Abstract

One of the major debates in pragmatics continues to concern the status of minimal semantic propositions. While Minimal Semanticists (Borg 2004, Cappelen & Lepore 2005) defend semantic truth-conditions, Contextualists (eg Carston 2002, Recanati 2004) argue that it is only utterances that have truth-conditional content. A further, related, issue is whether minimal propositions are psychologically real. Mostly, this is understood to relate to processes of utterance interpretation. In this talk, I will take a slightly different perspective on these questions and argue that a notion of minimal, but pragmatic, content is needed for another reason: to capture the speaker’s restricted overt commitment and accountability (cf Searle 1969, Liedtke 1995).

The number and kinds of constituents of the minimal pragmatic content of an utterance are determined by the lexical-conceptual structure of the sentence used, which is ‘conventional’ (ie context-independent) without necessarily corresponding to its syntactic structure. The content-side is determined not by convention, but by an individual’s information processing. It is argued that this entity enjoys another kind of psychological validity, which is often neglected. It is the entity that is accepted, and whose content is negotiated, by speaker and hearer in an interpersonal domain. Similar to Cappelen & Lepore’s idea of ‘shared content’ (2005, 2006), I assume that speech behaviour provides evidence for the boundaries of minimal content, although I disagree about which boundaries deserve attention.

In contrast to the semantic minimalists’ formally driven content, minimal pragmatic content is genuinely pragmatic. As a result, neither the difficulty of truth-evaluability nor Cappelen & Lepore’s problem of semantic content being both context-independent and asserted arises. At the same time, minimal pragmatic content significantly differs from the contextualists’ what is saidpragm, and indeed from their speaker meaning, in scope and function. I will argue that it would be not only inappropriate in view of the speaker’s commitment, but also unnecessary to postulate a richer notion of utterance content in the restricted interpersonal domain, which I regard as the proper domain of speaker meaning. In this way, the ‘slippery slope’ to unrestricted context-sensitvity that Cappelen & Lepore (2005) warn about is avoided for independent reasons.