganer (talk) 23:04, 5 March 2015 (UTC)\n\nOppose. Some mainpages like to add individual notices\/style. And it is also a problem with all the individual protections. And it is creating \/subpages like \/de \/fr but mainpages have a own naming here on commons, like \/de = Hauptseite. Not a good idea imho. --Steinsplitter (talk) 10:22, 6 March 2015 (UTC)\nUser:Kaganer said that the \u00abCurrent scheme with localized page names will be saved\u00bb; I assume that means that Hauptseite will become a redirect of Main_Page\/de, or one will be transcluded in the other. It should be transparent for users who access it, I think. -- Tuv\u00e1lkin 04:45, 7 March 2015 (UTC)\nYes, trancluded, as in Meta. See explanation. --Kaganer (talk) 00:38, 8 March 2015 (UTC)\n\n## Weird behaviour of MediaWiki message delivery\n\nIt just posted some notification at discussion page of my template in my userspace:User_talk:Pbm\/Credits? Why it posted it to template page instead my user page discussion? And how it's targeted - why I'm getting messages in Ukrainian? Pawe\u0142 'pbm' Szubert (talk) 17:08, 7 March 2015 (UTC)\n\nBecause you are listed on Commons talk:Wiki Loves Monuments in Ukraine\/3 years total number of objects pictured by uploader. Sent by @Ahonc:. Looks like a list generator error by the WLM Ukraine team. --Steinsplitter (talk) 17:12, 7 March 2015 (UTC)\nThat list was generated by user:Ilya, I only sent message.--Anatoliy (talk) 17:17, 7 March 2015 (UTC)\nOk, I think I know how I got into the list - I have some photos from Ukraine. And recenly I had User_talk:Pbm\/Credits as template in Author field (it's now fixed and moved to some other field). Thanks for explanation. Pawe\u0142 'pbm' Szubert (talk) 19:07, 7 March 2015 (UTC)\n\nHi! Can someone upload the new best version from same source of this file, please? --Micione (talk) 13:44, 7 March 2015 (UTC)\n\nDone by Shansov.net. Yann (talk) 16:50, 8 March 2015 (UTC)\nThanks, but why the other version? --Micione (talk) 17:19, 8 March 2015 (UTC)\nI tried to create manually a better version with the FireShot plugin, but it doesn't work as advertised. Yann (talk) 17:56, 8 March 2015 (UTC)\nSorry for my English. I wanted to say, why it is uploaded in this file, instead in this? Is from Google Art Project? --Micione (talk) 22:44, 8 March 2015 (UTC)\n\nWhat's the route for doing this these days?\n\nI've been trying to use http:\/\/tools.wmflabs.org\/geograph2commons\/ but had no joy from it today. Tried re-registering TUSC (which has been known to fix it in the past). Now can't login to TUSC.\n\nDoes any of this still work? Has it been replaced by something else?\n\nI'm after http:\/\/www.geograph.org.uk\/photo\/2682446 if anyone fancies a useful import test image!\n\nThanks Andy Dingley (talk) 12:26, 8 March 2015 (UTC)\n\nI have uploaded the file at File:Leawood Pump (geograph 2682446).jpg. I agree the geograph2commons tool does not appear to be working correctly at present. I just got it to work as far generating the file information which I then copied and pasted in to the basic upload form and manually uploaded it that way. There may well be a better or at least more convenient way of doing it that I am not aware of. Rept0n1x (talk) 13:15, 8 March 2015 (UTC)\nThanks Andy Dingley (talk) 00:27, 9 March 2015 (UTC)\n\n## French translator needed\n\nHi friends. There's an ongoing discussion regarding some emblems from France in Commons:Deletion requests\/Files uploaded by Oursmili. @Oursmili: has referred to a discussion in the French Wikipedia that, if I've understood it correctly, supports that this kind of emblems is in the public domain. However, I'm not being able to completely verify it, as I can't speak French at all. It would be helpful if some French speaker with some knowledge of the English language translated (or at least summarized) said discussion. Best regards --Discasto\u00a0talk | contr. | es.wiki analysis 20:40, 8 March 2015 (UTC)\n\n## Wikipedia Takes Manhattan project\n\nWhy do a bunch of categories have in their main page, \"This category has been improved by the Wikipedia Takes Manhattan project\"? I would think that, at most, if any thing like this is tracked it would belong on the talk page. I've improved literally thousands of category pages, but I don't go leaving marks like this on them. - Jmabel\u00a0! talk 15:40, 4 March 2015 (UTC)\n\nPinging @Pharos: the organizer of the project. --El Grafo (talk) 13:57, 5 March 2015 (UTC)\nYes, this project was from 7 years ago (!), so standards were not quite established then, and I agree it would be better to put it on the talk pages now. The reason for the effort is this was a large collective project with dozens of photographers participating.--Pharos (talk) 14:58, 9 March 2015 (UTC)\n\nMyself and a few others have found uploading to Commons to be rather slow, not sure if anyone else is experiencing the slowness but it is painful to just upload one 2 Meg file (in some cases it takes up to 10 minutes). Bidgee (talk) 06:58, 8 March 2015 (UTC)\n\n\u2022 I'm supposed to be getting 40kb\/s uploading, but Commons is not giving me that either. I had thought it was just my connection.\u00a0\u2014\u00a0Crisco 1492 (talk) 16:23, 9 March 2015 (UTC)\n\n## I Origins\n\nI recognised a Commons picture in this film (00:11:50) as I edited the related article once upon a time and I have an aptitude for facial recognition. They had special thanks to everyone at the finishing credits for permissions but not a mention of us even though User:che specifically says \"Please credit as \"Petr Nov\u00e1k, Wikipedia\" in case you use this outside Wikimedia projects.\" Naughty millionaire producers who download and use random pictures of the Internet!\n\nHere is a still image from the film. I guess it is fair usage now.--Abuk SABUK (talk) 01:15, 9 March 2015 (UTC)\n\nCreationists being dishonest? Why am I not surprised\u2026? -- Tuv\u00e1lkin 05:18, 9 March 2015 (UTC)\nWoww, you have a good eye for an eye\u00a0:)!! As this is notable for the image I added the information. See File:Eye iris.jpg#Usage. Sander.v.Ginkel (talk) 14:00, 9 March 2015 (UTC)\nNote that you can also use {{Published}} on a file talk page to indicate the usage of a file. \u2014 SMUconlaw (talk) 14:13, 9 March 2015 (UTC)\n\n## Commons:Photo challenge: Anybody interested in a challenge for analog (film) photography?\n\nLifeguard making sure film won't die while it's enjoying its autumn years at the beach (together with vinyl records and handwritten letters). By Christopher Crouzet, taken with a Hasselblad 500.\n\nDear all, most of you probably have heard of (or already participated in) our monthly Photo challenges. Recently, the idea has come up to do a special challenge that doesn't have a fixed subject picture-content wise, but would be restricted to photos taken with analog equipment. The challenge in a nutshell (see proposal for how it might look like + discussion):\n\nDe-dust whatever old analog photography equipment you can get your hands on, shoot whatever you like (must fit COM:SCOPE of course), digitize the results and enter them in the challenge.\n\nThe photo challenges were always intended to encourage people a) go shooting, b) try something new\/different and c) have fun doing it, and I think this challenge would fit this spirit perfectly. However, analog photography takes a lot of time: You need to finish a roll of film (unless you're shooting polaroid \u2013 which would be fine!), have it developed and digitize it. We would probably account for that by letting the challenge run longer than the usual month, but we are still a bit concerned that this may deter people from participating, so:\n\nWhat do you think? Could you imagine shooting some film in order to participate?\n\nThanks for your input, --El Grafo (talk) 15:07, 9 March 2015 (UTC)\n\n\u2022 Might sound odd, but I quite frankly can't think of a single place where I live that develops film.\u00a0\u2014\u00a0Crisco 1492 (talk) 15:16, 9 March 2015 (UTC)\n\u2022 No I can't imagine shooting some film, but I can imagine dusting off some old photographs and scanning them. --Jarekt (talk) 15:22, 9 March 2015 (UTC)\n\u2022 Most argentic pictures I have which have some educational value are already uploaded here. I don't have an an analog camera anymore, and I won't buy one, even for a contest.\u00a0;oD But otherwise, why not?... Yann (talk) 15:25, 9 March 2015 (UTC)\n\u2022 This would encourage shooting lower quality images than with digital equipment. Not to mention that majority of users already don't have film cameras. It would be surely better to ask for scanning of existing analog images --- [Tycho] talk 15:27, 9 March 2015 (UTC)\n\u2022 Existing analog images taken by the uploaders, naturally (as that is part of what the photo challenge is about).\u00a0\u2014\u00a0Crisco 1492 (talk) 16:22, 9 March 2015 (UTC)\n\n## 2D photo releases of 3D artworks\n\nAre there any good examples we can point to of 2D photo releases of 3D artworks? I.e., the underlying three dimensional artwork remains copyrighted, but the artist has agreed to copyleft a particular 2D photographic view of it. This is for a partnership project with a museum, fwiw.--Pharos (talk) 16:37, 9 March 2015 (UTC)\n\nI can not think of an example of a single photograph of a sculpture being approved for by the sculptor of still copyrighted sculpture, but I do not see an issue with it. I would suggest using {{Art photo}} template for description where you can most clearly separate 2 works (the sculpture and the photograph) and the 2 authors. We would need OTRS letter from the sculptor and possibly the photographer. --Jarekt (talk) 16:50, 9 March 2015 (UTC)\n\n## How do I upload a non-free company logo?\n\nI've searched all the village pump, and archives and help for specific instructions about non-free company logo images, and how (specifically) to upload them. The last time I tried to upload one, it was removed with speedy deletion because I'm sure I just filled out the information wrong. If someone can clearly explain step-by-step like say for example the Starbucks logo got put into Wikimedia Commons, I'd like to upload one just like that (same process). Thanks! Zul32 (talk) 20:26, 9 March 2015 (UTC)\n\nCommons doesn't accept non-free files. You'd need to convert it to a free logo by getting the company to officially release it under a free license. Alternatively you could try to upload it to a Wikipedia project instead using a fair-use rationale. --ghouston (talk) 23:31, 9 March 2015 (UTC)\nIf the Starbucks logo is accepted here, it means either it's not considered to be a non-free logo, or nobody got around to deleting it yet. --ghouston (talk) 23:36, 9 March 2015 (UTC)\nIf you are talking about this file, it's on Wikipedia, not Commons. --ghouston (talk) 00:00, 10 March 2015 (UTC)\nThere is a Category:Starbucks_logos here, but that's a logo I wouldn't upload here, the usual excuses like {{PD-textlogo}}, {{PD-shape}}, etc. aren't applicable for a seriously complex logo. As Ghouston already said, \"fair use\" with an upload on a Wikipedia (not commons) permitting \"fair use\" is a different story. \u2013Be..anyone (talk) 00:29, 10 March 2015 (UTC)\n\n## Christmas crossword\n\nEFF Crossword Puzzle 2014: The Year in Copyright News\n\nThe Electronic Frontier Foundation has published this (copyright related) crossword which you can play online at http:\/\/thedod.github.io\/eff-crossword-2014\/. It makes a nice break from feeding yourself with Christmas treats. Happy holidays everyone. (talk) \u00a0 16:57, 25 December 2014\u200e (UTC)\n\nCan be archived, I guess. --McZusatz (talk) 21:23, 10 March 2015 (UTC)\n\n## Turning an individual djvu-page\n\nHello, I need to have turned this djvu-page [9]]. Can someone do that? --Havang(nl) (talk) 12:51, 6 March 2015 (UTC)\n\n\u2022 Probably you can just make this request on the page with {{rotate}}, but if you don't think anyone will follow through on that for a djvu document, you could bring this to Commons:Graphic Lab\/Illustration workshop. Not exactly an illustration, but I think that's where you'd be most likely to find someone who knows djvu. - Jmabel\u00a0! talk 01:35, 7 March 2015 (UTC)\nMy understanding is that you have to 'decompile' the djvu to individual images, rotate the particular one, and then convert it back to a djvu... major pain. Hopefully someone knows an easier way. Revent (talk) 03:37, 7 March 2015 (UTC)\nThanks for the suggestions. I couldn't put rotate on the page, which should rotate all pages of the File. I finally choosed to download this one page, turn rotate and type the texte on my computer, and publish. --Havang(nl) (talk) 10:27, 10 March 2015 (UTC)\n\n## Fake claim of origin\n\nThe uploader claims that this image comes from a \"Offical Military Newspaper 1923\". It is however taken from the Andrew Mollo's book The Armed Forcds of World War II. Uniforms, insignia and organizations (Crown Publishers, New York) 1981.\n\nCreuzbourg (talk) 21:09, 9 March 2015 (UTC)\n\nSee: Royal Yugoslavian Air Force Rank Chart Creuzbourg (talk) 21:33, 9 March 2015 (UTC)\n\nCreuzbourg, this belongs in a deletion request, then. (And I think you\u2019re wrong, by the way.) -- Tuv\u00e1lkin 06:40, 10 March 2015 (UTC)\nAnd where would I find that? You Wiki-guys are really not weary user friendly, acting more like stereotype DMV bureaucrats. Creuzbourg (talk) 11:43, 10 March 2015 (UTC)\nYou\u2019re as much of a \u201cWiki-guy\u201d as anyone else here (including the bad temper and the creative rudeness, present at least in some of us). You nominate a file for deletion by clicking a link on its filepage that reads \"Nominate for deletion\" (it shows on the left side column on my screen, probably also in yours). That will open a separate discussion page where people can chime in to discuss the deletion request, present arguments, and support or oppose the request; after a week, give or take, an administrator will close the matter and delete, or keep, the file. (I guess a DMV bureaucrat would tell you to look it up\u2026) -- Tuv\u00e1lkin 12:26, 10 March 2015 (UTC)\n\n## SVG- and Commons-related travel grant (endorsements welcome)\n\nExample image drawn by Kevin\n\nHi community,\nit has come to my attention that there is quite an interesting Commons-related grant that's just been submitted as part of the Travel & Participation Grants programme on Meta (jointly organised by the Wikimedia Foundation, Wikimedia Deutschland and WMCH).\n\nIn short, Kelvin Song (@Kelvinsong) \u2014 whom some of you might recognize as a top-class creator of educational (and in, my opinion as a basic SVG creator, absolutely mind-blowing) SVG images \u2014 is asking for a grant to represent the Wikimedia community, and Commons in particular, at the Libre Graphics Meeting, which is due to take place in Toronto, Canada, at the end of April\/beginning of May this year.\n\nI don't usually follow TPS grant or advertise them in such a manner, but as a huge fan of Kelvin's outstanding work, and a SVG creator myself, I think this is an excellent way of at least attempting to bring new skills and people to an under-discovered area of Commons.\n\nThe grant page is on Meta; any help with copy-editing of the text as well as endorsements will be greatly appreciated.\n\nThanks! odder (talk) 22:47, 9 March 2015 (UTC)\n\nYes; Kelvinsong's illustrations are mind-blowing! Thanks for the info. Jee 12:24, 10 March 2015 (UTC)\nAlways positively delighted when I see these wonderful educative and artistic works. Especially because my attempts to draw SVGs hopelessly failed. -- Rillke(q?) 21:05, 10 March 2015 (UTC)\n\n## WMF to file suit against the NSA\n\nHi, I first checked that we were not on April 1st, but no, this is real. So the WMF decided to file suit against the NSA, the United States Department of Justice, and the United States Attorney General. I don't know which practical results that would bring, but bravo to the legal team to take up such a challenge.\n\nRegards, Yann (talk) 13:42, 10 March 2015 (UTC)\n\nReuters, The Independent, PC World. -- Cirt (talk) 14:16, 10 March 2015 (UTC)\n\n## Templates \"by year\" leading categories to show up as uncategorized\n\nHi, there are some categories \"by year\" that are listed at Special:UncategorizedCategories although they are categorized automatically by a template. Examples are\n\nCan this be fixed? Thanks, --Rudolph Buch (talk) 13:23, 10 March 2015 (UTC)\n\nThis has been a problem for a while not just for the categories but even for the files themselves. I think the only way to Fix this, because of the way that report is generated by the software, would be to create a sort of tracking category to account for it. Maybe something like Category:Categories with only template categories and Category:Articles with only template categories for the articles just so they will stop showing on the list. Adding the category to the ones that have any of those templates (or others) would be trivial to do with AWB or through any number of other methods and would greatly reduce the backlog of files in the Files with missing categories categories. Reguyla (talk) 13:59, 10 March 2015 (UTC)\nmoved from COM:AN --Steinsplitter (talk) 14:28, 10 March 2015 (UTC)\nIts because the categories for the page wasn't regenerated when the template changed (They are supposed to be regenerated whenever a template changes, but sometimes the job queue barfs). I've forced the categories for everything the first 545 entries on that list to be re-evaluated. In three days the list should be updated with such entries removed. Bawolff (talk) 19:21, 10 March 2015 (UTC)\nThat\u00b4s great, thank you. There are about a dozen categories with the same problem that I didn\u00b4t list above, please allow me to make you aware of them on your talk page after the next refresh of Special:UncategorizedCategories (but please don\u00b4t feel obliged, it surely is a low priority issue) --Rudolph Buch (talk) 19:39, 10 March 2015 (UTC)\nI'm basically just clicking to https:\/\/commons.wikimedia.org\/w\/api.php?action=purge&generator=querypage&gqppage=Uncategorizedcategories&gqplimit=30&forcelinkupdate=true&gqpoffset=0 , waiting about 30 seconds and then increasing the last number (gqpoffset=0) by 30, and repeating (Stupid rate limiters are making me only do about 30 at a time). So far I'm up to 305. This will force mediawiki to re-evaluate categories of everything on special:uncategorizedcategories (That's the gqppage=Uncategorizedcategories part of the url). Anyone should be able to do this. If you are in the admin group or the bot group then you could probably do all 5000 things on Special:Uncategorizedcategories at once. Bawolff (talk) 20:06, 10 March 2015 (UTC)\nSounds good, I\u00b4ll try this. I guess it\u00b4s going to be easier as soon as the special page has fewer entries which should be the case in a week or two. --Rudolph Buch (talk) 20:14, 10 March 2015 (UTC)\nI did up to 545 as well as the numeric section, but I've stopped for now (sorry, but it was getting tedious). Bawolff (talk) 20:23, 10 March 2015 (UTC)\n\n## Proposal: Hashtag for Twitter images\n\nMany people share images of what they see on Twitter, including images of notable current events (sport events, airplane crashes, etc.). As people want to show to the world what is happening, many of these people wouldn't care (or would be even proud) if these pictures are used by others. However it's not allowed to upload these pictures to Wiki Commons. Because of that I propose that an official hashtag would be created, that if people write this hashtag while posting the original image on Twitter, that it is allowed to upload the image to Wiki Commons. As there are several regularly used licenses Creative Commons licenses I would propose #CC-BY and #CC-BY-SA. Sander.v.Ginkel (talk) 13:37, 9 March 2015 (UTC)\n\nWhile I\u2019m all for social media pictures (on WP we desperately need images of taylor swift on her 1989 tour for example) I would be very surprised if \u201c#CC-BY-SA\u201d ever gains any usage outside the open source fandom. It simply contributes no meaning to your tweets (even though it might carry a lot of meaning commons-wise and legally). Also there\u2019s the issue of people tagging that who don\u2019t actually know what CC is\u2014 23:28, 11 March 2015 (UTC)\n\n# March 10\n\n## Untangling the Web\n\nFascinating publication, if anyone hasn't come across this yet. -- Cirt (talk) 18:46, 10 March 2015 (UTC)\n\n, I didn't know that a copy made it to commons. Related: WMF to file suit against the NSA. \u2013Be..anyone (talk) 20:15, 11 March 2015 (UTC)\n\n## Finding the rigth Berlin S-Bahn station\n\nI took some fascinating pictures of Murals in an Berlin S-Bahn station in 2008. Unfortunatly I dont remember wich station. Same station in pictures (Wall paintings S-Bahn Berlin 2008 2) and (... 2008 3).Smiley.toerist (talk) 00:17, 11 March 2015 (UTC)\n\nThese murals at Bahnhof Berlin Savignyplatz are indeed fascinating. -- Rillke(q?) 01:38, 11 March 2015 (UTC)\nThanks Smiley.toerist (talk) 08:36, 11 March 2015 (UTC)\n\n## We have a winner (25M)\n\nMain courtyard of the Mevlid-i Halil Mosque, \u015eanl\u0131urfa, Turkey\n\nWe reached a nice milestone today! Multichill (talk) 20:29, 11 March 2015 (UTC)\n\nNice! (How was the counting done?) -- Tuv\u00e1lkin 20:39, 11 March 2015 (UTC)\nCounting backwards and selecting the only file that didn't seem to be uploaded by a banned user. Multichill (talk) 21:02, 11 March 2015 (UTC)\nNice, I was so hoping that the 25M image wasn't porn related!. Great job, congrats. Reguyla (talk) 21:16, 11 March 2015 (UTC)\nWhy? Is there a lot of porn in Commons\u2019, however loosely defined? No there isn\u2019t. There are more photos of trams in Commons\u2019 than there are of naked people. You\u2019re watching too much Fox News if you think otherwise. (Not sure how Fox News feels about trams though, but I bet they aren\u2019t too hot on mosques, either\u2026) -- Tuv\u00e1lkin 22:02, 11 March 2015 (UTC)\nLooks like I was the first one to use this file to illustrate an article (on the English Wikipedia). I am sure we can improve this usage.--Ymblanter (talk) 21:44, 11 March 2015 (UTC)\n\n## 25 millionth file\n\nThe 25th millionth file will soon be uploaded, it'd be nice to mark the mile stone, is anyone thinking of doing something\u00a0? I was thinking of maybe doing an article on Wikinews, any other ideas?--KTo288 (talk) 22:17, 7 March 2015 (UTC)\n\nTwenty-five is a nice \u201cround\u201d number (in base 10, anyway), but the not less nice number 24 million went apparently unnoticed (*), and before that the 23-millionth file uploaded merited but a brief mention in this village pump\u2026 The problem is that the 25-millionth file may be an unremarklable item, as the 22-millionth was. -- Tuv\u00e1lkin 23:34, 7 March 2015 (UTC)\n(*) The (first) 24-millionth file was uploaded between November 30th and December 7th, 2014, yet no mention of it in Commons:Village pump\/Archive\/2014\/12. -- Tuv\u00e1lkin 05:02, 8 March 2015 (UTC)\nWell, we can estimate when it will be close to 25M and upload a lot of remarkable files in that minute\/hour\/day --- [Tycho] talk 09:59, 8 March 2015 (UTC)\nThe last time this was discussed, I found it distasteful in the results to see there was deliberate engineering of batch uploads as a form of carpetbagging. I would like to see those with special bot accounts and funded equipment refrain from this temptation and leave it as happen chance.\nLet's not let this become just a \"brand marketing opportunity\". Thanks -- (talk) 12:39, 8 March 2015 (UTC)\nAre you a bot, parent, guardian or dependent of a bot or an employee of a bot. Sadly, you are not eligible for this contest. This promotion void in New York, Oregon and Massachusetts!\u00a0:-)Reguyla (talk) 21:08, 8 March 2015 (UTC)\nThe last milestone to be marked on the community page was the 20 millionth file, last January. That's 5 million in just over a year. The difference between 24 and 25 doesn't seem to be that great, unless you work in base 12. There's always a number of files that the xth file can be, I guess we can try and pick the most \"likely\" candidate.--KTo288 (talk) 07:45, 10 March 2015 (UTC)\n24M was noted on Commons:Milestones. Jean-Fred (talk) 10:49, 10 March 2015 (UTC)\n\nSee #We have a winner (25M). Multichill (talk) 20:32, 11 March 2015 (UTC)\n\n# March 09\n\n## Help correcting image\n\nCould someone upload a new version of the following image. The nitrogen that's floating free isn't supposed to be there, there's supposed to be, where the 'N' is, a line going up to the 'R'. Other than that it's good. Nagelfar (talk) 16:51, 11 March 2015 (UTC)\n\nhttps:\/\/commons.wikimedia.org\/wiki\/File:219a-b.svg\n\nI\u2019m not sure what you\u2019re asking & I not a chemist but I did my best. BTW idk what app generated it but that the file is extremely poorly coded but whatevs\u2014 23:24, 11 March 2015 (UTC)\nThank you, *almost* perfect. I suppose the line on the other side should be closer 'up' toward the \"R\" but otherwise it is exactly what I requested so you have my thanks. (the other guy who made it for me dropped off the map upon finishing and never corrected the error) Nagelfar (talk) 01:16, 12 March 2015 (UTC)\nDone && np!!\u2014 02:03, 13 March 2015 (UTC)\n\n## Upload Wizard can't handle files with the same names but different file extensions?\n\nIt's common to have two files with the same name but different file extensions: File:Example.png and File:Example.jpg. I often upload large PNGs and accompany them with smaller JPGs, but the Upload Wizard doesn't allow this if they have the same file name. This sis a serious hassle, especially when uploading a large number of files. It's not Commons that disallows this\u2014it's Upload Wizard. Can Upload Wizard be altered to allow this? Curly Turkey (talk) 01:11, 12 March 2015 (UTC)\n\nThis is one of the ways phab:T48741 manifests. --Tgr (WMF) (talk) 04:07, 12 March 2015 (UTC)\nSo it's a bug? And one that not much progress has been made on from the looks of it. What a hassle. Curly Turkey (talk) 05:04, 12 March 2015 (UTC)\nYou can always ignore Upload Wizard and upload files using other tools. I, for one, have barely used it, if at all. -- Tuv\u00e1lkin 02:05, 13 March 2015 (UTC)\n\n# March 13\n\n## Upgrade of image rendering servers\n\nHi all,\n\nIts planned to upgrade the image rendering servers to Ubuntu trusty. This will hopefully fix some issues with some images. In particular:\n\nThis is also an important step towards making Opus audio tracks on video work.\n\nThe downside, is some large animated GIF files that were on the edge of rendering previously might stop rendering\n\nThe new image scalars will only be used for uncached renders (That is if nobody has looked at the particular image at that particular size, or if somebody has\u00a0?action=purge 'd the image recently). Additionally, at first (starting Thursday) only one server will be changed to make sure that there is no problems, so you will have a 1 in 9 chance of getting the new server.\n\nAnyways, in the unlikely event that you encounter any image not working, especially if it used to work, please report it here (or at phabricator). For the technically curious, the upgrade is tracked by phab:T84842. Bawolff (talk) 19:43, 10 March 2015 (UTC)\n\nThat's excellent news. Do you think it's feasible to code\/implement upload-through-stream? Like something sending an Opus-encoded stream (with or without container) but doesn't know its final (file) size while it is sending. -- Rillke(q?) 21:02, 10 March 2015 (UTC)\nI'm not sure. In principle I don't see why not (particularly if its not true streaming, but chunked upload that doesn't know the final size until the end). Bawolff (talk) 14:24, 13 March 2015 (UTC)\n\n# March 11\n\n## File:Shotokan Karate Union Logo Rising-sun-enso.gif\n\nImage was uploaded as \"own work\" using the \"CC 4.0 International\" license. Same image is also being used as the official logo of the Shotokan Karate Union. Uploader (Rachael reiko murakami) has stated here that she has no affiliation to the SKU. I opened a thread on here on user's Wikipedia user talk page to try and find out what is what, but I'm not sure what to do in the meantime. Should the file be tagged as COM:CSD#File? Should a license review request be made? Should a request for OTRS permission be made? Should I just wait to see how the uploader responds to my talk page post? - Thanks in advance. - Marchjuly (talk) 01:11, 13 March 2015 (UTC)\n\nRight, DR created: Commons:Deletion requests\/File:Shotokan Karate Union Logo Rising-sun-enso.gif\u200e. Regards, Yann (talk) 08:56, 13 March 2015 (UTC)\n\nThank you for the clarification Rachael. Basically, even if the SKU told you it was OK to use their image, Wikipedia Commons has no way of verifying such a thing. What the SKU needs to do is clearly let Wikipedia Commons know that they [SKU] intend to release the image for use under a free license. They can do this by email. Everything is explained at \"Licensing images: when do I contact OTRS?\" and \"If you are not the copyright holder\". Finally, when you sign your talk page posts please use four tilde (~~~~) and not {{u|Rachael reiko murakami}}. The four tilde not only add your username, but it also adds a time stamp to your posts. The template you used is not for signing posts. - Marchjuly (talk) 02:09, 14 March 2015 (UTC)\n\nThanks Marchjuly I have read your recent comment above several times and i think that i understand it now, and i will endeavour to retain your current input of information for future projects, as i have no need of it on this current project because as i stated above i have deleted it form the current project and i intend on replacing it at my convenience with an image that i personally have complete copyright over. Therefore I am also requesting its immediate deletion form wiki commons. What do i need to put a request in to do so\u00a0? or as you instigated the enquiry in the first instance then will that deletion request deal with it\u00a0?\n\nThis current faux pas of mine just goes to highlight my need for immediate assistance, and I refer you to the request for help that i sent you on my userpage.\n\nRegards Rachael Rachael reiko murakami (talk) 08:57, 14 March 2015 (UTC)\n\n## Template:\u0417\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f\n\nHi, I don't think it is a good idea to use such template. Could someone with a bot replace it by {{Information}}? Thanks, Yann (talk) 13:36, 13 March 2015 (UTC)\n\nI agree. It seems that this template was mostly used by User:UWCTransferBot, used to transfer images from the Ukrainian Wikipedia. I guess this template allows\/allowed to transfer the files easily, without the need to transcribe the Ukrainian information template. Nevertheless we should use the standard {{Information}} template here. --Sebari (talk) 18:13, 13 March 2015 (UTC)\n\n## Detrimental bot move\n\nCan anyone explain to me why File:Indu amerika rebuild-plant.jpg and File:Sarah Anderson Weiss.jpg were recently bot-moved from Category:Unidentified locations to Category:Unidentified countries? Both are clearly in the United States; the former even has a category saying as much. I'm not so much worried about these two files as that if this was a bot move probably the same incorrect move was made on a lot of other files, and I suspect that the activities of this bot in that time period should be investigated. - Jmabel\u00a0! talk 16:21, 13 March 2015 (UTC)\n\nI just noticed the same as a couple dozen files from of my (small) mass uploads were recategorized by ButkoBot away from Category:Unidentified locations arbitrarily to Category:Unidentified countries, and from there by user:Butko into an equaly arbitrary subcat \u2014 often slightly incorrect, some times grossly incorrect.\nI\u2019d say that Category:Unidentified locations is a legitimate categorization and the only unproblematic dissimination would be further into Category:Unidentified locations in Country (and still excluding international waters and off-Earth locations).\n-- Tuv\u00e1lkin 16:44, 13 March 2015 (UTC)\nSome contrasting examples:\nAlthough one wonders why the quirky fixation on countries, this is not wrong, as brick walls are seldom found on international waters or in outer space. Still, Category:Unidentified locations would still be a good categorization and these two moves did not add anything to Commons.\nNow that\u2019s a problem because people (unlike brick walls) are known to move about and it is unclear if the photographed subject is abroad or in her home country (which is known, trusting the description, and was already clearly identified as Category:Women of S\u00e3o Paulo (state), itself a subcat of Category:Women of Brazil).\nIn short, user:Butko should perhaps stop this bot and bring the matter of Category:Unidentified locations to discussion. -- Tuv\u00e1lkin 18:07, 13 March 2015 (UTC)\n\n## SUL finalization update\n\nHi all, please read this page for important information and an update involving SUL finalization, scheduled to take place in one month. Thanks. Keegan (WMF) (talk) 19:46, 13 March 2015 (UTC)\n\n## Someone with ChemDraw help please?\n\nI need a lot of images for the ton of work I've put into my page to go with the tables I've added: https:\/\/en.wikipedia.org\/wiki\/List_of_cocaine_analogues\n\nI just need someone to remake the following images, with suggested names (BME standing for Benzoyl Methyl Ecgonine, the numbers being S. Singh's alphanumeric for the compound, etc.):\n\n[10] BME401a-f\n[11] BMEnoncatalyticHapten394, BMEnoncatalyticHapten395, BMEnoncatalyticHapten396\n[12] 3alphaModifiedBenztropine\n\nI need a ton more, but this would be a start! Thanks Nagelfar (talk) 20:58, 13 March 2015 (UTC)\n\nI guess you need help from Wikipedia:WikiProject Chemistry, Commons:Graphic Lab or Commons:WikiProject Chemistry. -- Rillke(q?) 21:06, 13 March 2015 (UTC)\n\n## File:Frederic_and_Irene_Joliot-Curie.jpg\n\nI think there's good reason to think this is out of copyright, but can anyone figure out the exact rationale behind the Smithsonian's otherwise unelaborated statement that it is? Adam Cuerden (talk) 21:36, 13 March 2015 (UTC)\n\nWhy do you think it is out of copyright? Ruslik (talk) 07:27, 14 March 2015 (UTC)\nI'm guessing it'll be instrument of gift. I'm presuming the Smithsonian isn't just guessing. Which, while it could happen, seems unlikely. Adam Cuerden (talk) 03:33, 15 March 2015 (UTC)\n\n## File:Saturn diagram.svg tutorial on VectorTuts!\n\nIf anyone was wondering how I made my Saturn diagram, I\u2019ve written a tutorial on VectorTuts! \u2014 23:41, 13 March 2015 (UTC)\n\nTerrific image, many thanks! -- Tuv\u00e1lkin 01:48, 14 March 2015 (UTC)\nBrilliant. In your very last step please check if a slightly less dark grey for the irrelevant labels also works, they were hard to read on my laptop in a position not tuned for maximal contrast. \u2013Be..anyone (talk) 04:28, 15 March 2015 (UTC)\nDon't know if you notice that already, but the aurorae disappear when the SVG is opened by Firefox natively. -- Sameboat - \u540c\u821f (talk \u00b7 contri.)\n\n# March 14\n\n## President Obama Delivers Remarks on the 50th Anniversary of the Selma Marches\n\nThere are higher quality versions at YouTube, and also at www.whitehouse.gov.\n\nBut I wasn't sure if we still have a 100 MB upload limit these days.\n\nWould it be possible to upload a higher quality version?\n\nOr should we leave it as is for now?\n\nThank you,\n\n-- Cirt (talk) 04:14, 10 March 2015 (UTC)\n\nAlso, there's an audio file version in MP3 linked at http:\/\/www.whitehouse.gov\/photos-and-video\/video\/2015\/03\/07\/president-obama-delivers-remarks-50th-anniversary-selma-marches specifically at http:\/\/www.whitehouse.gov\/videos\/2015\/March\/030715_SelmaAL.mp3 -- but I was unable to convert it from MP3 to OGG. Can someone else upload that MP3 as an OGG file separately? Thank you, -- Cirt (talk) 04:25, 10 March 2015 (UTC)\n@Cirt: Technically the upload limit was raised to a 1000 MiB, but from my experience, it is rarely possible to upload files bigger than 250 MiB due to issues with server time outs and response times (it might even be 150 MiB if you're unlucky). The easiest way to avoid this limit is to try upload-by-url which you should be able to use, being an administrator. That said, however, the video of this speech at the highest quality provided by the White House is slightly above 1.1 GiB, so any method other than server-side uploads (requested through Phabricator) is unlikely to work. I'm currently downloading the file from the White House (and it's taking extremely long at a speed of around 56 KB\/s), and will convert it to WebM tonight and see what happens then. I can definitely upload an audio Ogg of the speech later today if you can wait\u00a0:-) odder (talk) 05:17, 10 March 2015 (UTC)\nWhat about using a chunked upload? \u2014 Cheers, JackLee talk 05:27, 10 March 2015 (UTC)\n\nUsing the Youtube API, you can find it already has the video transcoded as \"webm 1280x720 video 1648k , 30fps, video only, 272.07MiB\" and audio as \"webm audio only audio 99k , audio@128k (44100Hz), 20.45MiB\". I am having a go at converting it to vp8 rather than vp9 to make it \"Commons compatible\". -- (talk) 11:12, 10 March 2015 (UTC)\n\nHigh resolution version of Obama's Selma Marches speech in webm format. 499MB, taking around 9 hours to transcode on a volunteer's home desktop. Click here for 1,280\u00d7720 px playback.\nThanks for your help, Odder, Jacklee, and ! Yes, Odder, OGG audio of the speech would be awesome, at your convenience! I originally added the lowest-quality version of the file to Commons for the video because it was large, the FLV version at about 70 MB. So really anything more than that would be higher video quality. -- Cirt (talk) 14:14, 10 March 2015 (UTC)\nBTW, as an illustration of how bad things are when having to transcode files due to Commons not taking the most recent open standard codecs, I am now less than 30% done and it has been around 3 hours since I started the transcoding as a background task. To reiterate, I am actually having to transcode an already available webm file that I downloaded from Youtube in about a minute, as Commons cannot play the most current open standards that Youtube makes available.\nI could do a lot more in batch uploading video, but these problems make it an almost pointless and unsatisfying time-sink both for my processing time and volunteer time.\nWe have discussed this at length previously, and the way things are both with how the viewer front-end works and how the uploader back-end is problematic, I do not currently recommend institutions consider Commons as a video file repository. -- (talk) 14:26, 10 March 2015 (UTC)\nTranscoding from mp4 -> VP8 will probably be faster than VP9 -> VP8 (im not sure why i thought that. Encoding vp9 tends to be slow relative to other codecs in my experiance, but i dont imagine decoding for transcode will make much of a difference). I'm hopeful for a future where we support VP9. Bawolff (talk) 21:04, 10 March 2015 (UTC)\nDone New video uploaded, click the thumbnail above to see it. (Warning: It may take an hour or two after upload for Commons to transcode different sizes, this means that immediately after upload, the video may play back in at poor quality, try again later.)\nI got cold feet after about 4 hours of transcoding vp9 \u2192 vp8, thinking that Youtube probably already transcoded from mp4 \u2192 vp9 and \"double transcoding\" was probably a bad thing for size and quality reasons. I started again, and went direct from mp4 \u2192 webm (vp8). Considering this is a 30 minute video, it takes a ridiculously long time to process.\nThe path for doing all this was all free and open source:\n2. I used FFmpeg http:\/\/www.ffmpeg.org and free downloadable open source codecs with a bit of a Python wrapper script for convenience (which I have to hand from other Commons projects) http:\/\/www.python.org to make running the transcode from a command-line a bit easier,\n3. and thanks to Rillke's excellent User:Rillke\/bigChunkedUpload.js, pushing the 499 MB webm file onto Commons is pain free; I saw that the uploader handled a 503 error in the middle of the 120 upload chunks needed, pretty cool.\nIt is less fortunate that the reality is that number of unpaid volunteers that can grapple with how to meet the arbitrary restrictions on Commons to old codecs must be an eye-wateringly small proportion of those that would be potentially interested in contributing video. Even with my background of a couple of years playing with large uploads, I certainly found it hard to understand and slightly mind-numbingly complex to tease out how to set up FFmpeg to run a successful transcode to do this the first time around. It is no wonder that we see so few video related mass uploads on Commons, this project is just ready for them. -- (talk) 04:35, 11 March 2015 (UTC)\nThanks for all of your help, everyone above, this is most appreciated! -- Cirt (talk) 06:37, 11 March 2015 (UTC)\nI'm not so sure how good FFmpeg VP9 actually is (for 2.5, haven't tested 2.6 yet), but I think if you can get the best available MP4 from whitehouse.gov instead of YouTube, and convert it to OGV, where FFmpeg achieved perfection years ago, it should be as good as WebM, only the compression will be worse. Please correct me if that's completely wrong. Or slightly wrong. Or kind of correct missing the point.Be..anyone (talk) 16:11, 11 March 2015 (UTC)\nYou may be missing a key point, Commons cannot play VP9 video, only VP8. As for mpeg vs. ogv, I am not sure that the \"wrapper\" makes much difference if the underpinning codec is identical. Note that to create the video thumbnailed above, the best mp4 available from Youtube was used as the source to transcode to VP8. (Addendum I see that the White House video is larger in file size that the mp4 from Youtube, but I doubt that anyone would find any appreciable difference; especially as the video was taken at long zoom and suffers from noticeable chromatic aberration at full screen size.) -- (talk) 16:28, 11 March 2015 (UTC)\nI know its little consolation for the rather bad video support, but if converting the video on your local computer is problematic, you can always convert it on tools-dev.wmflabs.org (In a detached screen session perhaps, although perhaps tool labs folks would prefer it as a \"job\") and come back to the file when its done. As bonus points the upload would probably be a lot faster from tool labs as its going to have a better internet connection to wmf servers (being a wmf server) then your computer will have. Bawolff (talk) 14:14, 13 March 2015 (UTC)\nGoing on a slight tangent, but I would be happy to try something like this out, so long as chunked uploading were part of the pywikibot modules. Is it available now? Also is ffmpeg available? (it is). I found it a drag to sort out the codecs locally, so I would hesitate to start installing this all on my bit of labs. At the moment I'm parking two videos at a time to a USB stick and transcoding, but it is incredibly slow (right now I have a 500MB and a 800MB mp4 file on the go, I would expect them to take a few hours and these are part of a batch of a couple of hundred I hope to add to Category:Ebolavirus DoD videos). -- (talk) 12:28, 16 March 2015 (UTC)\n\n### Wikipedia video playback problem?\n\nThis probably has a technical cause, so I'm raising it as a sub-thread.\n\nI can playback the above video perfectly well from Commons, but when I launch it from the English Wikipedia article en:Selma_to_Montgomery_marches#Aftermath_and_historical_impact it takes around 5 seconds to start (I am faced with a black box where the video should be during those 5 seconds) and then \"stutters\" with a false start, pausing for 25 more seconds, before payback starts. This is using the standard pop-up video player defaulting to WebM 480P from within Firefox. I suspect that the ordinary public reader might not wait for 5 seconds or the following 25 seconds for the video to start. Is this a problem already identified or something limited to very large videos? Thanks -- (talk) 14:02, 12 March 2015 (UTC)\n\nFor me it doesn't play at all:\nError: cannot call methods on slider prior to initialization; attempted to call method 'value'\nLine 3\n\n-- Rillke(q?) 00:19, 13 March 2015 (UTC)\nFor me, I get non-smooth playback on both commons and wikipedia. One odd thing is it seems to be both loading the webm and ogg 480p transcodes when launched from the pop-up dialog (It should only be loading the webm afaik). Opening https:\/\/upload.wikimedia.org\/wikipedia\/commons\/transcoded\/1\/1a\/President_Obama_Delivers_Remarks_on_the_50th_Anniversary_of_the_Selma_Marches.webm\/President_Obama_Delivers_Remarks_on_the_50th_Anniversary_of_the_Selma_Marches.webm.480p.webm directly also takes a long time before starting (but no stuttering). Perhaps firefox issue? VLC seems to be able to open that url almost immediately. Google chrome also seems to be able to open almost immediately. Bawolff (talk) 14:11, 13 March 2015 (UTC)\nI am unsure what a next step looks like. It strikes me that:\n1. We could benefit if the guidance of COM:Video were to include recommendations for what the practical best sizes are for video, both in terms of file size and resolution. At least a case book of examples might help people have an idea of what the issues are if they expect to include video in articles.\n2. It might help if the various codecs and formats were formally tested by the WMF so that we can make a firm recommendation as to which are \"technically\" the most likely to have good results.\n3. If video is going to remain problematic with various browsers having mixed results, again COM:Video could benefit by explaining the issues and any recommended work-arounds (even if it boils down to \"use a different browser\").\n-- (talk) 14:49, 13 March 2015 (UTC)\nWe have some recommendation over at Help:Converting_video#General_conversion_tips which is linked from com:video.\nAlso, it's 2015 and we should stop figuring out work-arounds for displaying video. If there is a bug in firefox or mediawiki, they should be fixed instead. --McZusatz (talk) 20:55, 13 March 2015 (UTC)\n\n## Help needed with fixing files in Category:Pages using Information template with parsing errors\n\nAs a side effort related to m:File metadata cleanup drive we are tracking now files that do not have any of the standard infobox templates or templates derived from them, in Category:Media missing infobox template. A subset of those are files that seem to have parts of the {{Information}} template: those were placed in Category:Pages using Information template with parsing errors. Many files in that category started with a valid {{Information}} template but some edit to the wikitext broke it, for example this edit 7 year ago, and they just need a minor syntax correction. However since this is the first time we compiled such list there are a lot of files to fix and we could use some help with them. --Jarekt (talk) 18:55, 16 March 2015 (UTC)\n\n## Upload Wizard getting worse and worse\n\nHi! According to Commons stats I am the #3 uploader of Wikimedia Commons of all the times, appearantly the #1 for self-produced images. This is just to introduce my self and let you know that I know what I'm talking about. After WikiLovesMonuments last year I had a problem to my PC and I did not upload anything for some months, until December. Since that time Upload Wizard is getting every month worse and worse, taking so much more time to upload images for errors and disfuncioning. The process of uploading is getting more difficult and frustrating:\n\n1. Preview images do not show anymore, unless I add 3-by-3 images per time (or less). If I can't see the preview I cant' correctly describe the file and put proper categories.\n2. When I select images they do not come listed in alphabetical order anymore. For instance, if I upload 10 images named \"File 01.jpg, File 02.jpg, ... File 10.jpg\", during the passages of the wizard they come out all mixed up (\"File 10.jpg, File 07.jpg, File 02.jpg, File 09.jpg...). This makes much harder to write the correct descriptions and categories. If I have to copy and paste the same category to a serie of files that I named from 1 to 5, they come splitted among all the 50 files I am uploading, making it all extremely complicated and boring.\n3. In the very last week I am getting more trouble, since wizard gets error messages for 20% to 40% of the files I'm trying to upload. So I have to try and try again. Sometimes it also gets blocked for 1 or 2 files at the very last passage (pubblication), I wait and wait, but all I can finally do is just remove the file from the list and upload it manually with the basic form.\n\nI tried to upload from different computers, different Windows systems, and different connections with the same result, I have Adblock disabled, my camera is the same. Also, months ago I requested some easy impovements (like an alert, an extra button, an \"undo\" option for the dangerous \"Copy the title with automathic numeration\" button, which to my opinion should never be automatically checked). Nobody cared. Where are those programmers when you need them? Thanks for your attention. --Sailko (talk) 16:27, 14 March 2015 (UTC)\n\nWorrying. Thanks for highlighting the problem. My personal impression was that there was a healthy amount of interest from the WMF for improving the new user upload experience 2 and 3 years ago, but Commons has since then dropped down the \"food-chain\" or become less of a \"brand priority\" for the WMF. I don't really know how to change that perception, perhaps we should have a formally recognized place (on Commons rather than Phabricator) to collect problems that users experience and discuss how urgent they are for attention?\nI've experienced some of these problems too. I've filed a bug at the Phabricator (T92734). Feel free to add further comments there. \u2014 Cheers, JackLee talk 16:47, 14 March 2015 (UTC)\nPlease report undesired changes like these in Phabricator. They are here, because people are actively trying to rework the extension to something that is measurable, performs better and more reliable. This is VERY complex work, but is required before any improvements to it can be added. As you can see, there has been quite a bit of activity lately. If you experience any regressions due to this, please report them, NOW is the best time to bring them forward. \u2014TheDJ (talkcontribs) 21:16, 14 March 2015 (UTC)\nHi, I never use upload wizard, because I have a bad internet connection, so I can only upload a maximum of 4 images at a time, taking up to 10 minutes. But since yesterday evening it is impossible to upload anything with the basic upload formular. I uploaded 2 images, since then I click \"upload\", connexion starts and 2 minutes later I am back to the upload formular. Traumrune (talk) 21:45, 14 March 2015 (UTC)\nHi, after the basic upload somehow disappeared I may have used the wizard a couple of times - urgh. Switched over to Commonist, left it after some months to use Vicu\u00f1a- very happy with. --Jwh (talk) 15:18, 15 March 2015 (UTC)\nJwh, you mean this \u2014 Special:Upload? -- Tuv\u00e1lkin 14:02, 16 March 2015 (UTC)\nYes I think so, if I remember well there was a time it was more hidden and it was difficult to avoid the wizard. I tried the wizard several times as it allowed to upload multiple files in one transaction, but got often error messages and had to start all over. But that's tempi passati - as I mentioned I'm very happy now with Vicu\u00f1a. --Jwh (talk) 16:05, 16 March 2015 (UTC)\nSo, the Upload Wizard was created because all other tools were too geeky \u2014 it dumbed down uploading so that even the village idiot could use it (and they did!), and now it has problems that need the user to manually add quearies to the url and to file in phab tickets. That makes sense. -- Tuv\u00e1lkin 07:44, 16 March 2015 (UTC)\nYes yes, we all know it was very badly written, and that we are still paying the price for it so many years on, reiterating that isn't going to help in getting anything fixed. Getting those few people who know how to open a web inspector to use the debug flag MIGHT help however. \u2014TheDJ (talkcontribs) 14:28, 16 March 2015 (UTC)\nActually, I had another idea: Nuke it all from orbit, reinstate the previously offered tools, find out who decided it was needed and worked on it and who kept pushing it to be funded and developed instead of useful tools \u2014 and fire, block, office-ban them all, bury the key and and superprotect its grave. And then we can go back to work. -- Tuv\u00e1lkin 13:15, 17 March 2015 (UTC)\nI can confirm issues #1 and #2 from Sailkos post. These problems have been there for months now, especially the missing preview images are a big nuisance. I've had problem #3 a while ago, but the lastest uploads didn't produce such errors. --Magnus (talk) 08:38, 16 March 2015 (UTC)\n\n# March 15\n\n## File:Las mozas del c\u00e1ntaro.jpg\n\nWhat happened here, the picture is bluish on the Commons page and on the French and Spanish WP pages about the painting, but when downloaded the file has perfectly normal colours? Oliv0 (talk) 09:18, 16 March 2015 (UTC)\n\nThe file has an embedded colour profile, probably from the Imacon Flextight Precision scanner used to create the image. Most browsers will just ignore this. I'll apply the profile, if anyone really needs the original, it's still available in the file history. \u2014 Julian H. 10:17, 16 March 2015 (UTC)\n@Julian Herzog: Thank you\u00a0! Could the bad rendering of the original file be due to m:Tech\/News\/2015\/12 \/ Recent changes \"The servers that resize images are using new software. phab:T84842\"? Oliv0 (talk) 16:01, 16 March 2015 (UTC)\nHonestly, I don't know if anything was different before. But the thumbnail generation definitely doesn't do anything wrong, it keeps the colour profile from the original file. So technically, everything is correct, it's just not helpful because browsers don't use the profile. \u2014 Julian H. 16:14, 16 March 2015 (UTC)\nmost (not all. Especially not mobile phone) browsers use colour profiles. I dont think it has anything to do with image render upgrades. Probably either original had wrong colour profile, or there was something weird\/obscure with the profile and it got damaged during the shrinking of the image (ive heard of that happening on files with multiple conflicting colour info). This is speculation though, i havent looked at original file. Bawolff (talk) 23:43, 17 March 2015 (UTC)\n\n## Collapsing in-line text\n\nI know that we have templates to collapse cells inside of tables, but is there any way to collapse text that's in-line with non-hidden text? --Michaeldsuarez (talk) 17:11, 16 March 2015 (UTC)\n\nCan you give an example? Ruslik (talk) 20:00, 16 March 2015 (UTC)\nClick me! Have a look at mw:ResourceLoader\/Default_modules#jquery.makeCollapsible and build a template from it. Note that ID attributes must be, surprisingly, unique per page! Mind transclusion and other Wiki-magic. -- Rillke(q?) 12:14, 17 March 2015 (UTC)\n\n## Attention: new file moving errors\n\ni had 2 cases in the last hours where file moving produced two bad \"pages\" instead of one \"good page\" and a redirect. for further inspection i leave the following pages without a deletion request:\n\nin the move log there are more examples moved by other users, eg. see File:The Soviet Union 1971 CPA 4061 stamp (Order of the October Revolution and Building Construction) cancelled.jpg\n\ncan someone please take care? maybe also a sitenotice so that no further files are moved until the problem is solved? Holger1959 (talk) 03:23, 18 March 2015 (UTC)\n\nThis is phabricator:T93009. --Didym (talk) 03:28, 18 March 2015 (UTC)\nthank you, so manual purging the new page seems to help. good, i know now. Holger1959 (talk) 03:31, 18 March 2015 (UTC)\nNot only the file moves are affected, deletion without manual purging also does not hide files and pages. --Didym (talk) 03:32, 18 March 2015 (UTC)\n\nEven worse: When restoring deleted file, apparently only the description text versions get restored, not the actual file. At least that's what happened with File:H Steiner zug. - Entwurf zum Denkmal Heinrichs vom M\u00f6mpelgard FedZeich.aquar. ca1578 (ZaWH08).jpg. Prior to restoring, the file was still accessible; I still had it in an open browser tab und could re-upload from that. Can anybody else confirm such problems with restoring files? --Rosenzweig \u03c4 18:39, 18 March 2015 (UTC)\n\nIn my watchlist it says, among other things the file is not shown after undeletion. -- Rillke(q?) 18:41, 18 March 2015 (UTC)\nOK, the original file version is now back. Apparently some kind of delay. --Rosenzweig \u03c4 18:44, 18 March 2015 (UTC)\n\n# March 19\n\n## New file renaming criteria in place\n\nCommunity, I am here to notify you that the works around implementing Commons:Requests_for_comment\/File_renaming_criterion_2 are finished; a lot of translations are missing. Although I am not opposing development of policies and guidelines, the volume of work required due to multilingualism and integration into software was enormous and even the new criteria are image-centric. For the future, before starting up RfCs, please make sure there are sufficient resources for putting their results into place. Thank you. -- Rillke(q?) 01:51, 19 March 2015 (UTC)\n\n## Why does Commons host so few Public Library of Science PDFs?\n\nThe Public Library of Science is an open access collection of scientific journals. So far as I know, all of its contents are under Creative Commons attribution licenses that are compatible with being hosted on Wikimedia Commons. Naturally we host hundreds if not thousands of images and videos that were first published in a PLoS journal. I know Wikimedia Commons also hosts PDFs of freely license publications because Wikisource has transcribed some of them. However, we don't seem to have many, if any PDFs of actual PLoS articles and I was curious as to why. It doesn't seem to be lack of interest or awareness because as I mentioned earlier we host hundreds and hundreds of pictures and videos. Why not copies of the PDFs? Abyssal (talk) 20:55, 16 March 2015 (UTC)\n\nHi, Yes, the license allows these files to be hosted here, but what would be the objective to host them in quantity? Regards, Yann (talk) 21:00, 16 March 2015 (UTC)\n@Abyssal: @Yann: This is an excellent idea with lots of applications. There is a pilot of it at en:Wikisource:Wikisource:WikiProject Open Access\/Programmatic import from PubMed Central and related ideas at en:Wikisource:Wikisource:WikiProject Open Access. Wikimedia Commons may or may not be the right place to put PDFs; if the content where put into Wikisource then parts of it could be deconstructed and tagged with metadata, whereas an entire PDF file could not be easily taken apart and remixed. Some people have called for source content on Wikisource to be matched with a PDF upload on Commons but that may not make sense for digitally-born documents. I would be happy to talk this through with anyone. Blue Rasberry (talk) 21:35, 16 March 2015 (UTC)\nI can see the use of these files if they are transcribed on Wikisource, or used on Wikipedia (or Wikibooks, etc.), but I am not sure uploading thousands of them without any prior use in any Wikimedia projects is useful. I am ready to be proved otherwise. Regards, Yann (talk) 09:52, 17 March 2015 media\nYann When Wikipedia cites an open access paper there could be a bot which automatically migrates the paper to Wikisource and uploads all files from the paper to Commons. Getting the papers here means that the works can be more easily remixed, either with reuse of the files, translation, applying wikilinks to technical terms in the papers, packaging the papers in a way that the remix easier with wiki-content, placing papers in the web of citations to and from that paper, and otherwise further integrating them with other works.\nThere still is no leading contender for hosting commentary on all papers published. Some commentary can be objective, like \"this paper was retracted\", and a Wikimedia project could apply that metadata to all paper citations. Along with that open access papers could be heavily marked up on Wikipedia, such that a citation in Wikipedia could lead directly to a particular sentence in the cited paper rather than the entire work. Also someone could say how a paper is cited - like \"this paper confirms the result of that paper\" or \"this paper was only citing the methodology in that one, and not commenting on its results\". It is not inconceivable that Wikipedia, with better integration of open access papers, could pilot a project to set the standard for how metadata is used to remix other publishers CC-licensed works. Blue Rasberry (talk) 14:16, 19 March 2015 (UTC)\nWhile surely most if not all image and video files from PLOS publications are within COM:SCOPE, I question the utility and benefit of hosting complete PDFs on Commons and\/or transcribing them to Wikisource. For all practical purposes, the complete text of PLOS and other online open access journals are already fully digitized and completely machine searchable, unlike say old PD books and journals on archive.org, which while they may have minimal OCR scans, these often contain significant amounts of typos, poor formatting, electronic gibberish, and other impediments to easy online utilization. So rather than asking why Commons doesnt have PLOS PDFs, I'd ask why should Commons host PLOS PDFs?-Animalparty (talk) 04:43, 18 March 2015 (UTC)\nAnimalparty I do not think Commons should unless the text is in WikiSource. If the text is in Wikisource, a copy of the native form of publication (assuming that is an exported PDF) is warranted on Commons to back up the derivative form on Wikisource. The content on PLOS is not wiki-remixable, which it would be if it were exported here. I made some notes about why wiki-integration matters above.\nWe are soon coming to the day (almost certainly within 10 years) when it would be trivial to copy all open access papers to Wikimedia projects, if there would be any use in doing so. If it is useful in 1 of 1000 papers, then it might be easier to migrate all the papers, or it might be useful to migrate any paper as soon as any Wikimedia project cites or references it. Blue Rasberry (talk) 14:16, 19 March 2015 (UTC)\n\n# March 17\n\n## SUL finalization: Commons and usernames that are redirects\n\nHi all,\n\nI started contacting users that are slated to be renamed for single-user login finalization. Unfortunately the way the script was set out it followed redirects from old usernames to new usernames in some cases and warned users that they were to be renamed even if they were not scheduled to be but the old username is. The script was stopped the moment this was first reported and has been fixed. However, there are still a lot of messages in the queue following the old script that need cleared out before the new one starts. There will still be some more users that will be contacted that are not actually going to be renamed. My sincere apologies in advance for the confusion this is causing some affected users, please spread the word. Thanks, and again sorry for the trouble. Keegan (WMF) (talk) 21:35, 17 March 2015 (UTC)\n\nHi, I am one of the people who got this kind of message, which led me to make a suboptimal choice at Special:GlobalRenameRequest, and now I can't change it anymore. It would be nice if someone could stop this renaming process (Mate2code to WatchDuck) for me, or tell me how to do it. mate2code 13:59, 18 March 2015 (UTC)\nDammit. Now I have a new name because this fucked up notification made me choose one, and because this new kind of renaming request is not an edit that can be undone. mate2code 22:08, 19 March 2015 (UTC)\nHi, I also got several such messages. Could you send a message to all these cases please? That would make sure that the issue is fixed. Thanks, Yann (talk) 14:03, 18 March 2015 (UTC)\nReplied at phab:T90820#1130742. -- 23:55, 18 March 2015 (UTC)\n\n# March 18\n\n## What are Commons' weirdest photographs?\n\nCategory:Commons' weirdest photographs\n\nIf anyone remembers the weirdest photographs they have noticed on Commons, please add it to the above subjective category. Let's not fill it with genitals on the first day though. If there is an existing category that does the same job, feel free to move the contents.\n\nAs a project we tend to be literal in categorization, however I believe that some subjective categorization is usefully within our educational scope. Certainly Ripley's Believe It or Not! has never stopped being both popular and educational. -- (talk) 13:26, 19 March 2015 (UTC)\n\nAs a project we tend to be literal in categorization \u2013 says who? This sets a very bad precedent for a mess of other Commons' [\u2026]est [\u2026] categories and does not make any use of sub-\/parentcategorisation, so why don't you just use projectspace for your project? \u00a0\u00a0\u00a0FDMS\u00a0\u00a04\u00a0\u00a0\u00a0 13:48, 19 March 2015 (UTC)\nI often use hidden categories for projects, never a problem and a lot easier for the newer contributor to use than editing a gallery. My views on literalism is based on Commons category debates that have run for several years, excellent examples of taxonomy wars. I have been looking around for a parent category, that's tricky, suggestions appreciated. BTW, I tend to use the term Commons for the project space, the latter gets confused with being some sort of other namespace for wikiprojects. -- (talk) 13:58, 19 March 2015 (UTC)\nI read liberal, although the sentence makes far more sense with literal in it. \u00a0\u00a0\u00a0FDMS\u00a0\u00a04\u00a0\u00a0\u00a0 14:50, 19 March 2015 (UTC\nSounds like a bit of lighthearted fun, thanks. [though can I request we also avoid making it a people freak show]. Added one\u00a0:-) No, it's not a huge spurting penis. -- Colin (talk) 14:25, 19 March 2015 (UTC)\nCategorize it as a {{user category|F\u00e6}} and everything is fine - there are quite a lot of user categories where someone has made his own collection. --Rudolph Buch (talk) 14:38, 19 March 2015 (UTC)\nI would actually prefer this to be a gallery - so I could watch(list) it for new entries\u00a0;-) --El Grafo (talk) 14:43, 19 March 2015 (UTC)\nI like the idea and added some of the photographs I find strange, although \"Commons' weirdest photographs\" seems to be quite subjective. --Jarekt (talk) 15:00, 19 March 2015 (UTC)\nI would also prefer a gallery. - Jmabel\u00a0! talk 15:44, 19 March 2015 (UTC)\n\nA gallery would have two further benefits. The icon size for categories is ridiculously small, making it hard to enjoy the collection. The rather random nature of what people find funny is also likely to lead to a huge list where the nuggets of gold are hard to spot among the stones. What it needs is the ability to\u00a0!like an image. Or perhaps up\/down vote. Then a bot (could there be someone who writes bots anywhere around, by any chance???) could regularly move the most popular weird images to the top. Or have the ability to sort by date or popularity (separate pages?). More interaction would make it more fun. It may also help shift the really not very weird images off the bottom of the page, without anyone complaining they are censoring what they personally find weird. -- Colin (talk) 16:01, 19 March 2015 (UTC)\n\nI have created the gallery Commons' weirdest photographs, where images from the category are sorted by total views on Commons (since upload). This is probably the simplest measurement related to popularity. Obviously the gallery needs the category to generate itself. The top ten are currently:\n\nTo have an image move up the ranking, just start reusing it so that the public view it more often. I have set Faebot to update the gallery once a week. Tip: barnstars, userboxes and odd templates where readers are likely to click on an image to see it better, have a big impact on the number of image views. -- (talk) 12:23, 20 March 2015 (UTC)\n\n## Trains categories upper tree discussion\n\nRunning across an 'upside down scheme' circumstance, some of us have begun discussing how to reorganize the upper categories involving Trains under parent Rail transport. This is a mild long standing issue, first discussed in 2009... Apparently some terms (e.g. 'traincar') which would alleviate organizational grouping translate badly. In the interest of fixing things sensibly for the benefit of non-railfans, and to satisfy as many rail cultural backgrounds as possible\u2014your two-cents are welcome so come put them in!\n\n# March 21\n\n## Russavia related stuff\n\nIllustration of drama at Western College for Women, 1933. Uploaded to Commons today, Russavia claimed to be uninvolved.\n\nI know the whole Russavia topic is an open wound around here and I don't want to seem like I am grave-dancing here because I always liked the guy (I thought his prank on Jimbo was Epic personally) but I noticed a couple things related to him that I think might need some attention. First, there are a number of subpages under his username and since he is permabanned by the WMF and not likely to be returning anytime soon, I think it might be a good idea to browse those, clean them out and remove the subpages if they are no longer needed. Secondly, there are a couple categories directly associated to him. Of the ones I can find, Category:Files needing category checks (Russavia), Category:Files uploaded by Russavia (Eva Rinaldi), Category:Files uploaded by Russavia (cleanup). I was planning on fixing these cats myself but I do not nor will I be likely to get AWB rights here due to my standing on ENWP. I just wanted to mention these things so someone could address these issues. Reguyla (talk) 21:16, 11 March 2015 (UTC)\n\nHow about adding the subcategories to Category:Media needing categories and removing the notes from Russavia that he is still working on them? --ghouston (talk) 21:40, 11 March 2015 (UTC)\nThe subpages are these ones? [14]. I have no idea why they were created, I'd say just leave them there. --ghouston (talk) 21:48, 11 March 2015 (UTC)\nI see no reason to stir things up, or a need to wipe Russavia from Commons. Seek your LOLs on something more productive. -- (talk) 21:57, 11 March 2015 (UTC)\n@:, I appreciate your loyalty to your friend but I am not attempting to seek LOL's on anyone so can you assume some good faith please. I have been the target of severe harrassment on Wiki myself so I know how it feels and I am also not trying to wipe him from commons and if he were merely blocked or banned by an arbcom or some block happy admin I wouldn't even touch them. But Russavia is in fact blocked permanently by the WMF, an extremely rare fate that I have never seen anyone return from. As such, I really do not see any point in keeping to do categories with his name in them. The same is true of his subpages. I'm not trying to be a jerk, it just isn't necessary and is potentially confusing. Reguyla (talk) 01:04, 12 March 2015 (UTC)\nFiles that were uploaded by Russavia will always have been uploaded by Russavia. Nobody needs to be confused by that, and the actions recorded in the public logs should never be changed. -- (talk) 10:04, 12 March 2015 (UTC)\nI agree and that's not what I am saying. What I am saying is that I don't think we need a category that says Category:Files needing category checks (Russavia) or Category:Files uploaded by Russavia (cleanup). Reguyla (talk) 13:26, 12 March 2015 (UTC)\nBoth are populated. When project check\/clean-up categories are empty, then they can be considered for deletion. I suggest you work on meaningfully checking the files rather than fomenting a debate about the name. -- (talk) 13:37, 12 March 2015 (UTC)\nFae, please stop trying to turn this into a hurt feelings report about Russavia. It isn't and if you bothered to read the discussion I started at all rather than just scanning it fir keywords, which you clearly did not, you would see where I specifically said I think it might be a good idea to browse those, clean them out and remove the subpages if they are no longer needed. This applies to the Subpages and to the aforementioned categories. This isn't personal so please stop being so dramatic. Reguyla (talk) 19:12, 12 March 2015 (UTC)\nMeh, if there was never any drama here, it would be a very dull project. Doing the gardening would suddenly seem much more appealing. -- (talk) 19:30, 12 March 2015 (UTC)\nComment: Agree with (talk\u00a0\u00b7 contribs), the categories seem useful and helpful for tracking purposes. They are a valuable organization metric, and should be retained for the future. Thank you, -- Cirt (talk) 20:59, 12 March 2015 (UTC)\nFWIW I sort of agree with Reguyla (although it's not a huge issue now the user is banned) but I think that the best way to deal with this is to categorise all the images that were uploaded but not properly categorised, then deprecate the categories. Out of interest, what is the difference between \"Category:Images uploaded by X\" and the list provided by Special:Uploads? Chase me ladies, I'm the Cavalry (talk) 20:30, 14 March 2015 (UTC)\n@Chase me ladies, I'm the Cavalry: I have always viewed the category to be more of an easy way to view uploads by a user because it not only provides a more condensed version of Special:Upload, but it lists them alphabetically (useful if you are looking for something that begins with a name). While I find it odd that there are essentially two categories that both state that Russavia uploaded those images (but only because one could be renamed, \"Photographs by X\"), that's probably a discussion for another venue, since some of his categories have thousands of images within them and would require some work to rename them. Kevin Rutherford (talk) 23:34, 22 March 2015 (UTC)\nThis section was archived on a request by: Yann (talk) 12:41, 28 March 2015 (UTC)\n\n## Can anyone use free licence for a lower resolution of his work but copyright the higher one?\n\nThe question is about a real scenario that came up during the firsts conversations with an institution that has thousands of photographs. They ask me that because they usually sell the higher resolution ones and restrict the usage for only one publication, etc. They are afraid that once they release the lower resolution photos, that would mean that anyone that come across the higher resolution ones, can act as if they are released as well. Any ideas, thoughts or links to similar questions would be highly apreciated.--Zeroth (talk) 00:11, 20 March 2015 (UTC)\n\nThere was a discussion about this last year. Eventually somebody asked the Creative Commons people, and their opinion was that the high and low resolution versions may represent the same work under copyright law, so that anybody could potentially apply the low resolution license to the high resolution version.[15] --ghouston (talk) 02:54, 20 March 2015 (UTC)\nThis was also more recently raised by me on the OTRS noticeboard, under \"OTRS tickets for thumbnail versions of images\" on 13 Feb 2015. The response was that anyone releasing images on a free licence should understand that if you release a lower resolution image, you have legally released all resolutions. Secondly that OTRS volunteers take no responsibility to advise an uploader\/source donor of this fact, nor is there any expectation on them to do so. -- (talk) 12:47, 20 March 2015 (UTC)\nFae is right that we can't give legal advice. Pointing the institution at the CC FAQ pages would seem a helpful thing to do if they ask about it. If they are considering a large donation of images to Commons, then perhaps WMF Legal could give them specific advice.\nOne ongoing problem is that both CC and WMF heavily promoted for years, in official glossy literature, the idea that institutions and professional photographers\/videographers could release small low-quality copies as CC but retain larger high-quality versions for their paying clients. That was stupid, but don't hold your breath waiting for an apology. If I recall the discussions correctly, many people on Commons were uncomfortable to take advantage of naivety shown by people who followed this advice, or who thought (like many did) that CC applied to the File and not the Work of Copyright. Therefore, there was a strong reluctance to accept\/keep high-resolution files if they appeared to be \"all rights reserved\" and only a low-resolution file had an explicit CC licence. This was on both a ethical and possible-risk basis. -- Colin (talk) 13:44, 20 March 2015 (UTC)\nThanks a lot for your answers.--Zeroth (talk) 23:51, 21 March 2015 (UTC)\nThis section was archived on a request by: Yann (talk) 12:42, 28 March 2015 (UTC)\n\n## Portuguese Wikipedians\n\nIf there is someone who can add a 20 Escudo bank note from 1971 showing Garcia d'Orta, it would be very helpful for me as I am working on. We do not seem to have this one on Commons.\n\nGoogle Translation: (I hope this works) - Se h\u00e1 algu\u00e9m que pode adicionar uma nota de 20 Escudo banco desde 1971 mostrando Garcia d'Orta, seria muito \u00fatil para mim como eu estou trabalhando em. N\u00f3s n\u00e3o parecem ter este em Commons.\n\nShyamal (talk) 05:32, 21 March 2015 (UTC)\n\nThere you go. However, the licensing of all images at Category:Banknotes of Portugal is a mess (including of these two I just added), probably as a result of Portugal not being listed in Commons:Currency. Maybe the whole needs to be deleted as copyright violation \u2014 I await experts\u2019 input. -- Tuv\u00e1lkin 11:15, 21 March 2015 (UTC)\nInfo Commons:Deletion requests\/Files in Category:Banknotes of Portugal. Gunnex (talk) 11:55, 21 March 2015 (UTC)\nWhat a pity! I thought the rest of the images were standing on firm ground after it was replaced by the Euro. Shyamal (talk) 17:37, 21 March 2015 (UTC)\nThis section was archived on a request by: Yann (talk) 12:43, 28 March 2015 (UTC)\n\n## Brianboru100\n\nI've had a whole load of files marked for possible deletion by someone calle mattbuck and before I've had time to comment, the deletion debate has been closed. No reason for the possible deletion has been given. No indication of whether the possibility has been translated into a definite has been given either. I have an application that depends on the urls of these images. Not very friendly to a novice user who doesn't login every day and doesn't know how to follow any policy debates that are going on - just wants to share images.\n\n\u2014\u00a0Preceding unsigned comment added by Brianboru100 (talk\u00a0\u2022\u00a0contribs)\n@Brianboru100: Hi,\nIt seems that the deletion request is here: Commons:Deletion requests\/Files in Category:Mattbuck's temporary category. The last part is not closed yet, so you can answer there. Regards, Yann (talk) 20:26, 22 March 2015 (UTC)\n@Brianboru100: The reason for the deletion is that these are photographs of 2D graphic works (such as murals) in the UK. Various countries have rules which we refer to as \"freedom of panorama\", which generally means that if an artwork is permanently in a public space then it cannot be copyrighted. Unfortunately the UK version does not apply to 2-dimensional \"graphic works\" such as murals. This means that whoever created the mural holds copyright, and so we cannot accept images which show it. It seems a bit silly, but legally it's no different than taking a photo of a photo in a magazine. You have made a mistake that many many other people have made over the years, myself included. Copyright law is a web of confusion, mostly counterintuitive, and freedom of panorama laws vary wildly from country to country. -mattbuck (Talk) 23:34, 22 March 2015 (UTC)\n\nSo why couldn't this explanation be included in the first place? Still not a friendly process. When are the deletions due? I need time to get the images mmoved and my app to point to the new urls.\n\n# March 23\n\n## Cat-a-lot\n\nIn Category:Ancient Roman bronzes in the Museo archeologico nazionale (Florence) I created the sub-cat of bronze statuettes. Using Cat-a-lot I could move only 15 files; all others are rejected, unrecognized (\"they were skipped because the old category could not be found\"). Momentary disservice as it happens every now and then? Apparently not. After four days still it do not work. I should move them one by one. I wonder and ask: what have these files differently than other that prevents the action? These files are uploaded by the same user, and opening them in edit I do not see anything strange or different between the two groups. It's possible that they have some hidden element or sign that blocks the passage? Thanks for your answer, or the solution of the problem. Best regards, --Denghi\u00f9Comm (talk) 09:08, 22 March 2015 (UTC)\n\nI don't know why Cat-a-lot doesn't work, but alternatively you may try it with VisualFileChange: go to Category:Ancient Roman bronzes in the Museo archeologico nazionale (Florence), then left-hand side toolbar \"Perform batch task\", choose \"custom replace\", select files you want to move, and replace [[Category:Ancient Roman bronzes in the Museo archeologico nazionale (Florence)]] by [[Category:Ancient Roman bronze statuettes in the Museo archeologico nazionale (Florence)]] in the source text. --A.Savin 10:14, 22 March 2015 (UTC)\n\nNow it has unblocket\u00a0! Now it works, it has succeeded. Wonderful\u00a0! Thank you so much at all for your advices and for the solution of the problem\u00a0! Best regards, --Denghi\u00f9Comm (talk) 13:29, 22 March 2015 (UTC)\n\nJust in case you are wondering whether there is or not ... paste the text in question into that tool. -- Rillke(q?) 21:32, 23 March 2015 (UTC)\n\n## Renaming files\n\nCan I ask to rename file:1971. V \u043b\u0435\u0442\u043d\u044f\u044f \u0441\u043f\u0430\u0440\u0442\u0430\u043a\u0438\u0430\u0434\u0430 \u043d\u0430\u0440\u043e\u0434\u043e\u0432 \u0421\u0421\u0421\u0420. \u0411\u043e\u0440\u044c\u0431\u0430.jpg to the name \"The Soviet Union 1971 CPA 4016 stamp (Greco-Roman wrestling) cancelled.jpg\" for the following reason\u02d0 \"To harmonize the file names of a set of images (so that only one part of all names differs)\"? --Matsievsky (talk) 10:48, 23 March 2015 (UTC)\n\nParts that form a whole is explained as \"scans from the same book or large images that are divided into smaller portions due to Commons' upload size restriction\" (emphasis mine); this is not the case with these stamps, they are not pages from a book, or parts of an image that fit together to show a single image. They are single, complete, discrete, entities and therefore not eligible for harmonisation (each can be understood by themselves and do not need others to be understood). You may disagree with my reasons to decline, but without a valid reason to rename the default position is to decline. ColonialGrid (talk) 14:13, 23 March 2015 (UTC)\n\n(Edit conflict)\n\nMatsievsky, \u00a74 of COM:FR aims for generic filename schemas called up by templates. The mere contrastative use of these two filenames\n,\u00a0while helpful for human reading, doesn\u2019t qualify. Lacking support for \u00a74 what\u2019s left is a request to rename a file to a new name that is, as ColonialGrid correctly analysed,\n\u2022 just \u00aba bit better\u00bb (\u00a71), as it includes the CPA number (this should be in the description, anyway, and maybe in categories, not necessarily in the filename \u2014 and not sufficiently, too!). It is also a bit worse, as it lacks mention of the 5th Summer Spartakiad, and spells out the country name too verbosely (\u00abThe Soviet Union\u00a0\u00bb\u00ab\u00a0stamp\u00bb, seriously, as opposed to trimmer and clearer \u00abSoviet\u00a0\u00bb\u00ab\u00a0stamp\u00bb?)\u2026\n\u2022 a replacement of Russian with English \u2014 clearly against that clause \u00a72. Please note that even if this Soviet stamp image was filenamed in, say, Bengali, for any reason or no reason at all, changing it to Russian would still be against policy. In thsi case, moving it from the official language of the country the scanned object is an official document of \u2014 that\u2019s twice a bad idea, regardless of the prestige Shakespeare\u2019s Tongue may have in Mother Russia (I always observe with mirth how said prestige tends to be locally inversely proportional to actual command of English).\nIn terms of formal logic, one of the criteria being fulfilled would be enough for renaming, as they are disjunctive, but no criteria being fulfilled means no renaming per policy. -- Tuv\u00e1lkin 14:31, 23 March 2015 (UTC)\nTuv\u00e1lkin, my purpose - not transition from Russian into English, it isn't necessary to impose me your desires. The bad knowledge of language logically doesn't attract a mistake in reasonings, and the good knowledge of language logically doesn't attract permissiveness. I transfer Soviet stamp image names to the certain standard reached as the result of the compromise: \"The Soviet Union (Year) CPA (Number) stamp (Short description)\". --Matsievsky (talk) 16:53, 23 March 2015 (UTC)\nUse your standard for your own uploads, that\u2019s cool, but do not try to impose it on other files \u2014 COM:FR forbids it very clearly. That\u2019s all there is to the matter, really. -- Tuv\u00e1lkin 17:35, 23 March 2015 (UTC)\nThank's, it is news to me. Please, specify the concrete quote, I didn't find your information. --Matsievsky (talk) 19:15, 23 March 2015 (UTC)\nI see. By the way the both files are versions of the same stamp, therefore, it is two parts of a single whole - a stamp in its versions. \u00a74 perfectly works. --Matsievsky (talk) 09:34, 24 March 2015 (UTC)\nNo, you are wrong, they are not two parts of the one object, they are two versions of the same style of object, there is a difference. Think about is like pages in a street directory, each page is part of a single, whole map; individually they are without greater context, and should therefore be named in harmony. Please read the clause again in its entirety: \"Second, files that form parts of a whole (such as scans from the same book or large images that are divided into smaller portions due to Commons' upload size restriction) should follow the same naming convention so that they appear together, in order, in categories and lists.