{ "paper_id": "J77-3003", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T02:49:31.902466Z" }, "title": "NEWSLETTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUI S T I W ASPECTS OF ENGLISH SENTENCE STRESS, by Susan F. Schmerling, Reviewed by", "authors": [ { "first": "Sven", "middle": [], "last": "Jacobson", "suffix": "", "affiliation": {}, "email": "" }, { "first": "David", "middle": [ "G" ], "last": "Hays", "suffix": "", "affiliation": {}, "email": "" }, { "first": "William", "middle": [], "last": "Benzon", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Donald", "middle": [ "E" ], "last": "Walker", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Martin", "middle": [], "last": "Kay", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Xerox", "middle": [], "last": "P A L O Alto", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Research", "middle": [], "last": "Center", "suffix": "", "affiliation": {}, "email": "" }, { "first": "", "middle": [], "last": "Austin", "suffix": "", "affiliation": {}, "email": "" }, { "first": "American", "middle": [], "last": "Jaurn", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Computati~nd", "middle": [], "last": "Lingdttic~", "suffix": "", "affiliation": {}, "email": "" }, { "first": "M", "middle": [], "last": "Bruguiere", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Haut", "middle": [], "last": "Cornit6 De La Langue", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Eirancaise", "middle": [ "N" ], "last": "Chomsky", "suffix": "", "affiliation": {}, "email": "" }, { "first": "A", "middle": [], "last": "Culioli", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Paris", "middle": [ "A" ], "last": "Dugas", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Universitb", "middle": [], "last": "Du Qu6bec", "suffix": "", "affiliation": {}, "email": "" }, { "first": "B", "middle": [], "last": "Montrbal", "suffix": "", "affiliation": {}, "email": "" }, { "first": "-L", "middle": [], "last": "Grundt", "suffix": "", "affiliation": {}, "email": "" }, { "first": "R", "middle": [], "last": "Jakobson", "suffix": "", "affiliation": {}, "email": "" }, { "first": "S", "middle": [ "Y" ], "last": "Kuroda", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Umvedt6", "middle": [], "last": "De", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Californle", "middle": [], "last": "San", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Diego", "middle": [ "A" ], "last": "Lentin", "suffix": "", "affiliation": {}, "email": "" }, { "first": "M", "middle": [ "P" ], "last": "Schutzenberger", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Universitk", "middle": [], "last": "Paris", "suffix": "", "affiliation": {}, "email": "" }, { "first": "E", "middle": [], "last": "Spang-Hanssepj", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Univmitd", "middle": [], "last": "Tie Copesahague", "suffix": "", "affiliation": {}, "email": "" }, { "first": "E", "middle": [ "R" ], "last": "S~d", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Universitc", "middle": [], "last": "G~d", "suffix": "", "affiliation": {}, "email": "" }, { "first": "R.-L", "middle": [], "last": "De Lurid", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Ecble", "middle": [], "last": "Wagner", "suffix": "", "affiliation": {}, "email": "" }, { "first": "", "middle": [], "last": "Pratique", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Haute", "middle": [], "last": "Etudes", "suffix": "", "affiliation": {}, "email": "" }, { "first": "J", "middle": [], "last": "Dubois", "suffix": "", "affiliation": {}, "email": "" }, { "first": "A", "middle": [], "last": "Guillet", "suffix": "", "affiliation": {}, "email": "" }, { "first": "", "middle": [], "last": "Laboratoire D'automakque Documentaire", "suffix": "", "affiliation": {}, "email": "" }, { "first": "", "middle": [], "last": "Lmguistique", "suffix": "", "affiliation": {}, "email": "" }, { "first": "R", "middle": [ "S" ], "last": "Kayne", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Universitb", "middle": [], "last": "De", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Paris-Vhcennes- J .-C", "middle": [], "last": "Milner", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Universit6", "middle": [], "last": "Paris", "suffix": "", "affiliation": {}, "email": "" }, { "first": "N", "middle": [], "last": "R~t", "suffix": "", "affiliation": {}, "email": "" }, { "first": "J", "middle": [ "R" ], "last": "Vergnaud", "suffix": "", "affiliation": {}, "email": "" }, { "first": "", "middle": [], "last": "Laboratoire", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Documentain", "middle": [], "last": "Zinguistiquq", "suffix": "", "affiliation": {}, "email": "" }, { "first": "John", "middle": [ "W M" ], "last": "Verhaar", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Werner", "middle": [], "last": "Abraham", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Richard", "middle": [ "D" ], "last": "Brecht", "suffix": "", "affiliation": {}, "email": "" }, { "first": "K", "middle": [], "last": "Kunjunni Raja", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Bruoe", "middle": [], "last": "Fraser", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Benson", "middle": [], "last": "Mates", "suffix": "", "affiliation": {}, "email": "" }, { "first": "U", "middle": [], "last": "Of", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Morris", "middle": [], "last": "Halle", "suffix": "", "affiliation": {}, "email": "" }, { "first": "J", "middle": [ "F" ], "last": "Staal", "suffix": "", "affiliation": {}, "email": "" }, { "first": "-", "middle": [ "T" ], "last": "Usa", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Peter", "middle": [], "last": "Hartmann", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Pieter", "middle": [ "A" ], "last": "Verburg", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Univ", "middle": [], "last": "Konstanz", "suffix": "", "affiliation": {}, "email": "" }, { "first": "Arjen", "middle": [ "F" ], "last": "Florijn", "suffix": "", "affiliation": {}, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "AMERICAN JOURNAL OF COMPUTATIONAL LINGUISTICS is published by the Association for Computational Linguistics.", "pdf_parse": { "paper_id": "J77-3003", "_pdf_hash": "", "abstract": [ { "text": "AMERICAN JOURNAL OF COMPUTATIONAL LINGUISTICS is published by the Association for Computational Linguistics.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "T h e middle c h a p t e r s ( 3 , 4) examine certain assumptions that most studies of accent have made. Ch. 3 argues t h a t t h e r e is no such t h i n g as a n e u t r a l sentence w i t h a Hnormal~l-position for t h e accent, and o f f e r s t h e interesting explanation t h a t what t h e I1nor-ma1 accenttt advocates have been referring to, without realizing it, is t h e c i t a t i o n form of a sentence--that t h i s is t h e only t r u l y wanted such a pure context-free accent p a t t e r n , as S p o i n t s o u t , we could not g e t i t , because speakers will imagine a context according t~ c e r t a i n clues t h a t t h e sentence i t s e l f may o f f e r . Most w i l l read --John died with t h e accent on __I_ J o h n , but -My physics professor died w l t h t h e accent on died. (A neater p a i r m i g h t have been ---tlSilas Marner is hereft versus \"Silas Marner is hereu--to show mini m a l l y t h e e f f e c t of a personal name, m a t c h e d w i t h a book title.)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "It is c e r t a i n l y t r u e that \"neutral accentft has had t p o l i t t l e ... Those people just don't give a damn.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "The l a t t e r censure is milder--it m i g h t imply ' W h a t 1 t h e use of t h a n one, but t h e r e is no space to elaborate here.)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "The assumptions c r i t i c i z e d i n c h . 4 are two: that membership in a c e r t a i n category (say noun or pronoun) may tell w h e t h e r somet h i n g is accentable, and t h a t , t h e r e is a special kjnd of accent t h a t may be called oontrastive.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "S points w t that t h e r e are so many exceptions t o t;'h8 rep-dted accenting of nouns and no*-accenting of pronouns--even when one or t h e o t h e r is ~a p h o r i c --t h a t no generalizations can be made. She c i t e s the type -I know who's standfnqin f r o n tof Mary, -- -T h i s like the out t e n s e from verb> Why don't y o u write it? --I wrote it!' specific. In p l a c e of it, S presents a looser g e n e r a l i z a t i o n f o r which she uses such informal terms as t h a t w h i c h is remarkable, noteworthy (67), ornew (72), ~r r c u e positiye side, and lackof simificance (71) or taken f o r anted (71;) ont h e negative side. notion of presupposition .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "blit I don't know who -M", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "==== I d i d write i t ! ) -. L -- v - - Though S is undoubtedly r i g h t a", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "she", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "In a sentence Like -I didn'ttake.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "m g umbrella because -I didn't r e a l i z e -it w a s raininq t h e f a c t i v e verb realize e n t a i l s t h e factuality of 'raining' But t h i s has t o do w i t h t m t , h , not w i t h knowledge w h i c h t h e speaker and h e a r e r share--there is no reason to take i t f o r granted t h a t $he hearer knows it was r a i n i n g at t h a t place and time.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "We may now look back at t h e criticism l e v e l e d at B o l i r r g e r for using t h e term predictability. Abdicated is de-accented becauee \"The speaker is interested in t h e f a t e of kings and the rise of democracies. T h e r e is enough mutual underatanding between him and h i s interlocutor to m a k e him reasonably sure t h a t t h e mention of 'kings1--in t h e conkext of de-mocrac-y--will suggest 'abdication'\" (~o l i n g e r 1972, 635). The passage in which t h i s occurs began with t h e statement t h a t Itwhat counts is relative ;J=e;;=;==Ep= semantic weight\" (with emphasis as shown), and t h e only claim intended was that as between 'kings' tiad 'abdicatet, 'kings1 had more to say--the question was 'kingst ( onarc hy ) versus tdemacracy', not ' a b d i c a t e t versus 'remain' In S ' s terme, -w h i c h I conceive t o be naming t h e same thing t h a t I was t r y i n g to name, k i n would be t h e elemeqt designating t h a t w h i c h was most \"remarkablen.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "S. continues h e r critique w i t h t w o examples from h e r own ex- e r has to decide which element is more suggestive and which is less.