Philosophers have recently argued that self-fulfilling beliefs constitute an important counter-example to the widely accepted theses that we ought not and cannot believe at will. Cases of self-fulfilling belief are thought to constitute a special class where we enjoy the epistemic freedom to permissibly believe for pragmatic reasons, because whatever we choose to believe will end up true. In this paper, I argue that this view fails to distinguish between the aim of acquiring a true belief and the aim of believing what is true. While one cannot usually fail to establish that one will acquire a true belief without establishing the truth of the believed proposition, in the case self-fulfilling belief the two can come apart. I argue that insofar as the aim of belief has to do with determining whether the believed proposition is true, it will be both impossible and impermissible to believe for pragmatic reasons.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
A case adapted from Johnston (1989). I discuss other aspects of this case in Antill (forthcoming).
As Dahlback (forthcoming) shows, an analogous puzzle can be formulated in terms of credences and corresponding objective probabilities. There are also numerous real world examples. Cases of self-fulfilling belief span the quotidian—a child’s belief that she will be fed may prompt a parent to feed her—to more technical cases from social psychology—as in cases where students’ performances match the expectation of their teachers. Unlike the millionaire example above, it will be unclear in many of these real world cases whether—and to what extent—a belief in an outcome makes the outcome more probable. In this paper, I will focus on the millionaire case to avoid added complications concerning how to reason when you have evidence that a belief in a proposition might be self-fulfilling.
In fact, the view that self-fulfilling beliefs involve a distinctive form of epistemic freedom has enjoyed a recent resurgence. Some or all of Velleman’s position has been taken up by Dahlback (forthcoming), Drake (forthcoming), Raleigh (2017), McHugh (2015), Peels (2014), Reisner (2013) and Foley (1991). While I will reference some important contributions to Velleman’s underlying position from these works in Sects. 2 and 3, I will continue to focus on Velleman’s original argumentation, since it continues to be the fullest account and most robust defense of the position.
Nothing I say will count against these other ways of establishing our license to self-fulfilling belief. What I hope to argue is (a) that such a license cannot be grounded in an uncontroversial epistemic picture of the aim of belief and (b) that on a relatively attractive reading of the aim of belief, there are significant barriers to such beliefs ever being permissibly adopted.
I want to distinguish here two different claims about what might justify the self-fulfilling belief: the fact that the agent will have evidence for the belief, once held, and the fact that the belief will be true, if held. There is some textual evidence for attributing to Velleman each view (see, e.g. fns. 7, 14). While the positions are importantly different, these differences will not affect the discussion to follow.
There are two related questions we might ask about the believer’s theoretical reasoning: one normative, concerning the requirements governing how we should engage in theoretical reasoning; the second descriptive, concerning the limits on how it is possible to come to form beliefs through reasoning or other related processes. I will focus on the normative question in Sect. 2, and the descriptive question in Sect. 3.
While this does not require the strong ontological claim that reasons be the kinds of objects which can be used in reasoning directly, it does assume a tight connection between reasons and reasoning. I am assuming that reasons for belief—whether they are facts or mental states—must be relevant to the agent’s reasoning. See Mitova (2017), Way (2017), or Hieronymi (2005) for a defense of this position, as well as discussion of the some of the difficulties such a view faces.
Though there may be other important epistemological projects, this seems at least one important epistemic project with which we might be concerned. See Goldman (1980) for further discussion of the difference between this more regulative sense of ‘theoretical rationality’ and other, more evaluative, senses of ‘theoretical rationality.’
These obstacles are generated by the broadly-evidentialist assumption, noted earlier, that theoretical reasoning is concerned with truth, rather than with pragmatic or more general epistemic concerns. While I take this to be a common view about the nature of theoretical reasoning, it is not uncontroversial. A supporter of epistemic permissivism or doxastic voluntarism might simply reject this model of theoretical reasoning, and so deny that these objections need answering. However, part of the appeal of Velleman’s defense of epistemic freedom is that it appears to have persuasive answers to these objections, and so promises to support epistemic permissivisim or doxastic voluntarism, even for those who do find such evidentialist-leaning models of theoretical reasoning persuasive. (thanks to two anonymous reviewers for helping me to articulate this point more clearly).
One might worry that such an account will still involve an illicit bootstrapping, if your reason is the fact that you will have evidence for p once you believe that p, rather than the fact that you would have evidence for p were you to believe that p. This issue will resurface in Sect. 3.
The most plausible examples of this pattern of reasoning involve the (purportedly) self-fulfilling beliefs Velleman is most interested in: our beliefs about our own actions.