\" This explicitly states that to be parts of a whole they must be smaller portions that fit together to make a larger image (as a jigsaw puzzle does). The case you have provided has no valid rational for renaming under the current rename guidelines at Commons:File renaming, in fact, they fall under the first sentence of clause four: \"Just because images share a category or a subject does not mean that they are part of a set.\" They share a subject, but do not form a set under our definitions of being either pages in a book, or parts that combine to show a larger image. ColonialGrid (talk) 13:37, 24 March 2015 (UTC)\nStamps can also be considered as part of the catalog of stamps, as its increased images. --Matsievsky (talk) 16:51, 24 March 2015 (UTC)\nSure, they could, but in this context they aren't; you simply have to accept that your propsed rename isn't supported by policy. ColonialGrid (talk) 17:01, 24 March 2015 (UTC)\nIf it and so (what I doubt), change policy. --Matsievsky (talk) 22:59, 24 March 2015 (UTC)\n\n# March 24\n\n## New user edit-warring about a digitized audio clip from 1917\n\nCould someone help out at File:Tamo Daleko.ogg (File_talk:Tamo Daleko.ogg)? I was unsuccessful in trying to explain why it couldn't be attributed to him only, just because he digitized the song and put it through some filtering and general audio cleanup. - Anonimski (talk) 00:26, 24 March 2015 (UTC)\n\n@Anonimski: I left a long comment talking about the copyright issues. I also used one of my favorite underused templates, {{infosplit}}, to make it clear there are (at least potentially) two different copyright statuses involved. BTW, as I mentioned, pre-1972 sound recordings should use {{PD-US-record}}, not {{PD-US}}. Revent (talk) 01:51, 25 March 2015 (UTC)\n\nEvery time I try to use the \"Share images from Flickr\" option on the Upload Wizard, I get the message \"Unknown error: 'permissiondenied'.\" This would be a great option in lieu of uploading manually, but it simply does not work. Conifer (talk) 01:49, 24 March 2015 (UTC)\n\nThat option should only be visible to administrators and license reviewers for now. It looks like someone broke something again. LX (talk, contribs) 07:22, 24 March 2015 (UTC)\nComment There are other tools (Commons:Flickr files#Tools) that can upload from Flickr, such as the reasonably straightforward Flickr-2-Commons tool. ColonialGrid (talk) 15:53, 24 March 2015 (UTC)\nI used to have license reviewer permissions on my old account, User:David1217. However, when I got renamed on en.wp [16], I neglected to have the same rename here on Commons (I believe this was before global rename). Then when I created an SUL, it automatically made me this new account here, so now I have split accounts. Apparently I've had the right removed due to inactivity, but if there's some way to restore it on this account, I'd be very pleased, because it would make uploading much easier. Thanks, Conifer (talk) 19:26, 24 March 2015 (UTC)\nIt is more than 2 years since you last used the right and it was removed from your old account on 24 February, after notification, for this reason. You should raise a request formally on Commons:Requests_for_rights#Filemover. -- (talk) 19:47, 24 March 2015 (UTC)\nCommons:License review\/requests is probably the place F\u00e6 wanted to send you to. -- Rillke(q?) 20:23, 24 March 2015 (UTC)\n@Conifer: The license reviewer userright is irrelevant to uploading, as you are not allowed to review your own uploads (it does not make you 'exempt' from having them flagged for review). Revent (talk) 21:18, 24 March 2015 (UTC)\nI know that they still have to be reviewed; I meant that I could use the direct Flickr URL uploading instead of the manual method. Anyway, thanks to everyone for the help, since I'm now using Flickr2Commons, which works just as well. Conifer (talk) 21:25, 24 March 2015 (UTC)\n@Revent: We grant image reviewers the upload_by_url user right. With these rights, it is possible to transfer files from Flickr using Upload Wizard and issue uploads server side from several other sources. -- Rillke(q?) 23:07, 24 March 2015 (UTC)\nI was aware, but that slipped my mind when it came to the context of requesting the (group, not 'right', indeed), since it's not really the main 'purpose'. I was thinking he meant 'easier to upload from Flickr' in that sense, tho... my mistake. Revent (talk) 23:25, 24 March 2015 (UTC)\n\nHi all, is there a way to order tasks to Panoramio Upload Bot? I asked its operator and didn't receive any answer... Thanks --Discasto\u00a0talk | contr. | es.wiki analysis 23:22, 24 March 2015 (UTC)\n\nThis section was archived on a request by: \u00a0\u00a0\u00a0FDMS\u00a0\u00a04\u00a0\u00a0\u00a0 14:18, 31 March 2015 (UTC)\n\n## Request for help regarding an oversight issue & the Wikimedia Foundation\n\nThis section was archived on a request by: \u00a0\u00a0\u00a0FDMS\u00a0\u00a04\u00a0\u00a0\u00a0 14:18, 31 March 2015 (UTC)\n\n## Commons talk:We miss you#Should the people doing the missing be listed for each entry?\n\nI'll looking for people who are interested in having a discussion about the format of the Commons:We miss you page. If anyone is interested, please join in. --Michaeldsuarez (talk) 16:06, 19 March 2015 (UTC)\n\nI'm personally wondering about the inclusion criteria, because I saw someone add Penyulap the other day, and I don't think anyone missed them. -mattbuck (Talk) 12:29, 20 March 2015 (UTC)\nThat someone was me, and another user endorsed the entry via \"thanks\". I think that we should work out what the page should display before we work on the inclusion criteria for new entries. Abd suggested that entries should be seconded. I like that idea, but that's a discussion for another time. --Michaeldsuarez (talk) 14:18, 20 March 2015 (UTC)\nThe page is untenable if it is necessary to establish consensus for inclusion. Hence \"we miss you\" could simply be interpreted to mean that more than one user misses the person, thus \"we,\" and the first one as shown by an addition to the page, and the second, or more, as shown by listing additional users. If one person nominates, anyone may revert that, but then if another brings it back in, it should be accepted. Unless socking is shown, of course! The same person might have been seen as a PITA by nearly everyone else, might be blocked, banned, or excommunicated. Michael is correct that inclusion standards should be established, so that disruption is not caused by dispute over who is missed and who is not. The page should not become a debate. We can see a hint of this above, where clearly one person misses, and it was asserted as unlikely that anyone would miss that user. Obviously false, already known as such if anyone is paying attention. I don't know Pennylap from a HoleInTheWall, and don't need to. Let's keep it simple. --Abd (talk) 19:26, 25 March 2015 (UTC)\n\n# March 20\n\n## File:Cholula_Hot_Sauce.jpg\n\nAre personal photos taken of commercial products located on shelves in grocery stores OK to upload to Commons? I checked the archives and find quite a few threads about Coke bottles\/cans which say that the images are OK because the logo is no longer copyrighted and because of the simple design of the bottle\/can. COM:PACKAGING, however, seems to say that if the packaging of the product contains a printed design then it cannot be uploaded to Commons. Would that reasoning be applicable to this picture of these hot sauce bottles? Thanks in advance. - Marchjuly (talk) 02:25, 25 March 2015 (UTC)\n\n\u2022 Unless the picture of the woman is in the public domain (e.g. it's very old), File:Cholula_Hot_Sauce.jpg is derivative work of a copyrighted work, and we should not have this image on Commons. - Jmabel\u00a0! talk 04:55, 25 March 2015 (UTC)\nThanks Jmabel. What do you think is the best course of action. Tag it for speedy deletion per \"Apparent copyright violation\" or nominate it for deletion per COM:DR? -Marchjuly (talk) 05:53, 25 March 2015 (UTC)\nEither. No big deal which. - Jmabel\u00a0! talk 16:14, 25 March 2015 (UTC)\nHere's another photo which has been uploaded but does not seem to satisfy the criteria of COM:PACKAGING - Marchjuly (talk) 04:46, 25 March 2015 (UTC)\n\u2022 The Sriracha one is almost entirely text and simple shapes. The two tiny peppers on that are probably a small enough portion of the image to be de minimis. - Jmabel\u00a0! talk 04:57, 25 March 2015 (UTC)\nUnderstood. This image then does satisfy the exceptions of \"COM:PACKAGING\". Thanks for clarifying. - Marchjuly (talk) 05:54, 25 March 2015 (UTC)\n\n## Category:Files from the San Diego Museum of Art to be checked\n\nHi, Experts in Japanese and Chinese art needed.\u00a0;o) Thanks for your help, Yann (talk) 13:08, 25 March 2015 (UTC)\n\n## Ingrown trees\n\nI encountered this unusual tree + pole combination in Budapest. (also File:Ingrown tree close to Budafok kocsiszin tram depot (2).JPG and File:Ingrown tree close to Budafok kocsiszin tram depot (3).JPG. Can the tree species be determined?Smiley.toerist (talk) 10:10, 27 March 2015 (UTC)\n\nNot from that image; we'd need to see leaves and, ideally, flowers and\/or fruit. Then try Category:Unidentified trees in Hungary. Andy Mabbett (talk) 10:24, 27 March 2015 (UTC)\nOn reflection, the green bark suggests a dogwood, Cornus species. Andy Mabbett (talk) 10:26, 27 March 2015 (UTC)\nNo need for leaves etc. (though that would certainly make it easier). If you can make some good close-ups of the buds, someone with access to the right literature could probably identify it quite easily. From what I can see, it looks like it might have differently shaped buds for flowers and sprouts, which would fit the Cornus theory. --El Grafo (talk) 10:45, 27 March 2015 (UTC)\n\n## Type of metal?\n\nIs my guess that this is a cupronickel coin correct?Smiley.toerist (talk) 23:58, 25 March 2015 (UTC)\n\n@Smiley.toerist: https:\/\/en.wikipedia.org\/wiki\/Coins_of_Madagascar says it's stainless steel. If so, it's led a very hard life to get that banged up. Revent (talk) 05:22, 27 March 2015 (UTC)\nThere are very few coins in circulation so the few in use get a lot of use even with such a marginal value because people are very poor. As a tourist you get spend millions of Arials on consumptions, fees and souvenirs, but practicaly nothing in hard currency terms. This is in lots of banknotes wich everybody folds in bundels of ten. Only on the last days did I see any coins.Smiley.toerist (talk) 08:48, 27 March 2015 (UTC)\nAre there any other Septagonal coins? I had to create a new category.Smiley.toerist (talk) 08:53, 27 March 2015 (UTC)\n@Smiley.toerist: en:Fifty pence (British coin). Andy Mabbett (talk) 13:14, 27 March 2015 (UTC)\nOK:But these UK cannot be uploaded on the Commons.Smiley.toerist (talk) 08:08, 28 March 2015 (UTC)\n\n# March 27\n\n## Commons talk:We miss you#Should the people doing the missing be listed for each entry.3F 2\n\nWe have more or less finished discussing the structure and format of the page and have now moved onto discussing principles regarding how new entries will be handled, including their removal. We haven't created a solid proposal yet. We're still throwing around ideas and discussing them, and I like to hear from others. --Michaeldsuarez (talk) 14:31, 27 March 2015 (UTC)\n\n## Review request [derivative work?]\n\nHi. I'd like to request if you could assist me in determining if the following images are considered as derivative works?\n\nIf yes, i can then proceed with the deletion. Thanks. Ali Fazal (talk) 23:16, 27 March 2015 (UTC)\n\nMost of these are fine to be on Commons per {{PD-text}} and COM:FOP. Josve05a (talk) 23:33, 27 March 2015 (UTC)\n\n# March 28\n\n## Creating a free and open source typeface\n\nHi, I know this isn\u2019t the typical business of Commons, but I am currently designing a typeface and would like to release it as a free and open source font with help from the Commons community. I\u2019ve written a rationale for this font project that you can read at File:A proposed free and open source typeface.pdf (use pdf reader; pdf may not display with Firefox pdf.js). Typefaces are resources much like the images and video that Commons currently produces, and expanding into font creation I believe is a logical expansion of its mission. Please take a moment to take a look at my proposal, and perhaps test my font which lives on GitHub!\n\nIt is critical that we do not ignore the importance of type in the development of libre ecosystems. Typography has always been a stubborn holdout in this regard, and to this day there remain few free high-quality comprehensive text typefaces. Free type is mainly concentrated in a handful of flagship \u201csuperfonts\u201d that contain a staggering catalog of glyphs, but lack greatly in the quality of design and typographic styles and features seen in professional type. To my knowledge, there are currently just two great open source text families\u2014Gentium, which is still incomplete, and Linux Libertine, in addition to a few corporate gifts such as Adobe Source Serif and Bitstream Charter. To help fill the gap, I present my own original type design and ask for the Wikimedia projects\u2019 help in finishing and releasing my font to provide a quality free font choice\u2026\n\n15:59, 15 March 2015 (UTC)\n\nThis font looks really lovely texts are convenient to read in it - thus I hope I can read Wikipedia articles in that font one day - including mathematical formulae of course. ${\\displaystyle {\\frac {-4\\pm {\\sqrt {6^{2}-4\\times ac}}}{2a}}}$ -- Rillke(q?) 23:39, 15 March 2015 (UTC)\nI love it, except for the crossbar in the capital A, which is very distracting to me. \u2014TheDJ (talkcontribs) 14:23, 16 March 2015 (UTC)\nI\u2019m not sure exactly what you mean. It's right where crossbars on A\u2019s usually go. Is it too high? too low?\u2014 02:47, 17 March 2015 (UTC)\nThe crossbar in the A also looks messed up to me (using MacOS X). Kaldari (talk) 06:25, 17 March 2015 (UTC)\nI checked the PDF in Windows and iOS and fail to see any glitch of the capital A. Maybe a screen cap from MacOS would explain the issue better. -- Sameboat - \u540c\u821f (talk \u00b7 contri.) 07:43, 17 March 2015 (UTC)\nYou can see my problem in this screenshot. \u2014TheDJ (talkcontribs) 11:23, 17 March 2015 (UTC)\nI noticed that in the small f: phab:F97280 - installed the otf font files under Windows. -- Rillke(q?) 11:59, 17 March 2015 (UTC)\nThat is very strange, I have never seen it do that! It usually happens when there is a contradicting intersection, but shouldn\u2019t be happening there considering both contours are clockwise\u2014 22:32, 17 March 2015 (UTC)\n@Kaldari Rillke & User:TheDJ, I\u2019ve fused the A crossbar and the f crossbar & pushed updated font files to github. Pls download & check to see if the problem is still there on ur computers\u2014 22:55, 17 March 2015 (UTC)\nYeah looks like expected now. Although not that eye-catching the \"t\" glyph is also affected; interestingly only with smaller font sizes: phab:F99875 -- Rillke(q?) 13:36, 18 March 2015 (UTC)\n@Rillke Fixed 22:14, 18 March 2015 (UTC)\nI'm no font expert, but your font looks elegant and classy. -- Sameboat - \u540c\u821f (talk \u00b7 contri.) 02:14, 17 March 2015 (UTC)\nThanks sm!!\u2014 02:47, 17 March 2015 (UTC)\n@Kelvinsong: Sounds like a great idea. I'm going to let the WMF designers know about it in case they want to contribute. One thing to keep in mind: The SIL Open Font License (which is one of the most popular free font licenses) covers use and distribution of the font as a whole, not the individual glyphs. If you want to make sure that your font is completely free (both the software and the design elements), I would suggest using a CC0 or CC-BY license (or dual-licensing with both CC and SIL licenses). Kaldari (talk) 06:24, 17 March 2015 (UTC)\n@User:Kaldari idk I was going to use GPL font license to avoid the whole Charter parallel design mess\u2014 22:48, 17 March 2015 (UTC)\n@Kelvinsong: GPL+FE works too. Don't be afraid to multi-license though\u00a0:) Kaldari (talk) 23:08, 17 March 2015 (UTC)\nThis is a very nice font and well-designed. I hope I'll see this on Wikimedia projects at some point, and maybe even elsewhere on the web. Definitely my favorite custom serif font for paragraph texts. --GeorgeBarnick (talk) 07:00, 17 March 2015 (UTC)\nVery interesting and good-looking to my uneducated eye, but I guess that's up to the real font experts to judge (would it be possible to get feedback from a professional?). Just out of curiosity: What's wrong with Computer Modern\/BlueSky\/Latin Modern? --El Grafo (talk) 08:32, 17 March 2015 (UTC)\n\"It is critical that we do not ignore the importance of type in the development of libre ecosystems.\" YES! We discussed a lot about this topic and I'm personally very happy to see that you have stepped in so decidedly. Besides, I have seen several of your works without knowing that they came from the same designer. Congratulations for your skills, and thank you very much for your contributions.--Qgil-WMF (talk) 18:47, 17 March 2015 (UTC)\n@User:Kaldari & @User:GeorgeBarnick thank you sm for saying that!! \u263a\ufe0f\n@User:El Grafo I\u2019m not sure yet. All the major type designers communities went into strange decline in the past year but I\u2019ll send some samples out. && I don\u2019t want to get into a rant but Computer\/Latin Modern is jsut a dreadfully designed font. It was not even created by a human; it was made by a computer with only a rudimentary sense of curve aesthetics. The italics are half-decent but unstandard & so hard to read for long stretches. It is a decorative font at best, and is very illegible for body text. If you want a didone font; use Didot or New caledonia. It also gives off an impression of laziness on the part of the author, and a tone of dreary technicality on the content. The only thing it does well is it works well with TeX (I heard, since I don\u2019t use TeX).\n@Qgil-WMF Thanks sm!! & any hint if this is something WMF will be taking a lasting interest in?\u2014 22:44, 17 March 2015 (UTC)\n@Kelvinsong:, Vibhabamba is UX designer at the WMF, and I recommend you to follow up with her. She has posted some advice below already.--Qgil-WMF (talk) 11:54, 19 March 2015 (UTC)\nI read about Google's w:Noto fonts recently (Noto = \"No tofu\") - that already has 98 fonts completed. Is that (code, main site, Apache licence) something that would be compatible with our needs, and your (fantastic, as always) efforts? I hope we can avoid competing standards and mass-duplication of labour, as well as getting the largest possible global installation-base. It might be ideal to collaborate on this existing effort? Quiddity (talk) 22:49, 17 March 2015 (UTC)\n@Quiddity you are confusing fonts with apps. Fonts are in a way software. But they are not built nor used in the same way apps are. If I went out and built a new open source word processor, you might be justified in asking me why I didn\u2019t just contribute to LibreOffice (though if u ask me, LibreOffice is a mess, not as bad as GIMP but approaching). That\u2019s bc you only ever need one open source word processor & it\u2019s better to have one really good Libreoffice than two lesser rival apps that do the same thing. But fonts are not apps. For one I cannot contribute to an existing font project in a meaningful way. I do not know who designed Noto (it seems to be credited to one \u201cGoogle\u201d) but only that designer can make more Noto glyphs. I, with my typographical experience can offer suggestions and critiques on his (or her) typeface & fix bugs. But I cannot directly contribute to it. Only Noto\u2019s designer can design Noto Serif.\nMore importantly, diversity is a pro in type design, not a con. There is no such thing as \u201cduplication of labor\u201d or redundancy in type design, only lost potential. This is an extremely big issue & I could write a whole article about it. but anyway\u2014specific reasons why it makes sense to create a new font:\n\u2022 I don\u2019t really like Noto Serif\u00a0: This might be a bit subjective, but personally I am not a fan of its design (largely lifted from Droid serif). Droid serif is at perfunctory glance a more polished interpretation of the \u201ccomputer type\u201d families. In essence gluing serifs onto sans fonts. Sometimes that works, some people like that, but to me it makes a font that\u2019s uncomfortable to read. Don\u2019t get me wrong. Droid serif is not a bad font\u2014in fact it\u2019s better than the professional fonts some of my textbooks are set in\u2014just not my taste. It\u2019s not exactly a design I am enthusiastic to contribute to, uk? ofc that could just be my own typographer\u2019s bias\n\u2022 I couldn\u2019t contribute to if I wanted to\u00a0: basically see what I said before. Only Noto\u2019s designer can design Noto Serif. I have done such a thing before, contributing IPA glyphs & stuff to existing fonts. You can get a decent grip on what the original designer meant but it\u2019s difficult & basically what I would truly call wasted energy. \u2014 23:31, 17 March 2015 (UTC)\n\u2022 It wouldn\u2019t be \u201cour own\u201d\u00a0: Typefaces, even libre ones, have \u201cowners\u201d. Usually this is the company or organization that uses it the most. Google \u201cowns\u201d Roboto; Mozilla (w Google) \u201cowns\u201d Open Sans; Apple \u201cowns\u201d Helvetica, and big surprise, Google \u201cowns\u201d Droid\/Noto. It\u2019s a hard concept to put into words, but you get what I mean. Fira, Gentium, and Libertine don\u2019t have this problem. You can just kind of smell it.\n\u2022 Noto isn\u2019t really free\u00a0: No typeface is (or should be) free as in gratis, but you could argue that Google\u2019s superfonts aren\u2019t even free as in speech. They\u2019re more like legally-irrevocable gifts that we are allowed to use at Google\u2019s grace. && Google has a poor track record with its treatment of the type design craft & I\u2019m reluctant to give my labor to them. && see [22]\n\u2022 We still need new fonts\u00a0: Even if Google was the most angelic company in the world; even if Noto was the best designed font in the history of the planet; even if its designer\u2019s vision of the typeface was magically transferred to my heart, we would still need more choice in type. We\u2019re starting to reach saturation with Linux distributions. Fonts still have a long way to go. Feel free to google \u201cwhy we need new fonts\u201d, because every type designer on the planet has been asked this question at some point, and some have written extensively on it.\nI hope this makes sense I didn\u2019t want to spend too long writing a long explanation of this topic\u2014 23:31, 17 March 2015 (UTC)\n@Kelvinsong: That helps immensely, thank you for the details. This proposed project is intriguing, and I wish it great success.\n(Ramble: I adore the vast diversity of typefaces, and have spent many an hour browsing typography blogs\/libraries\/articles, and learning some of the nuances of the basics [Foundry:Family:Face:Font! But I still mix them up like a philistine, all too often >.< ], but most of my online font-usage-knowledge is still from circa '98-'02, when kottke's silkscreen was all the rage, hence I have somewhat outdated views particularly regarding embedded webfonts! Again, best of wishes for this proposal. I look forward to this elegant and accessible work of science and art.\u00a0:) Quiddity (talk) 06:05, 18 March 2015 (UTC)\n@User:Quiddity Thanks!! && btw a foundry is the font publisher (usually a company or artist collective; sometimes an individual). Family & face are the same thing; Font can either mean the same as Family or refer to a single instance of a family.\u2014 13:07, 18 March 2015 (UTC)\n@Kelvinsong: Hi! While I am taking a look at your typeface, it would be a good idea for you to submit it to Typographica. I could help connect you with Stephen Coles. Is there an email address where I can reach you? Thanks.\n\u2014\u00a0Preceding unsigned comment added by Vibhabamba (talk\u00a0\u2022\u00a0contribs) 08:03, 19 March 2015\u200e (UTC)\n@User:Vibhabamba Yes, thank you! I just sent you a message \u2014 00:59, 20 March 2015 (UTC)\n\n## Problems uploading photographs from SpaceX which are available under a Wikipedia compatible licence\n\nHi all\n\nI'm trying to upload images from SpaceX which are available on their official Flickr account under a WIkipedia compatible licence. The images were uploaded to Commons but then deleted (see discussion here), I then opened an undeletion request which has been refused and was told that the images would need to be uploaded again, however my attempts at uploading them are not working. Flickr2Commons doesn't work, it rejects uploading the files because they have previously been deleted, I have also tried the Commons Upload Wizard but it has rejected them because they have previously been deleted. Please can someone tell me how I can upload them? Bare in mind there are over 100.\n\nThanks very much\n\nMrjohncummings (talk) 22:41, 21 March 2015 (UTC)\n\nI suggest anyone tempted read up on Russavia's case first. I was recently falsely accused of being Russavia's meatpuppet, so I would not touch these with a barge pole. The images are being used as a political beach-ball where established Commons policies, such as the deletion policies, appear to be ignored in order to prove a point. -- (talk) 22:57, 21 March 2015 (UTC)\nIt appears that User:Huntster manually restored all 105 images of Category:Photographs by SpaceX by uploading them one by one - without using any \"wizards\" because apparently some control freaks in charge decided it would be good policy to prohibit people from re-uploading photos automatically if they've been deleted. Huntster's work is commendable, but at the rate things are going it sounds like we may need a new upload wizard - one written in Python to be easily run on users' PCs, intentionally designed not to be detectable as a bot, and distributed offsite. The existing architecture simply doesn't take into account that a file may be deleted not because it is a bad thing to have, but out of administrative pique. Wnt (talk) 12:27, 28 March 2015 (UTC)\n\n# March 22\n\n## MUTYALAPALEM-UPDATED FILE UPLOAD PROBLEMS -REG\n\nDEAR SIR,\n\nREGARDS\n\nA SWAMY NAIDU\n\n\u2014\u00a0Preceding unsigned comment added by Naidu.a2014 (talk\u00a0\u2022\u00a0contribs) 2015-03-23T03:44:10\u200e (UTC)\nLocate the key that looks like this\nor like this\nand press it once until the little light turns off.\n1. Please stop SCREAMING in our eyes.\n2. You didn't delete anything. Only administrators can delete content, and you're not an administrator.\n3. As noted on File:MUTYALAPALEM.pdf and File:Mutyalapalem 2 side brochure.pdf, they were deleted as a result of Commons:Deletion requests\/File:MUTYALAPALEM.pdf and Commons:Deletion requests\/File:Mutyalapalem 2 side brochure.pdf.\n4. You should not recreate previously deleted content. If you can explain how this content fits within Commons' project scope, you can request undeletion.\nLX (talk, contribs) 18:42, 23 March 2015 (UTC)\nUser:LX: I find this kind of snide response to be unwarranted. Caps Lock is not literal screaming; it doesn't damage your ears. I have no idea what kind of equipment the OP has. Are there old phones out there that can access Internet but can't do lowercase, or do it with too much trouble? Myself, I remember the days of the APPLE ][, so I can handle uppercase. Trust me, you can get used to it with just a little effort. Wnt (talk) 00:17, 24 March 2015 (UTC)\nUser:Naidu.a2014: You can access your contributions by the \"contributions\" tab that should be at the top of the page by your username (at least for me... it can vary depending on settings, I think). In your case this is https:\/\/commons.wikimedia.org\/wiki\/Special:Contributions\/Naidu.a2014 . You can click on each of the files you uploaded. Just under the section titled \"File History\" you can click on a link to \"Upload a new version of this file\". If you have further trouble let us know. Wnt (talk) 00:23, 24 March 2015 (UTC)\nYou may call my response snide if you like, but at least I addressed the actual question. The question was about a deleted brochure. That won't be listed in Special:Contributions\/Naidu.a2014, but in Special:Log\/Naidu.a2014, and you can't \"Upload a new version of this file\" for files that have been deleted. LX (talk, contribs) 07:11, 24 March 2015 (UTC)\nUser:LX: Alright, that's a good point! Wnt (talk) 12:14, 28 March 2015 (UTC)\n\n## Possibility? Viewing images in sub-categories at the same time as viewing parent categories\n\nIn a discussion on EN an editor argued that one should be able to \"make an option to view all the images of a category and all its sub-categories at one time.\" - To expand on that I can see how it can be difficult to view every single one of the subcategories at the same time, but one can pick and choose by \"expanding\" or \"collapsing\" the subcategories.\n\nIs there a system being developed that is like this?\n\nWhisperToMe (talk) 07:20, 28 March 2015 (UTC)\n\nSuch a system could be problematic if for instance there is a loop in the category tree. I don't know of such an item, but it is possible to create lists of images in categories and subcategories using AWB or catscan. -mattbuck (Talk) 16:30, 28 March 2015 (UTC)\nYes, there are some loops, but it wouldn't be that hard for the JavaScript to check for loops (and for two or more paths down to the same subcat) and fail to open any category twice on the same page. - Jmabel\u00a0! talk 16:41, 28 March 2015 (UTC)\nI don't think anyone is working on it, but I think that having categories returns results in BFS order would be both feasible (mostly. Worst case of hundreds of empty sub-categories would perhaps have to be excluded), and give results more in tune with expectations than the current system of only the current category. Bawolff (talk) 18:56, 28 March 2015 (UTC)\nFastCCI can make a list of images from subcategories. --ghouston (talk) 04:30, 29 March 2015 (UTC)\n\n## Link within langauge text and Valenciennes region\n\nI try to add the map link [23] but this doesnt work while in Category:Trams in the Valenciennes region the link works. I am proposing the create the Category (Buses in the Valenciennes region). The Tranvilles public transport network is much wider than Valenciennes city. Unfortunatly there is no wel defined area. The department Nord is to large and the Arrondissement de Valenciennes is to small, as there are some communities served wich are outside the arrondisement. We dont do much with the arrondissement in the Commons anyway.Smiley.toerist (talk) 10:04, 28 March 2015 (UTC)\n\nFixed with 1= for the 1st and only parameter, your external link contained a critical = in its ?name=value query part. The same trick also often helps with {{tlx|template|param1|3=foo=bar}} etc. \u2013Be..anyone (talk) 21:46, 28 March 2015 (UTC)\n\n## 30,000 New York crime scenes\n\nI hope someone is looking into this. [24] Jim.henderson (talk) 14:17, 28 March 2015 (UTC)\n\nAmazing pictures, but I am not sure about the copyright. These were presumably never published, so are still probably under copyright. Regards, Yann (talk) 14:31, 28 March 2015 (UTC)\nI hope you don't mean the ones from the 1920s-1930s. The photo of the 1935 New York book-burning is indeed disturbing... I hope we can do better than that. Wnt (talk) 23:51, 28 March 2015 (UTC)\nHave these been published before? Who are the authors? In the US, anonymous hurts us almost invariably, life+70 versus 120 years, but I suspect the NYPD has the names of the authors. If we assume they are anonymous and this is the first publication, then those before 2014 - 120 = 1894 are public domain.--Prosfilaes (talk) 10:42, 29 March 2015 (UTC)\nNow hold on a minute! This is absurd. To begin with, they're crime scene photos. That means (a) it probably wouldn't fly to tell a court you don't know who, where, when the picture came from, and (b) it would probably be taken askance if the photographer said hold on, this is my photo, you have to pay a copyright fee if you want to use this in the trial. I would suggest this is a \"work for hire\", and the image therefore was copyrighted by the police department from the beginning, but if not they know who held it. Also, we know that the police department felt comfortable publishing the digitized photos, which either means that they are Outrageous Pirates or else they were not taking material copyrighted by unknown authors. Also, I would assume that these have been published in court proceedings, which is what they were taken for - doesn't that count for something? Wnt (talk) 12:13, 29 March 2015 (UTC)\nI think some of these could fall under {{PD-NYCGov}} Yann (talk) 12:34, 29 March 2015 (UTC)\nSplendid. So, nevermind dates and previous publication, as these are works for hire with all rights released by the employer \/ owner, right? Jim.henderson (talk) 12:57, 29 March 2015 (UTC)\nActually, it is probably not so simple. Terms and Conditions from the NYPD clearly say that all documents are not automatically in the public domain. BTW did he get a fine for parking in the wrong place?\u00a0;oD Yann (talk) 13:05, 29 March 2015 (UTC)\nAh. Too bad, though the terms link is not linking for me. And wow, I bicycle up Sterling Place a dozen times most years. At least that link works. Jim.henderson (talk) 20:53, 29 March 2015 (UTC)\n\n# March 29\n\n## Help, please, re categorizing North America, South America, the Americas\n\nI noticed that North America and South America had disappeared from many continent categories, and \"the Americas\" had appeared as a continent in their place. Templates were also changed to follow this method. My understanding is that North and South America are continents, and \"the Americas\", if anything, is a region (similar to Eurasia). After doing some work to change some of this back, I noticed that a lot of these changes seemed to have been made by the same user, User:Verdy p. I left a message here on his\/her talk page, asking that he\/she not make such changes. The user left a lengthy reply. Not all the reply is clear, such as the talk about axes (plural of axis, not axe), but I believe the upshot is that he\/she disagrees, sees a great need to continue, and plans to do more.\n\nSo I'd appreciate input\/clarification from the community as to which is right. Thanks. --Auntof6 (talk) 05:28, 29 March 2015 (UTC)\n\nEnglish is not my native languages so excuse if I use \"axis\" as a plural.\nAnd I have said thgat only a few major topics need to reconciliate other axis than just North\/South (there are at least 4 major regions in the Americas, plus 2 with Nature topics and sometimes they need cross-references, but not below the political level of countries and dependencies). And this is NOT for a lot of topics.\nAlso the term \"continent\" taken strictly is geological (and geologically, Americas are not divided exactly like in political, natural, historical, and cultural topics). It's impossible to decide just one North\/South division for every topic.\nI have not removed any one of the North\/South categories they are still all there even if there are a few others using also Latin and Caribbean.\nAlso don't make false assumptions: the Caribbean is NOT just in North America (I've seen false categorisations such as sorting files about Venezuela, Trinidad and Tobago, and the Guianas (sometimes also Colombia) besing sorted incorrectly \"North America\". This is even more important for historical and natural topics. verdy_p (talk) 05:35, 29 March 2015 (UTC)\nFirst, \"axes\" can be the plural of either \"axe\" or \"axis\": I was clarifying which I meant. I wasn't saying that you spelled it wrong. What isn't clear to me is what you mean by \"axis\". I think you are probably not using it the way we would normally use it.\nYou are right that \"the Caribbean\" is not just North America, if you're talking about the region. However, the islands and nations in the Caribbean (as opposed to those on the edge of it) are categorized here as part of North America. That includes Trinidad and Tobago, even though it's very close to mainland South America. Sometimes we have to agree to categorize some things a certain way, even if it's arguable. That is why I mentioned that I know there are systems where \"the Americas\" is considered a continent. It's just that, in the Wikimedia projects I work on, North and South America are considered the continents.\nI know that you didn't remove the North America\/South America categories. You put them under \"the Americas\". However, they also need to be under continents.\nYou should have noticed that I have North\/South categories as direct members of continents. This is not a critical problem. Though we still have lots of medias that cannot be sorted that way (nature, culture, people, history and all related politics topics touching these domains).\nEven if you like it or not, there already existed categories for Caribbean and Latin America. But sorting them as direct children of North and South repectively is wrong and we get many files incorrectly categorized due to this confusion, or already sorted only in Caribbean or in Latin America that cannot be reached by looking them first by country and their history\/nature\/people\/culture subtopics. All I wanted to do was to do standard cross-categorization so that all of them are precisely categoprized geographically instead of remaining in \"limbos\" (far way in parent or children topics, for another American cultural or natural subregion unrelated to the North\/South division). verdy_p (talk) 18:00, 29 March 2015 (UTC)\nNot everything needs to have separate categories under the Americas. There are some things that make sense to group there, such as some geographical things and certain nature categories. \"The Americas\" is a region, just like Eurasia, and not that much different from a hemisphere. We don't have so many categories for those, and we don't need so many for the Americas. --Auntof6 (talk) 06:51, 29 March 2015 (UTC)\nI believe Auntof6 is entirely correct here. And, similarly, we do not want to turn Europe + Asia into Eurasia just because the boundary is unclear in several places. Please, Verdry p, do not continue farther in this direction except in the unlikely event that you can develop a consensus to do this. This is not a situation in which to be unilaterally \"bold\". Take it to COM:CFD or such if you want to try to develop such a consensus. - Jmabel\u00a0! talk 17:36, 29 March 2015 (UTC)\nAuntof6 \"Americans\" (funny), better, people from United Estates of America, \"study\" Geographic like \"USA and the rest\", in reflection of that, Central America do not exist and South America is a different continent. America is the continent for many geographers, and South\/North and Central Americas plus Caribbean islands are the subcontinent.\nThis is not Eurasia, this is what people outside USA study...\nAgain America, not Americas...\nBut I will not go further than this, this is a cultural issue way bigger than a simple category. -- RTA 06:25, 30 March 2015 (UTC)\nI realize that there are other ways of defining continents. en:Continent#Number of continents describes several methods. The issue is that Wikimedia Commons chose one method that everyone needs to follow, and that method uses the 7 continents that are shown at Category:Continents. That doesn't mean other methods are wrong, just that they aren't the method used here. Other things that are sometimes considered continents include the Americas, Eurasia, and Afro-Eurasia. Here on Commons, those are considered regions (intercontinental regions, to be exact), not continents.\nThe area that consists of North America and South America combined can be called either America or the Americas. Calling it America can be confusing because the United States is sometimes called \"America\" as well. Because of that, the Americas is clearer.\nThere are few categories that are meaningful at the continent level, even fewer if we start categorizing by something even bigger than the continents that we use now. --Auntof6 (talk) 07:20, 30 March 2015 (UTC)\n\n## Derivative work?\n\nWhat's the extend of the definition of \"derivative work\"? If a painter paints a picture after a (copyrighted) picture, does the painter need permission of the original copyright holder? Case in point: File:Jack Sels by Jules Grandgagnage.jpg is painting by a Wikimedia resident painter (the source link doesn't work for me; ymmv). But the real source of the image seems to be a Jack Sels record from 1961. Whaledad (talk) 21:14, 29 March 2015 (UTC)\n\nBrief answer, yes. If a work was inspired by, or based upon, some other work, then it is a derivative work. I suggest reading en:Derivative work, which is a not bad explanation. A derivative work cannot be 'more freely licensed' than the work it is based upon, and must give attribution to the original. Revent (talk) 05:49, 30 March 2015 (UTC)\nAs a further note, I'm strongly tempted to DR the work you linked on that basis, and would probably support deletion if it was well-reasoned... I didn't look at the copyright status of the original, but it seems to be an obvious DW, and should be attributed thus. Revent (talk) 05:52, 30 March 2015 (UTC)\nFormally, \"A \u201cderivative work\u201d is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a \u201cderivative work\u201d.\" 17 U.S.C. \u00a7 101 Revent (talk) 05:54, 30 March 2015 (UTC)\n\n# March 31\n\n## First concrete bridge and Budapest metro line 1\n\ndepicts a footbridge over an metro line. My guide told me this was the first bridge build with concrete in the world. This metroline construction was finished in 1896. Could a date be found for the bridge. On the en:Line 1 (Budapest Metro) I could not find any mention of the new aligment (straither line and underground) by the renovation in the 1970s?Smiley.toerist (talk) 23:56, 24 March 2015 (UTC)\n\nI also uploaded an excursion with the old metro car: File:Museum metro line 1 Budapest 01.JPG to File:Museum metro line 1 Budapest 04.JPG.Smiley.toerist (talk) 00:20, 25 March 2015 (UTC)\n\nSmiley.toerist It appears as though the Alvord Lake Bridge predates it by a few years, and that does not include Roman viaducts and infrastructure in that equation, as they also were made from concrete. Kevin Rutherford (talk) 22:13, 31 March 2015 (UTC)\n\n# March 25\n\n## Classifying hot steel transport\n\nI cant seem to find an adapted category for this kind of rail transport.Smiley.toerist (talk) 10:18, 29 March 2015 (UTC)\n\nI tried a few categories, none particularly relevant. It is incomplete, but sooner or later it will be improved. -- Tuv\u00e1lkin 19:06, 30 March 2015 (UTC)\nSo create one. They're obscure (but notable) so we probably don't have one as yet. I think they're generally called \"torpedo ladles\". The ones shaped like buckets are generally different as they're \"slag ladles\" for waste.\nBTW, it's unusual, but they have at times been used for long distances along main lines. [25] Andy Dingley (talk) 15:56, 31 March 2015 (UTC)\nAlready here at Category:Torpedo wagons Andy Dingley (talk) 16:03, 31 March 2015 (UTC)\n\n## Unifying emojis filenames\n\nHi! There are four different sets of emojis uploaded on Commons, each with its own naming scheme. I believe it would be good to rename them all so that the names are consistent across the different sets (and futures freely licensed sets that will appear one day!).\n\nHere\u2019s an instance of how the four styles of the smiling face emoji are named today:\n\n U+263A Emoji u263a.svg Twemoji 263a.svg Emojione 263A.svg PEO-white smiling face.svg\n\nAnd here are some proposed new names using the system Emoji <Name of the set> <uppercase Unicode number>.svg:\n\n U+263A Emoji Noto 263A.svg Emoji Twitter 263A.svg Emoji One 263A.svg Emoji Phantom 263A.svg","date":"2018-05-24 20:31:30","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 1, \"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\": 0.39605093002319336, \"perplexity\": 3007.9178394044516}, \"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-2018-22\/segments\/1526794866772.91\/warc\/CC-MAIN-20180524190036-20180524210036-00517.warc.gz\"}"}
| null | null |
Q: Toxiclibs.js extending particle class It may be a bit too specific, but maybe someone here has experience extending the particle class of toxiclib.js?
I keep getting:
TypeError: this.setWeight is not a function
I am doing a very simple thing (I am using Processing.js):
class Particle extends VerletParticle2D {
Particle(float x, float y) {
super(x,y);
}
void display() {
stroke(0);
ellipse(x,y,16,16);
}
}
A: Inheritance between Processing.js and other libraries is not all figured out yet. There is a workaround.
First thing is to refer to VerletParticle2D by its full namespace:
class Particle extends toxi.physics2d.VerletParticle2D
Second part is to add this to toxiclibs.js' constructor for VerletParticle2D (at the time of writing this, that is line #9941 of build/toxiclibs.js):
if(!(this instanceof VerletParticle2D)){
return new VerletParticle2D(x,y,w);
}
Third part is to add these 2 lines anywhere after you have defined your class. Unfortunately these 2 lines won't play nice with the Processing IDE:
Particle.prototype = new toxi.physics2d.VerletParticle2D();
Particle.prototype.constructor = Particle;
You can apply these 3 patterns to any other class as well to extend them.
Also, Processing 2.0 Alpha 5 released a new feature that will pull a .js file next to an accompanying .jar. So if you place toxiclibs.js next to toxiclibscore.jar and rename it, it will get exported with your sketch. This will make it easier to have one version that you make some modifications to for extending classes.
I have uploaded the web-exported sketch, with the modified toxiclibscore.js file here: http://haptic-data.com/toxiclibsjs/misc/ToxiclibsjsExtend.zip
best of luck!
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 5,603
|
"Attracting Abundance, Happiness & Love" Indian Agate Hand-Knotted Necklace with Opalized Ammonite
Featuring a hand-knotted (in blue natural silk thread) necklace of beautifully earthy Indian Agate beads (measuring 10mm in size), adorned with an ancient "opalized" Ammonite fossil wrapped in bronze wire. The iridescence of this jurassic Ammonite is so magical!
The length of the necklace measures 19" long and the pendant measures 1.7" long.
Part of my second "Elemental Earthchildren" collection, designed to embody different elements and become treasured talismans that provide a specific set of energetic properties for you and your journey.
This piece represents and promotes EARTH + WATER elements and energies ----->
--EARTH is the element responsible for stability, strength, grounding, vitality, confidence, dependability, caution, goal-orientation, organization, discipline, success, wealth, and a deep connection to Mama Gaia and Mother Nature.
--WATER is the element responsible for compassion, connection, devotion, forgiveness, love, psychic awareness, intimacy, emotional healing, healing others, stress relief, inner peace, self-expression, empathy, nurturance, tranquility, and dreamwork.
MANTRAS for this piece:
"I attract all the love and success that I dream of and deserve."
"I am confident in my self, my purpose, and all that I stand for."
"I radiate happiness, light, strength, and positivity."
"I embrace emotional healing and fully welcome inner peace with open arms."
"I live my life with a deep sense of connection, tranquility, and love."
Earth, Air, Fire, and Water are thought of (commonly in Western culture) as the four different elements that represent the nature of matter in our world on this physical plane. Hippocrates theorized that the temperaments and characteristics of these four elements must be in a balanced union in order for an individual to have well-being on a mental, physical, spiritual, emotional level. Depending on our own personal temperament and characteristics, we resonate most with specific earth elements at different points of our lives.
We are all simultaneously on our own path of spiritual evolution (whether we are aware of it or not), we're just in different developmental points of our human spirit. We each tend to align to a specific element (or set of elements) that helps to guide us at that specific point of our spiritual development. The element(s) that we align to and identify with can help us to be aware of our personal strengths and weaknesses; At times, however, we must incorporate other earth elements into our practice and conscious awareness, which can greatly help to create a more beneficial, stable, and better balance within ourselves and in our lives. This collection was carefully crafted to provide specific energetic enhancements to tailer what we may be in need of energetically within ourselves and in our lives!
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 6,094
|
/* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.camunda.bpm.dmn.xlsx;
import org.camunda.bpm.dmn.xlsx.api.Spreadsheet;
import org.camunda.bpm.dmn.xlsx.api.SpreadsheetCell;
import org.xlsx4j.sml.STCellType;
/**
* @author Thorben Lindhauer
*
*/
public class DmnValueNumberConverter implements CellContentHandler {
public boolean canConvert(SpreadsheetCell cell, Spreadsheet context) {
return STCellType.N.equals(cell.getRaw().getT());
}
public String convert(SpreadsheetCell cell, Spreadsheet context) {
return cell.getRaw().getV();
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 1,835
|
Q: Custom Callback is not working with a keras model I try to evaluate my tf.keras model within a Callback. Doing model.evaluate after fit works, but within the Callback not.
class Eval(Callback):
def __init__(self, model,validation_data,eval_batch=1):
super(Callback, self).__init__()
self.validation_data=validation_data
self.eval_batch=eval_batch
self.model=model
#self.X_val, self.y_val = validation_data
def on_epoch_end(self, epoch, logs={}):
if epoch % 1 == 0:
print(epoch)
print(self.validation_data)
res=self.model.evaluate(self.validation_data, batch_size=self.eval_batch)
print(res)
precisions.append(res[3])
print("Evaluation - epoch: {:d} - Eval_Prec: {:.6f}".format(epoch, res[3]))
Calling
evaluation=Eval(keras_model,eval_input)
keras_model.fit(
x_train,y_train,
epochs=args.num_epochs,
validation_data=(x_test,y_test),
batch_size=args.batch_size,
verbose=1,
callbacks=[evaluation])
And here the full traceback:
Traceback (most recent call last):
File "keras_cloud.py", line 435, in <module>
train_and_evaluate(args)
File "keras_cloud.py", line 421, in train_and_evaluate
callbacks=[lr_decay_cb,evaluation])#,tensorboard_cb])
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training.py", line 1639, in fit
validation_steps=validation_steps)
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training_arrays.py", line 239, in fit_loop
callbacks.on_epoch_end(epoch, epoch_logs)
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\callbacks.py", line 214, in on_epoch_end
callback.on_epoch_end(epoch, logs)
File "keras_cloud.py", line 312, in on_epoch_end
res=self.model.evaluate(self.validation_data, batch_size=self.eval_batch)
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training.py", line 1751, in evaluate
steps=steps)
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training.py", line 992, in _standardize_user_data
class_weight, batch_size)
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training.py", line 1117, in _standardize_weights
exception_prefix='input')
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training_utils.py", line 284, in standardize_input_data
data = [standardize_single_array(x) for x in data]
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training_utils.py", line 284, in <listcomp>
data = [standardize_single_array(x) for x in data]
File "C:\Users\\Anaconda3\lib\site-packages\tensorflow\python\keras\engine\training_utils.py", line 218, in standardize_single_array
if x.shape is not None and len(x.shape) == 1:
AttributeError: 'float' object has no attribute 'shape'
But the following works:
keras_model.fit(...)
print(keras_model.evaluate(eval_input, steps=1))
I have no clue what is going on with that Callback?
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 1,516
|
{"url":"http:\/\/mathhelpforum.com\/differential-geometry\/181451-isometric.html","text":"# Math Help - Isometric\n\n1. ## Isometric\n\nShow that C[0,1] and C[a,b] are isometric.\nSolution:\nA mapping T of X into X' is said to be isometric if for all x, y in X\nd'(Tx,Ty)=d(x,y)\nC is continous [0,1] and [a,b] closed interval..\nThe distance on Function space C[a,b] is d(x,y)=max|x(t)-y(t)|\n\nI couldnt procced from here....i need some help....\n\nthanks\n\n2. Originally Posted by kinkong\nShow that C[0,1] and C[a,b] are isometric.\nSolution:\nA mapping T of X into X' is said to be isometric if for all x, y in X\nd'(Tx,Ty)=d(x,y)\nC is continous [0,1] and [a,b] closed interval..\nThe distance on Function space C[a,b] is d(x,y)=max|x(t)-y(t)|\n\nI couldnt procced from here....i need some help....\n\nthanks\nWhat if you did a mapping that bijectively took $[a,b]\\to[0,1]$ in an increasing way, call this $g$, and you considered $T:C[a,b]\\to C[0,1]$ defined by $Tf(x)=f(g(x))$?\n\n3. Thank u very much...\nBut how can we prove that the map is bijective...i mean how can we see that it is one-to-one and onto...\n\n4. Originally Posted by kinkong\nThank u very much...\nBut how can we prove that the map is bijective...i mean how can we see that it is one-to-one and onto...\nHave you tried to prove it?\n\n5. yes i thought about it..but i couldnt find a way...please can u help me...","date":"2014-03-09 02:06:27","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\": 0.8048545718193054, \"perplexity\": 2208.411654080748}, \"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-2014-10\/segments\/1393999670363\/warc\/CC-MAIN-20140305060750-00043-ip-10-183-142-35.ec2.internal.warc.gz\"}"}
| null | null |
\section{Introduction}
Reconstructing the 3D shape of deformable objects from monocular image sequences is known as Non-Rigid Structure-from-Motion (NRSfM) and has applications in domains ranging from entertainment~\cite{Parashar19b} to medicine~\cite{Lamarca19}. Early methods relied on low-rank representations of the surfaces~\cite{Bregler00,DelBue06,Akhter08,Torresani01,Dai14,Lee16,Gotardo11a,Kumar19,Kumar19a}, while more recent ones exploit local surface properties to derive constraints and can handle larger deformations~\cite{Varol09,Vicente12,Taylor10a,Chhatkuli14,Chhatkuli17,Ji17}. Unfortunately, these constraints have to be enforced jointly on the entire set of reconstructed points for a whole sequence. Hence, the computational cost increases non-linearly with the number of images and quickly becomes prohibitive. Furthermore, a globally optimal solution is obtained using an iterative refinement, which requires a reliable initialization that is not always available. Finally, these {\it global} methods cannot handle missing data.
In earlier work~\cite{Parashar17,Parashar19a,Parashar20}, we have shown that {\it local} methods constitute a powerful alternative. Expressing isometry, conformality, or equiareality constraints in terms of differential properties makes the number of local variables remain fixed. Unfortunately, the systems of equations that arise in these computations are bivariate of high degree. They can have up to five real solutions. In theory, a unique solution can be obtained from 3 images, but this requires either a
complicated sum-of-squares formulation~\cite{Parashar17,Parashar19a} or reduction methodologies that add phantom solutions~\cite{Parashar20,Parashar21}. Hence, in practice, it takes more than 3 images to produce reliable estimates. Furthermore, when the motion between the frames is too small, the system becomes ill-posed and the estimates unreliable, without any mechanism to flag such situations as problematic.
In this paper, we introduce a new local method. Instead of inferring the depth derivatives, we estimate surface normals. More specifically, given a 2D warp between two images, we consider tangent planes at corresponding points. For each pair of points, we compute the homography relating the two planes and decompose it to compute the normals by solving local differential constraints~\cite{Parashar17,Parashar19a}. This has two solutions, instead of five in our earlier approaches~\cite{Parashar21}. For each plane, we pick the right one by enforcing an easy-to-compute measure of local smoothness. Furthermore, our formalism lets us assess how well-conditioned the problem was and, hence, how usable the resulting normals are. In other words, we can derive from an image pair a set of reliable normals and discard the others.
We will demonstrate on both synthetic and real data that we outperform state-of-the-art local and global methods at a fraction of the computational cost. Our contribution is therefore an approach to NRSfM that relies on solving in closed form a set of equations relating surface normals at corresponding points. Being entirely local, the computation is both fast and reliable. Although our solution is designed for isometric or conformal deformations, it yields good results for generic ones.
\section{Computing Normals from Two Images}
In earlier approaches~\cite{Parashar19a}, the NRSfM problem was addressed by solving the system of Eq.~\ref{eq:iso-nrsfm} under the isometry, conformality, and equiareality constraints of Eq.~\ref{eq:iso-nrsfm} with respect to the variables $k_1$ and $k_2$ of Eq.~\ref{eq:j_phi}. Here, we solve this system of equations directly in terms of the surface normals. We will show that, not only can this be done in closed form, but it also allows us to identify degenerate situations that result in unreliable estimates.
\parag{Differentiating the Warp.}
Let us consider a point $\overline{\mathbf{x}} = (\overline{u},\overline{v})^T$ in $\overline{\mathcal{I}}$ and its corresponding point $(u,v)^T=\eta(\overline{u},\overline{v})$ in $\mathcal{I}$, with corresponding points on surfaces $\mathbf{X}$ and $\overline{\mathbf{X}}$. Assuming the surfaces to be locally planar means that there is a $3 \times 3$ homography matrix $\mathbf{H} = [h_{ij}]_{1 \leq i,j \leq 3}$ such that $\mathbf{X} = \lambda \mathbf{H}\overline{\mathbf{X}}$. Since we assume a perspective projection for the camera, we write
\begin{equation}
\label{eq:homography}
\mathbf{x} = \frac{1}{\overline{s}}\mathbf{H}\overline{\mathbf{x}} \implies
\!\begin{pmatrix}
u \\ v \\ 1
\end{pmatrix}\!=\!\frac{1}{\overline{s}}\!\begin{pmatrix}
h_{11} \!& \!h_{12} \!&\! h_{13} \\
h_{21} \!&\! h_{22} \!&\! h_{23} \\
h_{31} \!&\! h_{32} \!&\! h_{33}
\end{pmatrix}\!\begin{pmatrix}
\overline{u} \\ \overline{v} \\ 1
\end{pmatrix}, \;
\end{equation}
where $\overline{s}=h_{31}\overline{u}\!+\!h_{32}\overline{v}\!+\!h_{33}$. The first- and second-order derivatives of $\eta$ can be computed as
\begin{align}
\label{eq:eta_der}
\mathbf{J}_\eta =
\begin{pmatrix}
\dfrac{\partial \eta}{\partial \overline{u}} & \dfrac{\partial \eta}{\partial \overline{v}}
\end{pmatrix}&=\frac{1}{\overline{s}}
\begin{pmatrix}
h_{11}\!-\!h_{31}u \!&\! h_{12}\!-\!h_{32}u \\
h_{21}\!-\!h_{31}v \!&\! h_{22}\!-\!h_{32}v
\end{pmatrix} \; , \nonumber \\[1mm]
\begin{pmatrix}
\dfrac{\partial^2 \eta}{ \partial \overline{u}^2} \! & \! \dfrac{\partial^2 \eta}{ \partial \overline{u}\overline{v}} \! & \! \dfrac{\partial^2 \eta}{ \partial \overline{v}^2}
\end{pmatrix}
&= -\frac{1}{\overline{s}} \mathbf{J}_\eta\begin{pmatrix}
2h_{31} \! & \! h_{32} \!& \! 0 \\ 0 \!& \! h_{31} &2h_{32}
\end{pmatrix}.
\end{align}
\parag{Image Embedding and Local Normal.}
The unit normal $\mathbf{n}$ at $\mathbf{x}$ is the cross product of the columns of the matrix $\mathbf{J}_\phi$ from~Eq.~\ref{eq:j_phi}. This lets us write
\begin{align}
\label{eq:normal}
\mathbf{n} & =\! \frac{1}{\beta^2\sqrt{\det \mathbf{g}}}\!\begin{pmatrix}
k_1 \\k_2 \\ 1\!-\!uk_1\!-\!vk_2
\end{pmatrix} \\
&=\!\frac{1}{\beta^2\sqrt{\det \mathbf{g}}}\!\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ -\mathbf{x}^\top &1
\end{pmatrix}\!\begin{pmatrix}
k_1 \\ k_2 \\ 1
\end{pmatrix}\!. \nonumber \\
\Rightarrow
\begin{pmatrix}
k_1 \\ k_2 \\ 1
\end{pmatrix} & =\!\beta^2\sqrt{\det \mathbf{g}}\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ \mathbf{x}^\top &1
\end{pmatrix}\mathbf{n}. \label{eq:k1k2}
\end{align}
Given the normal $\mathbf{n}$ of Eq.~\ref{eq:normal}, we rewrite the matrix $\mathbf{J}_\phi$ of Eq.~\ref{eq:j_phi} as
\begin{align}
\label{eq:j_phi_in_n}
\nonumber
\mathbf{J}_\phi \!&\!=\! \frac{1}{\beta}\!\begin{pmatrix}
0 & uk_1\!+\!vk_2\!-\!1 & k_2 \\
1\!-\!uk_1\!-\!vk_2 & 0 & \!-\!k_1 \\
\!-\!k_2 & k_1 & 0
\end{pmatrix}\!
\begin{pmatrix}
0 & 1 \\ \!-\!1 & 0 \\ v & \!-\!u
\end{pmatrix} \\
\!&\!=\! \beta\sqrt{\det \mathbf{g}}[\mathbf{n}]_{\times}\mathbf{E}.
\end{align}
We can now rewrite the differential constraints across images introduced in Section~\ref{sec:diffConstraints} in terms of the normals.
\parag{Linear Relation between Surface Normals.}
Given the $\eta$ derivatives from Eq.~\ref{eq:eta_der}, the linear relation of Eq.~\ref{eq:CS_rel} becomes
\begin{align}
\label{eq:CS_law}
\begin{pmatrix}
\overline{k}_1 \\ \overline{k}_2
\end{pmatrix}&=\!\mathbf{J}_\eta^\top\!\begin{pmatrix}
k_1 \\ k_2
\end{pmatrix}\!+\!\dfrac{1}{\overline{s}}\!\begin{pmatrix}
h_{31} \\h_{32}
\end{pmatrix} \; , \nonumber \\
\Rightarrow
\begin{pmatrix}
\overline{k}_1 \\ \overline{k}_2 \\ 1
\end{pmatrix}&=\!\begin{pmatrix}
\mathbf{J}_\eta^\top & \mathbf{m} \\ 0 & 1 \end{pmatrix}\!\begin{pmatrix}
k_1 \\k_2 \\1
\end{pmatrix}\!.
\end{align}
Using Eq.~\ref{eq:k1k2}, we rewrite the above expression as
\begin{align}
\label{eq:CS_norm}
\nonumber
\overline{\mathbf{n}}\!&\!=\!\frac{\beta^2}{\overline{\beta}^2}\!\sqrt{\frac{\det \mathbf{g}}{\det \overline{\mathbf{g}}}}\mathbf{T}\mathbf{n}\\
\nonumber \!&\!=\!\frac{\beta^2}{\overline{\beta}^2}\!\sqrt{\frac{\det \mathbf{g}}{\det \overline{\mathbf{g}}}}\!\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ -\overline{\mathbf{x}}^\top &1
\end{pmatrix}\!\begin{pmatrix}
\!\mathbf{J}_\eta^\top\!&\!\mathbf{m}\!\\ 0 & 1 \end{pmatrix}\!\begin{pmatrix}
\!\mathbf{I}_{2\times 2}\!&\!0\!\\ \mathbf{x}^\top &1
\end{pmatrix}\!\mathbf{n}
\\ \nonumber
\!&\!=\!\dfrac{ \beta^2}{\overline{s}\overline{\beta}^2}\!\sqrt{\frac{\det\!\mathbf{g}}{\det\!\overline{\mathbf{g}}}}\!\begin{pmatrix}
\!\mathbf{I}_{2\times 2}\!&\!0\!\\\!-\!\overline{\mathbf{x}}^\top\!&\!1\!
\end{pmatrix}\!\begin{pmatrix}
\!h_{11}\!&\!h_{21}\!&\!h_{31}\!\\
\!h_{12}\!&\!h_{22}\!&\!h_{32}\!\\
\!\overline{s}u\!&\!\overline{s}v\!&\!\overline{s}\!
\end{pmatrix}\!\mathbf{n}\! \\ &\!=\!\dfrac{ \beta^2}{\overline{s}\overline{\beta}^2}\sqrt{\frac{\det\!\mathbf{g}}{\det\!\overline{\mathbf{g}}}}\!\mathbf{H}^\top\!\mathbf{n} \; ,
\end{align}
which directly relates the two normals.
\parag{Metric Tensor.}
As shown in Fig.~\ref{fig:model}, we can write $\overline{\phi} = \psi \circ \phi \circ \eta$. Differentiating this expression and multiplying it by its transpose yields
\begin{equation}
\label{eq:mt_gen}
\overline{\mathbf{g}}=\mathbf{J}_{\overline{\phi}}^\top\mathbf{J}_{\overline{\phi}} = \mathbf{J}_\eta^\top\mathbf{J}_\phi^\top\mathbf{J}_\psi^\top\mathbf{J}_{\psi}\mathbf{J}_\phi\mathbf{J}_\eta.
\end{equation}
Using Eq~\ref{eq:j_phi_in_n}, we write $\mathbf{J}_\phi\mathbf{J}_\eta = \beta \sqrt{\det g}[\mathbf{n}]_\times \mathbf{E}\mathbf{J}_\eta$. Given the $\eta$ derivatives of Eq.~\ref{eq:eta_der}, we simplify $\mathbf{E}\mathbf{J}_\eta$ to $\frac{1}{\overline{s}}\begin{pmatrix}
\mathbf{h}_1 \times \hat{\mathbf{x}} & \mathbf{h}_2 \times \hat{\mathbf{x}}
\end{pmatrix}$, where $\mathbf{h}_1, \mathbf{h}_2$ are the first two columns of the homography matrix $\mathbf{H}$, and $\hat{\mathbf{x}} = \begin{pmatrix}
u & v & 1
\end{pmatrix}^\top$.
By writing $\mathbf{z}_1 = \mathbf{n} \times (\mathbf{h}_1 \times \hat{\mathbf{x}})$ and $\mathbf{z}_2 = \mathbf{n} \times (\mathbf{h}_2 \times \hat{\mathbf{x}})$, Eq.~\ref{eq:iso-nrsfm} reduces to
\begin{equation}
\label{eq:mt_norm_simp}
\begin{cases}
\overline{\mathbf{g}}\!=\!\frac{\lambda^2\beta^2\det (\mathbf{g})}{\overline{s}^2}\begin{pmatrix}
\mathbf{z}_1^\top \mathbf{z}_1 \!\! & \!\! \mathbf{z}_1^\top \mathbf{z}_2 \\ \mathbf{z}_1^\top \mathbf{z}_2 \!\! &\!\! \mathbf{z}_2^\top \mathbf{z}_2
\end{pmatrix} \!,\!
\\
\sqrt{\det(\overline{\mathbf{g}})}\
=\sqrt{\det(\mathbf{J}_{\eta}^\top \mathbf{g}\mathbf{J}_{\eta})}.\!
\end{cases}
\end{equation}
\parag{NRSfM from Isometric/Conformal Constraints.}
So far, we have expressed the metric preservation conditions in terms of the normals of the two surfaces under consideration. The only unknown left in the system is therefore $\mathbf{n}$. We now show that this unknown can in fact be computed in closed form.
Given the multiplicative
nature of the cross product, the constraints on the normals of Eq.~\ref{eq:CS_norm} imply that
\begin{equation}
[\overline{\mathbf{n}}]_\times=\dfrac{ \beta^2}{\overline{s}\overline{\beta}^2}\sqrt{\frac{\det\mathbf{g}}{\det\overline{\mathbf{g}}}}\det (\mathbf{H}^\top) \mathbf{H}^{-1}[\mathbf{n}]_\times \mathbf{H}^{-\top} \; .
\end{equation}
This lets us rewrite the matrix $\mathbf{J}_{\overline{\phi}}$ of Eq.~\ref{eq:j_phi_in_n} as
\begin{align}
\label{eq:jphi_n_overline}
\nonumber
\mathbf{J}_{\overline{\phi}}&\!=\!\frac{ \beta^2}{\overline{\beta}}\sqrt{\det \mathbf{g}} \mathbf{H}^{-1}\![\mathbf{n}]_\times\!\left(\!\frac{\det \mathbf{H}^\top}{\overline{s}}\!\mathbf{H}^{-\top}\overline{\mathbf{E}}\!\right)
\\
&\!=\!\frac{\ \beta^2\!\sqrt{\det \mathbf{g}}}{\overline{\beta}}\!\mathbf{H}^{\!-1}\![\mathbf{n}]_\times\!\begin{pmatrix}
\!\mathbf{h}_1\!\times\!\hat{\mathbf{x}}\!&\!\mathbf{h}_2\!\times\!\hat{\mathbf{x}}\!\end{pmatrix} \\ \nonumber
&\!=\!\frac{ \beta^2\!\sqrt{\det \mathbf{g}}}{\overline{\beta}}\!\mathbf{H}^{\!-1}\!\begin{pmatrix}
\!\mathbf{z}_1\!&\!\mathbf{z}_2\!
\end{pmatrix}\!.
\end{align}
Injecting this expression into the isometric/conformal metric tensor preservation relation of Eq.~\ref{eq:mt_norm_simp} yields
\begin{small}
\begin{align}
\label{eq:recon_eqn}
\begin{pmatrix}
\mathbf{z}_1^\top\mathbf{H}^{\!-\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_1\!&\!\mathbf{z}_1^\top\mathbf{H}^{\!-\!\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_2\! \\ \mathbf{z}_1^\top\mathbf{H}^{\!-\!\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_2\!&\!\mathbf{z}_2^\top\mathbf{H}^{\!-\!\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_2\end{pmatrix}&=\!\frac{\lambda ^2 \overline{\beta}^2}{\overline{s}^2\beta^2}\!\begin{pmatrix}
\mathbf{z}_1^\top \mathbf{z}_1 \!\!& \!\!\mathbf{z}_1^\top \mathbf{z}_2 \\ \mathbf{z}_1^\top \mathbf{z}_2 \!\!& \!\! \mathbf{z}_2^\top \mathbf{z}_2
\end{pmatrix}\!, \nonumber \\
\Rightarrow \mathbf{z}_i^\top \left(\overline{\mathbf{H}}^\top\overline{\mathbf{H}}-\frac{\lambda^2 \overline{\beta}^2}{\overline{s}^2\beta^2}\mathbf{I}_{3\times 3}\right)\mathbf{z}_j & = 0, \forall i,j \in \{1,2\},
\end{align}
\end{small}
where $\overline{\mathbf{H}}=\mathbf{H}^{-1}$. Assuming $\mathbf{H}$ to be normalized, that is, its second singular value to be 1, the relation between a 3D point observed in the two input images is given by $\phi(\mathbf{x}) = \mathbf{H} \phi(\overline{\mathbf{x}}) $. Using Eq.~\ref{eq:homography} yields $\overline{\beta} = \beta \overline{s}$. By writing $\mathbf{z}_i = [\mathbf{n}]_{\times}[\mathbf{h}_i]_\times \hat{\mathbf{x}}$, the above constraints further simplify to
\begin{equation}
\label{eq:recon_H}
[\mathbf{n}]_\times^\top (\overline{\mathbf{H}}^\top\overline{\mathbf{H}}-\lambda^2\mathbf{I}_{3\times 3})[\mathbf{n}]_\times = 0.
\end{equation}
Since $ \mathbf{H} \sim \alpha \mathbf{H},$ we divide the above expression by $\lambda^2$ and, with a slight abuse of notation, write $\frac{1}{\lambda}\overline{\mathbf{H}}$ as $\overline{\mathbf{H}}$.
This simplifies the above expressions to
\begin{equation}
\label{eq:final}
[\mathbf{n}]_\times^\top (\overline{\mathbf{H}}^\top\overline{\mathbf{H}}-\mathbf{I}_{3\times 3})[\mathbf{n}]_\times =[\mathbf{n}]_\times^\top \mathbf{S}[\mathbf{n}]_\times = 0.
\end{equation}
\parag{Degenerate Cases.} The system of Eq.~\ref{eq:final} holds as long as $\mathbf{S}$ is a non-null matrix, which means $\overline{\mathbf{H}}^\top \overline{\mathbf{H}} \neq \mathbf{I}_{3\times3}$. Therefore, $\overline{\mathbf{H}}$ should not be an orthogonal matrix, which makes pure rotations and reflections cause degeneracies.
\parag{Affine Stability.} Under affine imaging conditions, $h_{31}=h_{32}=0$, and $h_{33}=1$.
In this case, $\mathbf{z}_i$ and $\mathbf{S}$ remain non-null, and thus the system in Eq.~\ref{eq:final} does \emph{not} become degenerate, and we can still compute the normal.
\parag{Solution.} The solution to the system in Eq.~\ref{eq:final} can be obtained by homography decomposition~\cite{Malis07}. We give an overview of the solution here but recommend reading~\cite{Malis07} for more detail.
$\mathbf{S}=\{s_{ij}\}$ is a symmetric matrix expressed in terms of $\overline{\mathbf{H}}$. It can be numerically computed using $\eta$ and image observations $(\mathbf{x},\overline{\mathbf{x}})$. Specifically, Eq.~\ref{eq:CS_norm} gives the closed-form definition
$\mathbf{H}^\top =
\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ -\overline{\mathbf{x}}^\top &1
\end{pmatrix}\!\begin{pmatrix}
\!\mathbf{J}_\eta^\top\!&\!\mathbf{m}\!\\ 0 & 1 \end{pmatrix}\!\begin{pmatrix}
\!\mathbf{I}_{2\times 2}\!&\!0\!\\ \mathbf{x}^\top &1
\end{pmatrix}$. Let us write $\mathbf{n} = \begin{pmatrix}
n_1 & n_2 & n_3
\end{pmatrix}^\top$. Since $n_3 \neq 0$, we define $y_1 = \frac{n_1}{n_3}$ and $y_2 = \frac{n_2}{n_3}$ and expand the system in Eq.~\ref{eq:final} accordingly. This yields 6 constraints, out of which only 3 are unique. They are given by
\begin{align}
\label{eq:quad_recon}
\nonumber s_{33} y_2^2 - 2s_{23} y_2 + s_{22} &= 0\;, \\
\nonumber s_{33} y_1^2 - 2s_{13} y_1 + s_{11} &= 0\;, \\
s_{22} y_1^2 - 2s_{12} y_1 y_2 + s_{11} y_2^2 &= 0\;.
\end{align}
By solving the first two,
we obtain $y_1 = \dfrac{s_{13} \pm \sqrt{s_{13}^2-s_{33}s_{11}}}{s_{33}}$ and $y_2 =\dfrac{s_{23} \pm \sqrt{s_{23}^2-s_{33}s_{22}}}{s_{33}}$. We use the third expression to disambiguate the solutions. Ultimately, this gives us closed-form expressions
for the two potential solutions for the normal,
written as
\begin{align}
\label{eq:normal_sol}
\nonumber
&\mathbf{n}_a = \begin{pmatrix}
s_{13} + s\sqrt{s_{13}^2-s_{33}s_{11}}& s_{23} + \sqrt{s_{23}^2-s_{33}s_{22}}&s_{33}
\end{pmatrix}^\top\!, \\
\nonumber
&\mathbf{n}_b = \begin{pmatrix}
s_{13} - s\sqrt{s_{13}^2-s_{33}s_{11}}&s_{23} - \sqrt{s_{23}^2-s_{33}s_{22}}&s_{33}
\end{pmatrix}^\top\!, \\
&\text{where } s = sign(s_{23}s_{13}-s_{12}s_{33}).
\end{align}
\parag{Normal Validation.} The normals thus obtained must be visible to the camera. Given the analytical normal in Eq.~\ref{eq:normal}, $\mathbf{n}_a $ and $\mathbf{n}_b $ are visible if $\frac{s_{33}}{1\!-\!uk_1\!-\!vk_2} >0$, i.e., they have a similar orientation towards the camera. We discard the normals that do not meet the visibility constraint.
\parag{Normal Selection.} Using Eq.~\ref{eq:k1k2}, the local depth derivatives $(k_1,k_2)$ at $\mathbf{X}$ are given by $k_i=\frac{n_i}{u n_1 + vn_2 + n_3}$.
From the solution in Eq.~\ref{eq:normal_sol}, we thus obtain two possible solutions for the local depth derivatives
$(k_{1a},k_{2a})$ and $(k_{1b},k_{2b})$.
We pick the normal that minimizes the corresponding sum of squares of depth derivatives. That is, we compute the normal $\mathbf{n}$ as
\begin{equation}
\label{eq:nc}
\mathbf{n} = \begin{cases}
\mathbf{n}_a & \text{ if } k_{1a}^2 + k_{2a}^2 \leq k_{1b}^2 + k_{2b}^2 \!,\\
\mathbf{n}_b & \text{ otherwise. }
\end{cases}
\end{equation}
Following Eq.~\ref{eq:homography}, $\overline{\mathbf{n}}$ is then obtained as $\mathbf{H}^\top\mathbf{n}$.
\parag{Measure of Degeneracy.} In degenerate situations, the singular values $(\sigma
_1,\sigma_2,\sigma_3)$ of $\mathbf{H} $ are all one.
We use the ratio $\frac{\sigma_1}{\sigma_3}$ to quantify the degeneracy. Thus, we only reconstruct from $\mathbf{S}$ if $\frac{\sigma_1}{\sigma_3} > \tau$, and we set $\tau = 1.05$.
\parag{Surface Reconstruction.} We consider a planar surface and bend it to match the normals obtained using the homography decomposition mentioned above, as opposed to~\cite{Parashar19a,Parashar20,Parashar21} which integrate the normals on each surface. The upside of surface bending is that it does not require to set a smoothness parameter, which needs to be tuned for the normal integration. Furthermore, surface bending is much faster than its normal integration counterpart in the presence of dense data. It is also less affected by the noise in the normals corresponding to high-perspective image regions.
\comment{\parag{Equiareality.}
We write the area-preservation constraint of Eq.~\ref{eq:mt_norm_simp} using the $\mathbf{J}_{\phi}$ and $\mathbf{J}_{\overline{\phi}}$ given in Eqs.~\ref{eq:j_phi_in_n} and \ref{eq:jphi_n_overline}. The constraint is given by
\begin{equation}
\label{eq:eqar}
\begin{matrix}
&(\mathbf{z}_1 \times \mathbf{z}_2)^\top\frac{\beta^4 \det \mathbf{g}}{\overline{\beta}^2(\det \mathbf{H})^2} \mathbf{H}\mathbf{H}^\top(\mathbf{z}_1 \times \mathbf{z}_2) \\ &=
(\mathbf{z}_1 \times \mathbf{z}_2)^\top \frac{\beta^2\det \mathbf{g}}{\overline{s}^2}(\mathbf{z}_1 \times \mathbf{z}_2)
\end{matrix}
\end{equation}
Assuming that the homography is normalized, we write $\overline{\beta} = \overline{s}\beta$ and $\mathbf{H} \sim \frac{\mathbf{H}}{\det \mathbf{H}}$. The above equation is simplified to
\begin{equation}
\label{eq:eqar_mod}
(\mathbf{z}_1 \times \mathbf{z}_2)^\top\mathbf{H}\mathbf{H}^\top(\mathbf{z}_1 \times \mathbf{z}_2) =
(\mathbf{z}_1 \times \mathbf{z}_2)^\top (\mathbf{z}_1 \times \mathbf{z}_2).
\end{equation}
Here, $\mathbf{z}_i = \mathbf{n} \times( \mathbf{h}_i \times \hat{\mathbf{x}})$. We write $\hat{\mathbf{h}_i} = \mathbf{h} \times \hat{\mathbf{x}}$, and the cross product is given by $\mathbf{z}_1 \times \mathbf{z}_2 = (\mathbf{n}\cdot (\hat{\mathbf{h}}_1 \times \hat{\mathbf{h}}_2))\mathbf{n}$. Since $\mathbf{h}_1, \mathbf{h}_2$ are linearly independent, $\mathbf{n}. (\hat{\mathbf{h}}_1 \times \hat{\mathbf{h}}_2)$ is a non-zero scalar. Therefore, the area-preservation constraint is given by
\begin{equation}
\label{eq:eqar_simp}
\mathbf{n}^\top (\mathbf{H}\mathbf{H}^\top-\mathbf{I}_{3\times 3})\mathbf{n} = 0.
\end{equation}
This constraint is a consequence to isometric/conformal deformations in Eq.~\ref{eq:final}. Given that $\mathbf{H}$ is non-orthogonal both $\mathbf{n}_a,\mathbf{n}_b$ in Eq.~\ref{eq:normal_sol} satisfy this relation.
\parag{Diffeomorphism.} In a generic case of deformation where no metric quantity is preserved, we can assume that the areas are related by a scalar $\alpha$. Therefore, the area-preservation constraint in Eq.~\ref{eq:eqar_simp} can be re-written as
\begin{equation}
\label{eq:eqar_simp2}
\mathbf{n}^\top (\mathbf{H}\mathbf{H}^\top-\alpha\mathbf{I}_{3\times 3})\mathbf{n} = 0.
\end{equation}
Since $\mathbf{H} \sim \alpha \mathbf{H}$, this equation is also a consequence to our isometric/conformal NRSfM solution.
To summarise, a local closed-form solution to surface normals can be obtained by solving Eq.~\ref{eq:final}. There may be more solutions to generic deformation constraints that satisfy Eq.~\ref{eq:eqar_simp2}, the surface normals obtained by solving Eq.~\ref{eq:final}
is one of them.}
\section{Normals from Multiple Images}
Methods such as those of~\cite{Parashar19a,Parashar20,Parashar21} pick a reference image and formulate reconstruction constraints between it and the other images, which are then solved by solving a least-squares problem over the entire set of images. We use the same strategy, except that we reconstruct from image pairs, one of them being the reference image. Therefore, for $N$ images, we obtain $N-1$ estimates for the reference image and $1$ estimate for each of the non-reference images.
More formally, let $\{\mathbf{x} ^i_j\}, i\in [1,M], j\in[1,N]$, be a set of $N$ point correspondences between $M$ images. Our goal is to find the 3D point $\mathbf{X}^i_j$ and the normal $\mathbf{n}^i_j$ corresponding to each $\mathbf{x} ^i_j$. Using Eq.~\ref{eq:CS_norm}, we write the local homography for each point correspondence $\mathbf{H}^{ik}_j$ between image pairs $(i,k) \in [1,M], i \neq k$, using the warp $\eta$. Each local homography $\mathbf{H}^{ik}_j$ is normalized by dividing it by its second singular value. We compute $\mathbf{Hc}^{ik}_j$ given by the ratio of the first and third singular value, and the normals for each local homography $\mathbf{H}^{ik}_j$ using Eq.~\ref{eq:normal_sol}. We then pick a unique solution using Eq.~\ref{eq:nc}. The solution on the reference and non-reference image is given by $\mathbf{n}_j^{kk}$ and $\mathbf{n}_j^{ki}$, respectively. For non-degenerate cases, where $\frac{\sigma_1}{\sigma_3} \geq 1.05$, we compute the normal ${\bf n}^i_j$ by taking the median of the $\mathbf{n}_j^{ik}$s computed over $k$ reference images. We obtain a 3D surface by bending a planar surface to match the obtained normals on each surface.
We summarize our complete pipeline in Algorithm~\ref{alg:diff}.
\begin{algorithm}
\SetAlgoLined
\KwData{$\mathbf{x}^i_j, \mathbf{H}^{ik}_j$ and $\mathbf{Hc}^{ik}_j$}
\KwResult{$\mathbf{n}^i_j$
$\frac{\sigma_1}{\sigma_3} = 1.05$\;
\For{ each reference image $k = [1,M]$}{
\For{ each point $j= [1,N]$}{
\For{ images $i= [1,M], i \neq k$}{
\eIf{$\mathbf{Hc}^{ik}_j > \frac{\sigma_1}{\sigma_3}$}{
Compute normals using~\eqref{eq:normal_sol}\;
Pick a solution $\mathbf{n}_j^{kk}$ using~\eqref{eq:nc}\;
Write $\mathbf{n}_j^{ik} = (\mathbf{Hc}^{ik}_j)^\top \mathbf{n}_j^{kk}$\;
}
{ Set $\mathbf{n}_j^{ik}, \mathbf{n}_j^{kk}$ to zero\;
}
}
}
}
\For{ each point $j= [1,N]$}{
\For{ images $i= [1,M]$}{
Obtain $\mathbf{n}^i_j$ by as the median of the non-zero $\mathbf{n}^{ik}_j$s\;
}
}
\caption{Our NRSfM Algorithm}
\label{alg:diff}
\end{algorithm}
\section{Formalism and Assumptions}
\label{sec:formalism}
\input{fig/model}
At the heart of our approach is the fact that the normals at two different instants at a point on a deforming 3D surface can be computed given the point's projections in two images and a 2D warp between these images, under the sole assumptions of local surface planarity and deformation local linearity. In this section, we first introduce the NRSfM setup we will use in the rest of this paper, which is similar to the one of~\cite{Parashar19a}. We then explain what our assumptions mean and why they are widely applicable. Finally, we formulate the constraints we will use for reconstruction purposes.
\subsection{Setup}
Fig.~\ref{fig:model} depicts our setup when using only two images, $\mathcal{I}$ and $\overline{\mathcal{I}}$, acquired by a calibrated camera. In each one, we denote the deforming surface as $\mathcal{S}$ and $\overline{\mathcal{S}}$, respectively, and model it in terms of functions $\phi , \overline{\phi}: \mathbb{R}^2 \to \mathbb{R}^3$ that associate an image point to a surface point. Let us assume that we are given an image registration function $\eta: \mathbb{R}^2 \to \mathbb{R}^2$ that associates points in the first image to points in the second.
This is often referred to as a warp. In practice, it can be computed using standard image matching techniques, such as optical flow~\cite{Sundaram10,Sun18a} or SIFT~\cite{Lowe04}. These functions can be composed to create a mapping $\psi: \mathbb{R}^3 \to \mathbb{R}^3$ between 3D surface points seen in the two images. We use a parametric representation of $\eta$ and $\phi$ with B-splines~\cite{Bookstein89}, which allows us to accurately obtain first- and second-order derivatives of these functions. A finite-difference approach could also be used.
Given a point $\mathbf{x} = (u,v)$ on $ \mathcal{I}$ and its corresponding 3D point $\mathbf{X} = \phi(\mathbf{x})$ on $ \mathcal{S}$, we write $\phi(\mathbf{x})=\frac{1}{\beta(u,v)}\begin{pmatrix}
u & v & 1 \end{pmatrix}^\top$, where $\beta$ represents the inverse of the depth.
The Jacobian of $\phi$ is given by
\begin{equation}
\label{eq:j_phi}
\mathbf{J}_\phi = \frac{1}{\beta(u,v)}\begin{pmatrix}
1- uk_1 & -uk_2 \\ -vk_1 & 1-vk_2 \\ -k_1 & -k_2
\end{pmatrix},
\end{equation}
where $k_1=\frac{\partial_u \beta}{\beta}$ and $k_2=\frac{\partial_v \beta}{\beta}$. $\overline{u}$, $\overline{v}$, $\overline{\phi}$, $\overline{k_1}$, and $\overline{k_2}$ are defined similarly in $\overline{\mathcal{I}}$.
\subsection{Local Planarity and Linearity}
In this work, we assume local planarity of the 3D surfaces and local linearity of the deformations as described in~\cite{Lee97,Kock10}. We now describe these two assumptions and argue that they are weak ones that are generally applicable.
\parag{Surface Local Planarity.}
Let $\mathbf{x}_0$ be an image point with surface normal $\mathbf{n}$ at $\phi(\mathbf{x}_0)$. All points $\mathbf{x} = (u,v)$ sufficiently close to $\mathbf{x}_0$ can be accurately described as lying on the tangent plane. Hence, they satisfy $\mathbf{n}^\top\phi(\mathbf{x})+d=0$, where $d$ is a scalar, which we can rewrite as $\beta = -\frac{\mathbf{n}^\top}{d}\begin{pmatrix} u & v & 1 \end{pmatrix}^\top$. Therefore the inverse depth $\beta$ that appears in Eq.~\ref{eq:j_phi} is a linear function of $\mathbf{x}$ even though $\phi$ is not. Nevertheless, all higher-order derivatives of $\phi$ can be expressed in terms of $\beta$ and its first-order derivatives. This is widely viewed as a weak assumption that applies to most smooth manifolds~\cite{Lee97}. For example, our planet is a sphere that can be treated as locally planar.
\parag{Deformation Local Linearity.}
According to~\cite{Kock10}, every non-linear function can be approximated with an infinite number of linear functions. This assumption has been successfully used in shape-matching~\cite{Ovsjanikov12}. We assume the deformation $\psi$ that relates locally two planes to be smooth enough to be well described locally by its first-order approximation, so that we can ignore its second derivatives. In other words, we use a first-order approximation for the local deformations but a second-order one for the surface depth to allow for globally non-planar shapes. This is a looser set of assumptions than what is normally used in NRSfM. For example,~\cite{Lee16,Dai14} and other low-rank methods assume the deformation space to be small; physics-based methods that use inextensibility~\cite{Chhatkuli17,Vicente12} or piecewise-rigidity~\cite{Varol09,Taylor10a} make a much stronger assumption.
Under the assumption of local planarity, we have $\mathbf{X}$ and $\overline{\mathbf{X}}$ lying on a planar surface. A generic transformation between these two surfaces, which defines the deformation $\psi$, can be expressed as $\overline{\mathbf{X}}= \mathbf{S}\mathbb{R}\mathbf{X} + \mathbf{T}$, where $\mathbb{R}$ and $\mathbf{T}$ are rotation and translation and $\mathbf{S}$ is a scaling matrix. If $\mathbf{S}$ happens to be a purely diagonal matrix with equal entries, $\psi$ is a planar homography, and the resulting deformation is purely isometric or conformal. Nevertheless, $\psi$ is linear. Therefore, local planarity of surfaces implies local linearity of deformations. However, the reverse is not true.
\subsection{Differential Constraints across Images}
\label{sec:diffConstraints}
To express constraints between quantities computed in $\mathcal{I}$ and $\overline{\mathcal{I}}$, we define {\it metric tensors} and {\it connections} as described in~\cite{Lee97}.
\parag{Metric Tensors.}
The metric tensors $\mathbf{g}$ in $\mathcal{I}$ and $\overline{\mathbf{g}}$ in $\overline{\mathcal{I}}$ are first-order differential quantities that capture local distances and angles. They can be written as
\begin{equation}
\mathbf{g}=\mathbf{J}_\phi^\top\mathbf{J}_\phi \text{ and } \overline{\mathbf{g}}=\mathbf{J}_{\overline{\phi}}^\top \mathbf{J}_{\overline{\phi}} \;,
\label{eq:nrsfm}
\end{equation}
where $\mathbf{J}_\phi $ and $\mathbf{J}_{\overline{\phi}} $ are local surface jacobians computed according to Eq.~\ref{eq:j_phi}.
These tensors can be used to impose isometry, conformality, and equiareality constraints by forcing the scalars $k_1$ and $k_2$ of Eq.~\ref{eq:j_phi} to satisfy one of the three conditions below:
\begin{equation}
\label{eq:iso-nrsfm}
\begin{cases}
\overline{\mathbf{g}}\!=\!\mathbf{J}_\eta^\top \mathbf{g}\mathbf{J}_\eta,\!&\!\text{Isometry}
\\
\overline{\mathbf{g}}\!=\!\lambda^2\mathbf{J}_\eta^\top \mathbf{g}\mathbf{J}_\eta, \! \lambda^2\!\in\!\mathbb{R}^{+}\! -\! \{1\},\!&\!\text{Conformality}
\\
\sqrt{\det(\overline{\mathbf{g}})}\!=\!\sqrt{\det(\mathbf{J}_\eta^\top \mathbf{g}\mathbf{J}_\eta)},\!&\!\text{Equiareality}
\end{cases}
\end{equation}
where $\mathbf{J}_\eta$ is the Jacobian of the warp $\eta$.
\parag{Linear Relation between Surface Derivatives.}
Given $\mathbf{J}_\phi $, a local reference frame on the surfaces can be expressed with the column vectors as tangents and their cross product as normal. Connections are second-order differential quantities that express the rate of change of this local frame.
{Using connections under the assumption of local linearity as stated above, it can be shown~\cite{Parashar19a} that
\begin{equation}
\label{eq:CS_rel}
\begin{pmatrix}
\overline{k}_1 \\ \overline{k}_2
\end{pmatrix} = \mathbf{J}_\eta^\top\begin{pmatrix}
k_1 \\ k_2
\end{pmatrix} -\begin{pmatrix}
0 & 1 \\ 1 & 0
\end{pmatrix}\mathbf{J}_\eta^{-1}\dfrac{\partial^2 \eta}{ \partial \overline{u}\overline{v}} ,
\end{equation}
where $\dfrac{\partial^2 \eta}{ \partial \overline{u}\overline{v}} $ are the second-order derivatives of the warp.
Solutions to isometric, conformal and equiareal NRSfM can be obtained by solving the metric tensor preservation equations in Eq.~\ref{eq:iso-nrsfm} under the constraints of Eq.~\ref{eq:CS_rel}
\subsection{NRSfM Formalism}
Recall from Section~\ref{sec:outline} and Fig.~\ref{fig:model} that we are given an image registration function $\eta$ that associates points in the first image $\mathcal{I}$ to points in the second $\overline{\mathcal{I}}$ and we want to infer from it the functions $\phi , \overline{\phi}: \mathbb{R}^2 \to \mathbb{R}^3$ that associate a surface point to an image point.
