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

Foreword Classical NEG Raising: The First 900 Years Laurence R. Horn, Yale University The monograph by Chris Collins and Paul Postal (C&P) that you are about to read went to press on the Golden Anniversary of the first syntactic analysis for the “Transposition of NOT (EVER)” within a fragment of generative grammar: “Under certain conditions (e.g. after verbs like WANT or THINK which are themselves not negated), a NOT in the embedded sentence may be moved in front of the main verb . . . ” (Fillmore 1963:220). In motivating both his rule and the syntactic cycle itself, illustrated by the “repeated applications” of this movement rule involved in his proposed derivation of (1a) from the structure directly underlying (1b), Charles Fillmore did not cite any explicit evidence for a syntactic analysis. He relied instead on the (purported) paraphrase relation between (1b) and one reading of the putatively ambiguous (1a) (Fillmore 1963:220n12): (1) a. I don’t believe that he wants me to think that he did it. b. I believe that he wants me to think that he didn’t do it. The paraphrase relation posited here rests on somewhat shaky ground; Dwight Bolinger was widely credited for pointing out (in a 1967 letter to George Lakoff) that the force of negative commitment is weaker in (the relevant reading of) (1a) than it is in (1b). In this, Bolinger was recapitulating the remark of the grammarian Poutsma (1928:105), who had noted four decades earlier that “the shifting of not often has the effect of toning down the negativing of a sentence.” Poutsma’s dynamic metaphor is striking: even 30 years before transformational rules, it was natural to conceive of the placement of not in movement terms. After some years of neglect, Fillmore’s syntactic approach to what had come to be called negative transportation or NEG raising (NR), following the introduction of the NEG element in Klima 1964 (see Horn 1971 on the early terminological shifts), began to be supported with direct evidence. The most compelling argument was advanced by Robin Lakoff (1969b). Citing an earlier x Foreword personal communication from Masaru Kajita, Lakoff pointed out that strict or strong polarity items (e.g., until midnight, in weeks) normally requiring a tautoclausal negative licenser are nevertheless well-formed in the scope of NR predicates with higher negation, and furthermore (for some speakers) license positive opposite-polarity tags formed on the embedded clause: (2) a. I didn’t think John would leave until tomorrow. b. *I didn’t say John would leave until tomorrow. c. I don’t suppose the Yankees will win the pennant, will they? d. *I don’t suppose the Yankees will win the pennant, {won’t they?/do I?} These contrasts provide support for a grammatical rule of NR that accounts for the negative polarity item (NPI) licensing and tag formation before the NEG is raised into the main clause. The functional transparency of NR predicates (and only NR predicates) that overrides the locality restriction on strict NPIs turns out to be more complex than it first appeared. While Fillmore’s constraint that all predicates intervening between negative licenser and strict NPI must be NEG raisers is a necessary condition for grammaticality in sentences like (1a) and (2a), it is not sufficient, as seen in the licensing asymmetry between (3a) and (3b) from Horn 1971:120–121, an asymmetry for which Gajewski (2007) offers a new explanation: (3) a. I don’t believe John wanted Harry to die until Saturday. b. *I don’t want John to believe Harry died until Saturday. The polarity argument for NR, further chronicled in Horn 1978, is discussed insightfully and in great detail by C&P; see also Israel 2011 for general considerations on the character and distribution of strong polarity items. Over the years, empirical and theoretical considerations have gradually led linguists away from Fillmore and Lakoff (and the Horn of 1971) and down the trail blazed by Jackendoff (1971:291): “The synonymy between John thinks that Bill didn’t go and one reading of John doesn’t think that Bill went is inferential in character and has nothing to do with the syntactic component —it may even have nothing to do with the semantic component.” (This analytic trend is reviewed in Horn 1978 and Horn 1989 [2001:chap. 5].) The result has been a de facto rejection of the syntactic approach to NR, a rejection that C&P, on new grounds, urge us to reject. In essence, the pragmatic view of the NEG...

Share