Chinese verb subcategorization in a variant Lexical-Functional Grammar

全文

(1)Citation information: Her, One-Soon. (2008) Grammatical Functions and Verb Subcategorization in Mandarin Chinese, Revised edition, Taipei: Crane Publishing. The author has all the copy rights to this book and also the distribution rights to this electronic file, which is allowed to be freely distributed, subject to the Creative Commons restrictions below. Printed copies can be purchased from Crane Publishing. Information of the publisher is available at http://www.crane.com.tw..

(2)

(3) Taiwan Journal of Linguistics Book Series in Chinese Linguistics Editors: One-Soon Her & Kawai Chui. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN MANDARIN CHINESE. 漢語中的語法功能及動詞分類. One-Soon Her 何萬順.

(4)

(5) To the memory of my parents, HER Yu-Shou 何於壽 & KUO Wen-Ying 郭文英.

(6)

(7) v. PREFACE TO THE REVISED EDITION Since the book was first published in early 1991, I have received many comments, corrections, and suggestions. I myself have likewise noticed many typos. Also, back then the book was rather crudely typeset in the DOS version of WordPerfect. Therefore, when Angela Chen, one of Crane’s able editors, and Crane’s vice-president Paul Yang approached me about doing a revised edition, I was delighted. So they are the first two people I must thank, along with Fu-Gong Chang, Crane’s president. I thank Mr. Chang for his continuous trust and support. He has successfully established Crane to be the country’s most important bookseller and publisher in the field of linguistics and is thus to be credited for helping to promote the study of linguistics in Taiwan. This revised edition has thus undergone quite an overhaul in formatting and style. For this I thank my assistants Li-Hsin Ning and Yu-Ying Ho. Li-Hsin’s excellent skills with the word-processor Word, in particular, are largely responsible for the book’s fresh new look. My wonderful friend Karen Chung has found many typos for me, and likewise my other assistants, Yi-Ting Hsieh, Wan-Ju Huang, Hsiao-Chien Feng, and Guang-Chung Wu. The content of this revised edition, on the other hand, has not changed from the previous edition in any substantial way. The original verb subcategorization scheme proposed within Lexical-Functional Grammar, or LFG in short, in the book has since been reviewed extensively, from a Government and Binding (GB) perspective, by Feng-Fu Tsao in his 1996 Journal of Chinese Linguistics article, ‘On verb classification in Chinese’, where he has also proposed his own classification of Mandarin verbs under a GB/LFG hybrid framework. Meanwhile, a number of other important works also appeared, especially out of the Mainland China, dealing with the same general topic. The LFG framework, in the last two decades, has also made some significant advances and gained various new insights. To incorporate the advances in the LFG grammatical theory as well as in our knowledge of Mandarin verbs, I am now writing a new book, tentatively entitled Grammatical Classification of Mandarin Verbs, which should be published by Crane in 2010. However, this current book remains the first systematic classification of Mandarin verbs within a rigorously defined.

(8) vi. modern generative framework and as such, I believe, is worth the attention of theoretical linguists, Chinese grammarians, as well as lexicographers. I dedicate this revised edition to the loving memory of my parents, Yu-Shou Her and Wen-Ying Kuo..

(9) vii. ACKNOWLEDGEMENTS This book is a revised version of my doctoral dissertation. I often think that the writing of a dissertation, and the making of a PhD for that matter, is more a collective effort than most people realize. I certainly have many people to thank, who have helped me in different ways in writing and publishing this work. The first of them is no doubt Professor Robert Cheng, my dissertation adviser, who has graciously undertaken this laborious task and has provided me with numerous valuable suggestions, criticisms, and encouragements. Throughout my study at the University of Hawaii, he has always expressed his interest in my work and has always been generous in building up my confidence. The one teacher that has singularly influenced me the most in shaping my conception of how linguistics as a scientific discipline should be pursued is Professor Stanley Starosta. I certainly hope that this influence is evident in the pages that follow. Professor Hsin-I Hsieh, on the other hand, has always inspired me to view linguistics in the larger picture of life. I thank him for his insights and the many stimulating conversations we had. Professor John Haig has carefully read the first draft of the dissertation and given me many detailed constructive comments and useful suggestions. Professor David Stampe opened the door into the maze of computational linguistics for me, and that definitely led to a drastic change in my life, for the better I hope. I also enjoy his style of interacting with people. I hereupon express my gratitude to all these teachers for their positive influence and generous help. Thanks also due to Dr. Chu-Ren Huang of Academia Sinica for his many insightful comments and Professor Chin-Tang Lo for being the outside member of the dissertation committee. My special thanks go to Dr. Byron Bender, the chairman of the Department of Linguistics, who has provided me essential academic guidance over the years, to Claire Chang, who has very kindly helped me take care of so many administrative matters and has been a wonderful friend by being always supportive to me, to Professor Y. C. Li, an alumnus who has always been kind and encouraging to me and always showed interest in my work, and to my colleague Dr. Larry Browning, who has graciously read and provided detailed comments on most of my papers, including this dissertation, in terms of style and grammar. Also, I should not be thankless to Dr. Thomas Deterich of Portland State University, who turned me on to generative linguistics and.

(10) viii. showed me what a good taste of syntactic argumentation should be. I want to thank Wendy, the secretary of the Linguistics Department of UH, for her kindness and help in the past six years. For more than four years, I have been working for the company ECS, Inc. in Provo, Utah, developing machine translation systems. The president of the company, Mr. Larry Gibson, has been tremendously generous to me and most supportive to my efforts to complete my degree. I am sincerely grateful. I also want to extend my appreciation to my friends and colleagues at ECS, Dr. Larry Browning, Dr. Joseph Pentheroudakis, Dr. Dan Higinbotham, Hajime Wada (ABD), Jay Kim (ABD), Debby Masterson (ABD), Paul Warnick, M.A., Dave Rose, Reid Poole, Jim Hughes, Brice Poole, Kimbal Wirig, and Pat Reece. They are a wonderful group of people to work with. Many of the ideas in this dissertation were generated through my work experience here and conversations with my colleagues. I thank them very much for their emotional as well as intellectual support. Finally, there is my family; without them, life itself would not mean much..

(11) ix. ABSTRACT This book presents a classification of Mandarin verbs by the subcategorized grammatical functions within Lexical Functional Grammar (LFG). A brief description of LFG is provided. The following functions are identified as subcategorizable in Mandarin: SUBJ, OBJ, OBJ2, XCOMP, SCOMP, NCOMP, OBLLOCT, OBLGOAL, OBLTHME, OBLBNFC, and the syntactic encoding of these functions is discussed. Twenty-six subcategories are identified, and further subgroupings are made when different syntactic requirements of the subcategorized elements are observed. Additionally, four semantic subcategories are distinguished along the traditional distinction of action versus state. A number of our analyses of particular syntactic constructions in LFG differ from previous accounts. We consider topic a syntactic notion and argue against its subcategorizability in Mandarin. In relation to treating OBJ2, or indirect object, as encoded by the second, rather than the first, postverbal NP, we insist that [ba3 NP] should not be taken as object, rather it should be considered as an oblique function. To name one more, we reject bei4's status as either a subject marker or a preposition and analyze it as a verb requiring a pivotal construction subcategorizing for SUBJ, OBJ, and XCOMP. To improve efficiency in computational processing and linguistic generalizations, the formalism in which our analyses are formulated differs than the conventional LFG formalism in the notations of phrase structure rules and the feature inheritance structure. The later makes under-specified lexical entries and thus a lexicon of reduced size possible. Also in our variant formalism unification takes place at the same time when partial constituent structures are constructed. This allows earlier detection of functional structure violations, including incoherence of grammatical functions. While previous classifications of Mandarin verbs are often based on case roles and constituents, our classification is based on the notion of subcategorization for grammatical functions. Given the recognition of notions of subcategorization and grammatical relations in general linguistics, our study fills the gap in Chinese syntax due to the lack of a systematic, extensive account of Mandarin verbs concerning subcategorized grammatical functions..

