Pronominal Anaphora Research Articles

Splitting the reference time: the analogy between nominal and temporal anaphora revisited

The analysis in Partee (1984) of quantified sentences, introduced by a temporal connective, gives the wrong truth conditions when the connective is before or after. In this paper, we show how splitting the different roles of Reichenbach's reference time may be used in order to solve this problem. We further enhance the analogy between pronominal and temporal anaphora, by proposing an analog of plural NP-anaphora in the form of temporal anaphora involving multiple event antecedents and an analog of an E-type analysis of pronouns in the analysis of quantified narrative discourse. In Partee (1973), Partee introduced the notion of temporal anaphora to account for ways in which temporal expressions depend on surrounding elements in the discourse for their semantic contributio'n to the discourse. This analogy between NP anaphora and temporal anaphora has been an important factor in the analysis of temporal expressions in natural language semantics and a driving force in the early stages of the development of Discourse Representation Theory (DRT)2 (Kamp 1981, Heim 1982, Kamp & Reyle 1993) in particular. A factor vital to the analysis of temporal anaphora which is absent from the analysis of NP anaphora is the well-known notion of reference time (Reichenbach 1947). Thus the analysis of temporal anaphora has an added level of complexity relative to analyses of nominal anaphora. This may be seen in the unified treatment of temporal expressions in discourse given by Hinrichs (1981), Partee (1984), and Hinrichs (1986). In these works, a heavily burdened reference time plays an important role. In this paper, we examine temporal anaphora under two kinds of special circumstances: the interaction of temporal anaphora with quantification over eventualities and anaphoric links to multiple eventualities. Such circumstances, while interesting in their own right, are also good test-beds for theories of the semantic interpretation of temporal expressions.

An empirical study on the generation of anaphora in Chinese

The goal of this work is to study how to generate various kinds of anaphora in Chinese, including zero, pronominal, and nominal anaphora, from the syntactic and semantic representation of multisent...

The effect of discourse factors on backward anaphora

AbstractThis article investigates the conditions under which in English a pronoun and a following noun may be understood as having intended coreference when the pronoun lacks an immediate antecedent. The analysis presented supports the claim that, with the exception of certain syntactic restrictions, the possibility of such coreference is determined by discourse factors. The sensitivity of pronominal anaphora to various discourse-dependent stress patterns and syntactic structures is presented as evidence supporting the analysis offered. Finally, it is argued that apparent counterexamples represent a special use of pronouns and therefore do not refute the proposed framework.

Use of Domain Knowledge in Resolving Pronominal Anaphora

Abstract. The research reported here has been conducted in the context of the Plinius project, which aims at semi-automatic knowledge acquisition from short natural-language texts. In this framework, a system has been developed for finding the antecedents of pronominal anaphora, in particular 'it'- and 'its'- anaphora. The anaphora resolution module operates on parser output and can make use of information generated by the parser; the lexicon gives the conceptual representations corresponding to the words. The algorithm for anaphora resolution involves three steps: (i) Assemble: construct a list of discourse entities (DEs); (ii) Identify: identify anaphoric DEs; (iii) Select: select, for each anaphoric DE, another DE from the list of DEs as its antecedent. The third step applies four constraints, i.e. rules to which a DE must conform in order to be a valid candidate: (a) semantic type agreement; (b) number agreement; (c) projection constraint; (d) conceptual compatibility. Constraints (a, b, c) are linguistic, while (d) is domain-related. The algorithm has been tested on three texts. It turns out that applying (d) before (a, b, c) considerably improves efficiency.

L'interface pragmatique-sémantique et l'interprétation du discours

N. Asher, Université du Texas, Austin : « L'interface pragmatique-sémantique et l'interprétation du discours » The present paper extends dynamic semantics to take account of discourse structure and its semantic effects on discourse interpretation. It offers an introductory overview to the theory known as Segmented Discourse Representation Theory or SDRT, which has been applied to several domains : pronominal and temporal anaphora, lexical disambiguation, and spatial inferences in text.

SYNTAX AND LEXICAL STATISTICS IN ANAPHORA RESOLUTION

We describe a syntactically based salience algorithm for pronominal anaphora resolution and a procedure for reevaluating the decisions of the algorithm on the basis of statistically modeled lexical semantic/pragmatic preferences. We report the results of an extensive blind test of both systems on computer manual text. We discuss the implications of these results for the comparative roles of syntactically defined salience and statistically measured lexical preference in determining the references of pronouns in text.