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "To Bay t h a t ~Trumanl s d e a t h was expectedtt is to a t t a c h b t h e expect a t i o n t o t h e whole event, when it needs to be a t t a c h e d d i f f e r e nt i a l l y t o Truman or to death. The question is, given t h e t o t a l context, which word c a r r i e s greater semantic w e i g h t , is more newsw o r t h y , more remarkable--Truman ordied? Everybody h a s been t a l kwg about .Truman .so an ad@Tthmal mentf on of Truman is n o t part i c u l a r l y newsworthy. On t h e o t h e r hand, t h e question at t h e ~i m e was, will he l i v e through t h e day? will he linger on? w i l l h e perhaps r a l l y ? Q ' s same misconception regarding what t h e \"expectation\" a t t a c h e s to crops up d a t e r in h e r lttopic-commenttt Cases. She cites t h e example ~6 h n survived as one t h a t llcould be uttered only in a context where some such expectation on t h e p a r t of the audience was assumed t h e speaker know our friend has been ihvolved in a serious automobile a c c i d t n t ) \" ( 9 0 ) . Expect a t i o n of what? Not t h a t John would survive, but t h a t John would be mentioned. If we know that he has been involved in a serious 6 = = = = = = = = accident we no more expect him to a~r v i y e than t od i e or tobe knocked senseless or to -come out witbouka scratch--which possibility w i l l be t h e actuaX outcome i s precisely t h e question. Ano t h e r p a i r t h a t points up t h e contrast mort s h a r p l y :", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "What dp you have against smoking? --Itv s t h e nicotine t h a t it puts i n youp. body.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "", "sec_num": null }, { "text": "Given t h e knowledge of t h e world that speaker and Qearer share, (In t h e l a t t e r , it i o not even necessary t~ mention I _ .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "... It's t h e harm t h a t it does t o your body.", "sec_num": null }, { "text": "It is nob t h e newsworthin,e::s o f the. utterance as a w h o l e t h a t counts, h u t t h e relative newsworthiness of t h e itam to which r e l ative p i t c h prominence* is to be assigned. s misunderstanding was compounded of my c a r e l e s s wording a n & h e~ c~nfusion between overa l l and differential expectetian, o r so i t appears t6 me.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "door.)", "sec_num": null }, { "text": "S adds some examples of a t y p e where t h e r~~u n appears t o have l i t t i e semantic w e i g h t of i t s own, y e t c a r r i e s t h e accent, T h e y represent, I t h i n k , a balance between accent-for-prominence and 1. Bears a r e a c a n t~n k e r a u c s p e c i e s .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "door.)", "sec_num": null }, { "text": "2. Bears are a species t h a t is cantankerous.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "door.)", "sec_num": null }, { "text": "4; *Bears a r e -a lot t h a t ids cantankerous.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ". Bears are a cantankerous h t .", "sec_num": "3" }, { "text": "Bears are cantankerous. -example suggests t h a t you had b e t t e r heed the signal--if i t were", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "sai--'d -Look! -T h a t policeman is beckoning to you!, w i t h the verb de- v -", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "accented, the speaker would be implying something like 'Isn't t h a t", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "interest in&! But t h e most serious problem h a s t o do w i t h t h e notion of \"lev-els1I. By t r y i n g to t r e a t accent out of its intonational c o n t e x t j S leaves cases l i k e the following undefined, where clearly hot news is inv~lved (the speaker is r e p o r t i n g something he has just read):", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "Sa: did you see t h i s ? caped pris Some es oners have dered a wan.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "The ustressesty would be marked, I suppose, l i k e this: Some escaped prhsonershave murdered a man; but tb important thing is t h e na- The \"topic-comment\" notion is t m e by and m r g e , but again suffers by being divorced f r o m intonation. What can be tlassumed to be on t h e &dresscets mindl'may take e i t h e r of t w o forms. In one,", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "t h e topic is LO& resumptive--it adds no more information t h a n wou3d be contained in a de-accented pronoun. Typically t h e r e is a repetition, e i t h e r literal or by sense: In t h e first,the front window is a literal r e p e t i t i o n , and cont a i n s no p i t c h prominence whatever. I n t h e second, t h e speaker can assume t h a t t h e hearer will know t h a t your friends is t h e same as Jack, Marion, and Lucg, and again t h e r e need be no p i t c h prominence.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "I couldn't g e t in.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": ".", "sec_num": "5" }, { "text": "The trouble is that since p i t c h h e i g h t is ignored, t h e 'lstressl! markings w i l l be t h e same, for instance ont h e f r o n t window, as if an accent were a c t u a l l y t h e r e .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "The second kind of assumption t a k e s some familiarity for granted, but still finds it desirable to single out t h e t o p i c . T h i s is done b~r accenting it, but with a Lower pitch than the one on t h e comment:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "What were we t o do? Our g o l d had been confiscated. Our jewels had been stolen. O u r clothes were missing. We were al6ne in a strange land.