Given a point $\mathbf{x} = (u,v)$ on $ \mathcal{I}$ and its corresponding 3D point $\mathbf{X} = \phi(\mathbf{x})$ on $ \mathcal{S}$, we write $\phi(\mathbf{x})=\frac{1}{\beta(u,v)}\begin{pmatrix}
u & v & 1 \end{pmatrix}^\top$, where $\beta$ is a linear function that represents the inverse of depth. \PF{Why is $\beta$ linear?} The Jacobian of $\phi$ is given by
\begin{equation}
\label{eq:j_phi}
\mathbf{J}_\phi = \frac{1}{\beta(u,v)}\begin{pmatrix}
1- uk_1 & -uk_2 \\ -vk_1 & 1-vk_2 \\ -k_1 & -k_2
\end{pmatrix},
\end{equation}
where $k_1=\frac{\partial_u \beta}{\beta}$ and $k_2=\frac{\partial_v \beta}{\beta}\right)$. $\overline{u}$, $\overline{v}$, $\overline{\phi}$, $\overline{k_1}$, and $\overline{k_2}$ are defined similarly in $\overline{\mathcal{I}}$.
\parag{Metric Tensors.}
The metric tensors $\mathbf{g}$ in $ \mathcal{I}$ and $\overline{\mathbf{g}}$ in $\overline{\mathcal{I}}$ are first-order differential quantity that capture local distances and angles. They can be written as
\begin{equation}
\mathbf{g}=\mathbf{J}_\phi^\top\mathbf{J}_\phi \text{ and } \overline{\mathbf{g}}=\mathbf{J}_{\overline{\phi}}^\top \mathbf{J}_{\overline{\phi}} \; .
\end{equation}
These tensors can be used to imposed to impose isometric, conformal, and equiareal constraints by forcing the scalars $k_1$ and $k_2$ of Eq.~\ref{eq:j_phi} to satisfy one of the three conditions below:
\begin{equation}
\label{eq:iso-nrsfm}
\begin{cases}
\overline{\mathbf{g}}\!=\!\mathbf{J}_\eta^\top \mathbf{g}\mathbf{J}_\eta,\!&\!\text{Isometry}
\\
\overline{\mathbf{g}}\!=\!\lambda^2\mathbf{J}_\eta^\top \mathbf{g}\mathbf{J}_\eta, \! \lambda^2\!\in\!\mathbb{R}^{+}\! -\! {1},\!&\!\text{Conformality}
\\
\det(\overline{\mathbf{g}})\!=\!\det(\mathbf{J}_\eta^\top \mathbf{g}\mathbf{J}_\eta),\!&\!\text{Equiareality}
\end{cases}
\end{equation}
\parag{Connections.}
Connections are second-order differential quantities that express the rate of change of the metric tensors. \pf{Under the assumption of local linearity, it can be shown~\cite{Parashar19a}} that
\begin{equation}
\label{eq:CS_rel}
\begin{pmatrix}
\overline{k}_1 \\ \overline{k}_2
\end{pmatrix} = \mathbf{J}_\eta^\top\begin{pmatrix}
k_1 \\ k_2
\end{pmatrix} -\begin{pmatrix}
0 & 1 \\ 1 & 0
\end{pmatrix}\mathbf{J}_\eta^{-1}\dfrac{\partial^2 \eta}{ \partial \overline{u}\overline{v}} .
\end{equation}
The solutions to isometric, conformal and equiareal NRSfM are obtained by solving the metric tensor preservation equations of Eq.~\eqref{eq:iso-nrsfm} under the constraint of Eq.~\eqref{eq:CS_rel}.
\section{Conclusion}
We have proposed an approach to NRSfM that can estimate normals from image pairs given a 2D warp and point correspondences between the two images. It does so in closed form from individual correspondences and is therefore fast. Furthermore, it can estimate if these normals are reliable given the motion from one image to the next. When they are found to be, our experiments show that they are indeed very accurate. As a result, our method performs well with various deformation types and can reconstruct large and small deformations at a low computational cost.
Our next step will be to remove the dependency on expensive methods to compute warps and integrate normals so that a truly real-time application can be developed.
\section{Computing Normals from Two Images}
In earlier approaches the NRSfM problem was addressed by solving the system of equations of~\ref{eq:nrsfm} under the isometric, conformal, and equiareal constraints of Eq.~\ref{eq:iso-nrsfm} with respect the variables $k_1$ and $k_2$ of Eq.~\ref{eq:j_phi}. In our approach, we solve the system of equations directly in terms of the surface normals. We do this in closed form while simultaneously identifying degenerate conditions that result in unreliable estimates.
\parag{Differentiating the Warp.}
Let us consider a point $\overline{\mathbf{x}} = (\overline{u},\overline{v})^T$ in $\overline{\mathcal{I}}$ and its corresponding point $(u,v)^T=\eta(\overline{u},\overline{v})$ in $\mathcal{I}$. Assuming the surfaces to be infinitesimally planar means that there is a $3 \times 3$ homography matrix $\mathbf{H} = [h_{ij}]_{1 \leq i,j \leq 3}$ relating $\overline{\mathbf{x}}$ and $\mathbf{x}$. Assuming the homography to be normalized, that is, its second singular value to be 1, we write
\begin{equation}
\label{eq:homography}
\!\begin{pmatrix}
u \\ v \\ 1
\end{pmatrix}\!=\!\frac{1}{\overline{s}}\!\begin{pmatrix}
h_{11} \!& \!h_{12} \!&\! h_{13} \\
h_{21} \!&\! h_{22} \!&\! h_{23} \\
h_{31} \!&\! h_{32} \!&\! h_{33}
\end{pmatrix}\!\begin{pmatrix}
\overline{u} \\ \overline{v} \\ 1
\end{pmatrix}, \;
\end{equation}
where $\overline{s}=h_{31}\overline{u}\!+\!h_{32}\overline{v}\!+\!h_{33}$. Thus the first and second-order derivatives of $\eta$ can be computed as
\begin{align}
\label{eq:eta_der}
\begin{pmatrix}
\dfrac{\partial \eta}{\partial \overline{u}} & \dfrac{\partial \eta}{\partial \overline{v}}
\end{pmatrix}&=\frac{1}{\overline{s}}
\begin{pmatrix}
h_{11}\!-\!h_{31}u \!&\! h_{12}\!-\!h_{32}u \\
h_{21}\!-\!h_{31}v \!&\! h_{22}\!-\!h_{32}v
\end{pmatrix} = \mathbf{J}_\eta \; , \nonumber \\[1mm]
\begin{pmatrix}
\dfrac{\partial^2 \eta}{ \partial \overline{u}^2} & \dfrac{\partial^2 \eta}{ \partial \overline{u}\overline{v}} & \dfrac{\partial^2 \eta}{ \partial \overline{v}^2}
\end{pmatrix}
&= -\frac{1}{\overline{s}} \mathbf{J}_\eta\begin{pmatrix}
2h_{31} & h_{32} & 0 \\ 0 & h_{31} &2h_{32}
\end{pmatrix}.
\end{align}
\parag{Image Embedding and Local Normal.}
The unit normal $\mathbf{n}$ at $\mathbf{x}$ is the cross product of the columns of the matrix $\mathbf{J}_\phi$ from~Eq.~\ref{eq:j_phi}. We write
\begin{align}
\label{eq:normal}
\mathbf{n} & =\! \frac{1}{\beta^2\sqrt{\det \mathbf{g}}}\!\begin{pmatrix}
k_1 \\k_2 \\ 1\!-\!uk_1\!-\!vk_2
\end{pmatrix} \\
&=\!\frac{1}{\beta^2\sqrt{\det \mathbf{g}}}\!\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ -\mathbf{x}^\top &1
\end{pmatrix}\!\begin{pmatrix}
k_1 \\ k_2 \\ 1
\end{pmatrix}\!. \nonumber \\
\Rightarrow
\begin{pmatrix}
k_1 \\ k_2 \\ 1
\end{pmatrix} & =\!\beta^2\sqrt{\det \mathbf{g}}\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ \mathbf{x}^\top &1
\end{pmatrix}\mathbf{n}. \label{eq:k1k2}
\end{align}
Given the normal $\mathbf{n}$ as written in Eq.~\ref{eq:normal}, we re-write the matrix $\mathbf{J}_\phi$ of Eq.~\ref{eq:j_phi} as
\begin{align}
\label{eq:j_phi_in_n}
\nonumber
\mathbf{J}_\phi \!&\!=\! \beta\sqrt{\det \mathbf{g}}\!\begin{pmatrix}
0 & uk_1\!+\!vk_2\!-\!1 & k_2 \\
1\!-\!uk_1\!-\!vk_2 & 0 & \!-\!k_1 \\
\!-\!k_2 & k_1 & 0
\end{pmatrix}\!
\begin{pmatrix}
0 & 1 \\ \!-\!1 & 0 \\ v & \!-\!u
\end{pmatrix} \\
\!&\!=\! \beta\sqrt{\det \mathbf{g}}[\mathbf{n}]_{\times}\mathbf{E}.
\end{align}
We can now rewrite the differential constraints across images introduced in Section~\ref{sec:diffConstraints} in terms of the normals.
\parag{Connections.}
Given the $\eta$ derivatives~Eq.\ref{eq:eta_der}, the connections transfer relation of Eq.~\ref{eq:CS_rel} becomes
\begin{align}
\label{eq:CS_law}
\begin{pmatrix}
\overline{k}_1 \\ \overline{k}_2
\end{pmatrix}&=\!\mathbf{J}_\eta^\top\!\begin{pmatrix}
k_1 \\ k_2
\end{pmatrix}\!+\!\dfrac{1}{\overline{s}}\!\begin{pmatrix}
h_{31} \\h_{32}
\end{pmatrix} \; , \nonumber \\
\Rightarrow
\begin{pmatrix}
\overline{k}_1 \\ \overline{k}_2 \\ 1
\end{pmatrix}&=\!\begin{pmatrix}
\mathbf{J}_\eta^\top & \mathbf{m} \\ 0 & 1 \end{pmatrix}\!\begin{pmatrix}
k_1 \\k_2 \\1
\end{pmatrix}\!.
\end{align}
Using Eq.~\ref{eq:k1k2}, we re-write the above expression as
\begin{align}
\label{eq:CS_norm}
\nonumber
\overline{\mathbf{n}}\!&\!=\!\frac{\beta^2}{\overline{\beta}^2}\!\sqrt{\frac{\det \mathbf{g}}{\det \overline{\mathbf{g}}}}\mathbf{T}\mathbf{n}\\
\nonumber \!&\!=\!\frac{\beta^2}{\overline{\beta}^2}\!\sqrt{\frac{\det \mathbf{g}}{\det \overline{\mathbf{g}}}}\!\begin{pmatrix}
\mathbf{I}_{2\times 2} & 0 \\ -\overline{\mathbf{x}}^\top &1
\end{pmatrix}\!\begin{pmatrix}
\!\mathbf{J}_\eta^\top\!&\!\mathbf{m}\!\\ 0 & 1 \end{pmatrix}\!\begin{pmatrix}
\!\mathbf{I}_{2\times 2}\!&\!0\!\\ \mathbf{x}^\top &1
\end{pmatrix}\!\mathbf{n}
\\ \nonumber
\!&\!=\!\dfrac{ \beta^2}{\overline{s}\overline{\beta}^2}\!\sqrt{\frac{\det\!\mathbf{g}}{\det\!\overline{\mathbf{g}}}}\!\begin{pmatrix}
\!\mathbf{I}_{2\times 2}\!&\!0\!\\\!-\!\overline{\mathbf{x}}^\top\!&\!1\!
\end{pmatrix}\!\begin{pmatrix}
\!h_{11}\!&\!h_{21}\!&\!h_{31}\!\\
\!h_{12}\!&\!h_{22}\!&\!h_{32}\!\\
\!\overline{s}u\!&\!\overline{s}v\!&\!\overline{s}\!
\end{pmatrix}\!\mathbf{n}\! \\ &\!=\!\dfrac{ \beta^2}{\overline{s}\overline{\beta}^2}\sqrt{\frac{\det\!\mathbf{g}}{\det\!\overline{\mathbf{g}}}}\!\mathbf{H}^\top\!\mathbf{n} \; ,
\end{align}
which directly relates the two normals.
\parag{Metric Tensor.}
As can be seen in Fig.~\ref{fig:model}, we can write $\overline{\phi} = \psi \circ \phi \circ \eta$. Differentiating this expression and multiplying it by its transpose yields
\begin{equation}
\label{eq:mt_gen}
\overline{\mathbf{g}}=\mathbf{J}_{\overline{\phi}}^\top\mathbf{J}_{\overline{\phi}} = \mathbf{J}_\eta^\top\mathbf{J}_\phi^\top\mathbf{J}_\psi^\top\mathbf{J}_{\psi}\mathbf{J}_\phi\mathbf{J}_\eta.
\end{equation}
For isometric, conformal and equiareal deformations, the matrix $\mathbf{J}_\psi^\top\mathbf{J}_\psi$ is given by $\mathbf{I}_{3\times3}$, $\lambda^2\mathbf{I}_{3\times3}$ and $diag(\lambda^2_1, \lambda^2_2,(\lambda_1\lambda_2)^{-2})$, where $\lambda_1^2,\lambda_2^2 \in \mathbb{R}^{+}$, respectively which leads to the metric preservation relations of Eq.~\ref{eq:iso-nrsfm}.
When the deformation is diffeomorphic, $\mathbf{J}_\psi$ has 9 degrees of freedom. Using SVD, we write $\mathbf{J}_\psi = \mathbf{U} \Sigma \mathbf{V}^\top$, where $\mathbf{U}, \mathbf{V}$ are two orthogonal matrices and $\Sigma$ is a diagonal matrix. Therefore, $\mathbf{J}_\psi^\top\mathbf{J}_\psi = \mathbf{V}\Sigma^\top\Sigma\mathbf{V}^\top$. Since $\mathbf{V}$ is orthogonal, we can simplify Eq.~\eqref{eq:mt_gen} as
\begin{equation}
\label{eq:mt_gen_simp}
\overline{\mathbf{g}} = trace(\Sigma^\top\Sigma)\mathbf{J}_\eta^\top \mathbf{g} \mathbf{J}_\eta = \sigma^2\mathbf{J}_\eta^\top \mathbf{g} \mathbf{J}_\eta.
\end{equation}
Thus, the different conditions of Eq.~\ref{eq:iso-nrsfm} combined into one single diffeomorphic constraint.
We now use Eq~\eqref{eq:j_phi_in_n} to rewrite the metric preservation relation of Eq.~\eqref{eq:mt_gen_simp} as
\begin{small}
\begin{align}
\label{eq:mt_norm}
\!\overline{\beta}^2\!\det\!(\overline{\mathbf{g}})\overline{\mathbf{E}}^\top\![\overline{\mathbf{n}}]_\times^\top\![\overline{\mathbf{n}}]_\times\!\overline{\mathbf{E}}\!=\!\sigma^2\!\beta^2\!\det\!(\mathbf{g})\!\mathbf{J}_\eta^\top\!\mathbf{E}^\top\![\mathbf{n}]_\times^\top\![\mathbf{n}]_\times\!\mathbf{E}\!\mathbf{J}_\eta.
\end{align}
\end{small}
Given the $\eta$ derivatives of Eq-\eqref{eq:eta_der}, we simplify $\mathbf{E}\mathbf{J}_\eta$ to $\frac{1}{\overline{s}}\begin{pmatrix}
\mathbf{h}_1 \times \hat{\mathbf{x}} & \mathbf{h}_2 \times \hat{\mathbf{x}}
\end{pmatrix}$ where $\mathbf{h}_1, \mathbf{h}_2$ are the two columns of $\mathbf{H}$~\eqref{eq:homography}. By writing $\mathbf{z}_1 = \mathbf{n} \times (\mathbf{h}_1 \times \hat{\mathbf{x}})$ and $\mathbf{z}_2 = \mathbf{n} \times (\mathbf{h}_2 \times \hat{\mathbf{x}})$, Eq.~\eqref{eq:mt_norm} reduces to
\begin{small}
\begin{equation}
\label{eq:mt_norm_simp}
\overline{\beta}^2\det (\overline{\mathbf{g}})\overline{\mathbf{E}}^\top[\overline{\mathbf{n}}]_\times^\top[\overline{\mathbf{n}}]_\times\overline{\mathbf{E}} \! = \! \frac{\sigma^2\beta^2\det (\mathbf{g})}{\overline{s}^2}\begin{pmatrix}
\mathbf{z}_1^\top \mathbf{z}_1 \!\! & \!\! \mathbf{z}_1^\top \mathbf{z}_2 \\ \mathbf{z}_1^\top \mathbf{z}_2 \!\! &\!\! \mathbf{z}_2^\top \mathbf{z}_2
\end{pmatrix} \! .
\end{equation}
\end{small}
\parag{Diffeomorphic NRSfM.}
\pf{ So far, we have expressed the metric preservation conditions in terms of the normals of the two surfaces under consideration. The only unknown left in the system is therefore, $\mathbf{n}$. Hence, we now derive a closed-form solution} to diffeomorphic NRSfM by solving these equations in a closed-form. }
Given the multiplication property of the cross product, the constraints on the normals of Eq.~\ref{eq:CS_norm} imply that
\begin{equation}
[\overline{\mathbf{n}}]_\times=\dfrac{\gamma \beta^2}{\overline{s}\overline{\beta}^2}\sqrt{\frac{\det\mathbf{g}}{\det\overline{\mathbf{g}}}}\det (\mathbf{H}^\top) \mathbf{H}^{-1}[\mathbf{n}]_\times \mathbf{H}^{-\top} \; ,
\end{equation}
which lets us rewrite as the matrix $\mathbf{J}_{\overline{\phi}}$ of Eq.~\eqref{eq:j_phi_in_n} as
\begin{align}
\label{eq:jphi_n_overline}
\nonumber
\mathbf{J}_{\overline{\phi}}&\!=\!\frac{ \beta^2}{\overline{\beta}}\sqrt{\det \mathbf{g}} \mathbf{H}^{-1}\![\mathbf{n}]_\times\!\left(\!\frac{\det \mathbf{H}^\top}{\overline{s}}\!\mathbf{H}^{-\top}\overline{\mathbf{E}}\!\right)
\\
&\!=\!\frac{\ \beta^2\!\sqrt{\det \mathbf{g}}}{\overline{\beta}}\!\mathbf{H}^{\!-1}\![\mathbf{n}]_\times\!\begin{pmatrix}
\!\mathbf{h}_1\!\times\!\hat{\mathbf{x}}\!&\!\mathbf{h}_2\!\times\!\hat{\mathbf{x}}\!\end{pmatrix} \\ \nonumber
&\!=\!\frac{ \beta^2\!\sqrt{\det \mathbf{g}}}{\overline{\beta}}\!\mathbf{H}^{\!-1}\!\begin{pmatrix}
\!\mathbf{z}_1\!&\!\mathbf{z}_2\!
\end{pmatrix}\!.
\end{align}
Injecting this expression into the metric tensor preservation relation of Eq.~\ref{eq:mt_norm_simp} yields
\begin{small}
\begin{align}
\label{eq:recon_eqn}
\begin{pmatrix}
\mathbf{z}_1^\top\mathbf{H}^{\!-\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_1\!&\!\mathbf{z}_1^\top\mathbf{H}^{\!-\!\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_2\! \\ \mathbf{z}_1^\top\mathbf{H}^{\!-\!\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_2\!&\!\mathbf{z}_2^\top\mathbf{H}^{\!-\!\top}\!\mathbf{H}^{\!-\!1} \mathbf{z}_2\end{pmatrix}&=\!\frac{\sigma^2 \overline{\beta}^2}{\overline{s}^2\beta^2}\!\begin{pmatrix}
\mathbf{z}_1^\top \mathbf{z}_1 \!\!& \!\!\mathbf{z}_1^\top \mathbf{z}_2 \\ \mathbf{z}_1^\top \mathbf{z}_2 \!\!& \!\! \mathbf{z}_2^\top \mathbf{z}_2
\end{pmatrix}\!, \nonumber \\
\Rightarrow \mathbf{z}_i^\top \left(\overline{\mathbf{H}}^\top\overline{\mathbf{H}}-\frac{\sigma^2 \overline{\beta}^2}{\overline{s}^2\beta^2}\mathbf{I}_{3\times 3}\right)\mathbf{z}_j & = 0, \forall i,j \in [1,2],
\end{align}
\end{small}
where $\overline{\mathbf{H}}=\mathbf{H}^{-1}$. Assuming $\mathbf{H}$ to be normalized, that is, its second singular value to be 1, the relation between 3D points is given by $\phi(\mathbf{x}) = \mathbf{H} \phi(\overline{\mathbf{x}}) $. Using Eq.~\ref{eq:homography} yields $\overline{\beta} = \beta \overline{s}$. By writing $\mathbf{z}_i = [\mathbf{n}]_{\times}[\mathbf{h}_i]_\times \hat{\mathbf{x}}$, the above constraints further simplify to
\begin{equation}
\label{eq:recon_H}
[\mathbf{n}]_\times^\top (\overline{\mathbf{H}}^\top\overline{\mathbf{H}}-\sigma^2\mathbf{I}_{3\times 3})[\mathbf{n}]_\times = 0.
\end{equation}
Since $\alpha \mathbf{H}$ is equivalent to $\mathbf{H}$, we divide the above expression by $\sigma^2$ and write $\frac{1}{\sigma}\overline{\mathbf{H}}$ as $\overline{\mathbf{H}}$. This simplifies the above expressions to
\begin{equation}
\label{eq:final}
[\mathbf{n}]_\times^\top (\overline{\mathbf{H}}^\top\overline{\mathbf{H}}-\mathbf{I}_{3\times 3})[\mathbf{n}]_\times =[\mathbf{n}]_\times^\top \mathbf{S}[\mathbf{n}]_\times = 0.
\end{equation}
\parag{Degenerate cases.} The system of Eq.~\ref{eq:recon_H} holds as long as $\mathbf{S}$ is a non-null matrix, which means $\overline{\mathbf{H}}^\top \overline{\mathbf{H}} \neq \mathbf{I}_{3\times3}$. Since $\overline{\mathbf{H}}$ is a planar homography, we write $\overline{\mathbf{H}}=\mathbf{R}
+\frac{\mathbf{t}\mathbf{n}^\top}{d}$ where $\mathbf{R}$ rotation and $\mathbf{t}$ is translation. In case the relative motion between camera and object is purely rotational, $\overline{\mathbf{H}}^\top \overline{
\mathbf{H}} = \mathbf{I}_{3\times3}$ and the system~\eqref{eq:recon_H} does not hold. Another possibility is the mirror reflection about the plane, i.e. $\mathbf{t} = -2\mathbf{R}\mathbf{n}$ with $||\mathbf{t}||=2$ which is highly unlikely.
\parag{Affine stability.} Under affine imaging conditions, $h_{31}=h_{32}=0$ and $h_{33}=1$.
By making these changes, we see that $\mathbf{z}_i$ and $\mathbf{S}$ remain non-null. Therefore, the system~\eqref{eq:recon_H} holds and we can obtain the solution to normals.
\parag{Solution.} A solution to this system is obtained by homography decomposition~\cite{Malis07}. We give an overview of the solution here. We recommend reading~\cite{Malis07} for more details.
$\mathbf{S}={s_{ij}}$ is a symmetric matrix expressed in terms of $\mathbf{H}$ which can be numerically computed using $\eta$ and image observations $(\mathbf{x},\overline{\mathbf{x}})$. $\mathbf{T}$ in equation~\eqref{eq:CS_norm} shows a closed-form definition of $\mathbf{H}$. We write $\mathbf{n} = \begin{pmatrix}
n_1 & n_2 & n_3
\end{pmatrix}^\top$. Since $n_3 \neq 0$, we write $y_1 = \frac{n_1}{n_3}$ and $y_2 = \frac{n_2}{n_3}$ and expand the system~\eqref{eq:recon_H}. We obtain 6 constraints out of which, only 3 are unique. They are given by
\begin{align}
\label{eq:quad_recon}
\nonumber s_{33} y_2^2 - 2s_{23} y_2 + s_{22} &= 0 \\
\nonumber s_{33} y_1^2 - 2s_{13} y_1 + s_{11} &= 0 \\
s_{22} y_1^2 - 2s_{12} y_1 y_2 + s_{11} y_2^2 &= 0
\end{align}
By solving the first two quadratics, we obtain $y_1 = \dfrac{s_{13} \pm \sqrt{s_{13}^2-s_{33}s_{11}}}{s_{33}}$ and $y_2 =\dfrac{s_{23} \pm \sqrt{s_{23}^2-s_{33}s_{22}}}{s_{33}}$. We use the third expression to diambiguate the solutions. Thus, a closed form expression for the possible pair of normals is given by
\begin{align}
\label{eq:normal_sol}
\nonumber
&\mathbf{n}_a = \begin{pmatrix}
s_{13} + s\sqrt{s_{13}^2-s_{33}s_{11}}& s_{23} + \sqrt{s_{23}^2-s_{33}s_{22}}&s_{33}
\end{pmatrix}^\top\!, \\
\nonumber
&\mathbf{n}_b = \begin{pmatrix}
s_{13} - s\sqrt{s_{13}^2-s_{33}s_{11}}&s_{23} - \sqrt{s_{23}^2-s_{33}s_{22}}&s_{33}
\end{pmatrix}^\top\!, \\
&\text{where } s = sign(s_{23}s_{13}-s_{12}s_{33}).
\end{align}
\parag{Normal selection.} Using equation~\eqref{eq:k1k2}, the local depth derivatives $(k_1,k_2)$ at $\mathbf{X}$ are given by $\left(\frac{n_1}{n_3},\frac{n_2}{n_3}\right)$, $k_i=\frac{n_i}{u n_1 + vn_2 + n_3}$.
From the solution~\eqref{eq:normal_sol}, we obtain two possible solutions to local depth derivatives
$(k_{1a},k_{2a})$ and $(k_{1b},k_{2b})$.
We pick the normal that minimizes the corresponding sum of squares of depth derivatives. Our chosen normal $\mathbf{n}_c$ is given by
\begin{equation}
\label{eq:nc}
\mathbf{n}_c = \begin{cases}
\mathbf{n}_a & \text{ if } k_{1a}^2 + k_{2a}^2 \leq k_{1b}^2 + k_{2b}^2 \\
\mathbf{n}_b & \text{ otherwise }
\end{cases}
\end{equation}
As shown in equation~\eqref{eq:homography}, $\overline{\mathbf{n}}$ is given by $\mathbf{H}^\top\mathbf{n}_c$.
\parag{Measure of degeneracy.} In case of degenerate conditions, the singular values $(\lambda
_1,\lambda_2,\lambda_3)$ of $\mathbf{H} $ are all unity.
We use the ratio $\frac{\lambda_1}{\lambda_3}$ to quantify the degeneracy. Thus, we only reconstruct from $\mathbf{S}$, if $\frac{\lambda_1}{\lambda_3} > \tau$. We set $\tau = 1.2$. We put an upper limit $\tau = 10$ to ensure that the data is well-conditioned as we observed that a high value of $\tau$ leads to inaccuracies caused by measurement errors on $\eta$.
\section{Computing Normals from Multiple Images}
Let $\{\mathbf{x} ^i_j\}, i\in [1,M], j\in[1,N]$, denote a set of $N$ point correspondences between $M$ images. Our goal is to find the 3D point $\mathbf{X}^i_j$ and the normal $\mathbf{N}^i_j$ corresponding to each $\mathbf{x} ^i_j$.
Using equation~\eqref{eq:CS_norm}, we write the local homography for each point correspondence $\mathbf{H}^{ik}_j$ between image pairs $(i,k) \in [1,M], i \neq k$ using the warps $\eta$.Each local homography $\mathbf{H}^{ik}_j$ is normalized by dividing by their second singular value. We compute $\mathbf{Hc}^{ik}_j$ given by the ratio of first and third singular value. We compute normals for each local homography $\mathbf{Hc}^{ik}_j$ using equation~\eqref{eq:normal_sol} and pick a unique solution using equation~\eqref{eq:nc}. The solution on the reference and non-reference image is given by$\mathbf{N}_j^{kk}$ and $\mathbf{N}_j^{ki}$ respectively. For non-degenerate cases, where $\tau \in [1.2,10]$, we compute normals $N^i_j$ by taking the mean of $\mathbf{N}_j^{ik}$ computed over $k$ reference images. We obtain 3D point $\mathbf{X}^i_j$ by integrating the normals on each surface using the method described in~\cite{Parashar19a}. We summarise Diff-NRSfM in algorithm~\ref{alg:diff}.
\begin{algorithm}
\SetAlgoLined
\KwData{$\mathbf{x}^i_j, \mathbf{H}^{ik}_j$ and $\mathbf{Hc}^{ik}_j$}
\KwResult{ $\mathbf{X}^i_j$ and $\mathbf{N}^i_j$}
$\tau_{min} = 1.2$, $\tau_{max} = 10$ \;
\For{ each reference image $k = [1,M]$}{
\For{ each point $j= [1,N]$}{
\For{ images $i= [1,M], i \neq k$}{
\eIf{$\mathbf{Hc}^{ik}_j > \tau_{min}$ and $\mathbf{Hc}^{ik}_j < \tau_{max}$}{
Compute normals using~\eqref{eq:normal_sol}\;
Pick a solution $\mathbf{N}_j^{kk}$ using~\eqref{eq:nc}\;
Write $\mathbf{N}_j^{ik} = (\mathbf{Hc}^{ik}_j)^\top \mathbf{N}_j^{kk}$\;
}
{ Set $\mathbf{N}_j^{ik}, \mathbf{N}_j^{kk}$ to zero\;
}
}
}
}
\For{ each point $j= [1,N]$}{
\For{ images $i= [1,M]$}{
Obtain $\mathbf{N}^i_j$ by averaging the non-zero normals in $\mathbf{N}^{ik}_j$\;
}
}
\For{ images $i= [1,M]$}{
Integrate $\mathbf{N}^i_j$ to obtain $\mathbf{X^i_j}$.
}
\caption{Diff-NRSfM}
\label{alg:diff}
\end{algorithm}
\section{Related Work}
NRSfM was introduced in~\cite{Bregler00} and the ill-posedness of the problem was handled by constraining the deformations to lie on a low-dimensional manifold. Later variants introduced additional constraints for efficient low-rank factorization ~\cite{DelBue06,DelBue08,Akhter08,Gotardo11a,Gotardo11b,Torresani08a} or performed additional optimization~\cite{Dai14,Paladini09,Garg13a,Lee16,Kumar18,Kumar19,Zhu14} to improve the statistical modeling.
Learning-based techniques have been used to tune the dimensionality of the deformation space~\cite{Kong19,Pumarola18,Golyanik18} using a large amount of annotated data for supervision.~\cite{Novotny19,Sidhu20} formulated learning-based techniques in an unsupervised setting to reconstruct from sparse and dense data, respectively. However, this does not overcome the fundamental limitation of approaches relying on low-rank assumptions: they cannot model complex deformations. Furthermore, they do not naturally handle missing data and occlusions, and complex formulations~\cite{Golyanik17} are required to overcome this. As a result, these methods have been limited to objects that deform in a relatively predictable way, such as human faces. Recently, these limitations have been addressed by imposing constraints between corresponding points across images in one of the following ways.
\parag{Modeling Global Deformations.} Several methods seek to enforce physical properties on the deformation, such as isometry that preserves local distances on the deforming surface. They approximate isometry by inextensibility~\cite{Chhatkuli17,Ji17}, piece-wise inextensibility~\cite{Vicente12,Russell11,Russell14}, local or piece-wise rigidity~\cite{Taylor10a,Varol09,Chhatkuli14,Kumar19a}. A globally optimal solution is then found by jointly solving over all corresponding points. This requires a computationally expensive optimization, which makes this approach impractical for handling large numbers of images. To handle non-isometric surfaces, a mechanics-based approach is proposed in~\cite{Agudo14,Agudo15a,Agudo16}, introducing the forces required to compute the resulting shape. In any event, all these methods require an initialization, usually obtained using standard rigid-body reconstruction techniques. Furthermore, they are often inaccurate.
\parag{Modeling Local Deformations.} In earlier works, we have proposed methods that rely on formulating local deformation constraints in terms of algebraic expressions. This makes it possible to reconstruct each surface point independently by solving algebraic equations, which reduces the computation cost. Being local, these methods inherently handle missing data and occlusions. In~\cite{Parashar17}, we treated surfaces as locally planar (LP) and formulated local isometric constraints using metric tensors and connections representing the rate of change of metric tensors. In~\cite{Parashar19a}, we extended this deformation modeling to conformal and equiareal deformations by assuming the deformation to be locally linear (LL). For each pair of images, we obtained two cubic equations in two variables related to local depth derivatives with 9 possible solutions. In practice, up to 5 of them can be real. We found a unique solution by minimizing sum-of-squares of residuals over multiple images. In~\cite{Parashar21}, we proposed two fast solutions to the equations of isometric NRSfM~\cite{Parashar17}.
Using substitution and resultants, we converted the original bivariate equations to univariate ones that can be solved efficiently. However, this comes at the cost of adding phantom solutions that cannot be identified. We picked the solution that yields the smallest residual of the isometry constraints on the entire image set. In~\cite{Parashar20}, we proposed an NRSfM solution for generic deformations. It uses only connections to formulate constraints to enforce surface smoothness.
Table~\ref{table:summ} summarizes the characteristics of these local methods.
They tend to perform significantly better than their global counterparts but suffer from one key drawback: the local constraints are not always well-posed, leading to manyfold ambiguities or even degenerate solutions, without any mechanism for telling when this happens.
This is the problem we address in this paper.
\section{Method Framework}
\label{sec:outline}
\input{fig/model}
Fig~\ref{fig:model} depicts our setup when using only two images $\mathcal{I}$ and $\overline{\mathcal{I}}$ acquired by a calibrated camera. In each one, we denote the deforming surface as $\mathcal{S}$ and $\overline{\mathcal{S}}$, respectively, and model it in terms of functions $\phi , \overline{\phi}: \mathbb{R}^2 \to \mathbb{R}^3$ that associate a surface point to an image point. Let us assume that we are given an image registration function $\eta: \mathbb{R}^2 \to \mathbb{R}^2$ that associates points in the first image to points in the second. In practice, it can be computed using standard image matching techniques such as optical flow~\cite{Sundaram10,Sun18a} or SIFT~\cite{Lowe04}. These functions can be composed to create a mapping $\psi: \mathbb{R}^3 \to \mathbb{R}^3$ from 3D surface points seen in the two images. We use a parametric representation of $\eta$ and $\phi$ using thin-plate splines~\cite{Bookstein89}, which allows us to accurately obtain first-order derivatives of these functions. A finite-difference approach could also be used.
At the heart of our approach is the fact, which we will prove, that the normals at two corresponding points can be computed from $\eta$ and $\phi$ under the sole assumptions of local planarity and local linearity. \PF{Explain!}
The earlier methods~\cite{Parashar19a,Parashar20a} rely on a similar formalism, but our approach has two key advantages:
\begin{enumerate}
\item The system of equations we solve to compute the normals only has at most two solutions, instead of five when the system is expressed in terms of depth derivatives as in~\cite{Parashar19a,Parashar20a}. In practice, we pick the solution that minimizes the sum of squares of depth derivatives. In other words, we choose the normals that correspond to the smallest depth-change, thus promoting smoothness.
\item Given that not all motions are conducive to reliable normal estimates, our formulation enables us to explicitly quantify which normals can be trusted because the system of equations used to derive them was well-conditioned and which cannot. Hence, when integrating the normals to recover the 3D surfaces, we can discard the untrustworthy ones, which as we will see in the result section gives us a significant performance boost.
\end{enumerate}
\comment{
As in~\cite{Parashar20a}, we assume the deformations to be diffeomorphic but with two main differences:
\begin{enumerate}
\item We assume the surfaces to be infinitesimally planar instead of infinitesimally linear. This means that the second second-order derivatives can be expressed in terms of the first-order ones but are non-zero. \PF{Correct?}. \ShP{Yes}
\item In~\cite{Parashar20a} the diffeomorphic constraints are expressed solely in terms of connections \MS{We have not mentioned anything about connections so far. This will not be clear to the reader.} and the metric tensor is only used to enforce additional constraints, such as isometry, conformality or equiareality. By contrast, we use both the metric tensor and the connections and show that all possible deformation types are represented by only one constraint. \MS{This discussion is quite obscure to me. Is it really the right place to have this entire paragraph?} \PF{I find Mathieu's comment reassuring. I have absolutely no idea what this paragraph means ;-)} \ShP{I will take this paragraph as a discussion in the end of the paper}
\end{enumerate}
}
\section{Experiments}
We compare our method against state-of-the-art ones on both synthetic and real datasets with available ground truth.
\subsection{Datasets}
\parag{Synthetic Datasets.}
We created 3 smooth surfaces: a plane, a cylindrical surface and a stretched surface with 400 tracked correspondences, as shown in Figure~\ref{fig:syn_exp}.
\parag{Real Datasets from our Previous Work.}
These include the \textbf{Paper}~\cite{Salzmann07b}, \textbf{Rug}~\cite{Parashar17} and \textbf{Tshirt}~\cite{Chhatkuli14} datasets. \textbf{Paper} comprises 191 images from a video of a deforming sheet of paper with 1500 point correspondences. \textbf{Rug} comprises 159 images from a video of a deforming rug with 3900 point correspondences. \textbf{Tshirt} has 10 wide-baseline images with 85 point correspondences.
The correspondences in the {\bf Paper} dataset were obtained using SIFT with a manual supervision of accuracy and are thus highly accurate. By contrast, those in the {\bf Rug} dataset were computed using the dense optical flow method of~\cite{Garg13a} and contain errors due to optical drift and regional mismatches due to the lack of texture. The correspondences in {\bf Tshirt} are computed manually. The ground truth for \textbf{Paper} and \textbf{Rug} is obtained using kinect, which is very noisy and contains large, inconsistent depth variations. For an apt comparison, we refined the ground truth to obtain smooth surfaces. The ground truth for {\bf Tshirt} is computed using rigid reconstruction of each image from multiple views.
\parag{NRSfM Challenge Dataset.}
It consists of 5 image sequences depicted by Fig.~\ref{fig:nrsfm_ch_data}. They feature 5 kinds of non-rigid motions: articulated (piecewise-rigid) with 207 images and 69 point correspondences, balloon (conformal) with 51 images and 211 point correspondences, paper bending (isometric) with 40 images and 153 point correspondences, rubber (elastic) with 40 images and 481 point correspondences, and paper being torn with 432 images and 405 point correspondences. The dataset features images from 6 different camera motions and provides image points captured assuming both a perspective and an orthographic projection. It provides only one ground-truth surface for each of the sequences. The correspondences are sparse and not well-distributed across the images.
\parag{Datasets used by~\cite{Sidhu20}.}
\cite{Sidhu20} released the \textbf{Paper}, \textbf{Tshirt}, \textbf{Actor} and \textbf{Expressions} datasets, which have been widely used by many physics-based and low-rank constraints based methods. The \textbf{Paper} images are the same as the one used by us.~\cite{Sidhu20} uses 60K dense correspondences computed using optical flow~\cite{Garg13a} and the raw depth data from the kinect is considered as the ground truth. The \textbf{Tshirt} data has 300 images with 70K dense correspondences computed using~\cite{Garg13a}, with the kinect raw depth data as ground truth. To deal with the inconsistent depth variations of the raw kinect data,~\cite{Sidhu20} refines the raw data and focuses on small portions of theses datasets where the inconsistent depth variations are minimal, as shown in Fig~\ref{fig:dense_data}.
\textbf{Actor} contains 100 images of a deforming human face with 36K dense correspondences, and \textbf{Expressions} includes 384 3D shapes of a deforming human faces with 1000 point correspondences. The ground truth for both these datasets is synthetic. Fig~\ref{fig:dense_data} shows some samples.
\parag{Blue Sheet Dataset.}
Additionally, we recorded a video sequence featuring a textureless blue sheet deforming isometrically using a Kinect. It comprises 60 images and 7K point correspondences that were tracked using dense optical flow~\cite{Garg13a}. Optical flow on textureless surfaces is prone to large errors, and the flow we obtained confirms this.
\begin{figure*}
\centering
\includegraphics[width=\textwidth]{paper_normals.png}
\includegraphics[width=\textwidth]{rug_normals.png}
\includegraphics[width=\textwidth]{tshirt_normals.png}
\caption{ {\bf Datasets used in our previous work.} Reconstructed normals on three images. The ground-truth normals are shown in green, the ones predicted by \textbf{Ours} in blue, and those by \textbf{Pa21-R} in black. Note that our normals are far less noisy. }
\label{fig:paper_exp}
\end{figure*}
\input{fig/summ_tab_syn}
\subsection{Baselines and Metrics}
We compare our method to local linearity-based diffeomorphic NRSfM \textbf{Pa20}~\cite{Parashar20}, jointly solving isometric/conformal NRSfM \textbf{Pa19}~\cite{Parashar19a}, two fast solutions \textbf{Pa21-R} and \textbf{Pa21-S}~\cite{Parashar21} that transform the original constraints to univariate polynomials, which can be easily solved, and local and piecewise homography decomposition, \textbf{Ch14}~\cite{Chhatkuli14} and \textbf{Va09}~\cite{Varol09}, respectively. These are methods that, like ours, reconstruct local/piecewise surface normals and integrate them to obtain depth. Note that the solution to isometric NRSfM in~\cite{Parashar17} is the same as the one in \textbf{Pa19}. Therefore, there is no need for additional comparison.
We report errors in terms of accuracy of the normals \textit{En} and 3D points \textit{Ed}. \textit{En} is computed as the average dot product between ground-truth and computed normals. The normal integration done in the above methods yields a smooth reconstruction by enforcing a local smoothness on the normals. As a consequence, it improves the quality of the reconstructed normals. Therefore, we also report \textit{En (s)}, which is the error between the smoothened and the ground-truth normals.
\textit{Ed} is the mean RMSE between the ground-truth and computed 3D points.
We also compare our approach against three of the best global methods, \textbf{Ch17}~\cite{Chhatkuli17}, \textbf{Ji17}~\cite{Ji17} and \textbf{Lee16}~\cite{Lee16}, along with a dense method, \textbf{An17}~\cite{Ansari17}. They directly return 3D points. Hence, we only report \textit{Ed} for these methods.
While comparing on the datasets used by~\cite{Sidhu20}, we report $Ed$ as the mean 3D error, as computed in this method. Therefore, $Ed= \dfrac{1}{N}\sum_t \dfrac{|| P_{recon}-P_{GT} ||_2}{|| P_{GT}||_2}$, where $P_{recon}$ is the obtained reconstruction, $P_{GT}$ is the ground truth and $N$ is the number of images in the dataset.
In the remainder of this section we will refer to the method described in this paper as~\textbf{Ours}.
\begin{table*}
\caption{ (left) {\bf RMSE results on the datasets used in our previous work}. 'X' indicates that the method does not evaluate normals. '----' indicates that method failed to return a result due to its high computational complexity. (right) {\bf Computation times} as a function of the number of images and points used.}
\centering
\includegraphics[width=\textwidth]{Summary_rug_paper.png}
\label{fig:summary}
\end{table*}
\subsection{Comparative Results}
\parag{Results on Synthetic Data.} Fig.~\ref{fig:syn_exp} shows the generated surfaces. The performance of all methods is averaged over 10 trials with added gaussian noise with a 3 pixels standard deviation.
As \textbf{Ours} can reconstruct from two images only, we perform both pairwise reconstructions and joint reconstruction from the image triplet available for each surface. We report the results in Table~\ref{fig:summary_syn}. For methods that perform normal integration, we report errors of both computed and smoothened normals. The improvement in the normals due to smoothing is huge for \textbf{Ch14} and \textbf{Va09}, substantial for \textbf{Pa19}, \textbf{Pa20}, \textbf{Pa21-S} and \textbf{Pa21-R} and minor for our method. To truly compare the NRSfM techniques themselves, we therefore report the accuracy of the computed normals rather than the smoothed ones. We obtain a very accurate reconstruction from 2 images only. Beside \textbf{Ours}, \textbf{Va09} is the only baseline that can reconstruct from 2 images. However, it does not perform well on this data. \textbf{Lee16} and \textbf{An17} are designed for video sequences, and thus need more than 3 images to perform effectively. The remaining methods can operate on three images, but their accuracy is lower than ours, especially in terms of normal accuracy. Since we can discard the normals that have a low reliability, the accuracy of our reconstruction is strengthened using multiple images. Fig.~\ref{fig:syn_exp} further confirms the quality of our reconstructions by depicting the normals we obtain {\it without} any smoothing.
\begin{table}
\caption{ {\bf Computation times} as a function of the number of images and points used.}
\centering
\includegraphics[width=0.5\textwidth]{time.png}
\label{fig:time}
\end{table}
\parag{Results on the Datasets used in our Previous Work.}
Because the computational complexity of the global baselines grows rapidly with the number of correspondences, we evaluated all methods on the full set of correspondences and on a subset of 350 correspondences on {\bf Paper} and {\bf Rug}. For example, \textbf{Ch17}, \textbf{Ji17} have a cubic complexity and hence they yield a very high computation time when there are many correspondences. Their Matlab implementation crashes when using all correspondences, and using only 1000 correspondences still takes hours on a modern CPU.
Similarly \textbf{Ch14} and \textbf{Va09} take almost 1 hour to reconstruct 20 images and we therefore did not evaluate them on these datasets. The {\bf Tshirt} dataset has only 10 wide-baseline images. \textbf{Lee16} and \textbf{An17} are not designed to work on wide-baseline data, therefore we did not evaluate them on this dataset.
We report our quantitative results in Table~\ref{fig:summary}, and Figure~\ref{fig:paper_exp} depict qualitative ones. We outperform all baselines in terms of \textit{Ed} on the \textbf{Paper} and \textbf{Rug} dataset with partial and full correspondences. On the {\bf Tshirt} dataset, \textbf{Ch17} and \textbf{Ji17} perform better. Crucially, our performance is achieved at a much reduced computational cost by solving a set of equations in closed form, as opposed to invoking a complex solver. As a result, our approach is about 150 times faster than \textbf{Ch17} on 350 correspondences and can handle thousands whereas \textbf{Ch17} cannot. Furthermore, our approach is also 50 times faster than \textbf{Pa19}, the counterpart local approach which uses expensive polynomial solvers, because we do not have to derive a complicated formulation to obtain a unique solution for each correspondence.
Table~\ref{fig:time} provides a detailed analysis of the run-times of all the methods on 350 and 1500 points. We assume that the input point correspondences and their derivatives are pre-computed. Therefore, the timings only encode the computation of the normals or 3D points. Our approach yields the fastest run-times, seconded by \textbf{An17}. Note, however, that \textbf{An17} has a parallel implementation and is computationally optimized. By contrast, our approach, as all the other ones, is implemented in Matlab and not optimized for speed.
The relative slowness of the other local method arises from the local normal estimators of \textbf{Pa19} and \textbf{Pa20} having to minimize the sum of squares of polynomials,
which is expensive even if it has linear complexity. \textbf{Pa20} is further slowed down by having to transform polynomials into univariate expressions. \textbf{Pa21-S} and \textbf{Pa21-R} obtain analytical solutions but require a fairly expensive disambiguation. By contrast, our local normal estimator is computationally cheap as it has a closed-form solution.
\begin{figure*}
\centering
\includegraphics[width=\textwidth]{nrsfm_challenge.png}
\caption{NRSfM challenge dataset and some reconstructions using {\bf Ours}. Green indicates the ground truth and blue indicates our reconstruction.}\label{fig:nrsfm_ch_data}
\end{figure*}
\parag{Results on the NRSfM Challenge Dataset.}
Fig.~\ref{fig:challenge} compares the performance of \textbf{Ours} with that of other methods in terms of $Ed$, measured in mm, with {\bf Best} being the one that does best as reported in the benchmark statistics provided on the website. The local methods show a significant performance improvement compared to the other ones. \textbf{Pa19} uses second-order derivatives of the image registration $\eta$, which can be highly erroneous on this dataset. It uses an expensive polynomial solver, which cannot handle such large noise and fails on a large number of cases. \textbf{Pa21-S} and \textbf{Pa21-R} find an analytical solution to the isometric/conformal NRSfM posed in \textbf{Pa19}, which requires a non-linear refinement to obtain a unique solution; they obtain decent results on this dataset. \textbf{Pa20} solves NRSfM using diffeomorphic constraints, which uses only first-order derivatives of $\eta$, and is thus less impacted by the sparsity of the data and performs better than \textbf{Pa21-S} and \textbf{Pa21-R}. \textbf{Ours} requires second-order derivatives of the image registration, but it is equipped with a measure to compute the well-conditioning of the data. This lets us identify and discard the non-isometric/non-conformal data and reconstruct from as-isometric(or conformal)-as-possible data. As a result, \textbf{Ours} yields better results than \textbf{Pa20}. Fig.~\ref{fig:nrsfm_ch_data} shows some reconstructions obtained with our method.
\begin{table*}
\caption{ {\bf Results on the NRSfM challenge datasets.}}
\centering
\includegraphics[width=\textwidth]{challenge.png}
\label{fig:challenge}
\end{table*}
\parag{Results on the Blue Sheet Dataset and on the Datasets used by~\cite{Sidhu20}.}
These datasets are large in terms of the number of either point correspondences or images they contain. We compare the performance of \textbf{Ours} with \textbf{An17}, which is designed for reconstructing dense objects, however, it takes several hours to reconstruct. Additionally, we report the performance of our other local methods \textbf{Pa19}, \textbf{Pa21-S} and \textbf{Pa21-R}. In this case, we report the mean 3D error to be able to compare with the performance of~\cite{Sidhu20}, which has demonstrated best results on these datasets. Table~\ref{fig:dense} summarizes the results. \textbf{Ours} performs better than most of the methods on these datasets. The \textbf{Actor} and \textbf{Expressions} sequences are relatively simple, with small relative motion across images. All local methods therefore perform similarly on these sequences. \textbf{An17} performs better than \textbf{Ours} on the \textbf{Actor} sequence. However, the visual performance is quite similar as the error margin is very low, to the third decimal place. Fig.~\ref{fig:ac_exp} shows some reconstructions. Fig.s~\ref{fig:sheet_exp}~\ref{kin_exp} show the results on the \textbf{ Blue Sheet}, \textbf{Paper} and \textbf{Tshirt} datasets where \textbf{Ours} performs significantly better than the compared methods.
\begin{figure*}
\centering
\includegraphics[width=\textwidth]{sheet_depth.png}
\caption{ {\textbf{Blue Sheet dataset.}} Reconstructed surfaces for two images. The predictions of \textbf{Ours} are shown in blue, of \textbf{Pa21-R} in red, and of \textbf{An17} in black. Note that our reconstructions are less noisy and match the surface 3D shape much better. }
\label{fig:sheet_exp}
\end{figure*}
\begin{table}
\caption{ {\bf Performance on dense datasets.} }
\centering
\includegraphics[width=0.5\textwidth]{dense.png}
\label{fig:dense}
\end{table}
\begin{figure}
\centering
\includegraphics[width=0.5\textwidth]{actor_expr.png}
\caption{ {\textbf{Actor and Expressions datasets.}} Reconstructed surfaces for two images. The predictions of \textbf{Ours}, \textbf{Pa21-R} and of \textbf{An17} are quite similar.}
\label{fig:ac_exp}
\end{figure}
\begin{figure}
\centering
\includegraphics[width=0.5\textwidth]{kinect.png}
\caption{ {\textbf{Paper and Tshirt datasets.}} The predictions of \textbf{Ours} are shown in blue, of \textbf{Pa21-R} in red, and of \textbf{An17} in black. Note that our reconstructions are less noisy and match the surface 3D shape much better. }
\label{fig:kin_exp}
\end{figure}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 8,953
|
#ifndef __BTHIDD_H__
#define __BTHIDD_H__
#include "../default.h"
#define PROVIDE bthidd
#define REQUIRE { DAEMON, "hcsecd" }
#define BEFORE { LOGIN }
#define KEYWORD { NOJAIL, SHUTDOWN }
#define DEFAULT_BTHIDD_CONF "/etc/bluetooth/" NAME ".conf"
#define REQUIRED_FILES { DEFAULT_BTHIDD_CONF }
#define DEFAULT_HIDS_FILE "/var/db/" NAME ".hids"
#define PID_FILE "/var/run/" NAME ".pid"
#define BTHIDD_COMMAND_ARGS_FORMAT "-c %s -H %s -p " PID_FILE
#include "../common.h"
#endif
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 388
|
Content Not Copyrighted
Filed under: blogging, Christianity, writing — Tags: blessed to be a blessing, blogging, blogging etiquette, Christian blogging, Christian blogosphere, Christian ethics, Christian living, Christian writing, copyright, freely you have received, generosity, gifts from God, humility, intellectual property, Jesus' teachings, open source community, ownership of writing, possessions, pride, reblogging, royalties, sharing of ideas — paulthinkingoutloud @ 5:50 am
There is no limit on what can be done for God as long as it doesn't matter who is getting the earthly credit.
There's a worship song currently making the rounds that goes, "It's your breath, in our lungs, so we pour out our praise; pour out our praise…" To me, the song is a reminder that it's God who gives us breath, gives us abilities, gives us opportunities and one of the best uses of that is to offer back praise to him.
For the third time in nearly 2,000 posts, this week we got a take-down order at Christianity 201. Yes, it would be nice to have a staff and be able to contact writers in advance and say, "We think your writing would be a great addition to C201 and we'd like to include what you wrote last Tuesday in our gallery of devotional articles." But I just don't have that luxury. So we pay the highest compliments to our writers by encouraging our readers to check out their stuff at source, while at the same time archiving it for the many who we know statistically don't click through.
The one this week offered some lame excuse about how I was disturbing his Google analytics by publishing his works, and reminded me that he could sue me. Nice attitude, huh?
These days, most of the authors are appearing for the second, third or fourth time, and many write (both on and off the blog) to say how honored they are that we find their material helpful.
I honestly can't remember the name of the first two authors, but I know one had some recognition in Calvinist circles; so when the lightning struck again this week, I checked out the guy's Twitter to look for clues and guess what?
That got me thinking about something I wrote here about 16 months ago…
The Bible has a lot to say about the accumulation of wealth and the hoarding of possessions. Probably the classic statement of scripture on the matter is,
NASB Matt. 6:19 "Do not store up for yourselves treasures on earth, where moth and rust destroy, and where thieves break in and steal. 20 But store up for yourselves treasures in heaven, where neither moth nor rust destroys, and where thieves do not break in or steal…
MSG Matt. 6:19-21 "Don't hoard treasure down here where it gets eaten by moths and corroded by rust or—worse!—stolen by burglars. Stockpile treasure in heaven, where it's safe from moth and rust and burglars. It's obvious, isn't it? The place where your treasure is, is the place you will most want to be, and end up being.
The Bible doesn't say, 'Don't have any treasure whatsoever.' True, when Jesus sent his disciples out he told them to travel light, advice that extends through all of life:
NLT Matt. 10:9 "Don't take any money in your money belts—no gold, silver, or even copper coins. 10 Don't carry a traveler's bag with a change of clothes and sandals or even a walking stick.
But in everyday life, the Bibles teaching presuppose you will have a home or a donkey or bread that you may or may not choose to give your neighbor when he comes knocking late at night.
This week it occurred to me that at the time the Bible was written, one thing that we can possess that they didn't was intellectual property. There was no Copyright Act; no Letters Patent. Did Jesus' earthly father, Joseph the Carpenter have a special way of doing a table that would cause him great consternation if Murray the Carpenter down the road started copying the idea? You get the feeling that everything was open source.
I think it's interesting that in the prior verse of Matthew 10, Jesus makes the often-quoted statement, "Freely you have received, now freely give."
Personally, there's nothing on this blog that isn't up for grabs, provided it's cited properly and quoted properly and being used non-commercially. Like this article? Help yourself. Yes, in the past I have been paid to write and could thereby consider myself a professional writer; but this is only a blog and it's vital not to get too caught up in your own sense of self-importance; and I say that not out the spirit of someone who is loaded with wealth, but as a person who has had no specific fixed income for 19 years.
I also thought it was interesting that the one person who was so upset about the use of his material on other than his own website was complaining about a particular article that was about 50% scripture quotations. More than 50%, I believe. Oh, the irony. I can just hear Jesus saying, 'Uh, could you just link to my words in the Bible rather than print them out on your own website?'
That said, I am consciously aware that a double standard exists in the Christian blogosphere. We both permit and excuse the copying of text, but there is far less grace for poachers of cartoons and photographs. (I guess a picture really is worth a thousand words.) If you take what belongs to them, it's like trying to wrestle a t-bone from a pit-bull.
In the early days of this blog, the weekly link list included cartoons from Baptist Press. Not any more. Baptists can be very litigious, which is too bad, because the cartoons were worthy of an audience beyond a single denomination. Everybody loses, but that's the Baptist way, I guess.
Words are cheaper however. I respect intellectual property rights in general, but hey, guys, it's only a blog.
I really think when the writer is a little older, they will look back and see the foolishness of trying to hang on to what really isn't yours to begin with.
Think About It: Some things simply didn't exist when the Bible was written, such as smoking cigarettes or driving over the speed limit. It's the same with intellectual property. We have to appeal to the timeless, grand themes of scripture to make behavioral determinations.
The corollary to this is that if I do choose to copyright my blog writing here, I am basically saying this is mine; I wrote this, I created it, it was my talents and my gifts that went into creating it.
I'm glad the Biblical writers didn't feel that way. If you believe in plenary inspiration — that God birthed ideas within them but they stylized it and added their individual touch to the writing — then even if you hold that "all Scripture is inspired" (which I do) you could still make a case that they could copyright the particular words used.
But some would argue that even if you say, "This came entirely from God and I shouldn't really take any credit for it;" if you want your writing to reach the greatest number of people, then you've got to put somebody's name underneath the title.
That's essentially the case with Jesus Calling. I don't want to get into the larger debate on that book, because it's been done elsewhere (with many comments) but if, like the classic God Calling, the "authors" feel that this book is the equivalent to Dictation Theory in Biblical inspiration, realistically, nobody's name should appear on the cover. I wonder if "by Jesus" or "by God" would sell more or fewer copies than "by Sarah Young."
You can however engage the commercial marketplace and at the same time take no money (or very little) for your wares. Keith Green is a name that some of the younger generation don't know, but Keith basically said that if anyone couldn't afford his records or cassettes, he would send them copies free of charge. It was radical at the time — this was before free downloads — and Keith took ribbing that perhaps he was also going to ship stereo systems to people who had nothing on which to play the music.
Keith Green would have loved blogging — he'd have about ten of them — and would be fighting hard for the open source blogosphere mentioned above, and also when the first writer protested. (The post then was triggered by an irate blogger at C201 as well, so we're running one complaint every 700+ articles, which isn't bad.) In fact, Keith would argue for open source thinking in a variety of Christian media and art.
Bottom line: We have to be careful about holding too tightly to the things of this world including possessions that are tangible and those which are intangible such as intellectual property.
Moving forward: We'll try to stick to repeat authors and original devotional material. If you've ever wondered if you could write devotional material — and it's both a rare and challenging calling — check out the submissions guidelines at C201.
"It's your breath, in our lungs, so we pour out our praise…"
The Possession of Ideas, Part 2
Filed under: blogging, writing — Tags: Christian blogging, Christian ethics, Christian writing, freely you have received, generosity, Jesus' teachings, open source community, ownership of writing, possessions, sharing of ideas — paulthinkingoutloud @ 1:47 pm
The corollary to yesterday's discussion is that if I do choose to copyright my blog writing here, I am basically saying this is mine; I wrote this, I created it, it was my talents and my gifts that went into creating it.
That's essentially the case with Jesus Calling. I don't want to get into the larger debate on that book, because it's been done elsewhere (with 100+ comments) but if, like the classic God Calling, the "authors" feel that this book is the equivalent to Dictation Theory in Biblical inspiration, realistically, nobody's name should appear on the cover. I wonder if "by Jesus" or "by God" would sell more or fewer copies than "by Sarah Young."
You can however engage the commercial marketplace and at the same time take no money (or very little) for your wares. Keith Green is a name that some of the younger generation don't know, but Keith basically said that if anyone couldn't afford his records or cassettes, he would send them copies free of charge. It was radical at the time — and would be even more so today — and Keith took ribbing that perhaps he was also going to ship stereo systems to people who had nothing on which to play the music.
Keith Green would have loved blogging — he'd have about ten of them — and would be fighting hard for the open source blogosphere we talked about yesterday, and also almost exactly two years ago. (The post then was triggered by an irate blogger at C201 as well, so we're running one complaint every 700+ articles, which isn't bad.) In fact, Keith would argue for open source thinking in a variety of Christian media and art.
The Possession of Ideas
Filed under: blogging, writing — Tags: Christian blogging, Christian ethics, Christian writing, freely you have received, generosity, Jesus' teachings, open source community, ownership of writing, possessions, sharing of ideas — paulthinkingoutloud @ 10:16 am
The whole premise of the sister blog to this one, Christianity 201, is that we search the internet for sources of daily Bible exposition and discussion. Unlike the Wednesday Link List, where some people click and some people just read the list, I think it's important that these devotional meditations get seen in full, and statistics bear out the reality that most people don't click through.
Most of the bloggers are thrilled that their work is being recognized. C201 doesn't have quite the readership of Thinking Out Loud, but it possibly represents ten times as much as some of the writers see on their own pages. We get notes of appreciation, and a handful of readers also thank us regularly for putting them onto reading a particular writer.
So this week when, for the second time in about 1,450 posts someone strenuously objected to their material being reproduced in full — don't look for it, it's been removed — I started thinking about the whole intellectual property issue in the light of Jesus' teachings.
Personally, there's nothing on this blog that isn't up for grabs, provided it's cited properly and quoted properly and being used non-commercially. Like this article? Help yourself. Yes, I have been paid to write and could thereby consider myself a professional writer; but this is only a blog and it's vital not to get too caught up in your own sense of self-importance; and I say that not out the spirit of someone who is loaded with wealth, but as a person who has had no specific fixed income for 19 years.
I also thought it was interesting that the person who was so upset about the use of his material on other than his own website was complaining about a particular article that was about 50% scripture quotations. More than 50%, I believe. Oh, the irony. I can just hear Jesus saying, 'Uh, could you just link to my words in the Bible rather than print them out on your own website?'
There is a Part Two to this which appeared the next day.
Irony: The copyright symbol used today was already in my computer before I worried about such things…
Christian Blogging: Longing for Open Source Community
Filed under: blogging — Tags: blessed to be a blessing, blogging, blogging etiquette, Christian blogging, Christian blogosphere, Christian living, copyright, gifts from God, humility, intellectual property, ownership of writing, pride, reblogging, royalties — paulthinkingoutloud @ 6:58 am
Did God give me what I'm writing right now or am I making it up on my own strength?
That's a question it's fair to ask in all areas of Christian endeavor. Am I doing this 'on my own' or under God's power? What about the idea that 'all things come from God?' Do I really 'own' the concepts and insights shown here.
As we closed in on having 700 posts at Christianity 201 last week, for the first time we had a writer who objected to having his content used here. While blog etiquette dictates that you link back to writers' original pages, statistics bear out the idea that people read the teaser paragraph but don't click to continue reading. So C201 was created as a showcase — and a bit of a potpourri — of devotional and Bible study writing; much of it from previously obscure blogs that nobody had heard of, whose writers are thrilled to have an additional audience for their thoughts.
For several months, a music and book distributor for whom I was I was doing contract work assigned me to help out in royalty administration and distribution. I appreciate that those who have given themselves full-time to writing for major publishers derive their income from sales. I would never dream of photocopying an author's work and I have strong views about churches which project song lyrics on a screen at weekend services for which they haven't paid the appropriate license fees.
But a blog? Seriously?
When the attribution is clear, and the readers are given two separate opportunities — and sometimes additional inducements — to click to the original source page, I feel there is a legitimization of one-time use; though a few writers have been featured at C201 on two or three different occasions.
(Cartoonists however, seem to be another subject entirely. Despite having the largest treasure trove of Christian cartoons online, one denominational website had so many copyright warnings we decided they could just keep their comics to themselves, and stopped using them here at Thinking out Loud.)
The article in question had no copyright indicia, and no page dealing with reprints and permissions.
I would like to think that when God gives us an idea, he gives it to us not only to share, but to see disseminated as widely as possible. Someone once said,
Attribution's greatest value is that the people can go back to the same source for more insights. If I enjoy what "X" has to say today on this topic, then I may want to read what "X" has to say tomorrow about some other subject. In fact, I've had a handful of off-the-blog comments from people who are now regular readers of writers they heard about here at C201 and at Thinking out Loud.
In giving instructions to his disciples, Jesus said,
"And as you go, preach, saying, 'The kingdom of heaven is at hand.' "Heal the sick, raise the dead, cleanse the lepers, cast out demons. Freely you received, freely give. (Matt 10:7-8 NASB)
I've had content used (and misused) on other blogs, and at the end of the day, it doesn't really matter much. What does matter is how I respond to the "borrowings" at other websites. Do I say, "It's all good;" or do I fight for increasing my personal empire here at this website?
The writer in question also accused me of changing his content. I could see how that would be serious. But in fact, all I had done was to remove links to an online bookseller which left him, in one sentence, referring to "this book" with no remaining hint as to what that book might be; so I took the time to insert the title where the words "this book" had been.
I think it was with the objection to that change that the author really betrayed their true motives. Referrer fees from online sales can be fairly significant for a blogger at the end of the month; and I believe it can really cloud a writer's motives.
I simply won't do that here. I'm not trying to sell you anything. I'm not making money from this, and in fact I don't draw a salary from my "day job," so perhaps I have a different attitude toward the need to see everything I do as a line on a profit-and-loss balance sheet.
I wonder what the early church would think of what we've come to; a world where royalty administrators and agents hash out mechanical royalties and performance royalties and you buy a license in order to share the words to the latest worship songs. I wonder if the Apostle Paul were alive today if he would put a little copyright symbol at the end of each epistle? Would Matthew be expecting dividends from the sales of the Visual Bible DVDs that bear his name?
Freely we have received. Freely we give.
All that we have and are is a gift from God.
And we should keep it open source.
Ironically, trying to find a stylized copyright symbol to accompany this article was a challenge since nearly half of them were, in fact, copyrighted. This one above is from an article that also looks at this issue from a balanced Christian perspective.
It turns out the "There is no limit…" quotation is making its third appearance here. In addition to the reference linked above, I also used it in reference to Garrison Keillor at this post.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 9,000
|
'use strict';
/* global suite: false, setup: false, test: false,
teardown: false, suiteSetup: false, suiteTeardown: false */
var assert = require('assert');
var Walker = require('../');
suite('Walker.Types', function() {
var types;
test('new Walker.Types()', function() {
types = new Walker.Types();
});
test('Types#keys(\'Property\')', function() {
var keys = types.keys('Property');
assert.deepEqual(keys, ['key', 'value', 'kind']);
});
test('Types#keys(\'Property\', true)', function() {
var keys = types.keys('Property', true);
assert.deepEqual(keys, ['key', 'value']);
});
});
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3,055
|
Q: Run gitlab-ci.yml only when merge request to master made I am currently having my project in GitLab and Heroku. What I wanna do is as soon as I ask for merge request with my feature branch (let's call it crud-on-spaghetti), I want to automatically run the tests on this branch (npm test basically, using Mocha/Chai), and after they succeed, merge this crud-on-spaghetti with master, commit it and push it to origin/master (which is remote on GitLab) and after git push heroku master (basically, push it to the master branch in Heroku, where my app is stored). I have read several articles on GitLab CI and I think this is more suitable for me (rather than Heroku CI, because I do not have DEV and PROD instances).
So, as of now, I do this manually. And this is my .gitlab-ci.yml file now (which is not committed/pushed yet):
stages:
- test
- deploy
test_for_illegal_bugs:
stage: test
script:
- npm test
deploy_to_dev:
stage: deploy
only:
- origin master
script:
- git commit
- git push origin master
- git pull heroku master --rebase
- git push heroku master
Hence, my questions is: What do I exactly need to write in .gitlab-ci.yml in order to automate all these "manipulations" (above)?
PS. And another (theoretical) follow-up question: how is GitLab-CI Runner triggered? For instance, if I want it to trigger upon merge request with master, do I do that using only: ... in .gitlab-ci.yml?
A: Try
only:
- master
origin is just a name for a remote. master is the name of the branch.
The runner is triggered by GitLab-CI the moment that a commit is pushed to the repository, so alas not upon merge request.
You can use a trigger to trigger a pipeline and then call that trigger from a merge request event in integrations.
A: Restrict stages to Merge Requests:
To have your test stage only being executed when a Merge Request (MR) is opened, use
only:
- merge_requests
According to the Gitlab docs, you can further restrict this to only being executed for MRs with a certain target branch, e.g. only MRs for master
only:
- merge_requests
except:
variables:
- $CI_MERGE_REQUEST_TARGET_BRANCH_NAME != "master"
This adds an exception for all target branches that are not master.
Or use rules: for that:
rules:
- if: '$CI_PIPELINE_SOURCE == "merge_request_event" && $CI_MERGE_REQUEST_TARGET_BRANCH_NAME == "master"'
Restrict stages to Branches:
As already mentioned by @marcolz, this is achieved by
only:
- master
to only execute the stage for pushes to the master branch.
A: I am not sure, but I think that the rule one should use is to detect a "result merge event":
($CI_MERGE_REQUEST_EVENT_TYPE == "merged_result") &&
($CI_MERGE_REQUEST_TARGET_BRANCH_NAME == $CI_DEFAULT_BRANCH)
see https://docs.gitlab.com/ee/ci/pipelines/merged_results_pipelines.html
otherwise, the stage might run regardless if merge request was approved or not and finished successfully.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 582
|
Home » High-Volume Machines Slay the Variation Dragon
High-Volume Machines Slay the Variation Dragon
In Six Sigma machining operations, fighting variation is a never-ending quest. Just when production engineers and quality control specialists think that they have vanquished the variation dragon, the monster can reappear because of changing production requirements or other marketplace factors.
Now, however, with the help of a new breed of flexible rotary transfer machines that offer many of the benefits of multistation machining without some of the typical drawbacks, some shops are fighting back more effectively against variation. And while the variation beast can never be killed entirely, it is rearing its ugly head less often--in some shops, at least.
Among those most susceptible to variation problems are machine shops that deliver large volumes of parts to just-in-time manufacturing facilities in industries such as automotive and appliances. These shops must deploy machine tools that are agile enough to produce various parts in batches as the shops receive orders for them. Moreover, the systems must be flexible enough to run new jobs once the ever-shorter contracts common in industry today come to an end. Clearly, old-style, dedicated transfer lines that churned out the same parts day after day do not fit the bill.
For many machine shops, the strategy for offering agility at reasonable cost and risk has been to replace the dedicated machines with banks of machining centers. Another option is to insert machining centers into transfer lines. Many shops rely on a combination of both approaches.
These strategies do not come without tradeoffs, however. Although machining centers are flexible enough to cut any part that fits within their work envelopes, they are not as fast as dedicated transfer machines. Consequently, a number of them are necessary for each operation to create the required capacity.
Herein lies the quality control problem. Adding spindles increases the number of parallel streams in the production line, and more frequent changeovers multiply the number of setups. The opportunities for variation, therefore, also expand, because no two machines and no two setups can produce exactly the same part. Although diligence can keep the differences small and well within tolerance, the minute variation can be enough to complicate suppliers' efforts to adhere to the Six Sigma philosophy demanded increasingly by their customers.
The new hybrids
Technology advances eventually offer solutions to the problems that change brings with it. To make just-in-time production more compatible with the Six Sigma approach to quality control, some machine tool builders have developed a new breed of flexible rotary transfer machines that are essentially complete manufacturing cells made from three- to five-axis machining centers inserted into a common base. The rotary table at the center moves the parts from station to station, and a common controller coordinates all of the action. The result is a hybrid machine made by crossing the flexibility of a machining center with the economies of producing parts by transferring them among several spindles.
Among the advantages of these systems are fewer sources of error and simpler methods for controlling dimensional tolerances. Consequently, the accuracy and repeatability of these machines are typically much better than those possible from either conventional transfer machines or banks of machining centers. The reasons include fewer setups, automated material handling and software for error compensation. The rotary tables, the principal material handling device in these hybrid machines, fight error in various ways.
For one thing, they reduce the number of setups, which lessens the variation among parts that otherwise would occur when people or multiple material handling devices load parts into banks of machining centers. Automating parts handling among the sequence of operations and completing a part or a section of it on one machine without unclamping the parts eliminates an important source of stack-up error.
Despite the advantages, Arnold Jones, president of Kira America Inc. (Franks-ville, WI), a machine builder, urges engineers not to apply the single-clamping concept too slavishly. He points out that progressive clamping can be extremely competitive in mass production, as the Japanese demonstrated two decades ago when they were able to produce better cars at lower cost while moving parts manually from operation to operation. They had learned that the trick to controlling dimensions in high-volume sequential operations is to control the locating surface. "If each operation uses a unique fixture, you get the same result within millionths once you tune the station," says Jones.
He admits, however, that the need for a variety of fixtures can jinx the concept when several machining centers are performing identical operations in parallel. "You're going to have to deal with variation," he says. Minimizing the number of fixtures is one way of doing so, which is why his company also offers a hybrid machine, the Kira 4 Station CNC (computer numerical control) dial machine.
Minimize rechucks
David D'Aoust, vice president of machine builder Kaufman Manufacturing Co. (Manitowoc, WI), also recommends moderating the philosophy of chucking only once to one of rechucking parts as few times as possible. He notes that production volumes or part geometries often can preclude single-clamping operations. For example, one supplier to the automotive industry uses two of Kaufman's System 4 hybrid rotary machines to produce aluminum brake calipers. One machine performs Op 10, which is the machining of one side for locating the part for machining the other side in Op 20. "We grab the part only two times," says D'Aoust.
Jones points out another reason not to overemphasize single-clamping procedures. Unclamping parts sometimes boosts accuracy and repeatability, particularly when the machine is removing a large amount of material. "For any serious milling, unclamping is necessary to relieve stress," he says. "If you were to leave it clamped during finish machining, it would bow when it was released."
Relieving stress, however, does not necessarily mean unclamping parts, transferring them to another machine and relocating them there. Kaufman and other builders offer hydraulic and pneumatic distributors for their machines so that the CNC can vary the clamping pressure of the fixture at each station independently of the others. "We can chuck a part quite solidly during heavy roughing operations," explains D'Aoust. "When the fixture goes to the finishing stations where the chip loads are lighter, the CNC can reduce the chucking pressure."
The automotive industry uses this feature quite a bit for producing motor end shields and yokes for universal joints. "On the universal-joint yokes, the initial cross-drilling operation is done in solid forged steel, so you need a good grip on it," says D'Aoust about one application. "At the stations finishing the bore to size and putting in the snap-ring grooves, the CNC relieves the chucking pressure and lets the part relax." Otherwise the bore and groove will be shaped as an ellipse or have lobes once the pressure is relieved.
D'Aoust adds that the necessity of relieving stress depends upon three variables: the amount of material removed, the fragility of the remaining structure and the intensity of the clamping forces. "If you remove a lot of material, the part can spring out of shape," he says. "If you can't solve the problem by relieving the pressure, then you would have to put the part on another machine, relocating the part and rechucking it in an area that will not deform it."
Another way that the rotary tables on hybrid machines fight error is by being a simple transfer mechanism. Most no longer play a dominant role in the accuracy of rotary transfer machines, as they have in the past on conventional transfer machines. Rather than relying on the table and the position of each fixture on it, the locating accuracy comes entirely from coupling the fixture directly to the base of each machining center in the machine. The fixture typically fits into a Hirth ring or similar coupling mounted on the C-frame casting for each machining module.
According to engineers at Hydromat Inc. (St. Louis), a rotary transfer machine builder, this design using the Power Chuck from Erowa Technology Inc. (Arlington Heights, IL), a workholding device supplier, gives Hydromat engineers' new Advanced Technology rotary transfer machine 2-micron locating accuracy, station to station. "The Erowa Power Chuck operates on the same principle as a Hirth ring coupling," explains Martin Weber, vice president of manufacturing at Hydromat. "A number of precision-ground teeth arranged in a 90-degree pattern engage and lock to provide clamping accuracy and rigidity."
Besides making the machine more accurate and less time-consuming and costly to build and set up, the Power Chuck system makes post-machining inspection simpler and more accurate because only one set of measurements is necessary. The part, or a portion of it, can come off the machine completed. This means that inspectors or automation gaging stations can check the forms, dimensions and locations of different features on the part against each other with much greater certainty than they can when required to check one feature several times to an unmachined surface. Moreover, checking the part once takes less time and often reduces the number of dedicated functional gages necessary for the job.
Software magic
Another advantage of designing the new hybrid machines from independent CNC modules is that the machines can use modern software algorithms for compensating for various types of error, such as the slight deviations in dimension caused by tool wear. Unlike the mechanical modules in conventional transfer machines, which are set manually, the CNC modules inside the new hybrid machines can respond on the fly to feedback from off-line measurement devices. Based on changes in dimensions, the program developed during process design can adjust the pertinent offsets.
For example, a process engineer might specify that the parts go to a coordinate measuring machine or gaging station immediately after coming off the machine, while they are still clamped in the same fixture used for cutting. By not unclamping it, shop personnel can get a true measurement on the part. Although many shops ask the operators to adjust the offsets manually from time to time, the open architecture of the PC-based CNCs running on some of the machines also gives users the tools to establish continuous, closed-loop feedback and make adjustments automatically.
Most hybrid machines use the same technique to compensate for the thermal growth that occurs naturally throughout the day, rather than adjusting offsets based on temperature measurements from thermocouples or resistance tem-perature detectors at strategic points inside the machine, as many machining centers do today. D'Aoust reports that, in one application, one of his company's System 2 machines holds tolerances within 20 microns while producing 450 parts an hour. Based on feedback from a gaging station, the CNC calculates the necessary offsets to compensate for thermal growth and tool wear and adjusts the motion of the axes accordingly.
Another use for compensation software is to account for the slight variations in the fixtures fastened to the pallets, much like the pitch-error and thermal-growth compensation found in many machining centers today. "Because the fixtures on the pallets can never be exactly the same, the CNC moves the tool a few tenths to compensate for those differences," says Jones. "We 'tune' the fixtures ahead of time, and the CNC remembers the offsets." Such tools give the machine ample weapons for slaying the variation dragon.
Faster, High Tech Machines Demand Enhanced Safety Precautions
Probing the Limits: Variation as a Continual Improvement Tool
Gage R&R: The Key to Reducing Measurement Variation
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 9,785
|
describe Tracinho::WordClassifier do
describe 'full_classification' do
let(:classifications) do
[
'Segunda pessoa do singular do pretérito perfeito do indicativo do verbo mijar.',
'Primeira ou terceira pessoa do singular do pretérito imperfeito do conjuntivo do ' \
'verbo beber.',
'Primeira pessoa do plural do presente do indicativo ou pretérito perfeito do ' \
'indicativo do verbo conhecer.',
'Conjugação pronominal da segunda pessoa do singular do imperativo do verbo buscar.'
]
end
%w[mijaste bebesse conhecemos busca-mos].each_with_index do |word, index|
it "correctly returns the full classification of the word '#{word}'" do
classifier = described_class.new(Tracinho::Word.new(word))
expect(classifier.full_classification).to eq(classifications[index])
end
end
end
end
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 1,618
|
package com.vaadin.tests.components.draganddropwrapper;
import static org.junit.Assert.assertEquals;
import org.junit.Test;
import org.openqa.selenium.By;
import org.openqa.selenium.Dimension;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.interactions.Actions;
import com.vaadin.tests.tb3.MultiBrowserTest;
/**
* Test to check size of drag image element.
*
* @author Vaadin Ltd
*/
public class DragAndDropRelativeWidthTest extends MultiBrowserTest {
@Test
public void testDragImageElementSize() {
openTestURL();
WebElement label = getDriver().findElement(By.className("drag-source"));
Dimension size = label.getSize();
int height = size.getHeight();
int width = size.getWidth();
Actions actions = new Actions(getDriver());
actions.moveToElement(label);
actions.clickAndHold();
actions.moveByOffset(100, 100);
actions.build().perform();
WebElement dragImage = getDriver()
.findElement(By.className("v-drag-element"));
assertEquals("Drag image element height is unexpected", height,
dragImage.getSize().getHeight());
assertEquals("Drag image element width is unexpected", width,
dragImage.getSize().getWidth());
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 2,515
|
if (this.importScripts) {
importScripts('../../../resources/js-test.js');
importScripts('shared.js');
}
description("Test IndexedDB's basics.");
indexedDBTest(prepareDatabase);
function prepareDatabase(event)
{
trans = event.target.transaction;
db = event.target.result;
db.createObjectStore("objectStore");
objectStore = trans.objectStore("objectStore");
var req1 = objectStore.put({"key": "value"}, "object1");
var req2 = objectStore.put({"key": "value"}, "object2");
waitForRequests([req1, req2], createIndex);
}
function createIndex() {
// This will asynchronously abort in the backend because of constraint failures.
evalAndLog("objectStore.createIndex('index', 'key', {unique: true})");
// Now immediately delete it.
evalAndLog("objectStore.deleteIndex('index')");
// Delete it again: backend may have asynchronously aborted the
// index creation, but this makes sure the frontend doesn't get
// confused and crash, or think the index still exists.
evalAndExpectException("objectStore.deleteIndex('index')", "DOMException.NOT_FOUND_ERR", "'NotFoundError'");
debug("Now requesting object2");
var req3 = objectStore.get("object2");
req3.onsuccess = deleteIndexAfterGet;
req3.onerror = unexpectedErrorCallback;
debug("now we wait.");
}
function deleteIndexAfterGet() {
// so we will delete it next, but it should already be gone... right?
debug("deleteIndexAfterGet()");
// the index should still be gone, and this should not crash.
evalAndExpectException("objectStore.deleteIndex('index')", "DOMException.NOT_FOUND_ERR", "'NotFoundError'");
evalAndExpectException("objectStore.deleteIndex('index')", "DOMException.NOT_FOUND_ERR", "'NotFoundError'");
finishJSTest();
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 6,929
|
import datetime
from south.db import db
from south.v2 import SchemaMigration
from django.db import models
class Migration(SchemaMigration):
depends_on = (
("auth", "0001_initial"),
)
def forwards(self, orm):
# Adding model 'RegistrationProfile'
db.create_table(u'registration_registrationprofile', (
(u'id', self.gf('django.db.models.fields.AutoField')(primary_key=True)),
('user', self.gf('django.db.models.fields.related.ForeignKey')(to=orm['base.GoUser'], unique=True)),
('activation_key', self.gf('django.db.models.fields.CharField')(max_length=40)),
))
db.send_create_signal(u'registration', ['RegistrationProfile'])
def backwards(self, orm):
# Deleting model 'RegistrationProfile'
db.delete_table(u'registration_registrationprofile')
models = {
u'auth.group': {
'Meta': {'object_name': 'Group'},
u'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
'name': ('django.db.models.fields.CharField', [], {'unique': 'True', 'max_length': '80'}),
'permissions': ('django.db.models.fields.related.ManyToManyField', [], {'to': u"orm['auth.Permission']", 'symmetrical': 'False', 'blank': 'True'})
},
u'auth.permission': {
'Meta': {'ordering': "(u'content_type__app_label', u'content_type__model', u'codename')", 'unique_together': "((u'content_type', u'codename'),)", 'object_name': 'Permission'},
'codename': ('django.db.models.fields.CharField', [], {'max_length': '100'}),
'content_type': ('django.db.models.fields.related.ForeignKey', [], {'to': u"orm['contenttypes.ContentType']"}),
u'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
'name': ('django.db.models.fields.CharField', [], {'max_length': '50'})
},
u'base.gouser': {
'Meta': {'object_name': 'GoUser'},
'date_joined': ('django.db.models.fields.DateTimeField', [], {'default': 'datetime.datetime.now'}),
'email': ('django.db.models.fields.EmailField', [], {'unique': 'True', 'max_length': '254'}),
'first_name': ('django.db.models.fields.CharField', [], {'max_length': '254'}),
'groups': ('django.db.models.fields.related.ManyToManyField', [], {'to': u"orm['auth.Group']", 'symmetrical': 'False', 'blank': 'True'}),
u'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
'is_active': ('django.db.models.fields.BooleanField', [], {'default': 'True'}),
'is_staff': ('django.db.models.fields.BooleanField', [], {'default': 'False'}),
'is_superuser': ('django.db.models.fields.BooleanField', [], {'default': 'False'}),
'last_login': ('django.db.models.fields.DateTimeField', [], {'default': 'datetime.datetime.now'}),
'last_name': ('django.db.models.fields.CharField', [], {'max_length': '254'}),
'password': ('django.db.models.fields.CharField', [], {'max_length': '128'}),
'user_permissions': ('django.db.models.fields.related.ManyToManyField', [], {'to': u"orm['auth.Permission']", 'symmetrical': 'False', 'blank': 'True'})
},
u'contenttypes.contenttype': {
'Meta': {'ordering': "('name',)", 'unique_together': "(('app_label', 'model'),)", 'object_name': 'ContentType', 'db_table': "'django_content_type'"},
'app_label': ('django.db.models.fields.CharField', [], {'max_length': '100'}),
u'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
'model': ('django.db.models.fields.CharField', [], {'max_length': '100'}),
'name': ('django.db.models.fields.CharField', [], {'max_length': '100'})
},
u'registration.registrationprofile': {
'Meta': {'object_name': 'RegistrationProfile'},
'activation_key': ('django.db.models.fields.CharField', [], {'max_length': '40'}),
u'id': ('django.db.models.fields.AutoField', [], {'primary_key': 'True'}),
'user': ('django.db.models.fields.related.ForeignKey', [], {'to': u"orm['base.GoUser']", 'unique': 'True'})
}
}
complete_apps = ['registration']
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,193
|
Pure&Crafted Festival
The Pure&Crafted Festival's third edition went out with a bang under a starry late summer night sky. On the grounds of the Altes Kraftwerk Rummelsburg in Berlin, the festival appealed to its audience with a unique mix of motorcycle culture, lifestyle and music with Interpol, Car Seat Headrest and a variety of German artists taking the stage.