Conceptual Reference Points: A Cognitive Grammar Account of Pronominal Anaphora Constraints

This paper presents an analysis of the constraints on pronominal anaphora in English within the framework of Cognitive Grammar in terms of semantic distinctions between pronouns and full noun phrases. Semantic notions of prominence and conceptual interconnection are used to develop a model of CONCEPTUAL REFERENCE POINTS which defines the contexts within which coreference is acceptable or unacceptable. The analysis provides a conceptual-semantic account of the 'core' anaphora facts which have previously been explained in terms of structural notions such as c-command, as well as certain data which have long been problematic for structural accounts.*

Accessibilité Pronominale Des Dét. N1 de (DÉT.) N2

The analysis of pronominal anaphora in dét. N1 de (dét.) N2 requires on the one hand a preliminary study of anaphora and of the relations supported by de, and on the other hand, an examination of referential functioning of propositional constructions. The analysis of the relations supported by de is seldom enough to determine whether the whole structure, or just a part of it, is favoured. Thus, the notion of "head" which underlies the principle of pronoun accessing and the preselective clauses must take into account the way of determining the different components of the Dét. N1 de (dét.) N2. The role of determiners is therefore essential : those introducing N2 just as well as those preceding the whole prepositional construction because they all can contribute to making the whole complex structure, or part of it, salient. Examining the distribution of the determiners does not allow us to list all the possible and impossible anaphoras ; it simply aims at building up an access and preference scale based on the notions of salience and accessibility. The identification of references requires a treatment taking many other factors into account: conceptual agreement in gender, immediate or remote contexts, argumentative structure, shared information, extra-linguistic and encyclopedic knowledge...

An algorithm for pronominal anaphora resolution

This paper presents an algorithm for identifying the noun phrase antecedents of third person pronouns and lexical anaphors (reflexives and reciprocals). The algorithm applies to the syntactic representations generated by McCord's Slot Grammar parser and relies on salience measures derived from syntactic structure and a simple dynamic model of attentional state. Like the parser, the algorithm is implemented in Prolog. The authors have tested it extensively on computer manual texts and conducted a blind test on manual text containing 360 pronoun occurrences. The algorithm successfully identifies the antecedent of the pronoun for 86% of these pronoun occurrences. The relative contributions of the algorithm's components to its overall success rate in this blind test are examined. Experiments were conducted with an enhancement of the algorithm that contributes statistically modelled information concerning semantic and real-world relations to the algorithm's decision procedure. Interestingly, this enhancement only marginally improves the algorithm's performance (by 2%). The algorithm is compared with other approaches to anaphora resolution that have been proposed in the literature. In particular, the search procedure of Hobbs' algorithm was implemented in the Slot Grammar framework and applied to the sentences in teh blind test set. The authors' algorithm achieves a higher rate of success (4%) than Hobbs' algorithm. The relation of the algorithm to the centering approach is discussed, as well as to models of anaphora resolution that invoke a variety of informational factors in ranking antecedent candidates.

Intrasentential anaphora

Abstract In this paper, two types of intrasentential pronominal anaphora are examined: anaphora in a complex sentence and anaphora in a simple sentence. The theoretical framework is the Prague Functional Generative Description with its main principles: dependency grammar and a systematic description of the topic-focus articulation of the sentence including its relation to surface phenomena such as intonation. In this way, it is possible to describe a complete paradigm of types of sentences with anaphora, including those cases which have not yet been accounted for or which have been interpreted erroneously. The present approach is opposed to exclusively syntactic approaches (based on constituency/configurationality) as well as to exclusively pragmatic approaches.

Modal subordination and pronominal anaphora in discourse

Modal subordination and pronominal anaphora in discourse

What, when, and how?: Questions of immediacy in anaphoric reference resolution

Abstract This paper considers the question of the immediacy of interpretation of anaphors. Two aspects of immediacy are considered: (1) immediacy in terms of the initiation of processes that might be considered as supporting resolution, and (2) immediacy in terms of achieving resolution as an outcome. The functional justification, and the logic of these aspects are considered. The bulk of the paper is a review of pertinent experimental evidence. On the basis of studies both on speech and reading comprehension, we argue that the evidence favours immediate initiation. However, it is argued that immediacy of outcome must take into account the variety of different forms of definite anaphora: proper names, full definite descriptions, and pronouns. Whereas there is good evidence for early resolution in the case of the first two, it does not hold for all cases of pronominal anaphora. Furthermore, evidence from a variety of stylistic phenomena indicate the need to disentangle carefully the semantic constraints in...

