Here’s a quote from Scanlon:
A rational person who judges that there to be sufficient grounds for believing that P normally has that belief, and this judgment is normally sufficient explanation for so believing. There is no need to appeal to some further source of motivation such as “wanting to believe.” (Scanlon 1998, 33)
Immediately following this quote, Scanlon goes on to say that the same is true of intending to do x. We don’t need to appeal to a desire to explain why someone who judged that there was sufficient grounds for intending to do x intended to do x.
Now here’s an observation: no one, not even the Humean, would appeal to something like “wanting to intend to do x” (the analogue of Scanlon’s “wanting to believe”) in order to explain why someone has formed the intention to do x. I take it that if we do think that we need to appeal to some desire, it will be a desire for some states of affairs that the agent aims to bring about by doing x, not to a desire to intend to do x. The thought is that purposive actions aim at the realization of states of affairs and thus the motivation for such action lies with a desire for the state of affairs that’s being aimed at. This brings out an important difference between intending to do x and other attitudes such as desiring that P, believing that P, admiring S, etc. Whereas the attitude of intending to do x is telic, these other attitudes (desiring, believing, admiring) are not. For instance, when we believe that P, we don’t do so for the sake of realizing some end. By contrast, when we intend to do x, we do so for the sake of realizing some end.
Now here’s a question: Does the fact that the attitude of intending to do x is telic whereas these other attitudes (e.g., believing and desiring) are not give credence to the view that we must appeal to a desire for some end in explaining why someone has formed the intention to do x but not to explain why someone has formed these other attitudes (e.g., the belief that P or the desire that P)?
What Scanlon wants to oppose in the quote is the Humean principle which Gary Watson and others call the ‘desire-in, desire-out’ principle. The idea of this principle is that purpose action needs to be motivated by desires to reach certain ends and such desires can only be created by other prior desires. As Smith, Scanlon, Nagel and Watson others have argued, there seems to be no reason for thinking that beliefs about reasons could produce such desires in rational agents too independent of what their motivational set was before the reason-judgment. This need not deny that purpose action is always produced by desires for certain states of affairs.
Now, you seem to be thinking of a related principle ‘desire-in, intention-out’. I find it difficult to see why the same arguments would not count against this principle as the former principle. Your argument seems to rely on the idea that intention are telic states. I’m not sure what this is supposed to mean.
You say that ‘when we intend to do X, we do so for the sake of realizing some end’. This characterisation of intensions seems ambiguous. It’s true that intentions are desire-like in their direction of fit – they aim at bringing about certain states of affairs. But, if Scanlon and others are right, it’s not true that why we are in these states must always be that we have some other ends that we are disposed to realise. I might adopt the intention to go to the library because I judge that I have a reason to return certain book independent of whether I desire any further state of affairs. I don’t intend to go to the library here *for the sake’ of some end even though my intention does include a certain end. If this is possible then in addition to ‘desire-in, intention-out’ cases, we can have ‘beliefs-about-reasons-in, intentions-out’ cases.
Jussi,
I want to grant Scanlon his claim that to explain the fact that S believes (or desires) that P, we don’t need to appeal to any desire. The fact that S is rational and believes that she has sufficient reason to believe (or desire) that P is sufficient to explain why S believes (or desires) that P. Sometimes, though, I’m inclined to think that we do need to appeal to a desire to explain the fact that S intends to do x. If S intends to do x, then it seems that S must have some end that she hopes will be realized by her doing x. In your case where you intend to go to the library, isn’t your aim to realize the state of affairs where you’ve returned your library book? Could you intend to go to the library and not desire to realize some such state of affairs?
So one issue is whether beliefs can give rise to desires. I’m not denying that they can. Perhaps, there are motivated desires as Nagel and Scanlon claim. But another issue is whether someone can be moved to intentional do x without having a (motivated or unmotivated) desire that S believes would be fulfilled by her x-ing.
Hi Doug,
Nice question.
Consider this sequence:
(1) P deliberates by considering the reasons for and against performing an act of type A
(2) P judges there is most reason for him to perform an act of type A
(3) P intends to perform an act of type A
To show that there is something other than (2) that we need to advert to in explaining (3), Scanlon seems to think you need to produce a switching argument – you need a case where (2) but not (3) is true.
Wallace has produced one such argument in hopes of showing Scanlon needs to add something about the quality of the agent’s will. In his Arist. Society paper, Scanlon seems to concede the force of this point. He also rightly points out that reflective judgments like (2) need not be attributed in many cases.
My first question is whether you reject the need to provide a switching argument here if you are to establish the need to posit an additional mental state in order to explain or whether you have one in mind.
Second, your remarks suggest you think there is a tight conceptual connection between having an intention and having an aim – but if that is the case, then why think that the having of the aim is anything over and above the intention? If the connection is tight as can be, then it is hard to see why we need to introduce a “desire” in addition to the intention in order to explain the action.
Finally, I’m not sure about the claim that no one would appeal to a desire of the “want to intend” sort…unless you mean to allow for people like Vellleman and Broome, who have been taken to appeal to something like a higher order desire to intend as you judge there is reason to act (what Wallace calls “meta-internalism”)?
Brad,
Yes, I suppose that I would agree with this: “To show that there is something other than (2) that we need to advert to in explaining (3), …you need a case where (2) but not (3) is true.”