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "That t h e speaker had had gold and jewels i s assumed to be known and f o i l . Stealing; is not suoh a noteworthy a c t t h a t Lewels cannot otltweigh 3t;. g i n a l l y , the l a s t sentence contains t h r e e accents no one of which is l e s s prominent t h i u l t h e others--an example of something that 3's treatment does n o t to,uch.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "I conclude by t a k i n g t h r e e passages from S and testing t h e m for adequacy. The question is w h e t h e r news and topib-comment works b e t t e r , or relative semantib w e i g h t .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "The first involves t h e example John survived, w h i c h S says \"seems impossible as an out-of-the-blue P e p o r t f t ( 9 0 ) . The probl'em is t o find a context in w h i c h survived carries relatively l i t t l e", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "information. If Noah had h a d a son named John, and John had f i n a lly made it t o the a r k alone; w i t h . some o t h e r stragglers, Noah might have exclaimed -~b h n survived! --Tbank God! Given t h e f l o o d , sur-viva1 was on everyone's mind, and no p a r t i c u l a r attentidn needed to be c a l l e d to it--it would almost have bee-n enough to. say John!", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "--", "sec_num": null }, { "text": "The second involves t h e -ample What's wrong w i t h ~o h n ? --His ", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Thank -God!", "sec_num": null }, { "text": "---- d & -was -run", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Thank -God!", "sec_num": null }, { "text": "The seconZ would be a strange sllswer, n o t because sf t h e form, but because the wrong thing is h i g h l i g h t e d .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "know. )", "sec_num": null }, { "text": "(~o u n s usually code more information t h a n vefbs--a lexical c o a t will show t h a t t h e r e are vastly more nouns t h a n verbs, revealing a tendency f o r nouns in w h i c h the predicate has t h e f o r m of a noun phrase\" (94-5).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "know. )", "sec_num": null }, { "text": "Messis de-accented because it is \"not very hounym'. True, it is In contrast totheauthors cited before Barbara Partee claims t h a t \"preverb\" is not a s y n t a c t i c category but r a t h e r a f e a t u r e", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "know. )", "sec_num": null }, { "text": "d o t --i t is what I", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "know. )", "sec_num": null }, { "text": "f A C T , O R S I N F L U E N C I N G T H E P L A C E M E N T O F E N G L I S H A D V E R B S I N R E L A T I O N T 0", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "know. )", "sec_num": null }, { "text": "shared by a l l s e n t e n c e adverbs. 7Jacobson, i n a n earlier work It is not even clear that there exists a syntactically i n t e r e s t i n g group of adverbs called preverbs worth extended study as a u n i t , Jacobson has n o t given any evidence t h a t t h e r e is. For the time being, however, I will assume that it is worthwhile to consider \"preverbs\" as a group. where unbiased e m p i r i c a l data is essential. bhis. This shows a weakness i n his theoretical framework. to JAcobson ' s . (19) I t h i n k t h a t Jacobson's t y p e of t e s t is not", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "A U X I L I A R I E S : A S T U D Y I N V A R I A T I O N", "sec_num": null }, { "text": "adequate and needs to be replaced by other types of considerqtions. (20)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "A U X I L I A R I E S : A S T U D Y I N V A R I A T I O N", "sec_num": null }, { "text": "Also on the same pqge Jacobson claims t h a t it i-s clear that F'almost\" and \"neaylyl' are sometimes sentence adverbs. In contrast many authors have assumed t h a t \"almost\" and \"nearly\" SPe. always nonsentential adverbs. '21) Of course given Jacobson's t e s t f o r sentence adverbg \"almost\" and \" n e a r l y \" are indeed sometimes senteace adverbs, but as I mentioned his t e s t is not. acceptab1.e.. ", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "A U X I L I A R I E S : A S T U D Y I N V A R I A T I O N", "sec_num": null }, { "text": "Edward K l i r n a , (1964) . \"Neqation in) E n g l i s h \" In The S t r u c t u r e (1976) . \"On-the Form and F u n c t i o ni n g of T r a n s f o r m a t i o n s i n L i n g u i s t i c I n q u i r y . \" Vol. 7 , (pp. 3-4 0 ) , and Peter C u l i c o v e r , ( 1 9 7 7 ) . \"An I n (1976) . An Iptroduuctioh t o E n g l i s h T r a n s f o r m a t i o n a l S y n t a x . London. Longman, p . 8 7 .", "cite_spans": [ { "start": 85, "end": 91, "text": "(1976)", "ref_id": null }, { "start": 283, "end": 289, "text": "(1976)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "( 5 )~e e", "sec_num": null }, { "text": "v a l i d E v a l u a t i o n Metric!' i n L i n g u i s t i c A n a l y s i s . V", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "( 5 )~e e", "sec_num": null }, { "text": "( 1 2 )~a n y doubts remain, however, t h e ndtion of t r a n s p o r t a b i l i t y , which aacobson h i m s e l f a c c e p t s , makes t h e claim t h a t \"preve~b\" i s a n interesting s y n t a c t i c u n i t ( 1 9 7 6 ) . pp. 23-51 for a d e q u a t e ways t o d i s t i n g u i s h between s e n t e n c e and n o n -s e n t e n t i a l a d v e r b s . (15) See E n r i q u e D e l a c r u z , ( 1 9 7 6 ) . \" F a c t i v e s and Proposition Level C o n s t r u c t i o n s i n Montague Grammar\" i n Moptague Grammar., ed. by Barbara P a r t e e , New York Academic Press, (pp. 1 7 7 -1 9 9 . ) . p. 179.", "cite_spans": [ { "start": 361, "end": 365, "text": "(15)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "( 5 )~e e", "sec_num": null }, { "text": "(l6)s-e for example Yehoshua B a r -H i l l e l , Jonathan Malino, and Avishai Margal it, \"On Logic and T h e o r e t i c a l L i n g u i s t i cs. \" ( 1 9 7 5 ) i n c u r r e n t T r e n d s i n L i n g u i s t i c s , V o l . 