In his recent Moral Realism: A Defence, Russ Shafer-Landau argues that “moral facts are themselves intrinsically reason-giving, i.e. supply reasons for action regardless of the content of specific moral demands and their relation to other intrinsically or necessarily reason-giving kinds of considerations” (204). While he acknowledges that such intrinsically reasoning-giving facts may seem mysterious, Shafer-Landau attempts to show that we do, in fact, embrace such in other domains, particularly epistemology. I’m a bit suspicious of this move.
Here are a couple of representative passages from Shafer-Landau, defending brute epistemic normativity (which, if accepted, would make intrinsically reason-giving moral facts seem less unusual and improbable):
I think that intrinsic normativity is ineliminable. To see this, consider the parallels between conditions of epistemic and moral assessment. We say that agents, if they have reason to believe anything at all, have reasons to believe the truth, and to conform their reasoning to truth-preserving schemas, even if believing the truth is not conducive to the goals they set themselves (205).
I believe that there is intrinsic reason to think that two and two are four – the fact itself provides one with reason to believe it. […] The basic idea here is that certain things can be intrinsically normative – reason-giving independently of the value actually attached to them by agents (206, emphasis added).
Shafer-Landau’s example initially seems quite compelling – after all, don’t we have reason to believe the truth simply because it is the truth? (And this even if the epistemic can be overridden in some cases by pragmatic, moral, or other concerns…). Compare another case where we find intrinsic reasons. If you desire that p, then it seems that you have reason to promote p, other things being equal. That is, the very fact that you desire something gives you at least a prima facie reason to promote or pursue it; this is the nature of a desire. But now notice – there is an intrinsic reason here that arises out of the psychological state of desiring, not because of some external fact about the object of the desire. Simply insofar as you desire that p, you will have reason to promote p; we need not appeal to any additional, special normative properties of p itself. Similarly, in the case of belief, note that to believe that p is precisely to believe that p is true; it is inherent to the psychological state of belief that it aims at truth (Jonathan Adler and Ernest Sosa, among many others, argue for this claim). So while we have reason to believe the truth in general, and specifically that two plus two equals four, this is because the psychological state of believing by its very nature aims at truth (and it is true that two and two equal four). As in the case of desiring and its reasons, intrinsic reasons here arise out of the nature of the psychological state, and we need not posit any intrinsic reason-givingness that attaches to all true propositions as Shafer-Landau proposes. Thus, Shafer-Landau’s apparent appeal to epistemic normativity fails, I think, because he is actually focusing on the normativity of belief – and here the normativity is a matter of this psychological state, and not because of external reason-giving facts of the kind that Shafer-Landau is positing in the case of moral realism.
One could claim that just as beliefs aim at truth inherently, perhaps all actions must aim at moral rightness inherently. But this seems unlikely; there are many valid kinds of reasons for our actions, and morality does not play the same role for action as truth does for belief. We can properly perform actions for aesthetic, epistemic, pragmatic, and other non-moral reasons. For example, a musician can appropriately perform the action of tuning his guitar for purely aesthetic reasons. Again: to believe that p is to believe that it is true that p, and no such relation exists between acting and moral rightness. As such, I’m not sure that an appeal to our having reason to believe the truth provides support for Shafer-Landau’s proposed intrinsically reason-giving moral facts.
Jason,
Some questions or observations. I’ve not read Shafer-Landau’s book, but when he says that moral facts are intrinsically reason giving, is he claiming that they necessarily give us *moral* reasons to act in particular ways? And does he acknowledge the plurality of practical reason kinds you suggest? He doesn’t suggest anything in the quotation you provided about the sort of reasons that moral facts intrinsically provide, and it appears open for him to say that they intrinsically give us reasons for action (understanding reasons in a generic sense). If I’m understanding you correctly, your criticism seems to assume that the reasons in question are somehow specifically moral (perhaps meaning that they are categorical somehow?).
Michael,
Thanks for the questions!
S-L tends to speak of moral facts simply providing reasons for action (tout court). Still, he also states that “moral obligations do entail reasons, because they are a kind of (very powerful) practical reason-that an action is wrong, cruel, or vicious is in itself a reason for any one to refrain from it “(6, emphasis added). Elsewhere he writes “Since birds and trees cannot act on any epistemic (or moral) reasons, such reasons do not apply to them” (207). So I think S-L treats moral reasons as perhaps a species of practical reason.
