Doxastic Voluntarism

There are several questions about the relationship between belief and voluntariness that I’m planning to write about over the upcoming months. Several of these topics will be pretty familiar, but some might not be. (I’m interested in the role that doxastic voluntarism, or something like it, plays in Meditation Four, for instance, which is not as far as I can tell one of the big topics on the radar screen in contemporary philosophy.)

But those are for more serious posts. Today I just want to make a little observation. Philosophers often write as if it is obvious that we can’t decide to form beliefs. You might think that if this is obvious, then authors would never have characters, let alone narrators, decide to form beliefs. But if you did you’d be very badly mistaken.

For what it’s worth, I suspect the psychological assumptions these authors are making are quite plausible. When someone tells us something that is plausible, but not such that we should obviously trust them, we have to decide whether we will, on this occasion, trust them. If we have no other reason to believe what they say, but trusting them will involve (perhaps inter alia) believing what they say, then we are deciding to believe.

22 Replies to “Doxastic Voluntarism”

  1. I just tried a google book search on “I just knew that” and “but I was wrong”. Try it. I have limitless faith in the ability of the folk to assert falsehoods.

  2. one example given is
    I decided to believe Dvorah
    This sounds to me like “I decided to TRUST Dvorah” (I am not a native speaker of English, so correct me if I am wrong).
    Deciding to trust is a different cattle of fish than just forming a belief by will. But it is difficult issue. For instance it looks like it involves: “I will form a belief that Dvorah is reliable.” But it does not have to happen this way.

  3. Neil, I know people say things like that as well, but it’s easy to say what has gone wrong there – they’ve misused the word ‘know’. Or at least they have used it in a situation where it is descriptively inaccurate; maybe that’s compatible with using it properly.

    But what is the misdescription in the cases I mentioned? Is it that the agents didn’t really make a decision? Or is it that what they formed wasn’t a belief? Or is the situation is too incomprehensible to make sense of? I think the situations these authors describe seem easily imaginable, I think there’s nothing wrong with saying that they are describing a decision, and I think the result of that decision is a belief. I realise all three parts of this are contentious, but I think all three are plausible. That’s not the case in the ‘false knowledge’ cases – there it is easy to say what’s gone wrong.

    Nenad, I agree there is a belief/trust relation here, but I don’t see why it undermines the point. In most of the cases, the trust in question extends exactly as far as the particular claim mad. So if the characters had said “I decided to trust X”, it really would have meant “I decided to trust X on this occasion when she said that p”. And the distance between that and “I decided to believe that p” is pretty narrow.

  4. Here’s a suggestion in support of Nenad’s worry. (Hi Nenad!) When you decide to trust someone in respect to their claim that p and, as a result, come to believe that p, yes, your belief that p results from your decision. However, what you’ve decided isn’t to believe that p but, simply, that p. What we sometimes describe is as “deciding to believe” that p is just deciding that p.

    This is analogous to a point I once made (when working on self-deception) about wishful thinking. It is inaccurate to describe wishful thinking as believing “what you want to believe.” Rather, it’s believing what you want to be the case. For example, suppose you know that smoking is generally hazardous to people’s health but you wishfully think that it isn’t hazardous to yours, partly because (though not for the reason that) you love to smoke. So your belief is motivated. But what you want is not that you believe (or that it seems) that smoking isn’t hazardous to your health but that smoking (really) isn’t hazardous to your health.

    Anyway, Brian, I agree that your authors are describing a decision and that the result of that decision is a belief, but I wouldn’t describe it as a decision to believe.

  5. The misdescription is easily identified. The folk have forgotten their Cohen, and therefore the distinction between believing a proposition and accepting a proposition. Accepting a proposition does, as a matter of fact, often lead to believing it. But one doesn’t acquire the belief in these cases by deciding to believe it (except in some outre cases).

  6. I know two comments in a row is rather like bidding against oneself. Anyway..

    Compare these two cases. Bob is told a rather odd (but not incredible) story by Patty. He does not know Patty well enough to know whether she is reliable. But he, as he says, decides to believe her. Meanwhile, Bert is told the same story by Percy, whom he likewise does not know well. Bert is an epistemic duplicate of Bob. Bert, though, does not decide to beieve Percy; instead, he finds himself believing him (perhaps Bert is under cognitive load, and therefore defaults to acceptance of doubtful propositions [Daniel Gilbert. 1991. How Mental Systems Believe. American Psychologist 46 (2): 107-119]).

    Now, both Bert and Bob believe that it is extremely unlikely that there is any further evidence to be had for or against the truth of the story. But they\‘re wrong. A minute after Bob decides to believe the story, and Bert finds himself believing it, each encounters decisive evidence that it is in fact false.

    Now, will Bob be just as surprised as Bert? Surely, I claim (without any argument) Bert will be far more surprised. But if that\‘s right, then we have grounds for saying that whereas Bert believed the story, Bob merely accepted it.

