Phrase Structure Grammar Research Articles

An algebraic semantics for structured transition systems and its application to logic programs

We present a general methodology aimed at providing an algebraic semantics for a large class of formalisms. The methodology, which generalizes the algebraic treatment of Petri nets proposed by Meseguer and Montanari (1990) individuates three levels of description of a formalism (i.e., programs, structured transitions systems, and models), and defines two free constructions which generate in an automatic way the induced transition system and the free model of a program. These constructions are parametric with respect to the structure of states and transitions: instantiating them in various ways, different formalisms can be treated. The construction of the free model extends the algebraic structure of transitions to the computations of a program, thus producing a category having as arrows abstract computations, i.e., equivalence classes of concrete computations. Interestingly, the equivalence relation induced on computations captures some basic properties of true concurrency. Moreover, by general categorical properties the construction of the free model is compositional w.r.t. various forms of program union. As a running example, we apply the methodology to phrase structure grammars, while the main case study is logic programming. For both formalisms the free model is shown to include all the computations at a natural level of abstraction: the free model of a grammar includes all its derivation trees as arrows, while the arrows of the free model of a logic program are parallel refutations.

A spoken language translator for restricted-domain context-free languages

An effort is underway at AT&T Bell Laboratories and Telefónica Investigación y Desarrollo to build a restricted domain spoken language translation system, which we call VEST (Voice English/Spanish Translator). The eventual goal is a voice output translator which is speaker-independent, and has a vocabulary of several thousand words covering a specific application. This paper describes the first step of our research, a system which recognizes two speakers in each of Spanish and English and is limited to some four hundred words. The key new idea is that the speech recognition and the language analysis are tightly coupled by using the same language model, an augmented phrase-structure grammar, for both.

A computational representation for generalised phrase-structure grammars

Some modifications are suggested to recent (1985) generalised phrase-structure grammar which make the formalism more suitable to computational use, and at the same time provide a clear and elegant redefinition for parts of the formalism which are standardly complex and ill-defined. It is shown how the feature-instantiation principles can be represented as explicit rules in a format similar to metarules, and how a grammar of four parts, immediate-dominance rules, linear-precedence rules, metarules, and these new propagation rules, can be used to produce the ordinary GPSG analyses of English. Methods of computational implementation are discussed, in particular it is suggested that the parts of the grammar are most conveniently interpreted as instructions as to how to produce a set of context-free phrase-structure rules which can be used with a simple left-corner parser.

Formalisms for grammatical knowledge representation

Computational linguists require descriptively powerful, computationally effective formalisms for representing grammatical information or knowledge. A wide variety of formalisms have been employed in natural language processing systems over the past several decades, including simple phrase structure grammars, augmented transition networks, logic grammars and unification-based grammar formalisms. Until fairly recently however, comparatively little attention has been given to the issues which underly good grammar formalism design in computational linguistics.