Still, even if we treat moral facts as intrinsically (generic) reason-giving, I think the same problem arises in his appeal to the epistemic case (as an instance of facts being intrinsically reason-giving). The issue, as I see it, is that in the epistemic case, if we have reason to believe truths (and I agree that we do) this is because the psychological state of belief necessarily aims at truth. It is not because facts themselves give us reasons to believe them (or corresponding propositions). And I think S-L needs the epistemic case to be one where the facts themselves provide us with reasons (to show that we do embrace intrinsically reason-giving facts in domains outside of ethics).
I agree with Shafer-Landau that “moral facts are themselves intrinsically reason-giving.” If some state of affairs is, say, good, then that fact alone provides anyone who is in a position to bring about that state of affairs with a reason (albeit an overridable one) to do so. I’m inclined to think that this last claim is actually analytic – part of what it means for a state of affairs to be good. Thus, in my view, foes of intrinsic normativity must also be foes of objective value. But perhaps that not surprising.
Now, I can’t really get my head around this “belief aims at truth” stuff. So, if I may, I wanted to take a different tack in defending S-L’s claim. I think that *facts about personal welfare* are intrinsically reason-giving as well. But facts about welfare seem to be less controversial. Many people think intrinsic value simpliciter is “queer” and so don’t believe in it. But far fewer people have a problem with value for a person. They shy away from saying that some state of affair is just plain good, but don’t mind saying that some state of affairs is good for some person. But these less controversial facts about welfare are intrinsically reason-giving, too.
The idea is very simple. Suppose some state of affairs would be good for you. This provides you with a reason to bring about that state of affairs. And it does so “whether you like it or not,” as it were. That the state of affairs would be good itself provides the reason.
Now, you might be thinking that welfare is reducible to desire, and that if it is, then we can explain how facts about welfare give reasons by appeal to facts about one’s desires. Reasons are reducible to desire.
But I think this can be done only on what we might call an “actual, present desire theory of welfare.” But no one holds such a theory. It implies that what would be best for you – in the long run – is whatever would best satisfy your actual present desires. But that is silly. It is widely agreed that for a desire theory of welfare to have a chance, it must take into account either one’s future desires or one’s hypothetical desires (such as one’s fully-informed desires), or both.
But then a person can have a reason to bring about some state of affairs even though he has no (actual, present) desire for that state of affairs. So we seem to have an intrinsically reason-giving fact – that the person will, or would, desire some state of affairs. This fact is intrinsically normative: it gives the person a reason to act “whether he likes it or not.”
Some might be inclined to say that there is still in important difference between this case and the moral case. In this case, it’s not an “impersonal world” providing the reasons. It’s the person’s own future self, or his own counterfactual self, whose desire provides the reasons. But why does that make a difference? It still seems to me that we have this fact – that the person will, or would, desire the thing. And this fact is intrinsically normative: it provides the person a reason to get the thing whether the person (presently, actually) cares about the thing or not.
Hi, Jason. In your reply to Michael, you say, “The issue, as I see it, is that in the epistemic case, if we have reason to believe truths (and I agree that we do) this is because the psychological state of belief necessarily aims at truth. It is not because facts themselves give us reasons to believe them (or corresponding propositions).” I take it that “the psychological state of belief necessarily aims at truth” means “to believe that p is precisely to believe that p is true.”
Let’s grant that to believe that p is to believe that p is true. How does this provide a *reason* to believe that p? The truth-aimingness of belief might go some distance towards *explaining* why I believe that p (simpliciter). But I would think that the normative *reason*, or justification, for my belief that p would be something about p itself. The reason that I should believe that there’s a computer monitor in front of me is not because my belief that there is a monitor in front of me tacitly implies that I believe it’s true that there’s a monitor in front of me. Rather, the reason that I should believe that there’s a computer monitor in front of me is because there *is* a computer monitor in front of me. Right? But then aren’t the facts about my monitor providing me with a reason to believe? That is, aren’t they reason-giving?
Can someone clear up something about the analogy (between epistemic and moral reasons)? I do own the book but I have lent it to someone so I can’t check this myself.
