Gricean Quality

Abstract

According to the Knowledge Account of Assertion, assertions are governed by the rule that One must: assert that p only if one knows that p.1 Some philosophers oppose recent arguments for the Knowledge Account by claiming that assertion, being an act much like any other, will be subject to norms governing acts generally, such as those articulated by Grice for the purpose of successful, cooperative endeavours (Sosa 2009; Cappelen 2011; Goldberg 2013). Because asserting is part of a social cooperative activity aimed at varying goals, and because Grice's Cooperative Principle is a general principle meant to govern all rational social behavior (of which conversation is a subset), these philosophers suspect that Gricean resources will suffice for explaining any norm(s) relevant to assertion. In this paper I show how Grice is a traitor to their cause; or rather, they are his dissenters, not his disciples. Contrary to what they suppose, Grice thought of asserting as a special linguistic act in need of its own norm, and his discussions of the maxim of Quality point to knowledge.2 I develop an argument that the Quality maxim is not dependent, in the appropriate way, on the Cooperative Principle. If it is not thus dependent, then the Cooperative Principle cannot be the explanation of, or source of normativity for, the Quality maxim. Thus, not only is Grice of little help in resisting the Knowledge Account; in fact, leveraging powerful insights informing the maxim of Quality reveals new resources for a distinctive positive case that knowledge is the constitutive norm of assertion. Timothy Williamson's version of the Knowledge Account claims that the knowledge rule is constitutive of the speech act of assertion. The rule is constitutive of assertion in that it is essential to making the speech act what it is, much like the rules of a game constitute the game; and the rule is strict, in that "it imposes a perfect duty, one that applies strictly to each and every assertion" (Turri 2014a). Constitutivity of this rule implies strictness3 (though perhaps not vice versa); so a denial of its strictness entails a denial of constitutivity. consider giving up on the presupposition that there is a particular norm distinctive of assertion as such. It is normatively problematic even to risk causing someone to have a false belief, whether one takes that risk through assertion (ignorant or otherwise), through other forms of representation, or even through more direct means—by stimulating their cortex, for example. But this involves a norm that governs behavior in general: assertion is not normatively distinctive. (Sosa 2009, 272) Grice's maxims of conversation are not constitutive of the acts they govern. Grice takes them to be derived from general principles of rational cooperation. …They are norms that guide behavior, not norms that are essential to (or constitutive of) the behavior they guide. (2011, 24) part of being cooperative is to say only that for which one has adequate evidence (Quality), where what counts as adequate evidence is determined by mutual beliefs regarding the needs and expectations of the various participants in the conversation. (Goldberg 2013, 188, fn. 41) Sosa and Cappelen think that the norm of assertion is not constitutive of the practice because that practice can be explained by more general principles, in particular those due to Grice. Goldberg claims that Grice's maxim of Quality is itself a flexible norm of assertion not strictly requiring knowledge, and that its flexibility derives from the Cooperative Principle. In the next section I shall show that Goldberg's claim is not well motivated, and that Gricean Quality, in its most plausible version, does require knowledge. Then in §3., I present an argument for the conclusion that the maxim of Quality cannot be explained by, because it is not dependent on, more general conversational principles such as the Cooperative Principle; indeed, this a point which Grice himself came to appreciate. Such an argument puts considerable pressure on those dissenters, such as Cappelen, who wish to claim both that the norm of assertion is not constitutive and that Grice's system provides the resources to support such a claim. Given all this, it is better then to take a liberal reading of Quality and regard its sub-maxims as requiring one to assert only what one believes to be true, and only that for which one has adequate evidence; and if we charitably interpret Quality's supermaxim as being concerned with asserting truth rather than with trying to assert the truth, the norm is factive (indeed, as we'll see later in §3., Grice himself understands Quality as demanding truth and forbidding falsity). Putting these two together, we get the plausible norm of roughly justified true belief. Yet now we are very close to knowledge indeed.7 If this were all to be said in favor of the connection between Gricean Quality and the Knowledge Account, the case would be somewhat