Books and journals received

Atkinson, Martin, David Britain, Harold Clahsen, Andrew Radford, and Andrew Spencer. 1999. Linguistics: An introduction. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 438 pp., $59.95 (cloth), $22.95 (paper). Barrett, Martyn, ed. 1999. The development of language. Philadelphia: Psychology Press. 415 pp., $54.95 (cloth). Berlin, Brent, and Paul Kay. 1999. Basic color terms: Their universality and evolution. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 196 pp., $22.95 (paper). Blench, Roger, and Matthew Spriggs, eds. 1999. Archaeology and language III: Artefacts, languages and texts. New York: Routledge. 301 pp., $130.00 (cloth). Borsley, Robert D., and Adam Przepiorkowski. 1999. Slavic in Head-Driven Phrase Structure Grammar. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 345 pp., $64.95 (cloth), $24.95 (paper). Bouma, Gosse, Erhard Hinrichs, Geert-Jan M. Kruijff, and Richard Oehrle. 1999. Constraints and resources in natural language syntax and semantics. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 338 pp., $74.95 (cloth), $27.95 (paper). Butt, Miriam, Tracy Holloway King, Maria-Eugenia Nino, and Frederique Segond. 1999. A grammar writer’s cookbook. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 337 pp., $59.95 (cloth), $22.95 (paper). Byram, Michael, and Karen Risager. 1999. Language teachers, politics and cultures. Clevedon, England: Multilingual Matters. 206 pp., $39.95 (cloth). Carr, Philip. 1999. English phonetics and phonology: An introduction. Oxford: Blackwell. 169 pp., $24.95 (paper). Cattell, Ray. 2000. Children’s language: Consensus and controversy. London: Cassell. 276 pp., $82.95 (cloth), $26.95 (paper). Cenoz, Jasone, and Fred Genesee, eds. 1999. Beyond bilingualism: Multilingualism and multilingual education. Clevedon, England: Multilingual Matters. 269 pp., $39.95 (paper). Choi, Hye-won. 1999. Optimizing structure in context. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 223 pp., $59.95 (cloth), $22.95 (paper). Coates, Richard. 1999. Word structure. New York: Routledge. 100 pp., $16.99 (paper). Cognition. 70.2 (March 1999), 70.3 (April 1999), 71.1 (May 1999), 71.2 (June 1999), 71.3 (July 1999), 72.1 (August 1999), 72.2 (September 1999), 72.3 (October 1999), 73.1 (November 1999), 73.2 (December 1999), 73.3 (December 1999), 74.1 (January 2000), 74.2 (February 2000), 74.3 (March 2000), 75.1 (April 2000). Jacques Mehler, ed. The Netherlands: Elsevier. Cohen, Ariel. 1999. Think generic! The meaning and use of generic sentences. Stanford, Calif.: CSLI Publications. [Distributed by Cambridge University Press.] 208 pp., $59.95 (cloth), $22.95 (paper). Devitt, Michael, and Kim Sterelny. 1999. Language and reality: An introduction to the philosophy of language. 2nd ed. Cambridge, Mass.: MIT Press. 342 pp., $30.00 (cloth). Dixon, R. M. W., and Alexandra Y. Aikhenvald. 1999. The Amazonian languages. Cambridge: Cambridge University Press. 446 pp., $69.95 (cloth).

Normal forms for phrase-structure grammars

Quelques nouvelles formes normales pour grammaires de type 0 sont presentees. Chaque langage recursivement enumerable peut etre engendre par une grammaire ou les regles «context-free» sont de forme S→v, ou S est le symbole initial non terminal. En ce qui concerne les regles «non context-free», on a l'une des cinq situations suivantes: ou bien deux regles du type AB→e, CD→e, ou deux regles du type AB→e, CC→e, ou deux regles du type AA→e, BBB→e, ou une regle du type ABBBA→e, ou regle du type ABC→e. Dans tous les casx aucun symbole non terminal additionnel n'est necessaire

REFERENT GRAMMAR MACHINE TRANSLATION WITH DISCOURSE ANAPHORIC EXTENSION

While unification grammars have been in existence for some time and have given birth to grammar formalisms such as Generalized Phrase Structure Grammar (GPSG), it is only recently that computational tools such as Prolog Definite Clause Grammars (DCG) have allowed direct implementation of such grammars on personal computers. A very promising new formalism Referent Grammar is an outgrowth of these events. Referent Grammar (RG) is a sub‐class of GPSG which can be directly implemented in Prolog DCG and provides weak generative power at least equal to GPSG. This paper describes the Referent Grammar formalism and the research MT system developed using RG and anaphoric pronoun resolution extensions to RG. It also describes plans for combining sublanguage analysis with RG.

Linguistic processing related to speech understanding in spicos II

spicos II enables an on-going dialogue to take place with an office data-bank. This represents a further development of spicos I that, linguistically, was confined to a simple question-answer scheme. spicos II is speaker-adaptable and incorporates a vocabulary of some 1,200 words. The present article gives an overview of the elements connected with linguistic analysis. These elements are governed by a dialogue module that controls the dialogue and, by means of appropriate follow-up questions, avoids possible communications difficulties. The syntactic analysis is based on an augmented phrase-structure grammar. At the same time, semantic constraints are checked by means of a semantic network. Syntactic structure, together with semantic features and the results of the resolution of anaphoric bindings and a separate formal representation of discourse referents, is used to build formal-logic semantic representations of utterances. User presuppositions are also represented. From these formal representations date-bank queries are generated.