The idea that moral facts are intrinsically reason-giving is, as I understand it, the idea that facts like that I am obligated to return the library book provide me with reasons. But the example of an analogously reason-giving epistemic fact is that two plus two equals four. That is not analogous at all. An analogous epistemic reason-giving fact should be a normatively picked out fact, like that we are warranted in believing the results of the opinion poll.
The difference seems pretty important to me, especially if the point is supposed to have some bearing on whether realism is true (or plausible, whatever).
Jamie,
I don’t have my RS-L with me, but that sounds right. Maybe you could say a bit more about the implications for the broader point, though. After all, it seems intuitive to say that we are warranted in believing the poll results actually does give us a reason to believe those poll results. In which case, it shouldn’t seem uniquely strange to say that I am obligated to return the library’s book gives me a reason to return it.
At the same time, instead of saying that moral facts can generate reasons, perhaps (I’m flying by the seat of my pants here, since, again, I don’t have the book with me) RS-L should have said that morally relevant, but non-moral, facts, can be reason giving, and in that sense they’re like epistemically relevant but non-epistemic facts, such as that two plus two equals four. E.g., that I have promised to return the library’s book gives me a reason to return it, just as that two plus two equals four gives me a reason to believe that two plus two equals four.
Thanks for the comments everyone,
I’m only going to respond to one of Josh’s comments for now, as I’ve forgotten my S-L at my office, and I want to check on a couple of points before saying more…
Josh, in your comment you wrote:
“the reason that I should believe that there’s a computer monitor in front of me is because there *is* a computer monitor in front of me. Right? But then aren’t the facts about my monitor providing me with a reason to believe? That is, aren’t they reason-giving?”
I’d say the following: Yes, you have a reason to believe that there’s a monitor in front of you. But why? Because it is true that there is a monitor in front of you, and beliefs aim at the truth. [This is probably more confusing than helpful, but, we might say: If beliefs -somehow – didn’t aim at the truth, then the mere fact that there is a monitor in front of you would not give you a reason to believe that there is a monitor in front of you. It’s insofar as beliefs aim at the truth that you have reason to believe truths – including that there is a monitor in front of you… On the other hand, I’m not sure how to understand a ‘belief’ that doesn’t aim at the truth, so I’ll put this aside…]
When we turn to the moral case, this doesn’t seem to work [at least as I’ve initially understood it – I want to look over S-L again to consider Jamie’s question.] Actions don’t aim at moral rightness in the same way that beliefs necessarily aim at the truth. So while facts can give us reasons for belief (again, insofar as beliefs aim at the truth), I’m not seeing how moral facts (or facts like “I promised to do X”) give us similar reasons for action, as actions don’t have a necessary aim or target (unlike beliefs). To have a strictly parallel case, we’d need to hold that actions necessarily aim at moral rightness, or something like this.
In your response to Jamie, you write (as a possible modification of S-L):
that I have promised to return the library’s book gives me a reason to return it, just as that two plus two equals four gives me a reason to believe that two plus two equals four.
That 2+2=4 gives you reason to believe that 2+2=4 insofar as beliefs necessarily aim at truth (and it’s true that 2+2=4). But I’m not sure what to make of the other case – ‘that I have promised to return the library’s books’ gives me a reason to return them insofar as actions …..[fill in blank].
I assume S-L wants to say that nothing fills in the blank; that the promise just provides a reason for action (or – that I’m obligated to X provides a reason for action). Maybe this is so, but then this seems quite different from the case of belief and the kind of reasons we find there. And the ‘partners in crime’ strategy wouldn’t work.
Jason, I’m not sure what to think about your particular version of the argument, because I do not believe that something’s being true gives you any reason whatsoever to believe it. For example, I am pretty sure that I do not have any reason at all to believe that there are an odd number of stars in the Milky Way. I also think it is pretty doubtful that belief aims at the truth. Still, that at least seems like it must be related to something right. (Beliefs are correct when true, for instance.)
Josh,
Well, my own view about this is that the fact that we are obligated does not give us a reason. That, I think, would be a kind of double-counting. Whatever it is that grounds the obligation (the plain, non-normative fact), I think, is what gives us a reason. And, it seems to me, the same is true in the epistemic realm. Not the fact of warrant, but the fact doing the warranting, is what gives us a reason; it’s the evidence, not the fact that it is evidence, that provides me with a reason to believe a scientific hypothesis, for example.