weak. But Grice elsewhere makes the important connection between Quality and knowledge: the possibility of clashes between the demands of Quality and other maxims can not only generate conversational implicatures, but can also reveal knowledge as the core of Quality. this infringement of the first maxim of Quantity [make your contribution as informative as is required (for the current purposes of the exchange)] can be explained only by the supposition that B is aware that to be more informative would be to say something that infringed the second maxim of Quality, "Don't say what you lack adequate evidence for," so B implicates that he does not know in which town C lives. (1989, 32–33, emphasis mine) Another common response to a clash between Quantity and Quality is important here. Suppose a question is asked directly about whether a proposition q is the case: schematically, "Is q?" If one feels one cannot answer directly because one suspects that one does not fulfil the demand of Quality, one will often under such circumstances reply with "I don't know," or "I can't say".9 In unpublished papers,10 Grice regards the "I don't know" response as conforming to Quality at the expense of violating Quantity, for it would seem to provide too little information: it offers no help with respect to whether or not q. Yet if Quality did not demand knowledge, it is puzzling how the "I don't know" response could be perceived to be a clash between the requirements of Quantity and Quality. Furthermore, if, against the Knowledge Account, we took Quality to be lax enough to permit non-known contributions, then the "I don't know" reply would seem to violate the maxim of Relation ("Be relevant!"). For if Quality permitted assertions that are not known, it would seem irrelevant and unhelpful for one to point out that one doesn't know. But when prompted by a question, the answer to which one does not know (or takes oneself not to know), we do not judge the "I don't know" or "I can't say because I don't know" responses to be irrelevant or unhelpful. We deem them relevant because they signal to others the speaker's own deference to the epistemic demand of the maxim of Quality, and we deem them helpful because we normally don't want someone to answer with an outright assertion when he or she doesn't (take him- or herself to) know.11 I submit then that the most plausible interpretation of Grice's own discussion is that knowledge is at the core of the maxim of Quality. Even if he doesn't explicitly endorse this, following his material where it leads results in understanding Quality in terms of the Knowledge Account. For only on this interpretation can we well explain why knowledge (and not some other epistemic status) figures in conversational patterns surrounding prompts and their responses, as well as in the explanation of clashes between the demands of Quality and other maxims.13 In this section I argue that the maxim of Quality is not dependent, in the way that the other maxims are, on the Cooperative Principle.14 This argument is independent of the success of the claims made in the last section, that is, it is independent of thinking of Quality in terms of knowledge. And it is an argument that will hold interest for those who think of Grice's theory of communication as built up from fully general principles, governing all rational cooperative behavior, to more specific conversational maxims that hold in virtue of their relationship to the general principles.15 Make your contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged. (1989, 26) But the maxim of Quality seems to enjoy a distinctive status in comparison with the other conversational maxims: trangressing some of the other maxims (e.g. the maxims of Manner, or Quantity, by being overly prolix) may be thought to be less important, and subject to lighter criticism, than violating Quality. Thus Grice says, It is obvious that the observance of some of these maxims is a matter of less urgency than is the observance of others; a man who has expressed himself with undue prolixity would, in general, be open to milder comment than would a man who has said something he believes to be false. Indeed, it might be felt that the importance of at least the first maxim of Quality is such that it should not be included in a scheme of the kind I am constructing; other maxims come into operation only on the assumption that this maxim of Quality is satisfied.16 While this may be correct, so far as the generation of implicatures is concerned it seems to play a role not totally different from the other maxims, and it will be convenient, for the present at least, to treat it as a member of the list of maxims. (1989, 27, emphasis mine) Acknowledgement that Quality might best be given a class by itself, and that the other maxims operate only on the assumption that Quality is satisfied, is the first step toward realizing that the epistemic demands of Quality take a kind of priority over the demands of the