Remarks on the transparency of belief

1. In what conditions ought one to believe something? For example, ought one to believe that human-caused climate change is occurring? It’s sometimes suggested that the following answer should be accepted:

(Ought-1) One ought to believe P if and only if P.

That is,

(Ought-RL) If P, then one ought to believe P.

(Ought-LR) If not-P, then one ought not to believe P.

For example,

(1) If human-caused climate change is occurring, then one ought to believe that human caused climate change is occurring.

(2) If it’s not the case that human-caused climate change is occurring, then one ought not to believe that human caused climate change is occurring.

However, Ought-RL looks to be incorrect. For it claims that one ought to believe every truth. There are two central problems here.

The first problem rests on a very plausible—though controversial—claim, to the effect that ought implies can. According to the ought-implies-can claim, if one ought to A, then one can A. For instance, if one ought to make others happy, then one can make others happy. Crucially, it seems to follow that if one can’t A—if one can’t make others happy—then it isn’t the case that one ought to A—it isn’t the case that one ought to make others happy. Now it’s plausible that there are truths that are too long and complicated for any human to believe—say truths involving the conjunction of all currently known mathematical truths, or the conjunction of all the truths in all the books in the University of Warwick library. If it’s impossible for any human to believe such a truth, then the ought-implies-can claim delivers the result that no human ought to believe such a truth. It would follow that Ought-RL is false.

The second problem is less decisive, but more interesting. In order to avoid impossibility, suppose that someone set themselves the aim of getting as close as they could to knowing the conjunction of all the currently known mathematical truths, or all such truths contained in books within the University of Warwick library. The person in question has no ulterior motive for doing this, no further end for which this is a means. They are pursuing the end entirely for its own sake. They pursue the end for a month or so, and get themselves into a position in which they believe the conjunction of a large number of mathematical truths. Let’s call the conjunctive claim that they come to believe “Q”. We now have that it’s possible to believe Q, so the ought-implies-can claim fails to rule against this being something one ought to do. Nonetheless, it’s hard to believe that, all else being equal, one ought to do what one can to emulate this strange figure. We may admire their resolve, but in other respects it’s hard even to find intelligible the end that they have set themselves, given that it serves no further purposes. To that extent, it seems implausible to claim that, even where P is restricted to truths that it’s possible for one to believe, one ought to believe P just because it’s true.

2. Suppose that’s right. That is, suppose that it’s false that, for every P, regardless of its intrinsic interest or capacity to sub-serve one’s further ends, if P, then one ought to believe P. Now consider the following claim:

(Transparency) “the deliberative question whether to believe that P inevitably gives way to factual question whether P, because the answer to the latter question will determine the answer to the former question” (Shah and Vellemean 2005: 499, with inessential reformulation)

It’s somewhat plausible that one can’t answer the latter question—whether P—without coming to believe either P or not-P. In particular, if one comes up with the answer P, there’s some plausibility in the claim that that would be a way of coming to believe P. Suppose it so. Still, without more ado, coming to believe P doesn’t seem to decide the question whether to believe P. Similarly, if, in the course of deliberating whether to run down a hill, I stumble and begin so running, that isn’t a way of deciding to run down the hill. And in the latter case, it would seem that I might even decide, in the course of flailing wildly down the hill, not to do so, and so make every effort to stop. So, similarly, having come to believe P, I might keep open my deliberation over whether to do so, whether, that is, to continue to believe P. Now in fact, ceasing to believe P is, if anything, less within our control than is ceasing to run down a hill. And that is especially so in conditions where one has no reason to believe P is false. But it’s far from clear that, in either the case of activity or of attitude, the fact that one is saddled decides the deliberative question whether to acquiesce.

Consider, then, the following science fiction case. Suppose that, in 2035, neuroscientists have found a way to extinguish individual beliefs by the use of specially designed radioactive pills hooked up to some future analogue of a mobile phone app. Ishmael, pills dropped and phone in hand, is on a research mission at the University of Warwick library. He is on the hunt for a specific mathematical theorem that he knows is to be found in one of the maths books in the library. He doesn’t know the name of the theorem, but knows enough about it that he will be able to recognise it on presentation with a little reflection. Starting with the “A”s, Ishmael begins reading through books of mathematical theorems. Ishmael accepts as true each of the theorems that he comes across, but has no immediate interest in any of the theorems other than the one he seeks, and no interest that would be served by believing any of those other theorems. With respect to one of these theorems—let’s call it R—Ishmael deliberates over the question whether to believe R. Since he’s answered the factual question whether R, he’s come to believe R. The question driving Ishmael’s deliberation is whether to continue to believe R, given that he has the resources to cease so believing via a couple of key presses. Given that the mere fact that a claim is true does not suffice to make it so that one ought to believe P, it seems that the mere fact that R is true does not suffice to make it so that Ishmael ought to believe it. In particular, its truth does not suffice to make it so that Ishmael ought to continue to believe it. Thus, I think there is some reason to think that Transparency is false.


Dodd, J. (2013). ‘Jane Heal’s “Disinterested Search for Truth”.’ in G. Longworth ed. Proceedings of the Aristotelian Society, Virtual Issue 1, Truth: 138–144.

Heal, J. (1988). ‘The Disinterested Search for Truth.’ Proceedings of the Aristotelian Society, Volume LXXXVIII. Reprinted in G. Longworth ed. Proceedings of the Aristotelian Society, Virtual Issue 1, Truth: 125–136.

Shah, N. and Velleman, D. (2005). ‘Doxastic Deliberation.’ Philosophical Review 114: 497–534.