Language Transfer in Language Learning

1. List of Contributors 2. Preface 3. Introduction (by Gass, Susan M.) 4. A Role for the Mother Tongue (by Corder, S. Pit) 5. A New Account of Language Transfer (by Schachter, Jacquelyn) 6. Verification of Language Transfer (by Ard, Josh) 7. Nonobvious Transfer: On Predicting Epenthesis Errors (by Broselow, Ellen) 8. Language Transfer and the Acquisition of Pronominal Anaphora (by Gundel, Jeanette K.) 9. Transfer and Variability of Rhetorical Redundancy in Apachean English Interlanguage (by Bartelt, H. Guillermo) 10. Discourse Accent in Second Language Performance (by Scarcella, Robin C.) 11. Discourse Functions in Interlanguage Morphology (by Jordens, Peter) 12. Prior Linguistic Knowledge and the Conversation of the Learning Procedure: Grammaticality judgments of Unilingual and Multilingual Learners (by Zobl, Helmut) 13. Language Transfer And Fossilization: The Multiple Effects Principle (by Selinker, Larry) 14. Universal Grammar: Is it Just a New Name for Old Problems? 15. Afterword

Anaphora by pronouns

An adequate conception of anaphora is still a desideratum. Considering the anaphoric use of third-person personal pronouns, the present study contributes to the solution of the question of what anaphora is. Major tenets of generative approaches to pronominal anaphora are surveyed; descriptive and methodological problems with transformational as well as interpretive treatments are discussed. The prevailing assumption that anaphora is a syntactically based phenomenon is shown tobe inadequate. In particular, it is argued that pronominal anaphora does not constitute a case of eilher a syntactic ( agreement) relation or a semantic ( coreference) relation between antecedents and anaphors, i.e. linguistic expressions. Infact, there is no grammatical antecedent-anaphor relation that is essential to the description of pronouns. Pronouns are to be treated in their own right rather than by recourse to supposed antecedents. An account of the use of pronouns has to be based on a notion of speaker reference and on a unified description of lexical entries for pronouns that specify their meanings. Sampie entries for English are suggested. It is emphasized that pronoun meanings rejlect social, not biological, classifications of possible referents. To the extent that pronouns are used according to morphosyntactic features, as in languages like German or French, lexical entries for pronouns should specify the pronouns' 'associative potential'. Associative potential has the samefunction as conceptual meaning, viz. delimiting the associated extension. In addition to this, pronouns turn out to differ from 'normal definite nominals' only in the low conceptual content of their meanings. Pronoun occurrences that apparently agree with and are coreferential with referential antecedents are found to form a restricted subclass of pronoun use in generat as weil as of anaphoric pronoun use. Thus one must refrainfromforcing each and every pronoun occurrence into this mold. Instead, anaphora by pronouns is characterized as a type of use where pronouns serve to refer to referents that the speaker considers to be retrievable from the universe-of-discourse.

Syntactic Argumentation and the Structure of English

Syntactic Argumentation and the Structure of English (SASE) presents the major theoretical developments in generative syntax and the empirical arguments motivating them. Beautifully and lucidly written, it is an invaluable resource for working linguists as well as a pedagogical tool of unequaled depth and breadth. The chief focus of the book is syntactic argumentation. Beginning with the fundamentals of generative syntax, it proceeds by a series of gradually unfolding arguments to analyses of some of the most sophisticated proposals. It includes a wide variety of problems that guide the reader in constructing arguments deciding between alternative analyses of syntactic constructions and alternative theoretical formulations. Someone who has worked through the problems and arguments in this book will be able to apply the skills in argumentation it develops to novel issues in syntax. While teaching syntactic argumentation, SASE covers the major empirical results of generative syntax. Its contents include: Transformations in single-clause sentences; Complementation and multi-clause transformations; Universal principles governing rule interaction: the cycle and strict cyclicity; Movement rules; Ross' constraints; Pronominal reference and anaphora. SASE is an important book for several different audiences: for students, it is an introduction to syntax that teaches argumentation as well as a wide range of empirical results in the field; for linguists, it is a sourcebook of classical analyses and arguments, with some new arguments bearing on classical issues; and, for scholars, teachers, and students in related fields, it is a comprehensive guide to the major empirical and theoretical developments in generative syntax. SASE contains enough material for a two-semester or three-quarler sequence in syntax. Because it assumes no previous background, it can be used as the main text in an introduction to syntax. Since it covers a wide range of material not available in other texts, it is also suitable for intermediate and advanced syntax courses and as a supplementary source in more specialized courses and courses in other disciplines. A storehouse of classical and original arguments, SASE will prove to be of lasting value to the teacher, the student, and researchers in both linguistics and related fields.