Friday, May 22, 2009

Internalist supervenience for externalists

Suppose you know p and that you have an internal duplicate that believes p for roughly the same reasons. A commonly held intuition is that you and your duplicate are justified in believing p to the same degree. Suppose that’s right. A commonly drawn lesson is that justification supervenes on the intrinsic properties of a thinker. Suppose that’s right. A commonly drawn lesson is that the justification of our beliefs is determined by these intrinsic properties. That might be a mistake. Even if two intrinsic duplicates cannot have beliefs that differ in justificatory status, someone might say that the reason that the second subject is justified is that the second subject is an intrinsic duplicate of someone who has knowledge. That property might supervene on the intrinsic properties but there’s a perfectly good sense in which that property isn’t determined solely by that individual’s intrinsic properties. Suppose you hold the view that to be blameless before the law you don’t have to act like a lawful citizen, you only have to be motivated in the way that lawful citizens are. Maybe I intend to drive on the left and I have a very similar twin that intends to drive on the right. Whether one of us is the same on the inside as a lawful citizen depends (in part) upon the traffic regulations. Nevertheless, anyone who is just like the one of us that intends to drive on the correct side of the road is blameless before the law but only if they are the same on the inside as the one who intends to do something that is ‘turns out’ to be consistent with the regulations.

This might seem to be a minor point, but I think it matters in the epistemology of moral judgment. Someone might think that NED intuitions support the phenomenal conservative view that says that your judgment that p is true is justified if it seems that p is true and there’s no reason to think things are amiss. They don’t, however. Suppose someone’s moral judgment is based on an intuition where it could not be that the moral world is the way that the intuition makes it seem. According to PC, as long as they have no reason to think things are amiss it seems that the moral judgment based on the intuition is justified. According to externalism with supervenience internalism (ESI), that doesn’t follow. The person who has the sort of intuitions that lead to cannibalism isn’t the same on the inside as someone who knows what to do when trying to decide what to hunt up for dinner.

[Juan and I discussed this point earlier when I wasn't so sleepy and he has made a similar point in his Phil Perspectives paper.]


Pete Mandik said...

Hi Clayton,

This stuff about intrinsic supervenience with extrinsic determination sounds contradictory to me. If the second member of a pair has a property in virtue of being an intrinsic duplicate of the first member, then it strikes me as quite odd to say the property in question supervenes on intrinsic properties. Part of what I'm getting hung up on is that the property of being an intrinsic duplicate is not itself an intrinsic property, it's relational. (I'm assuming that being a duplicate supervenes in part on that which the duplicate is a duplicate of.)

Am I missing something here?

Mike Almeida said...

Take the lawful citizen whose lawfulness (allegedly) supervenes on his motivation. Let L0 and L1 be motivated to drive on the left. Let L2 be an intrinsic dupicate of L0 and L1. If L0 lives in the US and L1 in the UK, then (if your view is right) L2 both is and is not law abiding. Can't be right. The problem, I think, is that the L2's lawfulness is not supervening on his intrinsic properties. Minimally, supervenience has to ensure that, necessarily, L2 has intrinsic properties P0, . .Pn only if L2 has legal property Pn+1. But this obviously fails in this case. If lawfulness did supervene on intrinsic properties in these cases, then L2 would instantiate an impossible property.

Clayton said...

Hey Mike and Pete,

Here's what I have in mind. Suppose I know I ought to A and there's an intrinsic duplicate of me for whom everything seems to them the way it seems to me. Intuitively (say some) it seems that this subject is as justified in his judgments as I am in mine.

Two possible lessons.

1st: No J difference without an intrinsic difference, so any intrinsic duplicates are J-duplicates.

2nd: The J-facts are determined by facts about how things seem and not any further facts. So, if the it seems to the subject that p and there's no available reason for them to think things are amiss or reasons to think ~p, then they are justified in believing p.

The 2nd doesn't seem right. Suppose there's someone for whom it seems they should engage in terrorist activities. They aren't the same on the inside as anyone who knows that they ought to do what they judge they ought to do. So, you can have a view that accepts the 1st point but not the 2nd point. Everyone who is the same on the inside as a terrorist has unjustified beliefs. Everyone who is the same on the inside as a saint has justified beliefs. That you are the same on the inside as someone who makes a knowledgeable judgment depends upon something more than just the mental facts in virtue of which things seem to you a certain way.

Mike Almeida said...


If I've got this, you're claiming that, in the case I describe, L0 and L1 are not the same intrinsically. L1 has the wrong motive and L1 has the right motive, and that makes them intrinsically different. It seems like you want to say that relational properties of certain mental states make intrinsic differences to those states. But that's hard to believe as a general claim (as a claim that includes beliefs that are not de re and do not contain proper names). Suppose I believe that it's raining. Am I in different mental states as the weather changes?

Clayton said...

Hey Mike,

My initial example was poorly chosen precisely because of the case you describe. Here's a better way of putting the point.

Suppose we say that any two duplicates will try to perform actions that have the same deontic status.

If I know that I shouldn't A, then anyone who is a duplicate of me is right to refrain from A-ing.

While it's true that for any duplicate of me it will seem to them that they shouldn't A (say, in light of their beliefs and intuitions), but it doesn't follow that intrinsic properties determine that some particular agent performs an action that is permissible. It's possible that someone's intrinsic duplicates all perform impermissible actions. (Here, think about intrinsic duplicates of terrorists and cannibals.) So, _being the same on the inside as someone who acts permissibly_ is not an intrinsic property but if two subjects are the same on the inside there won't be a deontic difference.