(12) x. TABLE OF CONTENTS PREFACE TO THE REVISED EDITION .............................................. V ACKNOWLEDGEMENTS ................................................................... VII ABSTRACT..............................................................................................IX LIST OF TABLES.................................................................................. XV LIST OF FIGURES...............................................................................XVI LIST OF ABBREVIATIONS AND SYMBOLS ................................ XVII PREFACE..............................................................................................XXI CHAPTER 1 LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM ........................................................................ 1 1.1 CHINESE LINGUISTICS AND MODERN LINGUISTIC THEORIES ................1 1.2 LFG: A HISTORICAL AND CONTRASTIVE PERSPECTIVE.........................3 1.2.1 The Revolution: The Transformational Generative Grammar....... 4 1.2.2 The Second Stage: The Standard Theory ..................................... 4 1.2.3 The Third Stage: Government and Binding Theory...................... 5 1.2.4 Lexicase and the Formal Rigor .................................................... 6 1.2.5 LFG and the Emphasis on Processing .......................................... 6 1.2.6 Points of Convergence................................................................. 7 1.3 LFG: A SKETCH ........................................................................................9 1.3.1 The Lexicon ................................................................................ 9 1.3.2 The C-structure and Phrase Structure Rules ............................... 10 1.3.3 Unification: from C-structure to F-structure ............................... 12 1.3.4 From Form to Meaning ............................................................. 14 1.3.5 Further Readings ....................................................................... 15 1.4 VLFG: A VARIANT LFG FORMALISM ...................................................16 1.4.1 The Informational Organization of the Lexicon ......................... 17 1.4.2 The Phrase Structure Rules........................................................ 21 1.4.3 Unification: from C-structure to F-structure ............................... 23.

(13) xi. CHAPTER 2 GRAMMATICAL FUNCTIONS IN AN LFG GRAMMAR OF MANDARIN CHINESE............................................ 29 2.1 GRAMMATICAL FUNCTIONS AND F-STRUCTURE ..................................29 2.2 GRAMMATICAL FUNCTIONS AND THEMATIC ROLES ...........................30 2.3 SYNTACTIC ENCODING OF GRAMMATICAL FUNCTIONS ......................31 2.4 SUBCATEGORIZATION IN LFG ...............................................................33 2.5 CLASSIFICATION OF GRAMMATICAL FUNCTIONS.................................36 2.5.1 Subcategorizability.................................................................... 36 2.5.2 Semantic Restriction ................................................................. 38 2.6 SUBCATEGORIZABILITY OF GRAMMATICAL FUNCTIONS AND WELL-FORMEDNESS CONDITIONS ON F-STRUCTURE.................................39 2.7 GRAMMATICAL FUNCTIONS AS PRIMITIVES .........................................45 2.8 THE STATUS OF TOPIC AND SUBJECT ....................................................45 2.9 SUBCATEGORIZABILITY OF TOPIC IN CHINESE ...................................51 2.10 SYNTACTIC ENCODING OF SUBJ AND OBJ ........................................61 2.11 OBJ2 AS A SEMANTICALLY RESTRICTED FUNCTION .........................62 2.12 COMP FUNCTION AND SUBTYPES ......................................................65 2.13 OBLIQUE FUNCTION AND SUBTYPES...................................................66 2.14 SYNTACTIC ENCODING OF GRAMMATICAL FUNCTIONS IN CHINESE 67 CHAPTER 3 VERB SUBCATEGORIZATION IN MANDARIN CHINESE................................................................................................. 69 3.1 EVIDENCE FOR SUBCATEGORIZED ELEMENTS......................................69 3.1.1 Obligatory Co-occurrence ......................................................... 70 3.1.2 Ontological Necessity ............................................................... 71 3.1.3 Constancy of Semantic Content................................................. 72 3.1.4 Distributional Restrictions......................................................... 73 3.1.5 One per Sentence ...................................................................... 74 3.1.6 Proximity to the Head ............................................................... 76 3.1.7 Semantic Selectional Restrictions.............................................. 76 3.1.8 Possible Internal Gaps............................................................... 76 3.1.9 Ability of Being an Indirect Question ........................................ 77 3.1.10 Identification of a Subcategory................................................ 78 3.2 MANDARIN VERB SUBCATEGORIZATION IN VLFG ..............................78 3.2.1 Activity and Stativity ................................................................ 79 3.2.2 Syntactic Subcategorization ...................................................... 85.

(14) xii. 3.2.3 Optionality in Subcategorization................................................ 88 3.2.4 Derivational Relations between Verb Classes ............................ 89 3.3 SUBCATEGORIZATION OF GRAMMATICAL FUNCTIONS ........................93 3.3.1 <SUBJ> .................................................................................... 93 3.3.1.1 Intransitive Verbs............................................................... 93 3.3.1.2 Weather Verbs ................................................................... 96 3.3.2 <SUBJ , OBLLOCT> ................................................................... 97 3.3.3 <SUBJ , OBLGOAL>................................................................. 100 3.3.4 <SUBJ , OBJ>......................................................................... 103 3.3.4.1 Typical Transitive Verbs .................................................. 103 3.3.4.2 Verbs Requiring a Locative OBJ ...................................... 106 3.3.4.3 Locative Inversion Verbs.................................................. 108 3.3.4.4 You3 'exist, have' ............................................................. 109 3.3.4.5 Verbs Requiring TOPIC................................................... 125 3.3.5 <SUBJ , OBLTHME>................................................................. 126 3.3.5.1 Syntactic Category of ba3................................................. 127 3.3.5.2 The Double Direct Object Fallacy .................................... 128 3.3.5.3 Semantic Restrictions of [ba3 NP] .................................... 130 3.3.6 <SUBJ , OBLTHME , OBJ> ....................................................... 135 3.3.7 <SUBJ , OBLGOAL , OBJ> ....................................................... 141 3.3.8 <SUBJ , OBJ , OBLLOCT>........................................................ 142 3.3.9 <SUBJ , OBLTHME , OBLLOCT>................................................ 143 3.3.10 <SUBJ , OBLTHME , OBJ , OBLLOCT> .................................... 144 3.3.11 <SUBJ , NCOMP>................................................................ 145 3.3.11.1 Equational Verbs............................................................ 145 3.3.11.2 Verbs Denoting Inherent Quality.................................... 149 3.3.12 <SUBJ , XCOMP>................................................................ 151 3.3.12.1 Regular Type ................................................................. 151 3.3.12.2 Tough Construction........................................................ 154 3.3.12.3 Modal Verbs .................................................................. 159 3.3.12.4 Clause Union Verbs ....................................................... 166 3.3.13 <SUBJ , SCOMP> ................................................................ 169 3.3.13.1 Regular Type ................................................................. 169 3.3.13.2 Verbs Requiring an Interrogative SCOMP...................... 170 3.3.14 <SUBJ , OBJ , OBJ2> ........................................................... 173 3.3.14.1 Regular Type ................................................................. 173.