Having celebrated at Postbahnhof in the past two years, the Pure&Crafted Festival found its new home at Altes Kraftwerk Rummelsburg, a former power plant on the Spree river bank. Constructed in 1906, the iconic industrial building is a treat for the eyes; particularly its impressive, light-flooded machine hall. The hall served as the perfect backdrop for the General Store which housed about 30 brands presenting the latest new heritage trends.
Yet again, the Wheels Area was a centrepiece of the Pure&Crafted Festival. Marvelling, chatting and, of course, customising under the open sky, numerous custom workshops presented their motorcycles to the Berlin crowd. At BMW Werk Berlin's booth, it was all about delicate artistery where motorcycle lovers could have their bikes embellished with hand-painted pinstripes. The BMW Motorrad Truck invited festival-goers to take the latest models out fora spin.
Donald Ganslmeier and his team were the secret stars of the festival. Their Motodrom, the oldest running "Wall of Death", raised everyone's adrenaline levels significantly, leaving all festival-goers in awe. The festival's one-of-a-kind programme wouldn't have been complete without music. Interpol's performance was the undisputed festival highlight on Saturday. To celebrate the 15th anniversary of their debut record 'Turn On The Bright Lights', the band, a bunch of bike enthusiasts themselves, played the classic album in its entirety, bringing the sold out first day to a close.
Kytes, Pictures, Gurr, Razz, Abay and many others certainly proved that German musicians are not at all averse to guitar distortion. The festival's third edition made it clear: relaxing and Pure&Crafted go hand in hand. About 10,000 festival-goers ventured to Altes Kraftwerk Rummelsburg to celebrate the extraordinary atmosphere with family and friends over the weekend.
← Grand Prix Carraciola
The ten most expensive vehicles – auctioned at Pebble Beach →
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,030
|
require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
describe AgentAPIModel do
include APIModelMother
describe "should initialize correctly" do
it "should populate correct data" do
@agent_view_model = create_agent_model
agent_api = AgentAPIModel.new(@agent_view_model)
agent_api.uuid.should == "uuid3"
agent_api.agent_name.should == "CCeDev01"
agent_api.ip_address.should == "127.0.0.1"
agent_api.sandbox.should == "/var/lib/go-server"
agent_api.status.should == "Idle"
agent_api.build_locator.should == "/pipeline/1/stage/1/job"
agent_api.os.should == "Linux"
agent_api.free_space.should == "0 bytes"
agent_api.resources[0].should == "java"
agent_api.environments[0].should == "foo"
end
it "should handle empty data" do
@agent_view_model = create_empty_agent_model
agent_api = AgentAPIModel.new(@agent_view_model)
agent_api.uuid.should == nil
agent_api.agent_name.should == nil
agent_api.ip_address.should == nil
agent_api.sandbox.should == nil
agent_api.status.should == nil
agent_api.build_locator.should == nil
agent_api.os.should == nil
agent_api.free_space.should == nil
agent_api.resources.should == nil
agent_api.environments.should == nil
end
end
describe "should convert to json correctly" do
it "should have all fields correctly" do
@agent_view_model = create_agent_model
agents_api_arr = Array.new
agents_api_arr << AgentAPIModel.new(@agent_view_model)
ActiveSupport::JSON.decode(agents_api_arr.to_json).should == [
{
"agent_name" => "CCeDev01",
"free_space"=> "0 bytes",
"uuid" => "uuid3",
"sandbox" => "/var/lib/go-server",
"status" => "Idle",
"environments" => ["foo"],
"os" => "Linux",
"resources" => ["java"],
"ip_address" => "127.0.0.1",
"build_locator" => "/pipeline/1/stage/1/job"
}
]
end
end
end
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 2,291
|
Conostigmus canariensis is een vliesvleugelig insect uit de familie van de Megaspilidae. De wetenschappelijke naam van de soort is voor het eerst geldig gepubliceerd in 1986 door Dessart & Cancemi.
Megaspilidae
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 3,459
|
Caratteristiche tecniche
È un'ala destra.
Carriera
Cresciuto nel settore giovanile del , debutta in prima squadra il 22 dicembre 2020 giocando l'incontro di Série A perso 2-1 contro il .
Statistiche
Statistiche aggiornate al 22 gennaio 2021.
Presenze e reti nei club
Note
Collegamenti esterni
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 6,599
|
Flabellum atlanticum är en korallart som beskrevs av Stephen D. Cairns 1979. Flabellum atlanticum ingår i släktet Flabellum och familjen Flabellidae. Inga underarter finns listade i Catalogue of Life.
Källor
Stenkoraller
atlanticum
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 8,455
|
It's been a good ride
"Will I miss it? Sure" Speaking about his retirement from education, Blooming Prairie Schools Superintendent Barry Olson says his education career of 43 years "has been a good ride" for him. Olson officially steps down as Blooming Prairie School District superintendent on July 1. Enter Chris...
Owatonna voters reject new high school
By a slim margin, voters in the Owatonna School District rejected a plan to build a new high school. During a special election on Tuesday, May 14, the bond referendum failed by only 120 votes. Final results show 5,762 voters or 50.52% said no to the referendum while 5,642 voters or 49.47% said yes...
Special athletes compete in BP
It was an exciting day for 150 special education kids on Friday, May 10 at the Blooming Prairie High School track. These students in grades K-6 were the star attractions on this special Zumbro Education District (ZED) Activity Day. It was the 29th annual ZED Activity Day. Mostly sunny skies...
'Let me check my phone'
"How many days of sobriety are you at?" Destiny Reich, 28, of Owatonna, replied to a friend's inquiry, "Let me check my phone." Her research on her phone revealed 791 days. These 791 days reflected significant progress in the Steele Waseca Drug Court program and meant that she represents the 43rd...
Area students are getting a chance to explore art by getting their hands dirty and molding pottery. Last year Medford students were given the opportunity to participate in a pottery class at the Owatonna Arts Center, and this year it was Blooming Prairie students who tried their hand at the wheel....
Warrior Wagons help children's cancer victims
It's about building trust in the Lord. That was an emotional message brought to a group of mostly women gathered at First Lutheran Church of Blooming Prairie on a chilly May Day. Attendance was listed at 120. It was a third annual Guest Day where Lutheran Church Women of First Lutheran invited...
WORLD HALL OF FAME
Seventy-five years after he played for a Farmer's Union meeting with his button accordion at 8 years old, Luverne Wanous is being recognized for his contributions to performing old-time music and being hailed as a hall of famer. On Tuesday, May 14, Wanous, of Owatonna, will be inducted into the...
More support for new school
Just one week before the school referendum was set to go before voters, another business joined in support of a new high school for Owatonna. On May 6, Gopher Sport announced that if the referendum passes on May 14, it wants to contribute equipment to ensure students at the high school have more...
BP feeds starving children
"For I was hungry and you gave me food." -- Matthew 25:35 This Bible verse printed on Feed My Starving Children Mobilepack t-shirts symbolized a community-wide effort in Blooming Prairie on Friday and Saturday, May 10-11, to provide food for those in need around the world. First Lutheran Church...
BREAKING NEWS: Owatonna Bond Referendum Fails
By a narrow margin, the bond referendum for a new high school in Owatonna has failed. Voters in the Owatonna School District rejected the referendum during a special election Tuesday. The referendum was defeated with 5,762 no votes to 5,642 yes votes. The widest disparity of votes arose out of the...
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,536
|
<?xml version="1.0" encoding="utf-8"?>
<RelativeLayout xmlns:android="http://schemas.android.com/apk/res/android"
xmlns:tools="http://schemas.android.com/tools"
android:layout_width="match_parent"
android:layout_height="match_parent"
tools:context="com.smart.cloud.fire.activity.AddDev.ChioceDevTypeActivity">
<TextView
android:id="@+id/title_tv"
android:layout_width="match_parent"
android:layout_height="45dp"
android:text="添加设备"
android:textColor="#ffffff"
android:background="@color/login_btn"
android:gravity="center"
android:textSize="18sp"
/>
<ScrollView
android:layout_width="match_parent"
android:layout_height="wrap_content">
<RelativeLayout
android:layout_width="match_parent"
android:layout_height="match_parent"
android:layout_below="@id/title_tv"
android:background="@drawable/tianjia_bj">
<TextView
android:id="@+id/chioce_text"
android:layout_width="match_parent"
android:layout_height="wrap_content"
android:layout_marginTop="20dp"
android:text="选择设备类型:"
android:textSize="16sp"
android:gravity="center"/>
<RelativeLayout
android:layout_width="wrap_content"
android:layout_height="wrap_content"
android:layout_centerHorizontal="true"
android:layout_below="@id/chioce_text"
android:layout_marginTop="15dp">
<LinearLayout
android:id="@+id/sdtj_lin"
android:layout_width="150dp"
android:layout_height="wrap_content"
android:layout_marginTop="10dp"
android:orientation="vertical"
android:padding="20dp"
android:background="#fff"
android:layout_margin="5dp">
<ImageButton
android:id="@+id/sdsr_btn"
android:layout_width="60dp"
android:layout_height="60dp"
android:layout_gravity="center"
android:layout_margin="15dp"
android:background="@drawable/tianjia_shoudong"/>
<TextView
android:layout_width="100dp"
android:layout_height="wrap_content"
android:text="手动添加"
android:layout_marginTop="10dp"
android:gravity="center"
android:textSize="14sp"
android:textColor="#fff"
android:padding="10dp"
android:layout_margin="10dp"
android:layout_gravity="center"
android:background="@drawable/corner_blue"/>
</LinearLayout>
<LinearLayout
android:id="@+id/smsr_lin"
android:layout_width="150dp"
android:layout_height="wrap_content"
android:layout_marginTop="10dp"
android:layout_toRightOf="@id/sdtj_lin"
android:orientation="vertical"
android:padding="20dp"
android:background="#fff"
android:layout_margin="5dp">
<ImageButton
android:id="@+id/smsr_btn"
android:layout_width="60dp"
android:layout_height="60dp"
android:layout_gravity="center"
android:layout_margin="15dp"
android:background="@drawable/tianjia_saoma"/>
<TextView
android:layout_width="100dp"
android:layout_height="wrap_content"
android:text="扫码添加"
android:layout_marginTop="10dp"
android:gravity="center"
android:textSize="14sp"
android:textColor="#fff"
android:padding="10dp"
android:layout_margin="10dp"
android:layout_gravity="center"
android:background="@drawable/corner_blue"/>
</LinearLayout>
<LinearLayout
android:id="@+id/spjk_lin"
android:layout_width="150dp"
android:layout_height="wrap_content"
android:layout_marginTop="10dp"
android:layout_below="@id/sdtj_lin"
android:orientation="vertical"
android:padding="20dp"
android:background="#fff"
android:layout_margin="5dp">
<ImageButton
android:id="@+id/spjk_btn"
android:layout_width="60dp"
android:layout_height="60dp"
android:layout_gravity="center"
android:layout_margin="10dp"
android:background="@drawable/tianjia_camera"/>
<TextView
android:layout_width="100dp"
android:layout_height="wrap_content"
android:text="视频监控"
android:layout_marginTop="10dp"
android:gravity="center"
android:textSize="14sp"
android:textColor="#fff"
android:padding="10dp"
android:layout_margin="15dp"
android:layout_gravity="center"
android:background="@drawable/corner_blue"/>
</LinearLayout>
<LinearLayout
android:id="@+id/nfc_lin"
android:layout_width="150dp"
android:layout_height="wrap_content"
android:layout_marginTop="10dp"
android:layout_below="@id/sdtj_lin"
android:layout_toRightOf="@id/sdtj_lin"
android:orientation="vertical"
android:padding="20dp"
android:background="#fff"
android:layout_margin="5dp">
<ImageButton
android:id="@+id/nfc_btn"
android:layout_width="60dp"
android:layout_height="60dp"
android:layout_gravity="center"
android:layout_margin="10dp"
android:background="@drawable/tianjia_nfc2"/>
<TextView
android:layout_width="100dp"
android:layout_height="wrap_content"
android:text="NFC"
android:layout_marginTop="10dp"
android:gravity="center"
android:textSize="14sp"
android:textColor="#fff"
android:padding="10dp"
android:layout_margin="15dp"
android:layout_gravity="center"
android:background="@drawable/corner_blue"/>
</LinearLayout>
<LinearLayout
android:id="@+id/yongchuang_lin"
android:layout_width="150dp"
android:layout_height="wrap_content"
android:layout_marginTop="10dp"
android:layout_below="@id/spjk_lin"
android:orientation="vertical"
android:padding="20dp"
android:background="#fff"
android:layout_margin="5dp">
<ImageButton
android:id="@+id/yongchuan_btn"
android:layout_width="60dp"
android:layout_height="60dp"
android:layout_gravity="center"
android:layout_margin="10dp"
android:background="@drawable/yongchuan"/>
<TextView
android:layout_width="100dp"
android:layout_height="wrap_content"
android:text="传输装置"
android:layout_marginTop="10dp"
android:gravity="center"
android:textSize="14sp"
android:textColor="#fff"
android:padding="10dp"
android:layout_margin="15dp"
android:layout_gravity="center"
android:background="@drawable/corner_blue"/>
</LinearLayout>
<LinearLayout
android:id="@+id/virtual_point_lin"
android:layout_width="150dp"
android:layout_height="wrap_content"
android:layout_marginTop="10dp"
android:layout_below="@id/spjk_lin"
android:layout_toRightOf="@id/yongchuang_lin"
android:orientation="vertical"
android:padding="20dp"
android:background="#fff"
android:layout_margin="5dp">
<ImageButton
android:id="@+id/virtual_point_btn"
android:layout_width="60dp"
android:layout_height="60dp"
android:layout_gravity="center"
android:layout_margin="10dp"
android:background="@drawable/tianjia_nfc2"/>
<TextView
android:layout_width="100dp"
android:layout_height="wrap_content"
android:text="虚拟点位"
android:layout_marginTop="10dp"
android:gravity="center"
android:textSize="14sp"
android:textColor="#fff"
android:padding="10dp"
android:layout_margin="15dp"
android:layout_gravity="center"
android:background="@drawable/corner_blue"/>
</LinearLayout>
</RelativeLayout>
</RelativeLayout>
</ScrollView>
</RelativeLayout>
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3,689
|
namespace GUI
{
class Button : public Component
{
public:
typedef std::shared_ptr<Button> Ptr;
typedef std::function<void()> Callback;
enum Type
{
Normal,
Selected,
Pressed,
ButtonCount
};
public:
Button(const FontHolder& fonts, const TextureHolder& textures);
void setCallBack(Callback callback);
void setText(const std::string& text);
void setText(const std::string& text, int charSize);
void setToggle(bool flag);
virtual bool isSelectable() const;
virtual void select();
virtual void deselect();
virtual void activate();
virtual void deactivate();
virtual void handleEvent(const sf::Event& event);
private:
virtual void draw(sf::RenderTarget& target, sf::RenderStates states) const;
void changeTexture(Type buttonType);
private:
Callback mCallback;
sf::Sprite mSprite;
sf::Text mText;
bool mIsToggle;
};
}
#endif // BUTTON_HPP
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 169
|
Congratulations to Maj Steven DeFord, for completing his IFR Form 5. Out of 9 powered aircraft pilots, with Maj DeFord, Squadron 188 now has 5 pilots who have completed their IFR Form 5. In addition, the squadron has 2 glider pilots.
Also, thanks to Maj Jeff Ironfield for helping Maj DeFord with his training and to Capt Michelogiannakis, who was the IFR Form 5 check pilot.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,204
|
package org.apache.isis.core.metamodel.facets.collections.layout;
import org.apache.isis.applib.annotation.When;
import org.apache.isis.applib.annotation.Where;
import org.apache.isis.applib.layout.component.CollectionLayoutData;
import org.apache.isis.core.metamodel.adapter.ObjectAdapter;
import org.apache.isis.core.metamodel.facetapi.FacetHolder;
import org.apache.isis.core.metamodel.facets.all.hide.HiddenFacet;
import org.apache.isis.core.metamodel.facets.members.hidden.HiddenFacetAbstract;
public class HiddenFacetForCollectionXml extends HiddenFacetAbstract {
public static HiddenFacet create(final CollectionLayoutData collectionLayout, final FacetHolder holder) {
if (collectionLayout == null) {
return null;
}
final Where where = collectionLayout.getHidden();
return where != null && where != Where.NOT_SPECIFIED ? new HiddenFacetForCollectionXml(where, holder) : null;
}
private HiddenFacetForCollectionXml(final Where where, final FacetHolder holder) {
super(When.ALWAYS, where, holder);
}
@Override
public String hiddenReason(final ObjectAdapter targetAdapter, final Where whereContext) {
if(!where().includes(whereContext)) {
return null;
}
return "Hidden on " + where().getFriendlyName();
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,294
|
EXCLUSIVE: Anthony Yigit – "Still Hungry…"
By Craig Scott 16th October 2019
Sweden hasn't typically been a country known for its boxing, of late. Of course, former heavy-handed champion Ingemar Johansson of Gothenburg, shocked Floyd Patterson, stopping the American heavyweight sixty years ago. However, the country placed a thirty-six year ban on boxing, only lifting it partially in 2006.
More recently, Badou Jack has spilled plenty of blood, winning world titles in two divisions – yet is one of only three Swedish men to reach the pinnacle of the sport. A focus on other leisure activities, such as handball, football or golf have hindered the next generation of Swedish boxers.
As another eight-man tournament rapidly approaches, former world title challenger, Anthony Yigit (24-1, 8KOs) was excitable, when discussing his return to the ring. Starring in one of last year's most striking images, Yigit's eye had ballooned grotesquely during his World Boxing Super Series quarter-final with Belarusian former IBF champion, Ivan Baranchyk. But that was then and this is now.
Ahead of the next instalment of MTK's 'Golden Contract' tournament at light-welterweight, Yigit was keen to discuss his recovery from that injury and his introduction to boxing, whilst living in a city that often turned the other cheek. He is Swedish, by birth, of course. But a cocktail of different nationalities had given Yigit exposure to various walks of life.
"I'm going to tell you the whole story", he began, "I'm from a Russian family, they are Jewish-Russian, yeah. My Finnish family, they are originally Swedish and my Turkish family I think, obviously in the Middle East they moved around a lot from the Middle Ages and stuff like that, so my Turkish family we are originally from Syria I think, or Lebanon, I'm not quite sure. I don't remember, really."
"We are from this tribe and it's a whole messy story. Look, if I wanted to check my heritage it wouldn't be possible. But I'm always saying, 'I'm a Jewish-Muslim from Russia, Turkey and Sweden'. That's where I come from. You put some of this and some of that and you get a really nice drink, and that's me."
This Jewish-Muslim from Russia, Turkey and Sweden sounds like an intelligent Englishman at the other end of the phone, speaking as well as any international fighter. After hearing stories from former training partners or members of the media who'd dealt with Anthony, it was clear he understood the value of telling his own story. Boxing hadn't always been a part of his life, it drifted in and out as Yigit continued crossing borders.
"I was staying with my mom, because my mom and dad split up when I was young. She remarried and my sister came from that relationship, but then they broke up. She got married again and my two other siblings come from that [relationship]. She stayed with him for a long time, so he raised me and I always saw him as my father figure."
"When I was twelve, I got to meet my real father and he asked me if I was doing any sports. I was doing judo at the time, but my dad said, 'Listen, I know this boxing trainer'. I went there and I could train for free, but sometimes I'd train, sometimes I didn't. Then my mum moved us to Turkey for three years. When I was fifteen, I decided to move back to Sweden and live with my father – that's when I started boxing for real."
Yigit hit the ground running once boxing had become his priority, excelling in the Swedish amateur system. He'd been uprooted countless times, crossing borders and also suffering a spell of homelessness when ending his career in the unpaid ranks. His father kicked him out and the phone stopped ringing. Offers from overseas that seemed concrete, soon turned to quicksand which gave him an early insight into life as a professional.
His time as an Olympian at London 2012 seemed to count for little, and despite being the first Swedish fighter to win a bout at the Olympics in over fifteen years, he played the waiting game. Before taking matters into his own hands, he was linked with boxing super promotion Top Rank, but decided to reach out to Team Sauerland, far closer to home. A powerful operation in Europe, particularly in Scandinavia, the company have worked with their stylish prodigy, guiding him towards commercial success and world titles.
Never quite settling in one area throughout his childhood, the wandering Yigit has already been based in various countries as a professional fighter, such as Sweden, Germany, England and Spain. He couldn't enjoy home comforts – he was made to travel. In doing so, he'd beaten British duo Joe Hughes and Lenny Daws for his European title – bizarrely fighting the latter in little-heralded boxing town, Carshalton, Surrey.
The Swede's clean-cut image and immaculate behaviour could potentially fool the casual observer. Until his fight with Baranchyk, Yigit hadn't faced such adversity head-on between the ropes. That evening in New Orleans, 'Can You Dig It?' showed his mettle, disproving those who accused him of sidestepping Josh Taylor only a couple of years prior. The two could have infact been drawn together, however the IBF had opted for Baranchyk v Yigit to contest their vacant title – ironically now held by the Edinburgh-man.
Recently returning from his warm weather training camp in Las Palmas, Spain, he was now primed and ready to tackle some of the division's best fighters. Ohara Davies, Darren Surtees and Mohamed Mimoune are amongst potential opponents for Yigit, preparing to fight at an adopted second home in York Hall, Bethnal Green. Opponents didn't bother the former European champion and losing was just a consequence of hunting success. Strangely, though, a tattooed Northern Irishman had caught his eye during the build-up.
"Here we are now, approaching this 'Golden Contract' tournament and Tyrone McKenna, well actually he's quite new to me. I didn't know who he was before, but I watched him fight, he is a terrific fighter and I think the tournament is really gonna be a blast. Sky Sports are going to cover it, so I like that. It shows everyone that, okay, I'm still in the game, I'm still hungry and I'm still looking to be a world champion.
"That's what we do this for. We are supposed to fight each other", he added. "We are supposed to fight whoever comes in front of us. If you don't, you're not a real champion, and that's my mentality. I'm not scared about having a loss or two losses. For example, my loss is from this injury. A doctor stopped the fight, so I didn't lose. I didn't lose against one of the hardest hitters in this business. After the fight he was a world champion, and I didn't 'lose'.
"Would I be embarrassed to lose [again]? No. I'll make sure that, hey, I didn't lose because I'm bad, I'm losing because I'm one of the best fighting against the best, that's why. You'll always see me fight the best, because I belong there."
The sense of belonging wasn't familiar to Anthony, after spending so long finding his feet as an adolescent. To fully understand that boxing provided his stability had only strengthened his love for the sport. Yigit realised he could infact achieve what many others could only dream of.
The 'Golden Contract' is staged four weeks after the conclusion of the light-welterweight World Boxing Super Series, with only a twenty minute tube journey between both venues. Josh Taylor faces Regis Prograis at London's O2 in a unification clash, set for Box Office. In many ways, the Swedish family man's entry into the tournament had been a ticket to the big time – relatively short-lived.
As a beaten quarter-finalist, Yigit was looking forward to its conclusion, remembering his selection fondly. The beams of platinum light shooting upwards and the smoke-filled podiums had taken boxing by storm. It just didn't quite go to plan. Many felt he was edging himself into a slender lead during the fight, but the damage had already been sustained. It was painful, watching the enormous welt swelling by the second, continually being targeted by his opponent.
"It's always a dream come true just to partake in such a big event and the World Boxing Super Series seemed to be the center of attention", explained the twenty-eight year old. "Now it's going to be big again, when Josh Taylor fights Regis Prograis. It was just a big honour to be on that stage and to be fighting with these world-class fighters. I got to fight Ivan Baranchyk and it was a great fight."
"Watching it back, I can see some things that I need to work on, that I maybe overlooked. If I really want to be able to break through and make it in the big leagues, and be a world champion, I need to work on certain things that I didn't work on before. I just know that there is a reason I didn't win and I need to fix that."
"To be honest with you, we knew [the injury was coming]. I had problems with not only that eye, but my eyes, because if you look back on other fights, my eyes tend to swell up. We said before the Ivan Baranchyk fight, 'Yeah, okay, so we need to just put ice on the eyes from the beginning and make sure they don't swell up.'"
"You can see me complain to the doctor, it's because I was like, 'Hey, my eye's been shut for quite a while, so if you wanted to stop the fight why didn't you stop it earlier? If you didn't, then just let me continue.' My coach actually said, he was like, 'Well, listen to me, trust me, it's not looking good.' After, I went down to the changing room and I could see my eye, and then I was like, 'Oh, okay.'"
With his eye almost as big as his heart, the damaged fighter was determined to continue. It was never going to happen. At the time of writing, we keep Patrick Day in our thoughts, a fighter that battled and battled, ultimately taking far too much punishment. Boxing is a sport, but it's often a brutal meeting of machismo and misjudgement.
"I mean, now that I think back to it, I'm at peace with it. I'm not the kind of guy that has to have a spotless record and I just believe… it's not the end of the world. I still have the chance to be a world champion. I'm young, let's just take it from there."
"With that being said, would I take the chance [and continue]? I would, but if something would have happened I would have looked at someone else to blame. Obviously then it's not really my choice to make, is it? It is the doctor's choice to make and he made it."
That choice was taken from one of Yigit's dearest friends, countryman and mentor, Erik Skoglund. The pair had just finished training together in Spain, with former light-heavyweight and super-middleweight contender, Skoglund, continuing his remarkable rehabilitation. He suffered a serious brain injury after a routine training session, something that had shaken up boxing, especially in Sweden.
The nation that had previously banned boxing seemed vindicated as stories of Skoglund's condition circulated the national press. Erik was like an older brother to Anthony, with the pair standing side-by-side for the last decade. Given the severity of his own injury when fighting Ivan Baranchyk, it seemed strange that Yigit would have wanted to continue. But fighters were a strange breed after all.
"The weird part is, I wasn't afraid about him not waking up. I was more afraid about how he would feel when he did wake up. Because boxing has been his life and one day, you just wake up and they tell you that you can't do it anymore. It's heartbreaking. You invest your life into it and then somebody tells you, 'Okay – no more boxing'. I'm just sad about that. Erik really struggled coping with that.
"Bleeding isn't so hard. It's easy. You get punched and you bleed. The hard part about being a fighter is going through something like Erik has went through. Continuing to push through and look at him now. He's back and he's been training with me. He's crazy. That's a real fighter.
"I decided I would make him a part of my team and Erik tells me that he's never been more alive than when I bring him to the fights. He can be in the corner – even though he can never fight – but as long as he's in that environment, he feels as though it never left him."
With Erik Skoglund by his side, Yigit seemed stronger than ever. The inspiring story of the boy thrown from pillar-to-post, travelling from Sweden to Turkey and back again, was yet to reach its conclusion. He's been given a second chance, preparing to compete for the 'Golden Contract', yet boxing was deeper than that for Anthony.
He spoke with passion and dissected the sport's flaws, opening up over the struggles he'd faced mentally, stemming from the isolation forced upon fighters. When he wanted to socialise, nobody was interested. Friends and family had been unable to connect due to rigorous training schedules and time spent overseas, locked away, punishing both the body and the mind.
Becoming a world champion remained his priority – that much was clear. Yigit understands that fighters are often forgotten when the final bell tolls, but legacy can last forever. This seemed important to him. For his family, the wide range of ingredients that produced a 'nice cocktail', for Erik – who'd been stripped of the opportunity – and for himself. The journey was far from over.
"I want people to say that I never backed out of a fight. I want to make an impact, for the crowd. I don't want people to say, 'He was a champion, but he didn't fight the best', or, 'He was a coward'. I want them to say that I was an inspiration and that they wanted to watch me fight.
"Look at Arturo Gatti v Mickey Ward. Who won those fights? Nobody really cares. They just wanna watch those guys fight every time. They watch the highlights over and over again. That's what I want. I want them to YouTube me [when I retire] – whether I win or lose, I just want them to watch."
Interview written by: Craig Scott
Follow Craig on Twitter at: @craigscott209
Light Heavyweight Golden Contract Semi-Finals Set ...
Deion Jumah and Sam Hyde Clash for English Title i...
Jake Paul and AnEsonGib Clash in Fiery Press Confe...
Prospect Rivera has Chance to Impress Against Mald...
Unbeaten Danny Dignum and Alfredo Meli Collide on ...
Russell Jr Set To Face Nyambayar On February 8
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,168
|
#import "RNSVGNode.h"
#import "RNSVGContainer.h"
#import "RNSVGClipPath.h"
#import "RNSVGGroup.h"
#import "RNSVGGlyphContext.h"
@interface RNSVGNode()
@property (nonatomic, readwrite, weak) RNSVGSvgView *svgView;
@property (nonatomic, readwrite, weak) RNSVGGroup *textRoot;
@end
@implementation RNSVGNode
{
RNSVGGlyphContext *glyphContext;
BOOL _transparent;
RNSVGClipPath *_clipNode;
CGPathRef _cachedClipPath;
CGFloat canvasWidth;
CGFloat canvasHeight;
CGFloat canvasDiagonal;
}
CGFloat const RNSVG_M_SQRT1_2l = (CGFloat)0.707106781186547524400844362104849039;
CGFloat const RNSVG_DEFAULT_FONT_SIZE = 12;
- (instancetype)init
{
if (self = [super init]) {
self.opacity = 1;
self.opaque = false;
self.matrix = CGAffineTransformIdentity;
self.transforms = CGAffineTransformIdentity;
self.invTransform = CGAffineTransformIdentity;
_merging = false;
_dirty = false;
}
return self;
}
- (void)insertReactSubview:(RNSVGView *)subview atIndex:(NSInteger)atIndex
{
[super insertReactSubview:subview atIndex:atIndex];
[self insertSubview:subview atIndex:atIndex];
[self invalidate];
}
- (void)removeReactSubview:(RNSVGView *)subview
{
[super removeReactSubview:subview];
[self invalidate];
}
- (void)didUpdateReactSubviews
{
// Do nothing, as subviews are inserted by insertReactSubview:
}
- (void)invalidate
{
if (_dirty || _merging) {
return;
}
_dirty = true;
id<RNSVGContainer> container = (id<RNSVGContainer>)self.superview;
[container invalidate];
[self clearPath];
canvasWidth = -1;
canvasHeight = -1;
canvasDiagonal = -1;
}
- (void)clearPath
{
CGPathRelease(_path);
self.path = nil;
}
- (void)clearChildCache
{
[self clearPath];
for (__kindof RNSVGNode *node in self.subviews) {
if ([node isKindOfClass:[RNSVGNode class]]) {
[node clearChildCache];
}
}
}
- (void)clearParentCache
{
RNSVGNode* node = self;
while (node != nil) {
RNSVGPlatformView* parent = [node superview];
if (![parent isKindOfClass:[RNSVGNode class]]) {
return;
}
node = (RNSVGNode*)parent;
if (!node.path) {
return;
}
[node clearPath];
}
}
- (RNSVGGroup *)textRoot
{
if (_textRoot) {
return _textRoot;
}
RNSVGNode* node = self;
while (node != nil) {
if ([node isKindOfClass:[RNSVGGroup class]] && [((RNSVGGroup*) node) getGlyphContext] != nil) {
_textRoot = (RNSVGGroup*)node;
break;
}
RNSVGPlatformView* parent = [node superview];
if (![node isKindOfClass:[RNSVGNode class]]) {
node = nil;
} else {
node = (RNSVGNode*)parent;
}
}
return _textRoot;
}
- (RNSVGGroup *)getParentTextRoot
{
RNSVGNode* parent = (RNSVGGroup*)[self superview];
if (![parent isKindOfClass:[RNSVGGroup class]]) {
return nil;
} else {
return parent.textRoot;
}
}
- (CGFloat)getFontSizeFromContext
{
RNSVGGroup* root = self.textRoot;
if (root == nil) {
return RNSVG_DEFAULT_FONT_SIZE;
}
if (glyphContext == nil) {
glyphContext = [root getGlyphContext];
}
return [glyphContext getFontSize];
}
- (void)reactSetInheritedBackgroundColor:(RNSVGColor *)inheritedBackgroundColor
{
self.backgroundColor = inheritedBackgroundColor;
}
- (void)setPointerEvents:(RCTPointerEvents)pointerEvents
{
_pointerEvents = pointerEvents;
self.userInteractionEnabled = (pointerEvents != RCTPointerEventsNone);
if (pointerEvents == RCTPointerEventsBoxNone) {
#if TARGET_OS_OSX
self.accessibilityModal = NO;
#else
self.accessibilityViewIsModal = NO;
#endif
}
}
- (void)setName:(NSString *)name
{
if ([name isEqualToString:_name]) {
return;
}
[self invalidate];
_name = name;
}
- (void)setDisplay:(NSString *)display
{
if ([display isEqualToString:_display]) {
return;
}
[self invalidate];
_display = display;
}
- (void)setOpacity:(CGFloat)opacity
{
if (opacity == _opacity) {
return;
}
if (opacity <= 0) {
opacity = 0;
} else if (opacity > 1) {
opacity = 1;
}
[self invalidate];
_transparent = opacity < 1;
_opacity = opacity;
}
- (void)setMatrix:(CGAffineTransform)matrix
{
if (CGAffineTransformEqualToTransform(matrix, _matrix)) {
return;
}
_matrix = matrix;
_invmatrix = CGAffineTransformInvert(matrix);
id<RNSVGContainer> container = (id<RNSVGContainer>)self.superview;
[container invalidate];
}
- (void)setClientRect:(CGRect)clientRect {
if (CGRectEqualToRect(_clientRect, clientRect)) {
return;
}
_clientRect = clientRect;
if (self.onLayout) {
self.onLayout(@{
@"layout": @{
@"x": @(_clientRect.origin.x),
@"y": @(_clientRect.origin.y),
@"width": @(_clientRect.size.width),
@"height": @(_clientRect.size.height),
}
});
}
}
- (void)setClipPath:(NSString *)clipPath
{
if ([_clipPath isEqualToString:clipPath]) {
return;
}
CGPathRelease(_cachedClipPath);
_cachedClipPath = nil;
_clipPath = clipPath;
[self invalidate];
}
- (void)setClipRule:(RNSVGCGFCRule)clipRule
{
if (_clipRule == clipRule) {
return;
}
CGPathRelease(_cachedClipPath);
_cachedClipPath = nil;
_clipRule = clipRule;
[self invalidate];
}
- (void)setMask:(NSString *)mask
{
if ([_mask isEqualToString:mask]) {
return;
}
_mask = mask;
[self invalidate];
}
- (void)setMarkerStart:(NSString *)markerStart
{
if ([_markerStart isEqualToString:markerStart]) {
return;
}
_markerStart = markerStart;
[self invalidate];
}
- (void)setMarkerMid:(NSString *)markerMid
{
if ([_markerMid isEqualToString:markerMid]) {
return;
}
_markerMid = markerMid;
[self invalidate];
}
- (void)setMarkerEnd:(NSString *)markerEnd
{
if ([_markerEnd isEqualToString:markerEnd]) {
return;
}
_markerEnd = markerEnd;
[self invalidate];
}
- (void)beginTransparencyLayer:(CGContextRef)context
{
if (_transparent) {
CGContextBeginTransparencyLayer(context, NULL);
}
}
- (void)endTransparencyLayer:(CGContextRef)context
{
if (_transparent) {
CGContextEndTransparencyLayer(context);
}
}
- (void)renderTo:(CGContextRef)context rect:(CGRect)rect
{
self.dirty = false;
// abstract
}
- (CGPathRef)getClipPath
{
return _cachedClipPath;
}
- (CGPathRef)getClipPath:(CGContextRef)context
{
if (self.clipPath) {
_clipNode = (RNSVGClipPath*)[self.svgView getDefinedClipPath:self.clipPath];
if (_cachedClipPath) {
CGPathRelease(_cachedClipPath);
}
CGAffineTransform transform = CGAffineTransformConcat(_clipNode.matrix, _clipNode.transforms);
_cachedClipPath = CGPathCreateCopyByTransformingPath([_clipNode getPath:context], &transform);
}
return _cachedClipPath;
}
- (void)clip:(CGContextRef)context
{
CGPathRef clipPath = [self getClipPath:context];
if (clipPath) {
CGContextAddPath(context, clipPath);
if (_clipRule == kRNSVGCGFCRuleEvenodd) {
CGContextEOClip(context);
} else {
CGContextClip(context);
}
}
}
- (CGPathRef)getPath: (CGContextRef)context
{
// abstract
return nil;
}
- (void)renderLayerTo:(CGContextRef)context rect:(CGRect)rect
{
// abstract
}
// hitTest delagate
- (RNSVGPlatformView *)hitTest:(CGPoint)point withEvent:(UIEvent *)event
{
// abstract
return nil;
}
- (RNSVGSvgView *)svgView
{
if (_svgView) {
return _svgView;
}
__kindof RNSVGPlatformView *parent = self.superview;
if ([parent class] == [RNSVGSvgView class]) {
_svgView = parent;
} else if ([parent isKindOfClass:[RNSVGNode class]]) {
_svgView = ((RNSVGNode *)parent).svgView;
} else {
RCTLogError(@"RNSVG: %@ should be descendant of a SvgViewShadow.", NSStringFromClass(self.class));
}
return _svgView;
}
- (CGFloat)relativeOnWidthString:(NSString *)length
{
return [RNSVGPropHelper fromRelativeWithNSString:length
relative:[self getCanvasWidth]
fontSize:[self getFontSizeFromContext]];
}
- (CGFloat)relativeOnHeightString:(NSString *)length
{
return [RNSVGPropHelper fromRelativeWithNSString:length
relative:[self getCanvasHeight]
fontSize:[self getFontSizeFromContext]];
}
- (CGFloat)relativeOnOtherString:(NSString *)length
{
return [RNSVGPropHelper fromRelativeWithNSString:length
relative:[self getCanvasDiagonal]
fontSize:[self getFontSizeFromContext]];
}
- (CGFloat)relativeOn:(RNSVGLength *)length relative:(CGFloat)relative
{
RNSVGLengthUnitType unit = length.unit;
if (unit == SVG_LENGTHTYPE_NUMBER){
return length.value;
} else if (unit == SVG_LENGTHTYPE_PERCENTAGE){
return length.value / 100 * relative;
}
return [self fromRelative:length];
}
- (CGFloat)relativeOnWidth:(RNSVGLength *)length
{
RNSVGLengthUnitType unit = length.unit;
if (unit == SVG_LENGTHTYPE_NUMBER){
return length.value;
} else if (unit == SVG_LENGTHTYPE_PERCENTAGE){
return length.value / 100 * [self getCanvasWidth];
}
return [self fromRelative:length];
}
- (CGFloat)relativeOnHeight:(RNSVGLength *)length
{
RNSVGLengthUnitType unit = length.unit;
if (unit == SVG_LENGTHTYPE_NUMBER){
return length.value;
} else if (unit == SVG_LENGTHTYPE_PERCENTAGE){
return length.value / 100 * [self getCanvasHeight];
}
return [self fromRelative:length];
}
- (CGFloat)relativeOnOther:(RNSVGLength *)length
{
RNSVGLengthUnitType unit = length.unit;
if (unit == SVG_LENGTHTYPE_NUMBER){
return length.value;
} else if (unit == SVG_LENGTHTYPE_PERCENTAGE){
return length.value / 100 * [self getCanvasDiagonal];
}
return [self fromRelative:length];
}
- (CGFloat)fromRelative:(RNSVGLength*)length {
CGFloat unit;
switch (length.unit) {
case SVG_LENGTHTYPE_EMS:
unit = [self getFontSizeFromContext];
break;
case SVG_LENGTHTYPE_EXS:
unit = [self getFontSizeFromContext] / 2;
break;
case SVG_LENGTHTYPE_CM:
unit = (CGFloat)35.43307;
break;
case SVG_LENGTHTYPE_MM:
unit = (CGFloat)3.543307;
break;
case SVG_LENGTHTYPE_IN:
unit = 90;
break;
case SVG_LENGTHTYPE_PT:
unit = 1.25;
break;
case SVG_LENGTHTYPE_PC:
unit = 15;
break;
default:
unit = 1;
}
return length.value * unit;
}
- (CGRect)getContextBounds
{
return CGContextGetClipBoundingBox(UIGraphicsGetCurrentContext());
}
- (CGFloat)getContextWidth
{
return CGRectGetWidth([self getContextBounds]);
}
- (CGFloat)getContextHeight
{
return CGRectGetHeight([self getContextBounds]);
}
- (CGFloat)getContextDiagonal {
CGRect bounds = [self getContextBounds];
CGFloat width = CGRectGetWidth(bounds);
CGFloat height = CGRectGetHeight(bounds);
CGFloat powX = width * width;
CGFloat powY = height * height;
CGFloat r = sqrt(powX + powY) * RNSVG_M_SQRT1_2l;
return r;
}
- (CGFloat) getCanvasWidth {
if (canvasWidth != -1) {
return canvasWidth;
}
RNSVGGroup* root = [self textRoot];
if (root == nil) {
canvasWidth = [self getContextWidth];
} else {
canvasWidth = [[root getGlyphContext] getWidth];
}
return canvasWidth;
}
- (CGFloat) getCanvasHeight {
if (canvasHeight != -1) {
return canvasHeight;
}
RNSVGGroup* root = [self textRoot];
if (root == nil) {
canvasHeight = [self getContextHeight];
} else {
canvasHeight = [[root getGlyphContext] getHeight];
}
return canvasHeight;
}
- (CGFloat) getCanvasDiagonal {
if (canvasDiagonal != -1) {
return canvasDiagonal;
}
CGFloat width = [self getCanvasWidth];
CGFloat height = [self getCanvasHeight];
CGFloat powX = width * width;
CGFloat powY = height * height;
canvasDiagonal = sqrt(powX + powY) * RNSVG_M_SQRT1_2l;
return canvasDiagonal;
}
- (void)parseReference
{
self.dirty = false;
if (self.name) {
typeof(self) __weak weakSelf = self;
[self.svgView defineTemplate:weakSelf templateName:self.name];
}
}
- (void)traverseSubviews:(BOOL (^)(__kindof RNSVGView *node))block
{
for (RNSVGView *node in self.subviews) {
if (!block(node)) {
break;
}
}
}
- (void)dealloc
{
CGPathRelease(_cachedClipPath);
CGPathRelease(_strokePath);
CGPathRelease(_path);
}
@end
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3
|
```yaml
# app/config/config.yml
suncat_admin_page_board:
page_boards:
news_board: # page key/id
blocks:
-
position: left # block position: left,center,right, for sonata >= 2.3 top,bottom
type: suncat_admin.admin.block.news_detail # your `sonata.block` service
settings: # block settings
title: News detail
route:
name: admin_suncat_admin_news_board # route name (default )
path: /suncat/admin/news/{id}/board # *required
prefix: '/admin' # route path prefix (default: /admin)
routes_prefix_name: 'sonata_admin_page_board' # route name prefix (default)
host: 'site.com'
schemes: 'http,https'
requirements:
id: \d+
another_page:
blocks:
-
position: left
type: sonata.block.service.text
settings:
content: Some text
-
position: right
type: sonata.block.service.text
settings:
content: Some text
route:
name: admin_suncat_another_page
path: /suncat/admin/another/page
routes_loader_class: "Suncat\AdminPageBoardBundle\Routing\RouteLoader" # override route loader
routes_config_class: "Suncat\AdminPageBoardBundle\Routing\RouteConfig" # override route config
page_board_template: "SuncatAdminPageBoardBundle:Core:page_board.html.twig" # override board template
sonata_admin_layout_template: "SuncatAdminPageBoardBundle::layout.html.twig" # override sonata layout
sonata_admin_list_builder_class: "Suncat\AdminPageBoardBundle\Builder\ListBuilder" # override list builder
sonata_admin_version: 2.3 (default) # 2.2 left,center,right grid
# 2.3 top,left,center,right,bottom
```
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 8,640
|
Layers
---
1 Linux Kernal
===
* Security
* Memory & process management
* Power management
* Low memory killer
* Interprocess communiation
2 System libraries
===
* lots
3 Android Runtime
===
* Java [write]
* Dalvik Virtual Machine [execute]
4 Application Framework
===
* Package manager
* Window manager
* View system
* Resource manager
* Activity manager
* ContentProvider[Inter-application data sharing]
* Location manager
* Notification manager
Classes
---
Activity
***
[Takes input from user]
Service
***
[Runs in background]
1 Long running threads with no ui
2 Data between processes
BroadcastReiever
***
[Listens for and responds to events]
Content Providers
***
[Uses DB style memory]
Steps to creation
---
1 Define Resources
===
[At compilation time, resources generate the R.java class]
*Layout [.xml] @string/string_name Java: R.layout.layout_name
*Strings [.xml] @string/string_name Java: R.string.string_name
*Images
*Menus
1 Implement Application Classes
===
*onCreate()
1 Restore saved state
1 Set content view
1 Initialize UI elements
1 Link UI elements to code actions
1 Package Application
===
AndroidManifest.xml
1 Install & Run
===
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3,045
|
package org.diirt.datasource;
/**
* Processes the callback for events at the desired rate.
*
* @author carcassi
*/
interface DesiredRateEventListener {
/**
* New event to be processed.
*/
void desiredRateEvent(DesiredRateEvent event);
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 5,026
|
Rivers In India
Indus river
Krishna river
Mahanadi river
Brahmani river & Baitarani River
Kaveri River
Tapti river
Narmada River
All Major Rivers
Biggest states
Important States
National Parks Statewise
Most Imp. Protected Areas
Biosphere Reserves In India
Nanda Devi Biosphere Reserve
Panna Biosphere Reserve
Pachmarhi Biosphere Reserve
Nilgiri Biosphere Reserve
Achanakmar Amarkantak Biosphere Reserve
All Biosphere Reserves In India
52 Tiger reserves in IndiaMAP
32 Elephant Reserves in IndiaMAP
Complete Ramsar sites in IndiaMAP
UNESCO World Heritage Sites in IndiaMAP
Vulture in India 2021MAP
Critically endangered species in IndiaMAP
Mangroves Sites in India in 2021MAP
Master Biodiversity Hotspots in IndiaMAP
All Important Protected Areas
International Organisation
OPEC & OPEC
Indian Organisations
National Disaster Management Authority
Nuclear Power Plant in India
12 Major Ports in India
Oil imports by India countrywise
All International Organisations
India Art & Culture
Stupa | Master Stupas in India 2022 with detailed Map
The Great Amravati School of Art [2022]
A Beautiful Indus Valley Civilisation Map
Buddhism in India
40 UNESCO World Heritage Sites in India Beautiful MAP
unesco world heritage sites in india
A total of 40 UNESCO World Heritage Sites are found in India. In this article, we will have a look at all of these UNESCO World Heritage Sites in India. Starting with 32 Cultural UNESCO World Heritage Sites, then the 7 Natural UNESCO World Heritage Sites, and at last Mixed UNESCO World Heritage Site.
In August 2021, UNESCO has declared the Harappan city of Dholavira situated in Gujarat as India's 40th world heritage site.
Dholavira is the first site of Indus Valley Civilisation (IVC) in India to be included on the UNESCO world heritage site list.
unesco world heritage sites in india cultural
UNESCO World Heritage Sites in India Natural
UNESCO World Heritage Sites in India Mixed
Dholavira 2021
Agra Fort (1983)
Most Important Protected Areas
Ajanta Caves (1983)
Archaeological Site of Nalanda Mahavihara at Nalanda, Bihar (2016)
Buddhist Monuments at Sanchi (1989)
Champaner Pavagadh Archaeological Park, Gujrat (2004)
Chhatrapati Shivaji Terminus (2004)
Churches and Convents- Goa (1986)
Elephanta Caves (1987)
Ellora Caves
Fatehpur Sikri (1986)
The Great Living Chola Temples Thanjavur (1987, 2004)
Group of Monuments- Hampi (1986)
Group of Monuments in Mahabalipuram (1984)
Group of Monuments – Pattadakal (1987)
Hill Forts of Rajasthan (2013)
Historic City of Ahmedabad (2017)
Humayun's Tomb of Delhi (1993)
Jaipur City (2019)
Khajuraho Monuments (1986)
Mahabodhi Temple Complex – Bodh Gaya (2002)
Mountain Railways of India (1999, 2005, 2008)
Qutub Minar and its Monuments, Delhi (1993)
Rani-ki-Vav (the Queen's Stepwell) at Patan, Gujarat (2014)
Red Fort Complex (2007)
Rock Shelters of Bhimbetka (2003)
Sun Temple, Konarak (1984)
The Architectural Work of Le Corbusier (2016)
The Jantar Mantar, Jaipur (2010)
Victorian Gothic and Art Deco Ensembles of Mumbai (2018)
Great Himalayan National Park Conservation Area (2014)
Kaziranga National Park (1985)
Keoladeo National Park (1985)
Manas Wildlife Sanctuary (1985)
Nanda Devi and Valley of Flowers National Parks (1988, 2005)
Sundarban National Park (1987)
Western Ghats (2012)
Khangchendzonga National Park (2016)
Agra Fort -1983
Ajanta Caves -1983
Archaeological Site of Nalanda Mahavihara at Nalanda, Bihar -2016
Buddhist Monuments at Sanchi -1989
Champaner-Pavagadh Archaeological Park -2004
Chhatrapati Shivaji Terminus (formerly Victoria Terminus) -2004
Churches and Convents of Goa -1986
Dholavira: a Harappan City -2021
Elephanta Caves -1987
Ellora Caves -1983
Fatehpur Sikri -1986
Great Living Chola Temples -1987,2004
Group of Monuments at Hampi -1986
Group of Monuments at Mahabalipuram -1984
Group of Monuments at Pattadakal -1987
Hill Forts of Rajasthan -2013
Historic City of Ahmadabad -2017
Humayun's Tomb, Delhi -1993
Jaipur City, Rajasthan -2019
Kakatiya Rudreshwara (Ramappa) Temple, Telangana -2021
Khajuraho Group of Monuments -1986
Mahabodhi Temple Complex at Bodh Gaya -2002
Mountain Railways of India -1999,2005,2008
Qutb Minar and its Monuments, Delhi -1993
Rani-ki-Vav (the Queen's Stepwell) at Patan, Gujarat -2014
Red Fort Complex -2007
Rock Shelters of Bhimbetka -2003
Sun Temple, Konârak -1984
Taj Mahal -1983
The Architectural Work of Le Corbusier -2016
The Jantar Mantar, Jaipur -2010
Victorian Gothic and Art Deco Ensembles of Mumbai -2018
Great Himalayan National Park Conservation Area -2014
Kaziranga National Park -1985
Keoladeo National Park -1985
Manas Wildlife Sanctuary -1985
Nanda Devi and Valley of Flowers National Parks -1988,2005
Sundarbans National Park -1987
Western Ghats -2012
Khangchendzonga National Park 2016
Discovered by archaeologist Jagat Pati Joshi in 1968.
An exceptional and well-preserved ancient urban settlements.
Located in Kachchh District, of Gujarat, on Khadir bet island in the Kachchh Desert Wildlife Sanctuary in the Great Rann of Kachchh.
Dholavira is located at such location that Tropic of Cancer crosses it.
Harappan towns were normally built near to the rivers and perennial sources of water, but Dholavira is on the island of Khadir bet. Its unique location facilitated internal as well as external trade especially to the Magan (modern Oman peninsula)and Mesopotamian regions.
After Mohenjo-Daro, Harappa, Rakhigarhi, Ganweriwala, Dholavira holds the position of the fifth largest settlement of the Indus Valley Civilization (IVC).
Dholavira consisted of:
a walled city,
a castle,
two seasonal streams and
houses on different levels indicating a social hierarchy.
Dholavira shows three-stage settlements – the citadel, middle town & lower town – depicts the social hierarchy that prevailed there. Castle was for an important personality, while the middle town had rich merchants and the lower town housed the common people.
A series of reservoirs are found to the east and south of the Citadel, this showcase the efficient water management system of Dholavira.
located near the Taj Mahal is the 16th-century Mughal monument, the Red Fort of Agra.
Fortress of red sandstone was the imperial city of the Mughal rulers.
Consists of Jahangir Palace and Khas Mahal, built by Shah Jahan; and beautiful audience halls, like the Diwan-i-Khas.
32 Elephant Reserves in India Biodiversity hotspots in India
40 UNESCO World Heritage Sites India Critically Endangered Species in India
46 Ramsar sites in India Mangrove sites in India
52 Tiger reserves in India Vulture in India
All Biosphere Reserves
The Buddhist caves here date from the 2nd and 1st centuries B.C.
Some of the masterpieces of Buddhist religious art.
Location: Bihar.
The archaeological site of Monastic and Scholastic institutions dating from the 3rd century BCE to the 13th century CE.
Stupas, shrines, viharas along with artworks in stucco, stone and metal.
Nalanda holds the title of the most ancient university of the Indian Subcontinent.
It had a legacy of transmission of knowledge that prevailed for over 800 years.
It holds the distinction of being the oldest Buddhist sanctuary in existence.
It was a major Buddhist spot in India until the 12th century A.D.
Monolithic pillars, palaces, temples and monasteries remains are found here.
It is a Prehistoric site belonging to the chalcolithic phase
This area was the capital of the state of Gujarat. Now it is a hill fortress with archaeological remains.
Palaces, religious buildings, residential precincts, agricultural structures and water installations, belonging to the 8th to 14th centuries.
The Kalika Mata Temple, located at the top of Pavagadh Hill, is considered an important temple that attracts pilgrims all year round.
This site is the only remaining complete and unchanged Islamic pre-Mughal city.
Maharashtra India
It is an example of Victorian Gothic Revival architecture, mixed with themes from Indian traditional architecture.
Work of British architect F. W. Stevens.
Took over 10 years in making, starting in 1878.
The remarkable stone dome, turrets and pointed arches depict the traditional Indian palace architecture.
The churches and convents of Goa point towards the commencement of Evangelisation in Asia, especially the Basilica of Bom Jesus.
The Basilica of Bom Jesus also consists of the tomb of St. Francis Xavier.
The former capital of the Portuguese Indies, about 10 km east of the state capital Panjim.
Profound example of the work of missionaries in Asia.
Followed the architectural styles of Europe, adapted to native conditions through the use of local materials and artefacts.
These buildings were developed in Indo-Portuguese style during Portuguese control over the area, that lasted for 450 years until 1961.
Located on the island of Gharapuri (Elephanta Island) in the Sea of Oman, close to Mumbai.
Collection of rock arts related to the Shaivite cult.
A symbol of the vastness of Indian art, particularly the huge high reliefs in the main cave.
Constructed in the mid-5th to 6th centuries AD.
the great Cave 1 is the most important, it measures 39 metres from the front entrance to the back.
This cave closely resembles Dumar Lena cave at Ellora.
The seven-metre-high "Sadashiva" stand tall at the entrance to Cave 1.
This sculpture depicts three aspects of Shiva:
the Creator, the Preserver, and the Destroyer, represented by:
Aghora or Bhairava ( the left half of figure),
Taptapurusha or Mahadeva (the central full face of figure),
Vamadeva or Uma (the right half of the figure).
Nataraja, Yogishvara, Andhakasuravadha, Ardhanarishwara, Kalyanasundaramurti, Gangadharamurti, and Ravanaanugrahamurti are some of the masterpieces found here.
The tremendous 34 caves at Ellora at the Charanandri hills, Maharashtra depicts the spirit of co-existence and religious tolerance in architectural activities of that time. Three prominent religions architecture is found here
Brahmanism,
Jainism.
The rock-cut process was executed in three phases from the 6th century to the 12th century.
The earliest caves No- 1–12, excavated between the 5th & 8th centuries, displays the Mahayana philosophy of Buddhism.
The Brahmanical caves No- 13–29, including the world-famous Kailasa temple at cave no-16, was excavated between the 7th & 10th centuries.
The last phase of Caves belonging to the 9th and 12th centuries (caves 30–34) reflects Jaina philosophy.
Built by Emperor Akbar, Fatehpur Sikri or 'The City of Victory,' was the capital of the Mughal empire for a brief period.
It contains various monuments and temples, including one of the largest mosques of India- the Jama Masjid.
It was constructed between 1571 and 1573.
Fatehpur Sikri was the first planned city of the Mughals marked by
magnificent administrative, residential, and religious buildings.
Later the capital was shifted to Lahore in 1585, Fatehpur Sikri remained a key area for the visits by the Mughal emperors.
The Fatehpur Sikri is bounded on three sides by a wall 6 km long. It is fortified by towers and has nine gates.
The Fatehpur Sikri city was originally rectangular in plan, having a grid pattern of roads. It also featured efficient drainage and water management system.
The Jama Masjid is located in the centre of this city.
Red sandstone with little use of marble is the major building material used here.
The great Cholas set up a powerful monarchy in 9th CE in an area in and around Thanjavur.
Their rule lasted for four and a half centuries with various achievements, especially in architecture.
There are three remarkable living temples,
Brihadisvara at Thanjavur,
Brihadisvara at Gangaikondacholapuram
Airavatesvara at Darasuram.
The Temple of Gangaikondacholisvaram was built by Rajendra the First in 1035 and features a vimana (sanctum tower) of 53m
the Airavatesvara Temple was built by Rajaraja the Second, features a vimana (sanctum tower) of 24m.
These temple complexes form a unique group, demonstrating the development of Chola architecture.
world heritage sites in india
Hampi was the last capital of the Vijaynagar kingdom.
Dravidian temples and palaces here were built during the rule of Vijaynagar empire between the 14th and 16th centuries.
In 1565, the city was captured by Deccan Muslim Confederacy and pillaged for 6 months, before being abandoned
The site of Hampi mainly has the remnants of the Capital City of the Vijayanagara Empire.
It is an area located in the Tungabhadra river basin in the Bellary District of Central Karnataka.
The breathtaking landscape of Hampi is dominated by the Tungabhadra River, rocky mountain ranges and wide open plains with remnants of material remains..
The sophistication of the vibrant kingdom is evident from the more than 1600 surviving architectures that include forts, riverside features, royal and sacred complexes, many temples & shrines having pillared halls, mandapas, memorial structures, gateways, There are also defence check posts, stables, water structures, etc.
Name of few extraordinary architectures of Hampi site are:
Krishna temple complex,
Narasimha, Ganesa, Hemakuta group of temples,
Achyutaraya temple complex,
Vitthala temple complex,
Pattabhirama temple complex,
Lotus Mahal complex.
Dravidian architecture flourished under the Vijayanagara kingdom and its ultimate forms are represented by massive dimensions, cloistered enclosures, and lofty towers over the entrances encased by decorated pillars.
The Vitthla Temple is the most unique ornate structure in the area and the pinnacle of the Vijayanagara Temple architecture.
Vijayanagara architecture is also adopted some of the elements of Indo-Islamic Architecture in their secular buildings for example the Queen's Bath and the Elephant Stables, depicting a highly evolved multi-religious and multi-ethnic society.
Mahabalipuram (or Mamallapuram) Founded by Pallava kings in the 7th and 8th centuries these monuments are located along the Coromandel coast.
These temples have intricate and unique architectural styles represented in the form of-
rathas – the temples in the shape of chariots,
mandapas – the cave sanctuaries
Huge open-air reliefs such as 'Descent of the Ganges.'
Shore temple, with huge amount of sculptures attributed to the glory of Shiva is found here.
Pattadakal situated in Karnataka presents a unique blend of architectural forms of both North and South India.
Constructed by the Chalukya dynasty during the 7th and 8th centuries.
It contains nine Hindu temples as well as a Jain sanctuary.
A masterpiece – Temple of Virupaksha, built-in c.740 by Queen Lokamahadevi to commemorate the victory of her husband is present here.
This Hill forts includes 6 beautiful forts of the Rajasthan
Chittorgarh,
Kumbhalgarh,
Sawai Madhopur,
Jaisalmer,
Jhalawar.
Magnificent and stalwart forts showcase the lifestyle and nature of the Rajput rule on this land from the 8th to 18th centuries.
Enclosed inside the defensive walls of these forts are major urban centres, palaces, trading centres and various buildings including temples.
These Temples developed an elaborate courtly culture that supported learning, music and the arts.
These forts used the natural defences form offered by the landscape present here: the hills, deserts, rivers, and dense forests.
They also exhibit an extensive water harvesting system, largely still in use today.
Situated on the eastern bank of the Sabarmati river.
It was founded by Sultan Ahmad Shah in the 15th century.
Served as the capital of Gujarat for centuries.
This heritage embodies religious buildings along with the rich domestic wooden architecture:
The "Havelis"
"pols"- gated main streets,
Khadkis-inner entrances to the pols
The timber-based architecture of old Ahmedabad city is an exceptional and most unique aspect of this heritage.
Site exhibit a harmonious existence of diverse religions on this land, showcased by its architecture that includes the famous Bhadra citadel along with various mosques, tombs and numerous Hindu and Jain temples.
Humayun's Tomb was built in the 1560s, by the great Emperor Akbar (Humayun's son)
This is the first garden tomb built in India.
It was the inspiration for several architectural innovations, including the Miracle, Taj Mahal.
Persian and Indian craftsmen constructed this garden tomb,
It was far grander than any tomb built before in the Islamic world.
It is an example of the char bagh Style architecture – a four-quadrant garden with the four rivers of Quranic paradise represented with pools joined by channels.
The structure is made of red sandstone with white and black marble borders.
Humayun's garden-tomb is also known as the 'dormitory of the Mughals' as in the cells are buried over 150 Mughal family members.
The tomb stands in an extremely significant archaeological setting, centred at the Shrine of the 14th century Sufi Saint, Hazrat Nizamuddin Auliya.
Since it is considered auspicious to be buried near a saint's grave, seven centuries of tomb building has led to the area becoming the densest ensemble of medieval Islamic buildings in India.
Jaipur was founded in 1727 AD by the then Kachwaha Rajput ruler of Amber, Raja Sawai Jai Singh II.
The capital city of Rajasthan.
The Jaipur was developed on the plains and built based on a grid plan interpreted with knowledge of Vedic architecture.
Jaipur's urban planning is a mix of ideas from ancient Hindu and modern Mughal as well as Western cultures.
It was designed as commercial capital, and the city has maintained the streak of its local commercial, artisanal and cooperative traditions to this day.
The famous monuments of the city include the Govind Dev temple, City Palace, Jantar Mantar, Hawa Mahal etc.
Jaipur becomes the 2nd city of India after Ahmedabad to get the recognition of the UNESCO World Heritage Site.
The Khajuraho temples were built during the reign of the Chandella dynasty, reaching their peak between 950 and 1050.
The Khajuraho group of temples is the culmination of northern Indian temple art and architecture of the Chandella dynasty.
Distributed over an area of 6 square km in a picturesque landscape, the Khajuraho Group of Monuments are rare surviving examples that show the originality and high quality of Nagara-style temple architecture.
The Khajuraho Group of Monuments are regarded as the pinnacle of temple architectural development in northern India.
Built-in sandstone, the temples are profusely carved with motifs depicting sacred and secular themes.
Sculptures depicting acts of worship, minor clans and deities, and united couples, reflecting sacred belief systems.
Other themes depicting social life include home scenes, teachers and students, dancers and musicians, and couples in love.
The composition and finesse give the stone surfaces of the Khajuraho temples a rare vibrancy, an example of master craftsmanship
Only 20 temples remain, belonging to -Hinduism and Jainism, including the famous Temple of Kandariya decorated with intricately and beautifully carved sculptures.
The present Temple Complex at Bodh Gaya dates back to the 5th or 6th centuries, However, This Temple Complex in Bodh Gaya was built by Emperor Asoka in the 3rd century B.C.
One of the earliest Buddhist temples built with brick entirely.
Considered to be one of the four sacred places linked with the life of Gautama Buddha.
Sites are attached to the life of the Buddha, particularly to the attainment of Enlightenment.
Mahabodhi Temple Complex comprises the 50 m high grand Temple, the Vajrasana, sacred Bodhi Tree and other six sacred sites related to Buddha's enlightenment, surrounded by numerous ancient Votive stupas.
The seventh sacred place, the Lotus Pond, is outside this enclosure to the south.
This site includes three railways, All three railways are still fully operational.
Darjeeling Himalayan Railways
The Darjeeling Himalayan Railway was the first example of a hill passenger railway. Started in 1881, its design includes bold and ingenious engineering solutions for establishing an effective rail link across mountainous terrain.
Nilgiri Mountain Railways
It is a 46-km long metre-gauge single-track. Located in Tamil Nadu. It was first proposed in 1854, but due to its difficult location, the work was started in 1891 and was completed in 1908.
Kalka Shimla Railways
A 96-km long, single-track rail link was built in the mid-19th century. It provides service to the highland town of Shimla, an example of the technical and material efforts to mobilise mountain populations through the railway.
It was built in red sandstone at the beginning of the 13th century, in Delhi.
Qutub Minar is 72.5 m high, having diameters of 14.32 m and 2.75 m at its base and peak respectively.
The surrounding archaeological area has funerary buildings, including the magnificent Alai-Darwaza Gate, the masterpiece of Indo-Muslim art that was built in 1311, and two mosques, which includes the Quwwatu'l-Islam, the oldest in northern India, built of materials obtained from about 20 Brahman temples.
It is situated on the banks of the Saraswati river, It was built as a memorial to a king.
Stepwells have been constructed in the Indian subcontinent since the 3rd millennium B.C.
This stepwell represents the Maru-Gurjara architectural style,
It is designed in the form of an inverted temple to emphasize the sanctity of water and is endowed with more than a thousand sculptures.
The complex of Red Fort was built as the palace fort of Shahjahanabad – it was the new capital of the 5th Mughal Emperor of India, the Shah Jahan.
Got its name from massive enclosing walls of red sandstone. It is adjacent to another older fort, the Salimgarh, built by Islam Shah Suri in 1546, together they form the Red Fort Complex.
The private apartments inside the complex consist of a row of pavilions that are connected by a continuous water channel, known as the Nahr-i-Behisht (Stream of Paradise).
The Red Fort is considered to be an example of the zenith of Mughal creativity.
The architectural elements here are typical of a Mughal building. It reflects a fusion of Persian, Timurid and Hindu traditions
The Red Fort's planning and architectural forms, including the garden design, strongly influenced the later buildings and gardens built in Rajasthan, Delhi, Agra etc.
Located in the foothills of the Vindhya range, on the southern boundary of the central Indian plateau.
These natural rock shelters contain paintings that date back to the Mesolithic, and other periods succeeding it.
These rock shelters show the lifestyle of the earliest humans that lived in the Indian subcontinent.
The cultural practices of the inhabitants in the surrounding areas are similar to those displayed in these rock paintings.
The Sun Temple at Konarak, located on the eastern shores of India.
It is an outstanding testimony to the 13th-century kingdom of Orissa and a monument of Surya, the Sun God.
The Sun Temple is the zenith of Kalingan temple architecture.
A masterpiece, the temple flaunts a chariot of the Sun God, with twelve pairs of wheels drawn by seven horses moving across the heavens.
On the north and south sides of the temple are 24 carved wheels, each about 3 m in diameter, as well as symbolic motifs representing the cycle of the seasons and the months.
An immense mausoleum of white marble in Agra.
Built between 1631 and 1648 by order of the Mughal emperor Shah Jahan in memory of his favourite wife Mumtaz Mahal. It is located on the banks of the Yamuna River.
The Taj Mahal is the ultimate jewel of Muslim art in India and is a universally admired masterpiece of the world's heritage.
The Taj Mahal is a completely symmetrical building.
The mosque and guest house of the complex are built of red sandstone in contrast to the central marble mausoleum.
The four free-standing minarets at the corners of the platform add dimension to the Mughal architecture.
This transnational serial property has 17 sites spread across 7 countries Argentina, Belgium, France, Germany, India, Japan, and Switzerland.
It is a new architectural form that made a break with the past.
These sites were built for a half-century, in the course of what Le Corbusier described as "patient research".
These sites represent ideals of the Modern movement and are considered as a response to fundamental issues of architecture and society in the 20th century.
some of the sites included in this property are:
Complexe du Capitole of Chandigarh,
The Museum of Western Art of Tokyo
House of Dr Curutchet of La Plata in Argentina.
Unite habitation of Marseille in France.
Built at the beginning of the 18th century, Jantar Mantar is established to observe astronomical positions with the naked eye.
A set of 20 instruments are installed here to make an accurate observation.
Jantar Mantar is a manifestation of astronomical knowledge and skills of people dating back to the Mughal times.
It is the most significant, comprehensive and well preserved India's historic observatory.
It includes a collection of buildings designed in a Victorian NeoGothic style belonging to the 19th century and the Art Deco style of the 20th century.
Both these architectural styles are mixed with Indian architectural elements to create this architecture.
For example- The buildings created in Victorian Neo-Gothic styles are endowed with balconies and verandas.
The Indo- Deco term represents the style that evolved by merging Indian elements to Art Deco imagery and architecture.
National Parks in Himachal pradesh
Great Himalayan National Park
located in the Kullu region in Himachal Pradesh.
GHNP is on the list of UNESCO World Heritage Sites
This National park is at the junction of the world's two major faunal regions:
the oriental (Indomalayan) to the south
Palaearctic to the north.
the upper mountain glacial and snow meltwater is the source of several rivers and the catchment area of various rivers whose water supplies are vital to millions of downstream users.
The glacial and snow meltwater in the upper mountains of the Park is a source of various headwater tributaries to the River Beas:
Sainj, Jiwa Nal and Tirthan Rivers -westerly flowing
Parvati River flowing north-westerly
the Great Himalayan National Park have preserved the forests and alpine meadows of the Himalayas ranges which are sustaining a unique biota comprised of many distinct high altitude-sensitive ecosystems.
The rugged topography and inaccessibility of the park along with its location inside a much larger ecological complex of protected areas surrounding it have ensured its integrity.
The temperate forest flora-fauna of the Great Himalayan National Park is considered the westernmost extension of the Sino-Japanese Region.
The high elevation mountains in the park with a height up to 4100 meters house a diversity of zones with have their representative flora and fauna which include alpine, glacial, temperate, and subtropical forests.
Orang National Park
Kaziranga National Park in the north-eastern region of India Covering 429.96 Sq. Km. area and located in the State of Assam represents one of the last unmodified natural areas.
The Kaziranga National Park has one of the highest densities of tigers in India and has been declared a Tiger Reserve since 2007. (Learn About all 51 Tiger Reserves in India with Map)
Kaziranga National Park area is the single largest and undisturbed area lying in the Brahmaputra Valley floodplain.
The meandering of the Brahmaputra River creates spectacular examples of riverine and fluvial lands in this vast area.
This area comprises wet alluvial tall grassland, scattered with a large number of broad shallow pools fringed with vegetation patches of deciduous to semi-evergreen woodlands.
Riverbank erosion by the Brahmaputra river results in sedimentation and formation of new lands as well as new water-bodies
Succession between grasslands and woodlands in these newly formed sedimented lands represents outstanding examples of significant, continuous and dynamic ecological and biological processes.
Two-thirds of the Kaziranga National Park area is under Wet alluvial grasslands which are maintained by annual flooding and burning.
29 sq. Km
Formerly known as Bharatpur Bird Sanctuary.
UNESCO World Heritage site.
Birds paradise.
It is a man-made wetland area. The earlier ruler of this area Suraj Mal flooded this area by building Ajan bund, and water from nearby rivers Banganga and Gambhir were fed into this area having lowland topography.
The Keoladeo National park contains 10 artificial lagoons.
It is recognised as one of the world's most important breeding areas for birds.
One of the most famous bird species visiting this area in the winter season is the critically endangered Siberian Crane.
Other species found here are- cranes, pelican, Eagles, Flycatchers, Demoiselle cranes, falcons, jackals, chital, Nilgai, hyenas, porcupine etc.
The vegetation of this wetland is mostly Kadam, babul, kair, ber etc.
Manas National Park is located in the foothills of the Himalayas in the bhabar area of western Assam.
It spans the Manas river and is joined on the north by the Royal Manas National Park of Bhutan.
Manas National Park have six national and international designations
World Heritage Site UNESCO
Tiger Reserve
Biosphere Reserve (national)
Elephant Reserve
Important Bird Area
Manas National Park provides the highest legal protection to its species under the strong legislative framework of the provisions of the Indian Wildlife Protection Act 1972 and Indian Forest Act, 1927/Assam Forest Regulation 1891.
Manas was initially a game reserve since 1928
Manas became a Tiger Reserve in 1974 (Learn About all 51 Tiger Reserves in India with Map)
Manas became a World Heritage Site in 1985,
It became Biosphere Reserve in 1989.
Manas was declared as a National Park in 1990 with an area of 500 sq. km.
Manas is also the core area of the Chirang Ripu Elephant Reserve. (Learn About all 32 Elephants Reserves in India with Map)
To the North of the Manas National Park is the international border of Bhutan created by the imposing Bhutan hills.
Manas National Park spreads on either side of the majestic Manas river, the tumultuous river flows down the mountains in the backdrop of hills covered with forests coupled with the alluvial grasslands and tropical evergreen forests.
The monsoon and river system in the Manas National Park forms four principal geological habitats:
Bhabar savannah
Terai tract
marshlands
riverine tracts
Surface: 6407.03 Sq. Km.
Core area(s): 712.12 Sq. Km.
Buffer zone(s): 5148.57 Sq. Km.
Transition zone(s): 546.34 Sq. Km.
Designation date by UNESCO: 2004
Nanda Devi Biosphere Reserve dispersed across three districts of Uttarakhand viz. Chamoli, Pithoragarh, and Bageshwar.
Nanda Devi Biosphere Reserve in the Himalayan Mountains is located in the northern part of the country.
There are two core zones in Nanda Devi Biosphere Reserve, namely Nanda Devi National Park and Valley of Flower National Park. Both these National Parks are also designated as UNESCO World Heritage Sites.
Nanda Devi Biosphere Reserve has remained more or less intact because of its inaccessibility.
The snow-clad peaks in reserve with over 30 glaciers, charismatic animals and birds, and deep and vast valleys make the reserve unique. Meadows and rivers and the culture of the native communities make the Biosphere Reserve unique.
The Valley of Flowers National Park inside the Nanda Devi biosphere reserve is known for its meadows of endemic alpine flowers and outstanding natural beauty.
The biosphere reserve is a unique transition zone between the mountain ranges of the Zanskar and the Great Himalayas.
These National Parks contain the catchment area of Alaknanda River and its tributaries, including Rishi Ganga, Dhauli Ganga, Pushwapati, and Khir Ganga.
Lots of peaks line along the northern boundary of the core zone of the Nanda Devi National Park, which includes
Nanda Devi East (7434 m), Trishul (7120 m), Dunagiri (7066 m), Kalanka (6934 m), Changbang (6864 m) and Nanda Ghungti (6368 m).
Nanda Devi West Peak (7817 m) is within Nanda Devi National Park.
The area has an extensive altitudinal range (1,800 to 7,817 m) that is dominant with the peak of Nanda Devi. The unique topography, climate, soil, and biogeographical location of Nanda Devi Biosphere Reserve gives rise to diverse habitats. Unique communities and ecosystems and many ecologically and economically important species are part of this reserve.
The diverse topography of this area leads to distinctive micro-climates. The forest composition shows a change across the altitudinal gradient, starting from dry deciduous woods in the lower elevations to alpine meadows above the wood line.
The alpine vegetation of the reserve majorly comprises herbaceous species and scrub communities such as Rhododendron campanulatum, R. anthopogon, and Salix denticulata.
The percentage of native and endemic species is high in comparison to non-native species in reserve.
The Sundarban National Park contain the world's largest concentration of mangrove forests.
Situated at the Delta of the Ganges and Brahmaputra Rivers between India and Bangladesh.
The Sunderban mangroves host the single largest population of tigers in the world which have adapted and evolved to an almost amphibious life, Tigers here are capable of swimming for long distances and surviving on fish, crab and water monitor lizards.
The islands around here are of great economic value some of their benefits are as follow:
Act as a storm barrier, shore stabiliser, nutrient and sediment trap,
A source of timber and natural resources, and support a wide variety of aquatic, benthic and terrestrial organisms.
About 66% of the Sunderban mangrove forest area is in Bangladesh, and the remaining 34% area is in India.
Ecological processes observed here include-
monsoon rain flooding,
delta formation,
An area of 133,010 ha, with 55% forest land and 45% wetlands in the form of creeks, canals, tidal rivers and vast estuarine mouths of the river.
The Western Ghats are older than the Himalayas mountains.
The Western Ghats are a chain of mountains that runs parallel to India's West Coast and passes through the states of Kerala, Maharashtra, Goa, Gujarat, Tamil Nadu and Karnataka.
The Western Ghats cover an immense area and stretch for 1600 km broke only once by a 30km Palghat gap at around 11 degrees North.
Western Ghats influence the Indian monsoon heavily, it acts as a barrier to rain-laden monsoon winds that flows in from the southwest direction into the Indian Subcontinent.
The Western Ghats is home to tropical evergreen forests, with an Ultra high level of biodiversity and endemism and is recognized as one of the world's eight 'hottest hotspots' of biological diversity.(Biodiversity hotspots in India)
Khangchendzonga National Park hosts the world's third-highest peak, Mount Khangchendzonga.
The Park is full of steep-sided valleys, snow-clad mountains and lakes and glaciers. It includes the 26km long Zemu glacier, situated at the base of Mount Khangchendzonga.
Khangchendzonga National Park covers almost 25% of the state of Sikkim.
Provides a favourable environment to endemic, as well as threatened, plant and animal species.
Learn more about unesco world heritage sites in india from UNESCO
Checkout Other Beautiful Maps
Have a Look at these Beautiful Maps
Tiger reserves in India Map Elephant Reserves in India with Map
Mangrove sites in India Map Ramsar sites in India Map
Biodiversity hotspots in India Critically Endangered Species in India
Vulture in India
This was all about the UNESCO World Heritage Sites in India
# UNESCO World Heritage Sites# UNESCO World Heritage Sites in india# World Heritage Sites
Previous Post [MAP] Saddle peak national park | Rani jhansi national park | National Parks in Andaman and Nicobar 2022
Next Post Beautiful Black Necked Crane The State Bird of Ladhak [2022]
Thulasi MuthuDear Madam/Sir, Can u post( a state level)map which clearly shows the
Anek PorwalThank You, very much. This is an awesome compilation of Information. Appreciate
Akash KambleToo Good Mam,You Really Make good map which really helpful Thank you
Buddha FactsReally loved reading about Amaravathi
Thulasi MuthuSir plz.. uploaded tamilnadu national park map
APARNA A NAIRI need notification
Vandna Phogat, IFS
These quality articles are brought to you by Vandna Phogat, IFS Officer
For New content follow
Facebook Telegram YouTube Pinterest Twitter Instagram
Odhisa
Most Important Protected Areas in India
32 Elephant Reserves in India
40 UNESCO World Heritage Sites India
46 Ramsar sites in India
52 Tiger reserves in India
Biodiversity hotspots in India
Critically Endangered Species in India
Mangrove sites in India
Brahmani river | Baitarani River
Brahmaputra river
CDRI
IPCC report AR6 2021
Mekong Ganga Cooperation
OPEC | OPEC+
Taliban Terror History
Important Indian Organisations
Coal in India
Commission for Air Quality Management in NCR
Indian Space Association
National Disaster Managment Authority
Mountains In India
Purvanchal Hills| North East Mountain Ranges
Glaciers in India with a map
hills of peninsular India
Rajmahal Hills
Important Locations on Map
All Important Ocean currents
Danube river
Red Sea Countries
Amaravati school of arts
Stupas in India
Joint military exercise of India
Military exercise of India
Agni-V Missile
Important Schemes
e-RUPI Digital Payment system
Gram Ujala scheme
National Edible Oil Mission
National Policy for Rare Diseases 2021
New IT and Social Media Rules 2021
PM Atma Nirbhar Swasth Bharat Yojana
Unique Land Parcel Identification Number system
Important Species
Gangetic & Indus River Dolphin
Black Necked Crane
Vultures in India
Elephants in India
Tigers in India
Complete Indus river system and its tributaries [2022] with Map
Master 52 Tiger reserves in India with Map [2022]
Brahmaputra river and its tributaries Detailed [2022]
Master Complete Ramsar sites in India [2022] Map
Hemis National Park | Dachigam National Park | 5 National parks in jammu and kashmir Detailed [MAP]
IUCN Red List 2022 India | Critically endangered species in India 2022| Endangered Birds
National park in Madhya Pradesh with Detailed Map [2022]
Mollem National Park | Bhagwan Mahavir National Park Goa with detailed map 2022
Kaziranga National Park with Beautiful Map [2022]
Beautiful Guindy National Park with Map
All 5 National Park in Tamil Nadu with Map
Amrabad Tiger Reserve with MAP
Master Everything about Kaveri River UPSC [2022] MAP
Copyright © UPSC Colorfull notes 2022
Facebook Instagram Pinterest Twitter Telegram YouTube
Most Important Protected AreasIMP
HOT TOPICS 2021Hot
National Parks In India StatewiseMaps
Biosphere Reserves In IndiaMaps
Rivers In IndiaMaps
Mountains in IndiaMap
International OrganisationImp
Indian OrganisationsImp
Military Exercises of India
Important SchemesImp
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,845
|
Book Presentation: Religious Responses to Violence: Human Rights in Latin America Past and Present
Senior Scholar and Board Member, Latin American Program
Woodrow Wilson International Center for Scholars
Research Scholar in Residence
As Latin America has evolved from a region of political instability and frequent dictatorships into one of elected governments, high levels of violence have remained a persistent problem. Religious Responses to Violence: Human Rights in Latin America Past and Present, edited by Alexander Wilde, offers fresh insights into how religious actors have perceived and addressed violence, from the political and state violence of the 1970s and 1980s to the drug traffickers and youth gangs of today. Focusing on human rights, the moral dimensions of violence, and the consequences of human agency, the various contributors explore how Catholics and evangelicals have drawn on their faith to respond to violence through pastoral accompaniment in a variety of different contexts.