Grammar and Technical Writing

Harris argues that linguistic theory is useful for solving certain problems encountered in technical writing theory and pedagogy [1]. However, he undermines his purpose by introducing irrelevant distinctions between competing syntactic theories (Generalized Phrase Structure Grammar and Transformational Grammar) and by failing to exploit the full potential of the few applications he mentions. The passive rule is a case in point. It not only constitutes an operational test for identifying passive sentences, it also contributes to the flow of discourse by rearranging both thematic roles (e.g. agent and patient) and given/new information. The passive rule is only one of a class of noun phrase-moving operations that technical writing specialists may find useful.

The logic of unification in grammar

By unification, we understand a family of algorithms employed by compu tational versions of certain grammar formalisms to combine information in feature structures. The use of these formalisms has become widespread, and several extensions to the basic notion of feature structure have been proposed. Although algorithms for unification of these extended feature structures have been written, they are complicated, and a precise model of feature structures is desirable to give an adequate specification of what the algorithms do. We have developed a model in which descriptions of feature structures can be regarded as logical formulas, and interpreted by sets of directed graphs which satisfy them. We identify a feature structure with such a directed graph, but our mathematical work is facilitated by considering the graphs to be transition graphs for a special type of deter ministic finite automaton. This semantics for feature structures extends the ideas of Pereira and Shieber [11], by providing a way to model feature values which are speci fied by disjunctions and non-local path values embedded within disjunc tions. Our interpretation differs from that of Pereira and Shieber by using a logical model in place of a denotational semantics. The model yields a calculus of equivalences between formulas, which can be used to simplify them. A similar use of logic to describe feature structures was first pro posed in Generalized Phrase Structure Grammar (GPSG)[3], in order to describe feature co-occurrence restrictions. Our formulation, which was developed independently, is for the purpose of understanding the proper ties of unification. Recently Gazdar et al. [2] have extended their logic in order to give uniform descriptions of linguistic categories. The two logics have much in common when presented formally, and it should be possible to combine them in a uniform way. This, however, will be left for future work.

A string pattern recognition approach to polygon clipping

This paper describes a new approach to the clipping of a line segment by a simply-connected polygon. The polygon is not restricted to a convex polygon. The vertices and other key points of the polygon are given labels according to their relative locations to the line segment. When these labeled points are traversed in a predetermined order, a string of symbols is obtained. Different clipping of the line segment results in distinct string patterns of the symbols. By recognizing the different patterns, the nature of the intersection can be identified. We are interested in identifying four different types of clipped line segments: the interior line, the exterior line, the edge line, and the intersecting line. Using the concept of topologically equivalent polygon developed in this paper, a clipped polygon can be reduced to the simplest form without changing the type of the line segment. The string of symbols taken from this simplified polygon is easily classified into different line types. To expedite the recognition of string patterns, a phrase-structured string grammar is constructed. The major application of the clipping procedure is in the area of computer vision. A system capable of recognizing partially occluded planar objects uses the procedure described in this paper to identify the types of line segments connecting any two vertices in an object modeled as polygon. Other uses include hidden line removal in computer graphics and layout design in VLSI

Reflections on the history of dependency notions in linguistics

SummaryThis paper outlines the history of dependency notions from Antiquity to the present century. Although the notion of syntactic dependency was unknown in Antiquity, the idea of semantic dependency was foreshadowed in early definitions of the minor parts of speech, i.e., parts of speech other than the noun and the verb. In part, this happened because logicians had originally posited only the two major parts of speech, and grammarians then formulated their definitions of the minor parts of speech in relation to those of the noun and the verb. The adverb, for instance, was defined as augmenting or diminishing the meaning of the verb. The first writer who used (if not coined) a special term to refer to the notion that some words specify or ‘determine’ the meanings of the subject noun and the verbal predicate was Boethius (ca. 500 A.D.), and in this way the notion of ‘determination’ was launched. As a result of the subsequent popularity of Boethius’s logical works, ‘determination’ was adopted and extensively utilized by Latin grammarians from the 12th century on. In the 13th century, it was complemented by the term ‘dependency’, which was the logical converse of ‘determination’. Grammarians claimed that a dependency relation exists between the members of all constructions. The vogue of ‘dependency’ declined even before the advent of Renaissance humanism, while ‘determination’ survived. In the early modern period, the terminological repertory expanded. Thus, in the 18th century, French grammarians coined the terms ‘modification’ (Buffier) and ‘complement’ (Du Marsais). The 20th century has been marked by a further increase of new terms. Inspired by Tesnière’s posthumousElèments de syntaxe structurale(1959), some linguists have also proposed formalized dependency theories as alternatives to phrase-structure grammar.

