Skip to main content

How We Choose Our Beliefs


Recent years have seen increasing attacks on the "deontological" conception (or as we call it, the guidance conception) of epistemic justification, the view that epistemology offers advice to knowers in forming beliefs responsibly. Critics challenge an important presupposition of the guidance conception: doxastic voluntarism, the view that we choose our beliefs. We assume that epistemic guidance is indispensable, and seek to answer objections to doxastic voluntarism, most prominently William Alston's. We contend that Alston falsely assumes that choice of belief requires the assent to a specific propositional content. We argue that beliefs can be chosen under descriptions which do not specify their propositional content, but instead specify the mental actions by which they are formed and maintained. We argue that these actions partially constitute the beliefs and that is it in virtue of resulting from and being partially constituted by such actions that the beliefs are subject to epistemic appraisal.

This is a preview of subscription content, access via your institution.


  1. The guidance conception is sometimes referred to as “deontological” conception of justification (Alston 1988b, Plantinga 1992, Goldman 1999). The “deontological conception” is misleadingly named because epistemic guidance need not be deontological in the sense of that term familiar from ethics. On the distinction between deontological vs. teleological conceptions of epistemic justification, see Zagzebski (1996: 32–47).

  2. Foley (1999, 964)

  3. Traditionally internalism was motivated by the guidance conception of justification, so it arguably depends on voluntarism as well (Goldman 1999, 272–274). For this reason, opponents of voluntarism typically embrace some form of externalism (Alston 1988a; Plantinga 1993), or in the case of Conee and Feldman (2001), Feldman (2001, 2008), “mentalism”—a position that he describes as “internalist” though it is severed from the guidance conception. (See footnote 5.)

  4. For example, a popular book on finance tells its readers that “once a stock reaches your price target, unless you get new information […] you shouldn’t think the stock is going much higher” (Cramer and Mason (2006, 59)). In an essay about lesbianism, a teacher reports telling her students that “you shouldn’t think that people are bad just because they’re gay (Gillespie (1999: 171)). And a character in a 19th Century serial who wants his daughter to find a husband tells his friend, “That can’t be you, Duke […] I was wrong to think it could. One can’t dispose of other people’s hearts or indulge in cut-and-dried schemes for their futures” (Cook (1874: 159)). In each of these cases, a belief is censured for running afoul of some principle that the author thinks should guide people in the beliefs they accept about the relevant subject matter.

  5. Feldman (2001, 2008) argues that one is subject to certain obligations (“role ‘oughts’”) insofar as one fills a role, regardless of whether it is within one's power to discharge these obligations. One example Feldman gives is that teachers ought to clarify, and a teacher can be blamed for failing to do this, even if success is impossible to her (and even if the teacher has no choice about whether to be a teacher). Feldman may be right that there is a sense in which words like “ought” and “blame” can be applied in such cases, but, if so, they do not express prescriptions the point of which is to guide agents in action. Feldman (2008) is right that “role ‘oughts’” provide norms on the same order as those concerning ways we ought to chew and breathe, even if we do not choose our roles as a chewer or breather. But these norms are akin to those describing how we “ought” to circulate our blood or conduct cellular metabolism. Since the “role ‘oughts’” governing teachers are akin to the norms that apply to involuntary physiological processes, they are clearly not the sort of norms that provide guidance. In addition to these “oughts,” teachers clearly do need norms to guide them in fulfilling their role: indeed, the whole discipline of pedagogy is concerned with such norms, and myriad books and courses that purport to provide them. So Feldman’s examples of normative words being used in connection with things over which people have no choice, involve a different sense of these word than the one in which they are used both by traditional epistemologists and by the non-philosophers we quoted in the preceding note.

  6. According to Heil (1983) and Audi (2001), epistemology recommends certain procedures of gathering evidence, such as paying attention to logic, careful observation, identifying and scrutinizing the data on which a conclusion is based, etc., and these norms apply indirectly to belief states. See also Hilary Kornblith (1983) for a similar approach.

  7. Hieronymi (2006) distinguishes between “believing at will” and the voluntariness of our beliefs, conceding to Alston and Williams that we do not believe at will, but maintaining that our beliefs are nonetheless voluntary. Steup (2012) goes further and criticizes Alston’s arguments directly, drawing on a compatibilist account of freedom. For other compatibilist accounts, see also Ryan (2003) and Jäger (2004), and Steup (2008). We reject this position because we think that compatibilism is subject to insuperable objections that are intensified when applied to doxastic freedom. See Bayer (unpublished [b]) for a critique of the reasons-responsiveness account of compatibilism, which Steup applies to doxastic freedom.