Alexander Wilde directed a two-year research project at American University's Center for Latin American and Latino Studies that producedReligious Responses to Violence (University of Notre Dame Press, 2015). He is a senior scholar and board member of the Latin American Program at the Wilson Center and board member of Skylight Engagement, a human rights media organization. His research addresses religion, human rights, democracy, and historical memory in Latin America.
Over a long policy and academic career, Wilde has served in leadership positions with the Ford Foundation, the Washington Office on Latin America (WOLA), and the Wilson Center in addition to a variety of teaching posts. Formerly a senior fellow at the Kellogg Institute and member of its Advisory Council, Wilde was a key member of the early Kellogg leadership team. The research he conducted at Kellogg, while also associate professor of government at Notre Dame, resulted in a book coauthored with Faculty Fellow Scott Mainwaring: The Progressive Church in Latin America (University of Notre Dame Press, 1989). Wilde holds a PhD from Columbia University.
Speakers / Related People
Alexander Wilde
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 1,833
|
{"url":"https:\/\/www.eng-tips.com\/viewthread.cfm?qid=462119","text":"\u00d7\nINTELLIGENT WORK FORUMS\nFOR ENGINEERING PROFESSIONALS\n\nAre you an\nEngineering professional?\nJoin Eng-Tips Forums!\n\u2022 Talk With Other Members\n\u2022 Be Notified Of Responses\n\u2022 Keyword Search\nFavorite Forums\n\u2022 Automated Signatures\n\u2022 Best Of All, It's Free!\n\n*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.\n\n#### Posting Guidelines\n\nPromoting, selling, recruiting, coursework and thesis posting is forbidden.\n\n# Engineer needed for decorating trusses?3\n\n## Engineer needed for decorating trusses?\n\n(OP)\nHi all, engineer from another realm stopping in. I have a BSEE and Masters of Engineering in Materials Science, but neither of those qualify me to do structural engineering so I'm seeking some quick advice.\n\nI'm looking to decorate the exposed 40' trusses in my shop, the thought being:\n-attach dimensional lumber to the sides of the trusses to just make them appear to be the same thickness as the 6x6 poles.\n-cut the slop off the 2x4 lateral braces, and maybe thicken them for visual balance\n-add extra knee braces for symmetry\n-stain all the wood dark brown\n\nI've read that decorating trusses \"usually doesn't require engineer involvement\", and I'm removing nothing, only adding. But better safe than sorry, so I'm here to see if this gets a reaction of \"nah you're just adding decoration that, if anything, is extra bracing so nailgun away\" or if it's more like \"hire a structural engineer or your trusses will collapse under their own weight\"\n\nThanks for helping me get sorted!\n\n### RE: Engineer needed for decorating trusses?\n\n1) Correct answer = \"hire a structural engineer or your trusses will collapse under their own weight\".\n\n2) If this was my shop, I'd do this without batting an eye.\n\n3) Braces look like they could use a little beefing anyhow. Let us know if you'd like to do that strategically. It's not unheard of for the columns to be cantilevers such that the braces aren't required though so I wouldn't panic.\n\nIs there such a thing as snow where you live? The greater the snow, the less the impact of your plan.\n\n#### Quote (OP)\n\nHi all, engineer from another realm stopping in.\n\nIn exercising good judgment as you have, you've represented middle earth well.\n\n### RE: Engineer needed for decorating trusses?\n\nI'd not do it. The reason is you are then hiding important parts of the trusses that carry loading. Then, down the road there is some accident or fire, etc. Your claim for damages likely would be of no value since you modified he trusses somehow. I'd also check with the insurance company now. They may invalidate any coverage right now.\n\n### RE: Engineer needed for decorating trusses?\n\nAs far as attaching extra 2x's to the side of the trusses, it will add a small amount of extra weight. Your profile says you're in NY, so I'll assume the snow load is fairly substantial. If that's the case, the extra weight should be negligible. If you offset the joints in your added pieces on each side of the truss (start with a full 12' piece on one side and a 6' piece on the opposite side of the truss, and continue with 12' pieces), it will make the trusses a little stronger.\n\nAs far as the lateral braces, not all extra is 'slop'. There needs to be enough extending past the attachment point so that the end doesn't split and release the nails or screws. It's been a long time since timber design class for me, so I'm not sure how much you may need to leave.\n\nAdding the extra knee braces may stiffen the connection in ways that could cause some prying. I don't know if that's likely, or if there's a way to mitigate the effects. Hopefully, someone with more experience in this area will weigh in.\n\nI don't know of any detrimental effects from staining it.\n\nRod Smith, P.E., The artist formerly known as HotRod10\n\n### RE: Engineer needed for decorating trusses?\n\nLeave the wood work alone, don't do anything, but stain all wood to dark brown. It will be looks great then.\n\n### RE: Engineer needed for decorating trusses?\n\n(OP)\nThanks all!\nSnow load region of 20lb\/sqft, wind load region of 80mph, building 5 years old\n\nKootk, thanks for the honest input!\nI don't know exactly what the story with my knee braces is, every other barn of nearly identical construction in my area doesn't have them, I had one old house builder tell me if it was him he'd just remove them so they aren't in the way (I added that to the scary bad advice pile)\n\nIn any case, I was planning on just thickening the trusses and copying the braces on the other side, what would be a more strategic way?\n\nOldestguy, thank you for bringing insurance into the picture for me! Hadn't thought of that.\n\nBridgeSmith, good point on the overhang! Now my concern is that a lot of the joints have zero overhang, so the memo may have not gotten to the original builder :\/\n\n\n### RE: Engineer needed for decorating trusses?\n\n#### Quote (OP)\n\nIn any case, I was planning on just thickening the trusses and copying the braces on the other side, what would be a more strategic way?\n\nNothing, that's what I had in mind. Show off.\n\n### RE: Engineer needed for decorating trusses?\n\nPart of me wonders whether it may be possible to buy new trusses, with center-span field splices, and throw them up there for cheaper than hanging around on a rolling scaffold, scabbing 2x's on the trusses? Heck, then you get some added skookum factor as well. Just a thought...\n\nI'd be careful about adding knee braces. These can impose funny loads into the truss. If you do, add them so that they join the truss at a \"node\" (layman's terminology: where the sticks hit the shiny bits).\n\n### RE: Engineer needed for decorating trusses?\n\nI don't think the edge distance needs to be much (probably just a few inches), especially if it's just nails connecting it. I am hoping maybe somebody that's more familiar with this aspect will weigh in with something more definitive.\n\nRod Smith, P.E., The artist formerly known as HotRod10\n\n### RE: Engineer needed for decorating trusses?\n\nIf it was my shop, I would spend the time and money to insulate and sheetrock it so it is super comfortable to work in.\n(i assume there is no insulation on the exterior)\n\n### RE: Engineer needed for decorating trusses?\n\nrjpope42,\nAny chance that this building had to be permitted when it was built? If so the city\/county may have the truss drawings on file with the permit. Many jurisdictions that don't require full blown engineering for structures still require stamped truss drawings to be submitted with the permit application. If so, you should be able to look at the drawings to find the design criteria, many post frame trusses (not all) are designed for a 3-5psf bottom chord load to allow for ceilings, lights, etc. If the original design included this, then there really should be no issue with adding the architectural features you describe.\n\nThat said, I agree with KootK, if it were my shop I would do it without batting eye.\n\nBy looking at the construction of the building, I am fairly confident that it was built using \"standard\" techniques and there was no engineering done originally (other than the trusses). That said, the trusses were probably not designed to accommodate the knee braces, some builders put them in and some don't. In my experience, most engineered post frame buildings utilize the steel roof as a flexible diaphragm and the steel sidewalls as shear walls (sometimes the walls will need to have plywood\/osb installed on the inside if the wall is comprised mostly of openings) and don't require any braces. Both knee braces and wall bracing in post frame construction tend to suffer at their connections, most of them are installed without adequate fastening. You can read up more on knee braces here, https:\/\/www.hansenpolebuildings.com\/2012\/01\/post-f... (I have no connection with Hansen but they are very knowledgeable in engineered post frame buildings)\n\nBased on the testing cited in the link above, if it were my building, I would remove the knee braces unless I knew the original building was actually engineered to include them.\n\nIs the \"lateral bracing\" you are referring to, the truss bottom chord and web bracing that is shown in the bottom picture? If so, yes, you should be able to cut the excess ends off. I don't believe the NDS actually specifies a minimum edge or end distance for nails with a diameter less than 1\/4\", however maintaining a 1\" distance between the nails and the end of the board should be adequate.\n\n### RE: Engineer needed for decorating trusses?\n\nYou planning on holding dance parties or something? If not, I don't follow why you would a dime on adding wood to the existing trusses. This is what all pole buildings look like, and a bit of wood here and there is unlikely to change things much. Slap some paint on like suggested, and then spend money on some materials to improve the building envelope.\n\n### RE: Engineer needed for decorating trusses?\n\nI was curious about the application too. My guess: attractive but thermally uncomfortable in-law suite.\n\n### RE: Engineer needed for decorating trusses?\n\n(OP)\nCraig_H, that's an interesting idea, I'll be sure to consider it, thanks!\n\nDauwerda, thank you!! That's a lot of great information!!!\n\nI wasn't planning on making this a barn build thread, but since y'all'r curious, I plan to do 3\" closed cell foam on the walls, and have the finished interior wall set back a couple inches so the vertical poles are still visible. I was initially planning on doing a flat cieling under the trusses and faux wood beams to match the poles, but the wife really wants it vaulted and open. (Your next assumption that she's invading my man cave is incorrect, she spends just as much time in it as me).\n\nSo the outer cieling will be insulated with probably foam boards plus TBD then finished with the same material as the walls, leaving the majority of the trusswork exposed. The desire to thicken things and give a more \"timber frame\" type of look is for aesthetics. If I'm gonna stare at them all the time they might as well be prettier.\n\nAnd while I'm not planning on renting out my dream shop build for random weddings and dance parties, it's on a beautiful 8 acre hay field, and the thought has crossed my mind.\n\ntl;dr, I just want a really nice barn\n\n### RE: Engineer needed for decorating trusses?\n\nRjpope42:\nThat type bldg. is a real nasty volume to heat, so I would reconsider the flat clg. under the truss bot. chord, and insul. on its upper surface, a good vapor barrier and blown insul. Then vent the attic space. I certainly wouldn\u2019t paint or stain the trusses (framing up there) a dark color, since that just reduces reflected light to the living\/working area at fl. level. There just isn\u2019t much you can do to make those trusses and bracing in the roof system pretty. Those structures are what they are, and you shouldn\u2019t pee your money away trying to change that much, it just won\u2019t work very well. Paint everything above the truss bearing elev. white, and enjoy the lighter space, as is. Don\u2019t mess around with any of the truss bracing or the knee braces, unless you really understand the roof system design, and really know what you are doing. That bracing is usually an integral part of the roof system stability, and maybe the bldg. stability too. Any of your changes could mean you have assumed the responsibility for the roof system, if anything goes wrong. Also, regarding interior insul., the interior surface of the roof and siding metal tend to become condensing surfaces if the insul. and vapor barrier details are not handled very carefully. And, that can lead to a real mess. With the flat clg. you could install a fake (3 sided, 1x4 sides and 1x6 bot.) beam below each truss to match the 6x6 post, and flesh out the knee braces without changing them structurally. A vaulted clg. is nice, but doesn\u2019t work very well on this roof system.\n\n### RE: Engineer needed for decorating trusses?\n\nIf you want exposed wood 'beams' and you're going to put a ceiling in, you should consider screwing stained 1x6 or 2x6 boards flat along the bottom of the trusses and use them as ledges to support the ceiling panels, similar to a suspended ceiling, minus the ugly metal rails.\n\nRod Smith, P.E., The artist formerly known as HotRod10\n\n### RE: Engineer needed for decorating trusses?\n\nI'd be cautious about adding a ceiling to those trusses. They are optimized like crazy with connector plates that hardly make theoretical sense but are proven with testing. If it was designed with no ceiling in mind, there may not be enough capacity in the truss to do it. They're spaced out pretty far, so you'll have secondary ceiling framing to add as well as gyp or whatever might be used. That weight will add up quickly if it wasn't meant to be there...\n\n### RE: Engineer needed for decorating trusses?\n\n(OP)\nSo regarding the insulation and vapor barrier of the exterior vaulted ceiling walls, I've a friend that co-owns an insulation business, I'll just be copying what he did on his own 60'*140' vaulted shop roof. Hopefully I'll also be borrowing his lift.\n\nClimate control- yea it's more volume to pay for, but people climate control open spaces with vaulted ceilings all the time, that's not really insurmountable.\n\nI acknowledge flat ceiling would be more straightforward... but for the purposes of this discussion, we are assuming I'm gonna vault it anyway.\n\nThere is a cost consideration, but that's also outside the scope of this thread.\n\nI'm starting to feel bad letting this spiral into a DIY\/HGTV conversation, so let's bring it home with the topic at hand!\n\nIt sounds like most importantly, I need to check to see if it was permitted and if the plans are on file anywhere. (It should have been permitted) See if the plans have anything to say about knee braces and psf load of the trusses. (After phamENGs comment, it sounds like this is required even if I go with a contractor installed flat ceiling.)\n\n### RE: Engineer needed for decorating trusses?\n\n#### Quote (PhamENG)\n\nThey are optimized like crazy with connector plates that hardly make theoretical sense but are proven with testing.\n\nI designed these thing before and during college. And I worked for their trade association for a year after college. So, as a fairly knowledgeable prefab truss guy, I'm going to challenge you on some of this stuff.\n\n1) What makes you say that the connector plates hardly make theoretical sense? I see them as being not that much different than old school steel gusset pltes with a bunch of nails in them. They are really quite elegant in the sense that they do a much better job of distributing connection loads relatively uniformly about the connected members relative to other wood connections. The toothed plate system is a product of evolution which is usually a good thing.\nCorrosive environments can be a problem but won't be with this one. And don't kid yourself, all wood connections are validated through testing.\n\n2) I don't agree that prefab wood trusses are optimized like crazy. The reason for that is that the economic incentive for the kind detailed optimization that you see with PEMB buildings just isn't there. Consider:\n\na) The plates are conservatively over-sized because you have to account for a rather large potential for misplacement.\n\nb) Very often, the shop does not have exact plate sizes and will substitute much larger plates. Case in point: some of the plates on these trusses appear to be generously sized.\n\nc) The plates are designed assuming that they are already pulled out of the member by 1\/8\" which is usually not the case with the final product.\n\nd) Wood members used in trusses undergo a sort of unofficial, second grading process simply for the fact that curves, warps, large knots, and many other defects will render an ugly stick unusable in precise fabrication. As a result, the wood used in these trusses will typically outperform it's grade stress ratings.\n\ne) Trusses have been tested for 3D system effects that are unaccounted for in design and, as turns out, enhance performance substantially.\n\nf) The truss guys use about the same levels of load that your average, wood friendly EOR would use in the same situation. And they'll use whatever is specified if an EOR is involved in the project. This, because it doesn't make economic sense for fabricators to hyper-optimize given all of the stuff mentioned above.\n\nYour main threat with pre-fabricated wood trusses is really winding up with a hack supplier having his trusses designed by hack designer who was a checkout cashier this time last year. Even given that, however, if your truss designs bear an engineer's seal, you're probably good to go. The software is pretty dummy proof the important output is very easy for an experienced truss engineer to review.\n\n### RE: Engineer needed for decorating trusses?\n\nA house we lived in several years ago had a couple of exposed beams in the ceiling, and they either were, or were covered with, what was basically black Styrofoam. Looked okay, they were dark enough, you couldn't see a lot of detail anyway. But if that's a feasible approach, it would take care of the weight issue and add some sound dampening, for better or worse.\n\n### RE: Engineer needed for decorating trusses?\n\nKootK - thanks for sharing that experience and knowledge. I suppose I chose my words carelessly. When I said they didn't make much sense theoretically, I meant they're difficult to justify using typical wood connection principles. I had the chance to try a couple years ago - couldn't do it. The list values and the numbers we were calculating were way off. I understand that all wood connections and connectors have been validating by testing - the truss plates just feel a bit more like voodoo than anything else I've dealt with in wood. Though it is, as you said, an elegant solution to the problem of light frame wood truss connections.\n\nMost of the prefab truss buildings I've had to analyze for retrofit have required some sort of modification to get them to pass. I haven't done one exactly like this, so you're probably right that there's nothing to really worry about. I'd just approach it with plenty of caution.\n\n### RE: Engineer needed for decorating trusses?\n\nTo an extent, the retrofit issues are probably more about the nature of trusses rather than pre-fab trusses per se. With common, trabeate framimg, members usually have a handful of things to consider and are only governed by one concern with the others passing comfortably.\n\nBecause a truss is an assemblage of members and connections, there are a ton of things to check and a large set of them that may be designed marginally. I would submit that is also true of steel trusses, heavy timber trusses, and site built trusses as well. It's just sort of the cost of doing business when using such a highly efficient framing member typology. No free lunch.\n\nAssessing the plated connections is a challenge when approached from outside the industry for sure. How did you approach it? TPI code provisions and manufacturer plate data?\n\n### RE: Engineer needed for decorating trusses?\n\nValid point. Most of the heavy timber trusses I've worked with are well over 100 years old, so they were never \"analyzed\" in the first place. It's pretty hit or miss there in terms of modern building code compliance if such compliance is required by the work - some pass with excess capacity, some don't. The assemblies I've looked at from the 30s on get a bit tighter and more efficient.\n\nNo, we wanted to see how the data would compare to an equivalent nail pattern. Admittedly our procedure was a roughly approximate guess to begin with, but we were trying to draw some line of similarity as a baseline for understanding the TPI provisions and test data. Didn't really work.\n\n#### Red Flag This Post\n\nPlease let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.\n\n#### Red Flag Submitted\n\nThank you for helping keep Eng-Tips Forums free from inappropriate posts.\nThe Eng-Tips staff will check this out and take appropriate action.\n\n#### Resources\n\nWhite Paper - A Guide to 3D Printing Materials\nWhen it comes to using an FDM 3D printer effectively and efficiently, choosing the right material at the right time is essential. This 3D Printing Materials Guide will help give you and your team a basic understanding of some FDM 3D printing polymers and composites, their strengths and weaknesses, and when to use them. Download Now\n\nClose Box\n\n# Join Eng-Tips\u00ae Today!\n\nJoin your peers on the Internet's largest technical engineering professional community.\nIt's easy to join and it's free.\n\nHere's Why Members Love Eng-Tips Forums:\n\n\u2022 Talk To Other Members\n\u2022 Notification Of Responses To Questions\n\u2022 Favorite Forums One Click Access\n\u2022 Keyword Search Of All Posts, And More...\n\nRegister now while it's still free!","date":"2020-01-19 11:11:19","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\": 0.3562900125980377, \"perplexity\": 2562.653282150344}, \"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-2020-05\/segments\/1579250594391.21\/warc\/CC-MAIN-20200119093733-20200119121733-00416.warc.gz\"}"}
| null | null |
We had no idea that a company could pack this much power into a really slim laptop, but they've managed it — with dual NVIDIA GTX 765M graphics cards, no less. 17.3″ display, Intel Core i7, up to 32 GB of RAM, and dual SSD drives in a package that is 0.9″ thick. We're guessing the battery doesn't last very long.
If you haven't heard of this brand before, that's because it didn't exist before CES — they are backed by Gigabyte, kinda the same way Alienware is actually the gaming laptop side of Dell. At least that's what we've been told, we didn't get over to their press conference in time to hear the whole thing.
When we heard about this laptop, the first thing that came to mind was the cooling — how on earth can you keep a laptop cool with that much hardware inside? They've got 5 thermal pipes, 4 vents, and 2 fans, and they put it on the back of the laptop to keep it away from your wrists. We're guessing that you probably wouldn't want to hang out right behind this laptop, but on the plus side, with strategic placement of your coffee cup, you won't have to worry about it getting cold anymore.
The 17″ screen is a standard 1080p display, but it also supports up to 3 external displays with dual HDMI and display port — so you could turn it into a regular gaming desktop if you really wanted to. 802.11ac wireless, anti-ghosting keyboard, macros, and easy fan control are just a few of the gamer-friendly features this thing is packing.
Do you really need a laptop this crazy? Probably not, and it'll cost well over 2 grand, which would buy an impressive gaming desktop. But hey, CES isn't about things being practical, is it?
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,973
|
#ifndef RAPIDRADIO_H
#define RAPIDRADIO_H
#include <stdint.h>
#include "hardware.h"
namespace rapidradio
{
// High level functions -------------------------------------------------------------------------------------------------------------------------------------------
bool init();
void turnOn();
void turnOff();
#define DBMM10 0x00 // parameter for setPower(pwr): -10 dBm
#define DBMM5 0x01 // parameter for setPower(pwr): -5 dBm
#define DBM0 0x02 // parameter for setPower(pwr): 0 dBm
#define DBM5 0x03 // parameter for setPower(pwr): +5 dBm
void setPower(uint8_t pwr);
typedef enum
{
Success = 0,
FifoFull = 1,
MaxRT = 2,
Unknown = 3
} TransmitStatus;
typedef struct
{
TransmitStatus status;
size_t bytesSent;
} TransmitResult;
// if length is greater than 32 bytes, then buff will be splitted into 32-bytes long packets
TransmitResult send(const uint8_t *buff, const size_t &length, bool requestAck = true);
TransmitResult send(const uint32_t &targetAddress, const uint8_t *buff, const size_t &length, bool requestAck = true);
TransmitResult send(const uint8_t channel, const uint32_t &targetAddress, const uint8_t *buff, const size_t &length, bool requestAck = true);
void startListening(const uint32_t &localAddress);
void startListening(const uint8_t channel, const uint32_t &localAddress);
// buff has to be at least 32 bytes!
bool received(uint8_t *buff, uint8_t &length);
// can be used to check for IRQ when listening
extern volatile bool irq;
// Low level functions and definitions ---------------------------------------------------------------------------------------------------------------------
#define WITH_ACK 0x01 // parameter for sendPayload(..): send with ack expectation
#define NO_ACK 0x00 // parameter for sendPayload(..): send without ack expectation
#define MODE_PTX 0x00 // parameter for setMode(mode): set to transmitter
#define MODE_PRX 0x01 // parameter for setMode(mode): set to receiver
#define EN_AA 0x01 // parameter for configRxPipe(..): enable pipe auto ack
#define NO_AA 0x00 // parameter for configRxPipe(..): disable pipe auto ack
#define TX_DPL 0x01 // parameter for configTxPipe(..): enable dynamic payload for PTX
#define TX_SPL 0x00 // parameter for configTxPipe(..): enable static payload for PTX
#define CRC0 0x00 // parameter for configCRC(crc): disable CRC
#define CRC1 0x01 // parameter for configCRC(crc): 1 byte CRC
#define CRC2 0x02 // parameter for configCRC(crc): 2 byte CRC
#define MBPS1 0x01 // parameter for configSpeed(speed): 1Mbps
#define MBPS2 0x02 // parameter for configSpeed(speed): 2Mbps
#define PWR_BIT 0x02 // parameter for configSpeed(speed): 2Mbps
#define ADR_WIDTH3 0x03 // parameter for confAdrWidth(width): 3 byte
#define ADR_WIDTH4 0x03 // parameter for confAdrWidth(width): 4 byte
#define ADR_WIDTH5 0x03 // parameter for confAdrWidth(width): 5 byte
#define RFM73_MAX_PACKET_LEN 32// max value is 32
#define RFM73_BEGIN_INIT_WAIT_MS 0 // pause before Init Registers: 3000ms after RFM73 power up, but Raspberry Pi boots longer than 3s so can be set to 0ms here
#define RFM73_END_INIT_WAIT_MS 100 // pause after init registers
#define RFM73_CS_DELAY 0 // wait ms after CS pin state change
// SPI commands
#define RFM73_CMD_READ_REG 0x00 // Define read command to register
#define RFM73_CMD_WRITE_REG 0x20 // Define write command to register
#define RFM73_CMD_RD_RX_PLOAD 0x61 // Define RX payload command
#define RFM73_CMD_WR_TX_PLOAD 0xA0 // Define TX payload command
#define RFM73_CMD_FLUSH_TX 0xE1 // Define flush TX register command
#define RFM73_CMD_FLUSH_RX 0xE2 // Define flush RX register command
#define RFM73_CMD_REUSE_TX_PL 0xE3 // Define reuse TX payload register command
#define RFM73_CMD_W_TX_PAYLOAD_NOACK 0xb0 // Define TX payload NOACK command
#define RFM73_CMD_W_ACK_PAYLOAD 0xa8 // Define Write ack command
#define RFM73_CMD_ACTIVATE 0x50 // Define feature activation command
#define RFM73_CMD_RX_PL_WID 0x60 // Define received payload width command
#define RFM73_CMD_NOP_NOP 0xFF // Define No Operation, might be used to read status register
// Register addresses
#define RFM73_REG_CONFIG 0x00 // 'Config' register address
#define RFM73_REG_EN_AA 0x01 // 'Enable Auto Acknowledgment' register address
#define RFM73_REG_EN_RXADDR 0x02 // 'Enabled RX addresses' register address
#define RFM73_REG_SETUP_AW 0x03 // 'Setup address width' register address
#define RFM73_REG_SETUP_RETR 0x04 // 'Setup Auto. Retrans' register address
#define RFM73_REG_RF_CH 0x05 // 'RF channel' register address
#define RFM73_REG_RF_SETUP 0x06 // 'RF setup' register address
#define RFM73_REG_STATUS 0x07 // 'Status' register address
#define RFM73_REG_OBSERVE_TX 0x08 // 'Observe TX' register address
#define RFM73_REG_CD 0x09 // 'Carrier Detect' register address
#define RFM73_REG_RX_ADDR_P0 0x0A // 'RX address pipe0' register address
#define RFM73_REG_RX_ADDR_P1 0x0B // 'RX address pipe1' register address
#define RFM73_REG_RX_ADDR_P2 0x0C // 'RX address pipe2' register address
#define RFM73_REG_RX_ADDR_P3 0x0D // 'RX address pipe3' register address
#define RFM73_REG_RX_ADDR_P4 0x0E // 'RX address pipe4' register address
#define RFM73_REG_RX_ADDR_P5 0x0F // 'RX address pipe5' register address
#define RFM73_REG_TX_ADDR 0x10 // 'TX address' register address
#define RFM73_REG_RX_PW_P0 0x11 // 'RX payload width, pipe0' register address
#define RFM73_REG_RX_PW_P1 0x12 // 'RX payload width, pipe1' register address
#define RFM73_REG_RX_PW_P2 0x13 // 'RX payload width, pipe2' register address
#define RFM73_REG_RX_PW_P3 0x14 // 'RX payload width, pipe3' register address
#define RFM73_REG_RX_PW_P4 0x15 // 'RX payload width, pipe4' register address
#define RFM73_REG_RX_PW_P5 0x16 // 'RX payload width, pipe5' register address
#define RFM73_REG_FIFO_STATUS 0x17 // 'FIFO Status Register' register address
#define RFM73_REG_DYNPD 0x1c // 'Enable dynamic payload length' register address
#define RFM73_REG_FEATURE 0x1d // 'Feature' register address
// Interrupts
#define RFM73_IRQ_STATUS_RX_DR 0x40 // Status bit RX_DR IRQ
#define RFM73_IRQ_STATUS_TX_DS 0x20 // Status bit TX_DS IRQ
#define RFM73_IRQ_STATUS_MAX_RT 0x10 // Status bit MAX_RT IRQ
#define RFM73_IRQ_STATUS_TX_FULL 0x01
#define RFM73_PIN_PRIM_RX 0x01
#define RFM73_PIN_POWER 0x02
// FIFO status
#define RFM73_FIFO_STATUS_TX_REUSE 0x40
#define RFM73_FIFO_STATUS_TX_FULL 0x20
#define RFM73_FIFO_STATUS_TX_EMPTY 0x10
#define RFM73_FIFO_STATUS_RX_FULL 0x02
#define RFM73_FIFO_STATUS_RX_EMPTY 0x01
void setModeTX(void);
void setModeRX(void);
uint8_t getMode(void);
void setChannel(uint8_t cnum);
uint8_t getChannel(void);
uint8_t transmitSPI(uint8_t val);
uint8_t readRegVal(uint8_t cmd);
uint8_t writeRegVal(uint8_t cmd, uint8_t val);
uint8_t writeRegPgmBuf(uint8_t * cmdbuf, uint8_t len);
void readRegBuf(uint8_t reg, uint8_t * buf, uint8_t len);
void selectBank(uint8_t bank);
// adr has to be 5 bytes!
uint8_t configRxPipe(uint8_t pipe_nr, uint8_t *adr, uint8_t plLen, uint8_t en_aa);
void enableRxPipe(uint8_t pipe_nr);
uint8_t writeRegCmdBuf(uint8_t cmd, uint8_t * buf, uint8_t len);
void disableRxPipe(uint8_t pipe_nr);
// adr has to be 5 bytes!
void configTxPipe(uint8_t * adr, uint8_t pltype);
void flushTxFIFO();
void flushRxFIFO();
uint8_t receivePayload(uint8_t *payload);
uint8_t sendPayload(const uint8_t * payload, const uint8_t _len, const uint8_t toAck);
void SPI_MasterInit(void);
void SPI_MasterTransmit(char cData);
void setCE();
void resetCE();
void setCSN();
void resetCSN();
#define MIN(a,b) ((a) < (b) ? (a) : (b))
}
#endif
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,692
|
<?xml version="1.0" encoding="utf-8"?>
<LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"
android:orientation="horizontal"
android:layout_width="match_parent"
android:layout_height="match_parent"
android:paddingTop="4dip"
android:paddingLeft="2dip"
android:paddingBottom="4dip"
android:contentDescription="Podcast show information"
>
<ImageView android:layout_height="75dp" android:layout_width="75dp" android:id="@+id/thumbnailImage" android:src="@drawable/artwork_placeholder"></ImageView>
<LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"
android:orientation="vertical"
android:layout_width="0dip"
android:layout_weight="1"
android:layout_height="match_parent"
android:paddingLeft="4dip">
<TextView
xmlns:android="http://schemas.android.com/apk/res/android"
android:id="@+id/titleText"
android:layout_width="match_parent"
android:layout_height="0dip"
android:layout_weight="1"
android:singleLine="true"
android:textSize="18sp"
android:gravity="bottom"
android:textColor="@color/title_color"
android:paddingBottom="1dip">
</TextView>
<LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"
android:orientation="horizontal"
android:layout_width="match_parent"
android:layout_height="0dip"
android:layout_weight="1"
android:layout_gravity="top"
android:paddingTop="1dip"
android:paddingRight="0dip">
<TextView
xmlns:android="http://schemas.android.com/apk/res/android"
android:id="@+id/episodeText"
android:layout_width="0dip"
android:layout_height="match_parent"
android:layout_weight="1"
android:gravity="left"
android:singleLine="true"
android:textSize="14sp"
android:textColor="@color/artist_color">
</TextView>
<TextView
xmlns:android="http://schemas.android.com/apk/res/android"
android:id="@+id/showDate"
android:layout_width="0dip"
android:layout_height="match_parent"
android:layout_weight="1"
android:gravity="right"
android:singleLine="true"
android:textSize="14sp"
android:paddingRight="3dip"
android:textColor="@color/artist_color">
</TextView>
</LinearLayout>
</LinearLayout>
<ImageView
xmlns:android="http://schemas.android.com/apk/res/android"
android:id="@+id/showUnplayed"
android:layout_height="60dip"
android:layout_width="wrap_content"
android:paddingLeft="3dip"
android:paddingRight="0dip"
android:src="@drawable/green_bar"
android:layout_gravity="center_vertical|right">
</ImageView>
</LinearLayout>
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,418
|
To help meet the current needs of the community, the county is seeking citizen input for possible updates or amendments to the Hyde County Parks and Recreation Master Plan. We will hold a public comment session on Tuesday, September 4, 2018 at the Board of Commissioners meeting. The meeting will be held at the Hyde County Government Center and Ocracoke Community Center starting at 6 pm.
In 2014, the Hyde County Office of Planning and Economic Development completed the Hyde County Parks and Recreation Master Plan. Nine months of meetings, research, public workshops, and community surveys were conducted to gather input from the community.
The Hyde County Board of Commissioners officially adopted the plan on Monday, July 7, 2014.
Long-range plans are excellent comprehensive tools for guiding community development. The updated plan is complete with an existing asset summary, state recreation guidelines, community identified needs, and potential grant funding sources.
Results from Ocracoke and mainland Hyde are clearly separated, and countywide trends are also identified.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,012
|
{"url":"https:\/\/esp.mit.edu\/teach\/teachers\/rhkeeler\/bio.html","text":"# ESP Biography\n\n## RACHEL KEELER, ESP Teacher\n\nMajor: 12\n\nCollege\/Employer: MIT\n\nNot Available.\n\n## Past Classes\n\n(Clicking a class title will bring you to the course's section of the corresponding course catalog)\n\nX7322: Splash Contra Dance in Splash! 2013 (Nov. 23 - 24, 2013)\nCome learn to contra dance! Contra is a type of traditional social dancing that\u2019s high energy and a lot of fun. It\u2019s done as couples in a long line, so you end up dancing with everyone else in your set. We\u2019ll be starting from the very beginning, so no experience is needed, but come ready to dance; wear comfortable shoes and clothing and bring a water bottle if you can. We\u2019ll teach how contra works and go over some basic moves, then spend the rest of the time dancing.\n\nZ7832: Breathing is Fun, Kids: The Clean Air Act in Splash! 2013 (Nov. 23 - 24, 2013)\nIn 1948, smog descended on Donora, Pennsylvania. In less than three days, twenty people died and over a third of the town was sick; mortality rates were significantly higher than surrounding towns for many years. Smogs like this don't happen in the US anymore, because back when Congress was functional it passed this wonderful thing called the Clean Air Act. The CAA regulates airborne pollutants from asthma-causing O$_3$ and PM to acid-rain-forming SO$_2$. This class will cover a little bit of atmospheric chemistry, a little bit of history, and a little bit of regulatory and legal history, all focused around the CAA.\n\nX8043: How to run a Splash in Splash! 2013 (Nov. 23 - 24, 2013)\nStep 1: Get teachers. Step 2: Get students. Step 3: ??? Step 4: SPLASH! \u2026want to know what the \u201d???\u201d is? Come find out how we make Splash happen! Presented by the directors of several past Splashes.\n\nS6593: Biogeochemical Cycles: how the world recycles itself in Splash! 2012 (Nov. 17 - 18, 2012)\nEarth science: where chemistry, physics, and biology meet to teach us awesome things about the world around us. Come learn about nitrogen, carbon, ozone, and sulfates, where they are in the earth system and how they get from one place to another. In the process, we'll talk about how human activity changes the earth, atmosphere, and oceans around us.\n\nX5874: How to run a Splash! in Spark! 2012 (Mar. 10, 2012)\nStep 1: Get teachers. Step 2: Get students. Step 3: ??? Step 4: SPLASH! ...want to know what the \"???\" is? Come find out how we make Splash happen! Presented by the directors of Splash 2011.\n\nA4812: The Arts: Lecture Series in HSSP Summer 2011 (Jul. 10, 2011)\nDo you like to sing, dance, paint or write? Do you want to learn more about all of these artistic fields? \"The Arts\" will feature a different teacher each week, addressing a different artistic idea. Some weeks, teachers will lecture, and teach students about things like the history of jazz or film theory. Other weeks, teachers will lead students in art projects, performing activities, or music sampling. Exact class topics for this summer are still to be determined. Each class will be taught by a different teacher, so it will be more like a series of seminars on the arts rather than an actual class on a specific topic.\n\nE4813: Introduction to Engineering: Lecture Series in HSSP Summer 2011 (Jul. 10, 2011)\nDo you think you might want to be an engineer, but don\u2019t know much about what an engineer does? This class is designed to expose you to problems in various engineering fields. Past classes have included model rocket design, taught by an aerospace engineer and sorting algorithms taught by a software engineer. The following engineering disciplines will be addressed (subject to change). Mechanical Engineering Civil Engineering Software Engineering Chemical Engineering Electrical Engineering Each class will be taught by a different teacher, so it will be more like a series of seminars on engineering than an actual class.\n\nX4814: ESPrinkler in HSSP Summer 2011 (Jul. 10, 2011)\nDuring the third block, middle school students are given the opportunity to participate in ESPrinkler, a mini-program that consists of about five one-shot (Splash-style) classes and activities each week. This summer's offerings are still largely undetermined, though some confirmed classes are \"It's a Bird...It's a Plane...It's Bacteria?,\" \"Your classical intuition is wrong,\" and \"Storytime with Josh.\" While you don\u2019t have to individually register for each activity, you must register for this course (ESPrinkler) to partake in the different activities offered every week!\n\nW4717: Student lounge in Spark! 2011 (Mar. 12, 2011)\nCome hang out with other students and ESP volunteers. Eat snacks, relax, and chat about life, the universe, and everything!\n\nHow Not to Run a Splash in SPLASH (2011)\njmoldow is a baller","date":"2022-12-04 23:35:05","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\": 0.20197592675685883, \"perplexity\": 4208.312761591379}, \"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-2022-49\/segments\/1669446710980.82\/warc\/CC-MAIN-20221204204504-20221204234504-00835.warc.gz\"}"}
| null | null |
{"url":"https:\/\/brilliant.org\/problems\/sophie-germain\/","text":"# Sophie Germain?\n\nFind the sum of all positive integral $a<2014$ such that $ax^4+1$ can be factored to $(bx^2-cx+1)(bx^2+cx+1)$ for all $x$ where $b,c$ are integers.\n\n\u00d7","date":"2020-10-23 00:35:49","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 5, \"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\": 0.940018355846405, \"perplexity\": 264.61396628371074}, \"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-2020-45\/segments\/1603107880401.35\/warc\/CC-MAIN-20201022225046-20201023015046-00230.warc.gz\"}"}
| null | null |
Q: How do I ensure that objects are disposed of properly in .NET? I have created a Windows Forms application in .NET 2 using C# that runs continuously. For most accounts I am pleased with it, but it has been reported to me that it has failed occasionally. I am able to monitor its performance for 50% of the time and I have never noticed a failure.
At this point I am concerned that perhaps the program is using too many resources and is not disposing of resources when they are no longer required.
What are the best practices for properly disposing created objects that have created timers and graphical objects like graphics paths, SQL connections, etc. or can I rely on the dispose method to take care of all garbage collection?
Also:
Is there a way that I could monitor resources used by the application?
A: In addition to what's already been said, if you're using COM components it's a very good idea to make sure that they are fully released. I have a snippet that I use all the time for COM releases:
private void ReleaseCOMObject(object o)
{
Int32 countDown = 1;
while(countDown > 0)
countDown = System.Runtime.InteropServices.Marshal.ReleaseCOMObject(o);
}
A: You should call Dispose on scarce resources to free them. You can use a using statement for that matter:
using (var resource = new MyScarceObject())
{
// resource will be used here...
} // will free up the resources by calling Dispose automatically
A: There are a few ways of ensuring this. The main help I find is by utilising the "using" keyword. This is applied as such:
using(SqlConnection connection = new SqlConnection(myConnectionString))
{
/* utilise the connection here */
}
This basically translates into:
SqlConnection connection = null;
try
{
connection = new SqlConnection(myConnectionString);
}
finally
{
if(connection != null) connection.Dispose();
}
As such it only works with types that implement IDisposable.
This keyword is massively useful when dealing with GDI objects such as pens and brushes. However there are scenarios where you will want to hold onto resources for a longer period of time than just the current scope of a method. As a rule it's best to avoid this if possible but for example when dealing with SqlCe it's more performant to keep one connection to the db continuously open. Therefore one can't escape this need.
In this scenario you can't use the "using" but you still want to be able to easily reclaim the resources held by the connection.
There are two mechanisms that you can use to get these resources back.
One is via a finaliser. All managed objects that are out of scope are eventually collected by the garbage collector. If you have defined a finaliser then the GC will call this when collecting the object.
public class MyClassThatHoldsResources
{
private Brush myBrush;
// this is a finaliser
~MyClassThatHoldsResources()
{
if(myBrush != null) myBrush.Dispose();
}
}
However the above code is unfortunately crap. The reason is because at finalizing time you cannot guarantee which managed objects have been collected already and which have not. Ergo the "myBrush" in the above example may have already been discarded by the garbage collector. Therefore it isn't best to use a finaliser to collect managed objects, its use is to tidy up unmanaged resources.
Another issue with the finaliser is that it is not deterministic. Lets say for example I have a class that communicates via a serial port. Only one connection to a serial port can be open at one time. Therefore if I have the following class:
class MySerialPortAccessor
{
private SerialPort m_Port;
public MySerialPortAccessor(string port)
{
m_Port = new SerialPort(port);
m_Port.Open();
}
~MySerialPortAccessor()
{
if(m_Port != null) m_Port.Dispose();
}
}
Then if I used the object like this:
public static void Main()
{
Test1();
Test2();
}
private static void Test1()
{
MySerialPortAccessor port = new MySerialPortAccessor("COM1:");
// do stuff
}
private static void Test2()
{
MySerialPortAccessor port = new MySerialPortAccessor("COM1:");
// do stuff
}
I would have a problem. The issue is that the finaliser is not deterministic. That is to say I cannot guarantee when it will run and therefore get round to disposing my serial port object. Therefore when I run test 2 I might find that the port is still open.
While I could call GC.Collect() between Test1() and Test2() which would solve this problem it isn't recommended. If you want to get the best performance out of the collector then let it do its own thing.
Therefore what I really want to do is this:
class MySerialPortAccessor : IDispable
{
private SerialPort m_Port;
public MySerialPortAccessor(string port)
{
m_Port = new SerialPort(port);
m_Port.Open();
}
public void Dispose()
{
if(m_Port != null) m_Port.Dispose();
}
}
And i'll rewrite my test like this:
public static void Main()
{
Test1();
Test2();
}
private static void Test1()
{
using( MySerialPortAccessor port = new MySerialPortAccessor("COM1:"))
{
// do stuff
}
}
private static void Test2()
{
using( MySerialPortAccessor port = new MySerialPortAccessor("COM1:"))
{
// do stuff
}
}
This will now work.
So what of the finaliser? Why use it?
Unmanaged resources and possible implementations that don't call Dispose.
As the writer of a component library that others use; their code may forget to dispose of the object. It's possible that something else might kill the process and hence the .Dispose() would not occur. Due to these scenarios a finaliser should be implemented to clean any unmanaged resource as a "worst case" scenario but Dispose should also tidy these resources so you have your "deterministic clean up" routine.
So in closing, the pattern recommended in the .NET Framework Guidelines book is to implement both as follows:
public void SomeResourceHoggingClass, IDisposable
{
~SomeResourceHoggingClass()
{
Dispose(false);
}
public void Dispose()
{
Dispose(true);
}
// virtual so a sub class can override it and add its own stuff
//
protected virtual void Dispose(bool deterministicDispose)
{
// we can tidy managed objects
if(deterministicDispose)
{
someManagedObject.Parent.Dispose();
someManagedObject.Dispose();
}
DisposeUnmanagedResources();
// if we've been disposed by .Dispose()
// then we can tell the GC that it doesn't
// need to finalise this object (which saves it some time)
//
GC.SuppressFinalize(this);
}
}
A: Best practice is to make sure that all objects implementing the IDisposable interface is called Dispose on as soon as the object is no longer required.
This can be accomplished either with the using keyword or try/finally constructs.
Within a WinForms form that has resources allocated for the lifetime of the form, a somewhat different approach is necessary. Since the form itself implements IDisposable this is an indication that at some point in time Dispose will be called on this form. You want to make sure that your disposable resources gets disposed at the same time. To do this you should override the forms Dispose(bool disposing) method. The implementation should look something like this:
protected override void Dispose(bool disposing)
{
if (disposing)
{
// dispose managed resources here
}
// dispose unmanaged resources here
}
A note on Components in forms: if your object implements the IComponent interface you can place the instance in the forms Container. The container will take care of disposing components when the container itself is disposed.
A: A few tips:
-Take advantage of the Using() keyword whenever you can.
If you have unit tests in place I would recommend refactoring your code to implement this change.
http://msdn.microsoft.com/en-us/library/yh598w02.aspx
-Remember to explicitly unregister all event handlers and remove all objects from lists that live for the entire duration of your application. This is the most common mistake that programmers make in .NET that results in these items not being collected.
A: As for monitoring the built in perfmon (2) works fine for memory usage etc. If you're worried about file handles, dll handles, etc I recommend Process Explorer and Process Monitor
A: When an object is inaccessible, the Object.Finalize Method will get called. It is helpful to suppress this unnecessary call in your classes that implement IDisposable. You can do this by calling the GC.SuppressFinalize Method
public void Dispose()
{
// dispose resources here
GC.SuppressFinalize(this);
}
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 4,619
|
{"url":"https:\/\/actionmortgage.com\/fishy\/is\/lugz\/48929944a8dddad9db","text":"Select Page\n\nSolution: First write the generic expressions without the coefficients. As an online math tutor, I love teaching my students helpful shortcuts! As we have explained above, we can get the expansion of (a + b)4 and then we have to take positive and negative signs alternatively staring with positive sign for the first term So, the expansion is (a - b)4 = a4 Problem 1: Issa went to a shake kiosk and want to buy a milkshake. In this explainer, we will learn how to use Pascals triangle to find the coefficients of the algebraic expansion of any binomial expression of the form ( + ) . (X+Y)^2 has three terms. Each number shown in our Pascal's triangle calculator is given by the formula that your math teacher calls the binomial coefficient. Pascals triangle is useful in finding the binomial expansions for reasonably small values of $$n$$, it isnt practical for finding expansions for large values of $$n$$. When the exponent is 1, we get the original value, unchanged: (a+b) 1 = a+b. Each coefficient is achieved by adding two coefficients in the previous row, on the immediate left and immediate right. In Row 6, for example, 15 is the sum of 5 and 10, and 20 is the sum of 10 and 10. Math Example Problems with Pascal Triangle. Go to Pascals triangle to row 11, entry 3. F or 1500 years, mathematicians from many cultures have explored the patterns and relationships found in what we , which is called a binomial coe cient. 1+1. 8. If the exponent is relatively small, you can use a shortcut called Pascal's triangle to find these coefficients.If not, you can always rely on algebra! Whats Pascal's triangle then? 9.7 Pascals Formula and the Binomial Theorem 595 Pascals formula can be derived by two entirely different arguments. ). 1 4 6 4 1 Coefficients from Pascals Triangle. Solved Problems. add. If we denote the number of combinations of k elements from an n -element set as C (n,k), then. The name is not too important, but let's see what the computation looks like. Question: 8. Limitations of Pascals Triangle. We only want to find the coefficient of the term in x4 so we don't need the complete expansion. Since were The coefficients of the binomials in this expansion 1,4,6,4, and 1 forms the 5th degree of Pascals triangle. To find any binomial coefficient, we need the two coefficients just above it. Clearly, the first number on the nth line is 1. Definition: binomial . addition (of complex numbers) addition (of fractions) addition (of matrices) addition (of vectors) addition formula. Solution By construction, the value in row n, column r of Pascals triangle is the value of n r, for every pair of It's much simpler to use than the Binomial Theorem, which provides a formula for expanding binomials. asked Mar 3, 2014 in ALGEBRA 2 by harvy0496 Apprentice. Exercises: 1. How do I use Pascal's Triangle to expand these two binomials? Pascals triangle contains the values of the binomial coefficient of the expression. Pascals Triangle and Binomial Expansion Pascals triangles give us the coefficients of the binomial expansion of the form $$(a + b)^n$$ in the $${n^{{\\rm{th}}}}$$ row in the triangle. of a binomial form, this is called the Pascals Triangle, named after the French mathematician Blaise Pascal. The numbers in Pascals triangle form the coefficients in the binomial expansion. Background. Exponent of 1. Here you will explore patterns with binomial and polynomial expansion and find out how to get coefficients using Pascals Triangle. One is alge-braic; it uses the formula for the number of r-combinations obtained in Theorem 9.5.1. (b) (5 points) Write down Perfect Square Formula, i.e. To build the triangle, always start with \"1\" at the top, then continue placing numbers below it in a triangular Solution is simple. In Algebra II, we can use the binomial coefficients in Pascals triangle to raise a polynomial to a certain power. c) State a conjecture about the sum of the terms in The coefficients in the binomial expansion follow a specific pattern known as Pascal [s triangle . One of the most interesting Number Patterns is Pascal's Triangle. The exponents for a begin with 5 and decrease. One such use cases is binomial expansion. (x-6) ^ 6 (2x -3) ^ 4 Please explain the process if possible. Pascal's Triangle & the Binomial Theorem 1. The shake vendor told her that she can choose plain milk, or she can choose to combine any number of flavors in any way she want. Well, it is neat thanks to calculating the number of combinations, and visualizes binomial expansion. Exponent of 0. Pascal's Triangle is a triangle in which each row has one more entry than the preceding row, each row begins and ends with \"1,\" and the interior elements are found by adding the adjacent elements in the preceding row. The binomial expansion of terms can be represented using Pascal's triangle. Examples. This means the n th row of Pascals triangle comprises the It is, of course, often impractical to write out Pascal\"s triangle every time, when all that we need to know are the entries on the nth line. Concept Map. (a) (5 points) Write down the first 9 rows of Pascal's triangle. 1+2+1. Detailed step by step solutions to your Binomial Theorem problems online with our math solver and calculator. The Again, add the two numbers immediately above: 2 + 1 = 3. In mathematics, Pascals rule (or Pascals formula) is a combinatorial identity about binomial coefficients. Firstly, 1 is The Binomial Theorem First write the pattern for raising a binomial to the fourth power. Lets say we want to expand $(x+2)^3$. Like this: Example: What is (y+5) 4 . What is the formula for binomial expansion? (a + b) 2 = c 0 a 2 b 0 + c 1 a 1 b 1 + c 2 a 0 b 2. Background. If n is very large, then it is very difficult to find the coefficients. As mentioned in class, Pascal's triangle has a wide range of usefulness. Pascal's Triangle CalculatorWrite down and simplify the expression if needed. (a + b) 4Choose the number of row from the Pascal triangle to expand the expression with coefficients. Use the numbers in that row of the Pascal triangle as coefficients of a and b. Place the powers to the variables a and b. Power of a should go from 4 to 0 and power of b should go from 0 to 4. Coefficients are from Pascal's Triangle, or by calculation using n!k!(n-k)! 11\/3 = Pascal's Triangle & Binomial Expansion Explore and apply Pascal's Triangle and use a theorem to determine binomial expansions. For example, the 3 rd entry in Row 6 ( r = 3, n = 6) is C(6, 3 - 1) = C(6, 2) = = 15 . Binomial Theorem Calculator online with solution and steps. 2. Get instant feedback, extra help and step-by-step explanations. In Pascals triangle, each number in the triangle is the sum of the two digits directly above it. Dont be concerned, this idea doesn't require any area formulas or unit calculations like you'd expect for a traditional triangle. Binomial theorem. We can find any element of any row using the combination function. Solution : Already, we know (a + b) 4 = a 4 + 4a 3 b + 6a 2 b 2 + 4a b 3 + b 4. combinations formula. C (n,k) = n! The (n+1)th row is the row we need, and the 1st term in the row is the coe cient of 5.Expand (2a 3)5 using Pascals triangle. If you continue browsing the site, you agree to the use of cookies on this website. Any triangle probably seems irrelevant right now, especially Pascals. As mentioned in class, Pascal's triangle has a wide range of usefulness. In elementary algebra, the binomial The binomial coefficient appears as the k th entry in the n th row of Pascal's triangle (counting starts at 0 ). This way of obtaining a binomial expansion is seen to be quite rapid , once the Pascal triangle has been constructed. Any equation that contains one or more binomial is known as a binomial equation. Binomial coefficients are the positive coefficients that are present in the polynomial expansion of a binomial (two terms) power. I'm trying to answer a question using Pascal's triangle to expand binomial functions, and I know how to do it for cases such as (x+1) which is quite simple, but I'm having troubles understanding and looking For example, x+1 and 3x+2y are both binomial expressions. Use the Binomial Theorem to find the term that will give x4 in the expansion of (7x 3)5. adjacent side (in a triangle) adjacent sides Pascals triangle and the binomial theorem mc-TY-pascal-2009-1.1 A binomial expression is the sum, or dierence, of two terms. This is one warm-up that every student does without prompting. For example, x+1 and 3x+2y are both binomial expressions. Your calculator probably has a function to calculate binomial Specifically, the binomial coefficient, typically written as , tells us the b th entry of the n th row of Pascal's triangle, named after the famous mathematician Blaise Pascal, names the binomial coefficients for the binomial expansion. additive inverse. Example: (x+y) 4Since the power (n) = 4, we should have a look at the fifth (n+1) th row of the Pascal triangle. Therefore, 1 4 6 4 1 represent the coefficients of the terms of x & y after expansion of (x+y) 4.The answer: x 4 +4x 3 y+6x 2 y 2 +4xy 3 +y 4 When an exponent is 0, we get 1: (a+b) 0 = 1. There are some main properties of binomial expansion which are as follows:There are a total of (n+1) terms in the expansion of (x+y) nThe sum of the exponents of x and y is always n.nC0, nC1, nC2, CNN is called binomial coefficients and also represented by C0, C1, C2, CnThe binomial coefficients which are equidistant from the beginning and the ending are equal i.e. nC0 = can, nC1 = can 1, nC2 = in 2 .. etc. Binomial expansion. There are instances that the expansion of the binomial is so large that the Pascal's Triangle is not advisable to be used. Binomial Theorem.\n\nBy spotting patterns, or otherwise, find the values of , , , and . Binomial Expansion Formula; Binomial Probability Formula; Binomial Equation. For example, x+1, 3x+2y, a b We pick the coecients in the expansion The binomial theorem formula is used in the expansion of any power of a binomial in the form of a series. Notes include completing rows 0-6 of pascal's triangle, side by side comparison of multiplying binomials traditionally and by using the Binomial Theorem for (a+b)^2 and (a+b)^3, 2 examples of expanding binomials, 1 example of finding a coefficient, and 1 example of finding a term.Practice is a \"This or That\" activit Isaac Newton wrote a generalized form of the Binomial Theorem. A binomial expression is the sum or difference of two terms. To Let us start with an exponent of 0 and build upwards. Combinations are used to compute a term of Pascal's triangle, in statistics to compute the number an events, to identify the coefficients of a binomial expansion and here in the binomial formula used to answer probability and statistics questions. It is the coefficient of the x k term in the polynomial expansion of the binomial power (1 + x) n, and is given by the formula =!! A triangular array of the binomial coefficients of the expression is known as Pascals Triangle. One of the most interesting Number Patterns is Pascal's Triangle (named after Blaise Pascal, A Formula for Any Entry in The Triangle. The inductive proof of the binomial theorem is a bit messy, and that makes this a good time to introduce the idea of combinatorial proof. All the binomial coefficients follow a particular pattern which is known as Pascals Triangle. Algebra Examples. What is Pascal's Triangle Formula? Binomial Theorem\/Expansion is a great example of this! The Pascal's Triangle is probably the easiest way to expand binomials. Pascals triangle is a geometric arrangement of the binomial coefficients in the shape of a triangle. This algebra 2 video tutorial explains how to use the binomial theorem to foil and expand binomial expressions using pascal's triangle and combinations. Pascals Triangle is the triangular arrangement of numbers that gives the coefficients in the expansion of any binomial expression. If we want to raise a binomial expression to a power higher than 2 it is very cumbersome to addition sentence. Well (X+Y)^1 has two terms, it's a binomial. addition. addend. ), see Theorem 6.4.1. 1a5b0 + 5a4b1 + 10a3b2 + 10a2b3 + 5a1b4 + 1a0b5 The exponents for b begin with 0 and increase. 6th line of Pascals triangle is So the 4th term is (2x (3) = x2 The 4th term is The second method to work out the expansion of an expression like (ax + b)n uses binomial coe cients. Pascals Triangle gives us a very good method of finding the binomial coefficients but there are certain problems in this method: 1. Don't worry it will all be explained! Now on to the binomial. It gives a formula for the expansion of the powers of binomial expression. It tells you the coefficients of the progressive terms in the expansions.\n\nAny particular number on any row of the triangle can be found using the binomial coefficient. Algebra - Pascal's triangle and the binomial expansion; Pascal's Triangle & the Binomial Theorem 1. How many ways can you give 8 apples to 4 people? These coefficients for varying n and b can be arranged to form Pascal's triangle.These numbers also occur in combinatorics, where () gives the number of different combinations of b elements that can be chosen from an n-element set.Therefore () is often Find middle term of binomial expansion. Pascals triangle (1653) has been found in the works of mathematicians dating back before the 2nd century BC. Discover related concepts in Math and Science.\n\nFor natural numbers (taken to include 0) n and k, the binomial coefficient can be defined as the coefficient of the monomial Xk in the Simplify Pascal's Triangle and Binomial Expansion IBSL1 D Binomial Expansion Formula. Binomial Theorem II: The Binomial Expansion The Milk Shake Problem. Inquiry\/Problem Solving a) Build a new version of Pascals triangle, using the formula for t n, r on page 247, but start with t 0,0 = 2. b) Investigate this triangle and state a conjecture about its terms. Binomial. The formula is: Note that row and column notation begins with 0 rather than 1. If you wish to use Pascals triangle on an expansion of the form (ax + b)n, then some care is needed. adjacent faces. The triangle is symmetrical. Algebra 2 and Precalculus students, this one is for you. The 1, 4, 6, 4, 1 tell you the coefficents of the p 4, p 3 r, p 2 r 2, p r 3 and r 4 terms respectively, so the expansion is just. What is the Binomial Theorem? Lets expand (x+y). Pascal's triangle is triangular-shaped arrangement of numbers in rows (n) and columns (k) such that each number (a) in a given row and column is calculated as n factorial, divided by k factorial times n minus k factorial. The coefficients will correspond with line n+1 n + 1 of the triangle. on a left-aligned Pascal's triangle. We While Pascals triangle is useful in many different mathematical settings, it will be applied Binomial expansion. Pascals Triangle Binomial Expansion As we already know that pascals triangle defines the binomial coefficients of terms of binomial expression (x + y) n , So the expansion of (x + y) n is: (x It is named after Blaise Pascal. So this is going to have eight terms. 1+3+3+1. Recent Visits Use the binomial theorem to write the binomial expansion (X+2)^3. Binomials are Here you can navigate all 3369 (at last count) of my videos, including the most up to date and current A-Level Maths specification that has 1033 teaching videos - over 9 7 hours of content that works through the entire course. Now lets build a Pascals triangle for 3 rows to find out the coefficients. To find the numbers inside of Pascals Triangle, you can use the following formula: nCr = n-1Cr-1 + n-1Cr. The general form of the binomial expression is (x+a) and the expansion of :T E= ; , where n is a natural number, is called binomial theorem. https:\/\/www.khanacademy.org\/\/v\/pascals-triangle-binomial-theorem The binomial theorem is used to find coefficients of each row by using the formula (a+b)n. Binomial means adding two together. The following figure shows how to use Pascals Triangle for Binomial Expansion. (2 marks) Ans. If one looks at the magnitude of the integers in the kth row of the Pascal triangle as k Named posthumously for the French mathematician, physicist, philosopher, and monk Blaise Pascal, this table of binomial There are a total of (n+1) terms in the expansion of (x+y) n Then,the n row of Pascals triangle will be the expanded series coefficients when the terms are arranged.\n\nPascals Triangle. The coefficients in the binomial expansion follow a specific pattern known as Pascals triangle. In mathematics, the binomial coefficients are the positive integers that occur as coefficients in the binomial theorem.Commonly, a binomial coefficient is indexed by a pair of integers n k 0 and is written (). That pattern is summed up by the Binomial Theorem: The Binomial Theorem. Analyze powers of a binomial by Pascal's Triangle and by binomial coefficients. How to use the formula 1. A binomial expression is the sum or difference of two terms. We will use the simple binomial a+b, but it could be any binomial. Scroll down the page if you need more examples and solutions. The coefficients that appear in the binomials expansions can be defined by the Pascals triangle as well. Expand the factorials to see what factors can reduce to 1 3. Once that is done I introduce Binomial Expansion and tie that into Pascal's Triangle. The formula for Pascal's Binomial Theorem and Pascals Triangle: Pascals triangle is a triangular pattern of numbers formulated by Blaise Pascal. Write down the row numbers. Pascal's Triangle is the representation of the coefficients of each of the terms in a binomial expansion. For example, the first line of the triangle is a simple 1. Finish the row with 1. A binomial expansion is a method used to allow us to expand and simplify algebraic expressions in the form into a sum of terms of the form. These are associated with a mnemonic called Pascals Triangle and a powerful result called the Binomial Theorem, which makes it simple to compute powers of binomials. Using Pascals Triangle Use Pascals triangle to compute the values of 6 2 and 6 3 . It is especially useful when raising a binomial to lower degrees. Thanks. Write the rst 6 lines of Pascals triangle. If the binomial coefficients are arranged in rows for n = 0, 1, 2, a triangular structure known as Pascals triangle is obtained. Blaise Pascals Triangle Arithmtique (1665). Solved exercises of Binomial Theorem. ()!.For example, the fourth power of 1 + x is Binomials are expressions that looks like this: (a + b)\", where n can be any positive integer. Step 2. In this worksheet, we will practice using Pascals triangle to find the coefficients of the algebraic expansion of any binomial expression of the form (+). Q1: Shown is a partially filled-in picture of Pascals triangle. A binomial is an algebraic expression containing 2 terms. The coefficient is arranged in a triangular pattern, the first and last number in each row is 1 and number in each row is the sum of two numbers that lie diagonally above the number. * (n-k)! n C r has a mathematical formula: n C r = n! Chapter 08 of Mathematics ncert book titled - Binomial theorem for class 12 In this way, using pascal triangle to get expansion of a binomial with any exponent. We have a binomial raised to the power of 4 and so we look at the 4th row of the Pascals triangle to find the 5 coefficients of 1, 4, 6, 4 and 1. CK-12 Each entry is the sum of the two above it. additive identity. acute triangle. The first few binomial coefficients. The general form of the binomial expression is (x+a) and the expansion of , where n is a natural number, is called binomial theorem. It gives a formula for the expansion of the powers of binomial expression. It is important to keep the 2 term So we know the answer is . binomial-theorem; Binomial Expansion Using Pascals Triangle Example: It states that for positive natural numbers n and k, is a binomial coefficient; one interpretation of which is the coefficient of the xk term in the expansion of (1 + x)n. How is each row formed in Pascals Triangle? The rth element of Row n is given by: C(n, r - 1) =. The coefficient a in the term of ax b y c is known as the binomial coefficient or () (the two have the same value). To construct the next row, begin it with 1, and add the two numbers immediately above: 1 + 2. For example, (x + y) is a binomial. And you will learn lots of cool math symbols along the way. To find an expansion for (a + b) 8, we complete two more rows of Pascals triangle: Thus the expansion of is (a + b) 8 = a 8 + 8a 7 b + 28a 6 b 2 + 56a 5 b 3 + 70a 4 b 4 + 56a 3 b 5 + 28a 2 b 6 + 8ab 7 + b 8. We begin by considering the expansions of ( + ) for consecutive powers of , starting with = 0. Write 3. So the answer is: 3 3 + 3 (3 2 x) + 3 (x 2 3) + x 3 (we are replacing a by 3 and b by x in the expansion of (a + b) 3 above) Generally. The binomial expansion formula can simplify this method. Step 1. Bonus exercise for the OP: figure out why this works by starting Coefficients. Pascals Triangle. What is the general formula for binomial expansion? Pascal's Triangle. Practice Expanding Binomials Using Pascal's Triangle with practice problems and explanations. Substitute the values of n and r into the equation 2. I always introduce Binomial Expansion by first having my student complete an already started copy of Pascal's Triangle. This is the bucket, Design the formula how to find nth term from end . \/ ((n - r)!r! a) Find the first 4 terms in the expansion of (1 + x\/4) 8, giving each term in its simplest form. b) Use your expansion to estimate the value of (1.025) 8, giving your answer to 4 decimal places. In the binomial expansion of (2 - 5x) 20, find an expression for the coefficient of x 5. Comparing (3x + 4y) 4 and (a + b) 4, we get a = 3x and b = 4y Binomial Expansion. The other is combinatorial; it uses the denition of the number of r-combinations as the 2. And here comes Pascal's triangle. For any binomial expansion of (a+b) n, the coefficients for each term in the expansion are given by the nth row of Pascals triangle. Lets learn a binomial expansion shortcut. View more at http:\/\/www.MathAndScience.com.In this lesson, you will learn about Pascal's Triangle, which is a pattern of numbers that has many uses in math. \/ (k! Suppose you have the binomial ( x + y) and you want to raise it to a power such as 2 or 3. Pascal's triangle can be used to identify the coefficients when expanding a binomial. Another formula that can be used for Pascals Triangle is the binomial formula. However, Pascals triangle is very useful for binomial expansion. Blaise Pascals Triangle Arithmtique (1665). Exponent of 2 The passionately Row 5 Use Pascals Triangle to expand (x 3)4. Lets look at the expansion of (x + y)n (x + y)0 = 1 (x + y)1 = x + y (x + y)2 = x2 +2xy + y2 (x + y)3 = x3 + 3x2y + 3xy2 + y3 Binomial expansion using Pascal's triangle and binomial theorem SlideShare uses cookies to improve functionality and performance, and to provide you with relevant advertising. Let a = 7x b = 3 n = 5 n addition property of opposites.\n\nBinomial Theorem I: Milkshakes, Beads, and Pascals Triangle. We start with (2) 4. The sum of the powers of x and y in each term is equal to the power of the binomial i.e equal to n. The powers of x in the expansion of are in descending order while the powers of y are in ascending order. Pascals triangle determines the coefficients which arise in binomial expansion . The first remark of the binomial theorem was in the 4th century BC by the renowned Greek mathematician Euclids. For (a+b)6 ( a + b) 6, n = 6 n = 6 so the coefficients of the expansion will correspond with line 7 7. The triangle can be used to calculate the coefficients of the expansion of (a+b)n ( a + b) n by taking the exponent n n and adding 1 1. The Binomial Theorem and Binomial Expansions. However, for quite some time Pascal's Triangle had been well known as a way to expand binomials (Ironically enough, Pascal of the 17th century was not the first person to know","date":"2022-12-03 08:42:26","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\": 0.8085084557533264, \"perplexity\": 412.7537657123692}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"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-2022-49\/segments\/1669446710926.23\/warc\/CC-MAIN-20221203075717-20221203105717-00554.warc.gz\"}"}
| null | null |
Q: VS2010 ASPX Design View, vertical scrollbar disabled In VS2010 working on an ASP.NET project, when opening an ASPX page in the IDE and viewingin Design mode, the vertical scrollbar is disabled. The horizontal scrollbar remains enabled. It doesn't seem to matter what page I open. Note that I am using a single Master page.
I have tried searching other posts, including running "devenv /resetsettings"
http://forums.asp.net/t/1149955.aspx?Scroll+Bar+doesn+t+show+up+in+Design+View+
however this didn't solve it. This thread also makes mention of the actual content itself (ie "overflow-x: hidden;") being a potential cause, however that would be pathetic if source code affected how the IDE's vertical scrollbar behaves.
Any help would be most appreciated. Thanks.
David
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 8,199
|
\section{Introduction}
\label{sec-intro}
Apart from H$_2$ and (in some cases) HD, CO is the most abundant interstellar molecule.
Determinations of the abundance and rotational excitation of the dominant $^{12}$C$^{16}$O and its isotopomers thus can provide significant information about the physical conditions in Galactic and extragalactic molecular clouds.
The CO/H$_{2}$ abundance ratio is often assumed to be roughly constant in dense molecular gas, with a ``canonical'' value of order 10$^{-4}$ derived primarily from millimetric and radio observations of CO and inferred abundances of H$_{2}$ in Galactic molecular clouds (Young \& Scoville 1991).
Under that assumption, measurements of CO are used to estimate the total H$_2$ content when no data are available for H$_2$.
In more diffuse gas, both H$_2$ and CO were detected directly via their absorption bands in the far-UV using spectra from the {\it Copernicus} satellite; longer wavelength UV bands of CO were observed with {\it IUE}.
Analysis of the {\it Copernicus} spectra showed that the CO/H$_{2}$ ratio varied by two orders of magnitude (from about 5 $\times$ 10$^{-8}$ to 5 $\times$ 10$^{-6}$) (Federman et al. 1980) --- so that CO is not a good predictor for H$_{2}$ in more diffuse material.
Theoretical models of cloud chemistry suggested that such large variations in the CO/H$_{2}$ ratio could be expected in diffuse ($A_{\rm V}$ $\la$ 1 mag) and translucent (1 mag $\la$ $A_{\rm V}$ $\la$ 5 mag) regions still permeated by dissociating photons (e.g., van Dishoeck \& Black 1986a, 1988) as well as in regions perturbed by the recent passage of dissociative shocks (Neufeld \& Dalgarno 1989).
The relatively low sensitivity and resolution of {\it Copernicus} and {\it IUE}, however, allowed only limited investigation of more heavily reddened sight lines, where the CO/H$_2$ ratio might be expected to approach the dense cloud value.
Determining observationally the population distribution in the individual rotational levels of any molecular species can yield constraints on the physical properties characterizing the molecular gas (e.g., van Dishoeck et al. 1991; Federman et al. 1997b; Wannier et al. 1997).
Observations of the rotational excitation of C$_2$, for example, have yielded estimates for the temperatures and densities in diffuse molecular clouds (e.g., van Dishoeck \& de Zeeuw 1984); data for CN have been used both to constrain the temperature of the cosmic microwave background (e.g., Roth \& Meyer 1995) and to estimate local hydrogen densities (e.g., Black \& van Dishoeck 1991).
The pervasive abundance of the CO and H$_{2}$ molecules in the interstellar gas make those species especially important tools for investigations of the gas properties.
Unfortunately, however, the spectrographs on {\it Copernicus} and {\it IUE} could not resolve the rotational structure of the CO bands.
Isotopic ratio measurements provide another means to investigate the chemical processes occurring in the interstellar medium.
Carbon, whose isotopes $^{12}$C and $^{13}$C are products of primary and secondary nucleosynthesis processes (Langer \& Penzias 1990), respectively, is well suited for such studies.
Within about 1--2 kpc of the Sun, the current average $^{12}$C/$^{13}$C ratio is about 70 (e.g., Wannier et al. 1982; Stahl \& Wilson 1992; Lucas \& Liszt 1998), with some variations; there also is an apparent systematic decrease in the ratio toward the Galactic center (e.g., Wilson 1999; Milam et al. 2005).
When determined from molecules unaffected by chemical fractionation (e.g., CH$^{+}$), the $^{12}$C/$^{13}$C ratio is an indicator of Galactic stellar evolution.
The (optical) absorption lines from $^{13}$CH$^+$, however, are typically very weak.
Because chemical fractionation can occur for CO, the $^{12}$CO/$^{13}$CO ratio gives additional information on chemical processes in the molecular gas (e.g., Bally \& Langer 1982; Sheffer et al. 1992).
Again, however, the limited sensitivity of {\it Copernicus} and {\it IUE} made it difficult to obtain reliable measures of the generally weak $^{13}$CO absorption.
The higher far-UV sensitivity of the {\it FUSE} instrument (Moos et al. 2000), together with the higher sensitivity and resolution of the {\it HST} GHRS and STIS echelle modes in the near-UV, however, has allowed us to revisit these issues.
The {\it FUSE} spectra contain numerous absorption lines from the Lyman and Werner bands of molecular hydrogen, from which accurate total H$_{2}$ column densities can be derived even toward fainter, more heavily reddened stars (e.g., Rachford et al. 2002).
While the B-X (0-0) (1150 \AA), C-X (0-0) (1087 \AA), and E-X (0-0) (1076 \AA) bands arising from the $X ^{1}\Sigma^{+}$ ground electronic state of CO are also present within the {\it FUSE} wavelength coverage, an accurate determination of the CO column densities from the {\it FUSE} data alone is often hindered by line blending, saturation, and unresolved structure in the absorption profiles.
With GHRS and STIS, however, one could observe weaker bands in the CO A-X system between 1240 and 1550 \AA\ at high enough resolution and S/N to discern and measure the individual rotational ($J$) levels of the bands --- allowing accurate estimates of their individual column densities.
Moreover, some of the weaker CO spin-forbidden (triplet-singlet) intersystem bands between 1300 and 1550 \AA\ can be detected once the permitted bands become saturated (Morton \& Noreau 1994) --- enabling more reliable determinations of higher CO column densities.
Several recent studies, for example, have used UV spectra from {\it HST} and/or {\it FUSE} to determine accurate $N$(CO) toward the moderately reddened stars $\zeta$ Oph (Lambert et al. 1994), $\rho$ Oph and $\chi$ Oph (Federman et al. 2003), X Per (Sheffer, Federman, \& Lambert 2002a; Sheffer, Lambert, \& Federman 2002b), and HD 203374A (Sheffer, Federman, \& Andersson 2003); Pan et al. (2005) have reported $N$(H$_2$) and $N$(CO) toward a number of stars in Cep OB2 and Cep OB3.
High-resolution GHRS spectra of the C$_2$ D-X (Mulliken) (0-0) band toward $\zeta$ Oph have enabled measurements of the C$_2$ rotational levels up to $J$ = 24 (Lambert, Sheffer, \& Federman 1995) --- higher $J$ than had been obtainable from the weaker A-X (Phillips) bands in the optical and near-IR.
In this paper, we present a study of the $^{12}$CO, $^{13}$CO, and/or C$_{2}$ molecules toward ten moderately reddened [$E(B-V)$ = 0.37--0.72] Galactic stars, based on UV spectra obtained with {\it FUSE}, {\it IUE}, and/or {\it HST}; on higher resolution optical spectra of CH, CN, and \ion{K}{1}; and on very high S/N optical spectra of C$_2$.
This combination of UV and optical data has allowed us to perform detailed analyses of the CO and C$_{2}$ profiles and to derive column densities for the individual rotational levels of the ground electronic and vibrational states --- up to $J$ = 3 for $^{12}$CO, $J$ = 2 for $^{13}$CO, and $J$ = 18 for C$_2$.
In \S~\ref{sec-obsred}, we discuss the observational data and the data reduction procedures.
In \S~\ref{sec-analysis}, we describe the methods used to determine column densities from the spectra.
In \S~\ref{sec-phys}, we combine our new results with existing literature data for H$_2$, HD, CH, C$_2$, C$_3$, CN, and CO (for a total sample of 74 stars) and discuss the physical conditions characterizing the diffuse molecular gas using this expanded star sample.
Column densities of H$_2$, CH, C$_2$, CN, $^{12}$CO, and $^{13}$CO for that total sample are tabulated in the Appendix.
In \S~\ref{sec-summ}, we summarize the main results of this study.
\section{Observations and Data Reduction}
\label{sec-obsred}
In order to investigate the behavior of the CO/H$_2$ and $^{12}$CO/$^{13}$CO ratios at higher overall extinctions and column densities, we selected sight lines according to the following general criteria:
(1) inclusion in the {\it FUSE} survey of translucent sight lines (Rachford et al. 2002; for H$_2$ and CO);
(2) availability of higher resolution {\it HST} GHRS or STIS spectra of the CO A-X bands (for $^{12}$CO, $^{13}$CO, and the CO rotational structure); and
(3) availability of high-resolution optical spectra of \ion{K}{1}, CH, and CN (for information on component structure).
For the sight line toward HD~73882 (Snow et al. 2000), which was not observed with {\it HST}, and for several other sight lines with {\it HST} data, we also have analyzed archival high-dispersion {\it IUE} spectra of the CO A-X and intersystem bands.
The ten sight lines in the resulting sample (Table~\ref{tab:stars}) have $E(B-V)$ between 0.37 and 0.72 mag, visual extinction $A_{\rm V}$ between 0.83 and 2.36 mag, $N$(H$_2$) from 3--13 $\times$ 10$^{20}$ cm$^{-2}$, and molecular fractions $f$(H$_2$) between 0.14 and 0.76.
The ten sight lines also probe a variety of regions in the Galactic interstellar medium --- with two (HD~24534, HD~27778) in Taurus/Perseus, one (HD~147888) in Sco-Oph, three (HD~206267, HD~207198, HD~210839) in Cep OB2, and one (HD~210121) through a high-latitude molecular cloud.
\subsection{{\it FUSE} Data}
\label{sec-fuse}
{\it FUSE} observed our sample stars under programs P116 and X021 (PI: T. P. Snow), which were dedicated to the study of translucent sight lines (Rachford et al. 2002).
The exposures used for each sight line are listed in Table~\ref{tab:uvdata}.
The observations were obtained in time tag mode through the low-resolution aperture (LWRS) and were processed with version 2.0.5 of the CALFUSE pipeline (Kruk \& Murphy 2001; Sahnow et al. 2000).
Eight detector segments cover sections of the total wavelength range from 912 to 1187 \AA, with most spectral regions covered by two to four segments.
The resolution is $\sim$0.062 \AA\ (FWHM), corresponding to about $\sim$18 km~s$^{-1}$\, or $\sim$9 pixels.
For each segment, the exposures were co-added after cross-correlating and shifting the spectra (by at most 8 pixels) with respect to the brightest exposure.
The spectra were binned by 4 pixels (slightly less than one-half resolution element) in order to increase the signal-to-noise (S/N) ratio without degrading the resolution delivered by the spectrograph optics.
The co-added spectra typically have S/N of about 10 per resolution element in the LiF1A spectrum and up to 15 per resolution element in the LiF2A spectrum.
The LiF1B spectrum was excluded from the data analysis because of the presence of a well-known detector artifact which causes an artificial flux deficiency in this segment (Kruk \& Murphy 2001).
Figure~\ref{fig:fuse} presents the profiles of the CO E-X (0-0) (1076 \AA), C-X (0-0) (1087 \AA), and B-X (0-0) (1150 \AA) bands, as functions of heliocentric velocity, detected in the far-UV spectra of seven of the ten sight lines.
\subsection{{\it HST} Data}
\label{sec-hst}
The calibrated GHRS or STIS spectra listed in Table~\ref{tab:uvdata} were retrieved from the MAST archive located at the Space Telescope Science Institute.
All the STIS data were automatically reduced with the latest version of the CALSTIS pipeline.
The STIS spectra were obtained using either the FUV MAMA detector and E140H echelle mode or the NUV detector and E230H echelle mode, at resolutions of either 2.75 km~s$^{-1}$\, (for the 0.''2 $\times$ 0.''2 or 0.''2 $\times$ 0.''09 slits) or 1.5 km~s$^{-1}$\, (for the 0.''1 $\times$ 0.''03 slit).
The E140H spectra generally cover the wavelength range from about 1150 to 1370 \AA\ --- including the A-X (7-0) through (12-0) bands of $^{12}$CO and $^{13}$CO (between 1240 and 1350 \AA) and the F-X (0-0) and (1-0) bands of C$_2$ at 1341 \AA\ and 1314 \AA\ (as well as many atomic absorption lines).
Unfortunately, the E140H spectra generally do not include the strongest CO intersystem bands.
The more extensive E140H spectra of HD~24534, which do include those intersystem bands, are described by Sheffer et al. (2002a, 2002b).
The E230H spectra cover the range from about 2130 to 2400 \AA\ --- including the D-X (0-0) band of C$_2$ at 2313 \AA\ (and various atomic lines).
The extracted spectral orders were corrected for scattered light contamination using the procedures described by Howk \& Sembach (2000) or Valenti et al. (2002).
For the spectra of HD~206267 and HD~210839, which were obtained through the narrowest slit, additional processing was performed by E. Jenkins (private communication) in order to account for unbalancing of the even and odd pixels on the detector (see Jenkins \& Tripp 2001 for details).
The continua were normalized to unity using low-order polynomials.
In most cases, the S/N ratios (estimated from the fluctuations in the continuum regions) range from 16 to 36 per resolution element; the C$_2$ F-X band spectra for HD~24534, however, have S/N $\sim$ 130, and the C$_2$ D-X band spectra (for three stars) have S/N $\sim$ 85--120.
Figures~\ref{fig:stis1}--\ref{fig:stis3} show the profiles of some of the $^{12}$CO A-X bands toward six of the stars; Figure~\ref{fig:13co} shows the $^{13}$CO A-X (7-0) (1347 \AA) band profiles toward four of the stars and the $^{13}$CO A-X (4-0) (1421 \AA) band toward HD~24534.
The individual rotational levels ($J$ = 0, 1, 2, 3) of the ground vibrational state resolved in each CO band are numbered above the spectra; for $^{12}$CO, the R, Q, and P branches are left-to-right in each case.
The GHRS ECH-A spectra of HD~210121 cover the $^{12}$CO A-X (2-0) (1477 \AA) and (4-0) (1419 \AA) bands, the corresponding $^{13}$CO bands (1478 and 1421 \AA), and the relatively strong $^{12}$CO a'14 intersystem band (1419 \AA), at a resolution of about 3.5 km~s$^{-1}$.
The GHRS ECH-B spectra of HD~24534 cover the C$_2$ D-X (0-0) band at 2313 \AA.
Because the spectra were obtained at several fp-split offsets, the STSDAS routines POFFSETS, DOPOFF, and SPECALIGN were used to align and combine the individual spectra.
The S/N values range from 15 to 20 per resolution element for HD~210121 and are about 100 per resolution element for HD~24534.
\subsection{{\it IUE} Data}
\label{sec-iue}
High-dispersion (FWHM $\sim$ 25 km~s$^{-1}$) {\it IUE} spectra of HD~27778, HD~73882, and HD~207198 were retrieved from the archive (Table~\ref{tab:uvdata}).
For all three stars, multiple spectra had been obtained at several different offsets within the large aperture, in order to reduce the effects of fixed-pattern noise when the spectra are combined.
For these three stars, the S/Ns range from about 15--45 per resolution element in the aligned, co-added spectra, for wavelengths between about 1320 and 1550 \AA\ [covering the CO A-X bands (0-0) through (8-0)].
Figure~\ref{fig:iue} shows some of the stronger $^{12}$CO A-X bands toward HD~27778 and HD~73882 (see also Figure~7 of Morton \& Noreau 1994); the corresponding $^{13}$CO A-X bands and some of the intersystem bands of $^{12}$CO can also be discerned.
While the permitted A-X bands are saturated toward both of those stars, the stronger intersystem bands toward HD~73882 indicate a higher $N$(CO) in that line of sight.
The {\it IUE} data for HD~210121 were described by Welty \& Fowler (1992).
\subsection{Optical Data}
\label{sec-opt}
High-resolution (FWHM $\sim$ 0.6--3.6 km s$^{-1}$) spectra of \ion{K}{1} (7698 \AA), CH (4300 \AA), and CN (3874 \AA) were obtained with the Kitt Peak coud\'{e} feed telescope and the McDonald 2.7m telescope.
A detailed discussion of the reduction and analysis of the \ion{K}{1} spectra has been given by Welty \& Hobbs (2001); the CH and CN spectra, processed in similar fashion, will be discussed by Welty, Snow \& Morton (in preparation).
Profiles of the optical lines for six of the sight lines discussed in this paper are included in Figures~\ref{fig:stis1}--\ref{fig:stis3}; some have also been shown by Pan et al. (2004).
Medium resolution (FWHM $\sim$ 8 km s$^{-1}$) but very high S/N ($\ga$ 1000) optical spectra were acquired for nine of the stars with the 3.6m telescope and the ARC echelle spectrograph at Apache Point Observatory as part of an observing program designed to investigate the nature of the diffuse interstellar bands.
The details on the reduction and analysis of these spectra and, in particular, of the C$_{2}$ A-X (Phillips) bands detected in seven of the sight lines studied here, are described in Thorburn et al. (2003).