1 2 ; e d . by Thomas Sebeok. The Hague, Mouton. (pp. 37-101). p. 7 6 , B a r t s c h op. c i t . , Clark op. c i t . Frank Heny.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "( 5 )~e e", "sec_num": null }, { "text": "( 1 9 7 3 ) . \" S e n t e n c e and Predicate Modifiers i n E n g l i s h . \" i n S y n t a x and S e m a n t i c s , Vol. 2 '(pp. 217- ..,,,,.,., 83 HUMANITIES e b .~, e b , .~b , Tile operation of the recogni.tion automaton irlcliides a procedure which pi-cp:tses the description of !he pattern in such a way ns to exclt~dc the' influ;.nce of ~s o~~l o r p l f i c transformattons, that is 'the cl~a~lges of the i r~p~~t i-triage. A inethod i s propose(! which n~a k e s tllc recogr~itioi~ procedures indepc~iden t of tlie isomorphic t ransf or-illat io~ls. Ttie concept is based on a periodjcal rnod~tlation ir, the range of existence of the 1):1ttcr11 or 011 R ~x-i-iodicdl mocit~~nlion o f all the pitraltlcters distinguislietl. Al,plicat.io~is of this procedi~re are cti:,cussed with special referelice to the iso~liorphic fs:lnsf'orl~~:~tim of ihc type of \"gc~lcral i n t e~i s i t y illoclification\" of speccll ftqw for a probabilistic rccogt~itio~~ motlel. The lexical uniformity of a sample is considered to be no less importarit than its size, representiitiveness, etc. A method for determination of the lexical uniforlnity o f text with the aid of the computer is proposed (general flowchart vof the algorithnl i s given). The method is based on stntislical evaluation of the difference of the relative freqi~enci~es of word forms occuring i n two (or niore) portions of tcxt to bc compared as well as on the basis of thc evaluation of the general amount of coincident word forms. The application of the method is i1lustr:)tcd by compcrriso~l of two portio~~s of a hypothetiail text. \"['he word list of one portion covers up to 96% of the word occnrrenccs in tllc olllcr portiofi; only 70% of the coincidcnt words hove statistically insignificant differe~~ccs i n tile--relative frcquenclcs. This result is regarded as sufficient for concluding ns to the lexical unifortnity of B x t represented .", "cite_spans": [], "ref_spans": [ { "start": 136, "end": 182, "text": "..,,,,.,., 83 HUMANITIES e b .~, e b , .~b ,", "ref_id": null } ], "eq_spans": [], "section": "( 5 )~e e", "sec_num": null }, { "text": "by the two portions being compnred. 13 refs. For automatic syntactic analysis [he functions are dcscri bed of the brackct structures in English technical text on radiolocation (a f i l e of 2.000 scntences). Round brackcts are divided i i~ to unilateral and bjlatcral. U~~iiatcrnt round brackcts mark liun~bcrs and ictters in e~~llmerntions. Bilateral round brackets single out,: 1. a nurnbcr or ldtcr in enumeration ( i n the same way as unilateral ones, 2. a nunlbcr or letter i n a refere~~ce ( \\~s i~n l l y after a formula) which filnctions as the imnfiFying bausa [he ec-liolion for X is: l (=(l'r)\" ( 1 ) ; 3. a word (in particular, 31. 1 ab1)reviatint-Q 4. a scntetlce or il pair of sentences;", "cite_spans": [], "ref_spans": [ { "start": 597, "end": 611, "text": "(=(l'r)\" ( 1 )", "ref_id": null } ], "eq_spans": [], "section": "( 5 )~e e", "sec_num": null }, { "text": "5. .il phrase. Sentcnccs closed in brackets sllollrln bc sir~gled out as a segment. Therc are such. y-tlrascs which have t l~c snnll: f l~t~ctjon as tt\\r: sell tencc ; l n (~ shot~ld bc see~lrented correspondingly. The f o r r~~a l feature of this type of phrases is the vrrb to see it1 thc i~npcrative forin: 7'llc milnr i~~odulnfar(ssc~ Fig. 3) coir.sisis oJ.. or a psrsoilal vcrb wit11 a n S withailt :I preposition to the icft: Tire figure slrows rcrhere the ivrget is lnca,ed ( t h e cc2nler oaf the lurgct L s : .~)", "cite_spans": [], "ref_spans": [ { "start": 339, "end": 346, "text": "Fig. 3)", "ref_id": null } ], "eq_spans": [], "section": "LEXICOGRAPHY -LEXICOLOGY: TEXT I-IANDLING", "sec_num": null }, { "text": "Exalnples of phrosts that are n o t indcpc~ltlent scgmcnts are noun phrases Functioning as sentences and participle phr:lscs, Y7urthcrmore, rou~ld brackets as wcll as squ;lre brackets ant! braces are usctl in ma thematical expressions. (Teoriin vazyku i inzh. lingvisrika) , Lningrod: 66-16; , 1913. The notion. uf \"systemt' is considered as interprcted by solqe Soviet logicists. The set of linguistic objects is stated to be a system (at lcast \"badly organized\" or \"diffuse\" as cicfincd by V. V, Nniieov). The. linguost.atis~ic studies conducted by the \"Speech Statistics\" research group crnphalicolly do conform to the adopted approach to system analysis. 7' he grotrp lras pioi~eercd in co~~~prelre~~sive research on various linguistic systc~ns (in particular, owing to a uniform plan i~t~d method, some 80 frequency dict iu~~arics of various language subsets already cornpiled can be ir~tegruted in various combinations, etc.) 