Why Every Belief is a Choice: Descartes’ Doxastic Voluntarism Reconsidered

ABSTRACTDescartes appears to hold that everything we believe is the product of a voluntary choice. Scholars have been reluctant to take this particularly radical version of doxastic voluntarism as Descartes’ considered position. I argue that once Descartes’ compatibilist conception of free will as well as his position on the ‘freedom of indifference’ are taken into account, the primary motivations for the rejection of the aforementioned radical version of doxastic voluntarism lose their force. Consequently, we may take Descartes at his word when he maintains that everything we believe, we believe freely – even if we cannot believe anything we wish.KEYWORDS: Descartesdoxastic voluntarismvoluntary beliefwillassentepistemic deontologism Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1. As will be seen below, however, it is a matter of dispute whether Descartes means to be articulating a voluntarist position in the Fourth Meditation. Some scholars directly challenge this interpretation of the Fourth Meditation (O’Hear Citation1979). Others argue that Descartes endorsed the less extreme indirect version of voluntarism (Newman Citation2015; Cottingham Citation2008, ch. 11)—which, in spite of the name, amounts to a rejection of the view that belief is under the voluntary control of the will.2. I will follow that portion of the literature that refrains from drawing a distinction between assent, belief and judgment in Descartes. Della Rocca (Citation2006) makes this explicit. Schuessler (Citation2013, 150) uses the three terms interchangeably. Though Newman (Citation2015, 63) appears to understand judgment to be a discrete concept from belief and assent, he claims that the latter two notions are interchangeable. Cottingham (Citation2008, ch. 11) uses belief and assent interchangeably.3. Though Williams (Citation1978) and Alston (Citation1988) are most explicit about this, their influential interpretations have been taken up by a number of Descartes scholars. Alston, it should be noted, only mentions Descartes in passing.4. Newman (Citation2015, 65); Cottingham (Citation2008, 225-7) seems to reject this as a possibility as well. See also Grant (Citation1976).5. Vitz (Citation2010); Davies (Citation2001); Schuessler (Citation2013).6. Cottingham (Citation2008, ch. 8); Newman (Citation2015); Carriero (Citation2009, ch. 4); Kenny (Citation1973).7. Direct doxastic voluntarism can come in a variety of strengths—from holding that we may exercise direct control over some very specific subset of our beliefs to holding that we exercise direct control over all of our beliefs. Descartes, I argue, falls on the latter end of the spectrum.8. I am thinking here of the seminal attacks on doxastic voluntarism by Alston (Citation1988) and Williams (Citation1970). While there is a great deal of philosophically valuable material in both works, there are also systemic deficiencies such that even if one grants the premises of the arguments in both, the falsity (or even probable falsity) of doxastic voluntarism simply does not follow. This is mainly because of their lack of address of compatibilist conceptions of doxastic voluntarism, which, as we’ll see, Descartes endorsed. Recapitulations of such arguments tend to suffer from the same deficiencies. This is true of Buckareff’s attempt (Citation2014) to resuscitate Williams’ argument as well as Peels’ Alston-style argument (Citation2016, ch. 2).9. Descartes did not hold there is a distinction between an act being voluntary and its being free (as, for instance, Locke did (Essay II.xxi.5–11)). References to Locke, Essay, are to Locke (Citation1975), An Essay concerning Human Understanding, ed. P. H. Nidditch, cited in the text as ‘Essay‘, followed by book, chapter, section number. In his replies to Hobbes’ objections to the Meditations, Descartes claims: ‘If we simply consider ourselves, we will all realize in the light of our own experience that voluntariness and freedom are one and the same thing’ (Third Replies AT VII: 191; CSM II: 134). In other places he uses the two terms synonymously. For instance, in the Principles, ‘The supreme perfection of man is that he acts freely or voluntarily, and it is this which makes him deserve praise or blame’ (AT VIIIA: 19; CSM I: 205). The ‘or,’ here, is clearly epexegetical. For he goes on, ‘It is the supreme perfection in man that he acts voluntarily, that is, freely.’ Further on, he continues to use the two terms interchangeably. References to Descartes’ writings are to Descartes (Citation1964–76), Oeuvres de Descartes, ed. Charles Adam and Paul Tannery, 12 vols., cited as ‘AT’ followed by volume and page number, and to Descartes (Citation1984–91), The Philosophical Writings of Descartes, ed. and tr. John Cottingham, Robert Stoothoff, Dugald Murdoch (and Anthony Kenny, vol. 3), cited as ‘CSM’ for vols. 1 and 2, followed by volume and page number.10. Della Rocca (Citation2006, 153); Cottingham (Citation2008, 222). Contemporary expressions of this view are found in Steup (Citation2000, Citation2008, Citation2012, Citation2017) as well as Sharon Ryan (Citation2003, Citation2015).11. A number of authors have pointed out that this is starkly out of keeping with the scholastic conception of judgment and therefore should be viewed as one of the most interesting developments in the Fourth Meditation (Lennon Citation2014, 173; Alanen Citation2003, ch. 7). Yet, as Nuchelmans (Citation1983, 47–55) has pointed out, it is not terribly novel insofar as much the same idea was operative in stoic philosophy, which Descartes certainly admired (Rutherford Citation2014). For the stoic, judging or believing is a function of the will (E.g., Epictetus (Citation2004, 1)).12. Though this seems a rather strong way to put it, this appears to be what Descartes’ framework demands. Della Rocca, for instance, apparently interprets Descartes in precisely this fashion: ‘Descartes assimilates belief to ordinary action: believing that my stupid friends have locked the freezer and wielding an axe to break the lock are each actions in the same sense: they result from an appropriate combination of intellect and will, of representational and non-representational states’ (Citation2006, 124).13. It is not a promising strategy to suggest that Descartes meant that we have indirect control, but that he simply did not have the concepts at hand to adequately articulate this position. For proposals maintaining that the central locus of doxastic control consisted in indirect means had been articulated in scholastic philosophy at least as early as the fourteenth century by the eminent English Dominican, Robert Holcot. See Boespflug (Citation2018).14. This is Williams’ (Citation1970) way of characterizing direct control.15. Alston’s (Citation1988, 260) way of putting it is similar16. See Alston (Citation1988, 269) for a further discussion of this point.17. This is Alston’s example (Citation1988, 263).18. This is true of Alston (Citation1988, 261) and Williams (Citation1970). Plantinga (Citation1993, 23) may interpret Descartes this way as well, though it is less clear. Some of Wilson’s remarks suggest this position. In light of what she takes Descartes’ position, she demurs, ‘We can’t just decide to believe or assent to something, and forthwith believe it or assent to it’ (Citation1978, 145).19. Though Williams’ famous essay, ‘Deciding to Believe,’ is generally taken to fail in its aim to show that believing at will is metaphysically impossible, many authors hold that Williams’ claim about belief possessing a truth-aim does capture something significant about the nature of belief that is in tension with the idea that belief may be voluntary (Citation1970).20. Alston (Citation1988) takes a very different approach than Williams in arguing that doxastic voluntarism is false. Alston attempts to show that it is a mere contingent psychological fact that one cannot believe at will.21. Schuessler (Citation2013, 374–5) has noticed this. It is true that Alston is substantially more sensitive to different varieties of doxastic control than Williams, and deals with a number of them. Nevertheless, he never quite seems to address either Descartes’ view, that of the scholastics, or John Henry Newman (whom he cites).22. Yet, it is important to note that just because I cannot voluntarily believe that the U.S. is a colony of Great Britain, this does not necessarily mean that my believing that it is not a colony of Great Britain is not within my doxastic control (excuse the triple negative). After all, my inability to gouge my eye out with my pen does not mean that I lack voluntary control over not doing so. This is just to say that doxastic compulsion may be perfectly compatible with doxastic freedom or voluntary control. And some would say it certainly is. See, for instance, Steup (Citation2000, Citation2008).23. It is important to note that, even if Descartes’ voluntarism is restricted in the sense that there are some propositions that we are unable to assent to, it is unrestricted in the sense that everything we end up believing is voluntary.24. Della Rocca (Citation2006, 154); Ragland (Citation2006), unpublished manuscript; Patterson (Citation2011); Cottingham (Citation1986, 83, Citation2008, 348–9); Newman (Citation2015, 68).25. Moreover, as has been pointed out by several contemporary pragmatists, believing what is likely to be true is typically of most practical benefit to one (Rinard Citation2017; MacGuire and Wood Citation2020). Thus, the belief that is likely to be true is also probably most practically beneficial. A significant difficulty for the pragmatist position is the existence of beliefs that are the product of self-deception; these seem at once unlikely to be true, yet practically beneficial. This is an issue that requires of the pragmatist a considerable response.26. Descartes would certainly approve of Steup’s referring to this sort of position as a sort of chauvinism. ‘Let’s refer to [this] view as Chauvinism: Whereas responsiveness to practical reasons grounds freedom, responsiveness to epistemic reasons does not’ (Citation2008, 388). Locke, interestingly, holds basically the opposite position. He takes the mind’s natural proclivity toward believing what is likely to be true to exclude the voluntariness, freedom and the will’s direct involvement in general from the process of belief formation (Boespflug Citation2021).27. Newman (Citation2015, 69, 72), for instance, maintains that if one holds that clarity and distinctness determine assent, that assent cannot be voluntary.28. Della Rocca (Citation2006) and Curley (Citation1975) think similarly. There is, in addition, a substantial debate regarding whether Descartes was a compatibilist or libertarian regarding free will. Thomas Lennon (Citation2013, Citation2014) has been one of the most outspoken supporters of the compatibilist take on Descartes, though there are other supporters (e.g., Chappell Citation1994). C.P. Ragland (Citation2006, Citation2016), on the other hand, is a leading defender of interpreting Descartes as a libertarian. Other supporters of this view are Schmaltz (Citation2008) and Wee (Citation2014).29. Or any other plausible version of compatibilism for that matter.30. Cottingham (Citation2008, 355); Kenny (Citation1973, 109); Newman (Citation2014, Citation2015); Carriero (Citation2009, 60–3).31. Cottingham (Citation2008, 225–6). It should be noted that this criticism may be misplaced, given that Cottingham is ambiguous about what he means by Descartes being a doxastic determinist. In certain places he seems to be referring solely to clear and distinct perceptions while in others to all cases of belief formation. His connecting doxastic determinism to involuntarism—as well as his more general insistence that indirect control is the primary kind of doxastic control we have—suggests, however, that he has something fully general in mind. Newman (Citation2015), for instance, is fairly clear that we should interpret Descartes as holding that our assent is determined in that we must acquire a certain doxastic attitude once we direct our attention in certain ways, even if the directing of our attention is within our voluntary control Locke holds something close to the view being described: ‘It depends not on [Man’s] Will to see that Black, which appears Yellow; nor to persuade himself, that what actually scalds him, feels cold: The Earth will not appear painted with Flowers, nor Fields covered with Verdure, whenever he has a Mind to it: in the cold Winter, he cannot help seeing it white and hoary, if he will look abroad. Just thus is it with our Understanding…. As far as Men’s Senses are conversant about external Objects, the Mind cannot but receive those Ideas, which are presented by them, and be informed of the Existence of Things without…. For what a Man sees, he cannot but see; and what he perceives, he cannot but know that he perceives.’ (Essay IV.xiii.2)32. Curley articulates what seems to be a standard interpretation, namely, that the freedom of indifference has a broad scope: ‘Where the perception of the intellect is not clear and distinct, then the will is always more or less indifferent to what is proposed to it and therefore able to choose either’ (Citation1993, 165–6). So, it would seem that all perceptions that fail to be clear and distinct would have an element of indifference. Ragland adds that this sort of indifference is most fundamentally motivational (and, consequently, such situations may furnish a two-way power): ‘Humans who act from indifference are subject to a kind of internal motivational division or conflict rooted in their inability to see clearly’ (unpublished manuscript, 13).33. Principles I.45; AT VIIIa: 22; CSM I: 207–8. Mersenne (Second Objections AT VII: 126–127; CSM II: 90) claims the same thing in his