One of the central arguments for the knowledge norm of assertion appeals to the fact that it is typically legitimate to respond to an assertion with “How do you know?”, intended as a challenge to the assertion. The legitimacy of the challenge is taken as evidence in favor of the idea that permissible assertion requires knowledge. In this paper, I argue that if the legitimacy of “How do you know?” challenges supports a knowledge norm for assertion, it also supports the controversial KK thesis. I further argue that data from assertion prompts likewise supports KK. This is an unwelcome consequence for many proponents of the knowledge norm of assertion who reject KK. At the same time, it constitutes new evidence from conversational patterns in favor of the KK thesis.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
Here and elsewhere in the paper, ‘may’ and ‘permissible’ are meant to denote epistemic permissibility.
In response to this argument, one may raise the following worry: Don’t we often just mean “What are your reasons for thinking that?” when we ask someone how they know? If so, the “How do you know?” data may not, after all, support KNA. See Douven (2006), Lackey (2007), Kvanvig (2009) for versions of this objection, and see Williamson (2009), Benton (2011) for replies. I’m assuming throughout this paper that the “How do you know?” argument for KNA is sound, so I will not discuss this objection. However, I say more about how this idea is relevant to the present argument towards the end of Sect. 2.3.
Recent linguistic considerations in favor of KK include Dorst (2019), Haziza (forthcoming).
A standard distinction is between speaker and sentence presupposition. Sentence presupposition is, roughly, presupposition that is triggered by the semantic constituents of the sentence in a given linguistic context. A speaker presupposes p, I will assume, when (but perhaps not only when) the speaker uses a sentence that presupposes p. APK is about speaker presupposition. I say more about the relevant notion of speaker presupposition in Sect. 3. Thanks to an anonymous referee for pointing out the need to clarify this distinction.
I’m grateful to an anonymous referee for raising this concern.
Benton (2011), Benton (forthcoming) takes the incoherence of such exchanges to support KNA.
Greco (2015) argues that KK is required to explain the infelicity of the answer in such cases. He writes that, on the view that accepts KNA but rejects KK, “while [speakers who know P without knowing that they know] will be able to permissibly assert that P, if their permission to assert that P is challenged, they will not be able to permissibly defend themselves.” Although I am sympathetic to Greco’s view, I think this is mistaken, since speakers in such cases can defend themselves without asserting knowledge, e.g., by providing reasons for P. Cases like (1) do not support KK on their own.
Is (P2) is really necessary for the main argument of this section? It is, because without (P2) the conclusion (C2) becomes: One may assert p only if one asserts or presupposes Kp, instead of: One may assert p only if one may assert or presuppose Kp. That is, the speaker’s assertion or presupposition of Kp must itself be permissible. Only the latter can get us to KK.
Williamson (2000, p. 246).
This is somewhat simplified, but that’s the general idea. For a discussion of the semantics of how-questions, see Jaworski (2009) and Sæbø (2016). One may worry that the semantics of how-questions is not understood well enough to make such assumptions about it. But all I am assuming here about semantics is that “How p?” presupposes p, and nothing beyond that about its set of possible answers. This assumption is part of the “How do you know?” argument for KNA made by Williamson (2000), Williamson (2009) and others. As for the pragmatics of how-questions, I’m assuming that they are largely similar to other types of questions. In particular, that they can be answered (either fully or partially), declined, or have their presuppositions resisted. Thanks to an anonymous referee for raising these points.
I’m grateful to both referees for raising this point.
Thanks to one of the referees for examples on which these are based.
Recall that an acceptable answer to the challenge is one that would ordinarily tend to get the original assertion accepted.
It may not be possible to felicitously add “How do I know?” in this way to retracting or weakening answers as in (13) and (14), but those are not acceptable answers.
See Camp (2012) for an analysis of this example along these lines.
Cases of sarcasm and challenge questions are not the only ones in which we find this kind of pretense. For another example, consider a therapist who knows that the friend that her patient keeps talking about is not real. She might ask: “How is Steve this week?”, knowing that the presupposition that Steve exists is false. This too is a case of pretense: the therapist pretends that Steve exists in asking this question. Thanks to an anonymous referee for this example.
‘Kp’ denotes that the agent in question knows p.
To be clear, the logical form of APK is meant to be: one may assert p \(\rightarrow \) (one may presuppose Kp \(\vee \) one may assert Kp). Given that \(p \rightarrow (q \vee r), q \rightarrow r \vdash p \rightarrow r\), APK and PA entail AAK.
DeRose (2002, Sect. 2.3) considers AAK and rejects it on the basis of KNA and the denial of KK. He concedes, however, that “it’s hard to come up with clear counter-examples to [AAK]” (p. 185).
To see this, let ‘Ap’ mean ‘one may assert p’. Note first that KNA (\(Ap \rightarrow Kp\)) and AAK (\(Ap \rightarrow AKp\)) entail \(Ap \rightarrow KKp\). This and AAK jointly entail \(Ap \rightarrow KKKp\). This and AAK jointly entail \(Ap \rightarrow KKKKp\). This can be repeated indefinitely, yielding KNA*.
This argument also goes through with weaker versions of KNA-Suff, e.g., that third-order knowledge is sufficient for permissible assertion, which would be harder to deny. Given that KK failures, if they occur at all, can also occur at that level, KK again follows from KNA*.
Thanks to an anonymous referee for discussion on these points.
I’m grateful to two anonymous referees at this journal for comments which helped to improve this paper. A distant ancestor of this paper received helpful comments from Benj Hellie and participants in his 2019 seminar Knowledge Never.
Benton, M. A. (2011). Two more for the knowledge account of assertion. Analysis, 71(4), 684–687.
Benton, M. A. (2016). Expert opinion and second-hand knowledge. Philosophy and Phenomenological Research, 92(2), 492–508.
Benton, M. A. (Forthcoming). Knowledge is the norm of assertion. In E. Sosa, M. Steup, J. Turri, & B. Roeber (Eds.), Contemporary debates in epistemology (3rd ed.). Wiley.
Brown, J. (2010). Knowledge and assertion. Philosophy and Phenomenological Research, 81(3), 549–566.
Camp, E. (2012). Sarcasm, pretense, and the semantics/pragmatics distinction. Noûs, 46(4), 587–634.
Cross, C., & Roelofsen, F. (2020). Questions. In N. E. Zalta (Ed.), The Stanford encyclopedia of philosophy, fall 2020. Metaphysics Research Lab, Stanford University.
DeRose, K. (2002). Assertion, knowledge, and context. Philosophical Review, 111(2), 167–203.
Dorst, K. (2019). Abominable KK failures. Mind, 128(512), 1227–1259.
Douven, I. (2006). Assertion, knowledge, and rational credibility. Philosophical Review, 115(4), 449–485.
García-Carpintero, M. (2020). On the nature of presupposition: A normative speech act account. Erkenntnis, 85(2), 269–293.
Greco, D. (2015). Iteration and fragmentation. Philosophy and Phenomenological Research, 91(3), 656–673.
Hawthorne, J. (2012). Some comments on Fricker’s stating and insinuating. Aristotelian Society Supplementary, 86(1), 95–108.
Haziza, E. (Forthcoming). Assertion, implicature, and iterated knowledge. Ergo: An Open Access Journal of Philosophy.
Jaworski, W. (2009). The logic of how-questions. Synthese, 166(1), 133–155.
Johan, S. K. (2016). “How’’ questions and the manner-method distinction. Synthese, 193(10), 3169–3194.
Kelp, C., & Simion, M. (2017). Criticism and blame in action and assertion. Journal of Philosophy, 114(2), 76–93.
Kvanvig, J. (2009). Assertion, knowledge, and lotteries. In D. Pritchard & P. Greenough (Eds.), Williamson on knowledge (pp. 140–160). Oxford University Press.
Lackey, J. (2007). Norms of assertion. Noûs, 41(4), 594–626.
Lackey, Jennifer. (2011). Assertion and isolated second-hand knowledge. In Jessica Brown & Herman Cappelen (Eds.), Assertion: New philosophical essays (pp. 251–276). Oxford University Press.
Lackey, J. (2016). Assertion and expertise. Philosophy and Phenomenological Research, 92(2), 509–517.
Mandelkern, M., & Dorst, K. (Forthcoming). Assertion is weak. Philosophers’ Imprint.
McKinnon, R. (2012). How do you know that ‘how do you know?’ challenges a speaker’s knowledge? Pacific Philosophical Quarterly, 93(1), 65–83.
Simion, M. (2016). Assertion: Knowledge is enough. Synthese, 193(10), 3041–3056.
Turri, J. (2010). Prompting challenges. Analysis, 70(3), 456–462.
Unger, P. (1975). Ignorance: A case for scepticism. Oxford University Press.
Williamson, T. (2000). Knowledge and its limits. Oxford University Press.
Williamson, T. (2009). Replies to critics. In D. Pritchard & P. Greenough (Eds.), Williamson on knowledge (pp. 279–384). Oxford University Press.
Wiśniewski, A. (2015). Semantics of questions. The handbook of contemporary semantic theory (pp. 271–313). Wiley.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.