'l'hc author regrets that as, yet many linguisl~ havc failed to master the inelhod of statistirnl analysis. iekhnicheskoi inforrnutsii,\" 1973, srriia 2,) Symbol-by-langi~age replacement which is performed by a (context-free) si~hstilution does not depend 011 the con text of the symbol. A generalization o f context-f rce substi~ution is introduced so that the rcplncemerll i~ecomes associated with the context. It is proved that: (1) context-sensitive substitutions have, i n a sense, a wider scope limn context-free substitutipn~ (2) many classes of Inngunges src closcd under co~~tcxt-sensitive substitutions. 7 refs.", "cite_spans": [ { "start": 275, "end": 291, "text": "Lningrod: 66-16;", "ref_id": null }, { "start": 292, "end": 299, "text": ", 1913.", "ref_id": null } ], "ref_spans": [ { "start": 236, "end": 272, "text": "(Teoriin vazyku i inzh. lingvisrika)", "ref_id": null }, { "start": 1040, "end": 1085, "text": "iekhnicheskoi inforrnutsii,\" 1973, srriia 2,)", "ref_id": null } ], "eq_spans": [], "section": "LEXICOGRAPHY -LEXICOLOGY: TEXT I-IANDLING", "sec_num": null }, { "text": "(zagachakh rnatcmatichesltoi linyvistiki)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Objectives of Mathematical Linguistics", "sec_num": null }, { "text": "Linguistic problcnts requiring colist ruclion of rn:itl~cmatical l a~~g u ;~g e mqdels ore cliscussed: 1. Idwrtificatiun atid clas\\ification of the clcmentary li~igl~istic objects; 2. Study of thc types o f c~m h i n a b i l i ty of variotrs classes of morpl~cmes. Mathc~l~;~ticnl 11~ublcms fnci ng rnathcmdtical l'in'guistics arc c l~~c~~s s c t l :", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Objectives of Mathematical Linguistics", "sec_num": null }, { "text": "(a) 1nu1 hcmalical rlescr~p~ 1011 of the set of tcxts or a given Innguagc; study of thc i nv;~riancc condi tioils o f a language clescribcd will1 respect to crrtain transformaliotls; consttc~ctioti of at1 algorithm; (b) cornpitrlson o f sets of tekts its gc~icrnkd by certain strictly forin;ll inodcls-( c ) c;,nslruct~or~ of nlodcls for tril~~slat on of it) ngi~agcs; anti (d) siatislical st~ldies of Ii~t~gi~agcs. M;tlhernaiic;:J I~ngi~isiics IS intri~~sicnlly colinectid with tlicoretic:ll I~nguistics :~nd is i t s n:ltural c o m~~o~i c n t .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Objectives of Mathematical Linguistics", "sec_num": null }, { "text": "Matllelnatical l~ngi~istlcs deals wilh tnatlxnutir;ll nlodels uf li llguisiic plicno~nenn. The corlccp~ of \"111odel1 I llg\" is explainctl, disc~lss~tlg a rnngc of associntetl n o t i o~~s silcll as sin~ul;iliol~, aulom:ttion, a l g o r i~l~~n i~u t~~~~ and cornputer implerl~ent:ltion. 7 rcfs. Several independent kn~wledge sources (KSs) interact in a hypothesis-and-test paradigtn and are interfaced through a dytlalllic global data structure, tlre bliickbonrd. Hearsay I: KSa are activated in a lockstep sequence of three phases per cycle: pool, i~ypotlrrsize, fesi. All K% are activated at each phase and the next phasc doesn't start until all KSs have completcd the prcse~it one. A mediator module maintains thc blackboard, calcl~lates coinbitled ratings from ratings assigned to hypotheses by the i~ciividiinl KSs and decides wlren to stop and accept a solution (or to give up). The currently highest rctted hypollirsis is the one used as the context for the next cycle. Hearsay ti: The blnchi,oard has been extended nnd gcncralized to allow a) the representation of a l l levels of information (acoc~stic pl~onetic, syll~bic) in addition to thc lexical and sentence level of Hearsay I, and b) the explicit represenfiition of relationships among hypothesis. The overall control strategy is considerably Inore flexible than tlint for Hearsay I. ", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Objectives of Mathematical Linguistics", "sec_num": null }, { "text": "A method for-infor~nation efficiency lneast~re~nent of subject indexes to three separate series of VINITl's abstract journals is described, arltj the results of its evaluation are presented. The informatioil efficieiicy, y , i s shown to be convenicnt for evaluating the e f f~c i e n c y o f use of the sobject indcxes. The value of yl can be uscd as a r n~~s u r e of perfection of thc s~~b j e c t index vocabulary and, by that, as an objective basis for f u r t h e r improvement of the vocabulary. l text\" rnoclc!, c tc.). Different intcrprct;\\tiot~s of \"dcep structure\" and \"surf;tce structure\" arc considered. '1-he al~tHor suggests nsslgn ing the tern1 \"deep\" to semantic structure and \"surface\" to syntactic structure of the sentence. To map an intuitive theory into a mathematical one, corgcct defit~itior\\ of I e~r n s is rcquircd. A lcrln in physics shoi~ld be treated as corrcct if and only if for the quantity' ~t denotes thcre exists a ii~ensurenient proccA l t h o u g h t h i s", "num": null, "html": null }, "TABREF1": { "type_str": "table", "text": "Though everyone knows that bears a r e a", "content": "
and I;; t h i s t r e a t gerformative accent as a separ-
a t e e n t i t y . xf -= species is more accented t h a n cantankerous in 1,
i t is because of gradient. climax. It may e a s i l y be less accented,
t h a t is, lower in p i t c b
", "num": null, "html": null }, "TABREF2": { "type_str": "table", "text": "", "content": "
[ = accent] t h a n t h e subject
and t h e direct o b j e c t , if theye is one; in o t h e r words, apredi-
c a t e s receive lower stress [ = accent3 t h a n t h e i r arguments, ir-
respective of t h e i r linear position in surface structuren ( 5 2 ) .
