The handbook of contemporary semantic theory Ed. by Shalom Lappin (review)

REVIEWS171 Van Valin, Robert D., and William Foley. 1984. Functional syntax and universal grammar. Cambridge: Cambridge University Press. Department of American Indian Languages and Culture University of Copenhagen Njalsgade 80 DK-2300 Copenhagen S Denmark [soerenw@coco.ihi.ku.dk] The handbook of contemporary semantic theory. Ed. by Shalom Lappin. Oxford & Cambridge, MA: Blackwell, 1996. Pp. xvii, 670. Reviewed by Leo Obrst, The MITRE Corp., McLean, VA It is nearly impossible to convey the richness of the formal treatments of semantics and pragmatics presented in this book. No other text provides such an array of substantive and upto -date discussions of facets of natural language semantics. Twenty-two individual semanticists describe important issues in contemporary semantic theory by summarizing past arguments and detailing current research in virtually every subdiscipline of semantics. In 670 pages, eleven general topics are elucidated: formal semantics in linguistics; generalized quantifier theory; the interface between syntax and semantics; anaphora, discourse, and modality; focus, presupposition, and negation; tense; questions; plurals; computational semantics; lexical semantics; and semantics and related domains. Nearly the only subject areas in semantics which are underrepresented here are aspectuality (and a theory of events, though Fred Landman's paper sketches one such theory) and propositional attitudes. To these I would add a methodological topic which also warrants inclusion within these pages: compositionality, a principle with which any theoretical semantics must grapple. These gaps, however, are easily filled by other literature (e.g. aspectuality: Verkuyl 1996; compositionality: Janssen 1997 but see also Zadrozny 1994, whom John Nerbonne, p. 467, also cites) and are utterly inconsequential compared to the wealth of content on every other aspect of semantic theory in this volume. In the introduction, the editor characterizes the papers in the volume as exemplifying five themes found in contemporary semantic theory: (1) the increasing elaboration of the dynamic aspects of semantics; (2) an expansion of the formal repertoire of tools used in semantics, including extensions of type theory and model theory and the introduction of more complex algebraic structures into domains of models; (3) the tensions between the two perspectives on the syntactic-semantic interface, the one primarily syntactic in nature and exemplified by the abstract syntactic level of logical form of government-binding theory, and the other primarily semantic in nature and more surface-oriented, represented by categorial grammar (CG) and HPSG, for example; (4) the question as to whether there is an intermediate level of semantic representation between a given syntactic structure and its model-theoretic interpretation; and (5) the lack of a common theoretical framework to accommodate the research being performed in lexical semantics and that in formal semantics. Rather than attempt to do justice to the book's individual discussions within this limited space or even to map the individual chapters to the editor's themes, I will instead discuss five of the chapters which I believe illustrate multiple components of the five themes and hope that by this method I can best impart a sense of the richness of the volume as a whole. Of all the articles in this book, Barbara Partee's 'The development of formal semantics in linguistic theory' (11-38) is perhaps the most indispensable from the perspective of situating contemporary semantic theory in its brief historical context and, by the nature of its goal, the paper which touches all five themes. We are reminded, for example, that in the early days of semantics (in the generative paradigm) in the 1960s, there was no truth-conditional-based formal semantics at all. Investigation was primarily focused on lexical semantics and the identification 172LANGUAGE, VOLUME 74, NUMBER 1 (1998) of a minimal set of primitive semantic features into which meaning could be decomposed; the enterprise borrowed heavily from the methodologies ofpsychology and structuralist anthropology rather than from logic and the philosophy of language. Most linguists and logicians, in fact, did not see the logical methods developed for formal languages as being applicable to natural languages until Montague's (1973, 1974) work in the early 1970s. Employing model-theoretic techniques which had been developed in important work in modal logic, Montague's central premise was that a grammar could be constituted in terms of a syntax and a semantics...