  8. We focus on Alston rather than Williams, because it is Alston who connects the question of belief choice to the issue of epistemic norms. Williams is concerned with the narrower question of whether it is possible to self-consciously adopt a belief for pragmatic reasons. Doing so would require choosing the belief under the description that includes the belief’s content, and we agree that beliefs are not (ordinarily) chosen in this manner.

  9. Feldman (2001) later points out that if the proposition in question is “The switch has been flipped,” then one can quite literally choose to flip a switch in order to choose a belief.

  10. This is a point originally suggested by Chisholm (1968).

  11. A similar point is made by Feldman (2008: 345) against a proposal raised by Kelly (2002). Hieronymi (2006) urges that by positively considering evidence that bears on p, we therein believe that p, and she argues that this constitutes a form of “evaluative control” that we exercise over our beliefs. Still, Hieronymi agrees with Alston that this does not count as a form of believing at will because she thinks that acting at will involves an immediate responsiveness to practical reasons that evaluative control does not involve. Central to her argument is the assumption that no practical reason can be a reason whose consideration constitutes believing. But, as Bayer (unpublished [a]) has shown, this last assumption is mistaken. The attitude one adopts toward a proposition is an expression of one’s attitude towards the practical value of believing the truth.

  12. Likewise, Hieronymi (2006) has distinguished believing at will from what she calls “managerial control,” or the process acquiring a belief by placing oneself in a situation in which one is likely to form it. Unlike evaluative control, exercising this kind of control is responsive to practical reasons, but it counts merely as a kind of voluntarily bringing about a belief, not an act of believing at will.

  13. Indirect influence corresponds to what Pojman (1985) calls believing willingly, which he distinguishes from believing at will.

  14. This last point, which is important, is obscured in Alston’s example because we are told nothing about the employer’s reason for writing the program—he is treated as some sort of eccentric; but in fact there are many jobs that are essentially like the one Alston describes. For example, the person who sits behind the lighting console in a Broadway theatre presses buttons on the console at certain points during the play; the buttons initiate programs that turn on and off dozens of lights and alter their brightness. These programs were written by the lighting designer, and the person behind the console likely doesn’t know the effect of his action on each individual light. Nonetheless, he knows that if he doesn’t press the buttons when he should, the lighting will be wrong, and that it will be his fault if, because of this, the lead has to deliver a soliloquy in a shadow.

  15. Notice that the outcome of the choice is not the primary object of epistemological evaluation. This might be obscured by the fact that we can evaluate the subject’s beliefs merely on the grounds of their truth or falsehood. In the case of the servant employed to actuate the program opening and closing doors, the analogy to a well-supported but false belief would be a situation in which the servant runs the program, but (owing to a mistake in the instructions or a bug in the program) the doors end up in an undesirable configuration and the dog escapes. The dog’s escape, like the false belief, is bad. But this is irrelevant to our assessment of the servant as an employee. He did his job impeccably and (let us suppose) could not have prevented the escape. Similarly in the case of the well-supported but false belief, if the subject did his “job” as a thinker impeccably, he is not responsible for the false belief that he could not have prevented (or could not have prevented without taking special precautions that would have been unwarranted).

  16. In cases in which we do not make these choices with this goal in mind, we may form an unjustified belief.

  17. On certain theories, such mental states reduce to attitudes towards propositions. Though we disagree with such views, nothing here turns on the question of whether all content is propositional and all contentful states are propositional attitudes. Our point is simply that a mental state’s identity is not exhausted by its content. This view is compatible with a variety of views of the nature of states and their content.

  18. This position should not be confused with a pragmatism on which the content of the belief is somehow constituted by the activities in which the believing consists. For all we have said the content may exist mind-independently, and indeed (for the most part) we think that it does. Our point is only that the state of believing this content—of standing in a certain relation to it—is a complex one and is (at least partially) constituted by actions taken over time. The state of belief is like that of balancing a stick, which involves a succession of small actions. These actions are partially constitutive of the balancing, but (of course) they are in no way constitutive of the stick (which is analogous to the content of the belief.)

  19. For the same reason, Heil’s and Audi’s distinction between states of belief and acts of belief-formation is artificial.

