Abstract
Duncan Pritchard has argued that his basis-relative anti-luck construal of a safety condition on knowing avoids the problem with necessary truths that safety conditions are often thought to have, viz., that beliefs the contents of which are necessarily true are trivially safe. He has further argued that adding an ability condition to truth, belief, and his anti-luck safety conditions yields an adequate account of knowledge. In this paper, we argue that not only does Pritchard’s anti-luck safety condition have a problem with necessary truths, adding an ability condition is of no help. Indeed, the same sort of case that precipitates Pritchard’s introduction of an ability condition shows the inadequacy of his completed anti-luck account of knowledge. Moreover, reconstruing safety as an anti-risk condition as Pritchard has recently done does not fix the problem we’ve identified. We conclude by entertaining a radical suggestion to the effect that the failures of safety-based accounts of modal knowledge are due to failures of doxastic success rather than failures to satisfy an anti-luck (or anti-risk) condition. Accepting this radical suggestion makes available the view that there is, after all, no special problem between safety and necessary truths.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Here p is a variable ranging over declarative sentences and \(\langle p\rangle\) is the proposition expressed by p, read ‘that p’.
Since Pritchard’s arguments are most fully developed in (2012), we privilege this paper in our discussion.
See p. 3.
This case is due to Carl Ginet and was introduced into the literature by Alvin Goldman in (1976).
Cf. Pritchard (2009b, p. 34).
Cf. our (2011, p. 554).
Pritchard isn’t alone in adopting this version of safety, which sometimes goes under the rubric global safety. See, e.g., Brown (2000, p. 125), Manley (2007, pp. 403–404), Miščević, (2007, p. 60), Sainsbury, (1997, pp. 908–909), Weatherson, (2004, p. 377), and Williamson, (2000, p. 182). For criticism of this version of safety, see, e.g., Bernecker (2020), Paterson (2020), Roland and Cogburn (2011), and Zhao (2021). Hirvelä (2019) criticizes versions of safety akin to PPRS, yet also endorses modified global safety conditions (2019; 2022). With the exception of our earlier article, the critiques of global safety all differ from that in the present paper, most importantly in that none of them uses a Mathemae-type example (see the next section and following).
Cf. the helpful demon counterexample to simple (process) reliabilism by John Greco (1999, p. 286).
Our Mathemae example was arrived at independently around the same time as Pritchard’s Mathemi. See our (2011).
Cf. Pritchard (2009b, pp. 40–41).
Hudson (2014) makes a similar point against Pritchard, though for a different purpose. We return to this point at the end of this section.
See, e.g., Greco (1999, pp. 286–291).
See Pritchard (2010, p. 55).
Thanks to an anonymous referee for raising this possibility.
“It would be different if the world were engineered to guarantee you true beliefs within a certain domain (e.g., by a helpful demon). Now I would grant that your beliefs are safe, though they still aren’t knowledge” (Pritchard 2020, n. 20, original emphasis).
Note that Pritchard himself engages in the same sort of shorthand way of speaking as we do in this section in the context of discussing his Mathemi case.
Note that we do not here intend to suggest that, e.g., the truth of \(\langle p\rangle\) is a product of S’s relevant cognitive abilities. Clearly this will only be so in special cases. Rather, we’re here trying to get at the idea that S’s relevant cognitive abilities are in some way responsible for her having a belief which is both true and safe.
Note that these considerations apply also to other traditional methods of acquiring knowledge of necessary truths (e.g., logico-mathematical proof). So denying that an agent in a normal functioning calculator case satisfies (4\('\)) would likely threaten full-blown skepticism about knowledge of necessities on ALVE. This would be unpalatable to Pritchard.
This move smacks of something like the generality problem for process reliabilism. We reserve consideration of this issue for future work.
One might think that perhaps such a condition is needed to deal with other cases. Given the wide latitude for describing methods, it’s difficult to see why such cases couldn’t also be dealt with by describing the method at issue in such a way that the relevant belief turns out to be unsafe—just as this move does with the Temp case.