Since permissivism has been enjoying a bit of a resurgence in recent years, it is worth emphasizing precisely how radical this present form of permissivism would be. Some people think that there can be reasonable disagreement between epistemic peers, or, more radically, that the very same person might follow one of two or more equally reasonable principles for weighing reasons, and so might, depending on which principle they are following, be permitted in adopting one of two conflicting doxastic attitudes (see Schoenfield 2013). But while there is some debate over these weaker forms of permissivism, there is near universal support for the thesis that the very same rule, applied to the same person in the same context, should not permit both believing and disbelieving, as Velleman’s epistemic freedom would entail.
I contrast what would be true with what is true, because I assume that the truth of a proposition is eternal (though, in the case of self-fulfilling belief that p, p is eternally true only because you will eventually believe it). Nothing hangs on this assumption, however. If one accepts some thesis about the indeterminacy of future contingents, one can instead speak of the contrast between what would be true and what will be true. In either case, the central difference will not be temporal (whether we should be concerned with getting the world right now rather than at the time the belief is held) but modal (whether we should be concerned with establishing how the world actually is or will be, rather than establishing that we would have a true belief, whatever the world would turn out to be like (Thanks to an anonymous referee for encouraging me to clarify this important issue).
The differences between truth and the good only entail differences in the permissibility of acting and believing on the assumption that sufficient reasons for believing must be reasons sufficient to show p true (whereas two contrary actions might still be permissible, even on the assumption that reasons for acting must be sufficient to show the action choice-worthy or good). Thanks to an anonymous reviewer for helping me to see the importance of this point.
Not only are your reasons insufficient to show whether p is true, they are also insufficient to show whether p will be true. Because, while you can establish that a belief that p and a belief that not-p would both be correct, so that whatever you believe would be true, you have not yet established which of these two propositions will be true.
c/f Section 1 pp. 3–5.
This is particularly troubling given that many (though not all) of the seminal works on the aim of belief are explicitly proposition centered views. Wedgwood (2002) for example, glosses the claim that belief aims at the truth as the claim that “a belief is correct if and only if the proposition believed is true.” Indeed, Velleman himself appears, to endorse the second interpretation in later work, glossing the aim of true belief in Shah and Velleman (2005) as the aim of “believing p only if p.” If right, disentangling these two interpretations of the aim of belief allows us to see that Velleman’s defense of epistemic freedom requires an interpretation of the aim of belief which Velleman himself does not unequivocally endorse.
This sort of argument finds support in its ability to explain what Shah and Velleman (2005, p. 5) refer to as the ‘transparency’ of doxastic deliberation: the phenomenon by which “the deliberative question whether to believe that p inevitably gives way to the factual question whether p”.
An argument presented forcefully in Berker (2013). As Berker points out, self-fulfilling beliefs are compatible with a more narrow epistemic consequentialist requirement, restricted to producing correct doxastic states in individual propositions, which avoids some of these counter-examples. However, if the foundation of epistemic rationality is the aim of acquiring true belief, it seems, at the least, a further puzzle what would motivate restricting consequentialist requirements in such a way.
Raleigh (2017), for example, is explicit in his willingness to accept an epistemic consequentialist defense of epistemic permissivism.
Thanks to an anonymous reviewer for this helpful way of framing the point.
Though epistemic permissivism remains open to someone who accepts the state-centered conception of the aim of true belief, rejects the connection between reasons for belief and the aim of belief, or who denies belief aims at the truth at all. For a detailed discussion of epistemic consequentialism and a qualified defense of the state-centered conception of the aim of true belief, see Greaves (2013). For a more full-throated defense see Singer (forthcoming). For recent work rejecting the centrality of the aim of true belief in favor of a practical foundation to epistemic normativity, see McCormick (2015) and Rinard (2015, 2017).
That is to say, even if we accept that reasons which show a belief will be true might provide sufficient normative reason to permit the belief, there is a further question of whether it is possible to adopt such a belief—a question involving the believers possible operative reasons for believing. (Thanks to an anonymous reviewer for helping me see this important point more clearly.)
See for example, Reisner (2009).
This point requires a notion of intelligibility strong enough that it requires a practical reason must bear, for the agent, some minimal rational connection to the action, and by parallel, that theoretical reasons must bear some connection to the object belief. This should be acceptable, even to skeptics of ‘guise of the good’ theses, like Velleman (2000c) or Setiya (2007).
In the previous section, I assumed that an agent’s reasons to believe, or normative reasons, must bear some connection to the truth of the believed proposition. Here I am assuming a less controversial claim: that the reasons for which an agent believes—the agent’s operative reasons—must bear (or be taken to bear) some relation to the truth of that belief. For support of the assumption that practical considerations cannot constitute motivating reasons for belief, see Hieronymi (2005, 2006, 2009). For a contrasting view, see Leary (2017), who questions both the stronger claim about normative reasons for belief as well as the weaker claim that practical considerations cannot constitute motivating reasons for belief.
For a survey of historical thought on the issue, see Rescher (1960).
