Abstract
Can new evidence against what one knows defeat one’s knowledge? It depends on whom you ask. Strong views of knowledge claim that knowing cannot be defeated by counterevidence. Weak views of knowledge claim that knowing can be defeated by counterevidence. I discuss recent versions of those views of knowledge—Maria Lasonen-Aarnio’s and Peter Klein’s, respectively—and argue that both are wanting. Strong views have to account for the defeat intuition (i.e., the inclination some of us have, at least some of the time, to think that knowledge is defeated in particular cases). Weak views have to solve the indeterminacy problem (i.e., explaining, in a principled way, when and why knowledge is defeated). I argue that Lasonen-Aarnio’s view cannot account for the defeat intuition, and that Klein’s view cannot account for the indeterminacy problem. I then offer a novel version of the strong view of knowledge, one that improves on the version previously discussed.
I wrote this chapter to honor Peter Klein , the best philosopher I have ever met, and one of the best people I have ever met. Peter was my thesis advisor at Rutgers; he is also a mentor, and the best friend my work has ever had. Peter made me believe I could (sometimes) do philosophy, and that philosophy was something worth doing. During the eight years I was at Rutgers, Peter and I met once or twice every month during the academic year to discuss my work. We discussed every issue in this chapter (and more, much more) innumerable times over that period. We didn’t always agree, but Peter always taught me something valuable. I cherish those lessons, and I will be forever grateful to Peter for having given them to me
This chapter benefıted from the feedback of many people. Some of them read different drafts and were kind enough to send me comments. Among those, I am especially thankful to Peter Klein, Claudio de Almeida, Joao Fett, Felipe Medeiros and Gregory Gaboardi. Others let me pick their brains in conversations about some of the issues in the chapter. Those include Mike Veber, Nicola Salvatore, Fred Adams, and Rogel de Oliveira. Many audience members asked really smart and insightful questions about the presentations based on different parts of this chapter. I am especially thankful to the audiences at the following events: a meeting of the East Carolina University Philosophy Club, the 1st Colloquium on Contemporary Debates in Epistemology , the XIII Epistemology Colloquium, and the XVII Meeting of the National Association of Graduate Programs in Philosophy. I also gave a version of this paper during a visit to East Carolina University in January 2018. I am grateful especially to the faculty in the audience: George Bailey, Josh Collins, Nicholas Georgalis, Jay Newhard, and my host Mike Veber were particularly generous with their comments. I am grateful for their help. Part of the research for this chapter was funded by the Sao Paulo Research Foundation through a post-doctoral research grant. I am grateful for their support. Finally, many thanks to the two other editors for this volume, Cherie Braden and Branden Fitelson, for their support and patience as I failed to meet (almost) all the deadlines for this chapter. Cherie also sent me detailed written comments on a previous draft and saved me from making many mistakes. I am grateful for her help making this chapter better.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Plato’s discussion of the difference in value between knowledge and true belief famously relates knowledge and true belief to Daedalus statues, which, if not tethered, were said to run out and escape their base. A true belief, argued Plato, is like an untethered Daedalus statue—beautiful but prone to escaping its owner. An instance of knowledge, on the other hand, is like a Daedalus statue that is tethered—beautiful and not likely to escape its owner.
- 2.
I am aware of the pitfalls of trying to translate ‘aitias logismos,’ but I will not discuss them here. For an excellent discussion of some of the main exegetical issues surrounding the translation this expression, see Fine (2004).
- 3.
For example, in Republic (534b), Plato seems to argue that knowledge is immune to counterevidence , while mere true belief is not, because the latter, but not the former, may be essentially based on false beliefs. Because mere true beliefs can depend essentially on false beliefs, they may be undermined by new truths; this is not the case with knowledge, which depends exclusively on what is true. See Fine (2004, §viii) for a discussion of this point.
- 4.
