Abstract
A powerful objection to subject-sensitive invariantism (SSI) concerns various ‘strange-but-true’ (or “embarrassing”) conditionals. One popular response to this objection is to argue that strange-but-true conditionals pose a problem for non-sceptical epistemological theories in general. In the present paper, it is argued that strange-but-true conditionals are not a problem for contextualism about ‘know’. This observation undercuts the proposed defence of SSI, and supplies a surprising new argument for contextualism.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
(2) is adapted from an example in Cohen (2002).
See e.g. DeRose (2009: 194), Fantl and McGrath (2009a: 208), Hawthorne (2004: 177n), Schaffer (2006: 94n), and Stanley (2005: 106). Note that examples like (1)–(2) can sometimes seem true (and not strange). The problem is that SSI implies that examples like (1)–(2) can be true even if we suppose that the traditional ‘epistemic’ factors—the subject’s belief state, the reliability of her methods, etc.—remain constant across the actual situation and the relevant counterfactual situation in which the subject has more at stake, or is considering additional error possibilities. (For a more careful account of what to hold constant—one that aims to respect the fact that some proponents of SSI claim that there is a constitutive tie between ‘non-epistemic’ factors, such as stakes, and some traditional epistemic factors, such as evidence—see DeRose’s (2009: 194–196) appeal to “truth relevant” factors. Such subtleties need not concern us here.).
A similar problem arises regarding examples like the following:
(3)* If Jean is considering the possibility that the table is white but illuminated by red lights, she doesn’t know that the table is red. But if she isn’t considering improbable ways she might be mistaken, she knows that the table is red.
Note, though, that the most common forms of SSI tie knowledge to practical factors, such as stakes, but not to the error possibilities salient to the subject. These accounts can presumably avoid problems with examples like (2) and (3)*.
In Sect. 3, we shall look at some ‘metalinguistic’ examples, similar to (1)–(3), that might seem to pose a problem for contextualists about ‘know’.
See Goldman (1976) for the original presentation of the ‘fake barn’ case. (Goldman credits the example to Carl Ginet.) Some concerns have been raised regarding our judgments in the original fake barn case; see e.g. DeRose (2009: 23n), Gendler and Hawthorne (2005), and, for relevant empirical work, Colaço et al. (2014). But note that similar ‘strange-but-true’ conditionals plausibly arise in regard to other examples, such as the various ‘fake ring’ cases discussed in Gendler and Hawthorne (2005: 334–338); the arguments of the present paper could be recast in terms of those alternative examples. Of course, it should be acknowledged that further empirical work may ultimately upend claims about our judgments in all the relevant cases. This would serve to undercut the argument for contextualism given in the present work—but it would also undercut the defence of SSI being considered.
All else equal, a theory that results in fewer problematic judgments is presumably to be preferred. Advocates of SSI must presumably accept examples like (1)–(3) and (4)–(5) as strange-but-true, whereas rival non-sceptical theorists need (at most) accept only (4)–(5) as strange-but-true. The observation that (4)–(5) appear to pose a general problem for non-sceptical theorists therefore does not altogether remove the strike against SSI issuing from (1)–(3).
See Hawthorne (2004: 177n), Stanley (2005: 113–114), Fantl and McGrath (2009a: 209). The subject-sensitive invariantist response to examples like (1)–(3) given in the main text closely resembles Fantl and McGrath (2009a: 209). Stanley (2005: 113–114, drawing on p.c. with John Hawthorne) presents a slightly different response. Stanley suggests that examples like (4)–(5) are a problem for reliabilism, and since another popular view (reliabilism) has similar problems to SSI, strange-but-true conditionals are not a serious objection to SSI. This is clearly a much weaker response than the one presented in the main text. After all, if strange-but-true conditionals are problems merely for SSI and reliabilism, don’t such conditionals supply good reason to reject both those theories (DeRose 2009: 197–198)? The arguments to follow also further weaken Stanley’s argument, since it is argued that a contextualist—and so, plausibly, a reliabilist who endorses contextualism—need not hold that the envisaged utterances of (4)–(5) are strange-but-true. Thus, the arguments to follow suggest that (4)–(5) are not a problem for reliabilism per se (as Stanley suggests), but rather a problem for those reliabilists who embrace non-sceptical invariantism.
For relevant exposition and defence of contextualism about ‘know’, see e.g. Lewis (1996), Cohen (1988, 1999), DeRose (1995, 2009), Ichikawa (2011) and Blome-Tillmann (2014). To keep things simple, I am focusing on a simple-minded contextualist proposal that emphasises ruling out salient error possibilities (as in Lewis 1996). But I take it that an explanatory strategy similar to the one outlined in the main text is available to many other contextualists. See Sect. 3 for relevant discussion.
Parallel remarks apply to (4). For relevant discussion of how the contextualist takes ‘knows’ to behave when uttered in temporally/modally shifted contexts (as in (4)), see e.g. DeRose (2009: esp. 204–206).
