August 23, 2007

Reasons, Reasoning and Rationality

Today at ANU concluded the conference on Reasons, Reasoning and Rationality, Themes From the Work of John Broome. Speakers: Jamie Dreier (Brown), Nic Southwood (RSSS), Andrew Reisner (McGill), Geoffrey Brennan (RSSS), Garrett Cullity (Adelaide), Daniel Star (CAPPE), Wlodek Rabinowicz (Lund), John Broome (Oxford). I won't try to do justice to all of the interesting papers, but instead will touch on a couple.

Jamie Dreier examined tensions between various formulations of two principles, which he aimed to reformulate and vindicate:

(Buck-Passing) For something to be good is for it to have properties that provide sufficient reason to choose, prefer, ... or admire it.

(Subjectivism) R is a reason for S to Phi iff R explains why Phi-ing promotes something S wants.

We had the most fun thinking about the Narcissus Bomb Example:

Philosopher chemists at Washington University have invented a Narcissus Bomb. Once triggered, this bomb will explode unless it is in the presence of someone who admires it enormously. You are now in the presence of a triggered Narcissus Bomb and nobody else is in the room.

This part of the talk was designed to show the limitation of Buck-passing. The explosive potential of the bomb is sufficient reason to admire it, but the bomb is evil. Jamie relied here on a fix by Nomy Arpaly, which hypothesizes that the bomb is sufficient reason to make it the case that you admire it but not reason to admire it.

I don't know yet how to think about reasons, but arguably a reason for Phi-ing is a reasoning Psi-ing when Phi-ing entails Psi-ing. But then since 'makes it the case' is factive, 'making it the case that you admire the bomb' entails 'you admire the bomb'. Hence, by the above closure principle, a reason for the former is a reason for the latter, and we're back to the original problem.

Another exciting paper was by Wlodek Rabinowicz, in which he argued Incommensurability is possible if there is vagueness. Incommensurability obtains when two thing x and y are such that neither is better than the other, yet they are not equally good. Wlodek was responding to an argument that Incommensurability is not possible. The argument depended on the following symmetry claim: if it is indeterminate that x is better than y then it's indeterminate that y is better than x.

Berit had a great counterexample to symmetry. Consider: x seems to have the temperature absolute 0. It's determinate that y doesn't seem colder than x, because on all sharpenings of the vague predicate 'seems absolutely cold' it is true that y doesn't seems colder than x. After all, y can't seem colder than absolute 0. But it's indeterminate that x seems colder than y, because on some sharpenings of the predicate 'seems absolutely cold' y doesn't seem absolutely cold. Hence, symmetry fails.


Clayton said...


I'm not sure that I like the closure principle for reasons. You didn't restrict the actions to the same agent, but suppose Sam the security guard has a reason to catch Bill the burglar in the act. It's impossible that Sam catches Bill in the act unless Bill is stealing, but that there is a reason for Sam to catch Bill does not entail that there is a reason for Bill to do something that would get him caught.

If the actions are performed by one agent, I'm not sure that the problems go away. Suppose I've spilled something on your couch. I have reason to apologize sincerely for what I know I oughtn't to have done. I have no reason, however, to have done what I oughtn't to have done. So, it seems that the reasons generated by duties of reparation cause trouble for closure.

Joe said...

Thanks for this Clayton. If we are to analyze good simpliciter, then arguably there is a reason for Bill to do something that would get him caught. Indeed, he should turn himself in.

Can we simplify your second example? It would go something like this: I have a reason to apologize for killing your cat, but it doesn't follow that I have a reason to kill your cat.

Reply: Are apologies factive---in the sense that 's apologized for having brought it about that p' *entails* 's brought it about that p'? Can't i apologize for killing your cat, even though I did not in fact kill your cat. Maybe not, but i don't have strong intuitions here.

Nevertheless, I think your ultimate worry is justified. Putting it as follows seems to clinch it: I have a reason to apologize for doing the horrible thing that in fact I did do. It doesn't follow that I have a reason for doing the horrible thing that in fact I did do.

Clayton said...


I'm not sure whether you can apologize for that which you didn't actually do, but I think that if you can't, you can fix the example in the way you've just suggested.

btw, I'm really jealous. That conference sounds like it was incredibly good.