In lieu of an abstract, here is a brief excerpt of the content:

430 LANGUAGE, VOLUME 74, NUMBER 2 (1998) tion). Finally, there are the inevitable one or two papers whose circuitous paths or mismatched introductions and conclusions make them appear to have been written up in the hotel the night before. [Stephanie Gottwald and Margaret Thomas, Boston College.] Licensing empty nouns in French. By Petra Sleeman. The Hague: Holland Institute of Linguistics, 1996. Pp. 205. Petra Sleeman presents an interesting and thorough study of noun ellipsis in French, constructions in which N, dominated by DP, is empty, as in Trois pro arriveront demain "Three will arrive tomorrow'. S follows other work on ellipsis in claiming that 'noun ellipsis' actually involves an empty NP pro which must be licensed and identified. Her analysis departs from other approaches to pro, however, in proposing that the empty nominal in French DP is not licensed and identified by morphological agreement , it is licensed by the semantic feature [ + partitive ] and identified by discourse linking with an antecedent which must be specific in interpretation. S extends her account to certain related constructions in English, Dutch, Spanish, and Italian. Ch. 1 is a short introduction, with Ch. 2 providing the core of the analysis elucidated in subsequent chapters. S first develops a theory whereby different kinds of French prénommai adjectives and determiners are specified for the semantic feature [ + partitive ]. On her account, cardinals, superlatives, ordinals, seul/autre, etc., color adjectives, and certain adjectives of quality such as grand/petit are all partitive . Other adjectives of quality are nonpartitive and fail to license ellipsis. Thus, a nonpartitive adjective such as intéressantes is not a potential licenser of pro in *Je n 'ai pas entendu les deux intéressantes pro. ? have not heard the two interesting (ones).' Ellipsis is grammatical in cases in which pro is licensed by a numeral, a color adjective, or a partitive adjective of quality. Examples are, respectively: Trois pro arriveront demain "Three will arrive tomorrow .' Je prends la petite verte pro ? will take the small green (one).', and Je préfère les deux grands pro ? prefer the two tall (ones).' The requirement thatpro be identified by specificity is illustrated by the contrast between Trois pro arriveront demain and *J'ai lu trois pro ? have read three' . In the latter, the DP object containing pro is nonspecific, and identification fails. In the former, as in the other grammatical examples discussed above, DP is specific , and NP pro is both licensed and identified. S extends her analysis to English, Spanish, and Italian and argues that English differs from French in relaxing the requirement that pro in DP be specific, from which it follows that 7 have read three' in English is grammatical. In her briefdiscussion ofDutch, German , and the Scandinavian languages, S proposes that empty NP in these languages is licensed and identified through adjectival inflection, concluding that at least two different licensing and identification strategies for NP pro exist, one involving semantic features, and the other, morphological agreement features . A central tenet of S's analysis of the phrase structure of DP is that prenominal elements, including certain adjectives, numerals, etc., are specifiers of functional projections dominating NP. She does not elaborate on the nature of the functional heads of these projections and assumes that many such heads are phonologically empty with feature specifications, though exactly what these specifications are remains unclear. Proper government (licensing) of empty NP is through Spec-head agreement of a lexical element in Spec with an empty head, and empty Xo in such cases licenses pro. Numerals, for example, are not generated in Num but rather in the Spec of a higher functional QP. Num itself heads a lower NumP and is presumably the locus of features, though this is not made clear. Why numerals are not associated with the projection NumP is not addressed. In order to account for the distribution of personal and other pronouns in French, S proposes in Ch. 3 that these pronouns are actually DP dominating NP pro in which either D or its specifier are filled. Demonstratives such as celui, possessive pronouns such as sien, and quantifiers such as certains, plusieurs, chacun, and quelques-uns, among others, are pronouns generated in Spec of...

pdf

Share