Negative ... concord?

The main claim of this paper is that a general theory of negative concord (NC) should allow for the possibility of NC involving scoping of a universal quantifier above negation. I propose that Greek NC instantiates this option. Greek n-words will be analyzed as polarity sensitive universal quantifiers which need negation in order to be licensed, but must raise above negation in order to yield the scoping For All>(*) over bar * (sic). This gives the correct interpretation of NC structures as general negative statements. The effect is achieved by application of QR, and the account is fully compositional, as only sentence negation is the vehicle of logical negation (sic). Greek n-words are also compared to n-words in Romance, Slavic, and Hungarian. This analysis, if correct, has two important consequences. First, the analysis will provide a strong argument for retaining QR in the syntax-semantics mapping: we need it in order to interpret NC. Second, by employing a mechanism which is present in the grammar for the scope of quantifiers anyway, we have a simpler theory which makes NC look less anomalous; appeal to a mechanism invoked just to account for NC, as in the "negative absorption'' tradition, is thus rendered unnecessary.