Malcolm is making another distinction with his case. Although he is primarily interested in the synchronic features of the case (i.e., he is interested in what it is now rational for him to think about possible counterevidence to his knowledge that there is an ink-bottle in front of him), he acknowledges that one could also take an interest in the diachronic features of the case (i.e., one could be interested in what Malcolm would do, in the future, if he were confronted with evidence suggesting there was no ink-bottle in front of him). Malcolm says that he is not talking about the diachronic features of his situation, and he concedes that he cannot confıdently predict now how he would behave in the future were he to encounter evidence suggesting that there is no ink-bottle in front of him. For all he knows, says Malcolm, he would ‘become mad’ or ‘fall into a swoon’ if someone he trusts were to say that ink-bottle hallucinations are quite common for people in his circumstances. See Kripke (2011) and Borges (2015) for a discussion of how heeding this distinction may prevent one from misunderstanding the charge of dogmatism that is usually leveled against strong views of knowledge such as the one discussed by Malcolm.
- 5.
I don’t mean this case to presuppose that headaches are luminous states (i.e., states that are such that, if one is in one of them, then one is in a position to know one is in them). Maybe headaches are luminous; maybe they aren’t. The claim that they are is irrelevant for my case. Clearly, we are sometimes in a position to know we have a headache; I mean this case to be about one of those times. I mention this because Timothy Williamson (2000) has persuasively argued that there are no luminous states. His argument does not preclude the claim that one is sometimes in a position to know one is in a certain mental state, though. But see Greenough and Pritchard (2009) for criticism of Williamson’s argument against luminosity.
- 6.
Of course, I am also supposing that there is no funny business of the Gettier type going on here. The same is true of all the other cases in this chapter.
- 7.
For example, Jaakko Hintikka (1962, 20), claims that ‘whoever says “I know that p” proposes to disregard the possibility that further information would lead him to deny that p.’ Peter Klein (2017), and Roy Sorensen (2012), also endorse the view that knowledge is defeasible. I will discuss Peter Klein’s views below.
- 8.
Gilbert Harman (1973, 192) does just that in the following passage:
That undermining evidence … is relevant … to the acquisition of knowledge [and] also to its maintenance is clear from the … example concerning Tom and the library detective. The detective knows that he saw Tom steal the book and so he testifıes to the Judicial Council. After he leaves the hearing, Tom’s mother fabricates her story about Tom’s twin brother. Her lying testimony convinces the Judicial Council but is unknown to the detective back at his post in the library. [O]nce Tom’s mother has… testifıed, it is no longer true that the detective knows that Tom stole the book.
- 9.
Marbles is a variation of a case discussed by Williamson (2000, 205).
- 10.
N.B., I am not making a claim about exactly what impact each draw has on her justifıcation; the impact of each draw is plausibly really small. The point, rather, is that the sum of all one thousand draws intuitively does have a major impact on her justifıcation. Thanks to Cherie Braden for discussion here.
- 11.
Answer: small—0.51000.
- 12.
Be that as it may, all cases discussed in this chapter deal with categorical, rather than partial, defeat. That is, with respect to all cases discussed in this chapter, I am supposing that the potential defeating effect of counterevidence is always such that it lowers the subject’s degree of justifıcation below the threshold required for knowledge.
- 13.
There is another, less obvious, potential cost of accepting (C1). That is the danger of dogmatism and the rejection of an intuitive notion of intellectual humility : if knowledge is indefeasible, why should one pay attention to counterevidence to what one knows? Isn’t all counterevidence to what one knows misleading evidence? This problem for views that accept (C1) was put forward by Saul Kripke in a talk in the early 1970s. Harman (1973) made a similar problem popular. Kripke (2011) includes Kripke’s talk as well as an appendix explaining the difference between his version of the problem and Harman’s. Since I dealt with this issue in Borges (2015), I will not dis cuss it here again.
- 14.
