Triggering presuppositions

Abstract

While presuppositions are often thought to be lexically encoded, researchers have repeatedly argued for 'triggering algorithms' that productively classify certain entailments as presuppositions. We provide new evidence for this position and sketch a novel triggering rule. On the empirical side, we show that presuppositions are productively generated from iconic expressions (such as gestures) that one may not have seen before, which suggests that a triggering algorithm is indeed called for. Turning to normal words, we show that sometimes a presupposition p is triggered by a simple or complex expression that does not even entail p: it is only when contextual information guarantees that the entailment goes through that the presupposition emerges. On standard theories, this presupposition could not be hardwired, because if so it should make itself felt (by way of projection or accommodation) in all cases. Rather, a triggering algorithm seems to take as an input a contextual meaning, and to turn some contextual entailments into presuppositions. On the theoretical side, we propose that an entailment q (possibly a contextual one) of an expression qq' is treated as a presupposition if q is an epistemic precondition of the global meaning, in the following sense: usually, when one learns that qq' (e.g. x stops q-ing), one antecedently knows that q (e.g. x q-ed). Presuppositions thus arise from an attempt to ensure that information that is cognitively inert in general experience is also trivial relative to its linguistic environment. On various analyses, q is trivial in its linguistic environment just in case q is entailed by its local context; this provides a direct link between presupposition generation and presupposition projection. (An appendix discusses the relation between this proposal and an alternative one in terms of entailments that are in some sense counterfactually stable).