As most of you know, Daniel Star and Stephen Kearns have
recently made an interesting, original proposal of what reasons are generally
speaking (see, for instance, their ‘Reasons as Evidence’ in OSME 4).On their
view, to be a reason is to be evidence for an ought. I’ve started to wonder how
compatible this view would be with Dancy’s reasons-holism which attracts me.
So, I want to start from a triad of claims that seem inconsistent to me in the
light of certain plausible empirical and normative assumptions. I then want to
ask what should be given up.
Here are the three inconsistent claims:
as Evidence: Necessarily, a fact F is a reason for an agent A to phi iff F
is evidence that A ought to phi (where phi is either a belief or an
action) (Star & Kearns 2009, 216).
about evidence: q would be evidence for p just if q raises the probability
of p, that is, the conditional probability of p on q is higher than the
unconditional probability of p, Prob (p) < Prob (p | q) (Williamson
about Reasons: A feature that is a reason in one case may be no reason at
all, or an opposite reason, in another (Dancy 2004, 73).
inconsistency, we need a back-ground story with certain empirical and normative
assumptions. So, here’s one very rough example.
Suppose that I have borrowed Daniel £20. It must be true
that this raises the probability of that I ought to give £20 to Daniel. So, the
fact that I have borrowed Daniel £20 is a reason for me to give Daniel £20. And,
this is invariably so. The likelihood that I ought to give Daniel £20 is higher
in the set of worlds in which I have borrowed £20 from him than it is in all
Now, you might protest that it is not always even true that
I ought to give Daniel £20 if I have borrowed £20 from him. True. Imagine that
I knew that Daniel will use the £20 to build a bomb that he will use to kill a
lot of people. In that case, it is less likely than normally that I ought to
give Daniel £20. So the fact that [I have borrowed £20 from Daniel and that he
will use the money to build a bomb] is a reason not to give Daniel £20.
Imagine then that Daniel will not actually build a bomb with
the £20 (and that I know this). In this case, the fact that [I borrowed £20
from Daniel and he will not use it to build a bomb] makes it even likelier that
I ought to give Daniel £20. Thus, this more complex fact is better evidence and
a stronger reason. In fact, we can make the evidence even better and the reason
stronger. The fact that [I borrowed £20, I promised to pay it back, I have an
extra £20 in my wallet, no children are starving that I could help, Daniel will
not build a bomb…] will make it almost certain that I ought to give Daniel £20.
At this point, Dancy would of course protest that we have not only ruled out
holism but also lost the differences between the kinds of contributions which
different considerations can make normatively. We have included both the
‘favourers’ and the ‘enablers’ in the reason, and thus lost the distinction.
When I first presented this question to Daniel, he first
made an interesting observation. Daniel pointed out that it’s not clear whether
Dancy avoids this inconsistency either. At least in the case of theoretical
reasons, Dancy would accept that evidence for that (one ought to believe) p is a
reason to believe that p (where ‘is’ here is the ‘is’ predication and not that
of identity as for Star and Kearns
certainly does want to accept holism, and it seems likely in the light of
things he says in the end of his book that he accepts some form of probabilism.
So, Dancy too would have to give up one of the claims 1, 2, and 3, if Daniel is
right that he accepts them all.
Of course, that Dancy might be committed to inconsistent
claims doesn’t help Star and Kearns
one of the claims should they give up?
Well, of course, they could give up (3), i.e., holism. They could claim that the arguments for their view are sufficient to refute holism given the inconsistent triad. This
would mean accepting that it is always the complex considerations that are the
proper, invariant reasons, and that no distinction can be made between
favourers and enablers. Of course many people (such as Brad Hooker and Joseph
Raz) do say this. But, I do think this would be a theoretical cost of the view.
They could also give up probabilism. Before this, they could
first try to modify probabilism to more subjectivist version on which evidence
consists of considerations that only raise the probability of the ought
relative to what the agent already believes. They might also accept that ought
is likewise relative to agent’s epistemic situation. I’m not sure how this
helps though – the story about £20 and Daniel seems to still go through even if
we modify probabilism in these ways. We can consider the story from my ignorant
perspective from which I acquire new information. This would make all the
claims in the story still true.
Maybe Star and Kearns
accept some non-probabilistic account of evidence. This account should also be
independent of the notion of reasons for beliefs. I know there are such
alternatives but they are controversial. So, this seems like a theoretical cost
I might think that it would be easiest to give up 1 for
these reasons. 2 and 3 are thus reasons to give up 1 given the inconsistency of 1, 2, and 3. But I know that Daniel and Stephen won’t be convinced.