To secure this result, the contextualist might propose that an additional requirement on true application of ‘know’ (in any context) is that the subject be able to rule out the error possibilities that are ‘close’ or could easily obtain (cf. Sainsbury 1997; Sosa 1999; Williamson 2000). This requirement plausibly implies that if Henry is surrounded by fake barns, he cannot be truly said to ‘know’, in any context, that he is seeing a barn. But if Henry is not surrounded by fakes, this requirement will often leave open that Henry can be truly said to ‘know’. (Note, though, that the contextualist need not adopt this proposal. There may be some benefits to a contextualist account that allows that, in at least some contexts, Henry can be truly said to ‘know’ even if he is surrounded by fake barns; see e.g. DeRose (2009: 33n) and Greco (2017) for some relevant discussion).
The arguments of the present paper might suggest that those seeking to defend merely the anti-intellectualist element of SSI—viz. that knowledge is constitutively tied to ‘non-epistemic’ factors—could avoid problems with strange-but-true conditionals if they embrace contextualism about ‘know’. I cannot explore that possibility here.
These authors do not consider or comment on examples like (5). Note that Blome-Tillmann and DeRose’s remarks are (as far as I’m aware) the extant contextualist attempts to undercut the defence of SSI by appeal to examples like (4)–(5). The contention of the present paper is that there is a far better way to go—one that respects the fact that others find (4)–(5) genuinely strange, undercuts the proposed defence of SSI, and supplies a new argument for contextualism about ‘know’.
Note that this is not the only possible contextualist explanation for DeRose and Blome-Tillmann’s judgments; see the discussion of the next objection, and fn. 18.
Notice that if relevant utterances of (4) or (5) are to strike hearers as true, those hearers presumably need to appreciate that the reason why Henry lacks ‘knowledge’ if he is surrounded by fakes, but possesses ‘knowledge’ if he is not surrounded by fakes is that in the former case he is much more susceptible to fake barn deception than he is in the latter case. This would seem to suggest that for utterances of (4) and (5) to actually strike hearers in the context of utterance as true, the possibility of fake barn deception needs to be taken seriously.
A more recherché objection. A distinction is sometimes drawn between considering a possibility as actual (as a way the world might actually be) and considering it as counterfactual (as a way the world might have been)—see e.g. Chalmers (2002: §3). It might be suggested that when uttering (4) or (5) what most immediately happens is that we consider possibilities in which Henry is looking at a fake barn as ways things could have been for Henry, rather than as ways things might actually be for Henry. But it might be proposed that insofar as considering possibilities has an impact on what possibilities a subject must be able to rule out in order to ‘know’, it is only the possibilities that are being considered as actual that have an impact. (Note that the relevant ‘actual’ situation here is Henry’s actuality, not ours).
Once again, these reflections represent no serious cause for concern. One option for the contextualist is obviously just to deny that it is only possibilities that are being considered as actual that impact the requirements on ‘know’. But suppose that the proposal is right. Given how Henry’s actual situation is described (Henry looks out at a real barn as he is driving through the countryside), it seems plausible that even if, when (4)–(5) are uttered, we don’t consider possibilities in which Henry is looking at a fake barn as actual as immediately as we consider them as counterfactual, it nevertheless seems plausible that many people will consider them as actual fairly quickly. At the same time, as noted above, not all authors react to utterances of (4)–(5) by declaring that such utterances seem strange. A further explanation for their reactions could be that the proposal being considered is correct, and that these authors are imagining (4)–(5) being uttered in contexts where possibilities in which Henry is looking at a fake barn are considered merely as counterfactual and not as actual.
Non-sceptical invariantists are either classical or subject-sensitive invariantists. As noted in Sect. 1, classical invariantists—those non-sceptical invariantists who reject the claim that knowledge is constitutively tied to ‘non-epistemic’ factors—do not face a challenge from examples like (1)–(3). Nevertheless, such theorists presumably face a challenge from examples like (4)–(5).
For relativist treatment of an example similar to (6), see MacFarlane (2014: 197). For considerations of space, I cannot recount such an explanation here.
To any readers not sensing discomfort with (7), I reiterate Lewis’ (1996: 550) familiar advice to “hear it afresh”.
The argument presented in the main text is one way to argue that examples like (6) do not pose a similar problem for standard contextualism that examples like (1)–(3) and/or (4)–(5) apparently pose for non-sceptical invariantism. There are at least two further ways to argue for this conclusion. One way is to argue that it is less important for a theory to capture judgments regarding ‘metalinguistic’ examples (i.e. those examples, like (6), where the target expression is mentioned rather than used) since the meaning of a term is primarily determined by how it is used rather than how it is mentioned (cf. DeRose 2009: 153–154). Another way is to argue that the contextualist faces fewer strange-but-true conditionals than rival non-sceptical accounts, and that such conditionals are therefore a more serious problem for rival views than for contextualism (cf. fn. 9).
As indicated in Sect. 3, contextualists are also able to avoid problems with ‘metalinguistic’ examples similar to (1)–(3), such as (6).