\section{Data Analysis}
\label{sec-analysis}
Several factors have made it difficult to derive accurate CO column densities for more heavily reddened sight lines.
The lower resolution UV spectra from {\it Copernicus}, {\it IUE}, and {\it FUSE} yield equivalent widths for a number of the CO bands, but do not resolve either the (often complex) component structure or the rotational structure of the bands.
Blends with stellar lines can also complicate the interpretation of the spectra, especially when the S/N and/or $vsini$ are relatively low.
In addition, for these higher column density sight lines, many of the permitted CO bands that can be detected are significantly saturated.
Fortunately, the higher resolution (and often higher S/N) spectra obtained with the {\it HST} GHRS and STIS echelle modes at least partially resolve the CO rotational structure and enable reliable measurement of weaker CO A-X and intersystem bands even toward fainter, more heavily reddened targets.
Agreement between several recent observational and theoretical studies suggests that the $f$-values now available for the intersystem bands are reliable (Sheffer et al. 2002a; Eidelsberg \& Rostas 2003).
Finally, still higher resolution optical spectra of \ion{K}{1}, CH, and (especially) CN can provide useful information as to the velocity component structure --- enabling more accurate modeling of the observed CO line profiles.
Three methods were used to estimate the interstellar CO column densities.
First, empirical curves of growth were constructed using the measured equivalent widths of the CO bands.
The empirical curves were then compared with theoretical curves generated for single components with a range in the Doppler broadening parameter ($b$), for two representative CO rotational population distributions characterized by different excitation temperatures.
Second, the Apparent Optical Depth method (Hobbs 1971; Savage \& Sembach 1991) was applied to the observed CO line profiles.
Third, the absorption-line profiles were fitted using multi-component models derived from higher resolution optical spectra of \ion{K}{1}, CH, and CN.
Profile fitting was used to determine column densities from the UV and optical lines of C$_2$.
Because the individual gas components along these sight lines are at best partially resolved in the UV spectra, determination of the column densities for individual components is often uncertain with STIS data and essentially impossible with the lower resolution {\it FUSE} data.
We will therefore only discuss total sight line column densities in the present work.
\subsection{Curves of Growth}
\label{sec-cog}
In order to construct the empirical curves of growth (COG), total equivalent widths of the $^{12}$CO and $^{13}$CO bands seen in the normalized UV spectra were measured over velocity ranges which include all the detected rotational lines.
Table~\ref{tab:ew1} lists the CO equivalent widths found for the A-X, B-X, C-X, and E-X bands from {\it FUSE} and STIS data (eight sight lines); Table~\ref{tab:ew2} gives the values for the A-X and intersystem bands found from GHRS, STIS, and/or {\it IUE} data (six sight lines).
The CO equivalent widths for the well-studied sight lines toward X Per (HD 24534) and $\zeta$ Oph (HD 149757) are included in Table~\ref{tab:ew2} for comparison.
In all cases, the listed 1-$\sigma$ uncertainties include contributions from both photon noise and continuum-fitting uncertainty, added in quadrature; upper limits are 3-$\sigma$.
Because of the wide velocity range covered by the ensemble of rotational lines, the continuum-fitting uncertainties generally are dominant.
Inspection of the equivalent widths for HD~27778 and HD~207198 indicates that there is generally good agreement between the values determined from {\it IUE} and STIS spectra for the CO A-X (7-0) and (8-0) bands at 1322 and 1344 \AA.
The equivalent widths for HD~185418 and HD~192639 are consistent with those listed by Sonnentrucker et al. (2002, 2003).
Because the CO C-X (0-0) band at 1087 \AA\ is often blended with an adjacent \ion{Cl}{1} line (see Fig.~\ref{fig:fuse}) and, in the B stars in our sample, with a stellar absorption feature, it was not used in the construction of the curves of growth.
An estimate of its equivalent width is, nevertheless, given for the two stars toward which the blending was less severe.
The E-X (0-0) band at 1076 \AA\ is in the wing of the very strong H$_2$ (2-0) R(0) line at 1077 \AA, making the appropriate continuum difficult to estimate in some cases.
The rest wavelengths and oscillator strengths listed in Tables~\ref{tab:ew1} and \ref{tab:ew2} were taken from Morton \& Noreau (1994) for the CO A-X bands and from Federman et al. (2001) for the far-UV B-X (0-0) (1150 \AA), C-X (0-0) (1087 \AA), and E-X (0-0) (1076 \AA) bands.
For the CO intersystem bands, we have adopted the rest wavelengths of Eidelsberg \& Rostas (2003) and (where possible) the empirical $f$-values of Sheffer et al. (2002a), which yield smoother curves of growth for the equivalent widths measured toward both HD~24534 (X Per) and HD~149757 ($\zeta$ Oph) than do the corresponding $f$-values of Eidelsberg \& Rostas.
For the $^{12}$CO bands with both {\it FUSE} and STIS data, the range in $f\lambda$ can span nearly three orders of magnitude from the weakest A-X (12-0) (1246 \AA) to the strongest E-X (0-0) (1076 \AA) band; a similar range in $f\lambda$ is covered for sight lines with both STIS and {\it IUE} data.
In principle, the CO curve of growth depends on (1) the rotational structure of the individual bands, (2) the excitation temperatures characterizing the various rotational levels, and (3) the interstellar component structure (e.g., Black \& van Dishoeck 1988).
Theoretical curves of growth were generated for single interstellar components with a range of $b$-values (0.3--5.0 km~s$^{-1}$\ --- as an approximation to more complex component structures), using the CO rotational structure of the A-X (6-0) band (taken as representative of all the A-X bands).
Separate curves were generated for the far-UV E-X, C-X, and B-X bands, for which the rotational structure is somewhat different.
For $^{12}$CO, curves were generated for two sets of relative populations in the $J$ = 0--3 rotational levels.
The first had level populations 0/1/2/3 = 1.0/1.0/0.2/0.05, which are similar to the average ratios found for the sight lines in our small sample.
The relative populations in the $J$ = 0--2 levels, for example, correspond to an excitation temperature of about 5 K.
The second had level populations 0/1/2/3 = 1.0/0.5/0.02/0.0001, corresponding to an excitation temperature of about 3 K (typical of some other sight lines reported in the literature).
For $^{13}$CO, equal populations were assumed for $J$ = 0 and 1 (the only levels generally detected in the STIS spectra).
The sums of the $f$-values for the various individual transitions (R, Q, and P branches) for each $J$ are all equal to the $f$-value of the band as a whole, and the COG analysis of the integrated band equivalent widths yields the total column density over all $J$.
Columns (2) and (7) in Table~\ref{tab:nco} list the total $^{12}$CO and $^{13}$CO column densities (respectively) derived by comparing the empirical curves of growth with the theoretical single-component curves for $T_{\rm ex}$ $\sim$ 3 K and/or for $T_{\rm ex}$ $\sim$ 5 K; column (3) lists the best-fit ``effective'' $b$-value.
In each case, the uncertainty given for the column density corresponds to the range in $N$(CO) giving rms deviations less than 1.5 times that of the best ($N$,$b$) pair, for that best-fit $b$-value.
Figure~\ref{fig:cog} shows the $^{12}$CO curves of growth for the sight lines toward HD~27778 and HD~207198, with the adopted values for $N$ and $b$.
In the figure, the solid lines show the $T_{\rm ex}$ $\sim$ 5 K theoretical curves for the A-X bands (which in most cases should be adequate for the intersystem, B-X, and C-X bands); the dotted lines show the theoretical curves for the E-X (0-0) band at 1076 \AA, displaced downward by 1.0 for clarity.
The data point for the observed E-X (0-0) band is plotted against both sets of theoretical curves to show that the curves for the E-X (0-0) band lie slightly below those for the A-X bands.
For most of the sight lines, at least two CO bands lie on or near the linear part of the curve of growth.
Accurate total column densities can be derived from such optically thin bands with minimal assumptions regarding the gas velocity distribution.
When additional, stronger bands are measured, the COG analysis can (in principle) also yield some information regarding the overall gas distribution, as long as (1) the $f$-value range for the stronger bands is large enough that different degrees of saturation are observed and (2) the CO excitation temperature is reasonably well determined.
For eight of the nine sight lines analyzed, the effective $b$-values for the empirical curves, obtained by comparison with the theoretical curves, range from about 0.5 to 2.0 km~s$^{-1}$.
The very small effective $b$-values for HD~147888 and HD~210839 (0.5 km~s$^{-1}$) likely reflect the widths of the single dominant narrow components seen in CN in those two sight lines.
The higher values for HD~27778 and HD~206267 (both $\sim$ 1 km~s$^{-1}$) and for HD~207198 ($\sim$ 2.0 km~s$^{-1}$), however, presumably correspond to the multiple components seen in CN in those sight lines (rather than the widths of single components).
Slight differences between the empirical and theoretical curves for several sight lines may be due to the more complex component structure (seen in the higher resolution optical spectra) in those cases.
Comparisons of the observed equivalent widths with the theoretical curves for $T_{\rm ex}$ $\sim$ 3 K (which fall slightly below the corresponding curves for $T_{\rm ex}$ $\sim$ 5 K) generally exhibited the best agreement for somewhat higher column densities at the same effective $b$ or else for similar column densities and slightly higher effective $b$.
While the column densities derived via the COG thus depend on knowledge of $T_{\rm ex}$ (obtained via detailed fits to the high-resolution line profiles), they do provide useful confirmation of the total $N$(CO) derived in the profile fits.
\subsection{Apparent Optical Depth}
\label{sec-aod}
The Apparent Optical Depth (AOD) analysis was applied to all detected $^{12}$CO and $^{13}$CO bands, over the same velocity ranges used for the equivalent width measurements.
In the absence of unresolved saturated structure(s) within the line profile, the AOD method yields the ``apparent'' (instrumentally smeared) column density as a function of velocity, with some improvement over the simplest assumption of a linear relationship between equivalent width and column density.
If the line is somewhat saturated, the AOD analysis yields a lower limit to the actual column density, especially if the spectral resolution is not very high.
In principle, the degree of saturation in the profiles of a given species can be assessed if multiple lines, with $f$-values differing by at least a factor of 2, are available (Jenkins 1996).
For CO, the distribution of the total column density over multiple rotational levels (some blended) renders the specific $N$($v$) not very meaningful, but it does delay somewhat the onset of saturation.
As for the COG analysis, the total column density (over the whole profile and for all rotational levels) can still be estimated.
For most of the sight lines, smaller total column densities are obtained from the stronger CO bands (relative to those obtained from the weakest observed bands) --- indicative of more severe saturation in the stronger bands.
Toward HD~207198, for example, the AOD analysis of the strongest CO band observed with STIS [A-X (7-0) at 1344 \AA] yields a total $N$(CO) smaller by a factor of about 1.5 than that found for the weaker CO bands.
Toward HD~210839 and HD~206267, saturation effects are already apparent in the A-X (9-0) (1301 \AA) band, and the total column densities inferred from the A-X (7-0) band at 1344 \AA\ are factors of 3--4 lower than those obtained from the weakest bands.
In such cases, however, accurate column densities still can be determined by using only the weaker A-X (12-0), (11-0), and (10-0) bands (at 1246, 1263, and 1281 \AA), which yield concordant results.
The AOD column densities inferred from the weakest CO bands are listed in Columns (3) and (7) of Table~\ref{tab:nco}; the associated error bars include contributions from both photon noise and continuum-fitting uncertainty.
In most cases, the total column densities derived from the AOD analysis are consistent with those obtained with the COG method.
\subsection{Profile Fitting}
\label{sec-fits}
While the STIS resolution is high enough to at least partially resolve the individual rotational levels in the CO and C$_2$ band structures (Figs.~\ref{fig:stis1}--\ref{fig:13co} and \ref{fig:fit1}--\ref{fig:c2uv}), it is not sufficient to separate or resolve all of the multiple, narrow gas components that produce the observed absorption lines.
The blending between components --- both interstellar and rotational --- is particularly severe for the shorter wavelength R-branch lines of the CO A-X bands (Figs.~\ref{fig:stis1}--\ref{fig:stis3}) and the C$_2$ F-X bands (Fig.~\ref{fig:c2uv}).
In order to determine accurate (total) column densities for the individual rotational levels contributing to the molecular bands, the absorption-line profiles were fitted with multi-component models based on fits to higher resolution optical spectra.
The profile fitting program ``Owens'' (Lemoine et al. 2002) and a variant of the program FITS6P (e.g., Welty, Hobbs, \& Morton 2003) --- both of which perform iterative least-squares fits to the observed line profiles --- were used for the detailed component analyses of the optical and UV spectra.
While lines of molecular species such as CH, C$_2$, and CN are the most direct optical tracers of diffuse molecular gas, the good correlations observed among the column densities of H$_2$, CH, and \ion{K}{1} --- and the often striking correspondence between the line profiles of CH and \ion{K}{1} --- suggest that component information from \ion{K}{1} may be relevant as well (Welty \& Hobbs 2001).
More detailed structure is often discernible in the \ion{K}{1} line profiles because (1) the potassium atom is heavier than the CH molecule, (2) the hyperfine splitting of the \ion{K}{1} $\lambda$7698 line ($\sim$ 0.3 km~s$^{-1}$) is smaller than the separation between the lambda-doubled components of the CH $\lambda$4300 line (1.43 km~s$^{-1}$), and (3) the \ion{K}{1} line is typically stronger.
The more detailed \ion{K}{1} profile structures (i.e., the relative velocities and $b$-values for each discernible component) were therefore used to model the corresponding CH profiles.
For the ten sight lines in this study, between three and thirteen components were needed to fit the \ion{K}{1} profiles; not all were detected in CH, however.
Because CN is likely to be more concentrated in the colder, denser regions of each sight line, independent fits were performed to the CN profiles; one to four components (typically corresponding to the main components seen in \ion{K}{1} and CH) were required.
The structures found from the high-resolution optical spectra of \ion{K}{1}, CH, and/or CN (Appendix~\ref{sec-comp}; Table~\ref{tab:comp}) were then taken to represent the component distributions for the UV CO and C$_{2}$ bands in each sight line.
Knowledge of the component structure can be crucial for determining column densities from the stronger CO lines, but is generally less important for the corresponding analyses of the much weaker C$_2$ lines.
\subsubsection{CO Column Densities}
\label{sec-nco}
The component structures found for \ion{K}{1} and CH were used for the initial fits to the CO band profiles observed with GHRS and STIS.
In the Owens fits, the relative velocity and the $b$-value of each component were fixed, while the continuum placement and individual component column densities were allowed to vary.
The FITS6P fits were performed on the normalized spectra, fixing the relative velocities (usually) and the $b$-values but allowing the component column densities to vary.
In both cases, all the CO A-X band profiles were fitted simultaneously.
For sight lines where both Owens and FITS6P were used to analyze the CO bands, the two methods generally yielded consistent results (within the respective uncertainties) both for the column densities of the individual rotational levels ($J$ = 0--3) and for the total $N$(CO).
Toward HD~192639, HD~185418, and HD~207198, all the CO bands were adequately fitted using the \ion{K}{1} component structures.
Toward HD 27778, HD~206267, and HD~210839, however, fits using the \ion{K}{1} structures tended to under-produce the weakest bands and overproduce the strongest bands.
In an attempt to obtain mutually consistent column densities for all the bands in those sight lines, two additional series of fits were undertaken.
In the first additional series, the relative velocities of the various gas components were again fixed to those of \ion{K}{1}, but both the column densities and the $b$-values were allowed to vary.
For all three sight lines, the best fits to the CO bands were realized with $b$-values between 0.3 and 0.7 km~s$^{-1}$\ for the main components --- i.e., somewhat smaller than the 0.6--1.1 km~s$^{-1}$\ values found for \ion{K}{1}.
The second additional series of fits, using the (simpler) CN velocity structures (but with slightly smaller $b$-values), also yielded more consistent fits to the ensemble of observed profiles.
Inspection of the observed line profiles suggests that the profiles of the individual CO rotational lines seen in the higher resolution STIS spectra of HD~206267 and HD~210839 generally are more similar to the profiles of CN than to those of \ion{K}{1} and CH.
A close correspondence between the component structures for CO and CN has also been found by Pan et al. (2005) toward a number of stars in the Cep OB2 and Cep OB3 associations, including HD~206267, HD~207198, and HD~210839.
For those three lines of sight, our total CO column densities are consistent with those determined by Pan et al.
Examples of the fits to some of the CO A-X system profiles toward HD~27778 and HD~210839 are shown in Figures~\ref{fig:fit1} and \ref{fig:fit2}, respectively, together with the corresponding higher resolution optical profiles of \ion{K}{1}, CH, and CN.
The fits to the profiles for HD~27778 utilize two components separated by 2.3 km~s$^{-1}$, with $b$ $\sim$ 0.5--0.6 km~s$^{-1}$\ and relative column densities $\sim$ 3:1 --- i.e., similar to the structure found for CN, but with slightly smaller $b$-values and somewhat different relative strengths (Table~\ref{tab:comp}).
That structure (with the rotational level column densities $N_{J}$ obtained from fitting the ensemble of lines) provides acceptable fits to both weak and strong CO bands, including the stronger A-X bands observed at lower resolution with {\it IUE}.
The best fits to the profiles for HD~210839 employ two components separated by 2.1 km~s$^{-1}$, with $b$ $\sim$ 0.4 km~s$^{-1}$\ (slightly smaller than the $b$-values adopted for CN) and relative strengths roughly 3:1.
Because the lower resolution {\it IUE} spectra of HD~27778, HD~73882, HD~207198, and HD~210121 do not resolve the CO rotational structure, the line profiles and total equivalent widths of the various permitted A-X and intersystem CO bands listed in Table~\ref{tab:ew2} were fitted by assuming rotational excitation temperatures of 3 and 5 K, using the component structures determined for \ion{K}{1} and CN.
The resulting total sight line CO column densities toward HD~27778 and HD~207198 are consistent with those derived from the higher resolution STIS spectra, but differ somewhat from the values determined by Joseph et al. (1986) (and Federman et al. 1994a) from curve of growth analyses of more limited {\it IUE} spectra of the permitted A-X bands (only).
While no high-resolution CN spectra are available for HD~73882, three components are seen in both \ion{K}{1} absorption and mm-wave CO emission (van Dishoeck et al. 1991).
Three-component fits to the intersystem and weakest permitted A-X bands of $^{12}$CO, assuming $T_{\rm ex}$ = 5 K and several choices for the relative column densities in each component, yield column densities consistent both with those determined from the COG analysis and with the value estimated from the CO emission lines (van Dishoeck et al. 1991).
Because the higher resolution GHRS spectra of HD~210121 allow better recognition of narrow stellar absorption features blended with several of the interstellar CO lines, the column density of $^{13}$CO is significantly smaller than that estimated by Welty \& Fowler (1992) from the {\it IUE} spectra alone.
Adoption of a smaller CO $b$-value (0.5 km~s$^{-1}$ vs. 1.0 km~s$^{-1}$) for that sight line, however, yields a somewhat higher column density for $^{12}$CO.
The results of the CO profile fitting (FIT) analyses are summarized in Table~\ref{tab:nco}, which lists the total $^{12}$CO and $^{13}$CO column densities (for all gas components and all rotational levels), and in Table~\ref{tab:conj}, which lists the column densities $N_{J}$ for the individual rotational levels ($J$ = 0--3) for the eight sight lines with high-resolution UV spectra.
Where only $N_0$ has been measured for $^{13}$CO, we have assumed that $N_0$/$N_1$ (or the excitation temperature $T_{01}$) is the same as for $^{12}$CO.
The uncertainties in the $N_{J}$ include contributions from photon noise, continuum uncertainty, and uncertainties in the component $b$-values ($\pm$ 0.05--0.10 km~s$^{-1}$).
The continuum uncertainties are most significant for the weaker lines; the $b$-value uncertainties generally are most significant for the stronger lines (especially for $b$ $\la$ 0.5 km~s$^{-1}$).
While the column densities derived from the AOD, COG, and FIT methods agree within the mutual uncertainties, we have adopted the $N$(CO) determined in the profile fits, as they account explicitly for saturation (using the detailed component structures), they use information from all the observed transitions, and they yield values for the individual rotational levels.
The better fits to the CO bands obtained both with smaller $b$-values and with the CN component structure suggest that CO (like CN) may preferentially sample colder, denser gas than \ion{K}{1} and CH (see also Pan et al. 2005).
Discussions of the physical conditions in the gas in the following sections seem consistent with that interpretation.
\subsubsection{C$_{2}$ Column Densities}
\label{sec-nc2}
Nearly all of the existing studies of interstellar C$_2$ have been based on optical and/or near-IR spectra of the A-X (Phillips) bands [(1-0) at 10144 \AA, (2-0) at 8757 \AA, and (3-0) at 7719 \AA] (e.g., Souza \& Lutz 1977; Hobbs \& Campbell 1982; van Dishoeck \& de Zeeuw 1984; van Dishoeck \& Black 1986b; Federman et al. 1994a; Gredel 1999).
In one exception, high-resolution GHRS ECH-B spectra of the D-X (Mulliken) (0-0) band at 2313 \AA\ and lower resolution G160M spectra (FWHM $\sim$ 16 km~s$^{-1}$) of the F-X (0-0) band at 1341 \AA\ toward $\zeta$ Oph were used to obtain the C$_2$ rotational level populations up to $J$ = 24 and to estimate the relative $f$-values for the C$_2$ A-X, D-X, and F-X systems (Lambert, Sheffer, \& Federman 1995).
Many of the lines from the individual rotational levels in the F-X (0-0) band are inextricably blended in the G160M spectrum, however.
Kaczmarczyk (2000) has discussed GHRS echelle spectra of the F-X (0-0) and D-X (0-0) bands toward HD~24534.
In this paper, we present high-resolution profiles of the UV C$_{2}$ F-X (0-0) and (1-0) bands (as seen in the STIS E140H spectra of HD~24534, HD~27778, and HD~206267; Figures~\ref{fig:c2uv} and \ref{fig:fx00}), along with high-resolution GHRS ECH-B and STIS E230H spectra of the D-X (0-0) band toward HD~24534, HD~27778, HD~147888, and HD~207198 (Figure~\ref{fig:c2dx}) and high S/N optical spectra of the A-X (3-0) and (2-0) bands toward seven stars (Figures~\ref{fig:c2opt} and \ref{fig:c2opt3}).
All these UV and optical spectra at least partially resolve the rotational structure of the C$_2$ bands --- enabling detailed line profile analyses of the C$_2$ absorption.
The D-X (0-0) band is particularly useful, as the individual rotational lines from both the R and P branches are well separated and relatively strong.
Equivalent widths for the individual D-X (0-0) lines are given in Table~\ref{tab:c2dx}; equivalent widths for lines in the A-X (2-0) and (3-0) bands are listed in Table~\ref{tab:c2optw}.
Wavelengths and relative strengths for the individual rotational lines in the various C$_2$ bands --- as well as the adopted total band $f$-values --- are listed in Table~\ref{tab:c2bands} (Appendix~\ref{sec-c2bands}).
By comparing column densities derived from the various UV and optical bands, a set of mutually consistent band oscillator strengths may be determined.
Lambert et al. (1995) adopted a band oscillator strength $f$ = 0.0545 for the D-X (0-0) band, based on two independent (and closely concordant) theoretical estimates.
Comparison of the overall strengths of the D-X (0-0) and F-X (0-0) bands toward $\zeta$ Oph then yielded $f$ = 0.10$\pm$0.01 for the latter band --- in excellent agreement with the theoretical value 0.098 determined by Bruna \& Grein (2001).
With those $f$-values, fits to the profiles of the D-X (0-0) and F-X (0-0) bands toward HD~24534 and HD~27778 yield reasonably consistent C$_2$ column densities for rotational levels $J$ = 0--6 (Table~\ref{tab:nc2}), though there appear to be some systematic discrepancies for $J$ = 8 and 10 (see below).
Similar fits to the F-X (1-0) band at 1314 \AA\ toward HD~24534 then suggest a total oscillator strength $f$ $\sim$ 0.06 for that band.
As for CO, the \ion{K}{1} velocity structure was initially adopted to determine the column densities of the individual C$_2$ rotational levels present in the spectra of the three UV bands.
Additional fits were performed using either a single component (with an effective $b$) or the structure that best accounted for the saturation in the stronger CO bands; a separate three-component fit was made to the D-X (0-0) band toward HD~207198, where more complex structure (similar to that seen in \ion{K}{1} and CH) is evident (Figs.~\ref{fig:stis3}~and~\ref{fig:c2dx}).
In each case, the various fits led to similar results (within the uncertainties), indicating that the total C$_2$ column densities derived from the C$_{2}$ D-X (0-0) and F-X (0-0) and (1-0) bands generally are not very sensitive to the details of the gas velocity distribution along these lines of sight.
Column densities for individual C$_2$ rotational levels up to $J=$ 12 were thus obtained from the F-X band(s) toward HD~24534, HD~27778, and HD~206267 and up to $J=$ 18 from the D-X (0-0) band toward HD~24534, HD~27778, HD~147888, and HD~207198 (Table~\ref{tab:nc2}).
The listed uncertainties include contributions from photon noise, continuum-fitting uncertainty, and uncertainty in the effective $b$; the uncertainties for the D-X (0-0) band are similar to those that would be inferred from the uncertainties in the equivalent widths of the individual rotational lines.
The relative rotational level populations $N_{J}$/$N_2$ appear to be fairly similar for these five lines of sight (but see the caveats below).
Toward HD~210839, the C$_2$ F-X (0-0) band absorption is significantly weaker, so that detailed rotational distributions could not be determined from the UV spectra.
Detailed examination of the fits to the various C$_2$ band profiles and comparisons of the C$_2$ column densities obtained from those bands, however, indicate that the strength and structure of the UV C$_2$ bands --- particularly the F-X (0-0) band --- are not completely understood.
On the one hand, the fits to the D-X (0-0) and F-X (1-0) bands toward HD~24534 are quite good (Figs.~\ref{fig:c2dx} and \ref{fig:c2uv}) and yield mutually consistent column densities for the individual rotational levels $J$ = 0--10.
Moreover, while there are some ``discrepant'' $N_{J}$ for individual sight lines, there do not appear to be any systematic differences in the $N_{J}$ determined from the D-X (0-0), A-X (3-0), and A-X (2-0) bands (Table~\ref{tab:nc2}).
The fits to the F-X (0-0) band toward HD~24534, HD~27778, and HD~206267, however, all are rather poor near the ``pile-up'' of the R(6)--R(12) lines; some additional broad (?) absorption appears to be present there.
In addition, the calculated wavelengths of the clearly detected F-X (0-0) lines ($J$ $<$ 10) seem to be too small by about 36 m\AA\ (8 km~s$^{-1}$).
Finally, the lines from $J$ = 6--10 seem to be weaker in the F-X (0-0) band than in the intrinsically weaker F-X (1-0) band; compare, for example, the relatively unblended Q(8), Q(10), and P(8) lines of the two bands toward HD~24534 in the upper half of Figure~\ref{fig:c2uv}.
The discrepancies between the C$_2$ D-X (0-0) and F-X (0-0) bands previously noted in GHRS spectra of $\zeta$ Oph (Lambert et al. 1995) and HD~24534 (Kaczmarczyk 2000) may reflect similar effects.
Lambert et al. might not have recognized either the ``extra'' absorption near the F-X (0-0) band R-branch ``pile-up'' or the 8 km~s$^{-1}$ velocity offset in the lower resolution G160M spectrum.
Their fits to the whole F-X (0-0) band, which uniformly scaled the rotational level column densities determined from the D-X (0-0) band at 2313 \AA, would thus have tended to predict too much absorption over the rest of the band --- as seen, for example, for the Q(10)-P(4), Q(12)-P(6), and Q(14)-P(8) blends in their Figure~3.
Kaczmarczyk found that the $N_J$ derived from the D-X (0-0) band did not yield a good fit to the Q(12)-P(6) and Q(14)-P(8) blends in the F-X (0-0) band and noted (unspecified) problems with the F-X (0-0) band wavelengths.
While the ``extra'' absorption near the F-X (0-0) band R-branch ``pile-up'' is not evident in the lower S/N GHRS spectrum analyzed by Kaczmarczyk, he did mention a ``lower than expected'' continuum near several of the stronger lines.
The causes of the discrepant behavior observed for the F-X (0-0) band are not understood.
In principle, the ``extra'' broad absorption near the R-branch pile-up could be a stellar feature, but the three stars observed (HD~24534, HD~27778, and HD~206267) have somewhat different spectral types, radial velocities, and projected rotational velocities.
Moreover, the ``extra'' absorption toward HD~24534 is very similar in three separate sets of STIS spectra taken at epochs separated by about 3 and 33 days, respectively.
The wavelengths adopted for both of the F-X bands (Table~\ref{tab:c2bands}) were calculated from molecular constants derived from flash discharge spectra analyzed by Herzberg, Lagerqvist, \& Malmberg (1969).
The 8 km~s$^{-1}$ velocity offset for the clearly detected (and unblended) lines in the F-X (0-0) band (all with $J$ $<$ 10) does not appear to be due to errors in the STIS wavelengths, as the (0-0) band appears in two adjacent orders, one of which also contains the $^{12}$CO A-X (7-0) band at 1344 \AA\ and the \ion{Cl}{1} line at 1347 \AA.
While the velocity offset for those lower $J$ F-X (0-0) band lines would be consistent with a slightly smaller value for the upper level $\nu_0$ (74530.9 cm$^{-1}$ instead of 74532.9 cm$^{-1}$), the wave numbers for the higher $J$ lines measured by Herzberg et al. (1969) generally are consistent (within $\pm$ 0.2 cm$^{-1}$) with the values predicted using their tabulated constants.
Unfortunately, Herzberg et al. apparently could not measure the positions of any of the lower $J$ lines in the F-X (0-0) band, and there are no clear detections of any of the unblended higher $J$ lines in the spectra shown in Figure~\ref{fig:c2uv}.
There are, however, tantalizing hints of weak absorption at the predicted (i.e., unshifted) positions of several of the higher $J$ Q-branch lines in the spectrum of HD~24534 [which was taken from data set o66p01020, the longest single STIS exposure of HD~24534 covering the F-X (0-0) band].
When STIS spectra of HD~24534 obtained at two other epochs are combined with the spectrum shown in Figure~\ref{fig:c2uv} (more than doubling the total exposure time), those weak features persist.
Fits to that combined spectrum with only the lines from $J$ $<$ 10 shifted yield low values of $N_J$ for $J$ = 8--12, but quite reasonable values of $N_J$ for $J$ = 14--18 (Figure~\ref{fig:fx00}, Table~\ref{tab:nc2}).
Column densities for the individual C$_2$ rotational levels were also obtained from the A-X (2-0) and (3-0) bands at 8757 and 7719 \AA\ (Thorburn et al. 2003), using both the measured equivalent widths of the individual rotational lines in the medium-resolution ARCES spectra and fits to the ensemble of lines (including blended features).
Figure~\ref{fig:c2opt} shows normalized spectra of the optical C$_2$ A-X (2-0) band toward seven stars; Figure~\ref{fig:c2opt3} shows corresponding spectra of the A-X (3-0) band toward five stars.
The profiles observed toward HD~204827, which has high column densities of both C$_2$ and C$_3$ (Thorburn et al. 2003; Oka et al. 2003; \'{A}d\'{a}mkovics et al. 2003) are included to show the band structures more clearly.
Table~\ref{tab:c2optw} lists the wavelengths, log($f\lambda$) values, and equivalent widths for the A-X (2-0) and (3-0) band transitions toward seven of our stars (plus HD~204827); C$_2$ was not detected toward HD~185418 or HD~192639 and was not observed toward HD~73882 (though see Gredel et al. 1993).
The equivalent widths of A-X (2-0) R-branch lines located in the steep wing or deep core of the stellar Paschen 12 line of \ion{H}{1} can be rather uncertain; in some cases (e.g., toward HD~24534), some of those R-branch lines could not be reliably measured.
There is also a weak diffuse interstellar band nearly coincident with the A-X (3-0) Q(2) line in several cases (Herbig \& Leka 1991).
The equivalent widths in Table~\ref{tab:c2optw} are consistent with most of the (generally lower precision) values previously reported for HD~27778 (Federman et al. 1994a); HD~24534, HD~206267, HD~207198, and HD~210839 (Federman \& Lambert 1988); and HD~210121 (Gredel et al. 1992).
Unfortunately, the band oscillator strengths for the C$_2$ A-X bands have been somewhat uncertain, with persistent differences between theoretical and experimental estimates (e.g., van Dishoeck \& Black 1989; Langhoff et al. 1990; Lambert et al. 1995; though see Erman \& Iwamae 1995).
Lambert et al. (1995) adopted $f$ = 0.00123$\pm$0.00016 for the A-X (2-0) band, based on comparisons with the UV D-X (0-0) and F-X (0-0) bands toward $\zeta$ Oph and the D-X (0-0) band $f$-value noted above.
That empirical value lies roughly halfway between the theoretical estimate $f$ = 0.00144$\pm$0.00007 of Langhoff et al. (1990) and the experimental value $f$ = 0.00100 adopted by van Dishoeck \& Black (1989).
Similar comparisons based on our fits to the profiles of the F-X, D-X, and A-X bands observed toward HD~24534, HD~27778, HD~147888, HD~206267, and HD~207198 suggest a somewhat higher $f$-value for the A-X (2-0) band, however.
We have therefore adopted the value $f$ = 0.00140 --- an average of the $f$ = 0.00136$\pm$0.00015 measured by Erman \& Iwamae (1995) and the value calculated by Langhoff et al. (1990).
On average (toward HD~24534, HD~27778, HD~204827, HD~206267, HD~210121, and several other stars with data for these two Phillips bands), the strength of the C$_2$ A-X (3-0) band then is consistent with $f$ $\sim$ 0.00065 --- i.e., weaker than the (2-0) band by the expected factor $\sim$ 2.2 (van Dishoeck \& Black 1982).
The C$_2$ column densities derived from fits to the A-X (2-0) and (3-0) bands (with the adopted $f$-values) are listed in Table~\ref{tab:nc2}.
The values for $\zeta$ Oph given in Table~\ref{tab:nc2} were derived from published equivalent widths (Hobbs \& Campbell 1982; Danks \& Lambert 1983; van Dishoeck \& Black 1986b; Lambert et al. 1995).
The adopted C$_2$ band $f$-values yield quite good agreement for the column densities derived from the D-X (0-0), A-X (3-0), and A-X (2-0) bands toward $\zeta$~Oph as well.
For most of the sight lines in this study, observations of the UV and/or optical C$_2$ bands, together with the corresponding adopted $f$-values, appear to yield fairly reliable and mutually consistent column densities $N_{J}$ for the various rotational levels.
We have therefore adopted the weighted average $N_{J}$ listed in Table~\ref{tab:nc2} for sight lines with data for multiple bands.
Figure~\ref{fig:c2ex} compares the average C$_2$ rotational populations toward HD~24534 with those derived from the individual UV ({\it upper panel}) and optical ({\it lower panel}) bands.
In general, there is good agreement; the main exceptions are the low values found for $J$ = 8 and 10 from the F-X (0-0) band.
[In estimating the average C$_2$ in levels $J$ = 8 and $J$ = 10 toward HD~24534, HD~27778, and HD~206267, we have doubled the contributions from the F-X (0-0) band, in view of that apparent (and unexplained) deficiency, compared to the values obtained for those levels in the other C$_2$ bands.]
While we have C$_2$ data for $J$ = 0--8 for most of the sight lines in our sample, examination of the C$_2$ rotational population distributions toward HD~24534, HD~27778, and HD~207198 (for $J$ = 0--18) and $\zeta$ Oph (for $J$ = 0--24; Lambert et al. 1995) --- and extrapolation of those distributions to even higher $J$ --- suggests that a non-negligible fraction of the total C$_2$ is in rotational levels higher than $J$ = 8.
Toward HD~24534 and $\zeta$ Oph, for example, roughly 20\% and 40\% of the total C$_2$, respectively, appear to be in rotational levels above $J$ = 8.
In order to estimate the total C$_2$ column density in all rotational levels for each line of sight, we have multiplied the observed total for $J$ = 0--6 ($N_{0-6}$, which we have for all sight lines) by a factor $N_{0-20}$/$N_{0-6}$ derived from the theoretical rotational distribution that best fits the observed $N_{J}$ for that line of sight (see below, \S~\ref{sec-c2tex}).
Although the theoretical distributions are calculated only to $J$ = 20, $N_{0-20}$ is within 10\% of $N_{\rm tot}$ for all the sight lines in our sample; the factor $N_{0-20}$/$N_{0-6}$ ranges between about 1.1 and 2.3 for those sight lines.
The estimated total $N$(C$_2$) (listed in the last columns of Tables~\ref{tab:c2dx} and \ref{tab:nc2}) are generally somewhat smaller than the corresponding values given in Thorburn et al. (2003) --- due mostly to differences in the adopted A-X band $f$-values, but also to differences in the extrapolation to all $J$, in the estimation of saturation effects, and (in some cases) in the measured equivalent widths.
\subsubsection{Adopted Column Densities}
\label{sec-adopt}
Table~\ref{tab:ntot} summarizes the total column densities of H$_2$, $^{12}$CO, $^{13}$CO, and C$_2$ found for the ten lines of sight in this study, together with the resulting $^{12}$CO/H$_{2}$, $^{12}$CO/$^{13}$CO, and C$_2$/H$_2$ ratios.
The H$_2$ column densities are from Rachford et al. (2002), except for the value toward HD~147888 (Cartledge et al. 2004).
The column densities for CO and C$_2$ are from this paper, except for the $^{12}$CO and $^{13}$CO values toward HD~24534 (Sheffer et al. 2002a).
Because the STIS and {\it FUSE} data do not resolve the individual gas components seen in the higher resolution optical spectra, a component-by-component analysis is rendered uncertain (especially for H$_2$).
The discussions below therefore consider only the total column densities --- which should in any case represent the primarily molecular material in each line of sight.
\section{CO, C$_2$, and Physical Conditions in Translucent Sight lines}
\label{sec-phys}
The so-called ``translucent'' clouds are thought to be intermediate or transitional objects between diffuse, primarily atomic clouds and dense molecular clouds (e.g., van Dishoeck \& Black 1988, 1989).
It has been difficult, however, to isolate and characterize specific examples of ``true'' translucent clouds using optical/UV absorption lines, even with the more sensitive spectroscopic capabilities provided by {\it FUSE} and {\it HST}.
In a recent {\it FUSE} survey of interstellar H$_2$ in translucent sight lines, Rachford et al. (2002) obtained accurate measurements of the H$_{2}$ column densities in the $J=$ 0-1 rotational levels of the ground electronic and vibrational state for 23 lines of sight with $A_{\rm V}$ between 0.6 and 3.4 mag.
While all the sight lines exhibit strong H$_2$ absorption, with $N$(H$_2$) $>$ 3 $\times$ 10$^{20}$ cm$^{-2}$, none is characterized by the very high molecular fractions $f$(H$_2$) $\ga$ 0.8 expected for ``true'' translucent clouds.
Rachford et al. concluded that most of the observed sight lines contained multiple components consisting of mixtures of diffuse and denser gas, with ``little evidence for the presence of individual translucent clouds''.
That assessment has been confirmed toward HD~185418 and HD~192639 via detailed studies of those sight lines (Sonnentrucker et al. 2002, 2003).
Similarly, Snow, Rachford, \& Figoski (2002) did not find significantly enhanced depletion of iron (as would have been expected for translucent clouds), in essentially the same sample of sight lines (though it is possible that individual components with very severe depletions could be masked by other components with less severe depletions in the low-resolution {\it FUSE} spectra).
Nine of the ten stars considered here were included in the {\it FUSE} H$_2$ and \ion{Fe}{2} surveys.
\subsection{Behavior of CO}
\label{sec-co}
Because of the abundance of CO and its importance as a tracer and diagnostic of molecular gas, a number of studies have been undertaken to understand and predict its behavior in diffuse, translucent, and denser clouds.
Semi-analytic models (e.g., Federman \& Huntress 1989) identify the most significant channels for the formation and destruction of each molecular species and their dependences on the local physical conditions.
Comparisons of the predictions of such simplified chemical models with observed column densities, for an ensemble of sight lines, can be used to constrain (sometimes poorly known) rate coefficients and other parameters (e.g., the cosmic ray ionization rate) of the models.
More detailed numerical models (e.g., van Dishoeck \& Black 1988, 1989; Viala et al. 1988; Warin, Benayoun, \& Viala 1996; Le Petit et al. 2006) typically include more extensive reaction networks and radiative transfer effects, and attempt to determine the abundances of various species (including the individual rotational levels for H$_2$ and CO) as functions of depth within the model clouds.
All those detailed models, however, are for single, isolated plane-parallel clouds (often of uniform density).
In diffuse and translucent clouds with $A_{\rm V}$ $\la$ 3 mag, the primary route for the production of CO begins with the reaction C$^+$~+~OH~$\rightarrow$~CO$^+$~+~H.
The CO$^+$ then reacts with H$_2$ to form HCO$^+$, which subsequently forms CO via dissociative recombination.
In thicker, denser clouds, reactions involving O, CH, C$_2$, and other species may also contribute.
The destruction of CO is via line absorption into predissociated bound states; the relevant transitions are at far-UV wavelengths between 912 and 1100 \AA\ (e.g., van Dishoeck \& Black 1988; Viala et al. 1988; Warin et al. 1996).
At high enough column densities [$N$($^{12}$CO) $\ga$ 10$^{14}$ cm$^{-2}$], those transitions become saturated, self-shielding begins to reduce the photodissociation rate, and the column density of $^{12}$CO increases rapidly.
Recent observations of some of the far-UV CO bands suggest that the self-shielding by those bands may be more effective than previously thought (Sheffer, Federman, \& Andersson 2003).
In addition, some of the $^{12}$CO transitions coincide with strong absorption lines of \ion{H}{1} and/or H$_2$, which can dominate the shielding of $^{12}$CO at lower column densities.
The other isotopomers (e.g., $^{13}$CO, C$^{18}$O) are much less abundant, and so do not self shield (nor are they effectively shielded by either H$_2$ or $^{12}$CO).
At much higher column densities, reactions between CO and other species become the dominant destruction mechanism for CO.
In the sections below, the observed column densities of CO, H$_2$, and several other species will be compared with predictions from both the models of van Dishoeck \& Black (1988, 1989), Warin et al. (1996), and Le Petit et al. (2006).
The T, H, and I model sequences of van Dishoeck \& Black cover the range of extinctions and H$_2$ column densities relevant to most current observations of CO absorption ($A_{\rm V}$ $\la$ 2.5 mag).
The T models are characterized by the average interstellar radiation field, total hydrogen densities $n_{\rm H}$ = $n$(\ion{H}{1}) + 2$n$(H$_2$) of 300--1000 cm$^{-3}$, temperatures of 15--60 K, H$_2$ column densities of 0.5--4.0 $\times$10$^{21}$ cm$^{-2}$, and visual extinctions of 0.7--5.2 mag.
Several of the models allow for some variation in the temperature and density with depth.
The H models have half the average interstellar field, $n_{\rm H}$ = 500 cm$^{-3}$, $T$ = 40 K, $N$(H$_2$) = 0.3--3.0 $\times$ 10$^{21}$ cm$^{-2}$, and $A_{\rm V}$ = 0.4--3.8 mag.
The I models have ten times the average interstellar field, $n_{\rm H}$ = 2000 cm$^{-3}$, $T$ = 30 K, $N$(H$_2$) = 0.5--10.0 $\times$ 10$^{21}$ cm$^{-2}$, and $A_{\rm V}$ = 0.8--12.8 mag.
All three series assume depletions of carbon, nitrogen, and oxygen of 0.1--0.4, 0.5, and 0.6, respectively (with respect to the solar or ``cosmic'' abundances adopted at that time); all three use the average Galactic extinction curve [though van Dishoeck \& Black (1989) also considered the effects of curves that are shallower (HD 147889) and steeper (HD 204827) in the UV].
Warin et al. (1996) computed models for diffuse clouds [$A_{\rm V}$ = 1.09 mag, $N$(H) = 2 $\times$ 10$^{21}$ cm$^{-2}$, $n_{\rm H}$ = 500 cm$^{-3}$], translucent clouds [$A_{\rm V}$ = 4.34 mag, $N$(H) = 8.1 $\times$ 10$^{21}$ cm$^{-2}$, $n_{\rm H}$ = 1000 cm$^{-3}$], and dark clouds [$A_{\rm V}$ = 10.86 mag, $N$(H) = 20 $\times$ 10$^{21}$ cm$^{-2}$, $n_{\rm H}$ = 10$^4$ cm$^{-3}$] --- both at constant $T$ and in thermal equilibrium.
All the Warin et al. models are for uniform density slabs; all assume depletions of 0.1 for both carbon and oxygen.
Unlike the van Dishoeck \& Black models, the Warin et al. models allow for differences in rotational excitation temperature for different $J$ --- resulting from the coupling between level populations, self-shielding, and photodissociation rates.
The total unshielded photodissociation rate adopted by Warin et al. (1996), based on more recent CO spectroscopic data, is of order 20\% smaller than that used by van Dishoeck \& Black (1988).
Le Petit et al. (2006) give an overview of the current Meudon photon-dominated region (PDR) code (e.g., Le Bourlot et al. 1993), which computes the radiative transfer, thermal balance, and chemistry (including the excitation of H$_2$ and CO) for a plane-parallel cloud in steady state.
In principle, both temperature and density can vary with depth in the cloud.
While Le Petit et al. do not give extensive numerical results, they do provide a coarsely sampled table of column densities of H$_2$, CO, and several other species (for $n_{\rm H}$ = 100 cm$^{-3}$, the standard interstellar radiation field, and total $A_{\rm V}$ ranging from 0.2 to 7.0) and several figures showing the column densities of CH, CN, and CO, as functions of $n_{\rm H}$/$\chi$, for a set of diffuse cloud models [$A_{\rm V}$ = 1, $n_{\rm H}$ ranging from 20 to 2000 cm$^{-3}$, and $\chi$ (the scaling factor for the radiation field) from 0.1 to 10].
All the models use the current observed gas-phase interstellar abundances of carbon, nitrogen, and oxygen.
The various models of cloud chemistry suggest that the column density of CO generally will increase with increasing local density, H$_2$ column density, and optical depth.
For diffuse clouds, Federman et al. (1980) concluded that the local density of CO would be proportional to [$n$(H$_2$)]$^2$ (from formation) and to exp[$\tau$(OH)+$\tau$(CO)] (from destruction; the optical depths $\tau$ are for the wavelength ranges appropriate for photodissociation of OH and CO), but that the density dependence would be only linear with $n$(H$_2$) once reactions with other species dominate the destruction of CO.
Isobaric models with mean densities of 150 and 2500 cm$^{-3}$ appeared able to account for most of the range in $N$(CO) observed at any given $N$(H$_2$).
(The role of line absorption and self-shielding in the dissociation of CO was not yet appreciated, however.)
For clouds with $A_{\rm V}$ $\la$ 2.5, the more detailed models of van Dishoeck \& Black (1988), which include self-shielding, predict that:
(1) increasing the gas-phase carbon abundance by a factor of four [for constant density and $N$(H$_2$)] increases the total CO column density by a factor of 1.5--3.0 (models T1--T4);
(2) decreasing the radiation field by a factor of 2 [for constant depletion, density, and $N$(H$_2$)] increases $N$(CO) by a factor of 3--4 (models H2--H4 vs. models T1--T3); and
(3) increasing $N$(H$_2$) by a factor of 2 (for constant depletion and radiation field) increases $N$(CO) by more than an order of magnitude (models T1--T4, H2--H5, and I1--I4).
The Warin et al. (1996) translucent cloud model has twice the density, four times the H$_2$ column density, and roughly 3000 times the CO column density of the diffuse cloud model.
The earlier models suggest that CO becomes the dominant carbon-containing species for clouds with $A_{\rm V}$ $\ga$ 3, with the transition between C$^+$ (and C$^0$) and CO typically occurring at optical depths $\tau_{\rm V}$ between 0.5 and 1.0 (van Dishoeck \& Black 1988; Warin et al. 1996) --- where the hydrogen is already almost completely in molecular form.
The very high predicted molecular fractions $f$(H$_2$) $\ga$ 0.9 --- even for the total (integrated) column densities in the diffuse cloud models --- have not yet been observed, however (Rachford et al 2002).
In the Meudon PDR models with $n_{\rm H}$ =100 cm$^{-3}$ and $\chi$ = 1, C$^+$ remains the dominant form of carbon even for $A_{\rm V}$ = 7; in the diffuse cloud models, $N$(CO) does not show a very strong dependence on density, for $n_{\rm H}$/$\chi$ between 10 and 1000 cm$^{-3}$.
Additional aspects of the behavior of CO (e.g., rotational excitation, isotopic abundances), as predicted by the various models, will be discussed further below (\S~\ref{sec-cotex} and \S~\ref{sec-1213}, respectively).
\subsection{Behavior of C$_2$}
\label{sec-c2}
While C$_2$ is not as abundant as CO, it does play a role in the networks of chemical reactions occurring in diffuse and translucent clouds, and analyses of its absorption lines can provide constraints on the density, temperature, and radiation field in those clouds.
Federman \& Huntress (1989) have described the key reactions controlling the abundance of C$_2$; van Dishoeck \& Black (1982) have provided a detailed discussion of its excitation in diffuse and translucent clouds.
A number of papers (e.g., van Dishoeck \& de Zeeuw 1984; Federman \& Lambert 1988; van Dishoeck \& Black 1989; Federman et al. 1994a) have discussed observations of C$_2$ absorption in the context of models for its abundance and/or excitation.
In diffuse and translucent clouds, the formation of C$_2$ begins with the reaction C$^+$~+~CH~$\rightarrow$~C$_2^+$~+~H; subsequent hydrogen abstraction reactions and dissociative recombinations yield C$_2$ via several channels.
Destruction of C$_2$ can occur by photodissociation (more important at relatively low optical depths) or by reactions with atomic oxygen (yielding CO) or nitrogen (yielding CN), which become dominant at higher $\tau$.
The detailed models T1--T6 of van Dishoeck \& Black (1989), with total $A_{\rm V}$ = 0.7--5.2 mag, predict a roughly constant C$_2$/H$_2$ ratio $\sim$ 3.3--5.7 $\times$ 10$^{-8}$ over that range.
The C$_2$ abundance is somewhat enhanced if the oxygen depletion is more severe than the value of 0.5 usually assumed in the models, is essentially unaffected by the assumed nitrogen depletion, and is somewhat affected by the shape of the far-UV extinction.
As a ``second generation'' molecule dependent on the prior existence of H$_2$ and CH, C$_2$ is likely to be more concentrated in somewhat colder, denser regions.
The models of van Dishoeck et al. (1991) suggest that the average density traced by C$_2$ may be either higher or lower than that traced by CO, depending on the structure of the cloud and how it is illuminated by the ambient radiation field.
The rotational excitation of C$_2$ will be discussed below (\S~\ref{sec-c2tex}).
\subsection{Column Density Comparisons}
\label{sec-coldens}
In order to compare the column densities of $^{12}$CO and $^{13}$CO obtained in this paper with those found for other Galactic sight lines, we have collected CO column densities derived from UV absorption-line data from the literature.
Table~\ref{tab:coldens} (Appendix~\ref{sec-app}) lists total column densities for H$_2$, CH, C$_2$, CN, and CO for 74 sight lines for which CO has been measured (53 detections, 21 upper limits) --- a reasonably large sample which includes recently improved $N$(CO) values for both diffuse and more heavily reddened sight lines.
For a number of those sight lines, column densities also have been recently reported for HD (Lacour et al. 2004) and for C$_3$ (Maier et al. 2001; Roueff et al. 2002; Galazutdinov et al. 2002; Oka et al. 2003; \'{A}d\'{a}mkovics, Blake, \& McCall 2003).
Table~\ref{tab:dq} (Appendix) lists various derived quantities (e.g., excitation temperatures, column density ratios) for the sight lines with data for CO, H$_2$, CN, and CH.
There are now 19 sight lines with relatively well determined $N$(CO) $\ga$ 10$^{15}$ cm$^{-2}$ derived from UV absorption lines [plus a few others with values from mm-wave CO absorption (Liszt \& Lucas 1998, 2000)], compared to one such sight line ($\zeta$ Oph) in the early survey of Federman et al. (1980).
This enlarged higher column density sample allows more meaningful comparisons with the predictions of theoretical models of translucent clouds.
Comparisons among these molecular species (and of various ratios of these species) may provide some clues as to their behavior in different environments and to the physical conditions characterizing the clouds in which they reside (e.g., Federman et al. 1994a).
While there are likely to be some differences in the spatial distribution of these species, they all trace the main components in each sight line, where the bulk of the molecular material is located.
\subsubsection{$N$(CO) and $N$(C$_2$) vs. $N$(H$_2$)}
\label{sec-coh2}
Figure~\ref{fig:comod} shows plots of the column densities of $^{12}$CO and $^{13}$CO versus $N$(H$_2$).
In the plots, the different symbols refer to the sources of the CO data ({\it Copernicus}, {\it IUE}, {\it FUSE}, or {\it HST}); the filled points represent sight lines for which $^{13}$CO has been detected.
The dotted lines in both figures indicate the relationships predicted by the models of van Dishoeck \& Black (1988); series H, T, and I (increasing radiation field; all with carbon depletion $\delta_{\rm C}$ = 0.4) are left-to-right in each case.
The letters ``W'' indicate the values predicted by the diffuse, translucent, and dense cloud models of Warin et al. (1996).
The dashed lines show the predictions of the Meudon PDR models with $n_{\rm H}$ = 100 cm$^{-3}$, $\chi$ = 1, and $A_{\rm V}$ ranging from 0.2 to 7 (Le Petit et al. 2006).
The observed column densities confirm the steep increase of $N$(CO) with $N$(H$_2$), for $N$(H$_2$) $\ga$ 3 $\times$ 10$^{20}$ cm$^{-2}$, as predicted by the various models and as noted by Federman et al. (1980).
The much more extensive data for $N$($^{12}$CO) $\ga$ 10$^{15}$ cm$^{-2}$, however, appear to indicate a somewhat steeper increase for the higher column density sight lines than that predicted by Federman et al. --- presumably due to the effects of self-shielding.
The models of van Dishoeck \& Black (1988) appear to reproduce the trends in the observed data reasonably well, but predict too little $^{12}$CO by a factor of order 2--3 [for a given $N$(H$_2$)].
For the sight lines with detected $^{13}$CO, the observed $N$($^{13}$CO) appear to agree somewhat better with the model predictions than do the observed $N$($^{12}$CO).
The models of Warin et al. (1996) and Le Petit et al. (2006) predict much smaller CO column densities than the observed values.
While the density in these models from Le Petit et al. is relatively low, the Meudon diffuse cloud models ($A_{\rm V}$ = 1) do not show an increase in $N$(CO) until $n_{\rm H}$ is greater than about 1500 cm$^{-3}$ (for $\chi$ = 1).
More extensive exploration of the parameter space available in the Meudon code (though beyond the scope of this paper) might prove instructive.
As suggested by the three series of models (H, T, I), some of the scatter in the relationship between $N$(CO) and $N$(H$_2$) may be due to differences in the strength of the ambient radiation field.
For sight lines with column density data for the lowest few rotational levels of H$_2$, the relative amount in $J$ = 4 (which is assumed to be populated primarily via photon pumping) may be used to estimate the strength of the radiation field (Jura 1975; Lee et al. 2002; Welty et al. 2006).
The four open squares just to the left of the high radiation field (I) model sequence correspond to four sight lines in Cep OB3 (Pan et al. 2005).
For three of those sight lines, the H$_2$ rotational populations determined by Pan et al. do suggest fairly strong ambient fields.
The molecular material seen toward HD~37903 [open triangle near $N$(H$_2$) = 8 $\times$ 10$^{20}$ cm$^{-2}$ and $N$(CO) = 10$^{14}$ cm$^{-2}$)] also is likely subject to an enhanced field (Lee et al. 2002).
However, there are other sight lines with similarly high inferred fields whose CO column densities lie closer to the model sequences characterized by average or low fields.
In addition, Lacour et al. (2005b) obtained progressively higher $b$-values for the higher rotational levels of H$_2$ in several moderately reddened lines of sight (including HD~73882, HD~192639, and HD~206267) and argued that the column densities of those higher $J$ levels could be significantly overestimated under the usual assumption of constant $b$ for all $J$.
If $N_4$(H$_2$) is too high, then the radiation field will be similarly overestimated.
While differences in the radiation field probably do affect the abundances of CO observed in different sight lines, other factors must also play a role.
Increases in the gas-phase carbon and oxygen abundances and/or in the self-shielding of $^{12}$CO (Sheffer et al. 2003) in the models could yield somewhat better agreement between the observed $^{12}$CO column densities and those predicted by the models.
While interstellar \ion{C}{2} column density data are still rather limited, there does appear to be a fairly well-defined average value C/H = 161$\pm$17 ppm for both diffuse and translucent sight lines (Sofia et al. 2004).
For a total (gas + dust) carbon abundance of 245 ppm (Lodders 2003), the carbon depletion $\delta_{\rm C}$ is thus about 0.66 [or a logarithmic depletion D(C) $\sim$ $-$0.18 dex].
Only the sight line toward HD~27778 shows evidence for a gas-phase carbon abundance significantly below that average value [$\le$ 108 ppm (3-$\sigma$)]\footnotemark; the carbon abundance toward X~Per (155$\pm$45 ppm), with higher $A_{\rm V}$ and $f$(H$_2$) and similar molecular abundances, is quite consistent with the average (Sofia, Fitzpatrick, \& Meyer 1998).
\footnotetext{While the CO column density toward HD~27778 estimated by Federman et al. (1994a) seemed suggestive of an unusually high CO abundance in that line of sight (Sofia et al. 2004), our value for $N$(CO) is a factor of 2 smaller.}
The contribution of CO to the overall gas-phase carbon abundance in those two sight lines is only $\sim$ 6 ppm.
The Warin et al. (1996) models assumed a total carbon abundance of 360 ppm and $\delta_{\rm C}$ = 0.1.
The resulting gas-phase carbon abundance ($\sim$ 36 ppm) is a factor of $\sim$ 4.5 lower than the observed average interstellar value.
Increasing the gas-phase carbon abundance would move the predicted $N$(CO) closer to the observed values, but would generally not be enough to yield good agreement.
The van Dishoeck \& Black (1988, 1989) models assumed a total carbon abundance of 467 ppm and $\delta_{\rm C}$ = 0.1--0.4.
For $\delta_{\rm C}$ = 0.4, the resulting gas-phase carbon abundance ($\sim$ 190 ppm) is slightly higher than the average interstellar value, so that other adjustments (e.g., increasing the $^{12}$CO self-shielding) would be required to obtain better agreement with the observed $N$(CO).
Several other suggested modifications to the (relatively) simple early models may be able to increase the CO abundances predicted for diffuse and translucent clouds.
(1) Both van Dishoeck \& Black (1989) and Warin et al. (1996) noted that one way would be to reduce the (relatively poorly known) intensity of the radiation field below 1200 \AA.
The field would need to be weaker by an order of magnitude (or more), however, for the Warin et al. diffuse cloud model to reproduce the observed abundances of $^{12}$CO and $^{13}$CO (even with an increased carbon abundance).
(2) Liszt \& Lucas (1994) concluded that the low predicted values of $N$(CO) toward $\zeta$ Oph should be ascribed to insufficient production of CO (rather than excessive photodissociation), in view of the similar underprediction of its immediate precursor HCO$^+$.
An increase in the rate of the initiating reaction C$^+$~+~OH~$\rightarrow$~CO$^+$~+~H (e.g., Dubernet, Gargaud, \& McCarroll 1992; Federman et al. 2003) would yield corresponding increases in both HCO$^+$ and CO.
(3) In view of observational evidence for complex, inhomogeneous structure in molecular clouds, Spaans (1996) explored the consequences of having higher density clumps embedded in a lower density interclump medium.
On the one hand, dissociating radiation may penetrate further into the interior of such a clumpy structure than is the case for the previously modeled homogeneous slabs.
Both theoretical considerations (\S~\ref{sec-co}) and empirical evidence (\S~\ref{sec-var}), however, suggest that CO will be enhanced at higher densities.
Spaans found that CO (which is located mostly in the denser clumps) could actually be enhanced by factors of 5--6 in the clumpy models, compared to the abundances found for the corresponding homogeneous models.
(4) Because the photodissociation of CO is dominated by line absorption, it can depend on details of the velocity distribution in the molecular gas.
Consideration of the effects of a turbulent velocity field has suggested that resulting changes to the CO absorption-line profiles could significantly increase the self-shielding of CO --- thus increasing the CO abundance in moderately dense molecular clouds (e.g., R\"{o}llig, Hegmann, \& Kegel 2002).
Much smaller effects are seen, however, when the full CO rotational structure is incorporated (instead of the single-line approximation used in the initial calculations).
Finally, however, if many sight lines contain a significant amount of H$_2$ without much associated CO (e.g., in more diffuse molecular gas), then the agreement with the models would not be as good as Figure~\ref{fig:comod} would otherwise seem to suggest.
Detailed studies of individual lines of sight (to separate the contributions from different molecular components) and comparisons with more sophisticated models (incorporating these additional physical effects) may yield better agreement between observed and predicted CO abundances.
Figure~\ref{fig:c2mod} shows the relationship between the column densities of C$_2$ and H$_2$.
The solid lines are linear least-squares fits (weighted and unweighted) to the points representing detections of both quantities, taking into account that both quantities have associated uncertainties.\footnotemark\
\footnotetext{A slightly modified version of the subroutine regrwt.f, obtained from the statistical software archive maintained at Penn State (http://www.astro.psu.edu/statcodes), was used to perform the fits; see Welty \& Hobbs (2001).}
Table~\ref{tab:corr} lists the linear correlation coefficients ($r$), the slopes of the best-fit lines, and the residual scatter about the best-fit lines found for this comparison (and for others discussed below).
The dotted line, which shows the roughly linear relationship predicted by models T1--T6 of van Dishoeck \& Black (1988, 1989), falls slightly below many of the observed data points.
Much of the systematic difference between the models and the observed data may be ascribed to the depletion of carbon ($\delta_{\rm C}$ = 0.1--0.15) assumed in the models --- as the corresponding gas-phase carbon abundances are factors of 2--3 lower than the values observed for nearly all diffuse and translucent clouds.
Apart from that systematic offset, the observed data seem to follow a somewhat steeper relationship (slope $\sim$ 1.8) than the roughly linear predicted trend (though the ranges in both column densities are rather limited); the much steeper slope found for the unweighted fit is due largely to the point for 40 Per.
While Federman \& Huntress (1989) suggested progressively more severe depletion of carbon (to a factor of 15 by $A_{\rm V}$ = 4 mag.) was required to explain the C$_2$ abundances in translucent sight lines, none of the sight lines in Fig.~\ref{fig:c2mod} has $A_{\rm V}$ higher than 2.7 mag.
\subsubsection{$N$(CO) vs. Column Densities of CN, CH, C$_2$, C$_3$, HD, and \ion{K}{1}}
\label{sec-molec}
Figures~\ref{fig:cocnch} and \ref{fig:coc2k1} show plots of the column density of $^{12}$CO versus the column densities of CN, CH, C$_2$, and \ion{K}{1}; similar comparisons were also made with the column densities of C$_3$ and HD.
Again, the different symbols denote the sources of the CO data, with the filled points representing sight lines for which $^{13}$CO has been detected, and the solid lines are the linear least-squares fits to the data.
For the full sample, there are clear, significant correlations ($r$ from 0.733 to 0.905) with CN, CH, HD, and \ion{K}{1} (Table~\ref{tab:corr}).
The smaller correlation coefficients for C$_2$ and C$_3$ appear to be due to several sight lines (40 Per, HD~200775) with very high CO/C$_2$ ratios and to the small range in column density for the few sight lines with detections of C$_3$, respectively.
The strongest correlations are with CN ($r$ = 0.900; slope $\sim$ 1.50) and with HD ($r$ = 0.905; slope $\sim$ 1.75); the relationship with CN shows significantly smaller scatter, however.
The slopes of the relationships with CH, C$_2$, and \ion{K}{1} are much steeper (between about 3.1 and 4.0), though consideration of the upper limits for some sight lines suggests that the relationships may be somewhat shallower at lower column densities.
In view of the scatter in those relationships, it would be difficult to accurately predict $N$(CO) from the column densities of CH, C$_2$, or \ion{K}{1} for any individual sight line.
The dotted lines in Figures~\ref{fig:cocnch} and \ref{fig:coc2k1} give the relationships between the column density of $^{12}$CO and those of CN, CH, and C$_2$ predicted by models T1--T6 of van Dishoeck \& Black (1988, 1989).
In all three cases, the average slopes of the predicted relationships are very similar to those determined from the fits to the observed data, but the predicted $N$(CO) are lower than the observed (fitted) values by factors of 2--4.
To first order, the low values assumed for the carbon abundance in the models presumably should not significantly affect the predicted ratios of CO to CN or CH.
The gas-phase abundances assumed in the van Dishoeck \& Black (1988, 1989) models for oxygen and nitrogen are closer to the current observed values (Meyer, Cardelli, \& Sofia 1997; Cartledge et al. 2004).
The dashed lines in Figure~\ref{fig:cocnch} show the predictions of the Meudon diffuse cloud models (Le Petit et al. 2006), for $\chi$ = 1 and $n_{\rm H}$ ranging from 20 to 2000 cm$^{-3}$.
The predicted $N$(CO) would be consistent with the observed trends for densities of about 200--500 cm$^{-3}$ --- though there are sight lines with $A_{\rm V}$ $\la$ 1 and $N$(CO) much higher than those predicted values.
As was found from fitting the observed CO absorption-line profiles, these column density comparisons suggest that the distribution of CO is most similar to that of CN.
The correlation coefficient with CN is larger than those for CH, C$_2$, C$_3$, and \ion{K}{1}, and the slope of the relationship with CN is closer to unity.
Weak components in \ion{K}{1} and CH, for example, are probably much weaker in CO, though they may still be detectable in the strongest CO bands (e.g., Sheffer et al. 2003; see also the profiles for HD~206267 in Fig.~\ref{fig:stis2}).
While the abundances of both CO and CN appear to increase strongly with overall column density, however, the reasons for the increase may be different for the two species.
Because the production of CN depends on the prior existence of C$_2$, CH, and/or NH, the ``third-generation'' molecule CN is expected to trace colder, denser, more shielded regions where those precursor molecules have already become abundant (Federman, Danks, \& Lambert 1984; Federman et al. 1994).
On the other hand, the strong increase in $N$($^{12}$CO) is likely to be due both to increased formation at higher densities and to reduced photodissociation resulting from the onset of self-shielding, once sufficient $^{12}$CO is present [i.e., $N$($^{12}$CO) $\ga$ 10$^{14}$ cm$^{-2}$].
\subsection{Variation of the CO/H$_{2}$ Ratio}
\label{sec-var}
Both theoretical modeling and observations of the CO abundance indicate that the CO/H$_{2}$ column density ratio can vary by three orders of magnitude, from $\la$ 10$^{-7}$ in diffuse clouds to about 10$^{-4}$ in dense molecular clouds (van Dishoeck \& Black 1988; van Dishoeck \& Black 1990; Warin et al. 1996); the transition between those values occurs in the translucent regime.
The analogous ratio of $N$(H$_2$) to CO emission strength [the so-called ``conversion factor'' X$_{\rm CO}$ = $N$(H$_2$)/I($^{12}$CO), of order 2 $\times$ 10$^{20}$ cm$^{-2}$(K~km~s$^{-1}$)$^{-1}$ in our Galaxy] is often used to derive the H$_2$ column density in dense molecular clouds, since H$_2$ is often difficult to measure directly in such regions (Lacy et al. 1994).
That conversion factor reflects the relationship between the width of the (saturated) $^{12}$CO $J$ = 1--0 emission line and the mass of a gravitationally bound dense cloud.
A unique conversion factor cannot be defined, however, in diffuse and translucent clouds (e.g., Magnani \& Onello 1995) --- which are not gravitationally bound and where $N$(CO) is not high enough for the emission lines to be saturated.
Figure~\ref{fig:coh2mod} shows the same data as in Figure~\ref{fig:comod}, but plotted as $N$(CO)/$N$(H$_2$) versus $N$(H$_2$).
Again, the dotted lines show the relationships predicted by models H, T, and I of van Dishoeck \& Black (1988), the letters ``W'' show the values predicted by the models of Warin et al. (1996), and the dashed line (for $^{12}$CO) shows the predictions of the Meudon code (Le Petit et al. 2006).
In our sample of ten sight lines, the $^{12}$CO/H$_{2}$ ratio varies by a factor of about 100 --- from 2.8 $\times$ 10$^{-7}$ toward HD~192639 to 2.8 $\times$ 10$^{-5}$ toward HD~73882 (Table~\ref{tab:ntot}); values as low as 2.5 $\times$ 10$^{-8}$ are seen in the larger sample.
For our sight lines, the variations in the overall column densities of both H$_2$ and total hydrogen (\ion{H}{1} + 2H$_2$) are much smaller; for example, seven of the ten have $N$(H$_2$) in the range 4.9--7.2 $\times$ 10$^{20}$ cm$^{-2}$ (Table~\ref{tab:ntot}).
The differences in the CO/H$_2$ ratios in our sight lines thus are due primarily to variations in the CO column densities, which in turn depend on the physical conditions in the molecular gas in those particular sight lines.
Since the total hydrogen content is similar for those sight lines, variations in the shielding of the CO bands by H and H$_2$, lines cannot explain the differences in the CO column density.
It seems most likely that the range in $N$(CO) observed for our ten sight lines is due to differences in the CO photodissociation rate (ambient radiation field, self-shielding) and/or in the CO formation rate (local hydrogen density).
In all ten of our sight lines, CO contains only a minor fraction of the total carbon abundance.
For the current solar carbon abundance C/H $\sim$ 2.5 $\times$ 10$^{-4}$ (e.g., Asplund, Grevesse, \& Sauval 2005) and typical interstellar depletions ($\delta_{\rm C}$ $\sim$ 0.66), the total gas-phase ratio C/H$_2$ would be about 3.2 $\times$ 10$^{-4}$ (if hydrogen were fully molecular).
The highest values of the CO/H$_2$ ratio in our sample --- $\sim$ 2--3 $\times$ 10$^{-5}$ toward HD 27778 and HD 73882 --- are more than an order of magnitude smaller.
Most of the gas-phase carbon is still in the form of C$^+$ (Sofia et al. 2004; see also Liszt \& Lucas 1998).
In order to explore observationally the relationship between the CO/H$_2$ ratios and local physical conditions (e.g., $T$, $n_{\rm H}$, radiation field), empirical diagnostics of those parameters are needed.
The simplified chemical model of Federman \& Huntress (1989) implies, for example, that the CN/CH ratio in diffuse molecular gas should be roughly proportional to the square of the local hydrogen density ($n_{H}^2$) (Cardelli, Federman, \& Smith 1991) --- suggesting that the CN/CH column density ratio may be used as a density indicator.
The more detailed models of van Dishoeck \& Black (1986a, 1989), however, predict a somewhat weaker increase of $N$(CN)/$N$(CH) with $n_{\rm H}$.
While the densities estimated by Pan et al. (2005) for individual components toward stars in Cep OB2 suggest that the CN/CH ratio may be roughly proportional to $n^{1/2}_{\rm H}$ (with much scatter), the densities inferred below from C$_2$ (\S~\ref{sec-c2tex}) appear to indicate a somewhat steeper relationship.
The CN/CH ratio thus does appear to be correlated with density, but other factors (e.g., $A_{\rm V}$ and the steepness of the extinction curve in the far-UV) may contribute to the scatter (see also Cardelli 1988; Welty \& Fowler 1992).
Some CH may also be formed together with CH$^+$ in lower density gas (Federman, Welty, \& Cardelli 1997a; Zsarg\'{o} \& Federman 2003), but such contributions generally appear to be fairly minor for sight lines in which CN has been detected.
Figure~\ref{fig:coh2} shows the observed ratios $N$($^{12}$CO)/$N$(H$_2$) and $N$($^{13}$CO)/$N$(H$_2$) versus $N$(CN)/$N$(CH) for sight lines with data for those species (Appendix~\ref{sec-app}; Table~\ref{tab:dq}).
There are clear correlations in both cases ($r$ = 0.815 and 0.868); the solid lines show the linear least-squares fits to the data (with slopes of order 1.55--1.75; Table~\ref{tab:corr}).
The dotted line for $^{12}$CO/H$_2$, showing the relationship predicted by models T1--T6 of van Dishoeck \& Black (1989) [for $\delta_{\rm C}$ = 0.15, $\delta_{\rm N}$ = 0.6, $\delta_{\rm O}$ = 0.5, and grain model 2], is very close to the best fit to the observed data.
The letter ``M'', corresponding to the values predicted by the Meudon code for $A_{\rm V}$ = 1, $\chi$ = 1, and $n_{\rm H}$ = 100 cm$^{-3}$, also lies very close to the observed trend.
These correlations are perhaps not surprising --- in view of the good correlation between $N$($^{12}$CO) and $N$(CN) and the known roughly linear relationship between $N$(H$_2$) and $N$(CH) (e.g., Danks, Federman, \& Lambert 1984).
Nonetheless, the correlations do suggest that differences in the local density $n_{\rm H}$ [as indicated by $N$(CN)/$N$(CH)] may play a significant role in explaining the variations in the CO/H$_2$ ratios observed in diffuse and translucent molecular gas.
As the dispersions relative to the best fits to the data ($\sim$ 0.2 dex) are larger than the typical measurement uncertainties, it is likely that other factors (e.g., radiation field, UV extinction, cloud structure; see \S~\ref{sec-coh2}) also affect the CO abundances, however.
In contrast, the ratio $N$(C$_2$)/$N$(H$_2$) does not show any significant dependence on $N$(CN)/$N$(CH).
The column density ratios observed for our ten sight lines are quite consistent with the trends observed for the larger sample.
The largest ratios of both CO/H$_2$ and CN/CH are observed toward HD~73882 and HD~27778 (Table~\ref{tab:dq}); the smallest ratios are found toward HD~185418 and HD~192639 (where CN is weak).
For the three targets in the Cep OB2 association, the CN/CH ratio toward HD~206267 is higher than those toward HD~207198 and HD~210839 by factors of 1.4--2.1, while the CO/H$_2$ ratio is about 2.6--3.0 times larger.
Simple chemical modeling of the CH and CN absorption by Pan et al. (2005) yields densities of 525--1150 cm$^{-3}$ toward HD~206267, 225--425 cm$^{-3}$ toward HD~210839, and 150 cm$^{-3}$ toward HD~207198 for the main components seen in CN.
Analysis of the C$_2$ rotational excitation (see below) implies slightly higher densities toward HD~206267 than toward HD~207198.
For those three Cep OB2 sight lines, the CN/CH and CO/H$_2$ ratios thus both appear to increase with density.
\subsection{Rotational Excitation of CO and C$_2$}
\label{sec-tex}
Because the populations of the ground and excited rotational levels of molecules such as H$_2$, CO, and C$_2$ can depend on both collisional and radiative processes, comparison of the observed rotational excitation of those molecules with predictions from theoretical models can provide information on the density, temperature, and radiation environment characterizing the molecular gas.
In general, measurements of both lower and higher rotational states are needed to disentangle the collisional and radiative effects.
The relative rotational level populations are often expressed in terms of rotational excitation temperatures
$T_{ij}$ = [($E_j$/$k$)~$-$~($E_i$/$k$)]~/~[ln($N_i$/$g_i$)~-~ln($N_j$/$g_j$)] (usually for adjacent levels $i$ $<$ $j$), where $E_i$, $N_i$, and $g_i$ are the energy, column density, and statistical weight of level $i$.
Because H$_2$, CO, and C$_2$ each respond somewhat differently to the local physical conditions and may well trace somewhat different volumes of gas, determinations of their respective rotational excitations can provide both complementary constraints on those conditions and some information on the structure of the clouds.
Tables~\ref{tab:texco} and \ref{tab:texc2} list various excitation temperatures derived from the rotational level populations of CO, H$_2$, and C$_2$, which will be used to constrain the physical conditions in the molecular gas.
\subsubsection{CO Excitation}
\label{sec-cotex}
Because of the abundance and ubiquity of CO in interstellar space, observations of its emission lines have often been used to estimate the physical conditions in dense molecular gas ($n_{\rm H}$ $\ga$ 10$^{4}$ cm$^{-3}$, $T$ $\sim$ 10 K).
Although CO is a fairly good density and temperature diagnostic in such dense clouds, where collisional processes dominate, the use of CO as a diagnostic of the physical conditions is more problematic in diffuse and translucent clouds still permeated by UV photons (Wannier et al. 1997).
While the detailed models of van Dishoeck \& Black (1988) incorporated both chemical reactions and various radiative and collisional processes in order to predict both the total CO abundance and its rotational excitation, the excitation temperature was assumed to be the same for all rotational levels.
Warin et al. (1996) removed that constraint --- and were thus able to study the variations in the excitation of the various CO rotational levels for ''typical'' diffuse, translucent, and dense clouds.
For diffuse clouds, the Warin et al. (1996) models predict that the photodissociation rate remains both large throughout the cloud and globally independent of the $J$ level.
While CO in the excited rotational levels is easily photodissociated by UV photons, the relatively low gas densities do not allow efficient collisional population of those excited levels (achieved primarily via collisions with H$_2$ but also with H).
The rotational population distribution of $^{12}$CO (and its isotopes) is therefore sub-thermal --- i.e., the excitation temperatures $T_{ij}$ of the various excited rotational levels ($\la$ 10 K for $J$ = 1--4) are well below the assumed temperature of the gas (50 K) --- even though the energies of those levels are less than or comparable to the thermal energy ($kT$) at that temperature (Table~\ref{tab:conj}).
The models predict that the relative population of the $J=$ 1 level is $\sim$ 10--50\% higher than that of the $J=$ 0 level and that the populations of the higher levels decrease by nearly a factor 10 for each level above $J=$ 1.
For the typical column densities measured in diffuse clouds (and the sensitivity of the instrumentation available to this point), the $J=$ 0, 1, and (perhaps) 2 rotational levels of $^{12}$CO should be detectable.
Corresponding lines from the much less abundant isotopomers are expected to be weak in diffuse clouds.
For translucent clouds, the Warin et al. (1996) models predict a strong dependence of the photodissociation rate on the rotational level, as self-shielding becomes effective for the lowest-lying levels of $^{12}$CO.
Because of this differential photodissociation effect, the excitation temperatures are predicted to vary both with rotational level and with depth in the cloud.
Thermalization is still not achieved, however, and all excitation temperatures remain sub-thermal throughout the cloud.
For example, the isothermal model predicts that the relative populations in $J$ = 0/1/2/3 at optical depth (due to dust) $\tau_{\rm V}$ = 0.5 are $\sim$ 1.0/2.0/0.5/0.07, while the relative populations at the cloud core ($\tau_{\rm V}$ = 2.0) are $\sim$ 1.0/2.3/2.3/1.5.
The corresponding predicted excitation temperatures $T_{ij}$ for $J$ = 1--3 are 6--12 K at $\tau_{\rm V}$ = 0.5 and 19--21 K at the cloud core (compared to the assumed constant temperature $T$ = 25 K).
The $J$ = 0 and 1 levels of the isotopomers $^{13}$CO and C$^{18}$O also should be detectable, with the $^{13}$CO excitation temperature $T_{01}$ $\sim$ 9 K throughout the cloud.
Figure~\ref{fig:corot} shows the $^{12}$CO normalized excitation diagrams (ln[$N_{J}$/((2$J$+1)~$N_0$)] vs. $E_{J}$/k) for the eight sight lines whose rotational level column densities were derived from our fits to high-resolution UV spectra obtained with GHRS or STIS (Table~\ref{tab:conj}).
The dotted lines show the extrapolated straight-line fit to the points for $J$ = 0 and 1, the slope of which is the negative of the inverse of the excitation temperature $T_{01}$.
Table~\ref{tab:texco} lists the corresponding $^{12}$CO (and $^{13}$CO) excitation temperatures for those eight sight lines, together with the CO excitation temperatures reported in the literature for eight other sight lines and the average values derived from the CO rotational populations in the diffuse and translucent cloud models listed in Table~2 of Warin et al. (1996).
For $^{12}$CO, the previously reported $T_{01}$ are all between 2.7 and 3.6 K.
The $T_{01}$ derived from our profile fits range from $\sim$ 2.6 K toward HD~192639 and 3.3 K toward HD~185418 to $\sim$ 11 K toward HD~147888, with the values for the other five sight lines all between 3.8 and 6.3 K; similar values (generally between 3 and 8 K) have been derived from mm-wave CO absorption observations toward extragalactic radio continuum sources (Liszt \& Lucas 1998).
The $^{12}$CO $T_{01}$ values for all 16 observed sight lines are significantly lower than those predicted by the Warin et al. (1996) translucent cloud models --- which is perhaps not surprising, since none of the sight lines has $A_{\rm V}$ or $N$(H$_2$) as high as the values assumed for the models.
A number of the values determined in this paper are similar to those predicted for the thermal equilibrium diffuse cloud models.
Even for those models, however, the predicted dependence of excitation temperature on rotational level ($T_{12}$ $<$ $T_{01}$ $\la$ $T_{23}$) is generally not seen for the eleven sight lines with data for $J$ $>$ 1, many of which have $T_{01}$ $\la$ $T_{12}$ $\la$ $T_{23}$.
For most of the sight lines in our sample, those three observed excitation temperatures are equal, within the uncertainties.
Wannier et al. (1997) pointed out that CO line emission from dense molecular gas adjacent or near to that seen in absorption could in principle increase the populations of levels $J$ $>$ 0 above the values due to collisional and radiative processes in the observed gas alone.
Analyses of the CO rotational excitation which neglect that emission would thus overestimate the density in the gas seen in absorption.
For most of the sight lines in the upper part of Table~\ref{tab:texco}, Wannier et al. concluded that the proposed radiative excitation could account entirely for the observed rotational excitation --- which thus could not be used to infer $n_{\rm H}$ in the gas.
One consequence (and test) of the proposed radiative excitation is that the excitation temperatures for $^{13}$CO should be lower than the corresponding values for $^{12}$CO, as the $^{13}$CO emission lines should be much weaker.
That does not appear to be the case, however, for the seven sight lines in Table~\ref{tab:texco} with values for $T_{01}$($^{13}$CO).
In six of the sight lines, $T_{01}$($^{13}$CO) is nominally greater than $T_{01}$($^{12}$CO); in all cases, the two temperatures are equal, within the mutual uncertainties.
And while two of the six components detected in both $^{12}$CO and $^{13}$CO absorption by Liszt \& Lucas (1998) have $T_{01}$($^{12}$CO) $>$ $T_{01}$($^{13}$CO), the other four components have $T_{01}$($^{12}$CO) $\sim$ $T_{01}$($^{13}$CO).
The radiative excitation mechanism proposed by Wannier et al. thus may not be significant, in many cases.
It is not clear whether the higher values of $T_{ij}$ ($\sim$ 8--11 K) for $^{12}$CO observed toward HD~147888 can be associated with a higher density in the molecular gas in that sight line.
While a relatively high density could be suggested by the observed relatively strong absorption from the excited fine-structure states of \ion{C}{1} and \ion{O}{1}, the density inferred from the C$_2$ rotational excitation (\S~\ref{sec-c2tex}) is similar to the values found for many other sight lines.
Unfortunately, $^{13}$CO was not detected toward HD~147888.
\subsubsection{C$_2$ Excitation}
\label{sec-c2tex}
In diffuse and translucent clouds, most C$_2$ is in various rotational levels of the ground vibrational state of the ground $X ^1\Sigma^+_g$ electronic state.
Levels in upper electronic states are populated by absorption of either near-IR photons [to $A~^1\Pi_u$ (Phillips system)] or (less frequently) UV photons [to $D~^1\Sigma^+_u$ (Mulliken system) or $F~^1\Pi_u$].