A notable exception is Reisner (2013), who deals with the possibility of epistemic Buridan cases explicitly, in just these sorts of situations. Reisner (2013) presents an especially nuanced discussion of epistemic Buridan cases which merits further discussion. Though Reisner uses the possibility of Buridan cases to defend evidential permissivism, it is not clear that his conclusions are in conflict with those I make here. First, strictly speaking, Reisner (2013) does not argue that we do in fact have a capacity to believe at will, but rather that there are doxastic buridan cases where such a capacity would be the only way to secure a true belief. Second, while I take myself to have shown, in this section, that we cannot resolve Buridan cases by believing at will on the basis of pragmatic reasons, I have not argued against the possibility that there may be some other way to resolve Buridan cases to arrive at a true belief. (This is especially relevant since Reisner is particularly concerned with cases of self-fulfilling belief where there may be no practical upshot. In such cases it would be unclear how practical reasons would resolve things.) I discuss Reisner’s examples elsewhere in more detail in Antill (Forthcoming).
A point made in Feldman (2000, pp. 679–682).
I am grateful to Calvin Normore for this helpful example.
Since I am interested in the general import of Velleman’s account for the prospects of doxastic voluntarism more generally, I will avoid discussion of what is for Velleman, one of the strongest reasons in favor of thinking that Buridan cases must be resolvable, which stems from his background commitment to a cognitivist account of intention: if intentions are a species of self-fulfilling belief, and we determine which action to intend on the basis of practical reasons, then believing on the basis practical reasons must be possible, since it is actual. For criticism of Velleman’s argument in the context of his cognitivist account of intention, see Langton (2004).
A nomenclature originating with Morgenbesser and Ullmann-Margalit (1977).
An important point brought out by Hieronymi (2005).
We can consider, for comparison, the ways in which having a power to form an intention at will, rather than pick an action at will, would circumvent, rather than provide resolution to, an agent deciding which bale of hay to pursue. (An important lesson from Kavka 1983’s Toxin Puzzle case.).
This argument shows the difficulties in resolving doxastic Buridan cases by believing directly on the basis of pragmatic reasons. However, this does not mean that someone who rejects Velleman’s conception of epistemic freedom must hold that Buridan cases are unresolvable. It remains open that we might resolve such Buridan cases on the basis of some automatic belief formation mechanism. It also remains open that such automatic belief formation mechanisms may be affected by our practical reasons, so that we can indirectly get ourselves to believe on the basis of pragmatic reasons.
Al-Ghazali. (2000). Tahafut Al-Falasafia. In M. E. Marmura (ed. and trans.) (2nd ed.), Provo (Utah): Brigham Young University Press.
Anscombe, G. E. M. (1957). Intention. Cambridge: Harvard University Press.
Antill, G. E. (Forthcoming). Evidence and self-fulfilling belief. American Philosophical Quarterly.
Averroes. (1954). Tahafut al-Tahafut. In S. Van den Bergh (trans.) (Vol. 2). London: Luzac.
Bennett, J. (1990). Why is belief involuntary? Analysis, 50(2), 87–107.
Berker, S. (2013). The rejection of epistemic consequentialism. Philosophical Issues, 23(1), 363–387.
Bykvist, K., & Hattiangadi, A. (2014). Belief, truth, and blindspots. In T. Chan (Ed.), The aim of belief (p. 123). Oxford: Oxford University Press.
Dahlback, M. (Forthcoming). Infinitely permissive. Erkenntnis.
Davidson, D. (1963). Actions, reasons, and causes. Journal of Philosophy, 60(23), 685–700.
Drake, J. (Forthcoming). Doxastic permisiveness and the promise of truth. Synthese.
Greaves, H. (2013). Epistemic decision theory. Mind, 122(488), 915–952.
Feldman, R. (2000). The ethics of belief. Philosophy and Phenomenological Research, 60(3), 667–695.
Foley, R. (1991). Evidence and reasons for belief. Analysis, 51(2), 98–102.
Goldman, A. I. (1980). The internalist conception of justification. Midwest Studies in Philosophy, 5(1), 27–51.
Gombay, A. (1988). Some paradoxes of counter-privacy. Philosophy, 63(244), 191–210.
Harman, G. (1973). Thought. Princeton: Princeton University Press.
Hieronymi, P. (2005). The wrong kind of reason. Journal of Philosophy, 102(9), 437–457.
Hieronymi, P. (2006). Controlling attitudes. Pacific Philosophical Quarterly, 87(1), 45–74.
Hieronymi, P. (2009). Believing at will. Canadian Journal of Philosophy, Supplementary, 35(sup1), 149–187.
Johnston, M. (1989). Self-deception and the nature of mind. In A. Rorty & B. P. McLaughlin (Eds.), Perspectives on self-deception. Berkeley: University of California Press.
