Abstract
Akratic belief, or belief one believes one should not have, has often been thought to be impossible. I argue that the possibility of akratic belief should be accepted as a pre-theoretical datum. I distinguish intuitive, defensive, systematic, and diagnostic ways of arguing for this view, and offer an argument that combines them. After offering intuitive examples of akratic belief, I defend those examples against a common argument against the possibility of akratic belief, which I call the Nullification Argument. I then offer an Argument from Belief Attribution, using a discussion of the marks by which we typically attribute belief to defend attributions of akratic belief. I conclude by offering a way to explain what is puzzling about akratic belief, while allowing that it is possible.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
I will not repeatedly distinguish an “epistemic” and an “all-things-considered” “should”; for this distinction in the context of akratic belief, see Mele (1987, pp. 110–113). I have in mind an epistemic “should”; but what I say can be adapted to beliefs that use “should” in other senses. I believe what follows can also be adapted to closely related phenomena such as alienated belief (Hunter, 2011) and “naughty” beliefs, or beliefs one believes are false (Huddleston, 2012), but I will not argue this in the text.
The anorexic example, though troubling, would still be a relatively moderate case of anorexia. In many cases, an anorexic can be unable to hear a doctor’s advice, and can even refuse life-saving treatment. Bowden (2012), Arnold (2013), and Dunkle and Dunkle (2015) provide striking accounts, and Giordano (2005, p. 18) treats complete acceptance of the impulse to be thin as essential to full-blown anorexia. It may be more accurate to imagine the akratic “anorexic” as a borderline or partly recovered anorexic, though I will leave out this qualification. Though I will mostly use feminine pronouns as default pronouns, I changed the gender in the anorexic example to avoid perpetuating the perception that all anorexics are female.
See Elga (2005) and Greco (2012), respectively. On the gambler’s fallacy, see Tversky and Kahneman (1971). Some examples are of course controversial; but the range of apparent examples is wide. Just about anything one can believe, it seems, is something one can at the same time believe one should not believe.
On impossibility arguments, see n.4 above. For defenses of the possibility of akratic belief, see Rorty (1983), Heil (1984), Mele (1986, 1987, Ch. 8), Scanlon (1998, p. 35), Levy (2004), Tenenbaum (2007, Ch. 7), Borgoni (2015), and Chislenko (2016), and for a helpful, related defense of the possibility of beliefs we believe are false, see Huddleston (2012). I discuss many of these below. To my knowledge, none of these offer an extended discussion of belief attribution, puzzlement about akratic belief, or the status of akratic belief as a pre-theoretical datum. Still, Borgoni (2015) is particularly helpful, including a discussion of self-control and of kinds of akratic belief.
See Sect. 4 below.
The tendency to assume that akratic belief is possible, notable in discussions of the rationality of akratic belief, can itself be taken as a basis for a further argument that the possibility of akratic belief is intuitive. But I think the power of this particular argument is limited. It appeals to a relatively small statistical sample that is partly self-selecting, since skeptics about akratic belief are unlikely to debate its rationality. And some philosophers may be assuming a particular conception of belief, rather than treating akratic belief as a pre-theoretical datum.
The division into argument types is somewhat artificial, since examples and defense play a role in later sections. Rather than keep the arguments clearly separate, my aim is to combine them into a more powerful argument for the possibility of akratic belief.
Owens here relies on a variant of premise (1), but includes the caveat “freely and deliberately,” which makes his denial more limited. I return to such restrictions on what counts as akratic belief in Sect. 4.
One might think the argument can be made more compelling by revising the premises. For example, premise (1) could be changed to read: “To believe that p, a person must not believe there is conclusive reason against believing that p.” This would remove the regress. But it would also be more baldly question-begging, in the way I go on to argue that the Nullification Argument is. Whether we can believe against our own better—or in this case, conclusive—judgment is precisely what is at issue. Alternatively, one might try to make the Nullification Argument more plausible by appealing to a distinction between explicit and implicit, or conscious and unconscious, belief. Premise (1) may seem more plausible if it claims only that, to believe that p, a person must implicitly, or perhaps unconsciously, believe there is reason to believe that p. I think this does make premise (1) more plausible, but at the cost of making premise (2) even less plausible when the corresponding change is made there. It is especially hard to see why, when someone believes there is conclusive reason to believe not p, she cannot even implicitly or unconsciously believe she has reason to believe that p. This claim seems to idealize ordinary believers, in addition to raising the other problems I mention in the text.
On other impossibility arguments and responses to them, see nn. 4 and 6 above.
One might instead attempt an argument that appeals directly to marks of akratic belief, rather than marks of belief more generally. I do not rule out the possibility of such an argument, but I have two hesitations. First, I am not sure what these marks of akrasia would be, since features like inner conflict are also characteristic of related phenomena such as indecision and contradictory belief. Second, and relatedly, I am not sure the use of such marks is robust or widespread enough to support a convincing argument for the possibility of akratic belief, capable of dispelling doubts that these marks are proper bases for attributing akrasia. By contrast, it is hard to deny that beliefs more generally are often attributed based on some combination of sensitivity to evidence, recall, conviction, reporting or assertion, and further reasoning. So I find it more compelling to use the argumentative structure of applying widely used marks of belief to the specific case of akrasia.
Discussions of belief that emphasize responsiveness to evidence or reasons include Van Leeuwen (2014), Ichino (2019), and Ganapini (2020); discussions that emphasize assent include Cherniak (1986) and Kaplan (1996); discussions that emphasize availability for further reasoning include Stich (1978), Glüer and Wikforss (2013), and Mandelbaum (2014). Other views combine two or more marks, or remain uncommittal. Quine and Ullian (1970, pp. 3–4), for example, write: “Believing…is a disposition…to respond in certain ways when the appropriate issue arises. To believe that Hannibal crossed the Alps is to be disposed, among other things, to say ‘Yes’ when asked.” This passage focuses on reporting one’s belief or asserting its content, but may also have in mind other responses, including further reasoning. Scanlon (1998), discussed in the body text below, suggests a similar view. Some writers explicitly downplay the importance of particular marks; for an influential argument against treating language use as essential in attributing belief, see Marcus (1990). I make no claim that my list of marks is exhaustive; other features of belief, such as resistance to reconsideration or additional causal or functional roles, may be marks as well. I mean only to identify several fairly widespread and uncontroversial marks by which belief is often attributed.
There is a growing body of psychological research on these forms of belief attribution, as well as belief attribution in particular populations such as young children and autistic people, who at least often lack an explicit theoretical conception of the nature of belief. For a recent overview, see Curry (2018). Empirical findings provide further evidence for premise (1) in the Argument from Belief Attribution, and for the relative importance of particular marks. Rather than reviewing these findings, I will focus on premise (2) as the one in most need of defense.
Scanlon is interested in features of belief itself, whereas for our purposes, it is enough to focus on these characteristics merely as marks by which belief can be attributed. I return to this difference below.
For a useful recent survey of these and other views, see Schwitzgebel (2019). The Argument from Belief Attribution is consistent with a wide range of theories, but not with all of them; I consider conflicting theories in the next section.
The Argument from Belief Attribution is also consistent with a view of belief as requiring an ability to revise one’s beliefs in response to evidence. We can have difficulty exercising this ability, and we can sometimes fail; as Helton (2018, p. 23n24) puts it, “When this ability is masked, subjects can have conflicting beliefs.” The argument is also consistent with a view of belief in general as tending toward coherence, such that “failures to re-establish coherence are accompanied by a sense of distress” (Ganapini, 2020, p. 3273). It can be distressing—rather than impossible—to believe both that Jones is a loyal friend, and that one should not believe this.
Schwitzgebel’s focus is on attribution of contradictory beliefs, rather than akratic ones; I believe the response I give here can be applied in that context as well, but I leave that aside. For a response to Schwitzgebel, including a discussion of self-reports of belief in the cases he has in mind, see Borgoni (2016).
Thus, for example, we can agree with Ichino (2020, p. 216) that “many instances of superstitious attitudes are non-doxastic in nature”—just not all such instances, and in particular, not the ones that do not involve what Risen (2016) and Ichino (2020) call “acquiescence.” For helpful discussions of conflicting beliefs, see Huddleston (2012) and Helton (2018). For a more detailed discussion and defense of the possibility of evidence-resistant beliefs, see Viedge (2018). Akratic belief becomes easier to attribute if a set of beliefs is relatively “fragmented”; on belief fragmentation, see Bendana and Mandelbaum (forthcoming). To keep my claims modest, I avoid assuming any particular view about the fragmentation of belief.
Ganapini (2020, p. 3273), quoted earlier, emphasizes “a sense of distress” in cases of integration failure as a mark of belief. I focus here on partial inhibition of inferential integration, rather than on felt distress, since I think phenomena such as felt distress or cognitive dissonance are also characteristic of other, non-doxastic conflict.
Such statements will immediately strike some as similar to Moore-paradoxical assertions or beliefs such as “It’s raining, but I don’t believe it”. For discussion of this similarity, see Chislenko (2016).
For helpful discussions of self-ignorance and of mistaken meta-belief, see Schwitzgebel (2011) and Levy (2018), respectively. I take myself to agree with both of them, since neither claims that attributions of akratic belief are always mistaken; but Schwitzgebel (2010) can be extended to deny the possibility of akratic belief.
One remaining motivation for this insistence can be traced to the desire to interpret believers charitably. On some principles of charity, as Davidson (1984, p. 27) puts it, “we must maximize the self-consistency we attribute,” being careful to avoid attributing inconsistency whenever possible. Charity can seem to rule out attributing akratic belief. Although charity of interpretation deserves a longer treatment that I can give it here, it is worth briefly mentioning four points in reply. First, akratic belief is not strictly inconsistent; a principle of charity would need some further rationale for its extension to akratic belief. (One might then also want to know whether the principle rules out attributing akratic action.) Second, a principle of charity that rules out all inconsistency or conflict is implausibly restrictive, and seems, at least intuitively, to emphasize charity over accuracy; surely humans are at least occasionally inconsistent. Third, acceptance of such a restrictive principle of charity would likely have to be supported by some theoretical considerations. Akratic belief could then still be a pre-theoretical datum, which theoretical considerations might convince us to deny. Finally, in some cases, attributing akratic belief may be more charitable than the alternative. When Calvin tells many people that he needs to lose weight, insists that he should not believe it, and undertakes elaborate measures to undermine his own weight loss schemes, it may be more generous to see him as harboring one inconsistency—the akratic belief that he is fat—rather than hundreds of inconsistent or inexplicable judgments or perhaps no belief at all. In any case, we once again need good theoretical reason to accept a principle that entails that belief could never be akratic.
In this way, the Argument from Belief Attribution has a potential application to contradictory belief, and even to clear-eyed contradictory belief. Priest (2006, pp. 96–97), for example, writes: “I…believe that the Russell set is both a member of itself and not a member of itself…. Many arguments, most of which appear in this book, convinced me of it.” Priest reports his contradictory belief, gives extensive evidence for it, reasons from it, and seems quite sure of it. An extension of the Argument from Belief Attribution to contradictory belief would encourage us to take his self-report at face value. Still, contradictory belief may fare differently both with respect to theoretical considerations for denying its possibility, and with respect to intuitive, defensive, systematic, and diagnostic arguments for treating its possibility as a pre-theoretical datum. It thus needs a separate treatment.
See Ganapini (2020) for a recent argument against a closely related view, which she calls “Strong Traditionalism” (3265). Ganapini’s own view is that “beliefs are minimally rational in the sense that they respond to perceived irrationality by re-establishing coherence” (3272), and that “clear failures to re-establish coherence are accompanied by a sense of distress” (3273). This view allows akratic belief that is not clear-eyed, or not perceived as akratic, and clear-eyed akratic belief accompanied by a sense of distress, though it may rule out the possibility of non-distressed clear-eyed akratic belief. Other theories may deny that some particular examples are akratic, while accommodating my main concern: allowing that akratic belief is possible in principle. On denials that any beliefs exist at all, see Ramsey (2020).
Defending Bortolotti (2009), Bortolotti (2012, p. 39) encourages discussions of delusions and of beliefs more generally to follow recent psychological literature in accepting that “beliefs are often badly integrated with other beliefs, unsupported by evidence, resistant to change, and behaviourally inefficacious.” My argument is broadly in line with hers, though I do not take a stand on delusions in particular.
This discussion of our puzzlement is not meant as an argument for any conclusion about the rationality of akratic belief, or as a characterization of ways in which an akratic believer might herself experience puzzlement. I mean only to suggest some ways in which one might naturally find akratic belief puzzling, while resisting the conclusion that it is impossible.
For discussion of the relative “ease” of akratic action, see Pears (1982). Since Pears proceeds by considering particular theories of action, I leave out the details here. But it is interesting to note that Pears combines, and perhaps confuses, intuitions of impossibility and of difficulty when he writes that motivated, “full-blown” akratic belief is “scarcely” or “only marginally possible” (pp. 44, 46, 49, 50).
References
Adler, J. (2002a). Akratic believing? Philosophical Studies, 110, 1–27.
Adler, J. (2002b). Belief’s own ethics. MIT Press.
Arnold, C. (2013). Decoding anorexia. Routledge.
Aristotle. (1999). Nicomachean ethics, tr. Terence Irwin. Hackett.
Bendana, J., & Mandelbaum, E. (forthcoming). The fragmentation of belief. In C. Borgoni, D. Kindermann, & A. Onofri (Eds.), The fragmented mind. Oxford University Press.
Borgoni, C. (2015). Epistemic akrasia and mental agency. Review of Philosophy and Psychology, 6, 827–842.
Borgoni, C. (2016). Dissonance and irrationality: A criticism of the in-between account of dissonance cases. Pacific Philosophical Quarterly, 97, 48–57.
Bortolotti, L. (2009). Delusions and other irrational beliefs. Oxford University Press.
Bortolotti, L. (2012). In defence of modest doxasticism about delusions. Neuroethics, 5, 39–53.
Bowden, H. (2012). A phenomenological study of anorexia nervosa. Philosophy, Psychiatry, & Psychology, 19, 227–241.
Cherniak, C. (1986). Minimal rationality. The MIT Press.
Chislenko, E. (2016). Moore’s paradox and akratic belief. Philosophy and Phenomenological Research, 92, 669–690.
Cohen, L. J. (1992). An essay on belief and acceptance. Oxford University Press.
Christensen, D. (2021). Akratic (epistemic) modesty. Philosophical Studies, 178, 2191–2214.
Curry, D. S. (2018). Beliefs as inner causes: The (lack of) evidence. Philosophical Psychology, 31, 850–877.
Davidson, D. (1984). Truth and meaning. In D. Davidson (Ed.), Inquiries into truth and interpretation (pp. 17–36). Clarendon Press.
De Sousa, R. (1971). How to give a piece of your mind: Or, the logic of belief and assent. The Review of Metaphysics, 25, 52–79.
Dretske, F. (1971). Reasons, knowledge, and probability. Philosophy of Science, 38, 216–220.
Dunkle, E., & Dunkle, C. B. (2015). Elena vanishing. Chronicle Books.
Elga, A. (2005). On overrating oneself—And knowing it. Philosophical Studies, 123, 115–124.
Ganapini, M. (2020). Belief’s minimal rationality. Philosophical Studies, 177, 3263–3282.
Giordano, S. (2005). Understanding eating disorders. Oxford University Press.
Glüer, K., & Wikforss, Å. (2013). Aiming at truth: On the role of belief. Teorema, 32, 137–162.
Greco, D. (2012). The impossibility of skepticism. The Philosophical Review, 121, 317–358.
Greco, D. (2014). A puzzle about epistemic akrasia. Philosophical Studies, 167, 201–219.
Hare, R. M. (1963). Freedom and reason. Clarendon Press.
Heil, J. (1984). Doxastic incontinence. Mind, 93, 56–70.
Helton, G. (2018). If you can’t change what you believe, you don’t believe it. Noûs, 17, 1–26.
Hieronymi, P. (2006). Controlling attitudes. Pacific Philosophical Quarterly, 87, 45–74.
Horowitz, S. (2014). Epistemic akrasia. Noûs, 48, 718–744.
Huddleston, A. (2012). Naughty beliefs. Philosophical Studies, 160, 209–222.
Hume, D. (2000). A treatise of human nature. Oxford University Press.
Hunter, D. (2011). Alienated belief. Dialectica, 65(2), 221–240.
Hurley, S. (1989). Natural reasons. Oxford University Press.
Ichino, A. (2019). Imagination and belief in action. Philosophia, 47, 1517–1534.
Ichino, A. (2020). Superstitious confabulations. Topoi, 39, 203–217.
Kaplan, M. (1996). Decision theory as philosophy. Cambridge University Press.
Lasonen-Aarnio, M. (2020). Enkrasia or evidentialism? Learning to love mismatch. Philosophical Studies, 177, 597–632.
