2 Ibid., p. 148.

3 Williams has a second argument for this conclusion which is based on the fact that “factual beliefs can be based on evidence.” (The argument is given in “Deciding to Believe,” pp. 148–49. The foundation for this argument is laid on p. 141 in the discussion of the characteristic of being based on evidence.) I do not address that argument in this paper.

4 This locution of regarding a belief as something I take to be true, or as something one takes to be true, is Williams's. I follow him in its usage.

5 Ibid., p. 148.

6 The conclusion of this argument is stated in the first sentence of the paragraph quoted on p. 494, above: “[I]t is not a contingent fact that I cannot bring it about, just like that, that I believe something …” I have extracted the premises, i.e., (l)-(5), (8), and (10), as follows: (1) is nearly a quotation of “If I could acquire a belief at will, I could acquire it whether it was true or not; moreover I would know that I could acquire it whether it was true or not.” I have simply restated it in the third person. (2) also comes straight from the text: “[I]f I can acquire beliefs at will, I must know that I am able to do this …” (3) comes from “and could I know that I was capable of this feat, if with regard to every feat of this kind which I had performed I necessarily had to believe that it had not taken place?” (4) is suggested by “If I could will to acquire a ‘belief’ irrespective of its truth … there must be a restriction on what is the case after the event; since I could not then, in full consciousness, regard this as a belief of mine, i.e. something I take to be true …” (5) is implied by “I could not … regard this as a belief of mine, i.e. something I take to be true.” (8) is an implicit assumption necessary for the inference from (7) to (9). (10) is an implicit assumption necessary for the inference from (9) to (11).

With the exception of (6), the inferred lines are also stated more or less explicitly. (6), however, must be part of the argument because the explanation for why knowing a belief was acquired at will would be problematic is that it would be acquired whether or not it was true. (7) comes from “I could not … regard this as a belief of mine … and also know that I acquired it at will.” (11) is stated, “With regard to no belief could I know—or, if all this is to be done in full consciousness, even suspect—that I had acquired it at will.” (12) occurs at the end of the passage: “… and could I know that I was capable of this feat, if with regard to every feat of this kind which I had performed I necessarily had to believe that it had not taken place?”

7 These moves involve issues of self-knowledge in that they involve issues about one's knowledge of what one believes.

8 I believe that it was an omission on Williams's part, though minor, that he did not say that if one had the ability to acquire a belief at will then one would know that one had the ability to acquire it at will whether it was true or not. Williams cannot be assuming merely that one must have knowledge of having an ability to do something if one has the ability to do it at will, for then he would need to grant that one does have the requisite knowledge for acquiring beliefs at will. We know we can acquire beliefs, and we know we can acquire them whether they are true or not. On Williams's view, however, the problem is that we do not know that we have the ability to do this at will.

9 Ibid., p. 148. Emphasis added.

10 Ibid. Emphasis added.

11 In the argument Williams focuses our attention on cases of willing in full consciousness. It is not clear if he thinks willing must be done in full consciousness or if it must be possible for willing to be done in full consciousness. Given the strength of the Original Assumption, it is likely that he assumed the former. For it would be odd if he were to maintain that, for something to be willed, it must be possible for it to be willed in full consciousness and yet to maintain that for something to be willed one must know that one has the ability to do it at will.

12 This is vaguely suggested by Williams, for the general subject of the article is whether there is room for decision in belief acquisition. It is in the context of this issue that Williams argues that one cannot decide to acquire a belief at will.

13 Williams's view on this matter is shared by Winters, Barbara in “Believing at Will,” The Journal of Philosophy, 76 (1979): 245CrossRefGoogle Scholar; O'Shaugnessy, Brian in The Will: A Dual Aspect Theory (Cambridge: Cambridge University Press, 1980) pp. 21–28Google Scholar; Scott-Kakures, Dion in “On Belief and Captivity of the Will,” Philosophy and Phenomenological Research, 53 (1993): 77–104Google Scholar; and Pojman, Louis in “Believing and Willing,” Canadian Journal of Philosophy, 15 (1985): 37–55CrossRefGoogle Scholar. Pojman, for instance, writes that volitionalism in belief acquisition would require that “one must be fully conscious of what one is doing in acquiring a belief through an act of the will” (p. 39). Moreover, he maintains that in order to account for the difference between believing at will and believing willingly, “the volitionalist … must assert that the acts of will which produce beliefs are decisions of which we are fully aware” (p. 39). Part of the problem, I suspect, is genuine philosophical disagreement. However, I think it is just prejudice, albeit well-entrenched prejudice, to think that the example I have given would need to be described as just a case of reflex or as merely a case of scratching. One breathes willingly, but not normally at will. The scratching example strikes me as sufficiently different from breathing willingly to merit a stronger description even though the scratching does not issue from decision and is not done in full consciousness. A full consciousness requirement is not needed to make a conceptual distinction between believing at will and believing willingly. Pojman is offering a phenomenological argument against acquiring beliefs at will. Full consciousness is only needed for an instance of acquiring a belief at will to be immediately evident to the believer.