Kavka, G. S. (1983). The toxin puzzle. Analysis, 43(1), 33–36.
Langton, R. (2004). Intention as faith. Royal Institute of Philosophy Supplement, 55, 243–258.
Leary, S. (2017). In defense of practical reasons for belief. Australasian Journal of Philosophy, 95(3), 529–542.
McCormick, M. S. (2015). Believing against the evidence: Agency and the ethics of belief. New York: Routledge.
McHugh, C. (2015). The illusion of exclusivity. European Journal of Philosophy, 23(4), 1117–1136.
Mitova, V. (2017). Believable evidence. Cambridge: Cambridge University Press.
Morgenbesser, S., & Ullmann-Margalit, E. (1977). Picking and choosing. Social Research, 44(4), 757–785.
Normore, C. G. (1998). Picking and choosing: Anselm and Ockham on choice. Vivarium, 36(1), 23–39.
Owens, D. (2000). Reason without freedom: The problem of epistemic normativity. London: Routledge.
Owens, D. (2002). Epistemic akrasia. The Monist, 85(3), 381–397.
Parfit, D. (1984). Reasons and persons. Oxford: Oxford University Press.
Peels, R. (2014). Believing at will is possible. Australasian Journal of Philosophy, 93(3), 1–18.
Raleigh, T. (2017). Another argument against uniqueness. The Philosophical Quarterly, 67(267), 327–346.
Raz, J. (2001). When we are ourselves. In Engaging reason. Oxford: Oxford University Press.
Reisner, A. (2009). The possibility of pragmatic reasons for belief and the wrong kind of reasons problem. Philosophical Studies, 145(2), 257–272.
Reisner, A. (2013). Leaps of knowledge. In T. Chan (Ed.), The aim of belief (pp. 167–183). Oxford: Oxford University Press.
Rescher, N. (1960). Choice without preference. A study of the history and of the logic of the problem of ‘Buridan’s Ass’. Kant-Studien, 51(1–4), 142–175.
Rinard, S. (2015). Against the new evidentialists. Philosophical Issues, 25(1), 208–223.
Rinard, S. (2017). No exception for belief. Philosophy and Phenomenological Research, 94(1), 121–143.
Schoenfield, M. (2013). Permission to believe: Why permissivism is true and what it tells us about irrelevant influences on belief. Noûs, 47(1), 193–218.
Setiya, K. (2007). Reasons without rationalism. Princeton: Princeton University Press.
Setiya, K. (2008). Believing at will. Midwest Studies in Philosophy, 32(1), 36–52.
Shah, N., & Velleman, D. (2005). Doxastic deliberation. Philosophical Review, 114(4), 497–534.
Singer, D. J. (forthcoming). How to be an epistemic consequentialist. Philosophical Quarterly.
Steglich-Petersen, A. (2013). Truth as the aim of epistemic justification. In T. Chan (Ed.), The aim of belief. Oxford: Oxford University Press.
Velleman, D. (1989). Practical reflection. Princeton: Princeton University Press.
Velleman, D. (Ed.). (2000a). On the aim of belief. In The possibility of practical reason (pp. 244–281). Oxford: Oxford University Press.
Velleman, D. (Ed.). (2000b). Epistemic freedom. In The possibility of practical reason (pp. 32–53). Oxford: Oxford University Press.
Velleman, D. (Ed.). (2000c). On the guise of the good. In The possibility of practical reason (pp. 32–53). Oxford: Oxford University Press.
Way, J. (2017). Reasons as premises of good reasoning. Pacific Philosophical Quarterly, 98(2), 251–270.
Wedgwood, R. (2002). The aim of belief. Philosophical Perspectives, 16(s16), 267–297.
Wedgwood, R. (2013). What is the right thing to believe? In T. Chan (Ed.), The aim of belief (p. 123). Oxford: Oxford University Press.
Williams, B. (Ed.). (1973). Deciding to believe. In Problems of the self (pp. 136–151). Cambridge: Cambridge University Press.
Earlier versions of this material was presented to audiences at UCLA, Oxford, Notre Dame, and The National Autonomous University of Mexico. I am grateful to everyone who commented on those occasions, and in particular to Thomas Kelley, Barbara Herman, Lee-Ann Chae, Brian Hutler, and Andrew Ball. The material has also benefitted enormously from written commentary and conversation from Pamela Hieronymi, Andrew Hsu, Tyler Burge, Calvin Normore, Eric Schwitzgebel, Ralph Wedgwood, Stephen White, Andrew Jewell, Eileen Nutting, and Adam Masters.
About this article
Cite this article
Antill, G. Epistemic freedom revisited. Synthese (2018). https://doi.org/10.1007/s11229-018-1735-6
- Aim of belief
- Self-fulfilling belief
- Epistemic norms
- Reasons for belief