INTRODUCTION
Stephen Kearns and Daniel Star have recently proposed the following theory of reasons:
Reasons as Evidence: Necessarily, a fact F is a reason for an agent A to Φ iff F is evidence that A ought to Φ (where Φ is either a belief or an action).Footnote 1
In this article I present an objection, inspired by Mill's proof of the principle of utility, to the right-to-left reading of the biconditional. My claim is that the fact that you can perform some action can be evidence that you ought to do it without, itself, being a reason to do it. If this is true then Reasons as Evidence is false.Footnote 2
MILL'S PROOF
In chapter 4 of Utilitarianism, Mill argues for the hedonistic theory that pleasure alone is desirable. He claims that the only evidence that can be given for the view is that we desire pleasure alone.Footnote 3 One charitable way of reading Mill's argument is as combining the principle referred to as ‘Ought Implies Can’ with the claim that we can only desire pleasure in order to conclude that pleasure alone is desirable (given that assumption that something is desirable).
Mill's argument suggests an interesting kind of challenge to Reasons as Evidence. If the fact that you can Φ in C is evidence that you ought to Φ in C then, according to Reasons as Evidence, the fact that you can Φ in C is a reason to Φ in C. This is prima facie problematic. Even those who think reasons come cheaply (that we have many more reasons than we usually report and more than we often believe or take ourselves to have) will baulk at the idea that you have a reason to Φ in C simply because you can Φ in C.Footnote 4 And notice that Reasons as Evidence would not only be committed to the claim that when you can perform some action then there is a reason to do it. Rather, it would be committed to the claim that this fact – that you can perform the action – is a reason to do it (in any case where ability to Φ is evidence that one ought to Φ).
It is tempting to hold that the fact that you can Φ is never, itself, a reason to Φ.Footnote 5 However, for my purposes, I do not need to defend that claim. Here I claim only that there is at least one case where being able to Φ in C is not a reason to Φ in C, whilst nonetheless being evidence that you ought to Φ in C. I describe such a case below. Those who hold Reasons as Evidence must reply to the case in one of two ways. They must either (a) deny that being able to Φ is evidence that you ought to Φ in this case or (b) hold that being able to Φ is a reason to Φ in this case.
CAN THE ABILITY TO Φ BE EVIDENCE THAT YOU OUGHT TO Φ?
If an agent is unable to perform some action then it is not the case that she ought to perform that action.Footnote 6 This is a gloss on the principle ought-implies-can and some version of the principle seems undeniable for at least the kinds of ought claims relevant here, which are practical ought claims.Footnote 7 If we denied this principle we would be left seemingly unable to explain why it is not true that I ought to bring about world peace by clicking my fingers.Footnote 8 An analogous principle holds for reasons. I have no reason to use my microwave to travel back in time and bring about the early discovery of penicillin. The best explanation for this is the fact that I cannot perform this action.
Given that ought implies can, your being unable to Φ is conclusive evidence that it is not the case that you ought to Φ. If you know that you cannot perform some action then you know that it is not the case that you ought to perform that action. Because this is true, it is plausible to hold that – on at least some occasions – being able to Φ is evidence that one ought to Φ.Footnote 9
I grant that it does not always follow from P's implying Q that Q is evidence that P. For whether some fact is evidence for another fact is dependent upon the background facts, and presumably there could be unusual cases in which being able to perform some action is evidence that it is not the case that one ought to perform the action.Footnote 10 Nevertheless, if there is any case in which being able to Φ is evidence that one ought to Φ without at the same time being a reason to Φ, then Reasons as Evidence is in trouble. In this vein, consider the following example:
Bone Marrow: A renowned scientist, close to developing technology that would reverse climate-change a century from now, needs a bone-marrow transplant. A hospital informs you that you are the only possible remaining candidate to be a bone-marrow donor and asks if you would mind being screened. You agree to be screened. You later find out that you are a match.
My contention is that in this case your being able to donate bone marrow to the scientist is evidence that you ought to donate bone marrow to the scientist without being a reason to do so.
Can Reasons as Evidence take Horn A to avoid the case? Horn A was:
(a) Deny that being able to Φ is evidence that you ought to Φ in this case.
This reply is implausible. In this case the stakes were such that you knew that you ought to donate your bone marrow if you were a match. You also knew that if you were unable to donate, this would be decisive evidence that it was not the case that you ought to donate. So when you discovered that you are in fact a match, you acquired evidence that you ought to donate your bone marrow to the scientist. Thus, Reasons as Evidence cannot avoid this case by denying that ability was evidence in this case.
Can Reasons as Evidence take Horn B? Horn B was:
(b) Hold that being able to Φ is a reason to Φ in this case.
But this reply is implausible too. In discovering that you were able to donate bone marrow to the scientist you did not discover additional reasons to do so. You already knew what reasons there would be to donate (the climate-change and subsequent suffering that would be prevented) if it turned out that you could donate. If, before the screening, you were asked what reasons you would have to donate, should you turn out to be a match, it would be false to include among them the fact that you would be able to donate (even though this was evidence that you ought to donate). Reasons are facts that count in favour of actions whereas the fact that you can perform an action is, in and of itself, normatively neutral. If my claims about the case are correct then Bone Marrow is a case in which the fact that you can Φ is evidence that you ought to Φ but not, itself, a reason to Φ, pace Reasons as Evidence.
A general feature of Reasons as Evidence – one which explains the difficulty it has with Bone Marrow and why such a problem is generalizable – is that it cannot allow reasons to have enabling conditions whose obtaining is evidence that an agent ought to Φ. On the Reasons as Evidence view, such enabling conditions must, themselves, be reasons.Footnote 11 But ability to Φ is an enabling condition of reasons to Φ and so can be evidence that one ought to Φ without being a reason to Φ.Footnote 12
