Lexical-Functional Grammar: An Introduction to Parallel Constraint-Based Syntax (review)

Abstract

Reviewed by: Lexical-functional grammar: An introduction to parallel constraint-based syntax by Yehuda N. Falk Mark Miller Lexical-functional grammar: An introduction to parallel constraint-based syntax. By Yehuda N. Falk. Stanford: CSLI Publications, 2001. Pp. 237. ISBN 1575863405. $22. In this textbook on lexical-functional grammar (LFG), Falk addresses students already familiar with elementary generative syntactic theory (including transformational theory) and wishes ‘to motivate the concepts and formalisms of LFG in relation to derivational approaches’ (xi). His purpose in doing so ‘is to argue that LFG provides a better model of syntax’ (3) than derivational theories like government/binding theory and the minimalist program in its typological plausibility, its ease of conformity to what is known about the computation of language, and its ‘well-developed formalism’ (29). Throughout the book, F focuses on English, citing the pedagogical value of using a language already familiar to students and of providing a more cohesive treatment of the theory. F’s introductory chapter, ‘Welcome to lexical functional grammar’ (1–32), identifies LFG as ‘a variety of generative grammar’ though not of transformational theory (2). F shows how LFG is ‘lexical’ in that it obeys the lexical integrity principle (LIP) which claims that ‘syntax cannot see the internal structure of words’ (5). In contrast to derivational approaches, LFG is monotonic (in that information cannot be changed or deleted), is constraint-based, and has just one level of constituent structure. LFG is ‘functional’ in claiming that grammatical functions are independent of constituent structure configurations. It models a ‘parallel architecture’ (23) with three kinds of information existing simultaneously: c(onstituent) structure, f(unctional) structure, and a(rgument) structure. These layers of c-structure, f-structure, and a-structure are discussed in Chs. 2, 3, and 4, respectively. In Chs. 2(33–55), F explains how LFG, unlike transformationalist theories, prohibits nonfunctional empty categories and defines well-formed c-structures according to language-specific phrase structure rules. Ch. 3 (57–91), on f-structure, introduces the basic grammatical functions and demonstrates the formal mapping procedure from c-structure to f-structure. For F, ‘this mapping is the heart of the descriptive power of LFG, since it deals with the relationship between overt syntactic elements and the features they represent’ (64). The ‘mathematical relations of correspondence’ (66) involved in such mapping may prove formidable for some students, however, in the degree of abstraction involved, such as the use of metavariables in functional annotations to c-structure nodes. In Ch. 4 (93–120), F shows how a-structure mediates the mapping between thematic roles and grammatical functions, using lexical mapping theory (LMT) to analyze passives, unergatives, and unaccusatives. The remainder of the book analyzes several English constructions based on LFG formalisms: Ch. 5 on control (equi and raising constructions) (121–48); Ch. 6 on long distance dependencies such as islands and pied piping (149–72); and Ch. 7 on anaphora (173–91). Throughout these chapters, F uses empirical arguments to show how LFG better explains the facts of these constructions than does a transformationally based theory of syntax. For example, he notes that, because transformational theory forces an anaphoric analysis for equi constructions, the theory ‘cannot account for the properties of function-controlled equi’ (144) whereas LFG can. Features that should make this book useful to students include: exercises after every chapter, a glossary of technical terms (Appendix A), and a ‘Minigrammar of English’ (Appendix B) which represents the ‘final versions of all the rules and lexical entries’ developed in the preceding chapters (209). This textbook on LFG provides the student already trained in basic generative syntax with an alternative model to the standard transformational framework. For students unaccustomed to the conventions and precision of mathematical models, the LFG formalism may prove challenging. However, F demonstrates (usually convincingly) that, compared with a derivational approach, LFG offers a superior empirical, descriptive account of the facts of English and other languages. Mark Miller University of Texas, Arlington Copyright © 2003 Linguistic Society of America