other maxims. That is, Quality enjoys a special status from the other maxims in that the other maxims plausibly do not operate unless Quality is itself assumed to be satisfied. I assume that by a norm's being "satisfied", Grice does not mean to rule out cases where one conversationally implicates by flouting a maxim, for such implicatures depend for working them out that the other maxims are operative. To capture this notion, I shall instead talk of a maxim being respected when the maxim is operative and unviolated except by flouting/exploiting; in this sense one still respects a maxim when exploiting it to conversationally implicate, and one also trivially respects it when in engaging in other speech acts, such as asking a question. This concern about priority generates a reservation about the whole framework which sets up the maxims as dependent on the Cooperative Principle. Quality in particular looks to be constitutive of a certain kind of conversational contribution being what Grice calls a "genuine" contribution (as opposed to a "spurious" one17) in the first place: (1) The maxims do not seem to be coordinate. The maxim of Quality, enjoining the provision of contributions which are genuine rather than spurious (truthful rather than mendacious), does not seem to be just one among a number of recipes for producing contributions; it seems rather to spell out the difference between something's being, and (strictly speaking) failing to be, any kind of contribution at all. False information is not an inferior kind of information; it just is not information. (1989, 371) Taking his last line literally might raise a worry about how, if false claims fail even to be conversational contributions, they would fall under the purview of the Cooperative Principle, which is about contributions (for if they are not even contributions, they can neither violate nor conform to that Principle). But we needn't take this line literally, or follow Grice's assumption that false information is not information, to appreciate his point: by making as if to contribute in accord with Quality, when one is in fact not (strictly speaking) thus contributing, one is thereby not being cooperative. So the condition that helps us discern whether one is being cooperative in this sense is that given by Quality. Thus Quality is seen to be conceptually prior to the Cooperative Principle: if a (non-figurative18) contribution fails Quality, it ipso facto fails the Cooperative Principle by not being a cooperative contribution. We now have the makings of an Independence Argument for why (in non-figurative speech) the maxim of Quality is not dependent on the Cooperative Principle: Juror Martin: The defendant was innocent. Jane: What makes you think that? Martin: Oh, I don't believe that. Jane: Huh? Martin: But he was innocent.19 Notice that the conclusion of the Independence Argument is not committed to all false assertions being uncooperative conversational contributions. What it claims is that the Cooperative Principle cannot be operative (that is, available to the audience for working out what the speaker meant to implicate) unless Quality is presumed to be respected. However, one might worry that there are cases in which a speaker can assert falsely, and thereby violate Quality such that it is clearly not respected, but nevertheless still be making a contribution which depends on the operation of the other maxims for the audience being able to work it out. Wouldn't such a case cast doubt on Premise 1 of the Independence Argument? Teaching Suppose that Jenny is teaching a grade 10 science class. She wants to explain the structure of an atom and, more specifically, the electron configuration of different elements. Jenny is well aware that an early model of the electron structure of atoms, the Bohr model, is no longer considered accurate. Under the Bohr model, electrons travel in restricted orbits … [and are often depicted] as planets orbiting the nucleus of an atom. … More recently, though, the Bohr model has been replaced with the valence model. Under the valence model, due to incorporating principles of quantum mechanics such as Heisenberg's Uncertainty Principle, electron "orbits" are replaced with probability "clouds." Electrons are not restricted to rigid orbitals … So knowing all of this, Jenny also knows that her students aren't yet able to understand the valence model, but they are able to understand the Bohr model … and need to learn concepts such as the Bohr model as a stepping-stone. So when it's time to teach her students about the electron structure of atoms, she asserts, "Electrons behave according to the Bohr model." First, one might suspect that Jenny's utterance of (J) is actually not an assertion, or that she needn't have asserted (J) but instead could have tokened some other speech act to accomplish what she intends in the case. Against the idea that Jenny simply must assert (J) because (J) is the best assertive act she could make,22 it seems