For simplicity’s sake, in drawing this distinction I am presupposing that knowing that p entails believing that p. I think that even if one rejected this presupposition one could still draw the relevant distinction, although doing so would be more complicated. What I mean is that the distinction is compatible with the view, sometimes exposed by knowledge fırsters, that knowledge cannot be analyzed in terms of belief, truth, etc. The reason why this view is compatible with our distinction is that one may hold it and still think that whenever someone knows that something is the case that person also believes that it is the case; that is, one may consistently hold the view that knowledge cannot be analyzed in terms of belief while, at the same time, argue that knowing that p is always accompanied by believing that p (or, alternatively, that the latter state is always a by-product of the former state). If one had this kind of knowledge fırst view, one could consistently hold the distinction between what I will call below ‘psychological’ and ‘epistemic defeat.’ (Incidentally, Fine (2004) discusses a version of the knowledge fırst type of view I am sketching here and suggests one might attribute it to Plato.) I do not have the space to defend these claims here, however. I will come back to them some other day.
- 15.
- 16.
This is true not only of more traditional versions of the weak view of knowledge (i.e., weak views that take knowledge to be analyzed in part by justifıcation), but also of knowledge-fırst views such as the one in (Williamson 2000, Chapter 10), which says that knowledge is not analyzable but is nonetheless susceptible to epistemic defeat.
- 17.
- 18.
As far as I know, Peter also accepts (C2b), but I will not deal with that aspect of his view here.
- 19.
In Klein (2017, 54) Peter says that this latter truth ‘restores’ Ted’s justifıcation for believing that he has enough handouts. I fınd that confusing—if ‘It is not the case that there are 53 people in attendance’ is not really a defeater, then there is no damage to be restored by this truth. Instead, what seems to be going on here is that the destructive effect of ‘It is not the case that there are 53 people in attendance’ is preempted by Ted’s knowledge that there are 52 people in attendance.
- 20.
I have given a similar reply to Handouts myself in Borges (2017). See de Almeida (2017) for criticism of Peter’s way of handling cases of useful falsehoods . For alternative ways of dealing with Handouts as well as with other cases of knowledge from non-knowledge, see, among others, Ball and Blome-Tillmann (2014), Montminy (2014), Luzzi (2014), and Schnee (2015).
- 21.
It may be suggested that Peter should not have been so quick to assume that Ted is ‘epistemically close’ to what is obviously entailed by something he knows. After all, closure -deniers such as Fred Dretske and Robert Nozick can reasonably reject the claim about epistemic proximity Peter relies on. This is an important worry about Peter’s strategy, but I think it can be mitigated. I do not have space to give it a full treatment here; so, a couple of quick remarks will have to do—at least for now. The key point is this: the thing pushing the proposition beyond the subject’s epistemic reach in counterexamples to closure (e.g., Dretske’s zebra-in-the-zoo case) is not present in cases of useful falsehood , suggesting that closure does not fail in the latter cases (even if it fails in the former cases). According to Dretske, knowledge closure fails in his zebra-in-the-zoo case because, although the zoo-goer has evidence that is good enough to give him knowledge of ‘That’s a zebra’ (e.g., there’s a plaque saying that the animals in the pen are zebras), his evidence is not good enough to give him knowledge of something that is entailed by this truth, namely ‘That’s not a cleverly disguised mule.’ For Dretske (and, in related ways, for Nozick as well) knowledge closure fails in cases where the subject’s evidence, although good enough to produce knowledge of the entailing proposition, is not good enough to produce knowledge of the entailed one. That is the one thing pushing the entailed proposition beyond the subject’s epistemic reach. But Ted’s situation is different. The evidence he has for the falsehood ‘There are 53 people in attendance’ is also good enough to produce knowledge of the truth it entails (i.e., ‘There are approximately 53 in attendance’). Many thanks to Claudio de Almeida for discussion here.
- 22.
Peter used to call this problem the ‘problem of misleading defeaters’ but he has recently dropped this description of the problem, and for a good reason: since, according to Peter, misleading evidence against what one knows is just counterevidence that does not defeat knowledge; calling a misleading defeater a ‘defeater’ is just confusing, for defeaters, by defınition, do defeat and the so-called ‘misleading defeaters’ do not defeat (see Klein 2017).
- 23.
See, for example, Klein’s discussion of the Mr. Magic case in Klein (2004).