14 In some cases, it would also be correct to say that one cannot decide to do something if one lacks the ability to do it at will. For example, I have the ability to digest lactose, but digesting lactose is not something I do at will. When lactose is broken down by my digestive system it is not because I have decided that my digestive system should do this. Terrence Irwin discusses such cases in “Reason and Responsibility in Aristotle,” in Essays on Aristotle's Ethics, edited by Rorty, Amelie Oskenberg (Berkeley and Los Angeles: University of California Press), pp. 122–23.Google Scholar

15 The Original Assumption underpins premise (1) as well as premise (2), and so to determine that Williams is not in any way committed methodologically to the Original Assumption, we would need to examine whether a revision of premise (1) based on the Weakened Assumption is sufficiently strong to support (l)'s role in the argument. However, since, as I argue in part 4, the inference from (1) and (6) to (7) is invalid, I will not attempt to answer this question.

16 The inference to (7) relies on a confusion of two claims about knowledge. It confuses a claim about what one would know if one had the ability to acquire a particular belief at will with a claim about what one would know if one knew of a particular belief that it had been acquired at will. The confusion is inherent in the first premise. Premise (1) is a claim about what one would know if one knew one had the ability to acquire a particular belief at will. Williams maintains that the ability to do something at will requires knowledge of the relevant ability. But in (1) Williams partially defines acquiring a belief at will in terms of acquiring it whether it is true or not. From this he slides to assumption (*).

17 It merely provides background information for the argument about what, on Williams's view, it would be to acquire a belief at will. Since (1) drops out of the argument, we do not have to worry about weakening the premise as we did in part 2.

18 Scott-Kakures makes a similar, though subtler, mistake in “On Belief and Captivity of the Will” (p. 94) when he argues that one could not acquire a belief at will because one is precluded from forming an intention to do so by the fact that one's cognitive perspective would recognize there were no epistemic grounds on which to rationalize the belief

19 Williams, “Deciding to Believe,” p. 137. At present, I will not take issue with this claim.

20 Williams likens acquiring a belief by conscious decision (even if not at will) “because it is fashionable or comfortable or in accordance with the demands of social conformity to believe that thing” to a “project of trying to get oneself to believe the false” (ibid., pp. 150–51.) He then raises moral and practical objections to non-truth-centred motives for acquiring a belief. But having a non-truth-centred motive would not entail that the belief one acquired would be false.

21 Ibid., p. 137. According to Williams, part of aiming at truth is the fact that to say “I believe that p” is a way of asserting p. Williams claims that this fact is related to Moore's paradox; that is, to the paradox in one's saying “I believe that p, and p is false.”

22 Ibid., p. 149. This is the second reason Williams gives for thinking it is a necessary fact that a belief cannot be acquired at will.

23 These, of course, are not the only ways in which one could acquire a false belief. They are just stock explanations of how empirical beliefs can be false without this necessarily being attributable to any mistake of the believer. Moreover, that we are sometimes mistaken in our perceptions does not show that perception is an unreliable method for acquiring true beliefs.

24 Examples of beliefs that are acquired whether or not they are true, in the sense I attribute to Williams in part 4, are not difficult to come by, and do not necessarily involve acquiring the belief at will. So the sense of acquiring a belief whether or not it is true that I have attributed to Williams does not fully define what it would be to acquire a belief at will. It might be thought, therefore, that we need either to strengthen the notion of acquiring a belief whether or not it is true, or to augment this condition, to get at all that is problematic about the idea of acquiring a belief at will. We might also hold, for example, that if one could acquire a belief whether or not it was true, it would be possible for one to acquire a belief without the truth or falsity of the belief being pertinent; that is, having a true belief would not be a matter of concern to one in acquiring a belief. Considerations about the truth or falsity of the belief would have no bearing on one's decision.

25 1 am not certain that these facts would require explanation. We do not demand an explanation of how one can continue to hold a belief when one recognizes one has forgotten one's reasons for having that belief or cannot remember how one came to have that belief. Arguably, recognizing that a belief was acquired whether or not it was true is analogous to such a case insofar as, in both cases, the believer fails to discern a connection between having the belief and the relevant facts about the world.

26 Winters accepts this point. She argues, however, that one cannot recognize that a belief was acquired at will and is sustained at will (i.e., is sustained by no truth considerations). I do not wish to address this further issue at this point. But we are in agreement that the question of whether beliefs can be voluntarily sustained is distinct from that of whether they can be voluntarily acquired.

27 For instance, someone might argue that, unlike a belief acquired at will, a belief acquired in one of these other ways has its origins outside the person even if it is acquired whether or not it is true. And so there is, in some sense, still some reason for the person to hold the belief. But when one recognizes that a belief was acquired at will, one must recognize that there is no reason for one to hold the belief. This objection, however, relies on at least two mistakes. The first is that a reason for having a belief is provided merely by its having its origin outside the person; this is, rather, an explanation for the belief. The second is that a belief acquired at will is a belief acquired without reason. It would be acquired without conclusive evidential reasons for belief. But it would not necessarily be acquired without pragmatic reason. This is a complex issue that I cannot investigate here

28 My view at this point is different from Pojman's. In “Believing and Willing” Pojman concludes that “Volitional believing is not simply irrational believing, but is incoherently irrational, for it offers an account of believing that confuses the nature of believing” (p. 49). I do not maintain that it is incoherently irrational for the reason that I deny that a belief acquired at will must be acquired in full consciousness of the fact that it is acquired at will or independently of rational grounds for taking it to be true.