Abstract
Saul Kripke (Philosophical troubles, 2011) argued that the requirement that knowledge eliminate all possibilities of error leads to dogmatism (i.e., the view that, if one knows that p, then one may rationally decide now to disregard any future evidence against p one may encounter). According to this view, the dogmatism puzzle arises because of a requirement on knowledge that is too strong. The paper argues that dogmatism can be avoided even if we hold on to the strong requirement on knowledge. I show how the argument for dogmatism can be blocked and I argue that the only other approach to the puzzle in the literature is mistaken.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Harman (1973).
Kripke (2011).
Kripke (2011, p. 43) has the views of knowledge defended by Malcolm (1952) and Hintikka (1962) in mind here. In particular, Kripke considers a long passage from Malcolm (1952, pp. 185–186) in which Malcolm takes himself to know that there is an ink-bottle in front of him and wonders how he should respond to counterevidence suggesting he does not. Malcolm says his knowledge that there is an ink-bottle in front of him exemplifies a strong sense of “know,” a sense that denotes the kind of knowledge the knower might properly hold on to when faced with counterevidence.
cf. Harman (1973, p. 149). Throughout, “x is counterevidence to y” should be understood as synonymous with “x is evidence against y.”
Sharon and Spectre (2010) make a similar point.
Referee #3 raised two excellent objections to this argument. First, according to the referee the assumption that the entailment must be recognized before it can justify the discounting of counterevidence requires argument, for the truth of p may be sufficient for warranting the belief that evidence against p is misleading. The referee is right that an argument is needed here, and I offer one in Sect. 3 (roughly, there being a reason, r, to \(\phi \) is not enough to justify one’s \(\phi \)-ing—r needs to be one’s reason for \(\phi \)-ing in order for \(\phi \)-ing to be justified for one). On the other hand, it is not the case that an argument is needed because the truth of p is sufficient to warrant the belief that evidence against p is misleading. Consider: I wonder if the number of stars is even or odd; I flip a coin and on that basis arrive at the true belief that the number of stars is odd. I doubt anyone would say that the truth of this unjustified belief is sufficient to warrant me in disregarding rebutting evidence as misleading (e.g., rebutting evidence coming from astronomy). What’s more, Harman and Kripke were sensitive to the issue of unjustified or luckily acquired true belief and couched the problem in terms of knowledge rather than truth, and these are the versions of the argument I am interested in here. As I showed in footnote 3, for Kripke (and for Harman who followed him in this respect), the dogmatist paradox arises for certain analyses of knowledge (such as Hintikka’s) that take knowledge to exclude the possibility of error. (Kripke’s paper discussing the argument for dogmatism is, after all, called “Two Paradoxes of Knowledge.”) Second, the referee claims that Harman did take the relevant entailment to be recognized. In support of this point the reviewer quotes the passage where Harman (1973, p. 148) introduces the paradox: “If I know that h is true, I know that any evidence against h is evidence against something that is true; so I know that such evidence is misleading. But I should disregard evidence that I know is misleading. So, once I know that h is true, I am in a position simply to disregard any future evidence even though I do know a great many different things.” But in this passage Harman does not say that he is aware of or knows the relevant entailment—he simply states that his knowledge of the misleadingness of evidence is entailed by his knowledge that h. However, he does think this is enough to give him knowledge of the misleadingness of evidence against h. Harman is appealing to the following version of knowledge closure: If S knows that p and p entails q, then S knows that q. But, this version of closure is compatible with S coming to know the entailed proposition on the basis of something other than her knowledge of the entailment. In this sense, Harman’s presentation of the problem is importantly different from Kripke’s. Kripke presents his argument with a version of knowledge closure that appeals explicitly to the subject’s knowledge of the relevant entailment: “If A knows that p and A knows that p entails q, and, on the basis of such knowledge, A concludes that q, then A knows that q” [Kripke (2011, p. 43). See footnote 12 below as well].
From now on “dogmatism” will refer to synchronic dogmatism. As Kripke (2011, p. 49) makes clear, he is interested in this version of the puzzle.
Kripke (2011, p. 44) makes similar distinctions.
I am in debt to referee #3 for discussion on issues in this section.
Kripke (2011, p. 43).
For the most part I follow Kripke’s formulation of the steps in the argument. See Kripke (2011, pp. 43–44).
Kripke offers the following gloss of closure: “If A knows that p and A knows that p entails q, and, on the basis of such knowledge, A concludes that q, then A knows that q” (cf. Kripke 2011, p. 43). Like Kripke, I am assuming that the subject arrives at 4 on the basis of her knowledge of 1–3; thereby satisfying the antecedent of closure. I come back to the role of closure in the dogmatist’s argument below.
Sharon and Spectre (2010).
Sharon and Spectre (2010, p. 320).
Sharon and Spectre (2010, p. 309).
