Intonation systems: A survey of twenty languages Ed. by Daniel Hirst and Albert Di Cristo (review)

460LANGUAGE, VOLUME 76, NUMBER 2 (2000) 9 brings the associate relation to bear on the question of locality, that is, on how far a constituent can move without having moved too far. The sum of these four chapters offers a beautiful example oflinguistic argumentation, as S-Y Kuroda's blurb on the dustcover puts it, 'an excellent balance of theoretical finesse and descriptive craftsmanship'. The fact that I admire this book, and Cs work more generally, is not to say that I am entirely satisfied. Three fundamental and interrelated questions intruded on my reading. C takes a fairly hard line on what drives linguistic analysis. Her impulse is not to follow the peculiarities of the language under investigation where they might lead but to fix the realm of possible analyses by theoretical assumption. This impulse plays out most obviously in her general approach to morphology. Although a great deal is gained in Cs treatment ofChamorro by separating syntactic and morphological effects, some of this gain is made at the expense ofrelegating the morphological aspects of the language to that set beyond the limits of linguistic theory because they vary from language to language. One question has to do, then, with the idiosyncrasy of morphology. My own work on agreement has focused almost entirely on what C calls feature compatibility and, more particularly, on the morphological processes that underpin it. In response to the multiple 'feature compatibility' aspects of Luiseño—and almost as a mirror image to the approach taken by C—I tried to argue (Steele 1989, 1990) that these morphological details actually yield the syntactic structure. Should agreement depend on syntactic structures, one might ask about its necessity. But if agreement yields the structure, its necessity is obvious. Cs analysis of Chamorro might appear to undercut this conclusion because the agreement that rums on the associate relation serves to provide a kind of record of movement. But, then a question about the character of the associate relation arises. The associate relation, as noted, overlays a tree. Thus, we are again confronted with two geometries—the tree and the associate relation—and it is reasonable to ask about the necessity of both. If this line of thinking is right, it leads to the most general question: what is the role of idiosyncrasy in linguistic analysis? Or, better, what is the relationship between the particulars of a language and the universality of language? Is the associate relation the Chamorro equivalent of Luisefio's conditions on the distribution of person and number, in the sense that both are specific manifestations of a more general principle governing the construction of structure? My cavil is not that Cs answer would likely be no, but rather that such questions are not at issue. But the fact that this book raises such questions offers further evidence of its value. REFERENCES Chuno, Sandra. 1982. Unbounded dependencies in Chamorro grammar. Linguistic Inquiry 13.39-77. -----. 1983. The ECP and government in Chamorro. Natural Language and Linguistic Theory 1.207-44. -----. 1990. VPs and verb movement in Chamorro. Natural Language and Linguistic Theory 8.559-619. Kayne, Richard. 1994. The antisymmetry of syntax. Cambridge, MA: MTT Press. Steele, Susan. 1989. Subject values. Language 65.537-78. -----. 1990. Agreement and anti-agreement: A syntax of Luiseño. Natural Language and Linguistic Theory Book Series. Dordrecht: Kluwer. University of Connecticut U-171 Storrs, CT 06269-2171 [steele@gulley.vpa.uconn.edu] Intonation systems: A survey of twenty languages. Ed. by Daniel Hirst and Albert Di Cristo. Cambridge: Cambridge University Press, 1998. Pp. xii, 487. Reviewed by Amedeo De Dominicis, University ofBologna Fortunately, in the last fifteen years interest in linguistic investigation of intonation has significantly increased. This book treats twenty languages with specific empirical and theoretical in- REVIEWS461 sights on prosody. The book gives a measure of the force of attraction that the so-called 'school of Aix en Provence' has exerted in the international linguistic community. The turn came with new branches of generativism that have developed from the early 1980s under the names of metrical and autosegmental phonology. Intonation has been conceived as an autonomous set of features, linked to the segmental phonetics line by means of rules...