  7. When teaching, I often find myself asking my students, “What should we believe about this?” I think I mean this, and they hear this, as a practical question: what shall we believe? The assumption we are evidently making is that each of us can make a decision about what to believe.

    I also ask them to “Consider this argument.” I intend, and they know I intend, them to consider the argument as part of a procedure of fastening on a belief. The assumption here is that they can do something, namely considering, and that this will have some effect on what they believe. This is a less directly voluntarist assumption than the previous one, but it is still a form of doxastic voluntarism.

    In general, I don’t think the practice of teaching philosophy is well-described, nor understood by either teacher or student, as “We’ll voluntarily read and discuss some articles and arguments, and then you’ll involuntarily find yourself believing something.”

    Finally, FWIW, I can’t make sense out of Kent’s distinction between “deciding to believe that p” and “deciding that p”. In my mind, “deciding that p” is something I can only do when I think it is up to me whether p, like deciding that I’ll wear the blue shirt this morning.

  8. I think I see what’s bothering you, whitew. But I didn’t mean deciding-that as a kind of deciding-to. Nor was I thinking of deciding-that as what, for example, an umpire does when deciding that a pitch is a strike. As a result of his so deciding, it is a strike — even if it’s outside the strike zone. What I had in mind was judging that p, in the way that anyone could judge that a pitch is in the strike zone. Or, where inference is involved, one can decide that p in the sense of concluding that p. I think what I mean by deciding is what Neil means by accepting. As a result of judging or concluding or accepting that p, one may come to believe that p (not that this is necessary for believing for, as Neil points out, one might just “find oneself” believing something). Or one may not.

  9. Sorry if this piece is already known to all on this list, but I learnt a great deal from Shah and Velleman, ‘Doxastic Deliberation’, Vol 114 Philosophical Review, which is exactly on the current topic. I am eager to see how Brian’s views are different from theirs.

  10. … for example, an umpire does when deciding that a pitch is a strike. As a result of his so deciding, it is a strike — even if it’s outside the strike zone

    If that were true then we wouldn’t need a strike zone, would we? The rules let the ump’s judgment trump, but certainly the ump can get the call wrong. No?

  11. Right, the ump can get the call wrong. But it’s still a strike, i.e., counts as a strike. That’s unless the ump reverses himself, which generally happens only a check swing as judged by another ump. .

  12. Hi Brian,

    It seems to be pretty common ground that one can decide to do things as a result of which one will end up believing p (familiar e.g.: one can decide to go to church a lot with the aim of ending up with a belief in God). But that’s not normally taken as showing that one can decide to believe that p in the interesting sense at issue. How is your case different?

  13. Hi everyone, doesn’t trial deliberation involve something close to ‘deciding what to believe’? Take a murder trial. The prosecution claims and adduces evidence to show that the defendant caused the death of the victim in circumstances which amount to murder. The defendant says she didn’t do it. To leave out the difficulties of group dynamics, suppose a bench trial, so one person, the judge alone, decides on the verdict. At the end of the trial, she delivers a guilty verdict. This verdict is not a pure declaration (unlike, to use Austin’s example, the naming of a ship). It is partly assertive of propositions of antecedent facts. It asserts, among other things, either that:

    P1: the defendant did cause the death of victim, or, on a different interpretation,

    P2: it is proved according to the law (including the legal standard of proof and rules constraining evidential reasoning) that the defendant did cause the death of the victim

    In deciding what verdict to give, the judge is deciding whether she is in a position to assert P1/P2. So far as we expect the judge to be sincere, we expect her to believe the assertions she makes in the verdict she returns. Deliberation on the verdict therefore involves the judge having to decide whether P1/P2 is true. If she judges that P1/P2 is true, can she still not believe that P1/P2?

  14. I’m not sure that those who deny voluntarism regarding belief very often say that one cannot decide to believe. It is perfectly natural to say that one has decided or is trying to decide what to believe. I see no reason to give this a bent interpretation.

    The important question is whether deciding what to believe is interestingly analogous to deciding what to do. Most involuntarists regarding belief are interested in denying that we can choose what to believe in the way that we can ( or at least it feels that we can) choose what to do. (Actually, what is important to me is something more like the feel of things than the untimate nature. By this I mean, I do not believe issues regarding free will and determinism are really relevant here. It does not even feel like I choose what to believe, picking one proposition from a number of other possibilities that I could just as well have choosen. It does feel like I choose what to do in something like this way.)

    To report the way it seems to me, deciding what to believe is a matter of thinking about or considering various things, or perhaps doing various things, e.g., gathering data, in the hope that as a result of this I come to have a belief about something. Sometimes this process works, and I end up believing something, and sometimes it does not. In the latter case I have to flounder around somemore in the hope that I hit upon something that brings about a belief. But it never feels like I choose to believe one thing rather than another. It is more like falling asleep. I can do various things that are likely to lead to my sleeping, but it isn’t as if I choose to sleep now. It either happens or doesn’t in a way that is to an important extent outside my control. (If it were not so, there wouldn’t be such a market for sleeping pills!)