If this way of looking at things is right, then it’s hard for me to see how it is related to realism. Briefly: which things give me reasons is itself a normative question, so my answer to it should depend on my normative views, not on my meta-ethical views.
Jamie, thanks — that clears things up.
And, thanks for the reply, Jason. Jamie’s point, that something’s being true doesn’t seem to, of itself, generate any normative oomph, seems intuitive to me. I also wonder whether S-L couldn’t say something like the following:
Let’s grant your thesis that my reason to believe that 2+2=4 is generated by two facts — (1) the truth of 2+2=4, and (2) the fact that belief aims at the truth. Even so, notice that here normativity arises out of purely descriptive facts. (Again, and with Jamie I think, I’d want to insert a normative premise that I should believe the truth in order to get the normative claim that I should believe that 2+2=4, but let’s put that aside here.) And, that’s all that S-L needs to make his ‘partners in crime’ strategy work: the analogy of epistemic normativity with moral normativity. That is, S-L could maybe grant that the two kinds of normativity are not in all respects alike, but hold that they are alike in all relevant respects, in particular that both kinds of normativity are generated straight out of descriptive facts.
Apologies – this kind of rambles…
With respect to something’s being true giving you reason (or not) to believe it, I’d like to distinguish a couple of questions:
1) Do we have a reason to form a belief on a given issue or question? For example, does the mere fact that there are a certain number of grains of sand on a particular beach give me reason to investigate, and form a belief about this fact?
Here I’d agree, mere truths do not give us reason to believe them. [Perhaps there are various pragmatic concerns or what-have-you that will dictate whether we have reason to investigate and form a belief.]
2) Given that we are forming a belief with respect to some issue (say, whether or not it is true that p), does the fact that p is true give us reason to believe that p?
Here I’m inclined to say that the fact that p is true does give us reason to believe that p. A belief is correct /successful when it is true. [Jamie, could you say a bit more about why you think that it’s “pretty doubtful that belief aims at the truth”?]
Such a reason for belief would exist, even if we lack evidence. That is, if it were in fact true that whales are fish, I would have reason to believe that they are fish, even if all my available evidence suggests otherwise. [This would be akin to thinking that you have reason to do what is in fact in your best interest, even if your evidence suggests that something else would be in your best interest.]
There would be another kind of reason at stake too, as Jamie suggests. I have reason to believe what is indicated (to be true) by my evidence (even if the evidence is misleading). So, that the thermometer indicates that the temperature is 42 degrees, gives me reason to believe that the temperature is 42 degrees. My reason for believing (based on evidence) can, of course, come apart from my reason for believing what is in fact true.
There is the question of “Should I care about whether my beliefs are correct / successful?” or perhaps better put, “Even if the correctness of a belief is a matter of its being true, does this give us reason to form true beliefs?”
I’m not sure what to say here, but notice that if the correctness / success of a belief (i.e., its truth) only provides a reason for believing a truth given some further (non-epistemic) reason, then this would seem to show S-L to be in a more problematic situation. Because now facts / truths (even if we treat them in terms of evidence as Jamie suggests) would only provide reasons for belief (in some true proposition) given some further reason; and we would not obviously have intrinsically reason-giving facts of the kind S-L suggests.
whew.
Jason,
Given that we are forming a belief with respect to some issue (say, whether or not it is true that p), does the fact that p is true give us reason to believe that p?
No. Given that you are forming a belief that about the parity of the number of stars in the Milky Way, you have no reason to believe that there it is an even number. The mere fact that the number is even would in no way make it reasonable for you to believe that the number is even. If you did come to believe that the number is even, you would do so for no reason whatsoever.
That’s how it seems to me, anyway.
The thing about belief aiming at the truth is that it’s a metaphor, and I can’t see what unmetaphorical truth it might be expressing. Beliefs are correct when they are true. Insofar as the ‘aiming’ metaphor is supposed to say more than that, I bet it’s wrong.
Jamie,
In the case you mention, I’d grant that we lack a reason to form the belief, in the sense of internally-accessible evidence. So yes, we could deem such a belief (even if true) unreasonable in that respect (perhaps we could say ‘unjustified’).
