Integrating syntactic theory and variationist analysis: The structure of negative indefinites in regional dialects of British English

Abstract

This paper integrates syntactic theory and variationist analysis in an investigation of the variation between English not-negation (I don’t have any money), no-negation (I have no money) and negative concord (I don’t have no money). Using corpora of three varieties of UK English spoken in Glasgow, Tyneside and Salford respectively, I test two theoretical accounts of the variation. Account 1 applies Zeijlstra’s (2004) agreement-based theory of negative concord to all three variants, such that n-words (e.g. nobody) which feature in no-negation and negative concord are not inherently negative but agree with a negative operator in a higher NegP. Under Account 2, no-negation is instead derived via negative-marking within the DP followed by movement to the higher NegP for sentential scope (Kayne 1998; Svenonius 2002; Zeijlstra 2011). These accounts, together with observations about the raising properties of functional versus lexical verbs, lead to the formulation of different hypotheses about the distribution of variants in speech according to verb type, verb phrase complexity, and the discourse status of the propositions expressed. Results of distributional analysis and mixed-effects modelling support Account 2 of the variation over Account 1, suggesting structural identity between not-negation and negative concord (in contrast to no-negation). This supports Tubau’s (2016) proposal that English negative indefinites have two distinct structures: one in which negation is marked syntactically in the DP and one in which they agree with a syntactically-higher NegP.