- 24.
This is a version of a case presented by Harman (1973, 143–144).
- 25.
Objection: It is vague whether Smith knows or not in this case, and Defeasibility does exactly what we want it to do—it explains clear cases of knowledge/ignorance. Reply: The relevant accusation (what I mean by ‘less informative than one would have hoped for’) is that we look for defeaters only after we decide whether there’s knowledge in a relevant case. This gets things backwards, explanatorily speaking. As in science, we want philosophical theories that predict what type of evidence we are likely to fınd (whether we will fınd knowledge in a case or not) and not theories that only account for evidence we already have (whether cases that we already know have knowledge/ignorance conform to the theory). My claim is that Defeasibility has poor predictive power. Consider. The claim ‘All emeralds are green’ tells me that, for any x, if x is not green, then x is not an emerald. This conditional exploits the analysans (so to speak) of ‘emerald’ and suggests a method for fınding emeralds (i.e., looking at the color of things and disregarding things that are not green). Similarly, ‘All knowledge is undefeated belief’ suggests that, for any x, if x is defeated (if there is a defeater), then x is not knowledge. This conditional exploits the analysans of ‘knowledge’ and suggests a method for fınding knowledge (i.e., looking for defeaters and disregarding beliefs for which there are defeaters). However, while I can look for non-green things even if I do not know whether there are emeralds present or not, I do not see how I can look for defeaters without fırst knowing whether there is knowledge present or not. This is not good. It might not ‘refute’ Defeasibility, but it’s a problem that should be addressed, I think. Also, it is not so clear that it is vague whether the subject knows or not in Newspaper. Nozick (1981, 177), for example, thought it was a virtue of his proposed defınition of knowledge that it explained ‘why we are reluctant to say [the subject in Newspaper] knows the truth.’ According to him, condition 4 in his defınition of knowledge (i.e., if p, then S would believe that p) predicts the right result in that case—that the subject fails to know. What this tells me is that Nozick did not think the subject knew, that he thought that his theory explained why this was the case and that this conformed with what he took to be people’s general inclination about the case. Harman (1973, 145) also argued that Smith did not know and that this was a problem for his account. Thanks to Joao Fett for discussion here.
- 26.
- 27.
In fact, one might even say that this way of looking at the indeterminacy problem makes Defeasibility circular, for whether condition (4) is satisfıed depends in part on whether the subject knows. In my view, this limitation of belief-fırst versions of the weak view of knowledge gives credence to a knowledge fırst version of the weak view: instead of explaining knowledge in terms of defeat (i.e., what explains why one knows, when one does, is the fact that one’s justifıcation for believing is undefeated), we should explain defeat in terms of knowledge (i.e., what explains why one’s justifıcation is undefeated is the fact that one knows). I only mention this here, however, as I do not have the space to elaborate on this approach to defeat. See Baker-Hytch and Benton (2015) for a similar criticism of reliabilist versions of weak views.
- 28.
- 29.
This is similar to some versions of epistemological disjunctivism. For one such view, see Pritchard (2016).
- 30.
Williamson (Forthcoming) and Littlejohn (2012) make a similar distinction.
References
Baker-Hytch, M., & Benton, M. A. (2015). Defeatism defeated. Philosophical Perspectives, 29(1), 40–66.
Ball, B., & Blome-Tillmann, M. (2014). Counter closure and knowledge despite falsehood. The Philosophical Quarterly, 64, 552–568.
Borges, R. (2015). On synchronic dogmatism. Synthese, 192(11), 3677–3693.
Borges, R. (2017). Inferential knowledge and the Gettier conjecture. In R. Borges, C. de Almeida, & P. D. Klein (Eds.), Explaining knowledge (pp. 273–291). Oxford: Oxford University Press.
Borges, R., de Almeida, C., & Klein, P. (Eds.). (2017). Explaining knowledge: New essays on the Gettier problem. Oxford: Oxford University Press.
