Mind, Technology, and Society Talk Series

Speaker: Jonathan Cohen Professor of Philosophy, University of California, San Diego Time/Date: 3-4:30 p.m. Monday, March 30th, 2015

Location: Chancellor’s Conference Room, KL 232

Title: “Conversational Eliciture” Co-author: Andrew Kehler (UCSD Linguistics)

Abstract: There is a form of extrasemantically conveyed content that fails to fit neatly into any of the types of pragmatic enrichment thus far described in the literature. Consider, for example:

(1a) A jogger was hit by a car in Palo Alto last night.

(1b) A farmer was hit by a car in Palo Alto last night.

Although not entailed, the indefinite ‘a jogger’ in (1a) strongly invites the inference that the victim was jogging at the time of the accident. In contrast, the analogous inference for (1b) — that the farmer was farming at the time — is not normally evoked.

The phenomenon on display in (1a), which we label “eliciture”, shows that a speaker’s decision to use a particular way of referring to an entity over other alternatives invites the hearer to draw inferences that, crucially, are not triggered by any syntactic relationship or other type of felicity requirement on linguistic material.

We argue that elicitures cannot be reduced to other forms of pragmatic enrichment such as Gricean conversation implicature, Relevance Theoretic explicature, Bach’s impliciture, or any type of local pragmatic strengthening.

And we suggest a way of understanding the inferences underlying eliciture as following from a set of general cognitive (not specifically linguistic) strategies for building mental models of the world. In particular, we propose to explain these inferences in terms of hearers’ attempts to bring coherence to heard clauses by subsuming them under coherence relations mirroring Humean associative relationships between mental states.

Attention to conversational eliciture suggests that the forms of extrasemantic enrichment that have attracted the most theoretical attention to date (e.g., conversational implicature, scalar implicature) are in fact special cases of a more general phenomenon. However, the general phenomenon differs from the special cases. For one thing, the general phenomenon appears not to be triggered by a delimited range of linguistic lapses (e.g., unsaturated parameters, violations of cooperative norms) that drive the special cases. For another, the general phenomenon appears to draw in unconstrained ways on arbitrary non-linguistic beliefs about the world in ways that many of the special cases (e.g., scalar implicatures) arguably do not. We argue from these considerations that the special cases are atypical in their theoretical tractability, hence that theories of these cases cannot be scaled up to the general case. If so, then eliciture holds important negative lessons about the possibility of systematic, predictive theorizing in this area.

Bio (taken from http://aardvark.ucsd.edu/): Jonathan Cohen completed a B.A. and M.A. at the University of Chicago before earning a Ph.D. at Rutgers University followed by a one-year postdoc at the University of British Columbia. His principal interests are in philosophy of perception and philosophy of language, and especially on questions in these areas that interact with the cognitive sciences.