Passive and impersonal sentences Edited by Vincenzo Lo Cascio (review)

442LANGUAGE, VOLUME 53, NUMBER 2 (1977) Green, Georgia. 1976. Main clause phenomena in subordinate clauses. Lg. 52.382-97. Grice, H. Paul. 1975. Logic and conversation. In Cole & Morgan, 41-58. Hooper, Joan, and Sandra Thompson. 1973. On the applicability of root transformations . Linguistic Inquiry 4.465-98. Kraak, A. 1967. Presuppositions and the analysis of adverbs, ms, MIT. Lakoff, George. 1970a. Irregularity in syntax. New York: Holt. ------. 1970b. Linguistics and natural logic. Ann Arbor: Phonetics Laboratory, University of Michigan. [Reprinted in Semantics ofnatural language, ed. by Donald Davidson & Gilbert Harman, 545-665. Dordrecht: D. Reidel, 1972.] Lasnik, Howard. 1970. The scope of negation, ms. Smith, Steven Bradley. 1972. Relations of inclusion. Lg. 48.276-84. Vendler, Zeno. 1967. Verbs and times. In his Linguistics in philosophy, 97-121. Ithaca, NY: Cornell University Press. [Received 21 June 1976.] Passive and impersonal sentences. Edited by Vincenzo Lo Cascio. (Italian linguistics , 1.) Lisse: Peter de Ridder Press, 1976. Pp. 170. $6.50. Reviewed by Donna Jo Napoli, Georgetown University The appearance of this volume, the first of a new journal, is timely in that it coincides with the recent blossoming of much serious, modern linguistic work on Italian. Thejournal aims to present research by scholars ofdiverse linguistic schools and to promote comparative studies of Italian with other languages. This initial volume contains five essays, representing at least four different approaches to the study of language. As I discuss each of these in turn, it must be kept in mind that these articles were submitted in 1974. Thus some of the comments below relate to the state of the art then, as well as to the individual works. In the first article, 'Appropriateness conditions for the use of passives and impersonate in Italian' (11-32), Guglielmo Cinque supports the by now well-accepted hypothesis that in Italian, as in many other languages, word order is determined largely by such notions as old vs. new information, or topic vs. comment. He argues against the proposal that Passive is used to present the 'logical' subject as new information, showing instead that Passive topicalizes the 'logical' object. His analysis explains the infrequency of passives with inanimate surface subjects and human agents—as well as the high frequency of passives in relative clauses, where the logical object is coreferential with the head of the relative clause. Cinque's article is impressive for its organized and convincing presentation, as well as for its insight into problems of anaphora and intonation that are currently at the center ofcontroversy in generative studies. This article is typical of the high quality of work which Cinque habitually produces. The second article, 'On "linguistic variables" and primary object-topicalization in Italian' (33-76), by Vincenzo Lo Cascio—the principal editor of the journal, and the inspiration behind it—is an ambitious one, raising many important and difficult questions. Again and again, one is struck by Lo Cascio's ability to find examples that demonstrate the importance of an understanding of tense and modality for any description ofthe semantics ofItalian, or ofmany other languages for that matter. The organization and theoretical framework ofthe article, however, REVIEWS443 are not easily discernible; thus the subcategorization of various predicates is discussed with some attention to selectional restrictions,1 but then we find the claim that passive sentences do not derive from active ones with no explanation for the facts of subcategorization or selectional restrictions. At another point in the article, the difference between verbs like volere 'want' and the modals, with respect to the embedding of a passive sentence under them, is summed up with the observation that the surface subject is what is understood as the logical subject for volitional verbs, but not for the modals. No explanation is offered for this difference, which is easily explained in a generative framework that accepts underlying and superficial structures. Still, a very interesting discussion of 'momentary' events in contrast to 'situations', and of the possibility of Passive with respect to this distinction, forms one ofthe best parts ofthe article. The conclusion is that the more 'variable' arguments we have in a sentence—i.e., imprecise ones, or ones not lexically determined in the surface—the more 'universal, indistinct...