Thursday, October 22, 2009

Reasons as Propositions & Attitudes as Enablers

Someone could say that propositions are motivating reasons, but here's a worry:

Motivating reasons aren't necessary existents.
Propositions are necessary existents.
Motivating reasons aren't propositions.


This objection, the modal objection, seems to me to be pretty good. But, I can imagine a proponent of the view that identifies motivating reasons with propositions saying something like this. It's only when the subject has the right sort of propositional attitudes that the propositions can be the subject's reasons. This, I take it, is Miller's view from his recent Nous paper. Here's what he says:
Note that S’s bearing such an attitude towards p is itself a fact about p, and not a mental state. In addition, it is a fact which is not itself a part of the relevant motivating reason, but rather serves as one of that reason’s enabling conditions. Thus to use our previous example, the proposition there is widespread starvation in Iceland would not have served as one of my motivating reasons if I did not believe that there is widespread starvation in Iceland, even though strictly speaking it is the proposition which I believe rather than my belief itself that serves as my motivating reason (2009: 226).


I want to note two things. First, this response seems to concede that the motivating reasons that exist are necessary existents but then insists that it is contingent whether someone has a motivating reason and which motivating reasons they have. If it's intolerable for the motivating reasons themselves to be necessary existents, this response won't do. Maybe it's thought that it is tolerable for the motivating reasons to be necessary existents provided that the having of them is contingent. Note that what allows him to say that the having of them is contingent is the claim that it is only when we are in certain mental states that we have them and it is a contingent matter as to whether we are in such mental states. I think that's sort of right, but I think it's not going far enough.

It's clear that there's a difference between believing that the proposition p exists and believing p to be true. It's contingent whether someone believes any propositions to exist and it is contingent whether someone believes any proposition to be true. The question I have is simple. Why wouldn't knowing a reason to exist be sufficient for having the reason? If knowing a reason to exist is sufficient for having the reason, then I don't think that the account is going to get the modal profile of having a reason right. Here's why:

Knowing a motivating reason to exist is sufficient for having it.
Propositions are known to be necessary existents by the philosophically informed.
If reasons are propositions, the philosophically informed (who believe motivating reasons to be propositions) will know that they have all the same reasons in every possible world where they know propositions to be necessary existents.

5 comments:

Errol Lord said...

I'm sorry, but I'm really not sure how this is suppose to work. You ask:

Why wouldn't knowing a reason to exist be sufficient for having the reason?

It depends on what you mean by having and what you mean by reason. I assume that by having you mean something like this: A has motivating reason R to P just in case if A were to P, then a correct explanation of why A P-ed would cite R (this can't be right, but I take it this is the rough idea). I take it by reason you mean the things that are candidate motivating reasons (note that the way you put this is unfair to Miller. The propositions that would be motivating reasons if the background conditions were met are not motivating reasons if the background conditions aren't met. Compare: The food I'm going to eat for supper tonight isn't food I've eaten until the background condition that I eat it is met.)

If that is what you mean by have and by reason, then something like your argument works equally well against any non-skeptical view; for no matter what your view is about the ontology of motivating reasons, if you're not a skeptic, then whatever world you're in, you'll know that all the reasons that exist in that world exist in that world.

Knowing a motivating to exist is sufficient for having it.
No matter what world one is in, one knows that all the motivating reasons that exist in that world exist in that world.
Thus, no matter what world one is in, one has all the motivating reasons that exist in that world


But this is surely just as implausible as the conclusion of your argument. For I know that there are some things in this world that are the types of things that could be my motivating reasons even though I know I don't have them. I don't know everything, after all (side note: isn't this an objection to the view that motivating reasons are mental states? For isn't that person going to be fine with my argument above?). There are plenty of times where there are facts such that, if I were aware of those facts (and not just aware that they existed), I would act differently than I actually do.

Now that I've run through that, I'm thinking you must mean something different by 'knows that they exist'. But it's not like you know what the reason is about (the content if you go for propositions; the state of affairs represented if you go for facts) just because you know that it exists (which, it seems to me, is what you need).

Clayton said...

Errol,

Here's Miller's gloss on what a motivating reason is:
"motivating reasons are reasons by the agent’s own lights, and thus from the agent’s perspective serve to implicitly justify the action as well as the formation of mental states which bring it about. To take a simple example, what by a wife’s lights is the fact of her spouse’s infidelity may go a long way towards explaining why filing for divorce would seem to her to be worthwhile, as well as help justify her desire to do so."

I take it that Miller thinks that motivating reasons are propositions because he thinks that what the agent takes to be a justifying reason for an action will be a propositon. That seems to have the weird implication that the thing that the agent takes to actually justify A-ing (say, p when they believe p and believe p to count in favor of A-ing) is something that exists in all possible worlds regardless of whether p.

You're right that he rejects the view that the subject has these reasons when the subject doesn't have certain beliefs and you're right if you're saying that Miller isn't committed to the unfortunate view the agent's motivating reasons are the same in all possible worlds where we find the agent (or find the agent with the agent's actual aims). I don't deny that. I think the first bit of awkwardness is simply that the thing that Miller thinks the agent identifies as her reasons for A-ing are things that would have existed even if her attitudes were different and even if her beliefs were mistaken.