(15) xiii. 3.3.14.2 Idiomatic Expressions .................................................... 174 3.3.15 <SUBJ , OBJ , OBLBNFC> ..................................................... 176 3.3.16 <SUBJ , OBLTHME , OBJ , OBJ2> ......................................... 177 3.3.17 <SUBJ , OBLTHME , OBJ , OBLBNFC>.................................... 179 3.3.18 <SUBJ , OBLTHME , OBLBNFC>.............................................. 180 3.3.19 <SUBJ , OBLGOAL , OBLTHME> ............................................. 181 3.3.20 <SUBJ , OBLGOAL , OBLTHME , OBJ>.................................... 182 3.3.21 <SUBJ , OBJ , XCOMP>...................................................... 183 3.3.21.1 Regular Type ................................................................. 183 3.3.21.2 Clause Union Verbs ....................................................... 186 3.3.21.3 "Promise" Verbs ............................................................ 189 3.3.21.4 Tough Construction ....................................................... 191 3.3.21.5 Bei4............................................................................... 193 3.3.21.5.1 Existing Analyses for Chinese bei4......................... 194 3.3.21.5.2 Dismissal of bei4 as a Subject Marker..................... 196 3.3.21.5.3 Dismissal of bei4 as a Preposition ........................... 197 3.3.21.5.4 Bei4 as a Verb ........................................................ 202 3.3.21.5.5 Conclusion ............................................................. 220 3.3.22 <SUBJ , OBJ , SCOMP> ...................................................... 221 3.3.22.1 Regular Type ................................................................. 221 3.3.22.2 Verbs Requiring an interrogative SCOMP...................... 221 3.3.23 <SUBJ , OBLGOAL , XCOMP> .............................................. 222 3.3.23.1 Regular Type ................................................................. 222 3.3.23.2 Verbs Requiring an interrogative SCOMP...................... 224 3.3.24 <SUBJ , OBLGOAL , SCOMP>............................................... 225 3.3.24.1 Regular Type ................................................................. 225 3.3.24.2 Verbs Requiring an Interrogative SCOMP ..................... 226 3.3.25 <SUBJ , OBJ , OBJ2 , SCOMP>........................................... 226 3.3.26 <SUBJ , OBLGOAL , OBJ , SCOMP> ..................................... 227 3.4 SUMMARY OF SYNTACTIC SUBCATEGORIZATION OF VERBS.............228 3.5 COMPARISON WITH PREVIOUS CLASSIFICATIONS ..............................230 CHAPTER 4 GRAMMATICAL THEORIES AND NATURAL LANGUAGE PROCESSING............................................................... 237 4.1 LEXICASE, GOVERNMENT AND BINDING, AND LFG ..........................239 4.2 TASKS OF NATURAL LANGUAGE PROCESSING ...................................241.

(16) xiv. 4.3 PARSING NATURAL LANGUAGE ..........................................................244 4.4 NATURAL LANGUAGE GENERATION ...................................................250 4.5 MACHINE TRANSLATION......................................................................253 4.6 SUMMARY .............................................................................................256 CHAPTER 5 CONCLUSION............................................................... 257 5.1 THE VLFG FORMALISM .......................................................................257 5.2 MANDARIN VERB SUBCATEGORIZATION IN LFG...............................259 5.3 SUGGESTED AREAS FOR FURTHER RESEARCH ...................................262 BIBLIOGRAPHY .................................................................................. 265 APPENDIX FEATURE INHERITANCE ENTRIES........................... 277.

(17) xv. LIST OF TABLES Table 3.1 Semantic Subcategories of Mandarin Verbs................................ 80 Table 3.2 Classification of Mandarin Modal Verbs .................................. 165 Table 3.3 Syntactic Classification of Mandarin Verbs .............................. 228 Table 3.4 Chao's Classification of Mandarin Verbs (Chao 1968).............. 231 Table 3.5 Li's Classification of Mandarin Verbs (Li 1971) ....................... 232 Table 3.6 CKIP's Classification of Mandarin Verbs (CKIP 1989)............. 234.

(18) xvi. LIST OF FIGURES Figure 1.1 Co-description of C- and F-structure.......................................... 14 Figure 1.2 Correspondence of Linguistic Representations in LFG .............. 15 Figure 2.1 Correspondence between Conceptual Structure and Linguistic Structures.................................................................................. 31 Figure 2.2 Lexical and Syntactic Encoding of Grammatical Functions........ 33 Figure 2.3 Classification of grammatical functions in Mandarin ................. 38 Figure 3.1 Classification of Grammatical Functions.................................... 91 Figure 3.2 Possible Analyses for Chinese bei4.......................................... 195 Figure 4.1 NLP Application of Linguistics ............................................... 241 Figure 4.2 Correa's Variant GB Model (Correa 1987)............................... 245.

(19) xvii. LIST OF ABBREVIATIONS AND SYMBOLS A ADJ ag ATN ben BNFC CAT CLS COMP exp FI FS FUG GB go GPSG HPSG inst LFG loc LOCT LXC MT N NCOMP NLP NP OBJ OBJ2 OBLΘ P PAS pat PRED. Attributive Adjective in Chinese Adjunct Function in LFG agent Augmented Transition Network beneficiary beneficiary Category Classifier Inventory of Predicative Complements in LFG experiencer Feature Inheritance Entry Functional Structure Functional Unification Grammar Government and Binding Theory goal Generalized Phrase Structure Grammar Head-driven Phrase Structure Grammar instrument Lexical Functional Grammar locative location Lexicase Theory Machine Translation Noun Predicative Noun Complement in LFG Natural Language Processing Noun Phrase Object Function in LFG Second Object Function in LFG Inventory of Oblique Functions in LFG Preposition Predicate Argument Structure patient Predicates.

(20) xviii. PTCL PP PSR QTFR RELN RG S SCOMP SUBJ TG th THME TOPIC V vLFG VP XCOMP XADJ. Particle Prepositional Phrase Phrase Structure Rule Quantifier (Semantic) Relation Relational Grammar Sentence Sentential Complement Function in LFG Subject Function Transformational Grammar theme theme Topic Function in LFG Verb the variant LFG Formalism Verb Phrase Open Complement Function in LFG Open Adjunct Function in LFG. ↑. the level of f-structure corresponding to my mother node of the c-structure the level of f-structure corresponding to my own node of the c-structure unify with unify by default with must be identical with unify to be (usually a subcategorizable grammatical function) unify to be a member of (usually an adjunct or conjunct) optional constituent (in phrase structure rules); OR path (in f-structure) disjunction (in phrase structure rules and schemata); OR conglomerated list (in f-structure). ↓ = ≈ =c : ε () {}.

(21) xix. X+ X* <XY>. @. one or more instance of X (in c- or f-structure) zero or more instance of X (in c- or f-structure) X and Y in any order in vLFG PSR's; OR in LFG formalism, X and Y as subcategorized, thematically assigned functions, as part of the value of PRED what preceded is a non-thematically assigned function in vLFG.

(22)

(23) xxi. PREFACE This book first provides a brief description of the Lexical Functional Grammar (LFG) and a variant formalism of a unification grammar that most closely follows the theoretical constructs and formalism of LFG (Bresnan 1982, Sells 1985, Wescoat 1987, Kaplan and Zaenen 1987), including LFG's most distinctive feature: the division of constituent structure and functional structure. Within the LFG theory we will discuss the various grammatical functions in the grammar of Mandarin Chinese, and then, within this variant LFG formalism (vLFG), we will systematically present the various subcategories of verbs in Mandarin Chinese that require different subcategorization of grammatical functions. Our scheme of classification of Mandarin verbs within the vLFG formalism is therefore according to the grammatical functions that verbs subcategorize for. We shall argue for our analysis of grammatical functions and their subcategorizability in an LFG grammar of Mandarin Chinese. We shall also, in a relatively theory-independent manner, argue for our analysis of certain types of verbs of which there exist previous different accounts or whose syntactic constructions are known to be controversial in the literature of Chinese linguistics. Concerning the Mandarin data in this book, we should clarify that, in order to illustrate precisely and concisely the relevant points, we have selected example sentences that contain the minimum necessary elements. The judgement on the grammaticality or acceptability is primarily the author's, a native speaker of Mandarin Chinese college-educated in Taiwan whose parents are from the central and northern regions of China. Most of the several other native speakers of Mandarin who have read previous versions of the book also confirmed the author's judgement. Furthermore, most of the sentences we use in the discussions do comport with examples cited in other works in Chinese linguistics (e.g., Chao 1968 and Li and Thompson 1981). In CHAPTER 1, we will briefly review the historical development of Chinese linguistics in applying modern linguistic theories and the current state of Chinese linguistics in relation to contemporary grammatical theories and computational applications in processing Mandarin Chinese. We will briefly discuss the theory of LFG and its motivation in the context of the developmental stages of Chomsky's generative syntax of the past three decades and also in contrast with two other contemporary grammatical.