  20. Clifford (1901, 168–69) nicely describes this phenomenon when he says that each belief is “part of that aggregate of beliefs which is the link between sensation and action at every moment of all our lives, and which is so organized and compacted together that no part of it can be isolated from the rest, but every new addition modifies the structure of the whole. No real belief, however trifling and fragmentary it may seem, is ever truly insignificant; it prepares us to receive more of its like, confirms those which resembled it before, and weakens others; and so gradually it lays a stealthy train in our inmost thoughts, which may some day explode into overt action, and leave its stamp upon our character forever.” Such considerations about the interconnectedness of our beliefs are often cited in support of holism about semantic content. However, one can acknowledge the force of these considerations (as we do) without believing in semantic holism, so long as one does not identify the content of the belief with its function.

  21. See Harry Binswanger (1992).

  22. Alston counts checking to make sure a perceptual situation is normal as a type of voluntary influence that we have over our beliefs. But we have already argued that many of these “influences” are actually constitutive of believing.

  23. In an underappreciated portion of his Essay, Locke observes that we have a choice to “employ or withhold our faculties”: each of us must choose “whether he will curiously survey” the objects of perception, and whether he will “with an intent application, endeavour to observe accurately all that is visible in it” (Locke 1979/1690, 650). More recently, research on perception, in the tradition of J.J. Gibson has sensitized us to the variety and complexity of the activities that are involved in perceiving, and some of these activities (e.g. focusing our attention on one aspect of our sensory field rather than another) seem to be (at least often) under our direct control. (See Gibson (1986), Noë (2006) on the activity of perception; see Crowther (2009) for an account of how treating perception as “agential” is nonetheless consistent with its being metaphysically passive in the way needed for it to serve as a form of evidence.) Perceptual judgments include, in addition to the act of perception itself, the application of concepts. There are arguments from many quarters that the application and formation of concepts in response to sensory data is a complex process that involves voluntary components. See Rand (1990, 29), Peikoff (1991, 111-12), Brandom (1994, 85–89, 132–136; 2000, 163–5), Sloutsky (2003), Bayer (2011), Gotthelf (2013), Salmieri (2013).

  24. See McDowell (1996, 8).

  25. Each of us has addressed this question at length elsewhere. See Bayer (2011) and Salmieri (2013).


  • Alston, W. (1988a). Internalism and externalism in epistemology. In W. Alston (Ed.), Epistemic justification: Essays in the theory of knowledge (pp. 185–226). Ithaca: Cornell University Press.

    Google Scholar 

  • Alston, W. (1988b). The deontological conception of epistemic justification. Philosophical Perspectives, 2, 257–299.

    Article  Google Scholar 

  • Audi, R. (2001). Doxastic voluntarism and the ethics of belief. In M. Steup (Ed.), Knowledge, truth and duty (pp. 93–114). New York: Oxford University Press.

    Chapter  Google Scholar 

  • Bayer, B. (2011). A role for abstractionism in a direct realist foundationalism. Synthese, 180(3), 357–389.

    Article  Google Scholar 

  • Bayer, B. (unpublished (a)). Believing at will and the will to believe the truth. Accessed 17 Jul 2013.

  • Bayer, B. (unpublished (b)). The elusiveness of doxastic compatibilism. Accessed 17 Jul 2013.

  • Binswanger, H. (1992). Volition as cognitive self-regulation. Organizational Behavior and Human Decision Processes, 50(2), 165–178.

    Google Scholar 

  • Brandom, R. (1994). Making it explicit: Reasoning, representing, & discursive commitment. Cambridge: Harvard University Press.

    Google Scholar 

  • Brandom, R. (2000). Articulating reasons: An introduction to inferentialism. Cambridge: Harvard University Press.

    Google Scholar 

  • Chisholm, R. (1968). Lewis’ ethics of belief. In P. Schlipp (Ed.), The philosophy of C.I. Lewis (pp. 223–242). La Salle: Open Court Publishing Company.

    Google Scholar 

  • Clifford, W. (1901). The ethics of belief. In Lectures and essays, Vol. II. London: MacMillan and Co..

  • Conee, E., & Feldman, R. (2001). Internalism defended. American Philosophical Quarterly, 38(1), 1–18.

    Google Scholar 

  • Cook, E. (1874). Young Mr. Nightingale (Vol. III). London: Sampson, Low, Marston, Low, and Searle.

    Google Scholar 

  • Cramer, J., & Mason, C. (2006). Mad money: Watch TV, get rich. New York: Simon and Schuster.

    Google Scholar 

  • Crowther, T. (2009). Perceptual activity and the will. In L. O’Brien & M. Soteriou (Eds.), Mental actions (pp. 173–191). New York: Oxford University Press.

    Chapter  Google Scholar 

  • Feldman, R. (2001). Voluntary belief and epistemic evaluation. In M. Steup (Ed.), Knowledge, truth and duty (pp. 77–92). New York: Oxford University Press.