My second worry is just this. What entitles someone who holds the reasons as propositions view to say that the enabling condition required for p to be S's reason for A-ing to be (i) S's believing p to be true rather than (ii) S's believing p to exist?

Errol Lord said...

Maybe I should have led off with this, but I don't really find the consequence that the things that could have been reasons exist in all possible worlds to be that 'awkward.' Think about beliefs. Ian believes P and [if P, Q]. Because of this he comes to believe Q. Is it unpalatable in the same way as the practical case to point out that the very things that Ian thinks motivate a move to Q exist in all possible worlds? Not really because belief is presentational. I just don't see why this is that bad.

What entitles someone who holds the reasons as propositions view to say that the enabling condition required for p to be S's reason for A-ing to be (i) S's believing p to be true rather than (ii) S's believing p to exist?

I thought the aim of the game was to give the account that yields the best explanations of intentional action. It seems like the view that says the background condition is merely believing the reason to exist is really implausible. It predicts that sometimes we can explain why I phi-ed by appealing to propositions like ''p' exists.' 'P' and ''P exists' present very different contents as being true. Another reason why one can insist that the relevant background condition is belief in the proposition instead of a belief that the proposition exists is that it is constitutive of some propositional views that in order for some proposition to be a motivating reason, the proposition must be a normative reason if it were true. But the proposition ''p' exists' wouldn't be a normative reason if it were true.

Clayton said...

"Is it unpalatable in the same way as the practical case to point out that the very things that Ian thinks motivate a move to Q exist in all possible worlds? Not really because belief is presentational. I just don't see why this is that bad."

Assuming that the p's and q's are contingent, I guess I'd say that it seems bad and I don't know what the presentational nature of belief has to do with it. I think that the practical case is bad because I think the things we think move us are things we often think could have been different, would have been different if our opportunities were different, etc... That's why I thought that MRs are more plausibly taken to be facts, states of affairs, or maybe even states of mind.

"I thought the aim of the game was to give the account that yields the best explanations of intentional action."

Agreed.

"'P' and ''P exists' present very different contents as being true."

Agreed.

"the proposition must be a normative reason if it were true."

Is this assuming the identity theory of truth? I mean, I think that the identity of MR and NR is precisely why we should say that MR are facts when the agent is getting things right and that's why they aren't propositions. On the fact view, if I believe the fact to exist, I'll believe that I have the reason and believe a proposition to be true. On the proposition view, won't some of us realize (assuming the view is correct) that the thing that is my reason when I believe p and believe p to count in favor of A-ing is something I could have had even if ~p? That's part of what troubles me. Because if that's part of the view, I'm left wondering why the propositionalist gets to help himself to saying that you have to actually believe p to be true for p to be your reason when the reason is not the fact or the state of affairs that makes p true but something you know you could possibly have as your reason in worlds where ~p.

Errol Lord said...

Assuming that the p's and q's are contingent, I guess I'd say that it seems bad and I don't know what the presentational nature of belief has to do with it. I think that the practical case is bad because I think the things we think move us are things we often think could have been different, would have been different if our opportunities were different, etc... That's why I thought that MRs are more plausibly taken to be facts, states of affairs, or maybe even states of mind.

Wait, so are giving up the view that the contents of beliefs are propositions? Or are you claiming merely that facts are what 'move us' (whatever that amounts to) to accept Q when we have the other beliefs? If it's the former, then I think (as you know) that we have to think hard whether your view is throwing the baby out with the bath water. Also, the states of mind view can't really get the right results unless the state of mind one has in mind is factive.

My thought about the presentational nature of belief is simple: Due to the fact that belief is presentational, when we believe p, p is presented as true. That's why it doesn't seem so bad to me in the belief case (in either case, really) to say that the things that can be the reasons necessarily exist. It's because when you believe p you settle that this world (at least right now) is a p world. Once this is settled, who cares if there are some worlds that are not-p worlds where p exists?

On the fact view, if I believe the fact to exist, I'll believe that I have the reason and believe a proposition to be true. On the proposition view, won't some of us realize (assuming the view is correct) that the thing that is my reason when I believe p and believe p to count in favor of A-ing is something I could have had even if ~p? That's part of what troubles me. Because if that's part of the view, I'm left wondering why the propositionalist gets to help himself to saying that you have to actually believe p to be true for p to be your reason when the reason is not the fact or the state of affairs that makes p true but something you know you could possibly have as your reason in worlds where ~p.

Ok, I see an argument here that I agree brings out a problem. I didn't see before that the heart of one of the problems is the intuition that when I know some particular fact to exist I have that as a reason. It doesn't seem like the propositionalist can account for that intuition. But, of course, the propositionalist can say that whenever you know some fact to exist you have a reason (if some other constraints are met). Because she can insist that once you know that the fact exists you believe the proposition that represents the fact.

I still fail to see why the intuition shows that the the view that one has to believe the reason to be true is unmotivated.