(24) xxii. theories, Government and Binding Theory (GB) and Lexicase Theory (LXC). We will then sketch the theoretical constructs and formal notations of LFG as it is conventionally formulated. We then present the vLFG formalism employed in this book. We shall see that although the vLFG and the conventional LFG are very close in almost all aspects of linguistic theoretical constructs, the major differences between them include the timing of the building of the functional structure, the informational organization of the lexicon and lexical entries, and the notation of lexical entries and phrase structure rules. The vLFG formalism allows a feature inheritance structure in the lexicon, and unlike conventional LFG that builds a functional structure only after the entire constituent structure is built, the vLFG builds a partial functional structure whenever its corresponding constituent is properly formed. We will describe in detail the formalism and notations of grammar writing within the vLFG. In CHAPTER 2, we will first discuss what we consider grammatical functions in LFG and the different types among them. We will show how grammatical functions are lexically and syntactically encoded in a grammar. More specifically, we will identify a set of grammatical functions for the grammar of Mandarin Chinese and also justify our classification of subcategorizable and non-subcategorizable (adjunctive) functions in Mandarin. Among the subcategorizable functions, we will identify the semantically restricted ones versus the semantically unrestricted ones. We will devote a more lengthy discussion on the status of subject and topic, especially regarding the subcategorizability of TOPIC in Mandarin Chinese. The following subcategorizable functions are identified: SUBJ, OBJ, OBJ2, XCOMP, SCOMP, NCOMP, OBLLOCT, OBLGOAL, OBLTHME, OBLBNFC. We will also discuss the syntactic encoding of these functions. CHAPTER 3, the core of this book, first discusses the notion of subcategorization and the generally-accepted criteria for distinguishing subcategorized constituents from adjuncts. Then, we will present a semantic classification of Mandarin verbs based on two binary features of activity and process. After giving some reasons of not allowing optional arguments in subcategorization and introducing an example of morpholexical rule in LFG, we proceed to present our classification of Mandarin verbs according to the grammatical functions they subcategorize for. We will discuss each subcategory of verbs and argue for our analysis for certain verb classes when.

(25) xxiii. it differs from previous accounts. Noticeably, more detailed discussion is devoted to the following elements that have been known to be controversial: ba3, bei4, you3, and shi4 etc. Twenty-six subcategories are identified, and within several of them further subgroupings are made when differences in syntactic requirements of the subcategorized elements are observed. Finally we will offer a brief comparison of our classification schemes with three other existing ones, Chao (1968), Li (1971), and CKIP (1989). In CHAPTER 4 we will discuss the practical application and theoretical implication of the study in this book. We shall argue that the vLFG grammar has been shown to be suitable for the description of Mandarin verbs and that while its theory is generally constrained in similar ways as the LFG theory, its formalism is expressive and thus appropriate for explicit formulation of linguistic generalizations for computational applications. We shall examine in what ways LFG is relevant to the various tasks under the general domain of Natural Language Processing (NLP), such as parsing, generation, and machine translation. In contrast, we will also discuss the two other grammatical theories, Lexicase and Government and Binding Theory, regarding their NLP application. We will argue for the advantage of LFG, especially in the vLFG formalism we are adopting in this study, for computational applications of NLP. Finally, in CHAPTER 5, we will briefly reiterate the contributions we hope this book will make to the study of Mandarin Chinese and the study of LFG in terms of our subcategorization of Mandarin verbs, accounts of certain controversial syntactic constructions, and the several alternatives in the consideration of grammar formalism and processing procedure. We will point out areas for further research in order to gain a more complete account of Mandarin verbs in LFG and conclude the book..

(26)

(27) 1. CHAPTER 1 LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM In this chapter we will review briefly the development of linguistic approaches for analyzing Mandarin Chinese, and we will discuss briefly the current state of Chinese linguistics in relation to current linguistic theories and computational application. We will argue for the importance for linguists to work within and contribute to a linguistic framework that is truly "generative" in nature. We will then look at the development and the primary motivation of LFG in the context of the developmental stages of Chomsky's generative theory of language. Also, to provide different perspectives, we will compare LFG with two other current linguistic theories, GB and LXC in the context of these developmental stages. We will sketch the theoretical constructs and formalism of LFG as it is conventionally formulated. Then we will describe the variant formalism of LFG in which we will formulate our analyses. For ease of discussion, we will use the term "vLFG" to refer to our variant formalism, one that is different in the organization of the lexicon, notation of phrase structure rules and the timing of the building of functional structure.. 1.1 Chinese Linguistics and Modern Linguistic Theories For more than half a century Chinese linguists have applied modern linguistic theories to the analysis of Chinese languages, and most intensively to Mandarin Chinese, the language that has the largest number of native speakers on earth as reported by the 1989 World Almanac. Since almost all modern linguistic theories claim to follow western scientific principles and are developed out of the west, the application of new, innovative theories to Mandarin Chinese always comes noticeably more slowly, compared to their application to western languages. The traditional, indigent school of philology and the study of phonology, especially in the area of rhyme, existed in China for centuries before the western Structuralist school established linguistics as an independent discipline in modern sciences early in this century. This rich resource of traditional works has provided a solid basis for Chinese historical.

(28) 2. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. linguistics and dialectal studies and still occupies an important position in Chinese linguistics. The Structuralist framework dominated the modern linguistic scene of the study of Chinese from the late 1940's well into the 1970's, almost two decades after Chomsky revolutionized the study of language in the late 1950's with his generative theory of linguistics. In fact, even today a large number of Chinese linguists, especially in Mainland China, still publish papers in a general linguistic framework best described as Structuralist (e.g., Zheng and Chen 1989 and Yang 1989). Structuralist linguists have contributed greatly to the study of Mandarin Chinese; among them there is the most influential figure in Chinese linguistics to date, the late Dr. Y.R. Chao, whose monumental work laid the foundation for the study of modern Chinese. Many Chinese linguists worked diligently within the framework of Transformational Grammar, including its later version the Standard Theory, advocated by Chomsky as the first generative theory, in its heyday of the 60's and 70's (e.g., Huang 1966 and Li 1972). However, their number dwindled in the 80's as the Chomskyan school of linguistics came to be highly abstract and deviated from Chomsky's original vision of generative linguistics. In the meantime, alternative theories to the mainstream Chomskyan practice flourished, reacting to the inadequacies that many theorists perceived in the conventional Chomskyan theory. Many of these alternative theories insist strongly upon their "generative" nature, in its original sense of "formal" and "explicit" intended by Chomsky in his revolutionary work. Although the Chomskyan school of Government and Binding Theory is still often perceived as the mainstream and the application of the alternative theories in Chinese has come slowly, several of them have been applied in the analysis of Mandarin Chinese; they include Case Grammar (e.g., Li 1971 and Teng 1975), Lexicase (Starosta 1985 and Her 1985-6), Categorial Grammar (e.g., Liu 1987 and Zhang 1989), Generalized Phrase Structure Grammar (e.g., Huang 1987 and Sheu 1987), and Lexical Functional Grammar (e.g., Huang 1989 and Her 1989b). Having abandoned Transformational Grammar, which has been recognized as excessively powerful and thus non-revealing, many Chinese linguists, who followed Chomsky's earlier theory, stayed outside of the GB camp, and, like many other Chinese linguists, often resorted to analyzing Chinese linguistic data in a most general, undefined framework of linguistics, or one that.