Someone who has read Hirvelä’s (2019) might wonder whether the virtuous method approach to knowing necessary truths advanced there might help ALVE here. While this approach does not honor the letter of ALVE, it does honor the spirit of ALVE in incorporating both virtue-theoretic and safety-theoretic components into a modal epistemology. Maybe that’s enough for Pritchard and proponents of ALVE. Regardless, there’s no help here. Hirvelä acknowledges that the agent in his malfunction case (which is essentially Pritchard’s Mathema case), Paige, forms her belief using a virtuous method (p. 1180). Thus, so does Mathemae. But where Paige fails to have a virtuously safe belief because in some worlds close to hers she forms a false belief using the same method (p. 1181), Mathemae’s belief is virtuously safe, precisely because there are no close worlds where Mathemae forms a false belief using the same method. So though Hirvelä’s approach appears to give the right verdict on malfunction/Mathema, it gives the same wrong verdict as ALVE on Mathemae.
This notion of initial conditions is intuitively clear enough for present purposes. For some discussion, see Pritchard (2016, pp. 555–556).
See Pritchard (2016, p. 560).
See Pritchard (2020, p. 211).
Prime examples are the linguistic convention view of the logical positivists/empiricists and Gödel’s intuition-based views. For an overview of contemporary views and their issues, see Mallozzi et al. (2021).
In Quinean terms, this is like making adjustments closer the the center of our web of belief in the face of recalcitrant experience.
There is, of course, a justifiably enormous literature in the philosophy of language and mind concerning belief individuation and attribution. Epistemologists could do worse than start with the literature jointly inspired by Dennett (1989), Davidson (2001), and Stich (1996). In addition to standard Fregean worries about intensionality, and Davidsonian worries about normativity, issues concerning semantic internalism and externalism concern how much must be creditable to an agent for her to be a believer and the extent to which this is context relative, especially given well known underdetermination concerns. To the extent that one takes the above suggestion seriously, one might begin to suspect that one cannot be an externalist about both justification and belief simultaneously and in the same doxastic/epistemic contexts. How one comes down on this might very well (depending on whether or not contextualism can massage away the differences) then depend on whether one finds the arguments against internalism more compelling with respect to justification or belief.
If one is unsure that BonJour’s original Norman case supports this modal robustness, the case can esaily be modified to ensure it.
If one wants a case making this point where naturalistic qualms figure less prominently as a potential reason to reject the case, consider Keith Lehrer’s (2000) Truetemp. A similar disentangling modification of the case to Truetemp\(^\star\) is required. But it works to indicate the same kind of misdiagnosis of the problem with the case as with Norman and Norman\(^\star\).
Nothing we write in this paper requires that knowers have or use language. If a lion could speak, he would have a lot to say.
On this account, Mathemae arguably knows if she understands \(\langle 12 \times 13 = 156\rangle\) and thereby satisfies the belief condition on knowing. At least one of the current authors is not entirely comfortable with this result, and thinks that perhaps naturalistic considerations suggested by (1) above might need to be invoked to avoid it. Note, however, that regardless of whether Mathemae knows if she possesses the requisite understanding, the same cannot obviously be said for Temp. If Pritchard is correct in his view of Temp, pace Kelp and Hudson and as at least one of us believes (see n.13), then even if he possesses the requisite understanding to believe the relevant temperature propositions Temp’s knowing can still be ruled out using an appropriate ability condition. The Temp case exhibits the wrong direction of fit, and that’s sufficient to undermine Temp’s knowing. But as we argued in Sect. 5, the Mathemae case cannot exhibit the wrong direction of fit.
Notice that if the problem with a case really is in satisfaction of the belief condition, then it’s not really a false positive. So perhaps we should say ‘apparent false positive’ in this context.
Note that the manner in which PPRS quantifies over methods is something open to sensitivity theory, which we think has been abandoned with undue haste by modal epistemologists. This is one of the non-trivial conclusions or our (2011).
See especially in this context the discussion of externalism in Brandom (2000) as well as the demonstrated relevance of Fregean concerns to Gettierology in Heck (1995). For an antidote to the rebarbative things Brandom says about animals in these contexts, one especially promising with respect to the purported reallocation of the division of labor suggested above, see Okrent (2007).
Note that the problem of knowledge of necessary truths becomes, in the philosophy of language, the problem of individuating the content of necessary truths. This is not insuperable. See for example the brief for counterpossibles in individuating content in Priest (2002).
References
Bernecker, S. (2020). Against global method safety. Synthese, 197, 5101–5116.
BonJour, L. (1985). The structure of empirical knowledge. Cambridge University Press.
Brandom, R. (2000). Articulating reasons: An introduction to inferentialism. Harvard University Press.