But such a belief would be true (and correct). As such, it still seems to me that there is reason for you to form this belief concerning the parity of the number of stars.
I can’t quite put my finger on this, but – would it be strange that I can have reason to believe that p only when I have evidence that p is true, and will lack reason to believe that p (even though it is true that p) if I lack evidence? To be evidence is to be evidence that something is true. We care about having justified beliefs (or reasonable belief, in the sense of evidence)primarily because these are likely to be true.
How about this: the reason you have for believing propositions for which you have evidence is derivative from the reason you have to form true beliefs?
[Fair enough about beliefs ‘aiming’ at truth; I don’t know how to cash-out the metaphor, but it does seem to capture something of the nature of beliefs…]
Josh,
In your most recent comment, you suggest that in the epistemic case, we have a reason to believe generated by purely descriptive facts (that 2+2=4, and that belief aims at truth), and that this would be enough for S-L’s partners-in-crime strategy to work:
“That is, S-L could maybe grant that the two kinds of normativity are not in all respects alike, but hold that they are alike in all relevant respects, in particular that both kinds of normativity are generated straight out of descriptive facts.”
Hmm. I’m still not sure that this is enough. The differences between the cases still seem rather important. That is, it seems to me that the normative work in the epistemic case is being done by the descriptive fact of beliefs aiming at truth (and it just happens that 2+2=4 is a truth being considered); and without something parallel on the side of actions, the analogy really doesn’t hold.
If beliefs didn’t aim at truth, then I don’t think the fact that 2+2=4 would give you reason to believe it. So, without something parallel, I can’t quite see how the moral case is supposed to work.
On ‘belief aims at truth.’ Might the unmetaphorical truth it expresses or conveys just be that we (rational agents) aim to have true beliefs rather than false ones? That, in that sense at least, belief itself aims at truth? This certainly says more than just that beliefs are correct when true.
Actually, it seems this isn’t all that people have in mind by ‘belief aims at truth.’ It seems there’s the stronger suggestion that belief by its very nature is such that one can only (rationally) aim to acquire true beliefs not false ones. Is that what the idea that to believe something is to believe that it is true is supposed to be getting at? Come to think of it, ‘to believe that p is to believe that p is true’ does seem to say more than just that the belief that p is correct iff p is true.
Might the unmetaphorical truth it expresses or conveys just be that we (rational agents) aim to have true beliefs rather than false ones? That, in that sense at least, belief itself aims at truth? This certainly says more than just that beliefs are correct when true.
I don’t see it. What does it say, aside from the fact that beliefs are correct when true?
An archer aims at a target. But to do that, he has to be able to tell where the target is. Otherwise, he cannot aim at it. Compare a believer. Once a believer is able to tell which are the truths, his work is done. He doesn’t then have to aim at them, or aim to believe them, or anything like that. Once he can tell which are the truths, he already believes them.
Come to think of it, ‘to believe that p is to believe that p is true’ does seem to say more than just that the belief that p is correct iff p is true.
To me it seems to say less. The motto, ‘to believe that p is to believe that p is true’ seems to me to say absolutely nothing whatsoever about belief. Compare: to desire that p is to desire that p is true. To hope that p is to hope that p is true. To fear that p is to fear that p is true.
For anyone who is interested, Jon Kvanvig has recently put up a related post over at Certain Doubts.
Jason,
I wonder if you could clarify something for me. Is your point that facts do not in fact provide intrinsic reasons to believe things, but rather only provide epistemic reasons because they stand in some relation to something about the nature of belief? This was what I took to be the force of the point that facts provide reasons only because belief aims at truth but you also seem to say that belief aiming at truth explains how facts provide intrinsic reasons. But if facts are reasons only because of something about belief surely they aren’t intrinsic reasons after all?
Also, I didn’t quite get the point in the last paragraph either. Agreed, there are many different kinds of practical reasons, but suppose *one* of the things that action (or, perhaps, desire) aims at is moral rightness. Wouldn’t that be enough to establish, on your model of epistemic reasons, that there are moral reasons?
Hi Jonny,
Thanks – you’re right, I’ve sort of wavered around a bit on facts giving intrinsic reason for beliefs.