(29) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 3. Starosta (1988) describes as "eclecticism.” Truly, if linguistics is to be recognized as a discipline of modern science, it is crucial, or even mandatory, one might argue, for its workers to contribute collectively in a theoretically well-defined framework within this discipline. Furthermore, as a scientific pursuit, a sound linguistic theory has to adhere to the scientific methodology known as the hypothetico-deductive method (Starosta 1987). Logically, only a genuine generative linguistic theory where generalizations must be made formally and explicitly is falsifiable and thus compatible with the hypothetico-deductive spirit. The rapid advancement of computer technology and soaring demand of natural language processing in recent years have provided a serious testing ground for linguistic theories and analyses. An informal or inexplicit linguistic theory or analysis cannot survive the scrutiny of computational applications. Although, unfortunately, many projects of natural language processing do not utilize any well-established linguistic theory for their linguistic analysis, in recent years LFG has been gaining increasing popularity in computational applications, especially in the Republic of China (Huang and Chen 1989). We will discuss in greater details the relevance of linguistic theories in computational applications in CHAPTER 4. In terms of linguistic analysis, several areas of Chinese grammar have already been explored in the LFG framework, including Mandarin [NP-de] constructions (Huang 1987), sentence particles and questions (Shiu 1989, Shiu and Huang 1989), relative clauses (Hu 1989), topicalization (Chen 1989), bei4 sentences (Her 1989), lexical discontinuity (Huang 1988), and the linking relations between thematic roles and grammatical functions in Chinese (Huang 1989, 1989a). This present study of grammatical functions and verb subcategorization, which has been preliminarily reported in Her (1989a), will constitutes yet another significant step in the application of LFG in Mandarin Chinese.. 1.2 LFG: A Historical and Contrastive Perspective Lexical Functional Grammar, a generative lexicalist unification grammar theory, was first systematically introduced by Kaplan and Bresnan (1982). Like almost all other contemporary generative grammatical theories, LFG has its roots deeply embedded in Chomsky's early generative syntax and yet was.

(30) 4. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. developed as an improvement of and a reaction to some of the inadequacies that the LFG theorists observed in the directions that the mainstream Chomskyan grammarians chose to follow. To understand the motivation and the development of LFG, it should reveal meaningful insights to first look at LFG within the context of the developmental stages of Chomsky's generative syntactic theory and in contrast with other contemporary grammatical theories. Among other current syntactic theories we choose to compare LFG with Lexicase (LXC) and Government and Binding Theory (GB), which provide different perspectives than LFG and thus serve well as contrast. 1.2.1 The Revolution: The Transformational Generative Grammar Nearly all contemporary generative syntactic theories share a common ancestry: Chomsky's revolutionary work of generative transformational syntax of the late 1950's. We believe each of the three theories, in the areas it chooses to emphasize, represents a different reaction to, or extension of, Chomsky's earlier interpretation of syntax. From late 1950's to the present, Chomsky's syntactic theory has roughly undergone three perceivable developmental phases. Syntactic Structures (Chomsky 1957) revolutionized syntactic theorizing and trumpeted the advent of the era of generative grammar, and also firmly established the study of linguistics as a scientific pursuit. The proposal was that the objective of a grammar is to "generate" all and only the infinite grammatically acceptable strings of a natural language. Thus the key point is that grammars should be 'generative', in a mathematical sense. Therefore, the criterion of explicit, formal and falsifiable formulation of linguistic statements and generalizations was greatly emphasized. The claim was that this goal can be obtained with a transformational grammar but not with a phrase structure grammar nor the earlier Structuralist approach to language. 1.2.2 The Second Stage: The Standard Theory The 'Standard Theory' of Aspects of the Theory of Syntax (Chomsky 1965) posited two levels of syntactic representation: a deep structure as the basis for meaning interpretation and a surface structure as the basis for phonological interpretation. The crucial linking between the deep and surface structure is.

(31) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 5. accomplished by transformations. The focus of syntactic theory was that how a grammar should model the mental mapping between meaning and sound, and thus linguistic analyses should be psychologically real. The requirement of explicitness and formalness began to lose its earlier visibility. The Standard Theory soon divided into several revised and extended versions by the late 1970's. The most salient and probably also most important trend of syntactic research was the restraining of powerful theoretical constructs such as transformation rules. Within the Chomskyan camp of transformational grammar the greatest effort has been in the elimination of various structure-specific transformations; however, layers of highly abstract constraints have to be devised to allow for one single general transformation: Move-α. Other theories of different approaches, including LXC and LFG, on the other hand have totally ruled out the validity of transformations theoretically and treat syntax as a purely surface phenomenon. As a matter of fact, LXC, dating back to the early 1970's, was probably the first generative theory of syntax that was entirely transformationless. 1.2.3 The Third Stage: Government and Binding Theory Lectures on Government and Binding (Chomsky 1981), introducing the Government and Binding Theory, embodied research within the Standard Theory and extended models concerned with the constraining of transformations and the attaining of the explanatory power of how only grammars learnable based on the primary data should be allowed in the theory of language. In other words, GB attempts to provide a theory where a grammar of a natural language can be inferred through a set of universal principles and the setting of certain universal parameters. In the pursuit of linguistic parameters and universal grammar, standards of explicit, formal, and detailed formulation of analysis of specific syntactic constructions waned and were even reproved. While there is a considerable amount of GB cross-language research on parameters of language variation and universal principles, one rarely finds an explicit formulation of an analysis of a specific syntactic construction of a particular language..

(32) 6. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. 1.2.4 Lexicase and the Formal Rigor While GB has compromised considerably the standards of formal explicitness and psychological realism, LXC and LFG to a significant degree represent serious efforts to reemphasize some of the worthwhile characteristics of earlier phases of Chomsky's generative theory. Both LFG and LXC claim their respective linguistic theory to be a universal model for all languages and therefore do stress upon the standard of explicit rigor, psychological reality, and the universality of their theory. LFG however has strong emphasis on the computational and psycholinguistic processing of language, and the LXC literature has demonstrated a most serious commitment to the formal and explicit formulation of linguistic generalizations. One of LXC's strongest objection to GB is that it is not clear whether GB can still be considered 'generative' in the original sense intended by Chomsky in the first stage of the late 1950's. Starosta, the primary theorist of LXC who had substantial training in physics, devoted an entire volume readdressing the issue of 'generative grammar' as a hypothetico-deductive science (Starosta 1987), and on numerous occasions repeated the generative aspect of LXC and that the goal of a LXC grammar is to generate all and only the acceptable phrases of which sentences are a subset. The LXC literature is therefore largely composed of detailed, explicit accounts of grammatical phenomena of various natural languages, most of which are non-IndoEuropean. 1.2.5 LFG and the Emphasis on Processing LFG's concern for the processing aspect of language can no doubt be partially attributed to the two primary architects of the theory, Kaplan and Bresnan. The formal conception of LFG evolved in the mid-1970's from earlier work in both Transformational Grammar and computational linguistics. Kaplan was a psychologist and did experimental work on human sentence processing and computational natural language processing. He was one of the designers of Augmented Transition Network (ATN) grammar, a computationally-oriented grammar which also served as one of the precursors of LFG. When making a transition from Transformational Grammar to LFG, Bresnan (1978) argued that the model she was proposing was psychologically.

(33) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 7. more realistic. This point was again crucially emphasized in the most important compilation of LFG work, The Mental Representation of Grammatical Relations (Bresnan 1982). LFG thus differs in being a linguistic theory with the goal to also serve as the grammatical basis of a computationally precise and psychologically realistic model of natural languages. Consequently, many of the theoretical decisions have been influenced by this perspective (Bresnan 1982, Sells 1985). The fact that a great number of research projects of natural language processing employ LFG-style formalisms also reveals LFG's commitment as a processing-oriented syntactic theory. Another striking similarity that LFG shares with the Standard Theory is that as the Standard Theory identifies deep structure as the basis for semantic interpretation and surface structure as the basis for phonological interpretation, with transformational rules as the linkage, LFG also posits two levels of syntactic representation: the c-structure (constituent structure) which is the basis for phonological operation and f-structure (functional structure) from which the semantic representation is derived, and functional descriptions provide the linkage between c- and f-structures. However, it should be quickly pointed out that the similarity between deep/surface structure and c-/f-structure stops here. While deep and surface structures are two separate strata in the derivation process, c-structure and f-structure are associated with each other at any given point of the derivation and thus are co-descriptions of the same word string. 1.2.6 Points of Convergence Despite the fact that these three theories make different choices of what to emphasize and differ in the assumptions they make for the basis of a syntactic theory, and also employ drastically different formalism, there are two significant points of convergence among them and perhaps other contemporary theories as well: the reduced role of transformations and the increased role of the lexicon. How to limit or eliminate the power of transformations and the shifting of emphasis to the lexicon thus are the two major trends in the study of syntax in the past two decades. GB has reduced the earlier various ad hoc and powerful transformational rules to just one: Move-α (move anything to anywhere). LFG and LXC eliminated entirely the.