Quadrupole and intersystem (singlet-triplet) transitions then return the molecule to rotational levels in the ground vibrational state, whose populations are also affected by collisions with both H and H$_2$ (van Dishoeck \& Black 1982; van Dishoeck \& de Zeeuw 1984).
Because both collisional and radiative processes contribute to the population of the C$_2$ rotational levels in diffuse and translucent clouds, care must be taken when deriving the gas physical conditions from observations of C$_2$.
The effect of the radiative excitation (and subsequent cascades back to the ground electronic state) is to increase the populations of the excited rotational levels above their thermal equilibrium values --- thus yielding excitation temperatures for those levels that are greater than the actual kinetic temperature.
(While the energies of the excited C$_2$ rotational levels are similar to those of CO {\it for the same $J$}, the relative populations in those excited levels are much higher for C$_2$, primarily because C$_2$ has no dipole moment.)
Because the radiation-induced departures from thermal equilibrium increase with $J$, it is important (1) to get accurate measures of the lowest rotational levels (which yield the best estimates for the kinetic temperature) and (2) to measure as many of the higher $J$ levels (typically $J$ $\ga$ 8) as possible (to determine the extent to which radiative processes affect the rotational population distribution and to get tighter constraints on the density).
There are several ways of using the observed lower $J$ populations of C$_2$ to characterize the kinetic temperature $T_{\rm k}$.
In principle, the best estimate for $T_{\rm k}$ comes from $T_{02}$.
Unfortunately, however, $T_{02}$ can be somewhat uncertain, as $N_0$ is determined from the single, generally relatively weak R(0) line in each of the observed C$_2$ bands.
For the sight lines analyzed in this study, the relatively well determined $T_{02}$ range from about 20 K (HD~207198) to about 40 K (HD~206267, HD~210121, $\zeta$~Oph); the nominal values for HD~24534, HD~27778, and HD~147888 are all somewhat larger (but are also very uncertain).
Within the uncertainties in the $N_{J}$, it is often possible to fit a single excitation temperature to the lowest 3--4 levels (i.e., to $J$ = 0--4 or $J$ = 0--6), generally at a value somewhat larger than $T_{02}$ (e.g., for HD~24534 in Fig.~\ref{fig:c2ex}).
Toward HD~206267, for example, the excitation temperatures for the lowest two, three, and four levels are $T_{02}$ = 36$\pm$9 K, $T_{04}$ = 51$\pm$5 K, and $T_{06}$ = 53$\pm$3 K, respectively.
The populations of the higher $J$ levels, however, generally are increasingly higher than would be expected from thermal equilibrium at a local kinetic temperature of $T_{02}$ --- especially for $J$ $\ga$ 8.
As discussed above, such elevated populations of the higher $J$ levels can be produced by radiative excitation.
The set of observed C$_2$ rotational level populations may also be compared with models for the excitation in order to estimate both the actual kinetic temperature and the density (for an assumed ambient radiation field).
Figure~\ref{fig:c2rot} shows the normalized excitation diagrams (ln[(5~$N_{J}$)/((2$J$+1)~$N_2$)] vs. $E_{J}$/k) for six sight lines whose rotational level column densities were derived from our fits to the C$_2$ F-X, D-X, and/or A-X bands; the corresponding column densities were given in Table~\ref{tab:nc2}.
For each sight line, the observed relative C$_2$ rotational populations have been compared with those predicted by the models of van Dishoeck \& Black (1982; see also van Dishoeck \& de Zeeuw 1984), for $T$ ranging from 10 to 150 K (in steps of 5 K) and the density $n$ ranging from 50 to 1000 cm$^{-3}$ (in steps of 50 cm$^{-3}$)\footnotemark.
\footnotetext{The comparisons were performed using C$_2$ rotational populations calculated via a web-based tool (available at http://dibdata.org) developed by B. McCall (see \'{A}d\'{a}mkovics et al. 2003), which reproduces the relative $N_{J}$ in Table~7 of van Dishoeck \& Black (1982) for densities smaller by a factor of 1.35 (to account for differences in the adopted C$_2$ band $f$-values; see van Dishoeck \& de Zeeuw 1984).
Our derived densities were reduced by an additional factor of 1.2 to account for the C$_2$ $f$-values used in this study.}
The models may be characterized by the parameter $n\sigma$/$I$, which is a measure of the relative importance of collisional de-excitation and radiative excitation.
Here $n$ is the number density of collisional partners (H and H$_2$); $\sigma$ is the (assumed constant) collisional de-excitation cross-section for transitions from $J+2$ to $J$; and $I$ is a scaling factor for the strength of the interstellar radiation field in the near-IR (i.e., near the A-X bands), which is most important for the radiative excitation of C$_2$.
The cross-section $\sigma$ has not been measured for conditions appropriate for diffuse or translucent clouds, but has been estimated to be about 2$\times$10$^{-16}$ cm$^{2}$ from a detailed analysis of the $\zeta$ Per sight line (van Dishoeck \& Black 1982).
Calculations by Lavendy et al. (1991), however, suggest a somewhat higher value for the de-excitation cross-section for collisions with H$_2$ (Federman et al. 1994).
Higher values of $n\sigma/I$ --- i.e., higher $n$ and/or lower $I$ --- yield predicted rotational populations closer to the thermal equilibrium values.
The solid curves in Figure~\ref{fig:c2rot} show the populations predicted for the best-fit $T_{\rm k}$ and $n$ (assuming $I$ = 1).
For that best-fit temperature, the long-dashed straight lines show the rotational populations predicted for thermal equilibrium.
For the adopted cross-section $\sigma$, the A-X band $f$-values noted above, and $I$ = 1, the two dotted curves in each panel of Figure~\ref{fig:c2rot} correspond to $n$ = 1000 (near the thermal equilibrium line) and 100 cm$^{-3}$.
The two short-dashed curves correspond to the rotational populations predicted for slight differences from the best-fit $T_{\rm k}$ and $n$ --- for ($T_{\rm k}$ $+$ 10 , $n$ $+$ 50) and ($T_{\rm k}$ $-$ 10 , $n$ $-$ 50).
The values for $T_{\rm k}$ obtained in this way usually are lower than those for $T_{04}$, and are more comparable to those for $T_{02}$.
Estimates for $T_{02}$, $T_{04}$, and $T_{\rm k}$ (including values derived from equivalent widths or column densities reported in the literature for some additional sight lines) are listed in Table~\ref{tab:texc2}.
It can sometimes be difficult, however, to find a unique, tightly constrained combination of $T$ and $n\sigma/I$ for which the observed and predicted rotational populations agree within the (often considerable) observational uncertainties --- especially if data are not available for rotational levels above $J$ = 8.
In some cases, ``acceptable'' fits are possible for a fairly wide range of values, while in others, none of the models seem able to reproduce all the observed $N_{J}$.
Toward HD~206267, for example, the best fit to most of the $N_{J}$ (for these particular models) is for $T =$ 35 K (very similar to the derived $T_{02}$) and $n$ = $n$(H) + $n$(H$_2$) $\sim$ 250 cm$^{-3}$, assuming the average interstellar radiation field.
Within the uncertainties, however, acceptable agreement between the observed and predicted populations is also possible for slightly lower temperatures and densities and also for somewhat higher temperatures and densities.
[Note that $n$ (which counts collision partners) is smaller than $n_{\rm H}$ (which counts hydrogen nuclei) by a factor of (1 $-$ $f$(H$_2$)/2)$^{-1}$; estimates for both are given in columns (7) and (8) of Table~\ref{tab:texc2}.]
It seems somewhat unlikely, however, that $n_{\rm H}$ toward HD~206267 could be as high as 600--1200 cm$^{-3}$, as estimated from the CN/CH ratio (see \S~\ref{sec-var} above); CN may well trace denser gas, on average, than C$_2$.
If the (near-IR) radiation field is stronger than average, however, then the estimated densities would be proportionally higher.
The gas toward HD~206267 traced by H$_2$ probably has a somewhat higher temperature $\sim$ 65 K [$T_{01}$(H$_2$)] and somewhat lower average density $n_{\rm H}=$ 10-100 cm$^{-3}$ (obtained from an analysis of \ion{O}{1} fine-structure excitation, to be presented elsewhere).
Much of the H$_2$ may thus reside in more diffuse gas, with the contributions from the denser clumps masked by blending in the lower resolution {\it FUSE} spectra.
Of the ten sight lines with detected C$_2$ (mostly from this paper) listed in the bottom section of Table~\ref{tab:texc2}, seven (HD~24534, HD~27778, HD~147888, $\zeta$~Oph, HD~204827, HD~206267, and HD~210121) have relative rotational populations roughly consistent with $T$ $\sim$ 35--55 K and $n$ $\sim$ 150--300 cm$^{-3}$ (or $n_{\rm H}$ $\sim$ 200--400 cm$^{-3}$).
On average, the C$_2$-containing gas toward HD~207198 appears to be slightly warmer ($T_{\rm k}$ $\sim$ 60 K) but of comparable density;
the gas toward HD~73882 appears to be somewhat colder ($T_{\rm k}$ $\sim$ 20 K) and denser.
In many of the cases, somewhat lower $T_{\rm k}$ and slightly lower $n$ and/or somewhat higher $T_{\rm k}$ and higher $n$ would also be possible, within the uncertainties in the $N_{J}$.
(And recall that since the theoretical curves are for values of $n\sigma$/$I$, changes to the adopted $\sigma$ or $I$ will imply corresponding changes in $n$.)
The values for $T_{\rm k}$ and density in Table~\ref{tab:texc2} are consistent with those estimated by van Dishoeck et al. (1991), for nine sight lines in common, except that $T_{\rm k}$ is higher toward $\zeta$~Oph and HD~210121 (based on more recent C$_2$ data) and lower toward HD~62542.
We note that relatively few sight lines reported thus far in the literature have sufficiently accurate $N_{J}$ for high enough $J$ for very strong constraints to be placed on $T_{\rm k}$ and $n$; a number of these are included in the top part of Table~\ref{tab:texc2}.
More accurate column densities --- especially for $J$ = 0 and for the higher $J$ levels ($\ge$ 8) --- would be needed to better constrain the temperature and density in those sight lines.
Comparing the excitation temperatures derived for C$_2$ with the corresponding values for H$_2$ (Tables~\ref{tab:texc2} and \ref{tab:dq}) indicates that $T_{02}$(C$_2$) and $T_{\rm k}$(C$_2$) are almost invariably less than $T_{01}$(H$_2$) --- sometimes significantly so.
For the H$_2$ column density regime sampled by the sight lines with detected C$_2$, $T_{01}$(H$_2$) is thought to be an accurate measure of the local kinetic temperature, while (as noted above) $T_{02}$(C$_2$) may represent only an upper limit to $T_{\rm k}$ if radiative excitation is significant.
The systematically lower excitation and kinetic temperatures estimated for C$_2$ suggest that the C$_2$ is preferentially concentrated in somewhat colder, denser gas than that traced (on average) by H$_2$ --- consistent with predictions of the various models of cloud chemistry (e.g., van Dishoeck \& Black 1986a).
Given the indications that many (or most) sight lines are comprised of mixtures of gas with different characteristics --- and given the differences in physical conditions (e.g., density) often inferred from different tracers --- we may ask whether such mixtures can be detected and/or characterized via analyses of C$_2$ rotational excitation.
Cecchi-Pestellini \& Dalgarno (2002) have examined the behavior of the integrated sight line C$_2$ rotational populations when some fraction of the C$_2$ in the sight line is in dense clumps characterized by thermal equilibrium rotational populations.
Not surprisingly, the presence of such dense clumps is most noticeable when both the temperature and density of the more abundant diffuse gas are low --- e.g., for $T_{\rm k}$ $\la$ 20 K and $n$ $\la$ 100 cm$^{-3}$.
In such cases, the low-$J$ populations ($J$ $\la$ 8) for the sight line as a whole remain closer to the thermal equilibrium values, and the higher $J$ levels are somewhat enhanced as well.
Cecchi-Pestellini \& Dalgarno conclude, however, that such dense clumps will be very difficult to discern via analysis of the C$_2$ excitation (given the typical observational uncertainties) when the bulk of the more diffuse gas is warmer than $T_{\rm k}$ $\sim$ 20 K and/or denser than $n$ $\sim$ 100 cm$^{-3}$.
All of the sight lines in our primary sample (with the possible exception of that toward HD~210839) appear to fall in that latter category.
Several of the sight lines listed in the upper part of Table~\ref{tab:texc2} have $T_{\rm k}$ $\la$ 20 K, but $n$ $\ga$ 150 cm$^{-3}$; the sight line toward HD~110432 has $n$ $\sim$ 100 cm$^{-3}$, but uncertain temperature.
\subsection{Isotopic Ratio: $^{12}$C/$^{13}$C}
\label{sec-1213}
The carbon isotopic ratio $^{12}$C/$^{13}$C in the interstellar medium is often used as an indicator of the degree of stellar processing and, hence, of chemical evolution in the Galaxy.
It is thought that as the Galaxy evolves more $^{13}$C should be produced relative to the dominant isotope $^{12}$C.
While the current average local Galactic interstellar value of $^{12}$C/$^{13}$C is about 70 $\pm$ 3 (Stahl \& Wilson 1992), variations in that ratio of factors of 2--3 have been observed (Wilson 1999; Casassus, Stahl, \& Wilson 2005).
Estimates of this ratio have been based on observations of the isotopomers of species such as CH$^+$ (e.g. Hawkins et al. 1993), CO (Lambert et al. 1994), H$_2$CO (Henkel et al. 1985), and CN (Savage et al. 2002) --- each of which presents its own difficulties.
For example, while CH$^+$ is thought to be unaffected by chemical fractionation, measurement of $^{13}$CH$^+$ is often difficult because the (optical) lines can be quite weak.
CO, on the other hand, is subject to chemical fractionation --- either from isotope exchange reactions or from selective photodissociation.
Observations of mm-wave absorption spectra of the isotopomers of HCO$^+$, HCN, and HNC (which should not be significantly fractionated) have yielded $^{12}$C/$^{13}$C = 59 $\pm$ 2 (Lucas \& Liszt 1998).
The models of Warin et al. (1996) predict a somewhat complicated dependence of the $^{12}$CO/$^{13}$CO ratio on density, temperature, and $A_{\rm V}$ for clouds with $n_{\rm H}$ between 10$^2$ and 10$^4$ cm$^{-3}$ and $A_{\rm V}$ $\la$ 5 (see also van Dishoeck \& Black 1988).
The isotope exchange reaction $^{13}$C$^+$~+~$^{12}$CO $\rightarrow$ $^{13}$CO~+~$^{12}$C$^+$ [which is exothermic by only E/k $\sim$ 35 K (Watson, Huntress, \& Anicich 1979)] can enhance $^{13}$CO in low temperature gas at intermediate densities and $A_{\rm V}$ $\la$ 3 by as much as a factor of 5.
At higher temperatures, lower densities, and low $A_{\rm V}$, however, $^{13}$CO is more readily photodissociated than $^{12}$CO because of its much lower abundance, which does not allow effective self-shielding.
At higher densities ($\ga$ 10$^5$ cm$^{-3}$) and at higher $A_{\rm V}$, where chemical reactions dominate the destruction of CO and there is little $^{13}$C$^+$ present, the $^{12}$CO/$^{13}$CO ratio is restored to the average value corresponding to the carbon isotopic abundances.
Comparison of the $^{12}$CO/$^{13}$CO ratio with the corresponding average $^{12}$C/$^{13}$C ratio therefore yields indirect information concerning the physical conditions in the gas containing CO.
Reasonably reliable column densities of $^{13}$CO from UV absorption are now available for twelve sight lines (six from the literature and six from this paper; Table~\ref{tab:coldens}).
Of those twelve sight lines, three ($\rho$ Oph, $\chi$ Oph, $\zeta$ Oph) have $^{12}$CO/$^{13}$CO ratios significantly above the average value of 70, five (X Per, HD~27778, HD~207198, HD~210121, HD~210839) have $^{12}$CO/$^{13}$CO ratios consistent with the average value, and four (HD~73882, 20 Aql, HD~200775, HD~206267) have $^{12}$CO/$^{13}$CO ratios significantly below the average value.
Federman et al. (2003) ascribed the higher than average $^{12}$CO/$^{13}$CO ratios toward $\rho$ Oph, $\chi$ Oph, and $\zeta$ Oph to selective photodissociation and proposed that the ratio $N$($^{12}$CO)/$I_{\rm UV}$, where $I_{\rm UV}$ gives the relative strength of the UV radiation field, could be viewed as a measure of the degree of fractionation.
Liszt \& Lucas (1998) obtained $^{12}$CO/$^{13}$CO ratios between 15 and 54 (smaller than the average $^{12}$C/$^{13}$C ratio derived from other species) for a small set of mm-wave absorption components, with smaller ratios for higher $N$(CO).
Figure~\ref{fig:1213} plots the $^{12}$CO/$^{13}$CO ratios versus the corresponding $^{12}$CO column densities ({\it left}) and CN/CH ratios ({\it right}) for the twelve sight lines in Table~\ref{tab:coldens} with measured $N$($^{13}$CO).
For comparison, the asterisks in the left-hand panel show the total sight line values derived from mm-wave absorption by Liszt \& Lucas (1998).
For this relatively small sample of sight lines, the $^{12}$CO/$^{13}$CO ratio appears to be anti-correlated with both $N$($^{12}$CO) (as noted by Liszt \& Lucas 1998) and the CN/CH ratio, with correlation coefficients $r$ = $-$0.647 and $-$0.549, respectively (Table~\ref{tab:corr}).
If the CN/CH ratio is a density indicator, that anti-correlation would suggest that (in general) the $^{12}$CO/$^{13}$CO ratio decreases with increasing density --- presumably because the isotope exchange reaction has enhanced $^{13}$CO in denser, colder gas --- consistent with the behavior predicted by Warin et al. (1996) for densities in the range 100--1000 cm$^{-3}$.
The excitation and kinetic temperatures derived from observations of C$_2$ (Table~\ref{tab:dq}) suggest that $T$ might in fact be low enough in the clouds with low values of the $^{12}$CO/$^{13}$CO ratio for the isotope exchange process to be effective --- particularly if the effective temperatures are lower for the more centrally concentrated, density-sensitive CN and CO than for C$_2$.
Three of the four sight lines with lower than average $^{12}$CO/$^{13}$CO ratios (toward HD~73882, HD~200775, and HD~206267) have $T_{02}$(C$_2$) and/or $T_{\rm k}$(C$_2$) $\la$ 35 K; the four sight lines with average or above average $^{12}$CO/$^{13}$CO ratios and well-determined $T_{\rm k}$ all have $T_{\rm k}$ $\ga$ 45 K.
We note, however, that three of the four low $^{12}$CO/$^{13}$CO ratios (toward HD~73882, 20 Aql, and HD~200775) were derived from {\it IUE} spectra alone, and may be less reliable than the values obtained from the higher resolution, higher S/N {\it HST} spectra.
As there are also indications that the interstellar $^{12}$C/$^{13}$C ratio may be somewhat variable (e.g., Casassus et al. 2005), additional data on the $^{12}$CO/$^{13}$CO ratio and corresponding physical conditions are needed for a more secure understanding of the fractionation of CO in diffuse and translucent clouds.
Several of the other sight lines listed in the upper part of Table~\ref{tab:texc2} with $T_{\rm k}$(C$_2$) $\la$ 30 K might be interesting candidates for further study.
\section{Summary}
\label{sec-summ}
Archival UV spectra from {\it HST} (GHRS and STIS), {\it FUSE}, and/or {\it IUE}, together with high and medium resolution optical spectra, were analyzed to study the molecular abundances and physical conditions in the molecular gas along ten translucent sight lines.
Absorption from $^{12}$CO was detected in the far-UV E-X (0-0) (1076 \AA), C-X (0-0) (1087 \AA), and B-X (0-0) (1150 \AA) bands, in a number of bands in the A-X system (between 1240 and 1550 \AA), and/or in several of the stronger intersystem bands (between 1360 and 1550 \AA); corresponding A-X bands of $^{13}$CO were detected in six of the sight lines.
Accurate column densities were determined for $^{12}$CO and $^{13}$CO using a combination of curve of growth, apparent optical depth, and profile fitting methods.
High-resolution optical spectra of \ion{K}{1}, CH, and CN were used to determine the velocity structure of the gas along the sight lines in order to account for potential saturation effects in the UV spectra.
The UV C$_2$ F-X (0-0) absorption band at 1341 \AA\ was detected in high-resolution STIS echelle spectra of HD~24534, HD~27778, and HD~206267; the weaker F-X (1-0) band at 1314 \AA\ was detected toward HD~24534; and the stronger D-X (0-0) band at 2313 \AA\ was detected toward HD~24534, HD~27778, HD~147888, and HD~207198.
These constitute nearly all of the reported high-resolution spectra of the UV bands of C$_2$.
Absorption from the optical C$_2$ A-X (3-0) and/or (2-0) bands at 7719 and 8757 \AA\ was detected toward seven stars.
Comparisons among the various C$_2$ bands have yielded a (reasonably) mutually consistent set of band $f$-values; the value adopted for the A-X (2-0) band is consistent with several recent theoretical and experimental determinations.
There appear to be some discrepancies between the observed and predicted wavelengths and/or $f$-values for (at least) some individual lines in the F-X (0-0) band.
Reliable column densities (or limits) are now available for $^{12}$CO, CN, CH, C$_2$, and H$_2$ for at least 30 diffuse or translucent sight lines (11 with detections of $^{13}$CO; 13 with detections of HD; 7 with detections of C$_3$; 15 with $A_{\rm V}$ between 1.0 and 2.7 mag) --- enabling investigations of correlations between the various species and providing more stringent tests for models of the structure and chemistry of translucent clouds.
Of those molecular species, CN shows the best correlation with CO --- both in overall column density and in component structure.
Strong correlations between the ratios $^{12}$CO/H$_2$ and $^{13}$CO/H$_2$ and the density indicator CN/CH suggest that the abundances of both CO isotopomers increase with increasing local density.
Differences in the strength of the ambient radiation field (as inferred from H$_2$ rotational excitation) also are likely to affect the CO abundances.
For our sample of ten translucent sight lines, the $N$(CO)/$N$(H$_{2}$) ratio varies over a factor of 100, due primarily to differences in $N$(CO).
For the total sample, the $^{12}$CO/H$_2$ ratio ranges from 2.5 $\times$ 10$^{-8}$ to 2.8 $\times$ 10$^{-5}$; CO is not yet the dominant carbon-containing species in any of the sight lines.
The models of van Dishoeck \& Black (1988, 1989) predict relationships with H$_2$, CN, CH, and C$_2$ that have slopes similar to those obtained in fits to the observed data, but the predicted $N$(CO) generally are lower than the observed values by factors of 2--4.
The models of Warin et al. (1996) and Le Petit et al. (2006) predict much less CO than is observed, for a given $N$(H$_2$) --- though more of the parameter space available in the Meudon PDR code should be explored.
Inclusion of the effects of inhomogeneous/clumpy structure (e.g., Spaans 1996) and increasing the CO self-shielding and (for the Warin et al. models) the gas-phase carbon abundance should yield better agreement with the observed CO abundances.
Rotational level populations and corresponding excitation temperatures were determined up to $J$ = 3 for $^{12}$CO (nine sight lines) and up to $J$ = 2 for $^{13}$CO (four sight lines).
The $^{12}$CO excitation temperatures $T_{01}$ typically are between 3 and 7 K (well below the likely kinetic temperature of the gas); the corresponding values for the higher rotational levels are generally equal to $T_{01}$, within the uncertainties.
Some of these excitation temperatures are slightly higher than the values previously reported for eight other comparably reddened sight lines, but they are similar to the values found via mm-wave CO absorption by Liszt \& Lucas (1998) and to the values predicted for diffuse clouds with $A_{\rm V}$ $\sim$ 1.1 mag by Warin et al. (1996).
The excitation temperatures $T_{01}$ for $^{13}$CO are equal (within the mutual uncertainties) to the corresponding values for $^{12}$CO --- suggesting that the radiative excitation mechanism proposed by Wannier et al. (1997) may not be significant, in many cases.
Rotational level populations and excitation temperatures were also determined for C$_2$ --- up to $J$ = 12 from the UV F-X (0-0) and (1-0) bands (three sight lines), up to $J$ = 18 from the near-UV D-X (0-0) band (four sight lines), and up to $J$ = 12 from the optical A-X (2-0) and (3-0) bands (seven sight lines).
The excitation temperature for the lowest rotational level ($T_{02}$) ranges from about 20 to 45 K for those sight lines where it is reasonably well determined; corresponding values for the higher rotational levels increase with $J$ (especially for $J$ $\ga$ 8), due to radiative excitation of the upper states.
In most cases, the C$_2$ rotational excitation seems consistent with model predictions for kinetic temperatures of 35--55 K and local total hydrogen densities of 200--400 cm$^{-3}$, assuming the typical near-IR interstellar radiation field.
Of the twelve sight lines with measured $N$($^{13}$CO), three have $^{12}$CO/$^{13}$CO ratios significantly above the average Galactic $^{12}$C/$^{13}$C value of 70, five have $^{12}$CO/$^{13}$CO ratios consistent with the average value, and four have $^{12}$CO/$^{13}$CO ratios significantly below the average value.
There appears to be an anti-correlation between the $^{12}$CO/$^{13}$CO ratio and the CN/CH ratio --- suggestive of a decrease in $^{12}$CO/$^{13}$CO with increasing local density, as predicted by the models of Warin et al. (1996) for densities in the range 100-1000 cm$^{-3}$.
The $^{12}$CO/$^{13}$CO ratio is also anti-correlated with the column density of CO, as found (for a smaller sample) by Liszt \& Lucas (1998).
Sight lines with below average $^{12}$CO/$^{13}$CO ratios also appear to have kinetic temperatures (from analysis of the C$_2$ rotational excitation) less than about 35 K; sight lines with average or above average $^{12}$CO/$^{13}$CO ratios appear to be somewhat warmer.
While $^{12}$CO/$^{13}$CO ratios above the average value likely result from selective photodissociation (Federman et al. 2003), which favors the self-shielded $^{12}$CO, the below-average ratios probably are due to isotope exchange reactions, which favor $^{13}$CO in cold gas ($T$ $\la$ 35 K).
While the H$_2$ excitation temperature $T_{01}$ ranges from about 50 K to 110 K for our ten sight lines, there are several indications that C$_2$ and (especially) CO trace colder, denser gas in many cases.
The excitation temperature $T_{02}$ of C$_2$ (which provides an upper limit to the local kinetic temperature for the gas containing the C$_2$) and the $T_{\rm k}$ inferred from comparisons with models of the excitation are generally lower than $T_{01}$ for H$_2$ (which should equal the kinetic temperature of the gas traced by H$_2$).
Fits to the CO line profiles indicate that smaller $b$-values (0.3--0.7 km~s$^{-1}$) are needed than for fits to the profiles of \ion{K}{1}, CH, and CN; the CO structure also appears more similar to that of CN (which is concentrated in denser regions).
The column densities of both $^{12}$CO and $^{13}$CO are most tightly correlated with that of CN, and their abundances, relative to H$_2$, are correlated with the density indicator $N$(CN)/$N$(CH).
These results illustrate the utility of combining information from multiple species sensitive to different combinations of physical parameters in order to probe regions characterized by different properties (along a sight line or within a single ``cloud'').
They also emphasize the need for developing models capable of incorporating ranges and/or mixtures of physical conditions in order to accurately assess and predict the abundances of various atomic and molecular species in the diffuse ISM.
\acknowledgments
This work is based in part on data obtained for the Guaranteed Time Team by the NASA-CNES-CSA {\it FUSE} mission operated by the Johns Hopkins University.
Financial support to U. S. participants has been provided by NASA contract NAS5-32985.
P. Sonnentrucker acknowledges partial support from NASA LTSA grant NAG5-13114 to the Johns Hopkins University.
D. E. Welty acknowledges support from NASA LSTA grants NAG5-3228 and NAG5-11413 to the University of Chicago.
We thank E. Jenkins for providing processed STIS spectra of HD~206267 and HD~210839, S. Federman and K. Pan for providing results from their study of the Cep OB2 and OB3 regions in advance of publication, S. Federman for comments on an earlier version of the paper, B. McCall and T. Oka for advice on the analysis of C$_2$, and the anonymous referee for clarifying several points in the discussion.
This work has used the profile fitting procedure Owens.f developed by M. Lemoine and the {\it FUSE} French Team.
Facilities: \facility{FUSE}, \facility{HST (GHRS, STIS)}, \facility{IUE}, \facility{ARC}, \facility{KPNO:coud\'{e} feed}, \facility{McD:2.7m}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 4,692
|
\section{Introduction}
Inclusion of the dispersion interactions into the set of phenomena accounted for
by DFT models is recognized as one of the challenges in the
development of new density functional approximations~(DFAs).\cite{dobson2012dispersion,dobson2005soft,cohen2012challenges,dobson2012calculation}
Several
ways have been proposed to correct the currently available semilocal (SL) DFAs for the lacking nonlocal (NL) correlation contribution responsible for the dispersion interactions.\cite{cohen2012challenges,dobson2012dispersion}
Hereafter, global hybrid and range-separated
hybrid functionals will be called SL DFAs. Although the exchange parts of such functionals
are nonlocal, our focus will be on the correlation contributions, which in this
case depend on variables calculated at a single point of space.
The examples of such dispersion-corrected methods are:
\begin{inparaenum}[(i)]
\item the exchange-hole dipole method
(XDM),\cite{becke2005exchange,becke2005density,becke2007unified,hesselmann2009derivation,angyan2007exchange}
\item the atom pairwise additive schemes of Goerigk and Grimme, DFT-D3,\cite{grimme2010consistent} and Tkatchenko-Scheffler approach,\cite{tkatchenko2009accurate}
\item seamless van der Waals density functionals.\cite{dion2004van,lee2010higher,vydrov2009improving,vydrov2009nonlocal,vydrov2010nonlocal,vydrov2012nonlocal,dobson2012calculation}
\end{inparaenum}
It is clear that the accuracy of these methods depends not only on
a faithful representation of long-range electronic correlations, but also on
a consistent matching of a dispersion correction and the chosen SL complement.
Several groups have studied the conditions under which an SL functional
can be incorporated into a dispersion-corrected method.\cite{pernal2009dispersionless,rajchel2010derivation,kamiya2002density,murray2009investigation}
It has been concluded that the improper behavior of a GGA exchange functional in
the density tail (large reduced gradient regime) is responsible for
artificial exchange binding (as for the PBE\cite{perdew1996generalized} exchange) or overly repulsive
interaction (as for the B88\cite{becke1988density} exchange).\cite{kamiya2002density,murray2009investigation} Such
systematic errors may cause the NL correction to worsen the results
compared to the bare SL functional. The following exchange
functionals: PW86,\cite{perdew1986accurate}
refitted PW86,\cite{murray2009investigation} and range-separated hybrids\cite{kamiya2002density,vydrov2010nonlocal} were found to be free
from artificial binding, thus being consistent with NL dispersion
correction. Similarly, combining exact
exchange with NL correlation performs satisfactorily.\cite{vydrov2010implementation}
It has been observed that a failure to
satisfy the condition of vanishing correlation for a rapidly varying
density by SL correlation functionals\cite{perdew1996generalized} leads to a systematic overbinding of non-covalent complexes.\cite{kamiya2002density} The \emph{ad hoc} cure is to cancel the error of the correlation by the opposite-sign error of an exchange component.\cite{kamiya2002density} However, this does not resolve the problem of the double counting of SL and NL correlation. Several remedies have been proposed. For atom pairwise schemes, multiple damping functions have been devised.\cite{grimme2011effect} For VV09\cite{vydrov2009nonlocal}
and VV10\cite{vydrov2010nonlocal} density functionals the problem is avoided by demanding the NL constituent to vanish in
the homogeneous electron gas (HEG) limit, because the SL constituent is able
to describe the whole range of the electronic interactions in this limit.
Finally, \citet{pernal2009dispersionless} devised a procedure to reoptimize an existing SL
exchange-correlation functional so as to recover the dispersionless interaction energy.\cite{pernal2009dispersionless} The rationale of such an approach
is to let the SL functional contribute only the terms that it can describe reliably.
At this point we would like to shed light on the dispersion problem\cite{dobson2002prediction} in DFT. It has been well established that SL functionals fail to recover the long-range multipole-expanded dispersion energy. In fact, nearly all dispersion-corrected DFT approaches aim at recovering only long-range dispersion, roughly determined by
the leading terms of the
multipole expansion: $C_6$, and possibly $C_8$. The exceptions are
the approaches of \citet{pernal2009dispersionless} and
and \citet{rajchel2010derivation} which supplement an SL functional
with total non-expanded dispersion from SAPT. It is often overlooked that the long-range contribution, however, does not constitute the whole
dispersion energy at near-equilibrium distances. Setting aside the exchange-dispersion part, the dispersion energy,
as defined in the symmetry-adapted perturbation theory~(SAPT),\cite{hesselmann2003intermolecular,misquitta2005intermolecular} has a complex nature, and includes both long- and short-range contributions. This has first been observed by \citet{koide1976new} who quantified the short-distance behavior of the dispersion energy as $A + B R^2$, with $R$ being the intermonomer separation. The importance of the short-range correlation is also unambiguously, though indirectly, supported
by the significance of bond functions and explicitly
correlated Gaussian geminals in the dispersion energy calculations.\cite{burcl1995role}
A more direct argument points to
the existence of short-range terms in the exact angular expansion
of the dispersion energy. In the case of atomic interactions, the latter involves the interaction between $S$ states of monomers,\cite{koide1976new} which give no
contribution to the multipole expansion. Numerical results show that near the equilibrium bond length these terms, decaying
exponentially with the overlap density, can be comparable in magnitude to the
multipole expansion terms.\cite{koide1981second} As demonstrated by~\citet{dobson2002prediction} with the aid of simple models,
SL functionals cannot recover the
multipole expansion of the dispersion
energy. However, there is a good reason to believe that SL functionals are capable of describing
the terms that depend on overlap density.\cite{dobson2002prediction} Our model is intended to capture this contribution.
Recent thorough assessments of
DFT-D, XDM, and VV10 approaches have clearly shown
that a combination of an SL functional specifically designed for
a dispersion-corrected treatment with a dispersion correction improves both the description of
noncovalent
interactions\cite{burns2011density,goerigk2011thorough,hujo2011performance,vazquez2010assessment,thanthiriwatte2011assessment} and
general molecular properties.\cite{goerigk2011thorough,hujo2011performance} Among the functionals
that use atom-pairwise DFT-D correction,
B97-D,\cite{grimme2006semiempirical}
B97-D3,\cite{goerigk2011thorough}
and $\omega$-B97X-D\cite{chai2008long,chai2008systematic}
are characterized by one of the smallest magnitude and spread of
errors in interaction energies\cite{burns2011density} while performing
well in thermochemistry and reaction kinetics. The methods utilizing unaltered
conventional SL functional suffer from systematic errors. For example, PBE0-D2,
PBE-D3, and B3LYP-D3 tend to underbind dispersion-bound complexes and overbind
hydrogen-bonded systems.\cite{burns2011density}
The systematic overbinding of charge-transfer complexes within DFT is more pronounced for the dispersion-corrected approaches than for the pure DFAs.\cite{ruiz1996charge,sini2011evaluating}
See~Ref.\citenum{sini2011evaluating} and Table~2
in Ref.~\citenum{podeszwa2010extension}, where numerical
examples of huge overbinding by $\omega$-B97X-D and
B97-D functionals applied to charge-transfer complexes are given.
Although much attention has been devoted to the development of the
proper exchange contribution,\cite{kamiya2002density,murray2009investigation,swart2009new}
the theoretical effort to derive a dispersion-consistent SL correlation
functional thus far has been reduced to reoptimizing known expressions.
It has also been observed\cite{grimme2006semiempirical,chai2008systematic}
that fitting to empirical data coupled with
addition of higher-order terms in the B97 expansion\cite{becke1997density} does not systematically improve the performance as the saturation is approached. Clearly, there is a demand for the theoretical effort to overcome the problem.
The aim of this work is to develop an SL correlation functional
that can be matched with an arbitrary long-range dispersion correction
by optimizing a single parameter that has a simple physical meaning.
As a demonstration of this approach, we will combine our approximation with
DFT-D3 dispersion correction,\cite{grimme2010consistent} which contributes damped $C_6/R^6 + C_8/R^8$ terms, with
no short-range contributions. To avoid the systematic error of
spurious exchange attraction, full HF exchange will be used. To match SL and NL constituents,
a function which controls the spatial extent of our SL correlation hole will be
adjusted by minimization of errors in a relevant set of molecules.
\section{Theory}
We consider an electronic ground state of a finite
molecular system described by an electronic Hamiltonian of the form
\begin{equation}
\label{hamiltonian}
\hat{H}= \hat{T} + \sum_i \hat{v}_{\mathrm{ext}}(\mathbf{r}_i) + \hat{V}_\mathrm{ee},
\end{equation}
where $\hat{T}$ is the kinetic energy operator, the multiplicative external potential $\hat{v}_{\mathrm{ext}}$ is taken to be the
Coulomb potential of nuclear attraction, and $\hat{V}_\mathrm{ee}$ is the interelectronic repulsion. Atomic units are
assumed throughout this work. In constrained-search formulation\cite{levy1979universal}
of DFT\cite{hohenberg1964inhomogeneous} the ground state energy of electronic
system can be expressed as
\begin{equation}
\label{constrained-search}
E_0=\min_\rho \left[ \int \hat{v}_{\mathrm{ext}}(\mathbf{r}_1)\rho(\mathbf{r}_1)\mathrm{d}^3 \mathbf{r}_1 + \braket{\Psi^\mathrm{min}_\rho}{\hat{T}+\hat{V}_\mathrm{ee}} \right],
\end{equation}
where $\Psi^\mathrm{min}_\rho$ denotes an $N$-body wavefunction that yields electronic
density $\rho$ and simultaneously minimizes expectation
value of $\hat{T}+\hat{V}_\mathrm{ee}$. In Kohn-Sham scheme\cite{kohn1965self} the second
term on the right-hand side of \refr{constrained-search} is decomposed
into noninteracting kinetic, Hartree,
and exchange-correlation energies, respectively:
\begin{equation}
\braket{\Psi^\mathrm{min}_\rho}{\hat{T}+\hat{V}_\mathrm{ee}}=T_\mathrm{s}\left[\rho\right]+U\left[\rho\right]+E_{\mathrm{XC}}\left[\rho\right].
\end{equation}
Noninteracting kinetic energy $T_\mathrm{s}$ is known explicitly in terms of
the wavefunction of the KS system, denoted here as $\Phi^\mathrm{min}_\rho$,
which merely minimizes the expectation value of $\hat{T}$:
\begin{equation}
T_\mathrm{s}\left[\rho\right]=\braket{\Phi^\mathrm{min}_\rho}{\hat{T}}.
\end{equation}
Hartree energy is given by a classical formula
\begin{equation}
U\left[\rho\right]=\frac{1}{2}\iint \frac{\rho(\mathbf{r}_1)\rho(\mathbf{r}_2)}{r_{12}} \diff^3\ra\diff^3\rb.
\end{equation}
Exchange-correlation energy can be formally expressed through adiabatic connection
formula\cite{levy1996elementary}
\begin{equation}
\label{adiabatic-connection}
E_{\mathrm{XC}}\left[\rho\right]=\int_0^1 \braket{\Psi^\mathrm{min,\lambda}_\rho}{\hat{V}_\mathrm{ee}}\mathrm{d}\lambda -U\left[\rho\right],
\end{equation}
where $\Psi^\mathrm{min,\lambda}_\rho$ minimizes the expectation value of $\hat{T}+\lambda\hat{V}_\mathrm{ee}$ and yields
the same electronic density as wavefunction at $\lambda=1$.
\refr{adiabatic-connection} can be further decomposed so that the correlation energy
is separately expressed through coupling-constant integral
\begin{equation}
E_\mathrm{C}\left[\rho\right]=\int_0^1V_\mathrm{C}^\lambda\left[\rho\right] \mathrm{d}\lambda \label{lambda-average}
\end{equation}
where
\begin{equation}
V_\mathrm{C}^\lambda\left[\rho\right] = \braket{\Psi^\mathrm{min,\lambda}_\rho}{\hat{V}_\mathrm{ee}} - \braket{\Phi^\mathrm{min}_\rho}{\hat{V}_\mathrm{ee}}. \label{vc-def}
\end{equation}
Approximating $V_\mathrm{C}^\lambda$ is the primary objective of this work. Let us begin by expressing
$V_\mathrm{C}^\lambda$ in terms of a $\lambda$-dependent correlation hole
\begin{equation}
V_\mathrm{C}^\lambda=\frac{1}{2}\sum_{\sigma\sigma'}\iint \frac{\rho_\sigma(\mathbf{r}_1)h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)}{r_{12}}\diff^3\ra \diff^3\rb,
\end{equation}
where $\sigma$ denotes a spin variable,
\begin{equation}
h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2) = h_{\mathrm{XC}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2) - h_\mathrm{X}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2) \label{correlation-hole-def}
\end{equation}
and
\begin{align}
h_{\mathrm{XC}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)&=\frac{P_{2\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)}{\rho_{\sigma}(\mathbf{r}_1)}-\rho_{\sigma'}(\mathbf{r}_2), \\
h_\mathrm{X}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)&= -\delta_{\sigma\sigma'} \frac{\left|\sum_i^{N_\sigma}\psi_{i\sigma}^*(\mathbf{r}_1) \psi_{i\sigma}(\mathbf{r}_2)\right|^2}{\rho_\sigma(\mathbf{r}_1)}.
\end{align}
$N_\sigma$ is a number of $\sigma$-spin electrons and pair probability density, $P_{2\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)$, is defined as
\begin{align}
P_{2\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)&=N(N-1) \nonumber \\
& \times \sum_{\sigma_3\cdots\sigma_N}\int \Psi_\rho^{\min,\lambda*}(\mathbf{r}_1\sigma,\mathbf{r}_2\sigma',\ldots,\mathbf{r}_N\sigma_N)
\nonumber \\
& \times
\Psi_\rho^{\min,\lambda}(\mathbf{r}_1\sigma,\mathbf{r}_2\sigma',\ldots,\mathbf{r}_N\sigma_N)
\nonumber \\
& \times \mathrm{d}^3 \mathbf{r}_3 \cdots \mathrm{d}^3 \mathbf{r}_N.
\end{align}
Note that, due to the symmetry of $r_{\mathrm{ij}}^{-1}$ operator, it is the spherical average of exchange-correlation hole around
the reference electron that enters the energy expression:
\begin{equation}
\begin{split}
V_\mathrm{C}^{\sigma\sigma',\lambda} &= \frac 1 2 \iint \frac{\rho_\sigma(\mathbf{r}_1)h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,\mathbf{r}_2)}{r_{12}}\diff^3\ra \diff^3\rb \\
&= \frac 1 2 \int \diff^3\ra \int_0^\infty \frac{\rho_\sigma(\mathbf{r}_1)h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,s)}{s} 4\pi s^2 \mathrm{d} s \label{vcss-spherical}
\end{split}
\end{equation}
where the spherical average is implied by scalar argument $s$,
\begin{equation}
h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1, s) = \frac{1}{4\pi} \int_0^{2\pi} \mathrm{d} \phi_\mathbf{s} \int_0^\pi h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1, \mathbf{r}_1 + \mathbf{s})
\sin \theta_\mathbf{s} \mathrm{d} \theta_\mathbf{s}. \label{spherical-average}
\end{equation}
\refr{vcss-spherical} means that without loss of generality
we can focus our attention on approximating isotropic quantity defined in \refr{spherical-average}.
We postulate the following form of opposite-spin and same-spin correlation holes:
\begin{align}
h_{\mathrm{C}\lambda}^{\alpha\beta}(\mathbf{r}_1,s) &= (a_{\alpha\beta} + b_{\alpha\beta} s + c_{\alpha\beta} s^2) \exp(-d_{\alpha\beta} s), \label{opphole-def}\\
h_{\mathrm{C}\lambda}^{\alpha\alpha}(\mathbf{r}_1,s) &= s^2 (a_{\alpha\alpha} + b_{\alpha\alpha} s + c_{\alpha\alpha} s^2) \exp(-d_{\alpha\alpha} s). \label{parhole-def}
\end{align}
Unknown parameters appearing above
are actually functions of $\rho(\mathbf{r}_1)$, but we will not write that explicitly for the sake of brevity. Quadratic behavior of $h_{\mathrm{C}\lambda}^{\alpha\alpha}$ near the reference point
results from the Pauli exclusion principle and the cusp condition
discussed further below. The model of \refr{opphole-def} and \refr{parhole-def} cannot
recover radial dependence of the true hole at each point,
however, its simple form leads to reasonable shape of the
system-averaged correlation hole. We thus assume that the wealth of
features of the exact hole\cite{baerends1997quantum} is
averaged out, and use system-averaged function of shape similar to
\refr{opphole-def} and \refr{parhole-def} in the energy
expression. The argument involving system average was originally put forward
by Burke~et~al.\cite{burke1998semilocal} in
their discussion of the success of the local-density approximation~(LDA).
The shape of the correlation holes corresponding to Eqs.~\ref{opphole-def}
and~\ref{parhole-def} is illustrated
in~Fig.~\ref{fig-corrhole} (the details of parametrization are discussed below.) For any spin density
the qualitative picture is similar: both same-spin and opposite-spin
holes are removing electrons in the vicinity of the origin,
then cross the abscissa exactly once, and decay
exponentially. The fact that both model correlation holes~\eqref{opphole-def}--\eqref{parhole-def} change sign
exactly once can be readily proven.
\begin{figure}[p]
\includegraphics[width=1.0\textwidth]{corrhole.pdf}
\caption{Shape of the approximate correlation holes defined in
Eqs~\ref{opphole-def} and~\ref{parhole-def}. The functions
illustrated on the graph are $h_\mathrm{C}^{\alpha\beta,1}$
and $h_\mathrm{C}^{\alpha\alpha,1}$
multiplied by $4 \pi s^2$. The functions are
evaluated at $\rs=1$ in the spin-compensated HEG limit. Atomic units are used.}
\label{fig-corrhole}
\end{figure}
The form of correlation holes given in~\refr{opphole-def} and~\refr{parhole-def}
has been derived from observation of system-averaged correlation holes
(correlation intracules) in simple systems dominated by dynamical
correlation. Qualitatively, the shape of our model correlation hole is
similar to correlation intracules in \ce{He},\cite{sarsa1998correlated}
\ce{Ne},\cite{sarsa1998correlated} and \ce{H2}\cite{hollett2011nature} near
the equilibrium bond length. We note that there is a qualitative discrepancy
between our approximate correlation hole and the accurate one for systems like \ce{Li}\cite{sarsa1998correlated} or
\ce{Be}.\cite{sarsa1998correlated} These systems are characterized by a
significant contribution of static correlation. However, there
is a substantial cancellation between exchange and correlation holes in
systems of this type.
The cusp conditions for same-spin and opposite-spin
exchange-correlation holes\cite{rajagopal1978short} considerably
restrict the short-range expansion\cite{becke1988correlation} of $h_{\mathrm{XC}\lambda}^{\alpha\beta}$ and $h_{\mathrm{XC}\lambda}^{\alpha\alpha}$:
\begin{align}
h_{\mathrm{XC}\lambda}^{\alpha\beta}(\mathbf{r}_1,s) &= (B_{\alpha\beta} - \rho_{\beta}) + \lambda B_{\alpha\beta} s + \ldots, \label{hxcab-expansion}\\
h_{\mathrm{XC}\lambda}^{\alpha\alpha}(\mathbf{r}_1,s) &= -\rho_{\alpha} + (B_{\alpha\alpha} - \frac{1}{6} \nabla^2 \rho_{\alpha})s^2 + \frac{\lambda}{2}B_{\alpha\alpha} s^3 + \ldots \label{hxcaa-expansion}
\end{align}
\refr{correlation-hole-def} together with \refr{hxcab-expansion}, \refr{hxcaa-expansion}, and the expansion of
spherically-averaged exact exchange hole valid at zero current density,\cite{becke1988correlation,lee1987gaussian,becke1996current}
\begin{equation}
h_{\mathrm{X}}^{\alpha\alpha}(\mathbf{r}_1,s) = -\rho_{\alpha} - \frac{1}{6} \left[ \nabla^2\rho_{\alpha} -2 \tau_\alpha + \frac{1}{2} \frac{(\nabla\rho_{\alpha})^2}{\rho_{\alpha}} \right]s^2+\ldots, \label{exchange-hole}
\end{equation}
yields short-range expansions of correlation holes:\cite{becke1988correlation}
\begin{align}
h_{\mathrm{C}\lambda}^{\alpha\beta}(\mathbf{r}_1,s) &= (B_{\alpha\beta} - \rho_{\beta}) + \lambda B_{\alpha\beta} s + \ldots, \label{hcab-expansion}\\
h_{\mathrm{C}\lambda}^{\alpha\alpha}(\mathbf{r}_1,s) &= \left( B_{\alpha\alpha} - \frac 1 3 D_\alpha \right)s^2 + \frac \lambda 2 B_{\alpha\alpha} s^3 + \ldots \label{hcaa-expansion}
\end{align}
$D_\alpha$ is always non-negative and vanishes for single orbital densities:
\begin{equation}
D_\alpha = \tau_\alpha - \frac{|\nabla\rho_\alpha|^2}{4\rho_{\alpha}}, \label{d-inhom}
\end{equation}
where $\tau_\alpha$ is essentially the density of noninteracting kinetic energy
\begin{equation}
\tau_\alpha = \sum_i^{N_\alpha} |\nabla\psi_{i\alpha}|^2.
\end{equation}
We will adjust the unknown functions $B_{\alpha\beta}$ and $B_{\alpha\alpha}$ in \refr{hxcab-expansion} and \refr{hxcaa-expansion}
to recover short range expansion of spin-resolved pair distribution
function of the HEG developed by
Gori-Giorgi and Perdew.\cite{gori2001short} The pair distribution
function represents the
solution of the Overhauser model.\cite{overhauser1995pair}
$B_{\alpha\alpha}$ will be further modified to eliminate self-interaction error of the correlation functional.
We leave the on-top value of the correlation hole
(determined solely by $B_{\alpha\beta}$) unchanged by inhomogeneity corrections
because it is well-transferable from the HEG to real
systems.\cite{burke1998semilocal} For the discussion of the quality of
the HEG on-top hole density see the work of \citet{burke1998semilocal}.
Comparison of homogeneous density limit of \refr{hxcab-expansion} and \refr{hxcaa-expansion} with
short-range expansions of the spin-resolved HEG pair distribution
function\cite{gori2001short} yields
\begin{align}
\begin{split}
B_{\alpha\beta}(\rho_{\alpha}, \rho_{\beta},\lambda) &= \rho_{\beta} \left(1 + 0.0207\lambda\rsab + 0.08193(\lambda\rsab)^2 \right. \\
&\left. -0.01277(\lambda\rsab)^3+0.001859(\lambda\rsab)^4 \right)\exp(-0.7524\lambda \rsab),
\end{split} \label{bab-heg}
\\
\begin{split}
B_{\alpha\alpha}^\mathrm{HEG}(\rho_{\alpha},\lambda) &= \frac{D_\alpha^\mathrm{HEG}}{3} \left(1 -
0.01624\lambda\rsaa + 0.00264(\lambda\rsaa)^2 \right) \\
& \times \exp(-0.5566\lambda\rsaa), \label{baa-heg}
\end{split}
\end{align}
where $\rsaa$ and $\rsab$ introduce the dependence on electronic spin densities,
\begin{align}
\rsaa &= \frac{\left( 3/\pi \right)^{1/3}}{2
\rho_{\alpha}^{1/3}} \label{rsaa-def} \\
\rsab &= \frac{\left( 3/\pi \right)^{1/3}}{\rho_{\alpha}^{1/3} +
\rho_{\beta}^{1/3}} \label{rsab-def}
\end{align}
and each of them reduces to the Seitz radius
\begin{equation}
\rs = \left( \frac{3}{4\pi\rho} \right)^{1/3},
\end{equation}
for spin-compensated systems. The formulae~\eqref{bab-heg} and~\eqref{baa-heg} respect the exact high-density
expansion derived by \citet{rassolov2000reply}. The HEG limit of parameter~\eqref{d-inhom} in \eqref{baa-heg} reads\cite{becke1988correlation}
\begin{equation}
D^\mathrm{HEG}_\alpha=\frac{3}{5} (6\pi^2)^{2/3} \rho_\alpha^{5/3}.
\end{equation}
We substitute $D^\mathrm{HEG}_\alpha$ in \refr{baa-heg} for $D_\alpha$
of \refr{d-inhom} to get $B_{\alpha\alpha}$:
\begin{equation}
\begin{split}
B_{\alpha\alpha}(\rho_\alpha,|\nabla\rho_\alpha|,\tau_\alpha,\lambda) &= \frac{D_\alpha}{3} \left(1 -
0.01624\lambda\rsaa + 0.00264(\lambda\rsaa)^2 \right) \\
& \times \exp(-0.5566\lambda\rsaa). \label{baa-inhom}
\end{split}
\end{equation}
Such choice of the $B_{\alpha\alpha}$ function leads to vanishing parallel spin correlation
contribution for single orbital densities. In that sense no correlation
self-interaction error is present.
Restricting undetermined coefficients in \refr{opphole-def}
and \refr{parhole-def} to yield short-range expansions of \refr{hcab-expansion}
and \refr{hcaa-expansion}, respectively, gives
\begin{align}
a_{\alpha\beta} &= B_{\alpha\beta} - \rho_\beta, \\
b_{\alpha\beta} &= \lambda B_{\alpha\beta} + d_{\alpha\beta} a_{\alpha\beta}. \label{bopp-param}
\end{align}
Correct shape of $h_{\mathrm{C}\lambda}^{\alpha\beta}$ can be ensured requiring that the function satisfies the appropriate
sum rule,
\begin{equation}
4\pi\int_0^\infty h_{\mathrm{C}\lambda}^{\alpha\beta}(\mathbf{r}_1,s)s^2\mathrm{d} s = 0. \label{sum-rule}
\end{equation}
Consequently, coefficient $c_{\alpha\beta}$ is fixed for \refr{sum-rule} to hold for all densities:
\begin{equation}
c_{\alpha\beta} = -\frac{1}{12}(a_{\alpha\beta}d_{\alpha\beta}^2+3b_{\alpha\beta}d_{\alpha\beta}) \label{copp-param}.
\end{equation}
Analogously,
\begin{align}
a_{\alpha\alpha} &= B_{\alpha\alpha} - \frac{D_\alpha}{3}, \\
b_{\alpha\alpha} &= \frac{\lambda}{2}B_{\alpha\alpha} + a_{\alpha\alpha}d_{\alpha\alpha}, \label{bpar-param} \\
c_{\alpha\alpha} &= -\frac{1}{30}(a_{\alpha\alpha}d_{\alpha\alpha}^2 + 5b_{\alpha\alpha}d_{\alpha\alpha}) \label{cpar-param}.
\end{align}
With all but $d_{\alpha\beta}$ and $d_{\alpha\alpha}$ coefficients determined, spin resolved contributions
to $V_\mathrm{C}^\lambda$,
\begin{equation}
V_\mathrm{C}^{\sigma\sigma',\lambda} = \frac 1 2 \int \mathrm{d}^3 \mathbf{r}_1 \int_0^\infty \frac{\rho_{\sigma} h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\mathbf{r}_1,s)}{s} 4\pi s^2 \mathrm{d} s,
\end{equation}
can now be given as:
\begin{align}
V_\mathrm{C}^{\alpha\beta,\lambda} &= \int \mathrm{d}^3 \mathbf{r}_1 \rho_{\alpha} \pi \frac{b_{\alpha\beta} +
a_{\alpha\beta}d_{\alpha\beta}}{d_{\alpha\beta}^3}, \label{vcabl} \\
V_\mathrm{C}^{\alpha\alpha,\lambda} &= \int \mathrm{d}^3 \mathbf{r}_1 \rho_{\alpha} \pi \frac{8b_{\alpha\alpha} +
4a_{\alpha\alpha}d_{\alpha\alpha}}{d_{\alpha\alpha}^5}. \label{vcaal}
\end{align}
Several requisites for the exact exchange-correlation functional were derived using uniform coordinate scaling
technique,\cite{levy1991density,levy1985hellmann} i.e. by applying uniformly scaled density
\begin{equation}
\rho_\kappa(\mathbf{r}_1) = \kappa^3 \rho(\kappa \mathbf{r}_1).
\end{equation}
These relations are particularly valuable because they hold for arbitrary $N$-electron densities.
A density-scaling identity proved by Levy,\cite{levy1991density}
\begin{equation}
h_{\mathrm{C}\lambda}^{\sigma\sigma'}(\rho;\mathbf{r}_1,s) = \lambda^3 h_{\mathrm{C}\lambda'=1}^{\sigma\sigma'}%
(\rho_{1/\lambda}; \lambda\mathbf{r}_1,\lambda s),\label{lambda-dependence}
\end{equation}
constrains the set of admissible forms of $d_{\alpha\beta}$ and
$d_{\alpha\alpha}$. \refr{lambda-dependence} implies that
\begin{equation}
d_{\sigma\sigma'}(\rho,|\nabla\rho|,\lambda) = \lambda d_{\sigma\sigma'}%
\left(\frac{\rho}{\lambda^3},\frac{|\nabla\rho|}{\lambda^4},1\right).
\end{equation}
We propose the following simple function which satisfies the scaling condition:
\begin{equation}
d_{\sigma\sigma'} = \frac{F_{\sigma\sigma'}}{\rs^{\sigma\sigma'}}
+ \frac{G}{\rs} \frac{\nabla\rho \cdot \nabla\rho}{\rho^{8/3}}.
\label{dss-form}
\end{equation}
As~\refr{dss-form} is independent of $\lambda$, the coupling-constant
integration of~\refr{lambda-average} can be done analytically. The values
of $F_{\alpha\beta}$ and $F_{\alpha\alpha}$ were determined by least-squares fit of the HEG limit of
$V_\mathrm{C}^{\alpha\beta,\lambda=1}$ and~$V_\mathrm{C}^{\alpha\alpha,\lambda=1}$ to the reference values.\cite{gori2000analytic}
Opposite-spin and parallel-spin components were fit independently. Reference values of
$V_\mathrm{C}^{\sigma\sigma',\lambda=1}$ for the HEG were obtained by Gori-Giorgi et~al.\cite{gori2000analytic}
by integrating pair correlation functions from quantum Monte Carlo simulation.\cite{ortiz1999zero}
Our estimates of $V_\mathrm{C}^{\alpha\beta,\lambda=1}$
and~$V_\mathrm{C}^{\alpha\alpha,\lambda=1}$ were optimized to recover
the reference values for spin-compensated system at metallic
densities ($\rs = 1, 2, 3, \ldots, 10$). The resulting parameters are $F_{\alpha\beta}=2.1070$ and
$F_{\alpha\alpha}=2.6422$. The corresponding mean absolute percentage errors
(MAPE) of opposite-spin and parallel-spin components are $5.0\%$ and
$12.0\%$, respectively. The MAPE of total
$V_\mathrm{C}^{\lambda=1}$ is equal to $4.6\%$. See Fig.~\ref{fig:vc}
for comparison of our fit to the reference values. At high densities ($\rs < 1$) our model
does not reduce to the accurate correlation functional for the HEG as it
does not account for the logarithmic divergence of the correlation
energy density for $\rs\rightarrow 0$.\cite{gell1957correlation}
This is, however, a peculiarity of the HEG that
is not present in finite molecular systems.
\begin{figure}[p]
\includegraphics[width=1.0\textwidth]{vcabvcaa.pdf}
\caption{Comparison of the approximate
$V_\mathrm{C}^{\sigma\sigma',\lambda=1}$ energies in the HEG limit with reference
Monte Carlo data.\cite{gori2000analytic} Solid lines refer
to Eqs~\ref{vcabl} and~\ref{vcaal}. Circles and diamonds represent reference values.
The cause of discrepancy at low $\rs$ (high densities) is
discussed in the main text. Atomic units are used.}
\label{fig:vc}
\end{figure}
We supply our SL correlation functional with the DFT-D3
dispersion correction of~\citet{grimme2010consistent}, which contributes
damped terms of the multipole expansion of the dispersion energy.
The adjustment of the SL part to harmonize with the NL
correction is accomplished by optimization of the $G$
parameter, see~\eqref{dss-form}. The
$G$ parameter can be adjusted freely, without
interfering with any of the above-mentioned physical and formal constraints. In particular, it does not alter the first two terms in the short-range
spatial Taylor expansion of the correlation holes.
The value of $G$ can be chosen so that the correlation
contributions described by SL and NL parts do not overlap. As $G\rightarrow 0$,
our SL correlation model reduces to the correlation of the HEG
with self-interaction removed from parallel-spin part.
Our numerical results show that this leads to
a systematic overestimation of intermolecular interactions.
On the other hand, when $G\rightarrow \infty$,
the SL correlation vanishes, and the SL functional reduces to an exchange-only
approximation (without adding the dispersion correction). Provided that the
exchange functional is free from artificial binding,
the interaction energies should be underestimated in this limit. Between these two limits
lays the optimal $G$, which corresponds to an
interaction curve slightly shallower than the real one, for the addition
of the negative dispersion term should move the interaction energy towards
the accurate value.
The $G$ parameter of~Eq.~\ref{dss-form} and the two empirical parameters
present in DFT-D3 dispersion correction, $s_{r,6}$ and $s_8$, (see
Eqs. 3 and 4 in Ref.~\citenum{grimme2010consistent}) were
chosen to optimize mean absolute percentage error of binding energies
in S22 set of non-covalently bound
complexes.\cite{jurecka2006benchmark} The numerical optimization has
been carried out with the constraint that the dispersion-free energy
cannot fall below the reference total interaction energy. During the optimization process,
self-consistent KS calculations in aug-cc-pVTZ basis set were
performed using the molecular
structures published in Ref.~\citenum{jurecka2006benchmark}. Reference
interaction energies were taken from
Ref.~\citenum{podeszwa2010improved}. The resulting optimal values are
$G=0.096240$, $s_{r,6}=1.1882$, and
$s_8=0.65228$. The interaction energies in S22 set are presented in Table~\ref{s22-results}.
\section{Implementation}
The expression for the correlation energy is obtained after inserting~\refr{vcabl}
and~\refr{vcaal} into~\refr{lambda-average} and integrating with
respect to $\lambda$. Below we present $E_\mathrm{C}$ in a form convenient for implementation.
\begin{align}
E_\mathrm{C} &= E_\mathrm{C}^{\alpha\beta} + E_\mathrm{C}^{\beta\alpha} + E_\mathrm{C}^{\alpha\alpha} + E_\mathrm{C}^{\beta\beta}, \\
E_\mathrm{C}^{\alpha\beta} &= \int_0^1V_\mathrm{C}^{\alpha\beta,\lambda} \mathrm{d}\lambda = \int \mathrm{d}^3 \mathbf{r}_1 \rho_{\alpha} \pi
\frac{\mathcal{B}_{\alpha\beta} + \mathcal{A}_{\alpha\beta}d_{\alpha\beta}}{d_{\alpha\beta}^3}, \\
E_\mathrm{C}^{\alpha\alpha} &= \int_0^1V_\mathrm{C}^{\alpha\alpha,\lambda} \mathrm{d}\lambda = \int \mathrm{d}^3 \mathbf{r}_1 \rho_{\alpha} \pi
\frac{8\mathcal{B}_{\alpha\alpha} + 4\mathcal{A}_{\alpha\alpha}d_{\alpha\alpha}}{d_{\alpha\alpha}^5}, \\
\mathcal{A}_{\alpha\beta} &= \frac{\rho_{\beta}}{\rsab} \left[ \left(-P_0 + \sum_{k=1}^4 P_k
(\rsab)^k \right)\exp\left( -P_5 \rsab \right) + P_0 \right] -
\rho_{\beta} \label{Aopp-def}\\
\mathcal{B}_{\alpha\beta} &= \frac{\rho_{\beta}}{(\rsab)^2} \left[ \left(-Q_0 + \sum_{k=1}^5 Q_k (\rsab)^k
\right) \exp\left(-Q_6 \rsab \right) + Q_0 \right] + d_{\alpha\beta} \mathcal{A}_{\alpha\beta}
\label{Bopp-def}\\
\mathcal{A}_{\alpha\alpha} &= \frac{D_\alpha}{3 \rsaa}\left[ \left( -R_0 +
\sum_{k=1}^2 R_k (\rsaa)^k \right) \exp\left( -R_3 \rsaa \right)
+ R_0 \right] - \frac{D_\alpha}{3} \label{Apar-def}\\
\mathcal{B}_{\alpha\alpha} &= \frac{D_\alpha}{6(\rsaa)^2} \left[ \left( -S_0 +
\sum_{k=1}^3 S_k (\rsaa)^k \right)\exp\left( -S_4 \rsaa \right) +
S_0 \right] + d_{\alpha\alpha} \mathcal{A}_{\alpha\alpha} \label{Bpar-def}
\end{align}
Note that $E_\mathrm{C}^{\alpha\beta}=E_\mathrm{C}^{\beta\alpha}$. The formula for $E_\mathrm{C}^{\beta\beta}$ can be obtained by substitution
of spin indices in $E_\mathrm{C}^{\alpha\alpha}$. The values of the numerical constants appearing in Eqs.~\ref{Aopp-def}--\ref{Bpar-def} are
listed in~Table~\ref{numerical-constants}. The following functions:
$D_\alpha$, $\rsaa$, and $\rsab$ are defined in
Eqs~\ref{d-inhom}, \ref{rsaa-def}, \ref{rsab-def}, respectively. The
$d_{\sigma\sigma'}$ function, defined in Eq.~\ref{dss-form}, is
parametrized as follows:
\begin{align}
d_{\alpha\beta} &= \frac{2.1070}{\rs^{\alpha\beta}}
+ \frac{0.096240}{\rs} \frac{\nabla\rho \cdot \nabla\rho}{\rho^{8/3}}, \\
d_{\alpha\alpha} &= \frac{2.6422}{\rs^{\alpha\alpha}}
+ \frac{0.096240}{\rs} \frac{\nabla\rho \cdot \nabla\rho}{\rho^{8/3}}.
\end{align}
The parameters appearing in DFT-D3 correction\cite{grimme2010consistent} are $s_{r,6}=1.1882$ and
$s_8=0.65228$. Fortran code for numerical evaluation of the correlation
energy and its derivatives, together with the corresponding Mathematica\cite{mathematica7} notebook, can be obtained from the
authors by e-mail or from their webpage. The calculations presented in
this work were done using GAMESS program.\cite{schmidt1993general,gordon2005advances}
\begin{table}[p]
\caption{Ab initio numerical constants appearing in Eqs~\ref{Aopp-def}--\ref{Bpar-def}}
\label{numerical-constants}
\begin{tabular}{l|D{.}{.}{10} D{.}{.}{10} D{.}{.}{10} D{.}{.}{10}}
\hline \hline
k & P_k & Q_k & R_k & S_k \\
\hline
$0$ & 1.696 & 3.356 & 1.775 & 3.205 \\
$1$ & -0.2763 & -2.525 & 0.01213 & -1.784 \\
$2$ & -0.09359 & -0.4500 & -4.743\times 10^{-3} & 3.613\times 10^{-3} \\
$3$ & 3.837\times 10^{-3} & -0.1060 & 0.5566 & -4.743\times 10^{-3} \\
$4$ & -2.471\times 10^{-3} & 5.532\times 10^{-4} & & 0.5566 \\
$5$ & 0.7524 & -2.471\times 10^{-3} & & \\
$6$ & & 0.7524 & & \\
\hline \hline
\end{tabular}
\end{table}
\section{Discussion}
Similar strategy for designing a correlation functional, i.e., constructing
a real-space model for a spin-resolved correlation hole, was originally proposed
by~\citet{rajagopal1978short} with the first application by~\citet{becke1988correlation}, followed by the works of~\citet{proynov1994simple} and~\citet{tsuneda1999new}. A central role in those methods is played by \emph{correlation length}, a function completely determining both short- and long-range behavior of the correlation holes present in those models. Our approach has more degrees of freedom, as the short-range
behavior of $h_{\mathrm{C}\lambda}^{\sigma\sigma'}$ is decoupled from the choice of the
$d_{\sigma\sigma'}$ function which controls its decay. This flexibility allows us to adjust $d_{\sigma\sigma'}$ to
match a specific NL correction without sacrificing the short-range correlation that
can be accurately represented by an SL functional.
The ultimate goal is to develop a general-purpose functional that not
only yields satisfactory results for the dispersion interactions,
but also performs not worse than the existing approximations in
predicting other properties of chemical interest. To do so, the
inclusion of the NL constituent and the accompanying adjustment of the
SL part should not affect any of the energetically important
constraints already satisfied by the meta-GGA rung
functionals.\cite{perdew2005prescription} Among the formal
constraints, the most fundamental one is the non-positivity condition,
\begin{equation}
V_\mathrm{C}^{\sigma\sigma',\lambda}\left[\rho\right] \le 0, \label{non-positivity}
\end{equation}
which is obeyed by our model for every spin-density. Similarly,
the scaling conditions formulated by Levy,\cite{levy1991density} e.g.,
\begin{align}
\lim_{\kappa\rightarrow 0} \frac{E_\mathrm{C}\left[ \rho_\kappa
\right]}{\kappa} &=
\sum_{\sigma\sigma'}\lim_{\lambda\rightarrow\infty} \label{low-density-limit}
V_\mathrm{C}^{\sigma\sigma',\lambda}\left[\rho\right] > -\infty, \\
\frac{\partial V_\mathrm{C}^{\sigma\sigma',\lambda}\left[ \rho \right]}{\partial \lambda} &\le 0,
\end{align}
together with the high-density limit of the
correlation functional,\cite{levy1989asymptotic}
\begin{equation}
\lim_{\kappa\rightarrow\infty} E_\mathrm{C}\left[\rho_\kappa\right] >
-\infty, \label{high-density-limit}
\end{equation}
are satisfied. A failure to satisfy condition~\eqref{high-density-limit}
may contribute to overbinding of
molecules.\cite{levy1991density}
In addition to conditions~\eqref{non-positivity}--\eqref{high-density-limit}, our approximation satisfies a constraint which
has a direct connection to the prediction of interaction energies. It
was observed by~\citet{kamiya2002density} that if an SL
functional yields nonzero contributions to the correlation in the tail
of the density, then adding an NL correction may lead to a
severe overbinding.\cite{kamiya2002density} In the tail of electronic
density, where the reduced gradient is large, the $d_{\sigma\sigma'}$ function of~\refr{dss-form} goes to infinity,
thus our SL correlation correctly vanishes.
The correlation self-interaction error is corrected using $\tau_\sigma$
variable (the kinetic energy density), see~\refr{baa-inhom}, similarly to other meta-GGA correlation functionals\cite{becke1988correlation,perdew1999accurate,zhao2006design}. As a result, in our
model the parallel-spin correlation vanishes for single-orbital spin-compensated densities, and
the total correlation energy is zero for hydrogen atom.
The above-mentioned constraints are merely formal prerequisites for a high-quality
approximation to $E_\mathrm{C}$. A decent approximate model should also capture the physics of molecular
systems. Our model reflects the following physical properties:
\begin{enumerate}
\item Short-range electronic correlation is modeled by an expression
borrowed from the homogeneous electron gas, which is also appropriate
for real
systems.\cite{burke1994local,burke1998semilocal,henderson2004short,cancio2000exchange}
(See Eqs~\ref{bab-heg}, \ref{baa-heg}, and~\ref{baa-inhom}.) To
the best of our knowledge, we present the first beyond-LDA
functional which incorporates analytic
representation of the
short-range correlation function of the HEG developed
by~\citet{gori2001short}.
\item Long-range behavior of the correlation hole is governed by
$d_{\sigma\sigma'}$ function (see Eqs~\ref{opphole-def},
\ref{parhole-def}, and~\ref{dss-form}), which depends on both
density and its gradient at a reference point. This function accounts for damping
effect of density inhomogeneity ($\nabla\rho_\sigma$) on the correlation
hole. Furthermore, the $d_{\sigma\sigma'}$ function contains a free
parameter, $G$. It is used in tuning
the spatial range separation of the
correlation hole to properly blend with the long-range correlation correction.
\item Our model closely approximates
the exact correlation in the HEG regime at metallic densities. (See also
the discussion below~\refr{dss-form}.)
\end{enumerate}
To make our concept of stitching SL and NL correlation more transparent,
we briefly discuss it in the context of range-separated approach
of~\citet{kohn1998van}. It is possible to solve the dispersion problem
within DFT by partitioning the interelectron repulsion, $1/r$, into short-range
part, $\exp\left(-\mu r\right)/r$, and its long-range
complement.\cite{kohn1998van} ($\mu$ is a constant.) Both short-range exchange and correlation
are then treated at (semi)local level, and the
contributions originating from the long-range interaction are
approximated by a formula that is consistent with the
Casimir-Polder expression in the asymptotic region. Our treatment follows the same general
idea. The difference is as follows: the exponential factor that damps the interelectronic
interaction, $\exp\left(-\mu r \right)$, is
replaced by the $\exp\left(- d_{\sigma\sigma'} r \right)$ function of
Eqs~\ref{opphole-def}--\ref{parhole-def} which damps the short range
expansion of the approximate correlation holes. Thus, the $\mu$
constant is generalized into $d_{\sigma\sigma'}$ function, which depends on
density and its gradient at a reference point.
\section{Numerical results}
Our aim was to validate the correlation functional presented in this
work, preferably without the interference from the errors of
an exchange approximation. Therefore, we decided to perform
calculations using our correlation combined with full HF-like exchange,
and to compare it with other DFAs involving full HF-like
exchange. Although a general-purpose approximation cannot be formed by
combining semilocal DFT correlation with full exact exchange, it is a
demanding and useful test for a correlation functional. If the correlation
functional performs well with large portion of the exact exchange,
then there is much room for adjusting the exchange part of a global hybrid or
a range-separated hybrid exchange-correlation functional.
All DFT and HF calculations presented below were performed in
aug-cc-pVTZ basis. All energies are obtained from self-consistent calculations. Table~\ref{s22-results}
contains interaction energies for S22 set of
molecules.\cite{jurecka2006benchmark} The reference energies, $E_\mathrm{ref}$, are taken from
Ref.~\citenum{podeszwa2010improved}. $E_\mathrm{int}$ denotes
interaction energy calculated using the correlation functional described in this
work combined with $100\%$ HF-like exchange and DFT-D3
correction. All energies, as well as mean signed errors (MSE) and mean
unsigned errors (MUE) are given in kcal/mol. Mean absolute percentage
errors (MAPE) are given in percent. Our results (MUE=$0.46$ kcal/mol) compare rather favorably to
the other methods utilizing full HF-like exchange,
VV09\cite{vydrov2010implementation} (MUE=$0.90$ kcal/mol), M06HF\cite{zhao2008m06,goerigk2011thorough}
(MUE=$0.62$ kcal/mol), and M06HF-D3\cite{zhao2008m06,goerigk2011thorough} (MUE=$0.84$ kcal/mol). The
dispersion-free interaction energies are always significantly below the values
that would be obtained if the dispersion term as defined in SAPT was subtracted,
see supplementary information in~Ref.~\citenum{pernal2009dispersionless} and~Ref.~\citenum{modrzejewski2012dispersion}. This fact suggests that in our
model, at equilibrium distances, a large fraction of the dispersion
interaction is treated as short-range and accounted for by the SL
functional.
We further evaluate the performance of our approximation on the set of systems from nonbonded interaction
database of Zhao and Truhlar.\cite{zhao2005design,zhao2005benchmark} This
database gathers interacting dimers in subsets according to the dominant
character of the interaction: dispersion-dominated (WI7/05 and PPS5/05 subsets),
dipole interaction (DI6/04 subset), hydrogen-bonded (HB6/04), and charge transfer (CT7/04).
The results are presented in Tables~\ref{wi-results}, \ref{pps-results}, \ref{di-results},
\ref{hb-results}, and \ref{ct-results}, respectively.
The reference energies ($E_\mathrm{ref}$) are calculated at
CCSD(T)/mb-aug-cc-pVTZ level, see Ref.~\citenum{pernal2009dispersionless}. We compare
our approximation ($E_\mathrm{int}$) with M06HF functional\cite{zhao2008m06} ($E_\mathrm{M06HF}$) which combines
empirically-parametrized meta-GGA correlation with full HF-like exchange. As expected, our model
predicts interaction energies more accurately in cases where the dispersion interaction
dominates, see Tables~\ref{wi-results} and~\ref{pps-results}. In case of hydrogen bonded complexes,
Table~\ref{hb-results}, MAPE of either functional is close to $5\%$. Larger errors are present in DI6/04
and CT7/04 subsets. Although our approximation performs better that M06HF in case of DI6/04 dimers, the
error is rather large. In this case, as is seen in Table~\ref{di-results}, both functionals underestimate the interaction
strength and their errors are correlated. This fact suggests that the contribution coming from full HF-like
exchange is too repulsive, which cannot be counterbalanced by semilocal DFT correlation. Both functionals display
largest errors in charge-transfer complexes. Our approximation underestimates interaction for every CT complex.
This behavior to a large degree results from huge errors of the HF theory itself, see $E_\mathrm{HF}$ column in Table~\ref{ct-results}. As explained by~\citet{cohen2012challenges} this problem can be traced to the localization error of the HF theory, which makes electrons
excessively localized on the monomers. This error manifests itself as a concave curve of energy vs. fractional number of electrons, $E(N)$.\cite{cohen2012challenges} It is also known that pure semilocal DFT approximations
give convex $E(N)$.\cite{cohen2012challenges} See Ref.~\citenum{modrzejewski2012dispersion} for the relevant discussion of \ce{NH3\bond{...}ClF} dimer. Therefore, adding some amount of semilocal exchange
to our approximation should make the $E(N)$ dependence more linear and make the exchange contribution in CT interactions less repulsive, the step which will be undertaken in the future.
\begin{table}[p]
\caption{Interaction energies in S22 set (kcal/mol).}
\label{s22-results}
{\tiny
\begin{tabular}{lrrr}
\hline \hline
Dimer & $E_\mathrm{ref}$ & $E_\mathrm{int}$ & $E_\mathrm{dispfree}$ \\
\hline
\bf{Hydrogen-bonded} & & & \\
\ce{(NH3)2} & -3.145 & -2.75 & -2.17 \\
\ce{(H2O)2} & -5.004 & -4.86 & -4.41 \\
Formic acid dimer & -18.751 & -20.17 & -18.75 \\
Formamide dimer & -16.063 & -16.46 & -14.86 \\
Uracil dimer planar ($C_{2h}$) & -20.643 & -21.30 & -19.10 \\
2-pyridone $\cdot$ 2-aminopyridine & -16.938 & -16.33 & -13.67 \\
Adenine $\cdot$ thymine WC & -16.554 & -16.15 & -13.24 \\
\hline
MSE & -0.13 & & \\
MUE & 0.58 & & \\
MAPE & 5.0 & & \\
& & & \\
\bf{Predominant dispersion} & & & \\
\ce{(CH4)2} & -0.529 & -0.60 & 0.14 \\
\ce{(C2H4)2} & -1.482 & -1.52 & -0.15 \\
Benzene $\cdot$ \ce{CH4} & -1.448 & -1.45 & 0.10 \\
Benzene dimer parallel-displaced ($C_{2h}$) & -2.655 & -1.76 & 2.55 \\
Pyrazine dimer & -4.256 & -3.36 & 0.99 \\
Uracil dimer stacked ($C_2$) & -9.783 & -9.97 & -3.63 \\
Indole $\cdot$ benzene stacked & -4.523 & -3.25 & 2.73 \\
Adenine $\cdot$ thymine stacked & -11.857 & -11.63 & -3.10 \\
\hline
MSE & 0.37 & & \\
MUE & 0.45 & & \\
MAPE & 13 & & \\
& & & \\
{\bf Mixed interaction} & & & \\
Ethene $\cdot$ ethyne & -1.503 & -1.63 & -0.91 \\
Benzene \ce{H2O} & -3.280 & -3.78 & -2.19 \\
Benzene \ce{NH3} & -2.319 & -2.55 & -0.92 \\
Benzene \ce{HCN} & -4.540 & -5.68 & -4.02 \\
Benzene dimer T-shaped ($C_{2v}$) & -2.717 & -2.72 & -0.18 \\
Indole $\cdot$ benzene T-shaped & -5.627 & -5.89 & -2.45 \\
Phenol dimer & -7.097 & -6.87 & -3.96 \\
\hline
MSE & -0.29 & & \\
MUE & 0.36 & & \\
MAPE & 9.5 & & \\
& & & \\
MSE (total) & 0.002 & & \\
MUE (total) & 0.46 & & \\
MAPE (total) & 9.3 & & \\
\hline \hline
\end{tabular}
}
\end{table}
\begin{table}[p]
\caption{Interaction energies in WI7/05 set (kcal/mol).}
\label{wi-results}
\begin{tabular}{lrrr}
\hline \hline
Dimer & $E_\mathrm{ref}$ & $E_\mathrm{int}$ & $E_\mathrm{M06HF}$ \\
\hline
\ce{He\bond{...}Ne} & -0.041 & -0.037 & -0.13 \\
\ce{He\bond{...}Ar} & -0.058 & -0.045 & -0.085 \\
\ce{Ne\bond{...}Ne} & -0.086 & -0.064 & -0.13 \\
\ce{Ne\bond{...}Ar} & -0.13 & -0.07 & -0.15 \\
\ce{CH4\bond{...}Ne} & -0.18 & -0.18 & -0.20 \\
\ce{C6H6\bond{...}Ne} & -0.41 & -0.53 & -0.66 \\
\ce{CH4\bond{...}CH4} & -0.53 & -0.59 & -0.12 \\
\hline
MSE & & -0.01 & -0.006 \\
MUE & & 0.04 & 0.12 \\
MAPE & & 21 & 68 \\
\hline \hline
\end{tabular}
\end{table}
\begin{table}[p]
\caption{Interaction energies in PPS5/05 set (kcal/mol).}
\label{pps-results}
\begin{tabular}{lrrr}
\hline \hline
Dimer & $E_\mathrm{ref}$ & $E_\mathrm{int}$ & $E_\mathrm{M06HF}$ \\
\hline
\ce{(C2H2)2} & -1.36 & -1.47 & -1.06 \\
\ce{(C2H4)2} & -1.44 & -1.52 & -0.95 \\
Sandwich \ce{(C6H6)2} & -1.65 & -0.95 & 0.48 \\
T-shaped \ce{(C6H6)2} & -2.63 & -2.78 & -1.95 \\
Displaced \ce{(C6H6)2} & -2.59 & -2.06 & -0.94 \\
\hline
MSE & & 0.18 & 1.0 \\
MUE & & 0.31 & 1.0 \\
MAPE & & 16 & 55 \\
\hline \hline
\end{tabular}
\end{table}
\begin{table}[p]
\caption{Interaction energies in DI6/04 set (kcal/mol).}
\label{di-results}
\begin{tabular}{lrrr}
\hline \hline
Dimer & $E_\mathrm{ref}$ & $E_\mathrm{int}$ & $E_\mathrm{M06HF}$ \\
\hline
\ce{H2S\bond{...}H2S} & -1.62 & -1.20 & -0.82 \\
\ce{HCl\bond{...}HCl} & -1.91 & -1.40 & -0.99 \\
\ce{HCl\bond{...}H2S} & -3.26 & -2.74 & -2.48 \\
\ce{CH3Cl\bond{...}HCl} & -3.39 & -2.77 & -2.72 \\
\ce{CH3SH\bond{...}HCN} & -3.58 & -3.70 & -3.50 \\
\ce{CH3SH\bond{...}HCl} & -4.74 & -4.13 & -4.27 \\
\hline
MSE & & 0.43 & 0.62 \\
MUE & & 0.47 & 0.62 \\
MAPE & & 17 & 26 \\
\hline \hline
\end{tabular}
\end{table}
\begin{table}[p]
\caption{Interaction energies in HB6/04 set (kcal/mol).}
\label{hb-results}
\begin{tabular}{lrrr}
\hline \hline
Dimer & $E_\mathrm{ref}$ & $E_\mathrm{int}$ & $E_\mathrm{M06HF}$ \\
\hline
\ce{NH3\bond{...}NH3} & -3.09 & -2.82 & -2.53 \\
\ce{HF\bond{...}HF} & -4.49 & -4.63 & -4.27 \\
\ce{H2O\bond{...}H2O} & -4.91 & -4.90 & -4.72 \\
\ce{NH3\bond{...}H2O} & -6.38 & -6.28 & -6.35 \\
\ce{(HCONH2)2} & -15.41 & -16.39 & -15.72 \\
\ce{(HCOOH)2} & -17.60 & -19.57 & -19.33 \\
\hline
MSE & & -0.45 & 0.28 \\
MUE & & 0.58 & 0.30 \\
MAPE & & 5.2 & 4.6 \\
\hline \hline
\end{tabular}
\end{table}
\begin{table}[p]
\caption{Interaction energies CT7/04 (kcal/mol).}
\label{ct-results}
\begin{tabular}{lrrrr}
\hline \hline
Dimer & $E_\mathrm{ref}$ & $E_\mathrm{int}$ & $E_\mathrm{M06HF}$ & $E_\mathrm{HF}$ \\
\hline
\ce{C2H4\bond{...}F2} & -1.06 & -0.33 & -0.67 & 0.71 \\
\ce{NH3\bond{...}F2} & -1.80 & -0.74 & -0.90 & 0.19 \\
\ce{C2H2\bond{...}ClF} & -3.79 & -2.98 & -4.18 & -0.16 \\
\ce{HCN\bond{...}ClF} & -4.80 & -3.70 & -4.02 & -2.10 \\
\ce{NH3\bond{...}Cl2} & -4.85 & -3.50 & -4.00 & -1.12 \\
\ce{H2O\bond{...}ClF} & -5.20 & -4.69 & -5.26 & -2.91 \\
\ce{NH3\bond{...}ClF} & -11.17 & -10.53 & -11.92 & -5.49 \\
\hline
MSE & & 0.89 & 0.24 & \\
MUE & & 0.89 & 0.59 & \\
MAPE & & 31 & 20 & \\
\hline \hline
\end{tabular}
\end{table}
\section{Conclusions}
This paper presents a novel form of an SL correlation functional belonging to the meta-GGA rung that may be combined in an optimal way with the dispersion interaction component, either in the DFT+D manner or by incorporating a nonlocal potential. The important feature is that it is based on the first principles, in the form of a number of physical constraints imposed during the derivation. With minimal empiricism, our approximation is adjusted to a desired long-range dispersion correction by optimizing only a single empirical parameter. The parameter has a clear physical meaning: it governs the decay of the approximate correlation hole. Consequently, the correlation hole vanishes exponentially at large inter-electronic distances, which prevents double counting of the electron correlation effect that is already included when adding the long-range dispersion correction. An important and unique facet of our functional is that the adjustment of the empirical range-separation parameter has \emph{not} relaxed any of the physical constraints on which our model is based. The electron correlation is approximated by utilizing several numerical and analytical results of the HEG model. Most importantly, the HEG approximation to the short-range part of the correlation hole is rigorously conserved for arbitrary systems (only the self-interaction pertinent to the HEG model is removed from the parallel-spin correlation hole).