  1. So, firstly, I think all talk of (purely) epistemic oughts is a bit bonkers. “You ought to believe the truth” doesn’t seem to do much for me aside from in conjunction with things like “if you want to avoid making mistakes in life”, “if you want to be a good person in leading others to truth” and other pragmatic or ethical things. So all the stuff in 1. is right up my street

    And I don’t think the question whether to believe is tied to the truth.I may well realise that believing p will make me feel worse, or be a morally worse agent. So these other considerations have to have some weight. So if I come to believe that Wendy has run off with Clarence then I’ll be really upset because I love her a lot and was just building up the confidence to tell her and ask her to elope with me to some lovely little village in the Cotswolds. But I’m aware of this when I’m deciding whether I should believe: “Do I want to believe that Wendy has run off with Clarence?! NO!” and maybe at some stage I might be able to just stop any inquiry into the question and so not find out the truth of the matter; that answers the first question without even looking at truth, it does so solely in terms of my happiness.

    But, say I’ve found out p: she’s only gone and ran off with Clarence!
    What can I do? Well, I can still deliberate and decide it’s best not to believe p. But I can’t follow that through and not believe. If I think p is true I believe p, that’s simply what belief is; and if I believe p then I think it’s true. (I reckon… but then no one agrees on what belief is anyway). And so talk of deliberating whether to believe p or not doesn’t get us anywhere because I do believe p. I can deliberate on whether it would be better not to believe p, but I can’t deliberate on whether to believe p. In the earlier case, where I deliberate whether to believe p, I can do so by deliberating whether I want to believe p, and such things, and can answer that question without looking at the truth and thus get an answer not to believe p (so to believe ¬p? or to withhold? I’m not sure.. Perhaps if we decide “yes” when we ask “should I believe…?” then we do need to look at the truth. But point stands that we don’t need to look to the truth to answer the question). But given that I *do* have the truth, I can still do such a deliberation as to whether I *want* to believe p but I can’t help but reach the end result of believing p anyway and so there can’t be any deliberation over whether to believe p, because that’s just given by the fact I do believe p is true.

    Now, perhaps I deliberate and decide “I don’t want to believe this!” then perhaps I’m not beyond all hope. I currently think that it’s true that p. But I’ve not looked at all the evidence, and I might decide to inquire a bit more (perhaps there’s some doubt in my mind (there’d have to be to inquire, I’d say)) and consciously make sure that I look for evidence which will disabuse me of my belief. So I end up beliving ¬p (or ¬believing p, perhaps withholding) but I’ve only done that by changing the answer to the second question, by changing what I think in relation to “whether p?”; I just can’t see a way of doing otherwise. I’d say that Ishmael’s evidence for R being true is simply the fact he believes it, because he’s just gone and got the belief from the book. Get rid of that and he’s got little evidence left for R’s truth, none in fact; he might still be able to entertain R mentally, but he can’t believe it because he doesn’t think it’s true.

    I’d deny the claim that “the mere fact that R is true does not suffice to make it so that Ishmael ought to believe it” if by that you mean “the mere fact that R is true (for Ishmael, as in how he thinks of the world) does not suffice to make it so that Ishmael ought to believe it” (the brackets required because we’ve already denied that one *ought* to just believe something because it’s true) because ought implies can and he just does believe it so whether he ought to believe it or not just isn’t a question he can answer: the fact he thinks it’s true means he believes it, that’s all there is to it. The question’s answered for him.

    I’m not sure I’ve fully expressed and motivated my worry well enough, but I’ve spent enough time trying to so really should avoid saying any more or it’ll get unmanageably long. Do feel free to ignore this if what i’ve said is a rambling incoherent mess.

  2. Thanks for those helpful comments.

    1. The first point, that oughts have no place in thinking about belief, is delicate. One question here is whether that is a larger point about mental states–so a point that applies, for example, to intention, the outcomes of decision, dispositions of the will, &c. As a point specifically about belief, it may be that the idea is that what one believes is outside the scope of our control, and that only what we can control can be subject of oughts. Issues about control of belief are delicate. I don’t have settled views about whether oughts can apply outside the scope of control. I wouldn’t rule out, without further reflection, that oughts apply in a wide range of cases where we can think about proper function, defectiveness, &c. In that case, we might think that believing falsehoods, say, is a kind of defect, a way of being non-optimal and, to that extent, something one ought to avoid. But I don’t have firm views about this.

    2. The point that what one believes may be shaped by considerations other than ends concerning believing truths (or knowing) is important, but in the present context seems indecisive. It may be that ceteris paribus, one ought not to believe falsehoods, but like some other ceteris paribus oughts, that one can be outweighed by other considerations. Or, it may be that qua believer, one ought not to believe falsehoods, but qua agent more generally, it might be that doing so can be permissible.

    3. As you in effect suggest, the question is vexed, whether believing P is a matter of believing that P is true.

    4. The suggestion in the text is that deliberation over whether to believe P might withstand formation of belief. Consider, for example, the role of memory. I know the word count for a paper I’ve just read. So, I have a true belief about it. Should I retain that belief? In particular, should I do things that would help me to retain the belief, or should I let it naturally slide out of view? That question about what to believe seems–at least on the face of it–to withstand my presently having the belief. The science fiction version in the post appealed in effect to a means of controlling memory, via perfectly targeted forgetting. The question at that point is: given that I believe P, can it be reasonable to decide to lose that belief, given that I believe it to be true, that I believe P?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: