transparent gif


Ej inloggad.

Göteborgs universitets publikationer

The Pragmatics Of Insensitive Assessments

Författare och institution:
Gunnar Björnsson (Institutionen för filosofi, lingvistik och vetenskapsteori)
Publicerad i:
6th International Symposium of Cognition, Logic and Communi­cation Formal Semantics and Pragmatics: Discourse, Context, and Mod­els 19-21 November 2010, Riga, Latvia,
Konferensbidrag, övrigt
Sammanfattning (abstract):
The pragmatics of insensitive assessments Gunnar Björnsson, Linköping University, University of Gothenburg Alexander Almér, University of Gothenburg Natural language contains an array of phrases dedicated to expressing disagreement and agreement with and assessments of the veridicality of beliefs and utterances: “yes”, “no”, “that is true”, “that is false”, “you are wrong about that”, “you are mistaken”, “I agree”, “I disagree”, “she knows that”, “it is not the case that”, and so forth. When we assess utterances using such phrases, we normally (barring misunderstanding, etc) seem to assess whether the conditions that the utterer had been concerned to get across are satisfied. In many cases, this means being sensitive to the context of utterance and the communicative interests of the utterer. If we want to assess whether Dana was correct when she told Eric, “They live in the first big building to the right of the cathedral”, we need a rich understanding of the context to figure out what Dana referred to by “they”, what counts as a “big” building, relative to what perspective we should understand “right” and what cathedral is meant. Conversely, Dana is interested in communicating that certain conditions hold, and interested in using an expression that will communicate just that in the context of utterance. There is a convergence of speaker and assessor interests: in making her utterance Dana wants to get across that certain conditions are satisfied, and as assessors of the truth of what has been said, we want to determine whether those conditions are satisfied. Though this might be the typical case, the debate about so-called assessor relativism about expressions like “might” and “tasty” has been largely driven by cases in which seemingly felicitous assessments of utterances (or beliefs) are strikingly insensitive to aspects of the context of utterance that were highly relevant to the speaker’s choice of words. Here is one example, concerning “might”, where Alice has asked whether anyone has seen her keys. Bill, who left them in the car, answers: (1) “The keys might be in the car.” Bill’s concern in uttering (1) is presumably to convey that the key’s being in the car is not ruled out by information available to him, or to the parties of the conversation, and for that purpose, his utterance seems perfectly felicitous. Unknown to Alice and Bill, however, the neighbourhood girl, Emily, has just stolen the keys from the car and is hiding behind a bush, listening to the conversation. Here are two possible thoughts of hers in response to Bill’s utterance: (2) “No, they can’t be, because I have them in my pocket.” (3) *“That’s true, but I have them in my pocket.” Even though Emily had no grounds for thinking that the key’s being in the car wasn’t ruled out by what Bill and his interlocutors knew, (3) strikes people as defective whereas the insensitive assessment in (2) seems felicitous. Similar examples of seemingly felicitous insensitive assessments can be given for other expressions, such as “tasty”, “funny”, “ought”, “should”, “if” and graded adjectives like “rich”. Judging by speakers’ communicative concerns in using sentences like (1), the following contextualist analysis of “might” seems appropriate: MIGHT-C: An utterance of a sentence of the form P might be the case is true if and only if P is compatible with the body of information that is relevant in the context of utterance (cf. DeRose 1991, 1998; Bach 2008; Schaffer 2009). However, judging by typical reactions to (2) and (3), assessments of such utterances are insensitive to the truth-conditions postulated by MIGHT-C, instead following the pattern expected from the following assessment-relative analysis: MIGHT-R: An utterance of a sentence of the form P might be the case is true relative to a context of assessment if and only if P is compatible with the information available in that context (cf. MacFarlane 2005; Egan 2007) Given MIGHT-R, Emily is correct to reject Bill’s utterance since the keys’ being in the car is incompatible with the body of information available in the context where she assesses (1). The most prominent strategy for handling this assessor-relativist challenge within a contextualist framework is to reject the existence of felicitous insensitive assessments (Cappelen & Hawthorne 2009, pp. 110-111, and 118, e.g.). We argue that this strategy fails: such assessments are part and parcel of everyday discourse. In Almér and Björnsson (2009), we proposed another strategy, which allows the contextualist to mimic the relativist’s account. The central move is to reject the standard assumption that felicitous assessments of utterances need concern the satisfaction of the truth-conditions of those utterances, and substitute for it: C-ASSESSMENT: When we assess utterances using various assessment phrases we normally, (barring confusion, misunderstanding, etc) assess the satisfaction of the conditions that are made most immediately salient by the