(34) 8. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. theoretical validity of transformations and employed different analyses and/or lexical/morphological processes to account for syntactic phenomena that are previously accounted for by transformations. The lexicon plays a central role in Lexical Functional Grammar and Lexicase, as their names suggest. Lexicase takes the most extreme position in that the lexicon of a language is the entire grammar of that language and all linguistic generalizations are expressed by lexical rules, which must be feature-preserving, i.e., they can only add but cannot delete or change features. In LFG, every lexical entry has a set of functional expressions associated with it, and the f-structure of a phrase or a sentence is the result of unification of lexical functional structures according to the functional specifications associated with phrase structure rules that build the phrase or sentence. What most transformations used to perform now is handled by lexical rules, such as Passivization, Equi, and Raising. The significant similarity among all three theories is therefore that the clause structure of a verb to a large extend can be predicated by its semantic predicate structure. And the argument structure of a predicate is specified in the lexicon. The Projection Principle of GB and the Principle of Function-Argument Biuniqueness of LFG ensure that the predicate argument structure is realized structurally and may not be altered in essential ways. Such conditions are not necessary in LXC where contextual features associated with lexical items dictate entirely the possible phrase structures of a clause, and implied case relations, i.e., thematic relations roughly, are also specified in the lexical entry. To view LFG through this historical and comparative frame is interesting and revealing; it seems that each of the theories discussed here, in their particular emphasis, presents a different reaction to Chomsky's later directions and represents a developmental stage of Chomsky's theory of syntax. However, we by no means imply any accusation of LXC and LFG of reversing progress, for both theories do address the important issue of explanatory adequacy. Rather, we respect the commitment to formal rigor and processing efficiency in grammar writing on the part of LFG and LXC. We will come back to this point again when we discuss these three theories in CHAPTER 4 regarding their application in computational tasks of natural language processing..

(35) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 9. 1.3 LFG: A Sketch In this section, we will provide a brief description of the LFG theory and its formalism, based on Kaplan and Bresnan (1982), Sells (1985), Kaplan (1989), and Huang (1989c). This description thus serves as the basis for the comparison between the conventional LFG and our vLFG discussed in the next section. The best-known feature of LFG theory is probably that its grammar assigns two types of syntactic representations to a sentence or phrase, c-structure to reflect its constituent structure and f-structure for its grammatical/functional structure. The idea of c- and f-structure division is similar to that of the ID/LP format (ID = immediate dominance; LP = linear precedence) of stating phrase structure rules in GPSG. The difference is that while the ID rules are an abstraction of pure constituency away from LP rules, which describe the linear ordering of constituents, f-structure is an abstraction of the grammatical, relational, or functional information away from both phrasal constituency and ordering. Therefore, c-/f-structure division and ID/LP format are totally compatible. Many LFG practitioners do adapt the ID/LP format in stating phrase structure rules (e.g., Huang and Mangione 1985). The c-structure serves as the basis for phonological interpretation while the f-structure for semantic and discoursal interpretation. Lexical items have their functional information defined in the lexicon. The LFG context-free phrase structure rules are augmented with functional annotations. The c-structure formed by the PSR's is thus also annotated with functional expressions. The f-structure is then formed by instantiating the functional annotations on the c-structure. 1.3.1 The Lexicon Capitalizing on the linguistic trend of having lexical information account for various grammatical phenomena, in LFG the lexicon plays an essential role. All the functional or grammatical information of a sentence comes from the words in it and the functional specifications on phrase structure rules. In the lexicon, all lexical entries are fully inflected and thus have complete grammatical information. A lexical entry contains its categorial specification and a set of functional expressions known as "schemata," which are to be.

(36) 10. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. instantiated, or to take place, when the c-structure of the sentence is reached. The English word 'forces', as in a construction like "Mary forces John to go" would have the following entry: 1. forces. V, (↑ PRED) = 'force <(SUBJ) (OBJ) (XCOMP)>' (↑ OBJ) = (↑ XCOMP SUBJ) (↑ SUBJ PERSON) =c 3rd (↑ SUBJ NUMBER) =c SG. For practical purposes, we can interpret ↑ as "the level of f-structure associated with my mother node in the c-structure" and ↓ as "the level of f-structure associated with my own node in the c-structure.” The syntactic category of this lexical entry is V. The first functional equation defines (indicated by =) this entry's semantic form and the predicate structure. It requires three arguments, or subcategorizes for three grammatical functions, SUBJ, OBJ, and XCOMP. The second equations defines the "control" relation between the matrix object (John) and the missing subject of the embedded non-finite clause XCOMP, with the former being the controller and the later the controlled. The subject of "to go" is therefore to be identified (unified) with "John." The third and fourth equations are functional constraints (indicated by =c). Unlike an equation of definition (=), where unification actually takes place, a constraint (=c), where no unification is performed, only checks whether such attribute-value pair exists or not. Thus the two constraints in this entry make sure that its subject has the attribute PERSON with value 3rd and the attribute NUMBER with value SG. If there is any constraint not satisfied, the functional structure is considered ill-formed. Many agreement features are checked in this manner. As we shall see later, the functional equations, or schemata, on a lexical entry are instantiated only after the c-structure is constructed. 1.3.2 The C-structure and Phrase Structure Rules Basically an LFG grammar consists of a lexicon, a set of context-free phrase structure rules (PSR's) functionally annotated, and certain well-formedness conditions on f-structures. The PSR's describe the.

(37) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 11. constituent structure of a sentence, and in fact they may also be viewed as constraints on the well-formedness of c-structures. Only the syntactic categories, not the functional equations, of the words contained in a word string are relevant to its c-structure. Let's see a simple example, "Mary loves John" assuming the following lexicon and PSR's. 2. a. Mary. N, (↑ PRED) = 'Mary' (↑ PERSON) = 3rd (↑ NUMBER) = SG. b. John. N, (↑ PRED) = 'John' (↑ PERSON) = 3rd (↑ NUMBER) = SG. c. loves. V, (↑ PRED) = 'love <(SUBJ) (OBJ)>' (↑ SUBJ PERSON) =c 3rd (↑ SUBJ NUMBER) =c SG. 3. a. S →. NP VP (↑ SUBJ)=↓ ↑=↓. b. VP →. V (NP) ↑=↓ (↑ OBJ)=↓. (NP) (↑ OBJ2)=↓. c. NP →. (DET) (A) N ↑=↓ ↓ε(↑ ADJ) ↑=↓. If all we are concerned with is the c-structure of the sentence, both the functional equations on the lexical entries and the functional annotations, also known as "schemata," on the PSR's can be ignored. The only relevant feature is the lexical item's categorial specification indicating the preterminal category under which the lexical item may be inserted. Obviously then, the notion of c-structure is entirely compatible with that of the more commonly known tree.

(38) 12. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. structure. These PSR's thus assign the following tree structure correspondence to the word string. In the tree representation of c-structures, we will borrow the Lexicase convention that a straight vertical line indicates that the lower node is the head of its mother node, but this does not mean that we assume the Lexicase position that all c-structure constructions have a lexical head. 4.. S NP. VP. N. V. NP N. Mary. loves. John. 1.3.3 Unification: from C-structure to F-structure Of course c-structure rules in LFG are different from simple context-free PSR's in that they are augmented with functional annotations; therefore, they assign to a sentence its c-structure with particular specifications to regulate the manner of unification in constructing the f-structure of the sentence. Again, let's look at our previous example. The functionally annotated PSR's would assign the following c-structure with functional schemata to the sentence..

