Logical form: From GB to minimalism By Norbert Hornstein (review)

142LANGUAGE, VOLUME 74, NUMBER 1 (1998) language other than English than the BE students were when they entered school! H does not of course discuss the US case, but to his immense credit, he recognizes the great multilingual potential of France's own linguistic minorities—remember especially his assumption that all children would maintain their 'langue maternelle'; these minorities can assist importantly in H's ambitious multilingual educational program, in which every French schoolchild would attain some degree of communicative ability in at least three languages other than French. We can only hope that H's program will succeed and that its success will encourage efforts along these lines elsewhere in Europe and around the world. REFERENCES Freudenstein, Reinhold. 1996. Foreign language teaching after the year 2000. Georgetown University Round Table on languages and linguistics 1996, ed. by James E. Alatis, Carolyn A. Straehle, Maggie Ronkin, and Brent Gallenberger, 43-54. Washington, DC: Georgetown University Press. Hagège. Claude. 1996. Le français, histoire d'un combat. Boulogne: Editions Michel Hagège. Hamers, Josiane, and Michel Blanc. 1983. Bilingualité et bilinguisme. Bruxelles: P. Mardaga. Weinreich, Uriel. 1953. Languages in contact. New York: Linguistic Circle of New York. Department of Anthropology and Program in Linguistics Binghamton University (SUNY) PO Box 6000 Binghamton, NY 13902-6000 [straight@binghamton edu] Logical form: From GB to minimalism. By Norbert Hornstein. Cambridge, MA: Blackwell, 1995. Pp ix, 267. Reviewed by Daniel L. Everett, University of Pittsburgh This is a well written and thought-provoking introduction to the study ofmeaning in Chomsky's minimalist program (MP). It argues that MP enjoys significant conceptual and empirical advantages over both government and binding theory (GB) and nonderivational theories which place the burden of syntactic explanation on output conditions on syntactic representations. The range of coverage, general clarity of writing, importance of the subject matter, and status of the author make this an important book which every linguist interested in meaning should own. My overall evaluation of this book is positive. However, I am going to focus this review on a few (but nontrivial) objections that arose in my reading, because I believe that these negative aspects are more useful in understanding the place of this research program in modem linguistics research. The book begins with two bemusing statements. First, Homstein claims that in MP 'those phenomena ... characteristic of meaning are explicitly represented in a phrase marker whose structure is obtained from an SS phrase marker by successive application of Move-a' (5).1 In fact, only a small percentage of meaning-related phenomena can be derived in this way, namely (and somewhat circularly), structure-based meanings. Were s/he to take H literally here, the reader might get the erroneous impression that LF has something to say about, e.g. connotation vs. denotation, metaphor, figurative meaning, antinomy, etc. Second, H claims that LF is 'formally very similar to the logicians' logical form' (10) (what I will label If). But this is true only in the broadest and nearly meaningless of ways, namely, that If and LF both have recursion and syntax: If has no quantifier raising (QR), move-a, empty category principle, or full interpretation, to 1 H should not be blamed too much for these types of statements since they arise all too naturally in Chomskyan theory, which has generally been marked by an off-putting, glaring type of false advertising, namely, that it claims to be a theory of language (or ?-language'), when in fact it is nothing of the sort. Generative linguistics has never been a theory of language and always and only a theory of some aspects of grammar (namely, those for which no semantic, sociological, or functional factor can be proved to play a causal role). REVIEWS143 name a few obvious differences between If and LF. Moreover, If, unlike LF, lacks content items (i.e. words). I doubt if professional logicians would be very sanguine about identifying If and LF, even in the most general of ways. All but three of the nine chapters in this book are built around specific empirical issues that have traditionally received considerable attention in formal theories of grammar—general properties of logical syntax, antecedent-contained deletion, weak...