A new set of constraint-free character recognition grammars

A general pattern-recognition procedure for application to unconstrained alphanumeric characters is presented. The procedure is designed to allow hierarchical redescription of the input images in terms of significant elements only, and formal developments are given within the framework of elementary phrase-structure grammars. The extraction of the primitives associated with the terminal vocabularies of the grammars is always deterministic, and the productions of the parsers are characterized by a significant degree of topological fidelity. Preliminary experimental results indicate recognition rates comparable to the state of the art, but with a considerable reduction in computing time.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Mathematics of Language

1. Names and Addresses of Contributors 2. Introduction (by Manaster-Ramer, Alexis) 3. Computational Complexity, Mathematical Linguistics, and Linguistic Theory (by Berwick, Robert C.) 4. Exceptions to Generic Generalizations (by Carlson, Greg N.) 5. The Structure Underlying a Semantic Domain (by Davis, Anthony R.) 6. Binary Rules and Non-Binary Trees: Breaking Down the Concept of Phrase Structure (by Jensen, Karen) 7. An Introduction to Tree Adjoining Grammar (by Joshi, Aravind K.) 8. The Notion 'Rule of Grammar' Reconsidered (by Kac, Michael B.) 9. Unbounded Dependencies and Subjacency in a Tree Adjoining Grammar (by Kroch, Anthony S.) 10. A Topological Approach to Structural Equivalence of Formal Languages (by Kuroda, S.-Y.) 11. On the Design of Finite Transducers for Parsing Phrase-Structure Languages (by Langendoen, D. Terence) 12. Graphs and Grammars (by Marsh, William) 13. Discontinuity and Phrase Structure Grammar (by Ojedo, Almerindo E.) 14. Sources of Intractability in GPSG Theory (by Ristad, Eric Sven) 15. Formal Properties of Head Grammars (by Roach, Kelly) 16. Finding Natural Languages a Home in Formal Language Theory (by Rounds, William C.) 17. Theories of Language Learnability (by Savitch, Walter J.) 18. Remarks on Linguistic Semantics (by Thomason, Richmond H.) 19. On the Progression from Context-Free to Tree Adjoining Languages (by Vijay-Shanker, K.)

Generalized Phrase Structure Grammar.

Generalized Phrase Structure Grammar provides the definitive exposition of the theory of grammar originally proposed by Gerald Gazdar and developed during half a dozen years' work with his colleagues Ewan Klein, Geoffrey Pullum, and Ivan Sag. This long-awaited book contains both detailed specifications of the theory and extensive illustrations of its power to describe large parts of English grammar. Experts who wish to evaluate the theory and students learning GPSP for the first time will find this book an invaluable guide.The initial chapters lay out the theoretical machinery of GPSP in a readily intelligible way. Combining informal discussion with precise formalization, the authors describe all major aspects of their grammatical system, including a complete theory of syntactic features, phrase structure rules, meta rules, and feature instantiation principles. The book then shows just what a GPSP analysis of English syntax can accomplish. Topics include the internal structure of phrases, unbounded dependency constructions of many varieties, and coordinate conjunction a construction long considered the sticking point for phrase structure approaches to syntax.The book concludes with a well developed proposal for a model theoretic semantic system to go along with GPSP syntax. Throughout, the authors maintain the highest standards of explicitness and rigor in developing and assessing their grammatical system. Their aim is to provide the best possible test of the hypothesis that syntactic description can be accomplished in a single-level system. And more generally, it is their intention to formulate a grammatical framework in which linguistic universals follow directly from the form of the system and therefore require no explicit statement. Their book sets new methodological standards for work in generative grammar while presenting a grammatical system of extraordinary scope.