utterances in the context of assessment as we assess the utterances. In most cases, the conditions made most salient by an utterance are its truth-conditions. But in cases of felicitous insensitive assessments, such as Emily’s assessment of (1), thinking about the utterance in the context of assessment makes some other conditions more salient than the truth-conditions of the utterance. Expressions like “might”, “tasty”, etc provide philosophically interesting families of exception cases. Other exceptions are more mundane: (4) A: “I [wonder if / believe that] Bill lost the keys.” (5) B: “No, he didn’t, he had them in his pocket.” In most natural settings, a “no” in response to (4) would not express rejection of the claim that A wonders if (or believes that) Bill lost the keys, but rather rejection of the claim that Bill lost the keys. Various further considerations might determine whether this contextualist accommodation is preferable to the assessor relativist alternative. Whatever account one prefers, however, one needs an explanation of why these assessments are elicited by certain context sensitive or relative expressions but not by, say, paradigmatic indexicals and demonstratives. For assessor-relativists, this is the question of why only certain expressions have come to contribute to assessor-relative truth-conditions, and to do so in a variety of languages; for contextualists who appeal to C-ASSESSMENT, the question is why conditions other than the truth-conditions of utterances involving those expressions become salient in the context of assessment. Our task in this paper is to explain why certain expressions invite insensitive assessments whereas others do not. We sketch a pragmatic account of insensitive assessments compatible with both relativism and contextualism, and argue that it fits more naturally with the contextualist suggestion since pragmatic factors seem conversationally operative rather than ossified in an assessor-relative semantics. The pragmatic account identifies a variety of factors that affect the felicity of insensitive assessments. These include, among others, (i) presumptions of speaker authority, which we suggest are operative in explaining the felicity of (5), (ii) the accusatory connotations of everyday uses of “that’s false”, “you are wrong” or “that’s not true”, which invite assessment of utterances or beliefs from the speaker’s point of view, thus making it inappropriate for Emily to respond to (1) by any of these, and (iii) the role of explicit relativizations in sentences like “it is tasty for me” and “given what we know, the keys might be in the car”, which help keeping focus on the condition that concerned the speaker. Another factor, of crucial importance in explaining insensitive assessments of epistemic modals and conditionals as well as in the case of deontic judgments, is that some contexts might be systematically pragmatically superior to others relative to conversational interests. In the case of claims to the effect that P might be the case, the speaker has typically expressed an epistemic attitude of holding P open. When, as in most conversations involving epistemic “might”, it is an issue what epistemic attitude to hold toward P, it is highly relevant to reject holding P open if it is incompatible with information available to the assessor, whether it is available to the speaker or not. Moreover, when it is an issue what epistemic attitude to take towards P, it will typically be conversationally irrelevant to express an assessment of the information-relative claim that the speaker was concerned with if the speaker made that claim relative to inferior information. This explains why insensitive assessments of might-judgments are often quite natural from informationally superior positions but quite unnatural from informationally inferior positions or in contexts where the interest is not to decide whether to hold P open. All these factors, we argue, affect what conditions we should expect to be made immediately salient by utterances and beliefs as they are being assessed. This suggests that the pragmatic account supports C-ASSESSMENT. References Almér, A. and Björnsson, G. 2009. Relativism, Contextualism and Insensitive Assessments. Logique et Analyse 52:208 Bach, K. 2008. Perspectives on possibilities: Contextualism, Relativism, or what? In Egan and Weatherson forthcoming Cappelen, H. and Hawthorne J. 2009 Relativism and Monadic Truth, Oxford U. P. DeRose, K. 1991. Epistemic Possibilities. Philosophical Review 100, 581-605. DeRose, K. 1998. Simple Might’s, Indicative Possibilities, and the Open Future. The Philosophical Quarterly 48, 67-82. Egan, A. 2007. Epistemic modals, relativism, and assertion. Philosophical Studies 133, 1–22. Egan, A. & B. Weatherson, eds. Forthcoming. Epistemic Modality, Oxford U. P. MacFarlane, J. 2005. Making Sense of Relative Truth. Proceedings of the Aristotelian Society 105, 321–39. Schaffer, J. 2009. Contextualism for Taste Claims and Epistemic Modals. In Egan and Weatherson forthcoming.
Ämne (baseras på Högskoleverkets indelning av forskningsämnen):
Filosofi, etik och religion ->
Filosofi, etik och religion ->
Filosofi ->
Teoretisk filosofi
relativism, contextualism, insensitive assessments, MacFarlane, Egan
Postens nummer:
Posten skapad:
2010-11-26 14:41
Posten ändrad:
2010-11-26 18:50

Visa i Endnote-format

Göteborgs universitet • Tel. 031-786 0000
© Göteborgs universitet 2007