to me Jenny could make a number of other utterances, including some assertions, which are true, such as "We are now thinking of electrons as behaving as on the Bohr model. On this model, electrons behave thus… ." Similarly, Jenny might have, as is typical for a teacher, explicitly set up the didactic context using an imperative speech act calling for supposition: "Suppose that Bohr's theory holds: electrons behave according to Bohr's model. On this model… ." Having done this, all subsequent indicative claims in that context would be evaluated according to the supposition scenario, namely Bohr's model. But the context of supposition means that such utterances are not interpretable as flat-out assertions of Jenny's, just as the conditional statement If p, then q, is not an assertion that q, but at best a claim that q on the presupposition that p,23 in which case Jenny's real (conditional) claim seems true rather than false. Thus it's not at all clear that Jenny is, or needs to be, engaging in the speech act of asserting the proposition expressed by (J), for that proposition can be competently raised in didactic fashion without asserting (J) outright. But even if we accept (as the case stipulates) that Jenny is asserting this, it does not seem correct to say that Jenny is doing the best she could by asserting the false (J), for there are many other alternatives, even given her pedagogical aims, that she could have more appropriately asserted instead. Secondly, one might suspect that Jenny's assertion (J) is warranted in a sense which is not epistemic: it is pedagogically warranted, or maybe practically warranted, because it enables her students to learn and eventually gain knowledge or well-supported beliefs later in the learning process. But its being warranted in that sense does not tend to shed any light on whether the assertion is epistemically warranted along the dimension of normativity given by the norm of assertion or the maxim of Quality;24 the worry here then is that our judgment of Jenny's (J) being warranted (in the practical or pedagogical sense) is unduly affecting one's judgment of it as being epistemically warranted. This concludes my reservations about the Teaching case. Thus Premise 1, suitably cleaned up, should clarify that it is the audience, not the speaker, that must presume that Quality is respected, for it is the audience that will have to deploy the other maxims, or the Cooperative Principle, to make sense of what the speaker intends to communicate. Here then is the revised argument: Independence Argument* In Teaching, it is the students who are presuming that Jenny is respecting the maxim of Quality, and this is arguably crucial to their diachronic learning process: if the students didn't take their teacher to be asserting truthfully, and with good evidence in support of her claims, their learning process would be undermined.25 Presuming that their teacher is respecting the maxim of Quality in this context means that, unless the students have reason to believe Jenny is flouting or exploiting Quality so as to conversationally implicate some other proposition, Jenny's assertion of (J) puts forth the proposition (J) as true, worthy of their belief, available from which to draw inferences, and so on; this contributes to why it is the students learn the concepts of the Bohr model. Thus we can see that, while Teaching forced us to modify Premise 1 into Premise 1*, Teaching is not a counterexample to the latter premise; in fact, Premise 1* helps explain why Jenny's (J), though false, can be a cooperative conversational contribution.26 I do not take the Independence Argument* to show that one must think of the maxim of Quality as constitutive, in Williamson's specific sense,27 of the speech act of assertion (though it suggests at least a strict duty as discussed in §1.). I do, however, take it to put pressure on the dissenter to the Knowledge Account who wants to appeal to Grice as an ally. Such a dissenter cannot simply invoke contextually variable norms which are supposed to be explained, and justified, by their dependence upon something as general as the Cooperative Principle. For such a dissenter will need to provide compelling reasons for denying Premise 1* or 2 of the Independence Argument*, while also giving an alternative and unified explanation of all the conversational data which the Knowledge Account well-explains.28 Though some have thought that accepting Grice's communicative framework provides the resources for resisting the Knowledge Account of Assertion, in fact, embracing Grice's framework makes resisting the Knowledge Account far more difficult than it seems. This is because, first, Gricean Quality is best thought of in terms of knowledge. Second, on Grice's view the maxim of Quality is distinctive from the other conversational maxims, in ways that support rather than impede

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 51 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.