objections, and Descartes (Second Replies AT VII: 149; CSM II: 106) expresses no disagreement.34. While there are reasons to think that the scope of indifference is somewhat narrower than Curley suggests, it is quite difficult to say what exactly it is. For Descartes periodically identifies certain propositions that are not clear and distinct but that, nonetheless, he seems compelled to believe: for example, those things that ‘are quite impossible to doubt’ in the First Meditation, (AT VII: 18; CSM II: 17) and those that produce ‘a great propensity to believe’ in the Sixth (AT VII: 80-81; CSM II: 55–6). The examples Descartes gives are that material objects exist, that he has a body, and that it is as though he is intermingle (quasi permixtum) with that body. Jeremy Hyman offers a different interpretation of this passage, maintaining that Descartes’ reference to the ‘great propensity to believe’ means that such propositions are actually clear and distinct (Citation2019). In any case, whether there is clarity and distinctness on the scene or not, it seems that it would be misleading to maintain that the will is indifferent in such cases. What’s more, surely the freedom I have in believing, for instance, that I am intermingled with my body is at least not quite the lowest grade. My own sense is that we should regard perfect indifference and clarity and distinctness as occupying two ends of a normative, and correspondingly psychological, spectrum: in paradigm cases of clear and distinct perceptions, it is completely obvious which option I ought to choose—and psychological certainty is generated—while in paradigm cases of indifference, the options are on an absolute par—and perhaps suspension of assent is the psychological result. In these latter sorts of situations, we are as it were like Buridan’s ass between two piles of hay. But since clarity and distinctness as well as indifference come in degrees for Descartes, it is difficult to say where the one ends and the other begins. Yet, what does seem clear is that Descartes gives us compelling reason to think that he did not endorse doxastic determinism.35. Ragland offers an insightful discussion of both related problems for Lex Newman’s (Citation2015) indirect voluntarist reading of Descartes as well as some potential solutions.36. Especially Audi (Citation2001).37. Davies, (Citation2001, ch. 5); Vitz (Citation2010); Schuessler (Citation2013). It should be noted that Schuessler’s treatment adds a number of sophisticated nuances to Descartes’ conception of doxastic control. For instance, he maintains that the doxastic control involved in the First Meditation is merely indirect, having been generated by ‘skeptical techniques.’ Schuessler, likewise, isolates other sorts of control operative in other dimensions of Descartes’ philosophy (173). But he ultimately claims that ‘We may thus confidently saddle Descartes with a gatekeeper model of doxastic control’ (157).38. Vitz (Citation2010) offers a similar picture: Descartes held ‘the thesis that people have the ability to suspend, or to withhold, judgment directly by an act of will. Call this negative DDV [direct doxastic voluntarism]’ (108). Vitz also maintains that there is little textual evidence that Descartes held ‘that people have the ability to form a judgment directly by an act of will. Call this positive DDV’ (108). So, the will plays an essentially negative role. While it is able to insert itself into the process of belief formation by putting a halt to acquiring beliefs, it is not correspondingly able to cause beliefs to be formed.39. Though I allow that it is unclear whether Descartes holds that one has the power to suspend judgments regarding clear and distinct perceptions.40. Moreover, with the exception of Schuessler (Citation2013), the above authors could be substantially more sensitive to the indirect means that may be employed in generating the skeptical frame of mind in, for instance, the First Meditation.41. In spite of Vitz’s claim that there is little textual evidence that Descartes held ‘that people have the ability to form a judgment directly by an act of will’ (Citation2010, 115–6), Vitz says little in service of closing off the possibility that this straightforward reading is how Descartes ought to be interpreted.42. Esp. Vitz (Citation2010, section II).43. Vitz (Citation2010, section II).44. This article benefitted from discussions with Robert Pasnau, Matthias Steup, Scott Ragland, and Dan Kaufman. A special thanks also to the editors of this volume—Ruth Boeker and Graham Clay—and to two anonymous reviewers for a number of insightful recommendations.