While our new correlation functional can be combined with any of non-local dispersion models, for preliminary calculations of this paper, we employed the atom pairwise additive DFT-D3 dispersion correction. Given the fact that our correlation functional is combined with 100\% HF exchange -- far from an optimal choice in general case -- the numerical results are very encouraging. For the interaction energies of hydrogen-bonded complexes, the accuracy is on a par with that obtained with the M06HF functional, which is a highly parametrized empirical approximation containing full HF exchange. For dispersion-dominated complexes, the predictions of our model compare favorably with VV09 and M06HF. The results in the subsets of dipole-interaction and charge-transfer complexes are less satisfactory, which is easily explained by the inadequacy of the full HF-like exchange component: indeed, the signed errors correlate with the signed errors of the HF method. Obviously, much improvement may be expected when a more appropriate exchange part will be incorporated. Development of an optimal range-separated hybrid exchange approximation, appropriate for our new correlation functional, as well as implementation of non-local van der Waals correlation functionals are underway in our laboratory.
\section{Acknowledgments}
This work was supported by the Polish Ministry of Science and Higher
Education, Grant~N~N204~248440, and by the National Science Foundation
(US), Grant No. CHE-1152474.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 2,458
|
I thought it might be fun to review this like I'd never heard of it, the director or the genre before – "well, this is an early entry from a Pittsburgh independent filmmaker called George Romero, in what's come to be known as a 'zombie' movie" but I'm too lazy to keep it up all the way through. You don't need me to tell you about "Dawn Of The Dead", right? You've seen it? If you haven't, then go away immediately and watch it. It's as good as horror films have ever been, rich imagery, great performances, a plot with real depth to it; but if you're a fan of the sort of films we cover here, then this should be part of your DNA. Books and books have been written about it, which puts it a little outside our wheelhouse, but of the million great things written about it, picking one at random, THIS is excellent.
Why I'm doing this relates to our recent coverage of the movies of Bruno Mattei and Claudio Fragasso, and it's one of those stories that involves incoherent sequel numbering, a matter close to my heart. I'm guessing due to some contractual loophole, or weirdness in Italian copyright law, they started making sequels to this movie almost right away, only sequels with no returning cast or crew. Lucio Fulci, the legendary director of "The Beyond", "House By The Cemetery" and "New York Ripper", made part 2, also known in the UK as "Zombie Flesh Eaters" (all the sequels to that were part of the "Zombie Flesh Eaters" series in the UK).
By part 3, all bets were off. Fulci made most of "Zombi 3", but it was finished off by Mattei and Fragasso due to Fulci's failing health; but the most amazing thing is the sheer number of different movies that were released as "Zombi 3" in various parts of the world. "Nightmare City", "Let Sleeping Corpses Lie", "Zombie Holocaust", "Nights Of Terror / The Zombie Dead" and "The Hanging Woman" have all been subjected to it – "Nights Of Terror" is one of my favourite zombie movies ever, by the way – but they calmed down a bit by parts 4 and 5, neither of which bear any relation to the rest of the series or each other. Oh, and part 5 was made before part 4. Part 4 was also directed by Fragasso, but more on them when I get to reviewing them.
So, we'll be looking forward to Mattei and Fragasso's section of this franchise, and there's every chance that part 2 will be decent, as Fulci made some horror classics too. But we're here to talk about "Zombi". Firstly is why it's called that. Although I think we in the UK got Romero's version, with a few cuts for the more extreme gore, the rest of Europe got a version edited by Dario Argento, with a soundtrack comprised mostly of songs from his band, Goblin. Argento part-financed the movie, and acted as script editor, on the proviso he could re-edit the movie for release in the rest of the world, and his version ended up 9 minutes shorter, at 118 minutes.
If you'd like to read an extremely detailed breakdown of every difference in the versions (and talk of the "ultimate edition", which has all the footage from all the different versions and stuff from an edit that Romero prepared hurriedly for the Cannes Film Festival) then please go HERE, but if you don't, then I'll give you the highlights. Most of Argento's edits were to trim the odd second of fat from various scenes, and to remove some of the more overt comedy (the biker gang still have plenty of funny stuff to do, though). The guy who gets his head chopped off by the helicopter blades is absent from Argento's version, perhaps because he never liked the effect, and a few conversations are removed.
What's interesting, not so much the big stuff, which is a few light-hearted conversations, but the little things. There are hundreds of edits, a second here, a second there, and for a movie which was over 2 hours, I think – and this may be sacrilege to some people – Argento was right. His edit is fantastic, stripping fat from scenes and focusing it better; the original, and even the much longer version, are both masterpieces of cinema, but Argento's version might just be the best of the lot.
I think this depends on your attitude. I used to be "longer = better" when it came to director's cuts of my favourite movies, but I was cured of this when the "Redux" version of "Apocalypse Now" was released. I remember the acres of press coverage, the delight from movie fans that we were finally going to see Coppola's vision in full…and it ended up being unbearably dull. That plantation scene! Ye gods. So, since then, I've come to appreciate the work of a good editor, and there are very few films released today that wouldn't benefit from being 20 minutes shorter. It's still fun to see the extra stuff from your favourite films, but the number of deleted scenes that deserve to be put back in films is absolutely miniscule.
Well, that's a brief chat about "Zombi". Plot mockery and insulting cast and crew – the normal business of this site – will resume with "Zombi 2".
This entry was posted in Reviews and tagged all time classic, Dario Argento, George Romero, zombie movie. Bookmark the permalink.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 1,880
|
Q: What kind of encryption or packing method is used for the provided JavaScript code? I have placed the packed/encrypted JavaScript code presented below in an html in between script tags:
console.log(function(p, a, c, k, e, d) {
while (c--) {
if (k[c]) {
p = p.replace(new RegExp("\\b" + c + "\\b", "g"), k[c]);
}
}
return p;
}("23(16=(15='࠵ϾĹà{࢙Јà࠵Į26ࢤԩçòè࠾Ӑ࢚ϩƤ{⟭Ԫèþ⟬Љçú~✥Ԫðþ⟭ϪƛÝÓࡀӆóÌ࠾ӆƺÓࡊӆĮþࡀҼŃł⟭ϵƛÜÒࢭӆçþ⟭ϴêú~⟭Ԫêý⟭ϾƛÍÓࡉϴƒÖ{ࡉһèþË⟤ьƒú×⟢юçá⟬ҲƚõÒ⟘ӄĶýÉ⟬ҳƚôÎࢣӄêÌ⟢ӐƒÊ{⫺ϩçþéࡀӆçþ⟭јñú~⟭Ԫêþ⟭љƛÓࢤϼƒÖ{Ԫçü࢙ϫóĮÊ࠾ӆƺÓࢮӆĺÊÌࡈѠƺÓ✛ӆðþ⟭їĬúӅŘü࠶юçþ}⟢ҼƒĮÌࡉһéþ{⟭эƛÕ⟚ϳƛÞÓ✚ӆðþ⟚ϼƚÓࢮӆĮþ~⟭ϴƛ×ÎࢭӐƤĮËࡈӅêþ{⟭їƛÌ⟢ӐƒÊ|⫺ϩçþ}⟢ϳƛÌ✑ϪèÉÓࡊҼƤĮË⟤юƔÔÒ࠾ӆêþ⟭ϫƛÎࢮӅéý⟬эƔÔÒࡉӆńþ⟭ϴƛÓÓ✚ӆôþ⟭ЉƛÜÎࢮӆéþ⟭ЈƔÓÓ✐ӆçü⟭ьƛÊÓࢤӆŊþ⟭яƛÔÎࢮӅéý⟬ѡƚÜÎࢮӆĬþ⟭ѡƔÓÓࢢӆĺþ⟢ҼƤúÈ⭧ϲƤ7ࡈϾéĮ1ࡈѠƺÓࢮӆķþ⟭ЉĭúÓࡀҼĸ⟚Ϫƛ×Ì⟢Ͼƹ×࠶Ϫƛł}⟭ЉƛÞÓࢬЇƒ8}✚Ϫƹ').11,13=2,10='',9=0;9<16;9++)10+=14.27(24(15[9].25(0),!9||9%5?++13:13=3,21(18).11,(19(){20 22.28,40,48,46.44(49),43.45 47 42,41,33,14().32().31(),29 30,34,35.11,39,38,37})()));36(10)", 10, 50, "|z|||e||B|xz|x|i|s|length|g|d|String|uN|len|zz|false|function|return|uneval|Number|for|parseInt|charCodeAt|xy|fromCharCode|isInteger|typeof|Date|toUpperCase|toLowerCase|NaN|RegExp|Array|eval|true|undefined|null|Error|Boolean|File|Folder|ceil|Desktop|Math|instanceof|Object|Infinity".split("|")));
and when analyzed it in the Firefox console it outputs:
for(len=(uN='࠵ϾĹà{࢙Јà��࠵Į�xyࢤԩçòè࠾Ӑ���࢚ϩƤ�{⟭Ԫèþ�⟬Љçú~✥Ԫðþ�⟭ϪƛÝÓࡀӆó�Ì࠾ӆƺ�ÓࡊӆĮþ�ࡀҼŃł�⟭ϵƛÜÒࢭӆçþ�⟭ϴêú~⟭Ԫêý�⟭ϾƛÍÓæúÌ⫺ӐƤ8}✚Ϫƹ').11,13=2,10='',9=0;9<16;9++)10+=14.27(24(15[9].25(0),!9||9%5?++13:13=3,21(18).11,(19(){20 22.28,40,48,46.44(49),43.45 47 42,41,33,14().32().31(),29 30,34,35.11,39,38,37})()));36(10)", 10, 50, "|z|||e||B|xz|x|i|s|length|g|d|String|uN|len|zz|false|function|return|uneval|Number|for|parseInt|charCodeAt|xy|fromCharCode|isInteger|typeof|Date|toUpperCase|toLowerCase|NaN|RegExp|Array|eval|true|undefined|null|Error|Boolean|File|Folder|ceil|Desktop|Math|instanceof|Object|Infinity".split("|")));
I have changed eval(function(p, a, c, k, e, d) to console.log(function(p, a, c, k, e, d) so the code can be unpacked(decrypted?), but all I see is a next puzzle.
I have tried various unpackers and deobfuscators for the second piece of code but without success.
Will someone point me of what packing/encryption is used for the provided code?
P.S. The encrypted code is too, so I have included only a part of it with the readable content at the beginning and the end.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 447
|
Q: How to start and stop dictation in Apple Watch witout pressing button I wrote code to use dictation on my apple watch. I used presentTextInputControllerWithSuggestions without suggestions to directly start dictation.
But, I have two problem :
*
*I want to start dictation when my app starts. For this, I call my function in the willActivate method but with this, just a waiting image appears in my screen, not my first page with dictation.
*I want to stop dictation without press "Done" button. I don't know if it's possible and how can I make this.
There is my code :
func dictation(){
self.presentTextInputControllerWithSuggestions([], allowedInputMode: WKTextInputMode.Plain, completion:{
(results) -> Void in
//myCode
})
}
override func willActivate(){
super.willActivate()
dictation()
}
Do you have solutions ?
A: Thanks for your help @Feldur
I tried with delay and it seems to work
There is my code :
override init(){
super.init()
print("start init")
let seconds = 1.0
let delay = seconds * Double(NSEC_PER_SEC) // nanoseconds per seconds
let dispatchTime = dispatch_time(DISPATCH_TIME_NOW, Int64(delay))
dispatch_after(dispatchTime, dispatch_get_main_queue(), {
self.dictation()
})
print("end init")
}
There are my logs :
start init
end init
start awakeWithContext
end awakeWithContext
start willactivate
end willactivate
start didAppear
end didAppear
start dictation
My screen appears and after, my dictation starts.
Do you have an idea for stop dictation when user stops speaking ?
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 1,691
|
ACCEPTED
#### According to
Index Fungorum
#### Published in
Jber. schles. Ges. vaterl. Kultur 42: 117 (1875)
#### Original name
Puccinia pedunculata J. Schröt.
### Remarks
null
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 23
|
Video on the Hippie Ward
Posted on September 20, 2017 by Reps Admin
Televising Psyche: Therapy, Play, and the Seduction of Video
by Carmine Grimaldi
In 1967, the San Francisco Chronicle ran the panicked headline "HIPPIES WARN CITY—100,000 WILL INVADE HAIGHT ASHBURY THIS SUMMER." With the specter of homelessness, disease, addiction, and moral depravity looming, the city scrambled for some response: that summer, the San Francisco Assembly Committee on Public Health held a series of hearings to determine who exactly these new residents were, and how the anticipated crisis could be stanched.
Sitting before the committee, Dr. Ernest Dernburg, the director of psychiatric services at the Haight-Ashbury Free Medical Clinic, offered his professional opinion on the moral character of the hippie. He was working on the front lines of the crisis and would have been considered far more sympathetic to the counterculture than those working in traditional hospitals. And yet the portrait he drew was damning. The new generation, he explained to the committee, was "passive, withdrawn, emotionally unresponsive, drug dependent. They suffer . . . from massive psychological poverty." He then pointed to a potential culprit: "Keep in mind that this generation is the first to grow up in front of the television set. These children have been sitting passively before it, receiving stimulation from it, living mostly inside their heads, all during their period of development." This sentiment was hardly new; the moral panic about television has run in tandem with the medium's history. At the beginning of the decade, when many of Dernburg's patients were "growing up in front of the television," one widely read study warned that television "anesthetize[s] a person against pain and distress" and asked rhetorically, "In how many cases does television meet children's needs in the same way as alcohol or drugs might do so?" If one wanted to assign blame for the legions of white, middle-class kids who were dropping school and acid, television was as good a candidate as any.
Langley Porter Institute, San Francisco, 1966
Following Dernburg's rather alarmist evaluation, Dr. Harry Wilmer, then a professor of psychiatry at the University of California, San Francisco (UCSF), spoke before the committee. Like Dernburg, he had founded a new psychiatric clinic for hippie drug users, just twenty blocks from the Haight, at the Langley Porter Neuropsychiatric Institute. And like Dernburg, he agreed that television and mass media were largely responsible for the emergence of the counterculture. But unlike his colleague, who adhered to popular techniques of psychoanalysis and pharmacological remedies, Wilmer explained that he was developing a new approach that had only entered psychiatry in the last few years: it was a "pilot study" that sought to cure patients through the very medium that had originally harmed them—if used correctly, Wilmer explained, the television screen could be a potent medicine. In the next two years, he planned to use video feedback, made possible by videotape technology, to restructure the consciousness and sensorium of those who had fallen through the cracks of society. As he explained to the committee, "We are trying to reawaken the power to see and to interact." This Janus-faced power of television clearly caught the imagination of the public—in what was the first of many newspaper articles about Wilmer's project, the San Francisco Examiner's headline read, "How TV Produces and Heals a Drug Generation." Continue reading …
San Francisco's "Hippie Drug Ward," an experimental therapeutic clinic opened in 1967, sought to cure a wayward generation through an immersive multimedia environment. Examining archival records only recently made available, this paper explores the way the moving image—and in particular videotape—created a space in which style, affect, and psyche became commingled.
CARMINE GRIMALDI is a doctoral candidate in history at the University of Chicago and a fellow at Harvard's Film Study Center. His writings have appeared in the Atlantic and Filmmaker Magazine, and his films have screened widely at festivals in the United States and abroad.
This entry was posted in Essays and tagged Drug Therapy, Psychiatry, San Francisco, The Sixties, TV, Video, Wilmer by Reps Admin. Bookmark the permalink.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 1,835
|
\section{Introduction}
Spherical data is found in many applications (\figref{examples}).
Planetary data (such as meteorological or geological measurements) and brain activity are example of intrinsically spherical data.
The observation of the universe, LIDAR scans, and the digitalization of 3D objects are examples of projections due to observation.
Labels or variables are often to be inferred from them.
Examples are the inference of cosmological parameters from the distribution of mass in the universe \citep{perraudin2019deepspherecosmo}, the segmentation of omnidirectional images \citep{khasanova2017sphericalcnn}, and the segmentation of cyclones from Earth observation \citep{mudigonda2017climateevents}.
\begin{figure}[h]
\centering
\subfloat[]{
\includegraphics[height=0.20\linewidth,valign=t]{example_brain_meg}
} \hspace{0.1em}
\subfloat[]{
\includegraphics[height=0.20\linewidth,valign=t]{example_cosmo_cmb}
} \hfill
\subfloat[]{
\includegraphics[height=0.20\linewidth,valign=t]{example_climate_TMQ}
\label{fig:examples:climate}
} \hfill
\subfloat[]{
\includegraphics[height=0.20\linewidth,valign=t]{example_ghcn_daily_tmax}
\label{fig:examples:ghcn:tmax}
} \hfill
\subfloat[]{
\includegraphics[height=0.20\linewidth,valign=t]{example_ghcn_graph}
\label{fig:examples:ghcn:graph}
}
\caption{
Examples of spherical data:
(a) brain activity recorded through magnetoencephalography (MEG),\protect\footnotemark
(b) the cosmic microwave background (CMB) temperature from \citet{planck2015overview},
(c) hourly precipitation from a climate simulation \citep{jiang2019sphericalcnn},
(d) daily maximum temperature from the Global Historical Climatology Network (GHCN).\protect\footnotemark
A rigid full-sphere sampling is not ideal: brain activity is only measured on the scalp, the Milky Way's galactic plane masks observations, climate scientists desire a variable resolution, and the position of weather stations is arbitrary and changes over time.
(e) Graphs can faithfully and efficiently represent sampled spherical data by placing vertices where it matters.
}
\label{fig:examples}
\end{figure}
\footnotetext[1]{\scriptsize\url{https://martinos.org/mne/stable/auto_tutorials/plot_visualize_evoked.html}}
\footnotetext[2]{\scriptsize\url{https://www.ncdc.noaa.gov/ghcn-daily-description}}
As neural networks (NNs) have proved to be great tools for inference, variants have been developed to handle spherical data.
Exploiting the locally Euclidean property of the sphere, early attempts used standard 2D convolutions on a grid sampling of the sphere \citep{boomsma2017sphericalcnn, su2017sphericalcnn, coors2018sphericalcnn}.
While simple and efficient, those convolutions are not equivariant to rotations.
On the other side of this tradeoff, \citet{cohen2018sphericalcnn} and \citet{esteves2018sphericalcnn} proposed to perform proper spherical convolutions through the spherical harmonic transform.
While equivariant to rotations, those convolutions are expensive (\secref{method:convolution}).
As a lack of equivariance can penalize performance (\secref{exp:cosmo}) and expensive convolutions prohibit their application to some real-world problems, methods standing between these two extremes are desired.
\citet{cohen2019gauge} proposed to reduce costs by limiting the size of the representation of the symmetry group by projecting the data from the sphere to the icosahedron.
The distortions introduced by this projection might however hinder performance (\secref{exp:climate}).
Another approach is to represent the sampled sphere as a graph connecting pixels according to the distance between them \citep{bruna2013gnn, khasanova2017sphericalcnn, perraudin2019deepspherecosmo}.
While Laplacian-based graph convolutions are more efficient than spherical convolutions, they are not exactly equivariant \citep{defferrard2019deepsphereequiv}.
In this work, we argue that graph-based spherical CNNs strike an interesting balance, with a controllable tradeoff between cost and equivariance (which is linked to performance).
Experiments on multiple problems of practical interest show the competitiveness and flexibility of this approach.
\section{Method}
DeepSphere leverages graph convolutions to achieve the following properties: (i) computational efficiency, (ii) sampling flexibility, and (iii) rotation equivariance (\secref{equivariance}).
The main idea is to model the sampled sphere as a graph of connected pixels: the length of the shortest path between two pixels is an approximation of the geodesic distance between them.
We use the graph CNN formulation introduced in \citep{defferrard2016graphnn} and a pooling strategy that exploits hierarchical samplings of the sphere.
\paragraph{Sampling.}
A sampling scheme $\V = \{x_i \in \S^2\}_{i=1}^n$ is defined to be the discrete subset of the sphere containing the $n$ points where the values of the signals that we want to analyse are known. For a given continuous signal $f$, we represent such values in a vector $\b{f}\in\mathbb R^n$. As there is no analogue of uniform sampling on the sphere, many samplings have been proposed with different tradeoffs.
In this work, depending on the considered application, we will use the equiangular \citep{driscoll1994Fouriersphere}, HEALPix \citep{gorski2005healpix}, and icosahedral \citep{baumgardner1985icosahedral} samplings.
\paragraph{Graph.}
From $\V$, we construct a weighted undirected graph $\G = (\V, w)$, where the elements of $\V$ are the vertices and the weight $w_{ij} = w_{ji}$ is a similarity measure between vertices $x_i$ and $x_j$.
The combinatorial graph Laplacian $\b{L} \in \mathbb{R}^{n \times n}$ is defined as $\b{L} = \b{D} - \b{A}$, where $\b{A} = (w_{ij})$ is the weighted adjacency matrix, $\b{D} = (d_{ii})$ is the diagonal degree matrix, and $d_{ii} = \sum_j w_{ij}$ is the weighted degree of vertex $x_i$.
Given a sampling $\V$, usually fixed by the application or the available measurements, the freedom in constructing $\G$ is in setting $w$.
\Secref{equivariance} shows how to set $w$ to minimize the equivariance error.
\paragraph{Convolution.} \label{sec:method:convolution}
On Euclidean domains, convolutions are efficiently implemented by sliding a window in the signal domain.
On the sphere however, there is no straightforward way to implement a convolution in the signal domain due to non-uniform samplings.
Convolutions are most often performed in the spectral domain through a spherical harmonic transform (SHT).
That is the approach taken by \citet{cohen2018sphericalcnn} and \citet{esteves2018sphericalcnn}, which has a computational cost of $\mathcal{O}(n^{3/2})$ on isolatitude samplings (such as the HEALPix and equiangular samplings) and $\mathcal{O}(n^2)$ in general.
On the other hand, following \citet{defferrard2016graphnn}, graph convolutions can be defined as
\begin{equation} \label{eqn:graph_conv}
h(\b{L}) \b{f} = \left(\sum_{i=0}^P \alpha_i \b{L}^i\right) \b{f},
\end{equation}
where $P$ is the polynomial order (which corresponds to the filter's size) and $\alpha_i$ are the coefficients to be optimized during training.\footnote{In practice, training with Chebyshev polynomials (instead of monomials) is slightly more stable. We believe it to be due to their orthogonality and uniformity.}
Those convolutions are used by \citet{khasanova2017sphericalcnn} and \citet{perraudin2019deepspherecosmo} and cost $\mathcal{O}(n)$ operations through a recursive application of $\b{L}$.\footnote{As long as the graph is sparsified such that the number of edges, i.e., the number of non-zeros in $\b{A}$, is proportional to the number of vertices $n$. This can always be done as most weights are very small.}
\paragraph{Pooling.}
Down- and up-sampling is natural for hierarchical samplings,\footnote{The equiangular, HEALPix, and icosahedral samplings are of this kind.} where each subdivision divides a pixel in (an equal number of) child sub-pixels.
To pool (down-sample), the data supported on the sub-pixels is summarized by a permutation invariant function such as the maximum or the average.
To unpool (up-sample), the data supported on a pixel is copied to all its sub-pixels.
\paragraph{Architecture.}
All our NNs are fully convolutional, and employ a global average pooling (GAP) for rotation invariant tasks.
Graph convolutional layers are always followed by batch normalization and ReLU activation, except in the last layer.
Note that batch normalization and activation act on the elements of $\b{f}$ independently, and hence don't depend on the domain of $f$.
\section{Graph convolution and equivariance} \label{sec:equivariance}
While the graph framework offers great flexibility, its ability to faithfully represent the underlying sphere --- for graph convolutions to be rotation equivariant --- highly depends on the sampling locations and the graph construction.
\subsection{Problem formulation}
A continuous function $f : \mathcal C(\S^2) \supset F_\V \to \mathbb{R}$ is sampled as $T_\V(f) = \b{f}$ by the sampling operator $T_\V: C(\S^2) \supset F_\V \to \mathbb{R}^n$ defined as $\b{f}: f_i=f(x_i)$.
We require $F_\V$ to be a suitable subspace of continuous functions such that $T_\V$ is invertible, i.e., the function $f \in F_\V$ can be unambiguously reconstructed from its sampled values $\b{f}$.
The existence of such a subspace depends on the sampling $\V$ and its characterization is a common problem in signal processing \citep{driscoll1994Fouriersphere}.
For most samplings, it is not known if $F_\V$ exists and hence if $T_\V$ is invertible.
A special case is the equiangular sampling where a sampling theorem holds, and thus a closed-form of $T_\V^{-1}$ is known.
For samplings where no such sampling formula is available, we leverage the discrete SHT to reconstruct $f$ from $\b{f}=T_\V f$, thus approximating $T_\V^{-1}$.
For all theoretical considerations, we assume that $F_\V$ exists and $f \in F_\V$.
By definition, the (spherical) graph convolution is rotation equivariant if and only if it commutes with the rotation operator defined as $R(g), g\in SO(3)$: $R(g) f(x) = f\left(g^{-1} x \right)$.
In the context of this work, graph convolution is performed by recursive applications of the graph Laplacian \eqnref{graph_conv}.
Hence, if $R(g)$ commutes with $\b{L}$, then, by recursion, it will also commute with the convolution $h(\b{L})$.
As a result, $h(\b{L})$ is rotation equivariant if and only if
\begin{equation*}
\b{R}_\V(g) \b{L} \b{f} = \b{L} \b{R}_\V(g) \b{f}, \hspace{1cm} \forall f\in F_\V \text{ and } \forall g\in SO(3),
\end{equation*}
where $\b{R}_\V(g) = T_\V R(g) T_\V^{-1}$.
For an empirical evaluation of equivariance, we define the \textit{normalized equivariance error} for a signal $\b{f}$ and a rotation $g$ as
\begin{equation} \label{eq:equivariance error}
E_{\b{L}}(\b{f}, g) = \left(\frac{ \norm {\b{R}_\V(g) \b{L} \b{f} - \b{L} \b{R}_\V(g) \b{f}} }{\norm {\b{L} \b{f}}}\right)^2.
\end{equation}
More generally for a class of signals $f \in C \subset F_\V$, the \textit{mean equivariance error} defined as
\begin{equation} \label{eq:mean equivariance error}
\overline E_{\b{L}, C} = \mathbb E_{\b{f}\in C, g\in SO(3)} \ E_{\b{L}}(\b{f}, g)
\end{equation}
represents the overall equivariance error.
The expected value is obtained by averaging over a finite number of random functions and random rotations.
\subsection{Finding the optimal weighting scheme} \label{sec:optimal}
Considering the equiangular sampling and graphs where each vertex is connected to 4 neighbors (north, south, east, west), \citet{khasanova2017sphericalcnn} designed a weighting scheme to minimize \eqref{eq:mean equivariance error} for longitudinal and latitudinal rotations\footnote{Equivariance to longitudinal rotation is essentially given by the equiangular sampling.}.
Their solution gives weights inversely proportional to Euclidean distances:
\begin{equation} \label{eqn:weights:khasanova}
w_{ij} = \frac{1}{\norm{x_i-x_j}}.
\end{equation}
While the resulting convolution is not equivariant to the whole of $SO(3)$ (\figref{equivariance_error}), it is enough for omnidirectional imaging because, as gravity consistently orients the sphere, objects only rotate longitudinally or latitudinally.
To achieve equivariance to all rotations, we take inspiration from \citet{belkin2005towards}.
They prove that for a \emph{random uniform sampling}, the graph Laplacian $\b{L}$ built from weights
\begin{equation} \label{eqn:weights:belkin}
w_{ij} = e^{- \frac{1}{4t} \norm{x_i - x_j}^2}
\end{equation}
converges to the Laplace-Beltrami operator $\Delta_{\mathbb{S}^2}$ as the number of samples grows to infinity.
This result is a good starting point as $\Delta_{\mathbb{S}^2}$ commutes with rotation, i.e., $\Delta_{\mathbb{S}^2}R(g) = R(g)\Delta_{\mathbb{S}^2}$.
While the weighting scheme is full (i.e., every vertex is connected to every other vertex), most weights are small due to the exponential.
We hence make an approximation to limit the cost of the convolution \eqnref{graph_conv} by only considering the $k$ nearest neighbors ($k$-NN) of each vertex.
Given $k$, the optimal kernel width $t$ is found by searching for the minimizer of \eqref{eq:mean equivariance error}.
\Figref{kernel_widths} shows the optimal kernel widths found for various resolutions of the HEALPix sampling.
As predicted by the theory, $t_n \propto n^\beta, \beta \in \mathbb{R}$.
Importantly however, the optimal $t$ also depends on the number of neighbors $k$.
Considering the HEALPix sampling, \citet{perraudin2019deepspherecosmo} connected each vertex to their 8 adjacent vertices in the tiling of the sphere, computed the weights with \eqnref{weights:belkin}, and heuristically set $t$ to half the average squared Euclidean distance between connected vertices.
This heuristic however over-estimates $t$ (\figref{kernel_widths}) and leads to an increased equivariance error (\figref{equivariance_error}).
\begin{figure}
\begin{minipage}{0.6\linewidth}
\centering
\includegraphics[width=\linewidth]{equivariance_error}
\caption{
Mean equivariance error \eqref{eq:mean equivariance error}.
There is a clear tradeoff between equivariance and computational cost, governed by the number of vertices $n$ and edges $kn$.
}
\label{fig:equivariance_error}
\end{minipage}
\hfill
\begin{minipage}{0.35\linewidth}
\centering
\includegraphics[width=\linewidth]{kernel_widths}
\caption{Kernel widths.}
\label{fig:kernel_widths}
\vspace{1em}
\includegraphics[height=4.5em]{lamp_000018}
\hfill
\includegraphics[height=4.5em]{lamp_000018_sphere_nobar}
\caption{3D object represented as a spherical depth map.}
\label{fig:depthmap}
\vspace{1em}
\includegraphics[width=\linewidth]{spectrum}
\caption{Power spectral densities.}
\label{fig:spectrum}
\end{minipage}
\end{figure}
\subsection{Analysis of the proposed weighting scheme}
We analyze the proposed weighting scheme both theoretically and empirically.
\paragraph{Theoretical convergence.}
\begin{wrapfigure}[10]{r}{3cm}
\centering
\includegraphics[width=0.9\linewidth]{patch}
\caption{Patch.}
\label{fig:patch}
\end{wrapfigure}
We extend the work of \citep{belkin2005towards} to a sufficiently regular, deterministic sampling.
Following their setting, we work with the \emph{extended graph Laplacian} operator as the linear operator $L_n^t: L^{2}(\S^2) \rightarrow L^{2}(\S^2)$ such that
\begin{equation} \label{eq:Heat Kernel Graph Laplacian operator}
L_n^t f(y) := \frac{1}{n}\sum_{i=1}^{n} e^{ -\frac{\|x_i-y\|^2}{4t}} \left(f(y)-f(x_i)\right).
\end{equation}
This operator extends the graph Laplacian with the weighting scheme \eqnref{weights:belkin} to each point of the sphere (i.e., $\b{L}_n^t \b{f} = T_\V L_n^t f$).
As the radius of the kernel $t$ will be adapted to the number of samples, we scale the operator as $\hat{L}_n^t := |\S^2| (4\pi t^2)^{-1} L_n^t$.
Given a sampling $\V$, we define $\sigma_i$ to be the patch of the surface of the sphere corresponding to $x_i$, $A_i$ its corresponding area, and $d_i$ the largest distance between the center $x_i$ and any point on the surface $\sigma_i$.
Define $d^{(n)} := \max_{i=1, \dots, n} d_i$ and $A^{(n)} := \max_{i=1, \dots, n} A_i$.
\begin{theorem}
For a sampling $\V$ of the sphere that is equi-area and such that $d^{(n)} \leq \frac{C}{n^\alpha}, \ \alpha\in (0,\linefrac{1}{2}]$, for all $f: \S^2 \rightarrow \mathbb{R}$ Lipschitz with respect to the Euclidean distance in $\mathbb{R}^3$, for all $y\in\S^2$, there exists a sequence $t_n = n^\beta$, $\beta\in\mathbb R$ such that
\begin{equation*}
\lim_{n\to\infty}\hat{L}_n^{t_n}f(y) = \Delta_{\mathbb S^2} f(y).
\end{equation*}
\label{theo:pointwise convergence for a regular sampling}
\end{theorem}
This is a major step towards equivariance, as the Laplace-Beltrami operator commutes with rotation.
Based on this property, we show the equivariance of the scaled extended graph Laplacian.
\begin{theorem}\label{theo:equivariance}
Under the hypothesis of theorem~\ref{theo:pointwise convergence for a regular sampling}, the scaled graph Laplacian commutes with any rotation, in the limit of infinite sampling, i.e.,
\begin{equation*}
\forall y\in\mathbb S^2\quad\left| R(g) \hat L_n^{t_n} f (y) - \hat L_n^{t_n} R(g) f(y) \right| \xrightarrow{n\to\infty}0.
\end{equation*}
\end{theorem}
From this theorem, it follows that the discrete graph Laplacian will be equivariant in the limit of $n \rightarrow \infty$ as by construction $\b{L}_n^t \b{f} = T_\V L_n^t f$ and as the scaling does not affect the equivariance property of $\b{L}_n^t$.
Importantly, the proof of Theorem~\ref{theo:pointwise convergence for a regular sampling} (in Appendix~\ref{sec: appendix: proof of theorem}) inspires our construction of the graph Laplacian.
In particular, it tells us that $t$ should scale as $n^\beta$, which has been empirically verified (\figref{kernel_widths}).
Nevertheless, it is important to keep in mind the limits of Theorem \ref{theo:pointwise convergence for a regular sampling} and \ref{theo:equivariance}.
Both theorems present asymptotic results, but in practice we will always work with finite samplings.
Furthermore, since this method is based on the capability of the eigenvectors of the graph Laplacian to approximate the spherical harmonics,
a stronger type of convergence of the graph Laplacian would be preferable, i.e., spectral convergence (that is proved for a full graph in the case of random sampling for a class of Lipschitz functions in \citep{belkin2007convergence}).
Finally, while we do not have a formal proof for it, we strongly believe that the HEALPix sampling does satisfy the hypothesis $d^{(n)}\leq \frac{C}{n^\alpha}, \ \alpha\in (0,\linefrac{1}{2}]$, with $\alpha$ very close or equal to $\linefrac{1}{2}$.
The empirical results discussed in the next paragraph also points in this direction.
This is further discussed in Appendix \ref{sec: appendix: proof of theorem}.
\paragraph{Empirical convergence.}
\Figref{equivariance_error} shows the equivariance error \eqref{eq:mean equivariance error} for different parameter sets of DeepSphere for the HEALPix sampling as well as for the graph construction of \citet{khasanova2017sphericalcnn} for the equiangular sampling.
The error is estimated as a function of the sampling resolution and signal frequency.
The resolution is controlled by the number of pixels $n = 12N_{side}^2$ for HEALPix and $n = 4b^2$ for the equiangular sampling.
The frequency is controlled by setting the set $C$ to functions $f$ made of spherical harmonics of a single degree $\ell$.
To allow for an almost perfect implementation (up to numerical errors) of the operator $\b{R}_\V$, the degree $\ell$ was chosen in the range $(0, 3N_{side}-1)$ for HEALPix and $(0, b)$ for the equiangular sampling \citep{gorski1999healpixprimer}.
Using these parameters, the measured error is mostly due to imperfections in the empirical approximation of the Laplace-Beltrami operator and not to the sampling.
\Figref{equivariance_error} shows that the weighting scheme \eqnref{weights:khasanova} from \citep{khasanova2017sphericalcnn} does indeed not lead to a convolution that is equivariant to all rotations $g \in SO(3)$.\footnote{We however verified that the convolution is equivariant to longitudinal and latitudinal rotations, as intended.}
For $k=8$ neighbors, selecting the optimal kernel width $t$ improves on \citep{perraudin2019deepspherecosmo} at no cost, highlighting the importance of this parameter.
Increasing the resolution decreases the equivariance error in the high frequencies, an effect most probably due to the sampling.
Most importantly, the equivariance error decreases when connecting more neighbors.
Hence, the number of neighbors $k$ gives us a precise control of the tradeoff between cost and equivariance.
\section{Experiments}
\subsection{3D objects recognition} \label{sec:exp:shrec}
The recognition of 3D shapes is a rotation invariant task: rotating an object doesn't change its nature.
While 3D shapes are usually represented as meshes or point clouds, representing them as spherical maps (\figref{depthmap}) naturally allows a rotation invariant treatment.
The SHREC'17 shape retrieval contest \citep{shrec17} contains 51,300 randomly oriented 3D models from ShapeNet \citep{shapenet}, to be classified in 55 categories (tables, lamps, airplanes, etc.).
As in \citep{cohen2018sphericalcnn}, objects are represented by 6 spherical maps.
At each pixel, a ray is traced towards the center of the sphere.
The distance from the sphere to the object forms a depth map.
The $\cos$ and $\sin$ of the surface angle forms two normal maps.
The same is done for the object's convex hull.\footnote{Albeit we didn't observe much improvement by using the convex hull.}
The maps are sampled by an equiangular sampling with bandwidth $b = 64$ ($n = 4 b^2 = 16,384$ pixels) or an HEALPix sampling with $N_{side} = 32$ ($n = 12 N_{side}^2 = 12,288$ pixels).
The equiangular graph is built with \eqnref{weights:khasanova} and $k = 4$ neighbors \citep[following][]{khasanova2017sphericalcnn}.
The HEALPix graph is built with \eqnref{weights:belkin}, $k = 8$, and a kernel width $t$ set to the average of the distances \citep[following][]{perraudin2019deepspherecosmo}.
The NN is made of $5$ graph convolutional layers, each followed by a max pooling layer which down-samples by $4$.
A GAP and a fully connected layer with softmax follow.
The polynomials are all of order $P=3$ and the number of channels per layer is $16, 32, 64, 128, 256$, respectively.
Following \citet{esteves2018sphericalcnn}, the cross-entropy plus a triplet loss is optimized with Adam for 30 epochs on the dataset augmented by 3 random translations.
The learning rate is $5 \cdot 10^{-2}$ and the batch size is 32.
\begin{table}
\centering
\begin{tabular}{l cc r rS[table-format=2.0]}
\toprule
& \multicolumn{2}{c}{performance} & \multicolumn{1}{c}{size} & \multicolumn{2}{c}{speed} \\
\cmidrule(lr){2-3} \cmidrule(lr){4-4} \cmidrule(lr){5-6}
& F1 & mAP & params & \multicolumn{1}{c}{inference} & \multicolumn{1}{c}{training} \\
\midrule
\citet{cohen2018sphericalcnn} ($b=128$) & - & 67.6 & 1400\,k & 38.0\,ms & 50\,h \\
\citet{cohen2018sphericalcnn} (simplified,\protect\footnotemark $b=64$) & 78.9 & 66.5 & 400\,k & 12.0\,ms & 32\,h \\
\citet{esteves2018sphericalcnn} ($b=64$) & 79.4 & 68.5 & 500\,k & 9.8\,ms & 3\,h \\
DeepSphere (equiangular, $b=64$) & 79.4 & 66.5 & 190\,k & 0.9\,ms & 50\,m \\
DeepSphere (HEALPix, $N_{side}=32$) & 80.7 & 68.6 & 190\,k & 0.9\,ms & 50\,m \\
\bottomrule
\end{tabular}
\caption{
Results on SHREC'17 (3D shapes). DeepSphere achieves similar performance at a much lower cost, suggesting that anisotropic filters are an unnecessary price to pay.
}
\label{tab:shrec17}
\end{table}
\footnotetext[7]{As implemented in \url{https://github.com/jonas-koehler/s2cnn}.}
Results are shown in \tabref{shrec17}.
As the network is trained for shape classification rather than retrieval, we report the classification F1 alongside the mAP used in the retrieval contest.\footnote{We omit the F1 for \citet{cohen2018sphericalcnn} as we didn't get the mAP reported in the paper when running it.}
DeepSphere achieves the same performance as \citet{cohen2018sphericalcnn} and \citet{esteves2018sphericalcnn} at a much lower cost, suggesting that anisotropic filters are an unnecessary price to pay.
As the information in those spherical maps resides in the low frequencies (\figref{spectrum}), reducing the equivariance error didn't translate into improved performance.
For the same reason, using the more uniform HEALPix sampling or lowering the resolution down to $N_{side} = 8$ ($n=768$ pixels) didn't impact performance either.
\subsection{Cosmological model classification} \label{sec:exp:cosmo}
Given observations, cosmologists estimate the posterior probability of cosmological parameters, such as the matter density $\Omega_m$ and the normalization of the matter power spectrum $\sigma_8$.
Those parameters are estimated by likelihood-free inference, which requires a method to extract summary statistics to compare simulations and observations.
As the sufficient and most concise summary statistics are the parameters themselves, one desires a method to predict them from simulations.
As that is complicated to setup, prediction methods are typically benchmarked on the classification of spherical maps instead \citep{schmelze2017cosmologicalmodel}.
We used the same task, data, and setup as \citet{perraudin2019deepspherecosmo}: the classification of $720$ partial convergence maps made of $n \approx 10^6$ pixels ($1/12 \approx 8\%$ of a sphere at $N_{side} = 1024$) from two $\Lambda$CDM cosmological models, ($\Omega_m = 0.31$, $\sigma_8 = 0.82)$ and ($\Omega_m = 0.26$, $\sigma_8 = 0.91)$, at a relative noise level of $3.5$ (i.e., the signal is hidden in noise of $3.5$ times higher standard deviation).
Convergence maps represent the distribution of over- and under-densities of mass in the universe \citep[see][for a review of gravitational lensing]{bartelman2010gravitationallensing}.
Graphs are built with \eqnref{weights:belkin}, $k = 8, 20, 40$ neighbors, and the corresponding optimal kernel widths $t$ given in \secref{optimal}.
Following \citet{perraudin2019deepspherecosmo}, the NN is made of $5$ graph convolutional layers, each followed by a max pooling layer which down-samples by $4$.
A GAP and a fully connected layer with softmax follow.
The polynomials are all of order $P=4$ and the number of channels per layer is $16, 32, 64, 64, 64$, respectively.
The cross-entropy loss is optimized with Adam for 80 epochs.
The learning rate is $2 \cdot 10^{-4} \cdot 0.999^{\textrm{step}}$ and the batch size is 8.
\begin{table}
\begin{minipage}[b]{0.7\linewidth}
\centering
\begin{tabular}{l c c}
\toprule
& accuracy & time \\
\midrule
\citet{perraudin2019deepspherecosmo}, 2D CNN baseline & 54.2 & 104\,ms \\
\citet{perraudin2019deepspherecosmo}, CNN variant, $k=8$ & 62.1 & 185\,ms \\
\citet{perraudin2019deepspherecosmo}, FCN variant, $k=8$ & 83.8 & 185\,ms \\
$k=8$ neighbors, $t$ from \secref{optimal} & 87.1 & 185\,ms \\
$k=20$ neighbors, $t$ from \secref{optimal} & 91.3 & 250\,ms \\
$k=40$ neighbors, $t$ from \secref{optimal} & 92.5 & 363\,ms \\
\bottomrule
\end{tabular}
\caption{
Results on the classification of partial convergence maps.
Lower equivariance error translates to higher performance.
} \label{tab:cosmo}
\end{minipage} \hfill
\begin{minipage}[b]{0.25\linewidth}
\includegraphics[width=\linewidth]{cosmo_cost_accuracy_tradeoff}
\captionof{figure}{Tradeoff between cost and accuracy.}
\label{fig:cosmo:tradeoff}
\end{minipage}
\end{table}
Unlike on SHREC'17, results (\tabref{cosmo}) show that a lower equivariance error on the convolutions translates to higher performance.
That is probably due to the high frequency content of those maps (\figref{spectrum}).
There is a clear cost-accuracy tradeoff, controlled by the number of neighbors $k$ (\figref{cosmo:tradeoff}).
This experiment moreover demonstrates DeepSphere's flexibility (using partial spherical maps) and scalability (competing spherical CNNs were tested on maps of at most $10,000$ pixels).
\subsection{Climate event segmentation} \label{sec:exp:climate}
We evaluate our method on a task proposed by \citep{mudigonda2017climateevents}: the segmentation of extreme climate events, Tropical Cyclones (TC) and Atmospheric Rivers (AR), in global climate simulations (\figref{examples:climate}).
The data was produced by a 20-year run of the Community Atmospheric Model v5 (CAM5) and consists of 16 channels such as temperature, wind, humidity, and pressure at multiple altitudes.
We used the pre-processed dataset from \citep{jiang2019sphericalcnn}.\footnote{Available at \url{http://island.me.berkeley.edu/ugscnn/data}.}
There is 1,072,805 spherical maps, down-sampled to a level-5 icosahedral sampling ($n = 10 \cdot 4^l + 2 = 10,242$ pixels).
The labels are heavily unbalanced with 0.1\% TC, 2.2\% AR, and 97.7\% background (BG) pixels.
The graph is built with \eqnref{weights:belkin}, $k = 6$ neighbors, and a kernel width $t$ set to the average of the distances.
Following \citet{jiang2019sphericalcnn}, the NN is an encoder-decoder with skip connections.
Details in \secref{climate:appendix}.
The polynomials are all of order $P=3$.
The cross-entropy loss (weighted or non-weighted) is optimized with Adam for 30 epochs.
The learning rate is $1 \cdot 10{-3}$ and the batch size is 64.
\begin{table}
\centering
\begin{tabular}{l l l}
\toprule
& accuracy & mAP \\
\midrule
\citet{jiang2019sphericalcnn} (rerun) & 94.95 & 38.41 \\
\citet{cohen2019gauge} (S2R) & 97.5 & 68.6 \\
\citet{cohen2019gauge} (R2R) & 97.7 & 75.9 \\
DeepSphere (weighted loss) & $97.8\pm 0.3$ & $77.15\pm 1.94$ \\
DeepSphere (non-weighted loss) & $87.8\pm 0.5$ & $89.16\pm 1.37$ \\
\bottomrule
\end{tabular}
\caption{
Results on climate event segmentation: mean accuracy (over TC, AR, BG) and mean average precision (over TC and AR).
DeepSphere achieves state-of-the-art performance.
}
\label{tab:climate}
\end{table}
Results are shown in \tabref{climate} (details in tables~\ref{tab:climate:accuracy}, \ref{tab:climate:map} and~\ref{tab:climate:speed}).
The mean and standard deviation are computed over 5 runs.
Note that while \citet{jiang2019sphericalcnn} and \citet{cohen2019gauge} use a weighted cross-entropy loss, that is a suboptimal proxy for the mAP metric.
DeepSphere achieves state-of-the-art performance, suggesting again that anisotropic filters are unnecessary.
Note that results from \citet{mudigonda2017climateevents} cannot be directly compared as they don't use the same input channels.
Compared to \citet{cohen2019gauge}'s conclusion, it is surprising that S2R does worse than DeepSphere (which is limited to S2S).
Potential explanations are (i) that their icosahedral projection introduces harmful distortions, or (ii) that a larger architecture can compensate for the lack of generality.
We indeed observed that more feature maps and depth led to higher performance (\secref{climate:appendix}).
\subsection{Uneven sampling} \label{sec:exp:ghcn}
To demonstrate the flexibility of modeling the sampled sphere by a graph, we collected historical measurements from $n \approx 10,000$ weather stations scattered across the Earth.\footnote{\url{https://www.ncdc.noaa.gov/ghcn-daily-description}}
The spherical data is heavily non-uniformly sampled, with a much higher density of weather stations over North America than the Pacific (\figref{examples:ghcn:tmax}).
For illustration, we devised two artificial tasks.
A dense regression: predict the temperature on a given day knowing the temperature on the previous 5 days.
A global regression: predict the day (represented as one period of a sine over the year) from temperature or precipitations.
Predicting from temperature is much easier as it has a clear yearly pattern.
The graph is built with \eqnref{weights:belkin}, $k = 5$ neighbors, and a kernel width $t$ set to the average of the distances.
The equivariance property of the resulting graph has not been tested, and we don't expect it to be good due to the heavily non-uniform sampling.
The NN is made of $3$ graph convolutional layers.
The polynomials are all of order $P=0$ or $4$ and the number of channels per layer is $50, 100, 100$, respectively.
For the global regression, a GAP and a fully connected layer follow.
For the dense regression, a graph convolutional layer follows instead.
The MSE loss is optimized with RMSprop for 250 epochs.
The learning rate is $1 \cdot 10^{-3}$ and the batch size is 64.
\begin{table}
\centering
\begin{tabular}{
c
S[table-format=2.2]
S[table-format=1.2]
S[table-format=1.3]
S[table-format=1.2]
S[table-format=1.2]
S[table-format=1.3]
S[table-format=1.2]
S[table-format=1.2]
S[table-format=-1.3]
}
\toprule
& \multicolumn{3}{c}{temp. (from past temp.)} & \multicolumn{3}{c}{day (from temperature)} & \multicolumn{3}{c}{day (from precipitations)} \\
\cmidrule(lr){2-4} \cmidrule(lr){5-7} \cmidrule(lr){8-10}
order $P$ & {MSE} & {MAE} & {R2} & {MSE} & {MAE} & {R2} & {MSE} & {MAE} & {R2} \\
\midrule
$0$ & 10.88 & 2.42 & 0.896 & 0.10 & 0.10 & 0.882 & 0.58 & 0.42 & -0.980 \\
$4$ & 8.20 & 2.11 & 0.919 & 0.05 & 0.05 & 0.969 & 0.50 & 0.18 & 0.597 \\
\bottomrule
\end{tabular}
\caption{
Prediction results on data from weather stations.
Structure always improves performance.
}
\label{tab:ghcn}
\end{table}
Results are shown in \tabref{ghcn}.
While using a polynomial order $P=0$ is like modeling each time series independently with an MLP, orders $P>0$ integrate neighborhood information.
Results show that using the structure induced by the spherical geometry always yields better performance.
\section{Conclusion}
This work showed that DeepSphere strikes an interesting, and we think currently optimal, balance between desiderata for a spherical CNN.
A single parameter, the number of neighbors $k$ a pixel is connected to in the graph, controls the tradeoff between cost and equivariance (which is linked to performance).
As computational cost and memory consumption scales linearly with the number of pixels, DeepSphere scales to spherical maps made of millions of pixels, a required resolution to faithfully represent cosmological and climate data.
Also relevant in scientific applications is the flexibility offered by a graph representation (for partial coverage, missing data, and non-uniform samplings).
Finally, the implementation of the graph convolution is straightforward, and the ubiquity of graph neural networks --- pushing for their first-class support in DL frameworks --- will make implementations even easier and more efficient.
A potential drawback of graph Laplacian-based approaches is the isotropy of graph filters, reducing in principle the expressive power of the NN.
Experiments from \citet{cohen2019gauge} and \citet{boscaini2016anisotropicgraphnn} indeed suggest that more general convolutions achieve better performance.
Our experiments on 3D shapes (\secref{exp:shrec}) and climate (\secref{exp:climate}) however show that DeepSphere's isotropic filters do not hinder performance.
Possible explanations for this discrepancy are that NNs somehow compensate for the lack of anisotropic filters, or that some tasks can be solved with isotropic filters.
The distortions induced by the icosahedral projection in \citep{cohen2019gauge} or the leakage of curvature information in \citep{boscaini2016anisotropicgraphnn} might also alter performance.
Developing graph convolutions on irregular samplings that respect the geometry of the sphere is another research direction of importance.
Practitioners currently interpolate their measurements (coming from arbitrarily positioned weather stations, satellites or telescopes) to regular samplings.
This practice either results in a waste of resolution or computational and storage resources.
Our ultimate goal is for practitioners to be able to work directly on their measurements, however distributed.
\subsubsection*{Acknowledgments}
We thank Pierre Vandergheynst for advices,
and Taco Cohen for his inputs on the intriguing results of our comparison with \citet{cohen2019gauge}.
We thank the anonymous reviewers for their constructive feedback.
The following software packages were used for computation and plotting: PyGSP \citep{pygsp}, healpy \citep{healpy}, matplotlib \citep{matplotlib}, SciPy \citep{scipy}, NumPy \citep{numpy}, TensorFlow \citep{tensorflow}.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 7,902
|
Q: "No newline at end of file" compiler warning What is the reason for the following warning in some C++ compilers?
No newline at end of file
Why should I have an empty line at the end of a source/header file?
A: It isn't referring to a blank line, it's whether the last line (which can have content in it) is terminated with a newline.
Most text editors will put a newline at the end of the last line of a file, so if the last line doesn't have one, there is a risk that the file has been truncated. However, there are valid reasons why you might not want the newline so it is only a warning, not an error.
A: #include will replace its line with the literal contents of the file. If the file does not end with a newline, the line containing the #include that pulled it in will merge with the next line.
A: The requirement that every source file end with a non-escaped newline was removed in C++11. The specification now reads:
A source file that is not empty and that does not end in a new-line character, or that ends in a new-line character immediately preceded by a backslash character before any such splicing takes place, shall be processed as if an additional new-line character were appended to the file (C++11 §2.2/1).
A conforming compiler should no longer issue this warning (at least not when compiling in C++11 mode, if the compiler has modes for different revisions of the language specification).
A:
Of course in practice every compiler adds a new line after the #include. Thankfully. – @mxcl
not specific C/C++ but a C dialect: when using the GL_ARB_shading_language_include extension the glsl compiler on OS X warns you NOT about a missing newline. So you can write a MyHeader.h file with a header guard which ends with #endif // __MY_HEADER_H__ and you will lose the line after the #include "MyHeader.h" for sure.
A: C++03 Standard [2.1.1.2] declares:
... If a source file that is not empty does not end in a new-line character, or ends in a new-line character
immediately preceded by a backslash character before any such splicing takes place, the behavior is undefined.
A: Think of some of the problems that can occur if there is no newline. According to the ANSI standard the #include of a file at the beginning inserts the file exactly as it is to the front of the file and does not insert the new line after the #include <foo.h> after the contents of the file. So if you include a file with no newline at the end to the parser it will be viewed as if the last line of foo.h is on the same line as the first line of foo.cpp. What if the last line of foo.h was a comment without a new line? Now the first line of foo.cpp is commented out. These are just a couple of examples of the types of problems that can creep up.
Just wanted to point any interested parties to James' answer below. While the above answer is still correct for C, the new C++ standard (C++11) has been changed so that this warning should no longer be issued if using C++ and a compiler conforming to C++11.
From C++11 standard via James' post:
A source file that is not empty and that does not end in a new-line character, or that ends in a new-line character immediately preceded by a backslash character before any such splicing takes place, shall be processed as if an additional new-line character were appended to the file (C++11 §2.2/1).
A: I am using c-free IDE version 5.0,in my progrm either of 'c++' or 'c' language i was getting same problem.Just at the end of the program i.e. last line of the program(after braces of function it may be main or any function),press enter-line no. will be increased by 1.then execute the same program,it will run without error.
A: Because the behavior differs between C/C++ versions if file does not end with new-line. Especially nasty is older C++-versions, fx in C++ 03 the standard says (translation phases):
If a source file that is not empty does not end in a new-line
character, or ends in a new-line character immediately preceded by a
backslash character, the behavior is undefined.
Undefined behavior is bad: a standard conforming compiler could do more or less what it wants here (insert malicous code or whatever) - clearly a reason for warning.
While the situation is better in C++11 it is a good idea to avoid situations where the behavior is undefined in earlier versions. The C++03 specification is worse than C99 which outright prohibits such files (behavior is then defined).
A: The answer for the "obedient" is "because the C++03 Standard says the behavior of a program not ending in newline is undefined" (paraphrased).
The answer for the curious is here: http://gcc.gnu.org/ml/gcc/2001-07/msg01120.html.
A: This warning might also help to indicate that a file could have been truncated somehow. It's true that the compiler will probably throw a compiler error anyway - especially if it's in the middle of a function - or perhaps a linker error, but these could be more cryptic, and aren't guaranteed to occur.
Of course this warning also isn't guaranteed if the file is truncated immediately after a newline, but it could still catch some cases that other errors might miss, and gives a stronger hint to the problem.
A: In my case, I use KOTLIN Language and the compiler is on IntelliJ. Also, I am using a docker container with LINT to fix possible issues with typos, imports, code usage, etc. This error is coming from these lint fixes, most probably - I mean surely.
In short, the error says, 'Add a new line at the end of the file' That is it.
Before there was NO extra empty line:
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 6,425
|
#ifndef itkMetaSceneConverter_h
#define itkMetaSceneConverter_h
#include "metaScene.h"
#include "itkMetaEvent.h"
#include "itkSceneSpatialObject.h"
#include "itkDefaultStaticMeshTraits.h"
#include "itkMetaConverterBase.h"
#include <string>
#include <map>
namespace itk
{
/** \class MetaSceneConverter
* \brief Converts between MetaObject and SpaitalObject scenes.
*
* SpatialObject hierarchies are written to disk using the MetaIO
* library. This class is responsible for converting between MetaIO
* scenes and SpatialObject scenes
*
* \sa MetaConverterBase
* \ingroup ITKSpatialObjects
*/
template< unsigned int NDimensions,
typename PixelType = unsigned char,
typename TMeshTraits =
DefaultStaticMeshTraits< PixelType, NDimensions, NDimensions >
>
class ITK_TEMPLATE_EXPORT MetaSceneConverter
{
public:
/** SpatialObject Scene types */
using SceneType = itk::SceneSpatialObject< NDimensions >;
using ScenePointer = typename SceneType::Pointer;
/** Typedef for auxiliary conversion classes */
using MetaConverterBaseType = MetaConverterBase< NDimensions >;
using MetaConverterPointer = typename MetaConverterBaseType::Pointer;
using ConverterMapType = std::map< std::string, MetaConverterPointer >;
MetaSceneConverter();
~MetaSceneConverter();
static constexpr unsigned int MaximumDepth = 9999999;
/** Read a MetaFile and create a Scene SpatialObject */
ScenePointer ReadMeta(const char *name);
/** write out a Scene SpatialObject */
bool WriteMeta(SceneType *scene, const char *fileName,
unsigned int depth = MaximumDepth,
char *spatialObjectTypeName = nullptr);
const MetaEvent * GetEvent() const { return m_Event; }
void SetEvent(MetaEvent *event) { m_Event = event; }
/** Set if the points should be saved in binary/ASCII */
void SetBinaryPoints(bool binary) { m_BinaryPoints = binary; }
/** set/get the precision for writing out numbers as plain text */
void SetTransformPrecision(unsigned int precision)
{
m_TransformPrecision = precision;
}
unsigned int GetTransformPrecision(){ return m_TransformPrecision; }
/** Set if the images should be written in different files */
void SetWriteImagesInSeparateFile(bool separate)
{
m_WriteImagesInSeparateFile = separate;
}
/** add new SpatialObject/MetaObject converters at runtime
*
* Every Converter is mapped to both a metaObject type name
* and a spatialObject type name -- these need to match what
* gets read from & written to the MetaIO file
*/
void RegisterMetaConverter(const char *metaTypeName,
const char *spatialObjectTypeName,
MetaConverterBaseType *converter);
MetaScene * CreateMetaScene(SceneType *scene,
unsigned int depth = MaximumDepth,
char *name = nullptr);
ScenePointer CreateSpatialObjectScene(MetaScene *scene);
private:
using SpatialObjectType = itk::SpatialObject< NDimensions >;
using SpatialObjectPointer = typename SpatialObjectType::Pointer;
using TransformType = typename SpatialObjectType::TransformType;
using MetaObjectListType = std::list< MetaObject * >;
template <typename TConverter>
MetaObject *SpatialObjectToMetaObject(SpatialObjectPointer &so)
{
typename TConverter::Pointer converter = TConverter::New();
// needed just for Image & ImageMask
converter->SetWriteImagesInSeparateFile(this->m_WriteImagesInSeparateFile);
return converter->SpatialObjectToMetaObject(so);
}
template <typename TConverter>
SpatialObjectPointer MetaObjectToSpatialObject(const MetaObject *mo)
{
typename TConverter::Pointer converter = TConverter::New();
return converter->MetaObjectToSpatialObject(mo);
}
void SetTransform(MetaObject *obj, TransformType *transform);
void SetTransform(SpatialObjectType *so, MetaObject *obj);
double m_Orientation[100];
double m_Position[10];
double m_CenterOfRotation[10];
MetaEvent * m_Event;
bool m_BinaryPoints;
bool m_WriteImagesInSeparateFile;
unsigned int m_TransformPrecision;
ConverterMapType m_ConverterMap;
};
} // end namespace itk
#ifndef ITK_MANUAL_INSTANTIATION
#include "itkMetaSceneConverter.hxx"
#endif
#endif
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,262
|
Home » Archives » U.C. Bulletin July 2009
U.C. Bulletin July 2009
The Cambodia Weekly Announces New Name: The Southeast Asia Weekly
Anticipating the grand opening of the new T.V. station (Southeast Asia) and coinciding with the up and running radio station (FM 106), The Cambodia Weekly has decided to under go a name change. It will now be known as The Southeast Asia Weekly.
Sixth Annual Fashion Show Proves to be More Than Ordinary
As Term 2 comes to an end, everyone finds themselves gathered at the UC Conference Room for the Sixth Annual Fashion Show in celebration of the founding of The University of Cambodia on Sunday, June 21st, 2009.
The University of Cambodia Hosts Event for Nou Hach Literary Association
By Christopher Smith and Phon Chanvutty
On Monday, June 15, The University of Cambodia hosted an event for the Nou Hach Literary Association. There were writers from Cambodia and Northern Europe who worked together to present a feast of poetry for a discerning audience of students and guests.
Customer Service Training Workshop
As part of our step to better care for our students and staff, the Asian Leadership Center offered a customer service training section to provide the importance of its needs to the University.
What is a Virus? How Can We Protect Our Computers From Such A Threat?
On June 15 and 17, 2009 the Chhay Hok computer company came and explained about the basics of a virus and how it eventually affects our computers.
UC Participates in "Southeast Asian Service-Learning Institute"
Part of a four and a half day event, June 29-July 3, UC became part of a seminar that taught us about the importance of Community Service-Learning.
From the Dean's Desk: UC: We Build Tomorrow's Leaders!
Vannarith Chheang, Associate Dean of the Graduate School talks about how UC has a vision and mission to produce tomorrow's leaders.
AFD and CES Prepares for Graduation
In preparation for July 15, 2009 graduation, AFD and CES are working very hard to get everything finalized for the big event for our fellow graduates.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,688
|
Q: Arduino: how to move a servo based on difference of resistance in photoresistors I am trying to make a solar panel that tracks the sun and in order to do that I am using 4 photoresistors. I am very new to coding but I have already coded the part that determines the resistance of the photoresistors (for now I'm only trying with 2). I need to get a servo to move in a certain direction depending on which photoresistor has a larger resistance. Because the resistance will not be constant even when light is, I also need to have a margin where as long as the difference of both resistances is within it, the servo will not move. I tried copying over one of the examples for servo movement into my code but I can't get it to work.
Here is what I have:
const int sensorPin = A0;
const int sensorPin1 = A1;
int sensorValue = 0;
int sensorValue1 = 0;
float Vin = 5;
float Vout = 0;
float Vout1 = 0;
float Rref = 2180;
float R = 0;
float R1 = 0;
#include <Servo.h>
Servo myservo;
int pos = 0;
void setup() {
Serial.begin(9600);
myservo.attach(9);
}
void loop() {
sensorValue = analogRead(sensorPin);
Vout = (Vin * sensorValue) / 1023;
R = Rref * (1 / ((Vin / Vout) - 1));
Serial.print("R: ");
Serial.println(R);
delay(1000);
sensorValue1 = analogRead(sensorPin1);
Vout1 = (Vin * sensorValue1) / 1023;
R1 = Rref * (1 / ((Vin / Vout1) -1));
Serial.print("R1: ");
Serial.println(R1);
delay(1000);
for (R1 > R; pos = 180; pos <= 0; pos += 1) {
myservo.write(pos);
delay(15);
}
for (R1 < R; pos = 0; pos >= 180; pos -= 1) {
myservo.write(pos);
delay(15);
}
}
A: Your for loop is wrong, it has four arguments. I would suggest putting the R1 < R part around your for-loop in an if-condition. Right now, your for loop also starts at pos = 180, is incremented, but the condition is pos <= 0. Vice versa for the second loop. Your increment/decrements and conditions are wrong. This loop never executes, even if the fourth argument is removed.
For-loop: https://www.programiz.com/c-programming/c-for-loop
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 5,963
|
If you went to buy a consumer good a few years ago, depending on the urgency, you would leverage your contacts to get the best advice. Maybe you would start with a chat with friends over a coffee, discussing what they use. Then maybe you would talk to professionals in the area, like service people, about their recommendations. You would endlessly walk around the shops talking with sales people, researching as you went.
Maybe you still do things this way. Nothing wrong with that, especially if you aren't constantly online.
Working in the online world, I often see trends both social and technological well before they go mainstream. For example, the now common place usage of online shops and social media. This is an advantage you get working in the web industry.
However often things that I consider common place I not always such for the rest of the community.
The world is getting faster. Yes granted, but to different degrees for different people.
Take this point in case. We need to purchase a consumer good on the weekend.
By using my online social networks and research on various online forums both general and vendor specific we where able to do all our research and sourcing of recommendations from contacts and experts within 90 minutes of starting to look for the information.
Two hours later we had the consumer good in the back of our car on the drive home. What would have taken weeks was now down to a few hours.
However it's not about the time. It's about what we did.
The use of our online research on forums, looking at the user experiences that people had with their consumer goods. Looking to the communities for opinions. Something that is more often than not outside of the control of the manufacturers. I knew that from these communities I would get a truer opinion, than from the manufacturers.
Now this is something your have to consider if your business is online. Don't think old school. You control nothing.
You don't control your message anymore – the consumer does.
Similarly asking for reviews from my online social network. These reviews I trust, as I know the people, they are my peers. These are the testimonials that I know are not hand selected, I trust that they will be unbiased.
The point here is that the social network is MY social network, I control it, not the consumer goods manufacturer. Interestingly the only manufacturer to contact (spam) me did so 36 hours after the event (I don't think they get it).
Again the testimonial isn't controlled by the manufacturer, it's controlled by me!
Something to think about, the online world and it's hyper-connectivity is starting to influence us in many different ways.
Now I know that the average consumer doesn't operate in this hyper-connected online world that I do. However consider that this lifestyle is just around the corner for everyone.
The online social world is now becoming ever important. Previously having a web site used to be all you need to do online. The times are changing. Social media needs to be considered. Question is are you ready, is your business ready. Are you prepared to commit resources to it, or not?
I'm not saying rush out and get a Twitter account or a Facebook page. However you need to be aware of your brands online footprint.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,491
|
// ----------------------------------------------------------------------------------
//
// Copyright Microsoft Corporation
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
// http://www.apache.org/licenses/LICENSE-2.0
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
// ----------------------------------------------------------------------------------
using AutoMapper;
using Microsoft.Azure.Commands.Compute.Common;
using Microsoft.Azure.Commands.Compute.Models;
using Microsoft.Azure.Commands.ResourceManager.Common.ArgumentCompleters;
using Microsoft.Azure.Management.Compute;
using Microsoft.Azure.Management.Compute.Models;
using Microsoft.Rest.Azure;
using System;
using System.Collections;
using System.Globalization;
using System.Management.Automation;
namespace Microsoft.Azure.Commands.Compute.Extension.AzureDiskEncryption
{
[Cmdlet("Disable", ResourceManager.Common.AzureRMConstants.AzureRMPrefix + "VMDiskEncryption",SupportsShouldProcess = true)]
[OutputType(typeof(PSAzureOperationResponse))]
public class DisableAzureDiskEncryptionCommand : VirtualMachineExtensionBaseCmdlet
{
[Parameter(
Mandatory = true,
Position = 0,
ValueFromPipelineByPropertyName = true,
HelpMessage = "The resource group name to which the VM belongs to")]
[ResourceGroupCompleter]
[ValidateNotNullOrEmpty]
public string ResourceGroupName { get; set; }
[Alias("ResourceName")]
[Parameter(
Mandatory = true,
Position = 1,
ValueFromPipelineByPropertyName = true,
HelpMessage = "Name of the virtual machine")]
[ResourceNameCompleter("Microsoft.Compute/virtualMachines", "ResourceGroupName")]
[ValidateNotNullOrEmpty]
public string VMName { get; set; }
[Parameter(
Mandatory = false,
Position = 2,
ValueFromPipelineByPropertyName = true,
HelpMessage = "Type of the volume (OS, Data or All) to perform decryption operation")]
[ValidateSet(
AzureDiskEncryptionExtensionContext.VolumeTypeOS,
AzureDiskEncryptionExtensionContext.VolumeTypeData,
AzureDiskEncryptionExtensionContext.VolumeTypeAll)]
public string VolumeType { get; set; }
[Alias("ExtensionName")]
[Parameter(
Mandatory = false,
Position = 3,
ValueFromPipelineByPropertyName = true,
HelpMessage = "The extension name. If this parameter is not specified, default values used are AzureDiskEncryption for windows VMs and AzureDiskEncryptionForLinux for Linux VMs")]
[ResourceNameCompleter("Microsoft.Compute/virtualMachines/extensions", "ResourceGroupName", "VMName")]
[ValidateNotNullOrEmpty]
public string Name { get; set; }
[Alias("HandlerVersion", "Version")]
[Parameter(
Mandatory = false,
Position = 4,
ValueFromPipelineByPropertyName = true,
HelpMessage = "The type handler version of AzureDiskEncryption VM extension that is used to disable encryption on the VM.")]
[ValidateNotNullOrEmpty]
public string TypeHandlerVersion { get; set; }
[Parameter(HelpMessage = "To force the operation of decrypting the virtual machine.")]
[ValidateNotNullOrEmpty]
public SwitchParameter Force { get; set; }
[Parameter(
Mandatory = false,
ValueFromPipelineByPropertyName = true,
HelpMessage = "Disable auto-upgrade of minor version")]
public SwitchParameter DisableAutoUpgradeMinorVersion { get; set; }
[Parameter(
Mandatory = false,
ValueFromPipelineByPropertyName = true,
HelpMessage = "The extension type. Specify this parameter to override its default value of \"AzureDiskEncryption\" for Windows VMs and \"AzureDiskEncryptionForLinux\" for Linux VMs.")]
[ValidateNotNullOrEmpty]
public string ExtensionType { get; set; }
[Parameter(
Mandatory = false,
ValueFromPipelineByPropertyName = true,
HelpMessage = "The extension publisher name. Specify this parameter only to override the default value of \"Microsoft.Azure.Security\".")]
[ValidateNotNullOrEmpty]
public string ExtensionPublisherName { get; set; }
private OperatingSystemTypes? currentOSType = null;
private Hashtable GetExtensionPublicSettings()
{
Hashtable publicSettings = new Hashtable();
// Generate a new sequence version everytime to force run the extension.
// This is to bypass CRP & Guest Agent's optimization of not re-running the extension when there is no config change
string sequenceVersion = Guid.NewGuid().ToString();
publicSettings.Add(AzureDiskEncryptionExtensionConstants.volumeTypeKey, VolumeType ?? String.Empty);
publicSettings.Add(AzureDiskEncryptionExtensionConstants.encryptionOperationKey, AzureDiskEncryptionExtensionConstants.disableEncryptionOperation);
publicSettings.Add(AzureDiskEncryptionExtensionConstants.sequenceVersionKey, sequenceVersion);
return publicSettings;
}
private Hashtable GetExtensionProtectedSettings()
{
return null;
}
private bool IsVmModelEncryptionSet(VirtualMachine vm)
{
return (vm != null &&
vm.StorageProfile != null &&
vm.StorageProfile.OsDisk != null &&
vm.StorageProfile.OsDisk.EncryptionSettings != null);
}
private string GetVersionForDisable(VirtualMachine vm)
{
// if there is currently an extension installed, use that version for disable as well
if (vm.Resources != null)
{
foreach (VirtualMachineExtension vme in vm.Resources)
{
if (vme.Publisher.Equals(AzureDiskEncryptionExtensionContext.ExtensionDefaultPublisher) &&
(vme.VirtualMachineExtensionType.Equals(AzureDiskEncryptionExtensionContext.ExtensionDefaultType) ||
vme.VirtualMachineExtensionType.Equals(AzureDiskEncryptionExtensionContext.LinuxExtensionDefaultType)))
{
return vme.TypeHandlerVersion;
}
}
}
// If we reach this point, no extension is currently installed, even if the VM is encrypted.
// To disable encryption, we will select the extension version matching the encryption
// settings present on the VM and then use that version to issue the disable operation.
if (IsVmModelEncryptionSet(vm))
{
// encryption settings present in VM model, use default dual pass version
if (currentOSType == OperatingSystemTypes.Linux)
{
return AzureDiskEncryptionExtensionContext.LinuxExtensionDefaultVersion;
}
else
{
return AzureDiskEncryptionExtensionContext.ExtensionDefaultVersion;
}
}
else
{
// encryption settings not present in the VM model, use single pass version
if (currentOSType == OperatingSystemTypes.Linux)
{
return AzureDiskEncryptionExtensionContext.LinuxExtensionSinglePassVersion;
}
else
{
return AzureDiskEncryptionExtensionContext.ExtensionSinglePassVersion;
}
}
}
private VirtualMachineExtension GetVmExtensionParameters(VirtualMachine vmParameters)
{
Hashtable SettingString = GetExtensionPublicSettings();
Hashtable ProtectedSettingString = GetExtensionProtectedSettings();
if (vmParameters == null)
{
ThrowTerminatingError(
new ErrorRecord(
new ApplicationException(
string.Format(
CultureInfo.CurrentUICulture,
"Disable-AzureDiskEncryption can disable encryption only on a VM that was already created ")),
"InvalidResult",
ErrorCategory.InvalidResult,
null));
}
VirtualMachineExtension vmExtensionParameters = null;
if (OperatingSystemTypes.Windows.Equals(currentOSType))
{
this.Name = this.Name ?? AzureDiskEncryptionExtensionContext.ExtensionDefaultName;
vmExtensionParameters = new VirtualMachineExtension
{
Location = vmParameters.Location,
Publisher = this.ExtensionPublisherName ?? AzureDiskEncryptionExtensionContext.ExtensionDefaultPublisher,
VirtualMachineExtensionType = this.ExtensionType ?? AzureDiskEncryptionExtensionContext.ExtensionDefaultType,
TypeHandlerVersion = GetVersionForDisable(vmParameters),
Settings = SettingString,
ProtectedSettings = ProtectedSettingString,
AutoUpgradeMinorVersion = !DisableAutoUpgradeMinorVersion.IsPresent
};
}
if (OperatingSystemTypes.Linux.Equals(currentOSType))
{
this.Name = this.Name ?? AzureDiskEncryptionExtensionContext.LinuxExtensionDefaultName;
vmExtensionParameters = new VirtualMachineExtension
{
Location = vmParameters.Location,
Publisher = this.ExtensionPublisherName ?? AzureDiskEncryptionExtensionContext.LinuxExtensionDefaultPublisher,
VirtualMachineExtensionType = this.ExtensionType ?? AzureDiskEncryptionExtensionContext.LinuxExtensionDefaultType,
TypeHandlerVersion = GetVersionForDisable(vmParameters),
Settings = SettingString,
ProtectedSettings = ProtectedSettingString,
AutoUpgradeMinorVersion = !DisableAutoUpgradeMinorVersion.IsPresent
};
}
return vmExtensionParameters;
}
/// <summary>
/// This function gets the VM model, fills in the OSDisk properties with encryptionSettings and does an UpdateVM
/// </summary>
private AzureOperationResponse<VirtualMachine> UpdateVmEncryptionSettings()
{
var vmParameters = (this.ComputeClient.ComputeManagementClient.VirtualMachines.Get(
this.ResourceGroupName, this.VMName));
if ((vmParameters == null) ||
(vmParameters.StorageProfile == null) ||
(vmParameters.StorageProfile.OsDisk == null))
{
// VM should have been created and have valid storageProfile and OSDisk by now
ThrowTerminatingError(
new ErrorRecord(
new ApplicationException(
string.Format(
CultureInfo.CurrentUICulture,
"Set-AzureDiskEncryptionExtension can enable encryption only on a VM that was already created and has appropriate storageProfile and OS disk")),
"InvalidResult",
ErrorCategory.InvalidResult,
null));
}
DiskEncryptionSettings encryptionSettings = new DiskEncryptionSettings();
encryptionSettings.Enabled = false;
vmParameters.StorageProfile.OsDisk.EncryptionSettings = encryptionSettings;
var parameters = new VirtualMachine
{
DiagnosticsProfile = vmParameters.DiagnosticsProfile,
HardwareProfile = vmParameters.HardwareProfile,
StorageProfile = vmParameters.StorageProfile,
NetworkProfile = vmParameters.NetworkProfile,
OsProfile = vmParameters.OsProfile,
Plan = vmParameters.Plan,
AvailabilitySet = vmParameters.AvailabilitySet,
Location = vmParameters.Location,
Tags = vmParameters.Tags
};
return this.ComputeClient.ComputeManagementClient.VirtualMachines.CreateOrUpdateWithHttpMessagesAsync(
this.ResourceGroupName,
vmParameters.Name,
parameters).GetAwaiter().GetResult();
}
public override void ExecuteCmdlet()
{
base.ExecuteCmdlet();
ExecuteClientAction(() =>
{
VirtualMachine virtualMachineResponse = (this.ComputeClient.ComputeManagementClient.VirtualMachines.Get(this.ResourceGroupName, this.VMName, InstanceViewTypes.InstanceView));
if ((virtualMachineResponse == null) ||
(virtualMachineResponse.StorageProfile == null) ||
(virtualMachineResponse.StorageProfile.OsDisk == null))
{
// VM should have been created and have valid storageProfile and OSDisk by now
ThrowTerminatingError(
new ErrorRecord(
new ApplicationException(
string.Format(
CultureInfo.CurrentUICulture,
"Disable-AzureDiskEncryption can disable encryption only on a VM that was already created and has appropriate storageProfile and OS disk")),
"InvalidResult",
ErrorCategory.InvalidResult,
null));
}
currentOSType = virtualMachineResponse.StorageProfile.OsDisk.OsType;
if (OperatingSystemTypes.Linux.Equals(currentOSType) &&
!AzureDiskEncryptionExtensionContext.VolumeTypeData.Equals(VolumeType, StringComparison.InvariantCultureIgnoreCase))
{
ThrowTerminatingError(
new ErrorRecord(
new ArgumentException(
string.Format(
CultureInfo.CurrentUICulture,
"Disabling encryption is only allowed on Data volumes for Linux VMs.")),
"InvalidType",
ErrorCategory.NotImplemented,
null));
}
if (this.ShouldProcess(VMName, Properties.Resources.DisableDiskEncryptionAction)
&& (this.Force.IsPresent ||
this.ShouldContinue(Properties.Resources.DisableAzureDiskEncryptionConfirmation, Properties.Resources.DisableAzureDiskEncryptionCaption)))
{
VirtualMachineExtension parameters = GetVmExtensionParameters(virtualMachineResponse);
var opExt = this.VirtualMachineExtensionClient.CreateOrUpdateWithHttpMessagesAsync(
this.ResourceGroupName,
this.VMName,
this.Name,
parameters).GetAwaiter().GetResult();
// +---------+---------------+----------------------------+
// | OSType | VolumeType | UpdateVmEncryptionSettings |
// +---------+---------------+----------------------------+
// | Windows | OS | Yes |
// | Windows | Data | No |
// | Windows | Not Specified | Yes |
// | Linux | OS | N/A |
// | Linux | Data | Yes |
// | Linux | Not Specified | N/A |
// +---------+---------------+----------------------------+
if ((OperatingSystemTypes.Windows.Equals(currentOSType) && parameters.TypeHandlerVersion.Equals(AzureDiskEncryptionExtensionContext.ExtensionSinglePassVersion)) ||
(OperatingSystemTypes.Linux.Equals(currentOSType) && parameters.TypeHandlerVersion.Equals(AzureDiskEncryptionExtensionContext.LinuxExtensionSinglePassVersion)) ||
(OperatingSystemTypes.Windows.Equals(currentOSType) && !string.IsNullOrEmpty(VolumeType) && VolumeType.Equals(AzureDiskEncryptionExtensionContext.VolumeTypeData, StringComparison.InvariantCultureIgnoreCase)))
{
var result = ComputeAutoMapperProfile.Mapper.Map<PSAzureOperationResponse>(opExt);
WriteObject(result);
}
else
{
var opVm = UpdateVmEncryptionSettings();
var result = ComputeAutoMapperProfile.Mapper.Map<PSAzureOperationResponse>(opVm);
WriteObject(result);
}
}
});
}
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,119
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.