The Computation of Scalar Implicatures: Pragmatic, Lexical or Grammatical?

Abstract

The problem that I focus on in this paper is whether scalar implicatures are truly derived through pragmatic means. I provide an overview of the alternatives to the pragmatic approach that I have come across, as well as provide my own thoughts on the matter and offer some questions for further examination. I begin with an introduction for the non-linguist of scalar implicatures and the (neo-)Gricean theory as described by Geurts (2011), after which I identify two other competing theories laid out by Sauerland (2012), focusing on some-all implicatures. I then explain the advantages that the grammatical theory has over the pragmatic theory, with an emphasis on SIs embedded in disjunctions that Sauerland and Chierchia et al. (2013) point out. Considering these arguments, I offer a critical review of these papers and conclude that, despite the pragmatic theory facing challenges still, it seems that some degree or pragmatic reasoning is essential to understanding scalar implicatures, even if that means accepting a pragmatic-lexical hybrid approach.