To direct future research, it may be useful to think of the positive argument for contextualism offered here as presenting a kind of challenge for the contextualist’s non-sceptical opponents. As noted in the discussion of (6), it seems plausible that strange-but-true conditionals do not pose a genuine problem for a theory if some suitably motivated explanation can be found for why those conditionals seem strange despite being true. The challenge for the contextualist’s non-sceptical opponents would therefore seem to be to provide a suitably motivated explanation for why examples like (1)–(3) and/or (4)–(5) seem strange despite being true. If that challenge can be met, it would undercut the positive argument for contextualism presented above; if not, that argument remains in force.
References
Blome-Tillmann, M. (2009). Contextualism, subject-sensitive invariantism, and the interaction of ‘knowledge’-ascriptions with modal and temporal operators. Philosophy and Phenomenological Research, 79, 315–331.
Blome-Tillmann, M. (2012). Presuppositional epistemic contextualism and the problem of known presuppositions. In J. Brown & M. Gerken (Eds.), Knowledge ascriptions (pp. 104–119). Oxford: Oxford University Press.
Blome-Tillmann, M. (2014). Knowledge and presuppositions. Oxford: Oxford University Press.
Chalmers, D. (2002). Does conceivability entail possibility? In T. Gendler & J. Hawthorne (Eds.), Conceivability and possibility (pp. 145–200). Oxford: Oxford University Press.
Cohen, S. (1988). How to be a fallibilist. Philosophical Perspectives, 2, 91–123.
Cohen, S. (1999). Contextualism, skepticism, and the structure of reasons. Philosophical Perspectives, 13, 57–89.
Cohen, S. (2001). Contextualism defended: Comments on Richard Feldman’s skeptical problems, contextualist solutions. Philosophical Studies, 103, 87–98.
Cohen, S. (2002). Basic knowledge and the problem of easy knowledge. Philosophy and Phenomenological Research, 65, 302–329.
Colaço, D., Buckwalter, W., Stich, S., & Machery, E. (2014). Epistemic intuitions in fake-barn thought experiments. Episteme, 11, 199–212.
DeRose, K. (1995). Solving the skeptical problem. The Philosophical Review, 104, 1–52.
DeRose, K. (2009). The case for contextualism (Vol. 1). Oxford: Clarendon Press.
Fantl, J., & McGrath, M. (2007). On pragmatic encroachment in epistemology. Philosophy and Phenomenological Research, 75, 558–589.
Fantl, J., & McGrath, M. (2009a). Knowledge in an uncertain world. Oxford: Oxford University Press.
Fantl, J., & McGrath, M. (2009b). Critical study of John Hawthorne’s knowledge and lotteries and Jason Stanley’s knowledge and practical interests. Noûs, 43, 178–192.
Gendler, T., & Hawthorne, J. (2005). The real guide to fake barns: A catalogue of gifts from your epistemic enemies. Philosophical Studies, 124, 331–352.
Goldman, A. (1976). Discrimination and perceptual knowledge. The Journal of Philosophy, 73, 771–791.
Greco, J. (2017). Contextualism and Gettier cases. In J. Jenkins Ichikawa (Ed.), The Routledge handbook of epistemic contextualism (pp. 190–202). Oxford: Routledge.
Hawthorne, J. (2004). Knowledge and lotteries. Oxford: Clarendon.
Ichikawa, J. (2011). Quantifiers and epistemic contextualism. Philosophical Studies, 155, 383–398.
Lewis, D. (1979). Scorekeeping in a language game. Journal of Philosophical Logic, 8, 339–359.
Lewis, D. (1996). Elusive knowledge. Australasian Journal of Philosophy, 74, 549–567.
MacFarlane, J. (2005). The assessment sensitivity of knowledge attributions. In T. S. Gendler & J. Hawthorne (Eds.), Oxford studies in epistemology (Vol. 1, pp. 197–233). Oxford: Oxford University Press.
MacFarlane, J. (2014). Assessment sensitivity: Relative truth and its applications. Oxford: Oxford University Press.
Pritchard, D. (2016). Epistemic risk. The Journal of Philosophy, 113, 550–571.
Sainsbury, M. (1997). Easy possibilities. Philosophy and Phenomenological Research, 57, 907–919.
Schaffer, J. (2006). The irrelevance of the subject: Against subject sensitive invariantism. Philosophical Studies, 127, 87–107.
Sosa, E. (1999). How to defeat opposition to Moore. Philosophical Perspectives, 13, 141–153.
Stanley, J. (2005). Knowledge and practical interests. Oxford: Clarendon Press.
Weatherson, B. (2012). Knowledge, bets, and interests. In J. Brown & M. Gerken (Eds.), Knowledge ascriptions (pp. 75–103). Oxford: Oxford University Press.
Weatherson, B. (2017). Interest-relative invariantism. In J. Jenkins Ichikawa (Ed.), The Routledge handbook of epistemic contextualism (pp. 250–254). Oxford: Routledge.
Williamson, T. (2000). Knowledge and its limits. Oxford: Oxford University Press.
Acknowledgements
Thanks to Jessica Brown, Patrick Greenough, Dirk Kindermann, Daniele Sgaravatti, and an anonymous referee for helpful comments and discussion.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dimmock, P. Strange-but-true: a (quick) new argument for contextualism about ‘know’. Philos Stud 175, 2005–2015 (2018). https://doi.org/10.1007/s11098-017-0947-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-017-0947-9