Brown, J. (2000). Reliabilism, knowledge, and mental content. Proceedings of the Aristotelian Society, 100, 115–135.
Davidson, D. (2001). Inquiries into truth and interpretation. Clarendon Press.
Dennett, D. (1989). The intentional stance. MIT Press.
Goldman, A. (1976). “Discrimination and perceptual knowledge". In Liasons: Philosophy meets the cognitive sciences (pp. 85–103). MIT Press.
Greco, J. (1999). Agent reliabilism. Philosophical Perspectives, 13, 273–296.
Greco, J. (2003). Knowledge as credit for true belief. In L. Z. Michael DePaul (Ed.), Intellectual virtue: Perspectives from ethics and epistemology (pp. 111–134). Oxford University Press.
Heck, R. (1995). The sense of communication. Mind, 104, 79–106.
Hirvelä, J. (2019). Global safety: How to deal with necessary truths. Synthese, 196, 1167–1186.
Hirvelä, J. (2022). Justification and the knowledge-connection. Philosophical Studies, 179, 1973–1995.
Hudson, R. (2014). Saving Pritchard’s anti-luck virtue epistemology. Synthese, 191, 801–815.
Kelp, C. (2013). Knowledge: The safe-apt view. Australasian Journal of Philosophy, 91, 265–278.
Lackey, J. (2007). Why we don’t deserve credit for everything we know. Synthese, 158, 345–361.
Lehrer, K. (2000). Externalism and epistemology naturalized. In E. Sosa & J. Kim (Eds.), Epistemology: An anthology (pp. 387–400). Blackwell Publishers.
Mallozzi, A., Vaidya, A., & Wallner, M. (2021). The epistemology of modality. In Zalta, E. N., editor, The stanford encyclopedia of philosophy.
Manley, D. (2007). Safety, content, apriority, self-knowledge. Journal of Philosophy, 104, 403–423.
Miščević, N. (2007). Armchair luck. Acta Analytica, 22, 1–22.
Okrent, M. (2007). Rational animals: The teleological roots of intentionality. Ohio University Press.
Paterson, N. J. (2020). Safety and necessity. Erkenntnis. https://doi.org/10.1007/s10670-020-00231-6.
Priest, G. (2002). In contradiction: A study of the transconsistent. Oxford University Press.
Pritchard, D. (2007). Anti-luck epistemology. Synthese, 158, 277–297.
Pritchard, D. (2009a). Knowledge. Palgrave MacMillan.
Pritchard, D. (2009b). Safety-based epistemology: Whither now? Journal of Philosophical Research, 34, 33–45.
Pritchard, D. (2010). Knowledge and Understanding. In D. Pritchard, A. Millar, & A. Haddock (Eds.), The nature and value of knowledge (pp. 1–88). Oxford University Press.
Pritchard, D. (2012). Anti-luck virtue epistemology. Journal of Philosophy, 109(3), 247–279.
Pritchard, D. (2016). Epistemic risk. Journal of Philosophy, 113, 550–571.
Pritchard, D. (2020). Anti-risk virtue epistemology. In C. Kelp & J. Greco (Eds.), Virtue theoretic epistemology (pp. 203–224). Cambridge University Press.
Pritchard, D. (2021). Varieties of epistemic risk. Acta Analytica. https://doi.org/10.1007/s12136-021-00489-7.
Roland, J. W. & Cogburn, J. (2011). Anti-luck epistemologies and necessary truths. Philosophia, 39, 547–561.
Sainsbury, M. (1997). Easy possibilities. Philosophy and Phenomenological Research, 57, 909–919.
Sosa, E. (2007). A virtue epistemology: Apt belief and reflective knowlewdge. Oxford University Press.
Stich, S. (1996). Deconstructing the mind. Oxford University Press.
Weatherson, B. (2004). Luminous margins. Australasian Journal of Philosophy, 82, 373–383.
Williamson, T. (2000). Knowledge and its limits. Oxford University Press.
Zhao, H. (2021). Better virtuous than safe. Synthese, 198, 6969–6991.
Acknowledgements
Thanks to five anonymous referees for this journal for valuable comments. A distant ancestor of this paper was presented at the Annual Meeting of the Alabama Philosophical Society in 2012. Thanks to the participants of that session.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Roland, J.W., Cogburn, J. Pritchard’s Epistemology and Necessary Truths. Erkenn (2022). https://doi.org/10.1007/s10670-022-00636-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10670-022-00636-5