Conversational maxims as social norms

ABSTRACT

I argue that although Paul Grice’s picture of conversational maxims and conversational implicature is an immensely useful theoretical tool, his view about the nature of the maxims is misguided. Grice portrays conversational maxims as tenets of rationality, but I will contend that they are best seen as social norms. I develop this proposal in connection to Philip Pettit’s account of social norms, with the result that conversational maxims are seen as grounded in practices of social approval and disapproval within a given group. This shift to seeing conversational maxims as social norms has several advantages. First, it allows us to neatly accommodate possible variation with respect to the maxims across well-functioning linguistic groups. Second, it facilitates a more psychologically plausible account of flouting. And third, it generates insights about the nature of social norms themselves.

KEYWORDS:

• Paul Grice
• conversational maxims
• social norms
• flouting
• conversational implicature

Acknowledgements

For helpful conversation related to the ideas in this paper, I am grateful to Daniel Hieber, Meredith McFadden, the students in my Winter 2018 graduate seminar in philosophy of language at McMaster University, and an audience at Western University. I am also grateful to several anonymous referees for helpful feedback on the paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 For an actual case of intergroup variation, consider Min-Sun Kim and Steven Wilson’s (1994, 223, 225–226) finding that native speakers of Korean judge direct forms of requests to be clearer but less effective than indirect forms, whereas native speakers of American English judge direct forms of requests to be both clearer and more effective. American English speakers thus seem to judge perspicuity to be a more effective conversational strategy than Korean speakers do, with respect to requests. This suggests some subtle intergroup variation with respect to the ‘Be perspicuous’ maxim (cf. 227–228).

2 This case is not hypothetical. Eve Danziger (2010, 211) contends that among the Mopan Maya people of Central America, the maxim is ‘Make your contribution one that is true’ rather than ‘Try to make your contribution one that is true.’ Her justification for this claim is that in a study she conducted, Mopan respondents did not seem to consider speakers’ intentions and beliefs to be relevant to the blameworthiness of false utterances (210).

3 Elinor Ochs Keenan (1976) describes an actual case that she interprets as variation with respect to Quantity. She observed that among the Malagasy people of Madagascar, if a man is asked where his mother is, he might reply, ‘She is either in the house or at the market,’ even if he knows for certain that she is in the house. This kind of response ‘is not usually taken to imply that [the speaker] is unable to provide more specific information needed by the hearer,’ suggesting genuine variation with respect to the ‘Be informative’ maxim (70). Keenan offers several other examples of contexts in which Malagasy practices with respect to providing information are more conservative than within the groups with which Grice was familiar. Also with respect to Quantity, Anna Wierzbicka (1991) describes differences in what gets communicated by utterances of tautologies within different linguistic groups. For instance, translations of ‘War is war’ do not seem to have the same significance among Russian speakers or French speakers as ‘War is war’ has among English speakers (394). This might be taken to suggest variation with respect to the Quantity maxims, although Wierzbicka herself takes it to imply that the significance of particular tautologies is a semantic matter (392).

4 Going forward, I will use the term ‘intergroup variation’ as shorthand for ‘variation across well-functioning linguistic groups.’

5 This example is closely modeled on one I heard from Peter Graham.

6 For more on this notion of arbitrariness, see Stotts 2017.

7 I should note that I am not the first to portray conversational principles as something other than tenets of rationality. Dan Sperber and Deirdre Wilson (1986, 32; cf. Lepore and Stone 2015, 60) portray the single relevance principle with which they would replace Grice’s maxims as a psychological law, not a tenet of rationality. And Laurence Horn (1984, 1989) and Stephen Levinson (2000) each portray their Grice-inspired sets of conversational principles as linguistic rules (at least as Lepore and Stone (2015, Ch. 3) interpret them). These views are not direct competitors to mine because they include deep alterations to the content and structure of Grice’s conversational principles, which my view does not. Arguing for my approach of treating conversational principles as social norms with content based closely on Grice’s original set of maxims over these other approaches would be another paper in its own right, so I leave these other views aside in the main text. But I must also acknowledge a broader debt to Lepore and Stone (2015), whose perspicuous discussion of how central Grice’s view about the rationality of the maxims is to his overall program, and of the ways in which he has been challenged on that point, was part of the impetus for this paper.

8 Pettit (1990, 751) eventually adds a common belief requirement to this account. I haven’t used that version because I find his reasons for adding common belief uncompelling and thus, in my judgement, the resulting account requires too much mental activity on the part of population members. I also want to acknowledge that the idea of conversational maxims as social norms in Pettit’s sense has cropped up elsewhere, in relation to the Quality maxims. In the context of using the Quality maxims within an account of lying, Don Fallis (2009, 2012) makes it clear that he thinks of Grice’s conversational maxims as social norms in accordance with Pettit’s account (see especially 2012, 565). Additionally, Peter Graham (2015, 260) has discussed Pettit’s definition of social norms in connection to the Quality maxims, although he treats the relevant social norms as additional norms that have ‘the same content as Grice’s maxims,’ where the maxims themselves are still tenets of rationality.

9 Thus, seeing conversational principles as social norms still allows conversational implicatures to be calculable (Grice 1989, 31).

10 This proposed revision to Pettit’s account is modeled on one condition in Bicchieri’s (2006, 11) account of social norms, but her condition is ultimately quite different because the belief in conformity component occurs as part of the antecedent of a conditional preference for conforming.

11 This gloss of prima facie approval is closely modeled on Andrew Reisner’s (2013) definition of a prima facie ought.

12 This belief may never be brought to consciousness. But the idea is that with a bit of prompting, people in my linguistic group could be brought to assent to similar claims about how much effort toward truthfulness is required in their group, perhaps by drawing on comparisons to other actual or imagined groups.