Sub-Sententials: Pragmatics or Semantics?

Stainton points out that speakers “can make assertions while speaking sub-sententially”. He argues for a “pragmatics-oriented approach” to these phenomena and against a “semantics-oriented approach”. In contrast, I argue for a largely semantics-oriented approach: typically, sub-sentential utterances assert a truth-conditional proposition in virtue of exploiting a semantic convention. Thus, there is an “implicit-demonstrative convention” in English of expressing a thought that a particular object in mind is F by saying simply ‘F’. I note also that some sub-sentential assertions include demonstrations and argue that these exploit another semantic convention for expressing a thought with a particular object in mind. I consider four objections that Stainton has to a semantics- oriented approach. The most interesting is the “syntactic ellipsis” objection, which rests on two planks: (A) the assumption that this approach must claim that what appears on the surface to be a sub-sentential is, at some deeper level of syntactic analysis, really a sentence; (B) the claim that there is no such syntactic ellipsis in these sub-sentential utterances. I argue that (A) is wrong and that (B) may well be. I also reject the other three objections: “too much ambiguity”; “no explanatory work”; and “fails a Kripkean test”. Nonetheless, occasionally, sub-sentential utterances semantically assert only a fragment of a truth-conditional proposition. This fragment needs to be pragmatically enriched to yield a propositional message. To this extent a pragmatics-oriented approach is correct.