    Deciding to act does not feel that way at all. I may consider all kinds of reasons for and against performing a certain action in a way that is superficially similar to the way I consider reasons for and against the truth of a proposition. But in the end I choose to act or not — or so it feels — whereas I form a belief (or not) whithout my choosing.

    Sorry if this sounds rather dogmatic, but I am mostly just reporting how it all feels to me. I have some entitlement to be dogmatic about that.

  15. First of all, there is the issue of the usage of the phrase ‘I decided to believe’ and the like. Personally, I am wary of deriving any deep insights into the nature of the objects of philosophical study from beta simply because authors (especially authors of fiction) often operate with license that goes beyond even that of conversational implicature. But even when we control for problems such as this one, it leaves open the possibility that ‘deciding to’ in ‘deciding to believe’ is used differently than it is in other circumstances.
    (1) ‘I decided that I shall go shopping.’
    which seems more or less equivalent to
    (2) ‘I decided to go shopping.’
    Now compare it to
    (3) ‘I decided that the answer to the problem is 12.’
    which does not appear to have a similar equivalent. Most candidates report decisions that presuppose (that 3) is already among the stock of one’s beliefs. At the very least, it is not clear that (3) needs to involve any voluntary component in order to be used correctly. In fact, as I think someone has already noted, we speak of computers deciding in much the same sense.
    (4) ‘I decided to believe S.’
    is equivalent to
    (5) ‘I decided that S is believable/trustworthy with respect to p.’
    which is a lot more like (3) than it is like (2).
    or perhaps
    (6) ‘I decided that I already believe S with respect to p.’
    ‘Deciding’ in (3), as well as in (6), bears a strong similarity to ‘discovering’. And discovery, of course, need not be voluntary. We discover all sorts of things in spite of ourselves.
    Quite a few of the results of the beta search, taken together with bits of their contexts, support that this is at least one of the senses of ‘deciding to believe’:
    “I decided to believe my family. Why would they lie to me?” (Inference to the best explanation.)
    “First, I decided to believe in myself. If Jane Bell believed in me, I should be
    able to believe in me too.” (Expert testimony.)
    And so on.
    I wonder whether this distinction between deciding that and deciding to, together with the fact that believing/accepting and believing/trusting locutions are used quite liberally in the colloquial English, cannot account for much of the empirical evidence that beta, or any other source, can provide.

  16. Dear fellas,

    I think the right stance to take on this issue is the following:

    YES..There is a perfectly sound everyday sense in which we sometimes decide to believe this or that.

    YES..This sense is at least rougly analogous to one perfectly sound everyday sense in which we sometimes decide to DO this or that (eg. kick a ball).

    BUT no sense of “decide” invoked above has anything to do with a willing/intention to produce the attitude/action in question.

    The opposite impression, in my view, mainly derive from the fact that, in the practical realm, we often use “decide” as synonymous to “intend” as in: Why did she kick the ball? Because she decided to kick it!

    There is, however a second sense of “decide” closely attached to deliberation in which we may deliberate about what to do in a certain scenario, reach a conclusion about which option would be the best to realize, and hence decide for that option without necessarily intending to realize it.

    In this sense, we also sometimes deliberate about which of more contents is best supported by our epistemic reasons, and, in so far as we reach a conclusion, decide to believe that option.

    The crucial difference to the practical realm is the process by way of which our “decision” may ultimately result in an attitude/action. In the doxastic realm we may sometimes actually come to believe the conclusion of our deliberation about what to believe. But in so far as this happens, the process is fairly automatic. If it misfires, there is nothing we can do – by way of intentions to form the belief in question – in order to alter the result.

    In the practical realm, in stark contrast, an intention to perform the action is needed on top of the conclusion of our deliberation in order to produce the relevant action.

    Cf. Kent Bach above, we could try to disambiguate the relevant senses of “decision” by talking of a decision for p, rather than a decision to believe p (which, if effecient would vindicate doxastic voluntarism). Or we could differentiate: deciding for believing p/deciding to believe p.

    However, I take it that neither disambiguation attempt is very helpful. Talking about “decisions” will always provoke voluntarist connotations. Sadly enough, we will probably have to live with our discourse as it is.

    I develop these thoughts in greater detail in my “The analogy argument for doxastic voluntarism”; accepted and bound to appear some time in Phil.Stud. I will be happy to supply an electronic copy to interested parties.

  17. Carl Ginet has a nice paper on this. “Deciding to Believe” in Matthias Steup (ed)., Knowledge, Truth and Duty: Essays on Epistemic Justification, Responsibility and Virtue, Oxford: UP (2001), pp. 63-76.

Leave a Reply