De Almeida, C. (2017). Knowledge, benign falsehoods, and the Gettier problem. In R. Borges, C. de Almeida, & P. D. Klein (Eds.), Explaining knowledge (pp. 292–311). Oxford: Oxford University Press.
Dretske, F. (1971). Conclusive reasons. Australasian Journal of Philosophy, 49(1), 1–22.
Dretske, F. (1981). The pragmatic dimension of knowledge. Philosophical Studies, 40(3), 363–378.
Fine, G. (2004). Knowledge and true belief in the Meno. In D. Sedley (Ed.), Oxford studies in ancient philosophy (Vol. XXVII, pp. 41–81). Oxford: Oxford University Press.
Greenough, P., & Pritchard, D. (Eds.). (2009). Williamson on knowledge. Oxford: Oxford University Press.
Harman, G. (1973). Thought. Princeton: Princeton University Press.
Hintikka, J. (1962). Knowledge and belief. Ithaca: Cornell University Press.
Klein, P. D. (1971). A proposed defınition of propositional knowledge. The Journal of Philosophy, 68(16), 471–482.
Klein, P. D. (1981). Certainty: A refutation of scepticism. Minneapolis: University of Minnesota Press.
Klein, P. D. (2004). Knowledge is true, non-defeated belief. In S. Luper (Ed.), Essential knowledge: Readings in epistemology (pp. 124–135). New York: Longman.
Klein, P. D. (2008). Useful false beliefs. In Q. Smith (Ed.), New essays in epistemology (pp. 25–63). Oxford: Oxford University Press.
Klein, P. D. (2017). The nature of knowledge. In R. Borges, C. de Almeida, & P. D. Klein (Eds.), Explaining knowledge (pp. 35–56). Oxford: Oxford University Press.
Kripke, S. (2011). Philosophical troubles. Oxford: Oxford University Press.
Lasonen-Aarnio, M. (2010). Unreasonable knowledge. Philosophical Perspectives, 24, 1–21.
Lehrer, K., & Paxson, T. (1969). Knowledge: Undefeated justifıed true belief. The Journal of Philosophy, 66(8), 225–237.
Littlejohn, C. (2012). Justification and the truth-connection. Cambridge: Cambridge University Press.
Luzzi, F. (2014). What does knowledge-yielding deduction require of its premises? Episteme, 11(3), 1–15.
Malcolm, N. (1952). Knowledge and belief. Mind, 61(242), 178–189.
Montminy, M. (2014). Knowledge despite falsehood. Pacific Philosophical Quarterly, 44(3–4), 463–475.
Nozick, R. (1981). Philosophical explanations. Cambridge, MA: Belknap Press of Harvard University Press.
Pritchard, D. (2016). Epistemic angst: Radical skepticism and the groundlessness of our believing. Princeton: Princeton University Press.
Schnee, I. (2015). There is no knowledge from falsehood. Episteme, 12(1), 53–74.
Sorensen, R. (2012). Epistemic paradoxes. In N. Z. Edward (Ed.), The Stanford encyclopedia of philosophy. Winter 2011 Edition.
Sutton, J. (2007). Without justification. Cambridge, MA: MIT Press.
Warfıeld, T. A. (2005). Knowledge from falsehood. Philosophical Perspectives, 19(1), 405–416.
Williamson, T. (2000). Knowledge and its limits. Oxford/New York: Oxford University Press.
Williamson, T. (2011). Knowledge first epistemology. In S. Bernecker & D. Pritchard (Eds.), The Routledge companion to epistemology (pp. 208–218). New York: Routledge.
Williamson, T. (Forthcoming). Justifıcations, excuses, and sceptical scenarios. In J. Dutant & F. Dorsch (Eds.), The new evil demon. Oxford: Oxford University Press.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Borges, R. (2019). Knowledge, despite Evidence to the Contrary. In: Fitelson, B., Borges, R., Braden, C. (eds) Themes from Klein. Synthese Library, vol 404. Springer, Cham. https://doi.org/10.1007/978-3-030-04522-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-04522-7_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-04521-0
Online ISBN: 978-3-030-04522-7
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)