T h i s concerns %irnple 'news ~entences'~', whme 'Itbe sgeaker as-
sumes no particular expectations with regard t a t h e information
content on t h e part of his audiencen (81). Examples: ~6 h n hi,$
sill; ~6 h n d%ed. The p r i n c i p l e d o e s not a p p l y t o sentences l i k e
--l r e a o y
been introduced.
T h e main problem is t h a t \"news sentencesw are not sufficiently
defined. Fake S t s p r i n c i p l e in t h e narrowest sense, whepe news
means what WcCawley has c a l l e d l1hot newsff--something being men-
tioned for t h e first time:
1. \\Phat do you have a g a i n s t Smithers? --Oh, h e ' d e s e r t e d t h e i r p o s t s .
", "num": null, "html": null }, "TABREF3": { "type_str": "table", "text": "", "content": "
-* -w---
Another consideratiofi is t h e nature of the subject. Even w i t h
hot news, indefinites may well not carry t h e main accent: Hey,
morn! Something's burn in^!; Hey, mom! Some p~ ---here ! -qcreaminq out _._
S t i l l ----L--
", "num": null, "html": null }, "TABREF5": { "type_str": "table", "text": "", "content": "", "num": null, "html": null }, "TABREF6": { "type_str": "table", "text": "to falsify any explicitly formulated grammar by considering examplesathat have not been understood by the t h e o r y constructor, and hence are not adequately accounted for by h i s theory. It i s He finds the relation between preverbs and auxiliares to be especially interesting. Hence to reiterate he intends to present a survey of how preverbs are actually used by carrying out a ment in surface structure; the taxonomic aim, to give the necessary discrete categories is subsidiary to this. I n his study he finds that there is no significant difference between American and British ~n g l i s h . The tests on which Jacobson bases h i s conclusions repre-", "content": "
common knowledge that as yet no adequate grammar for any natural language has been c o n s t r u c t e d , and t h a t everybody is a l o n g way His book begins with a s h o r t p r e s e n t a t i o n of t h e aims of t h e Work. quantative study, illustrated by examples. The s y n t a c t i c aim of cussion of some details of Jacobson's work. Jacobson remarks on p. 7 that the relations of preverbs to auxiliaries is especailly interesting. It seems to me, however, that he fails to distinguish between stylistic and syntactic or based t h e since I am not qualified to do so. I now want to turn to a d i s -results.
Jacobson also has a problem with his discrete categories. On
p. 9 and p. 49 he states that h i s taxonomy requires countable items
and cannot therefore use the concept of a continumn. Hence Jacobson
has to find a way to classify p r e v e r b s that gives rise to r a t h e r
sharp b o u n d a r i e s , but I do not think that he has succeeded in d 6 i n g
", "num": null, "html": null }, "TABREF7": { "type_str": "table", "text": ". . , , , 76 SPEECH UNDERSTANDING , , . . . 77 DOCUMENTATI M\\l t I WEXI NG . , 77 TRANSLATION . . . , , . . .. . . . . .", "content": "", "num": null, "html": null }, "TABREF8": { "type_str": "table", "text": "Device for Agton~atic Recognition of Phonemes (Ustrovistvo dlia recognition dev,ice which isolates.phone~nes ffonl the ~o n t~n u o u s speech flow. A pho~leme is defined as a set of physical realizations grouped together by their common .role in the verbal rl~essage i n a given language. Each physical realization of n p h o l~e n~e i s described by a set o f mensitrable features. which are tile intensities i n sepnr;lte freriucncy bands of the acuustic spectrunl. 'T'lle device has 30\u00b0frequency baiids erico~npnssing the range from 60 cycles to 8.3 kHz. A recogl~ition [nodel i s based 011 statistical tlleory of' p:tttern rccogr~iliot~. A flow chart of tilt (levice is presented, d e z~r i bing the operat ions o f the 1 ndividual con] ponents, the input amplifier, the band filters, amplifiers of the. fil'ter chant~els, detectors, and the fi~iictiunaI converter. -15 refs, Method for hflodification of the Overall Intensity o f Sound I tervals for zvukovvkh otrezkov vtseliakh ikh mashinnoso rassoznavaniia) J Machine Recognition (Ob odnorn sposobe izrnencniia. nbshchei tensivnosli", "content": "
PHCBVETICS-PHONOLOGY: RECOGNITIONMicrofiche 68: 51
A a vtorna ticheskogo ~aspoznavaniia fonern)
I. a. Strel'nikov
Questions on the Atln1,ysis of Speech (Yopr. nnaliza r e~h i ) ,~4 Tbilisi. \"Metsnierebd\": 145-
153, 1973.In t h i s issue and the next
AJCL will be featuring abstracts
of Russian language material
covering the l a s t four years.