But I thought that the fact that there are such cases is just given. The problem is that that alone won’t do it. Scanlon agrees that there are cases where (2) but not (3) is true, but wants to account for the lack of (3) in such cases, not in terms of a lack of some desire, but in terms of how vividly the considerations in favor of (2) strike the agent.
Which article of Wallace are you referring to? I should read it (or read it again).
Regarding why think that the having of the aim is anything over and above the intention, I’m not at all sure about this stuff, but I guess the thought would be that the intention has an act as its object whereas the aim has a distinct state of affairs as its object.
Fair enough regarding meta-internalism.
Doug,
I think this
“If S intends to do x, then it seems that S must have some end that she hopes will be realized by her doing x.”
is fine and I don’t think Scanlon would have any objection against it either. I think the issue is rather what constraints there are for adopting the intention which comes with end that is desire to be realised. The Humeans want to claim that a necessary requirement for adopting an intention is that the agent had some other ends that are served by the ends that come with the intention. Scanlon and others want to deny this. They want to say that if you believe you have reason to realise such ends then this is sufficient for being able to intend to satisfy them.
One way to put this is that Scanlon and the Humeans agree on principles like ‘intentions as desire-like states in, actions out’. The question is what comes before the intentions that are constituted of desiring some ends. Are reason-judgments enough for acquiring those ends or do we always need more general, antecedent desires?
Doug,
Perhaps the most accessible place to look for that argument is section 2 of “Normativity, Commitment, and Reason,” which you can get at Philosopher’s Imprint. There is also the discussion in part one of his author-critics piece on WWO (Ethics (Apr 02): 429-70).
I agree with what you say about the interpretation of switching cases being the key. Scanlon connects talk of how vividly considerations strike us with the attribution of desires “in the directed attention sense,” but he pictures them as intervening factors whose absence need not be cited as part of the explanation when rationality rules the roost.
So then the question becomes: Why think this is wrong?
Brad and Jussi,
Thanks to both of you for your very helpful comments. I think that I’m much clearer on what the relevant issues are.
Brad: I don’t have any good reason to think that it is wrong, although I’m not sure that I’m convinced either. I’m undecided.
“This brings out an important difference between intending to do x and other attitudes such as desiring that P, believing that P, admiring S, etc. Whereas the attitude of intending to do x is telic, these other attitudes (desiring, believing, admiring) are not. For instance, when we believe that P, we don’t do so for the sake of realizing some end.”
This is slightly off topic, but I’m not quite sure why we should say we don’t believe for the sake of realizing some end. I get the idea that we don’t desire for the sake of realizing some end, but it seems not obviously wrong to say that we believe for the sake of some end (e.g., settling a question).
Clayton,
I’m glad that you brought this up. After I wrote the post, I had the same worry. Philosophers often say “belief aims at truth.” Nevertheless, I think that there’s a difference. When I intend to do x, it seems that I must aim at bringing about some state of affairs (even if it’s just the state of affairs where I’ve performed x). This is not the case with belief. When I believe that P, I needn’t have been aiming at bringing about any state of affairs (not even the one where I believe what’s true). Think of perceptual beliefs. In believing that the cursor that I see before me is blinking, I needn’t desire that I have true beliefs; I needn’t have been aiming at truth. I could want to believe false things and yet perceptual beliefs will arise in me unbidden nevertheless. I could be convinced that radical sceptism is true and thus that believing what I perceive is not necessarily true and yet I will still acquire these perceptual beliefs non-voluntarily. So whereas a condition for correctly ascribing to someone the intention to do x is that they have some aim that she believes x-ing will realize, the same doesn’t seem to hold for correctly ascribing belief or desire. If I believe that P, it doesn’t follow that I must have some aim that I believe will be realized by my believing that P. I hope that this seems plausible.
“I could want to believe false things and yet perceptual beliefs will arise in me unbidden nevertheless.”
Quick question: is this want de dicto or de re? I could want to believe that Kerry won the election without wanting to believe false things (because I also want it to be true that he won). Clayton might be driving at the view that belief aims at truth de dicto.
Very interesting post.
Justin,
I’m thinking that we can ascribe a belief to a subject without attributing to him or her either a de dicto or de re desire to realize any state of affairs (even the one where the proposition believed is true).
Hi Doug,
Fair enough. But for what it’s worth, I’m still not certain about this “perceptual belief arises unbidden” bit. I’ll grant that some sort of mental phenomenon arises unbidden, but I’m not sure why we should call it a belief. Surely there can be perceptual phenomena without belief, as when the moon appears implausibly large when it’s close to the horizon. I also wonder whether you would say that when a fish smells food nearby the fish necessarily believes there is food nearby.
My view is that we can’t settle this question a priori. We might necessarily believe what we see initially (and only revise our belief about the moon as an afterthought). But if so, it wouldn’t be necessary in any way that our bare intuitions would be capable of demonstrating by themselves. It might be psychologically necessary, for example. But somehow I doubt that it would be psychologically necessary for the fish.
Therefore it’s at least possible that belief requires some further desire. Perhaps a de dicto desire to realize the truth, or perhaps just a desire to “settle the matter” as Clayton suggested. (Although I’m still attached to the former.)
This is a bit of a sideshow, and I do appreciate the force of your question about desires and intentions.
Justin,
I take your point. I’ll need to think about it.