See Hawthorne (2005) for some of the costs of denying SPC. Also, even if knowledge-closure is not valid (i.e., even if it has some false instances), it does not follow that all instantiations of the principle are false. Hence, knowledge-closure deniers owe us a story about what makes some instantiations of the principle true and others false. Nozick (1981) tells us that the true instantiations preserve tracking while the false ones do not. For a recent, alternative story, see de Almeida (2011).
cf. Klein (1995).
The case is adapted from Dretske (2005, pp. 14–15).
But see de Almeida (2011) for purported counterexamples to SPC that take into account the distinction between TES and SPC. deAlmeida is careful enough to rest his case against SPC on examples that do not feature a failure of TES. I am not completely convinced that the cases he discusses show that SPC is false, but, even if he is right, my case against Sharon and Spectre still stands, for it does not claim that SPC is true but, rather, that their diagnostic of the dogmatism paradox as a problem for SPC is less than conclusive because it can be plausibly seen as a problem for TES instead.
Thanks to referee #3 for prompting me to be more explicit about this.
The arrows represent the “x is a source of positive epistemic status for y” relation.
Speaking of ravens, SPC, and TES, it seems that the Raven Paradox introduced by Carl Hempel also provides an example in which TES fails while SPC holds. Assume: (e) a list of the positive instances of Ravens that are black; (p) All Ravens are black; (q) All non-black things are non-Ravens. It seems that e is a source of positive epistemic status for p, p is a source of positive epistemic status for q, but e does not seem to be a source of positive epistemic status for q. This amounts to a failure of TES. Plausibly, SPC still holds here, however, for p is (intuitively) itself a source of positive epistemic status for q. I am grateful to reviewer #3 for giving me the opportunity to expand on the differences between TES and SPC. Thanks also to Peter Klein for discussion here.
This example is originally from Sorensen (1998).
Sharon and Spectre (2010, p. 320).
Even if I am wrong about all of this, Sharon and Spectre’s reply to dogmatism does not generalize to include cases of dogmatism that do not explicitly rely on TES or SPC. For example, their strategy does not apply to a case discussed by Hawthorne (2004, p. 181) where a subject is offered a “dogmatist pill” capable of causing her to disregard all counterevidence to what she knows. In contrast to their view, the view I defend in the next section can account for cases such as Hawthorne’s. I have benefited from discussion with referees #1 and #3 here.
As I show in Borges (2013), the same considerations I will present below apply to the diachronic version of the puzzle as well. The idea is that if it is not rational for me to be dogmatic about my knowledge that p now, when I have no knowledge of counterevidence against p, how could it be rational for me to be dogmatic about my knowledge that p later, when I am presented with counterevidence for p?
Pollock and Cruz (1999, pp. 196–197).
Referee #1 raised an important issue concerning this case. He/She argued that this case of inductive knowledge might be in tension with the view the dogmatist adopts (and that we are assuming in this work, for the sake of argument). If one knows that p only if one’s evidence makes it impossible that \(\lnot \)p, then one knows that p only if one’s evidence entails that p. Thus, the view seems to rule out the possibility of inductive knowledge. This is not the time or the place to discuss this worry, however. My goal here is to remove the dogmatist roadblock to the infallibilist view. My goal is not to remove all roadblocks to this view. That would require a much longer work and, honestly, I am not sure infallibilism can remove all of them. However, this is an exciting topic, and I hope to come back to it in future work of mine. I am grateful to referee #1 for raising this issue.
Thanks to referee #2 for urging me to clarify this case.
Although Kripke (2011, p. 44) seems to appreciate at least one of the points I am making here (i.e., that misleading evidence might lead to suspension of judgment rather than to a false belief), the argument suffers from problems that are deeper than the ones emerging from Kripke’s casual presentation. The dogmatism puzzle emerges, to a great extent, because of the naive view of evidence the argument conceals. As I will argue below, this naive view neglects the fact that r justifies one in \(\phi \)-ing only if one knows r. The genius of Kripke was to highlight the epistemological damage caused by a naive view of evidence if we couple this view with an infallibilist account of knowledge such as Hintikka’s. I thank referee #1 for prompting me to address this issue here.
See footnote 3.
I owe this example to Ernie Sosa.
Although this is the most charitable reading of premise 5, from a purely exegetical point of view it is most likely not the reading Kripke himself had in mind. Kripke (2011, p. 44) explicitly denies that one could maintain the resolution to avoid places, people, and things potentially containing misleading evidence. According to him, it should be possible for us to “ignore” misleading evidence, regardless of whether we wanted to be confronted with it or not. This suggestion also betrays what Kripke takes to be the motivating thought behind his argument for dogmatism: the thought, defended by Malcolm (1952, pp. 185–186) and Hintikka (1962, pp. 20–21), that there is a sense in which knowledge is conclusive [see Kripke (2011, p. 39)].
One might object that premise five fails to comply with the injunction that epistemic “ought” implies epistemic “can” (Wedgwood 2013): one does not have direct control over one’s doxastic states and, thus, cannot satisfy the intention of not forming a false belief. But, if whether one ought to form an intention depends on whether one can satisfy that intention, then it seems that, contrary to what premise 5 says, one ought not to form that intention. I am sure there are things the dogmatist can say here, but I will not pursue this line of inquiry any further. The paper is rather long as it is. I am in debt to referee #1 here.
Thanks to Duncan Pritchard for discussion here.
This is in line with the knowledge-first approach to practical reasons defended by Jason Stanley and John Hawthorne in Hawthorne and Stanley (2008). Also, if Timothy Williamson’s E=K account of evidence is correct (cf. Williamson 2000), then r is a justifying reason for believing only if r is known.
Referee #3 raised the following challenge. I am thankful for the opportunity to discuss it here. The claim that knowledge of p is not sufficient to warrant disregarding of misleading evidence is mistaken: p and SPC are sufficient to warrant one in taking counterevidence against p to be misleading. However, despite its initial plausibility this is not correct. As I emphasized in footnote 6, not only did Kripke and Harman explicitly state the dogmatist argument using the premise that one knows that p, it is easy to cook-up cases where truth is clearly not sufficient to warrant disregarding of counterevidence. I gave one such example in footnote 6. Also, the motivation for the dogmatist position is a desire to prevent the loss of knowledge, not the loss of true belief.
A comment from referee #1 helped me improve this passage significantly.
The case I am about to present is a modified version of the Assassination Case discussed in Harman (1973, pp. 143–144).
Thanks to John Hawthorne and Peter Klein for pressing this point in conversation.
References
de Almeida, C. (2011). Epistemic closure, skepticism and defeasibility. Synthese, 188, 197–215.
Borges R (2013). Dogmatism depuzzled. Manuscript.
Dretske, F. (1970). Epistemic operators. The Journal of Philosophy, 67(24), 1007–1023.
Dretske, F. (2005). Is knowledge closed under known entailment? The case against closure. In M. Steup & E. Sosa (Eds.), Contemporary debates in epistemology. Oxford: Blackwell.
Harman, G. (1973). Thought. Princeton: Princeton University Press.
Hawthorne, J. (2004). Knowledge and lotteries. Oxford: Oxford University Press.
Hawthorne, J. (2005). The case for closure. In M. Steup & E. Sosa (Eds.), Contemporary debates in epistemology. Oxford: Blackwell.
Hawthorne, J., & Stanley, J. (2008). Knowledge and action. Journal of Philosophy, 105(10), 571–590.
Hintikka, J. (1962). Knowledge and belief. Ithaca: Cornell University Press.
Huemer, M. (2000). Direct realism and the brain-in-a-vat argument. Philosophy and Phenomenological Research, 2(61), 397–413.
Klein, P. D. (1981). Certainty: A refutation of scepticism. Minneapolis: University of Minnesota Press.
Klein, P. D. (1995). Skepticism and closure: Why the evil genius argument fails. Philosophical Topics, 23, 212–236.
Kripke, S. (2011). Philosophical troubles. Oxford: Oxford University Press.
Malcolm, N. (1952). Knowledge and delief. Mind, 61(242), 178–189.
Nozick, R. (1981). Philosophical explanations. Cambridge: Belknap Press of Harvard University Press.
Pollock, J., & Cruz, J. (1999). Contemporary theories of knowledge. Lanham: Rowman and Littlefield.
Sharon, A., & Spectre, L. (2010). Dogmatism repuzzled. Philosophical Studies, 148, 307–321.
Sorensen, R. (1998). Blindspots. Oxford: Oxford University Press.
Wedgwood, R. (2013). Rational “ought” implies “can”. Philosophical Issues, 23(1), 70–92.
Williamson, T. (2000). Knowledge and its limits. Oxford: Oxford University Press.
Wright, C. (2002). (Anti-)sceptics simple and subtle: G.E. Moore and John McDowell. Philosophy and Phenomenological Research, 65, 330–348.
Acknowledgments
Different drafts of this paper benefited from the generous feedback of many people. Peter Klein read all of them and his help and support were invaluable. Duncan Pritchard, Claudio de Almeida and Ernest Sosa also commented on different drafts of this paper. I am very grateful for their help and advice. I must also thank John Hawthorne, Roy Sorensen, and Mike Veber for discussing with me an earlier draft of the paper. The paper also benefited from conversations with Matt Benton, David Black, Guido Melchior, Carlotta Pavese, Luis Rosa, Daniel Rubio, and Kurt Sylvan. The research for this paper was partially funded by the CAPES/Fulbright Commission. I am grateful for their support.
Conflict of interest
The author declares that he has no conflict of interest.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Borges, R. On synchronic dogmatism. Synthese 192, 3677–3693 (2015). https://doi.org/10.1007/s11229-015-0715-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-015-0715-3