    Chapter  Google Scholar 

  • Feldman, R. (2008). Modest deontologism in epistemology. Synthese, 161, 339–355.

    Article  Google Scholar 

  • Foley, R. (1999). Voluntarism. In R. Audi (Ed.), The Cambridge dictionary of philosophy (2nd ed., p. 964). New York: Cambridge University Press.

    Google Scholar 

  • Gibson, J. (1986). The ecological approach to visual perception. Hillsdale: Lawrence Erlbaum Associates.

    Google Scholar 

  • Gillespie, P. (Ed.). (1999). Love makes a family: Portraits of lesbian, gay, bisexual, and transgender parents and their families. Amherst: University of Massachussets Press.

    Google Scholar 

  • Goldman, A. (1999). Internalism exposed. Journal of Philosophy, 96(6), 271–293.

    Article  Google Scholar 

  • Gotthelf, A. (2013). Ayn Rand’s theory of concepts: Rethinking abstraction and essence. In A. Gotthelf & J. Lennox (Eds.), Concepts and their role in knowledge: Reflections on Objectivist epistemology. Pittsburgh: University of Pittsburgh Press.

    Google Scholar 

  • Heil, J. (1983). Doxastic agency. Philosophical Studies, 43, 355–364.

    Article  Google Scholar 

  • Hieronymi, P. (2006). Controlling attitudes. Pacific Philosophical Quarterly, 87, 45–74.

    Article  Google Scholar 

  • Jäger, C. (2004). Epistemic deontology, doxastic voluntarism, and the principle of alternate possibilities. In W. Löffler & P. Weingartner (Eds.), Knowledge and belief: Proceedings of the 26th international Wittgenstein symposium. Wien: ÖBV & HPT.

  • Kelly, T. (2002). The rationality of belief and some other propositional attitudes. Philosophical Studies, 110, 163–196.

    Article  Google Scholar 

  • Kornblith, H. (1983). Justified belief and epistemically responsible action. Philosophical Review, 92(1), 33–48.

    Article  Google Scholar 

  • Locke, J. (1979/1690). An essay concerning human understanding. (P. H. Nidditch, Ed.) Oxford: Clarendon Press.

  • Noë, A. (2006). Action in perception. Cambridge: MIT Press.

    Google Scholar 

  • McDowell, J. (1996). Mind and World. Cambridge: Harvard University Press.

  • Peikoff, L. (1991). Objectivism: The Philosophy of Ayn Rand. New York: Dutton.

  • Plantinga, A. (1992). Justification in the 20th Century. Philosophical Issues, 2, 43–77.

    Article  Google Scholar 

  • Plantinga, A. (1993). Warrant: The current debate. New York: Oxford University Press.

    Book  Google Scholar 

  • Pojman, L. (1985). Believing and willing. Canadian Journal of Philosophy, 15(1), 37–55.

    Google Scholar 

  • Rand, A. (1990). Introduction to Objectivist epistemology (2nd ed.). In. H. Binswanger, & L. Peikoff, (Eds.) New York: Meridian.

  • Ryan, S. (2003). Doxastic compatibilism and the ethics of belief. Philosophical Studies, 114, 47–79.

    Article  Google Scholar 

  • Salmieri, G. (2013). Conceptualization and justification. In A. Gotthelf & J. Lennox (Eds.), Concepts and their role in knowledge: Reflections on Objectivist epistemology. Pittsburgh: University of Pittsburgh Press.

    Google Scholar 

  • Sloutsky, V. (2003). The role of similarity in the development of categorization. Trends in Cognitive Sciences, 7(6), 246–251.

    Article  Google Scholar 

  • Steup, M. (2008). Doxastic freedom. Synthese, 161, 375–392.

    Article  Google Scholar 

  • Steup, M. (2012). Belief control and intentionality. Synthese, 188, 145–163.

    Google Scholar 

  • Williams, B. (1973). Deciding to believe. In B. Williams (Ed.), Problems of the self (pp. 136–151). Cambridge: Cambridge University Press.

    Chapter  Google Scholar 

  • Zagzebski, L. (1996). Virtues of the mind. New York: Cambridge University Press.

    Book  Google Scholar 

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Benjamin Bayer.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Salmieri, G., Bayer, B. How We Choose Our Beliefs. Philosophia 42, 41–53 (2014).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Doxastic voluntarism
  • Belief
  • Mental states
  • Mental actions
  • Epistemic normativity
  • Guidance conception of epistemology
  • Epistemic deontologism