(39) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 5. NP (↑ SUBJ)=↓ N (↑ NUMBER)=SG (↑ PERSON)=3rd (↑ PRED)='Mary'. Mary. 13. S │ VP ↑=↓ V (↑ PRED)='love<...>' (↑ SUBJ NUMBER)=cSG (↑ SUBJ PERSON)=c3rd. loves. NP (↑ OBJ)=↓ N (↑NUMBER)=SG (↑PERSON)=3rd (↑PRED)='John' John. Note that the functional schemata originating in the lexicon are not formally distinct from the ones coming from the augmented PSR's. Through the instantiation of the functional schemata and the operation of unification, the f-structure of the sentences is then composed. Therefore, f-structure is constructed only after the c-structure is constructed. Readers interested in the formal, detailed procedures of deriving f-structure correspondence from c-structure through functional descriptions should refer to Kaplan and Bresnan (1982) or Wescoat (1987). We will now illustrate in the figure below the correspondence between the c- and f-structure, or the co-description of cand f-structures, of the sentence "Mary loves John.".

(40) 14. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. S NP. VP. N. V. PRED ‘<(SUBJ) (OBJ)>’ SUBJ. NUMBER SG PERSON 3rd PRED ‘Mary’. OBJ. NUMBER SG PERSON 3rd PRED ‘John’. NP N. Mary. loves. John. Figure 1.1 Co-description of C- and F-structure As mentioned before, LFG also posits certain well-formedness conditions on f-structures. A final f-structure associated with a final c-structure still needs to satisfy all the well-formedness conditions. We will discuss these conditions together with grammatical functions in CHAPTER 2, since most of the conditions make reference to the notion of "subcategorizable functions." Thus, in LFG, both phrase structure rules and information associated with lexical items contribute a local co-description of partial information to the final c- and f-structure of the sentence. The advantage of this division is that it allows separate encoding of external constituent structure, which varies across different languages, and the internal relational structure of grammatical functions which is largely invariable universally. The two structures are two independent but parallel planes associated to each other. Therefore, the mapping from a c-structure to an f-structure is purely procedural and not derivational, which is the crucial difference between c-/f-structure division and the bistratal deep/surface structure distinction. 1.3.4 From Form to Meaning The notion of correspondence in LFG is an important one: it provides a general way of correlating different kinds of linguistic information in separate domains through modular specifications. We have seen that the word string,.

(41) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 15. through the mediation of PSR's, first maps to a c-structure correspondence, which is then mapped to an f-structure correspondence by the instantiation of functional schemata and the operation of unification. Kaplan (1989) suggests that the correspondence idea may be extended for the entire mapping between the external form of an utterance and its internal representation of meaning, as shown in the following configuration. utterance mediation of PSR’s c-structure. unification specified by functional annotations. f-structure. anaphoric structure. semantic stucture. discourse structure. meaning Figure 1.2 Correspondence of Linguistic Representations in LFG By extending the same conceptual framework of description and correspondence, the same existing mathematical and computational techniques in mapping c- to f-structure can be applied to other systems of linguistic information. For the semantic component in LFG, refer to Halvorsen (1983). 1.3.5 Further Readings The most important compilation of works in LFG to date is still Bresnan (1982), of which Chapter 4 (i.e., Kaplan and Bresnan 1982) describes.

(42) 16. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. systematically LFG theory and formalism. Two other chapters that are also important in terms of theoretical significance are Chapter 1 (Bresnan 1982a), which justifies the advantages of lexical processes in a lexical theory over transformational analyses, in the case of passivization, and Chapter 5 (Bresnan 1982b) where a theory of control relations and complementation in LFG is presented. The section on LFG of Sells (1985) constitutes a well-balanced and comprehensive introduction to the theory and formalism. Likewise, Chapter 4 of Horrocks (1987) is a clearly illustrated and explicated introduction on LFG. As for detailed, step-by-step instructions on working with LFG's formalism and unification process, Wescoat (1987) is a most practical manual. Kaplan (1989) provides a description of LFG's formal architecture and a summary of the recent developments of LFG's formalism since Kaplan and Bresnan (1982): functional uncertainty, functional precedence, and the natural-class organization of grammatical functions (also known as the Lexical Mapping Theory). The Lexical Mapping Theory started with the pioneering work of Levin (1986) and was formally introduced in Bresnan and Kanerva (1989). The extended technique of functional uncertainty is described in Kaplan and Zaenen (1989a), and Kaplan and Zaenen (1989) present a discussion on functional precedence in LFG. For the operation of unification, Shieber (1986) is still the best reference. 1.4 vLFG: A Variant LFG Formalism In this section, we will describe the formalism and specific notations of grammar writing within the vLFG employed in this study. We will illustrate that vLFG maintains most of LFG's theoretical constructs, including its most distinctive feature: the division of constituent structure and functional structure. The major differences of vLFG include the timing of building the functional structure, the notation of lexical entries and phrase structure rules, and the informational organization of the lexicon. The syntactic component of a vLFG grammar, like LFG, contains a lexicon, a set of phrase structure rules augmented with functional annotations, and a set of well-formedness constraints on f-structures. To reach a c-structure and a corresponding f-structure for a word string, individual lexical items contained in the string are first looked up in the lexicon. When a matching lexical entry.

(43) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 17. is found, all feature inheritance (FI) entries specified in it will be expanded. When the lexical item is fully specified, it can then be inserted under a category equivalent to its categorial specification. When phrase structure rules apply to assign a partial c-structure of the sentence, unification takes place at the same time to assign a partial f-structure to the c-structure. Thus, whenever a final c-structure is reached, a final f-structure is also reached, at the same time. We will now step through each one of these stages and describe the various notations of lexical entry, feature inheritance entry, and augmented phrase structure rules, and also the operative mechanisms associated with them. 1.4.1 The Informational Organization of the Lexicon The vLFG utilizes a feature inheritance structure in the lexicon that does not exist in LFG. A vLFG lexicon is therefore composed of lexical entries and feature inheritance entries. Another grammatical theory that fully takes advantage of such a scheme of feature inheritance in the informational structure of its lexicon is the Lexicase framework (Starosta 1988). We will illustrate the different structures of lexical entries in LFG and vLFG. 6. LFG: a. WORD X, (↑FEATURE1) = VALUE1 (↑FEATURE2) = VALUE2 (↑FEATURE3) = VALUE3. `categorial feature `functional schemata. b. forces V, (↑ PRED) = 'force <(SUBJ) (OBJ) (XCOMP)>' (↑ OBJ) = (↑ XCOMP SUBJ) (↑ SUBJ PERSON) =c 3rd (↑ SUBJ NUMBER) =c SG.

(44) 18. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. 7. vLFG: a. WORD: CAT X FS FEATURE1 VALUE1 FEATURE2 VALUE2 FI-1. `keyword string `dag `category (CAT) `f-structure (FS) `feature ìnheritance èntry (FI). b. forces V, FS FORM ‘force’ FI-V-PIVOTAL FI-3RD.SG c. FI-V-PIVOTAL: CAT V FS PRED <SUBJ , OBJ , XCOMP> (↑ OBJ) = (↑ XCOMP SUBJ) d. FI-V-3RD.SG: FS PERSON 3RD NUMBER SG.

(45) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 19. A lexical entry in vLFG is composed of two fields: one, the keyword string, and two, the definition of the word, i.e., the grammatical information associated with that word, organized in a bracket format. Since the information specified in this format is completely equivalent to a directed acyclic graph, or "dag" in short, we will refer to them as dags, following the common usage in the literature of unification grammars. Thus, a lexical entry is composed of a keyword string and a dag depicting its grammatical information, which includes the categorial specification of the lexical item and its f-structure. The information represented in the above bracket format is identical with the one below; however, for the sake of typographical ease we will adopt the following format. 8. a.. WORD: [ CAT X FS [ FEATURE1 VALUE1 FEATURE2 VALUE2 ] FI-1 ]. b.. 'keyword string 'dag. forces: [ FS [ FORM 'force' ] FI-V-PIVOTAL FI-3RD.SG ]. Information that is idiosyncratic to the lexical item is specified in the word dag. Information that is shared with a word class, be it functional or otherwise, can be specified through feature inheritance entries. A feature inheritance (FI) entry is very similar in structure to a lexical entry, except that it contains information common to a class of lexical items, as illustrated below. The information contained in the entry FI-1 is shared by a class of words that have this FI-1 specified in their word entries..