Our regular coverage w i l l be
back to normal with the next
issue (the current 'sltnnp' is
the result of the pressures
imposed on the U b l i o g r a p t h e r
by dissertation writing) .
PHONETICS-PHONOLOGY: R~CO'GNITION
A Qllestions on ihe utlalysis of Speech (Vopr. Atluliza R r c / l i 4 Tbilisi \" M o i s~~i e r e h n \" : 154-159,
j973.
b , 84
ANALYSIS m~e~e e e m e e e m m~e e m 86
", "num": null, "html": null }, "TABREF9": { "type_str": "table", "text": "", "content": "
LEXICOGRAPHY-LEXICOLOGY: TEXT HANDLING
A Method of
", "num": null, "html": null }, "TABREF10": { "type_str": "table", "text": "On Some Procedures for Obtaining Reduced Word Codes (0 neltotorykh metodov polucheniia s vernutylth kodov slovThe importance of reducing the volu~ne of input information ljy efficacious codl~ig is strcsretl. The gcneral t~chnicli~es of word form code rcduction arc discusseti. The tecllliiclues are sllown to be efficient for retlncing the length o f ~npiit text hut they fail to provide for i~~~amblguous identification o f the word form. It is proved that in principle such n soltltion of the rcduction prohlem exists which cnsurcs u n n m biguous identi fica~ion. This solution is supposud lo be b.:~scJ on stotistic;~l-distrib~~tioi~ features of texl. 4 Russian verbal word forms can proceed in three stages: 1) isolation (in ttlr word form) of segments, 2) determination of t h e sytltagrnadc position of the segment depending on the meaning. it expresses and establishment o f the -type of meanin35 expressed by the segment, (ill this way morphs are singlor. out i n the word form), , 3) rncrke~ of individual positions in one o.rder if their nieanings coincide, but independent of the expression of these meanings. ~n * o r d e r is defined as thc sum of positions of thc ~norphs that are combiacd within one morphernc. Nine orderi are introduced for personnl verbs; they are lil~car~zed in strict sequence. Cc)inpulsory/optional presence, in a word form, of morphs of this or that order and colnbihtlbjIitics of individual orders arc discussed. Conclusions regarding the clependence of the meaning of a morph on its position in the word form are made. Exanlples of segment analysis of verbal words f o r t~i s arc given.An algorithm of morphological analysis of Russian is described that has been developed at the Mathematical Linguistics G r o \\~p of the Corn puting Ccntrc of L.en~ngratl U~liversi ly. 7'hc input dilta for the algorithm are the Russian cnIry word form and a tabular dictionary describing the Russian nrurphology in terms of a f i~l i t e at1to1n:t ton moclel. l'hc ta bulor ,dictionary fa11s iato a dictio~iery of stems and the tablc o f affixcs. 1 he alp01 itt1111 scans the nlterriative paths for generi~tion o f the word fornr. 'The structure of the tabular dlctionury i s described (each l inc'contai~ls a string genen ted) a s arc the preceding and su bsccjuen't states of the automaton. A list of morphologicai feilturcs for tlifferenl parls of spccch is giveti.For automatic semantic analysis, text units ate classified which signal the srlurce of message (equalized to an individual sentence): (1) reference to the aotl~or's knowledge (1 say that...);( 2 ) reference to collective knowledge (According to q~mntum theory,...); ( 3 ) reference to a particular source (from the Rol tzm:intl formula it i s obvious that ..., Ru therford supposed", "content": "
funktsional'nogo modelirovaniia rechevoi deitel'/losri,) Leningmd Universily: 140-144, 1973.
'nogo modelirovoriiia prechevoi i e i c t ' s i ) , Letlingrad U n i v e r s i !~: 145-174,
/973.
", "num": null, "html": null }, "TABREF11": { "type_str": "table", "text": "LINGUISTICS: METHODS:", "content": "", "num": null, "html": null }, "TABREF12": { "type_str": "table", "text": "Three common types of A l hubris are offered for cot~sideration. 1) Wist~jul rnr~ernatrics: Calling the maill loop o f , a program \"UNDERSTAND\" ~nakes it loo easy to beg the itnportant theoretical it un'derstnnd'? W1sllf111 labeling ot links Tea& to the s:me type of questionbegging (the ubiqt~itous IS-A link is discussed). 2) Unrrccirtral Innglrccge. A l workers tend both to oversimplify and ovcrglorify the proble~ns of NL processing. N", "content": "
", "num": null, "html": null }, "TABREF13": { "type_str": "table", "text": "", "content": "
", "num": null, "html": null }, "TABREF14": { "type_str": "table", "text": "", "content": "
", "num": null, "html": null } } } }