Levy, N. (2004). Epistemic akrasia and the subsumption of evidence: A reconsideration. Croatian Journal of Philosophy, 4, 149–156.
Levy, N. (2018). You meta believe it. European Journal of Philosophy, 26, 814–826.
Mandelbaum, E. (2014). Thinking is believing. Inquiry, 57, 55–96.
Marcus, R. B. (1990). Some revisionary proposals about belief and believing. Philosophy and Phenomenological Research, 50, 132–153.
Mele, A. (1986). Incontinent believing. The Philosophical Quarterly, 36, 212–222.
Mele, A. (1987). Irrationality. Oxford University Press.
Owens, D. (2002). Epistemic akrasia. The Monist, 85, 381–397.
Pears, D. (1982). How easy is akrasia? Philosophia, 11, 33–50.
Priest, G. (2006). In contradiction (2nd ed.). Clarendon Press.
Quine, W. V. O., & Ullian, J. S. (1970). The web of belief. Random House.
Ramsey, W. (2020). Eliminative materialism. In E. N. Zalta (Ed.) The Stanford encyclopedia of philosophy, Summer 2020 Edition. https://plato.stanford.edu/archives/sum2020/entries/materialism-eliminative/
Raz, J. (2009). Reasons: Practical and adaptive. In D. Sobel & S. Wall (Eds.), Reasons for action (pp. 37–57). Cambridge University Press.
Risen, J. (2016). Believing what we do not believe: Acquiescence to superstitious beliefs and other powerful intuitions. Psychological Review, 123(2), 182–207.
Rorty, A. (1983). Akratic believers. American Philosophical Quarterly, 20, 175–183.
Scanlon, T. M. (1998). What we owe to each other. Harvard University Press.
Schwitzgebel, E. (2010). Acting contrary to our professed beliefs or the gulf between occurrent judgment and dispositional belief. Pacific Philosophical Quarterly, 91, 531–553.
Schwitzgebel, E. (2011). Self-ignorance. In J. L. Liu & J. Perry (Eds.), Consciousness and the self: New essays (pp. 184–197). Cambridge University Press.
Schwitzgebel, E. (2019). Belief. In E. N. Zalta (Ed.) The Stanford encyclopedia of philosophy, Summer 2019 Edition. https://plato.stanford.edu/archives/sum2019/entries/belief/.
Setiya, K. (2007). Reasons without rationalism. Princeton University Press.
Skipper, M., & Steglich-Petersen, A. (Eds.). (2019). Higher-order evidence: New essays. Oxford University Press.
Stich, S. (1978). Beliefs and subdoxastic states. Philosophy of Science, 45, 499–518.
Tenenbaum, S. (2007). Appearances of the good. Cambridge University Press.
Tversky, A., & Kahneman, D. (1971). Belief in the law of small numbers. Psychological Bulletin, 76(2), 105–110.
Van Leeuwen, N. (2014). Religious credence is not factual belief. Cognition, 133, 698–715.
Velleman, J. D. (1989). Practical reflection. Princeton University Press.
Viedge, N. (2018). Defending evidence-resistant beliefs. Pacific Philosophical Quarterly, 99, 517–537.
Watson, G. (1977). Skepticism about weakness of will. The Philosophical Review, 86, 316–339.
Williamson, T. (2000). Knowledge and its limits. Oxford University Press.
Acknowledgements
For helpful feedback on earlier versions of this paper, I am grateful to Lee-Ann Chae, Colin Chamberlain, Julia Chislenko, Philip Clark, Hannah Ginsborg, August Gorman, Jim Hutchinson, Mandana Kamangar, Arthur Krieger, Adam Marushak, Barry Stroud, Benjamin Wald, R. Jay Wallace, my anonymous reviewers, and audiences at the Berkeley-London Philosophy Conference, Temple University, UC Berkeley, and the Western Canadian Philosophy Association. Research on this project was supported by a Mabelle McLeod Lewis Memorial Fund fellowship, a Temple University Summer Research Award, and a Temple University College of Liberal Arts Research Award. A special thanks to Charles Goldhaber for extremely helpful feedback on several drafts.
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
About this article
Cite this article
Chislenko, E. How can belief be akratic?. Synthese 199, 13925–13948 (2021). https://doi.org/10.1007/s11229-021-03404-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-021-03404-0