(46) 20. 9.. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. FI-1: [ FS [ FEATURE3 VALUE3 FEATURE4 VALUE4 ] ]. èntry name `dag. It is easily noticeable that an entry of feature inheritance is very similar in structure as a lexical entry. When a lexical entry has its FI entries expanded to become fully specified, the expansion is achieved through the unification of the lexical dag and the FI dags. It is crucial to point out that the unification between a lexical dag and a feature inheritance dag is by way of default, or extension as it is sometimes called. It is also equivalent to the LFG mechanism of priority union proposed in Kaplan (1987). If there is a conflict in terms of the value of certain feature, unification by extension will not fail; rather, the value of the lexical dag is preserved and the conflicting value of the feature inheritance entry will be overwritten. Assuming that value1 does not equal value2 below, we will contrast unification, signaled by =, with the operation of extension, signaled by ≈: 10.a. Unification (=) [ feature1 value1 ] = [ feature1 value2 ] → fails, no unification b. Extension (≈) [ feature1 value1 ] ≈ [ feature1 value2 ] → [ feature1 value1 ] By allowing unification by extension, idiosyncratic behaviors of a lexical item can be fully accounted for and the generalizations can be stated in a maximally general way. A lexical entry may inherit information from more than one feature inheritance entry, and a feature inheritance entry may in turn call upon other feature inheritance entries and acquire more information from them. A feature inheritance structure thus not only maximizes the economy of the size of the lexicon but also fully captures the generalizations of word.

(47) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 21. classes while still allows the idiosyncrasies of individual lexical items. A homograph is indicated by a disjunction mark {}: 11.. WORD: { [ CAT X FS [ FEATURE1 FEATURE2 ] FI-1 ] [ CAT Y FS [ FEATURE1 FEATURE3 ] FI-2 ] }. 'keyword string 'homonym 1 VALUE1 VALUE2. 'homonym 2 VALUE1 VALUE3. 1.4.2 The Phrase Structure Rules A syntactic rule in the vLFG consists of three parts: rule name, pattern matching, and actions. Like LFG, phrase structure rules are augmented with functional expressions. 12. PSR-1: X Y: → Z. <↓ FEATURE1> =c VALUE1 GRAMMATICAL FUNCTION. 'rule name 'pattern 'action. The pattern matching section specifies what kinds of syntactic elements are needed in order to build a higher category. A constituent may assign a certain grammatical function, indicated by either: or ε followed by a function name; the former usually assigns a subcategorizable function, while the latter assigns an adjunctive function. We will illustrate these two symbols with their LFG equivalents below. The only required part of the action section is the.

(48) 22. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. specification of the higher category that is to be built. When the higher category is built, in the meantime its corresponding f-structure is also built. In the event that the building of the f-structure fails, the higher category will not be built. Other actions, if any, then perform the specified operations on the f-structure thus far built by the rule. Again, if any of the operation of the actions fails, the higher category still will not be built. We will illustrate the differences between the notation of LFG and vLFG below. 13. LFG: a. S → b. VP → c. NP →. NP (↑ SUBJ)=↓ V ↑=↓. VP ↑=↓ (NP) (↑ OBJ)=↓. (DET) (A) ↑=↓ ↓ε(↑ ADJ). 14. vLFG: a.. PSR-1: NP: SUBJ VP → S. b.. PSR-2: V (NP): OBJ → VP. N ↑=↓.

(49) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. c.. 23. PSR-3: (DET) (A) ε ADJ N → NP. Given a lexicalist linguistic theory like LFG, we feel that it is more suitable intuitively to express in a phrase structure rule the constituents first: X Y -> Z, rather than Z -> X Y. The former indicates that when two elements, X and Y, are found, Z is built and that between X and Y the one not assigned with any grammatical function is the head. In terms of both parsing and generation, this scheme comports better with the lexicalist view of grammar where it is the actual lexical items that dictate the structure of a string and activate the parser or generator for processing. The pattern section allows optionality and disjunction, and the specification of partial ordering is also facilitated. 15.. a. [ X Y ] b. < X Y Z > c. { X Y } d. < 1-2; X Y Z > e. (X) f. X+ g. X*. `X immediately precedes y `X, Y and Z in any order èither X or Y, but not both `X, Y and Z in any order except that `X must precede y; this specifies `partial ordering `X is optional òne or more X `zero or more X; equivalent to (X+). 1.4.3 Unification: from C-structure to F-structure In conventional LFG, phrase structure rules first build the entire c-structure of the sentence. Unification then proceeds according to the specified functional equations on the c-structure to build a corresponding f-structure (Wescoat 1987). In the vLFG however, as we have just seen from the vLFG rule format, when any legal portion of the c-structure of a sentence is built, it must have simultaneously an f-structure built to correspond to it. In other words, when a phrase structure rule applies to build a higher category, the.

(50) 24. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. functional equations specified by that rule are applied at the same time. And only when the f-structure associated with the c-structure is also sound, can the higher category be built. As explicated by Figure V in Wescoat (1987:10), LFG does allow this alternative. We will give a simple example to illustrate some of the points we have covered thus far, assuming the rules in 12 and 13 above, the following lexicon in vLFG, and the lexicon of 2 in LFG, given the sentence: "John loves Mary." 16. a.. Mary: [ CAT N FS [ FORM 'Mary' PERSON 3rd NUMBER SG ] ]. b.. John: [ CAT N FS [ FORM 'John' PERSON 3rd NUMBER SG ] ]. c.. loves: [ CAT V FS [ PRED <SUBJ , OBJ> FORM 'love' TENSE PRESENT SUBJ [ PERSON 3rd NUMBER SG ] ] ].

(51) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 17. LFG: a.. NP. `no unification `no f-structure. ↑=↓ N John <↑PRED>='John' b.. NP. `no unification `no f-structure. ↑=↓ N Mary <↑PRED>='Mary' c.. VP. `no unification `no f-structure. ↑=↓ V. <↑OBJ>=↓ NP. loves <↑PRED>= 'love <(SUBJ)(OBJ)>'. ↑=↓ N. Mary <↑PRED>='Mary'. 25.

(52) 26. GRAMMATICAL FUNCTIONS AND VERB SUBCATEGORIZATION IN CHINESE. d.. S <↑SUBJ>=↓ NP. ↑=↓ VP ↑=↓ V. ↑=↓ N. John <↑PRED>='John'. loves <↑PRED>= 'love<(SUBJ)(OBJ)>'. ùnification `f-structure. <↑OBJ>=↓ NP ↑=↓ N Mary <↑PRED>='Mary'. Only when a c-structure that expands the whole word string is reached does unification proceed to build the f-structure according to the functional annotations on the final c-structure: 18. PRED 'love <(SUBJ)(OBJ)>' SUBJ [ PRED 'John'] OBJ [ PRED 'Mary'] The vLFG on the other hand builds a c-structure and a corresponding f-structure simultaneously in every step:.

(53) LEXICAL FUNCTIONAL GRAMMAR AND A VARIANT FORMALISM. 19.. vLFG: a.. c-structure NP │ N │ John. b.. NP │ N │ Mary. c.. VP. 27. f-structure [ FORM 'John' NUMBER SG PERSON 3rd ] [ FORM 'Mary' NUMBER SG PERSON 3rd ]. V. NP. loves. N Mary. [ PRED <SUBJ , OBJ> FORM ‘love’ TENSE PRESENT SUBJ [ NUMBER SG PERSON 3rd ] OBJ [ PERSON 3rd NUMBER SG FORM 'Mary' ] ].