Historical Romance morphology By James M. Anderson and Bernard Rochet (review)

424LANGUAGE, VOLUME 58, NUMBER 2 (1982) In an example like 3 (not Ladd's), an analysis of accent concentrating on contrast would have little explanation regarding what to actually contrasts with: d (3) We thought about what we could o o to but then we realized there was nothing do. The accent on to is not so much an accent which properly belongs to the word to, but rather to the phrase to do; however, do is anaphorically de-accented, and the accent has no place to go but onto the to. Ladd's discussion is much more intricate than this simple example might suggest, and quite interesting. In conclusion, Ladd's book contains many extended and thoughtful discussions of issues that should be faced more frequently than they are these days. It definitely should be read. REFERENCES Goldsmith, John. 1976. Autosegmental phonology. MIT dissertation. [Published, New York: Garland Press, 1979.] Ladd, D. Robert. 1978. Stylized intonation. Lg. 54.517-40. Leben, William. 1976. The tones in English intonation. Linguistic Analysis 2.69-107. Liberman, Mark. 1975. The intonational system of English. MIT dissertation. [Distributed by Indiana University Linguistics Club.] Walton, A. Ronald. 1976. Syllable and tone in Chinese. Cornell University dissertation. Yip, Moira. 1980. The tonal phonology of Chinese. MIT dissertation. [Distributed by Indiana University Linguistics Club.] [Received 23 July 1981.] Historical Romance morphology. By James M. Anderson and Bernard Rochet. Ann Arbor: Published for the University of Calgary by University Microfilms International, 1979. Pp. 395. $20.00. Reviewed by Suzanne Fleischman, University of California, Berkeley It has been a number of years since anyone has undertaken to write a new, genuinely comparative, historical morphology of Romance. Among 'classics' of the genre, the first (1836-44) was appropriately from the pen of the founder of Romance linguistics, Friedrich Diez, followed a half century or so later by Meyer-Lübke's celebrated Grammatik der romanischen Sprachen (1890-1902)— which, though bearing the same title as its predecessor, reflects very solidly the Neogrammarian ideology in vogue during the latter part of the century. These monuments of 19th century philology in the German tradition continue to be cited today, if perhaps more commonly in French translation. The early 20th century saw the appearance of the first major comparative venture written in a Romance tongue, Bourciez's Eléments de linguistique romane (1910), which for all its defects is now in a fifth edition (1967)—which serves as the principal information source for the volume being reviewed here. Two more enterprises deserve mention in this brief history of the tradition in which the REVIEWS425 A&R volume is set: Tagliavini's Le origini délie lingue neolatine (1949) and Lausberg's Romanische Sprachwissenschaft (1962-63), both now available in Spanish translation, and both more up-to-date and more rigorous than Bourciez .1 It should be noted that all the above in effect go beyond the confines of historical morphology and are properly classed as historical grammars, though three of the five works (Diez, Meyer-Lübke, Lausberg) contain a volume devoted entirely to morphological evolution. In view of this spectrum of available and, on the whole, very respectable historical morphologies, one might legitimately ask: Why yet another one? What do A&R have to offer that their predecessors did not? The answer is: relatively little, inasmuch as their material derives wholly from other sources, and they offer nothing really innovative in theory or methodology. Yet they do bring together here, for the first time in English and in a relatively nontechnical presentation, extensive data from five 'major' Romance speech areas, and they organize these data so as to maximally facilitate cross-language comparison . There is clearly something to be said for such an undertaking, ifcarried out successfully; and A&R have achieved a measure of success. The major strength of the book lies in making available a vast array of data drawn from five continental Romance standards: French, Spanish, Portuguese, Italian, and Rumanian. The book is thus ideal in principle for anyone wishing to compare or contrast the various Romance speech areas with respect to a particular formation or category of grammar. The difficulty, however, is that the reader not already thoroughly conversant with Romance historical grammar...