And yes, I’d want to say that it is only because of the kind of mental states that beliefs are that we have any reason to believe that any facts obtain (where this is because beliefs aim at truth). So any normative work here is being done by the kind of state beliefs are, not by the facts themselves (e.g. that 2+2=4). In a few comments I carelessly suggested that facts themselves give you reason to believe them.
As to moral rightness being *one* of the things at which actions (or perhaps desires) aim. Here I might point to the possibility of amoralists, whose actions do not seem to be aimed at moral rightness at all. [I can’t really picture a parallel in the case of belief.] Certainly some actions aim at moral rightness, but I don’t see that they all must so aim -even as just one aim among many (which is a significant difference from the case of belief). As another case, imagine a basketball player passing the ball; I don’t see that his action is aiming at moral rightness in any way.
[I suppose at some point, we’ll need to clarify the aiming at stake in the moral case – moral rightness as motive, or moral rightness as an end, etc.]
I guess the idea would be that desire or action aims at the good in some very general sense, encompassing moral goods, prudential goods, aesthetic goods and so on and so forth. A non-instrumental desire then counts as correct just in case it’s for something that’s actually good, in this broad sense. Because moral ends are one of the good things it follows that if X-ing promotes some moral end then there is a reason to X, thus there are moral reasons. The amoralist is only a counter-example if his actions/ desires don’t aim at (what he takes to be) good in any sense whatsover. I take it that’s not plausible – he can still think prudential/egoistic ends are good, for example. Similarly we don’t need every action undertaken to be (believed) morally good, we only need that one of the ways in which an action/ desire can be correct is if it’s morally correct.
I realise that this leaves a whole bunch of issues unresolved, and certain important disanalogies with the belief case remain, and I’ve also just ignored all the legitimate worries that Jamie and others have raised about the usefulness of the whole “aiming at” talk, but this is how I think the outline of the view would go.
Jonny,
Sorry for the delayed response – break and all…
I think I see what you’re getting at, but the cases still seem relevantly different to me.
In the moral case, it seems up to the agent whether or not she wishes to act morally (she might be focused on other goals – aesthetic, or what-have-you).
In the epistemic case, if an agent is forming beliefs, they necessarily aim at truth, regardless of the agent’s interests. No matter the agent, or her goals, simply because of the sort of state belief is, it will aim at the truth.
You might say that there is still reason for an agent to act morally in a case where she chooses to act ‘aesthetically’ (that is, the moral fact(s) are intrinsically reason-giving even if the agent chooses to act otherwise). Perhaps, but then it seems like you’d need some additional argument for this; and this is relevantly different from the epistemic case that S-L is appealing to as a partner in crime (again, in the epistemic case, every agent’s beliefs aim at truth because of the nature of the state).
As a general parting note, I had been speaking of having intrinsic reason to believe truths (given that one is forming a belief on a certain question). To that extent, I had been in agreement with S-L (I was just arguing that the epistemic case was unlike the moral case that S-L was defending).
Given Jamie’s comments, and what Jon Kvanvig has said over at Certain Doubts, well, I’m not sure what to say now. I still think that there are important differences between the moral and epistemic cases that S-L appeals to, but I’m not sure that I’d still want to speak of in terms of reasons to believe truths, rather than just saying that beliefs necessarily aim at truth (however this is cashed-out), or that beliefs are correct/successful when true [whereas I don’t think there is a strict moral parallel in the case of actions].
I have a related question regarding SL’s “Moral Realism: A Defense.” He alludes to a problem for noncognitivists, namely, the multiplicity of attitudes problem. It goes something like this:
If noncognitivism is true, then to say something is “right” or “good” is just expressing a favorable attitude. But what of other pro-attitudes, such as, “X is just” “X is brave” “X is wise” etc. The noncognitivist seems to be committed to either claiming that all of these moral attributes reduce to a pro-attitude (a troublesome option), or to make a distinction between a pro-bravery attitude, a pro-wise attitude, etc – which also is troublesome for the non-cog. because we can imagine cases where “X deserves justice” is a compliment (in a case in which we have a pro-attitude toward someone getting her wallet back) and “X deserves justice” as a con-attitude if we imagine that x is dahmer and justice is death.
Does anyone know of any responses noncognitivists to this general problem??? Thanks, Damien