Negative indefinites (review)

Abstract

Reviewed by: Negative indefinites Elena Herburger Negative indefinites. By Doris Penka. (Oxford studies in theoretical linguistics.) Oxford: Oxford University Press, 2011. Pp. 264. ISBN 9780199567270. $45. In Negative indefinites Doris Penka focuses on three topics in the syntax and semantics of negation: negative concord, the split-scope readings of negative quantifiers in German, and the restricted distribution of quantifiers in Scandinavian. P argues that in each instance what looks like a negative quantifier is really an indefinite whose apparent negative force derives from a separate— at times tacit, at times overt—negation operator with which the indefinite enters an agreement relation. In addition to methodically making a case for analyzing negative quantifiers as NEGATIVE INDEFINITES, the book also discusses a large number of existing analyses and paradigms, offering a valuable and timely survey of some of the current literature. The first third of the book is devoted to negative concord, where what looks like one and the same expression (N-WORD) appears to have negative force in certain contexts and merely existential force in others. P rejects various previous accounts, including the view that so-called n-words like Spanish nadie or Russian nikto are negative quantifiers (e.g. Zanuttini 1991), the view that they are their negative polarity counterparts (e.g. Laka 1990, Ladusaw 1992), and the view that they are ambiguous between the two (Herburger 2001). Adopting Zeijlstra's (2004) approach instead, she analyzes n-words like nadie/nikto as existential quantifiers in need of syntactic agreement with an overt or tacit negation operator. P then extends the analysis to French, which poses particular challenges. Since the account of negative concord P adopts bears architectural resemblances to the negative polarity item (NPI) accounts of negative concord, it faces similar challenges. To explain the occurrence of n-words without negation (e.g. in elliptical answers and in preverbal positions in nonstrict negative concord languages like Spanish) it posits a silent negation operator. To constrain the distribution of this operator, P appeals to an economy condition that forces the insertion of a silent negation operator if the derivation would otherwise crash. Because P assumes that the negation with which the n-word agrees always takes scope over the event operator, her account does not capture the occurrence of n-words that appear in postverbal position without a negation and that are interpreted as narrow-scope negative quantifiers (Herburger 2001). In connection with the general resemblance of the negative indefinite and the NPI accounts of negative concord just pointed out, it is worth noting that P insists that negative concord and NPI [End Page 663] licensing are separate phenomena. One argument is that n-words do not fit neatly into van der Wouden's (1997) classification of NPIs. Another reason is that unlicensed n-words result in ungrammaticality, whereas unlicensed NPIs according to the analyses advanced by Kadmon and Landman (1993) and Lahiri (1998), for example, are merely pragmatically infelicitous. It may be, however, that the fact that the analyses in question rule out 1 as pragmatically bizarre and not as ungrammatical is a drawback of these analyses and not an independently established fact one should confidently base an argument on (Herburger 2011). 1. *She ever said that. Another reason a reader might remain a bit skeptical of P's assertion that negative concord has nothing to do with NPI licensing is examples where n-words are licensed in comparatives or by predicates like doubt. These require P to say that the relevant feature ([iNEG]) is not limited to (c)overt negation operators but can also appear on other elements, which happen to be NPI licensors. In one of the last chapters P discussed the various similarities of the account of negative concord she adopts with that put forth in Ladusaw 1992. In that context she makes important observations about how the unselective binding analysis of indefinites that Ladusaw adopts must be modified in order to capture all scopal interactions between the negation operator and the n-word. P herself, however, does not commit to any specific semantic analysis of indefinites. The second third of the book deals with the split-scope readings of negative and other downward-entailing quantifiers. The...