An HPSG approach to Welsh

Welsh differs from English in a number of ways. The most obvious point is that it is a VSO language, but it also has distinctive agreement phenomena and clitics. For this reason, it is natural to ask of any theory of syntax that has been developed primarily on the basis of English: how can it handle Welsh? Welsh has had fairly extensive attention within the Government-Binding theory (see, for example, Harlow, 1981; Sproat, 1985; Sadler, 1988, and Hendrick, 1988). It has also had some attention within Generalized Phrase Structure Grammar (GPSG) (see Harlow, 1983; Borsley, 1983; 1988a). In this paper, I will consider how some of the central features of Welsh can be accommodated within Head-driven Phrase Structure Grammar (HPSG). This is a framework developed over the last few years by Carl Pollard, Ivan Sag and others, which seeks to combine the insights of GPSG, categorial grammar and certain other theories (see Pollard, 1985, 1988; Sag &amp; Pollard, 1987, and Pollard &amp; Sag, 1988). In fact, I will be mainly concerned with the version of HPSG developed in Borsley (1986, 1987, 1988 b), but I will also have something to say about standard HPSG.

Multiple heads and minor categories in generalized phrase-structure grammar

Multiple heads and minor categories in generalized phrase-structure grammar

Assessing Acceptability in Translated Children's Books

The purpose of the study outlined in this article is to compare the acceptability of style in two contemporary Finnish translations of L. Frank Baum's The Wizard of Oz with special reference to the effect of sentence structure on acceptability. For the purposes of this study, acceptability is defined in terms of the norms and expectations prevailing in the language and style of Fin-nish children's fiction which include the requirements of a high degree of readability and natural style. The hypothesis is that the use of complex nonfinite con-structions in one translation decreases readability and consequently lowers acceptability, whereas the other translation is expected to be more readable and acceptable due to its natural, dynamic style characterized by simple finite cons-tructions. The syntactic structures relevant to the readability of the translations are analysed with the help of two linguistic models: a modification of phrasestructure grammar and a propositional model. Three empirical tests, i.e. a cloze test, subjective assessment and a reading test, provide information about the readability and acceptability of the translations.

Coordination and grammatical relations

The most serious recent work on the theory of coordination has probably been done in terms of three theories of grammatical structure: Generalized Phrase Structure Grammar (GPSG–see especially Gazdar, 1981; Gazdaret al., 1982; 1985; Saget al., 1985; Schachter &amp; Mordechay, 1983), Categorial Grammar (CG–see especially Steedman, 1985; Dowty, 1985) and Transformational Grammar (TG–notably Williams, 1978, 1981; Neijt, 1979; van Oirsouw, 1985, 1987). Each of these approaches is different in important respects: for instance, according to whether or not they allow deletion rules, and according to the kinds of information which they allow to be encoded in syntactic features. However, behind these differences lies an important similarity: in each case the theory concerned makes two assumptions about grammatical structure in general (i.e. about all structures, including coordinate ones):I The basic syntagmatic relations in sentence-structure are part-whole relations (consituent structure) and temporal order; note that this is true whether or not syntactic structure is seen as a ‘projection’ of lexical properties, since these lexical properies are themselves defined in terms of constituent structure and temporal order.

Linguistics, Technical Writing, and Generalized Phrase Structure Grammar

Linguistics has been largely misunderstood in writing pedagogy. After Chomsky's revolution, it was widely touted as a panacea; now it is widely flogged as a pariah. Both attitudes are extreme. It has a number of applications in the writing classroom, and it is particularly ripe for technical writing students, who have more sophistication with formalism than their humanities counterparts. Moreover, although few scholars outside of linguistics are aware of it, Transformational Grammar is virtually obsolete; most grammatical models are organized around principled aversions to the transformation, and even Chomsky has little use for his most famous innovation these days. Among the more recent developments is Generalized Phrase Structure Grammar, a model with distinct formal and pedagogical advantages over Chomsky's early transformational work.