Sunday, October 11, 2009

Xenophobia, partisanship, and epistemic peer disagreement

[Cross-posted from my political blog, Donkeylicious]

A lot of smart people in America are uncomfortable with the idea that they should treat similarly educated folks from other advanced democracies as generally ignorant, deluded, or crazy on global political issues. Instead, we should treat them as 'epistemic peers' -- people just as intelligent as us, working from the same body of evidence, who are roughly our equals in ability to know the truth. In the case at hand, the bodies of evidence differ somewhat, since we have different news sources. But we can mostly solve this problem by sharing our evidence in discussion. If the evidence conflicts and we try to argue that their news sources are unreliable, all we usually have to go on is our news sources, and they can argue the same against us.

Similarly, a lot of smart Democrats are uncomfortable with the idea that they should regard Republicans as generally ignorant, deluded or crazy on global political issues. Many of the same considerations apply here. If we argue that their news sources are unreliable, they can argue that ours are, they're in possession of a basically isomorphic argument. Ordinarily, treating Republicans as epistemic peers would be a reasonable position, just as treating foreigners that way is. But the trouble at our historical moment is that we're no longer able to treat Republicans and educated people throughout the world as epistemic peers at the same time.

I think the following is a fair characterization of the reasoning that resulted in Obama's recent honor: The Republican Party has gone mad and become so destructive of world peace that you get a Nobel Peace Prize for removing them from power. That's an incredibly strong way to to put the point, and I don't know if the consensus of educated people outside America is willing to go quite that far. But if it stops short, it doesn't stop too far short. The 2008-2009 jump in favorable views of America, especially in Western Europe but including many other nations, is a sign of how differently people see Obama-era America from what preceded it. Foreigners will have many different views of what exactly is going on, but they're generally going to include the idea that Republican views on foreign policy are so tainted by the xenophobia, bloodthirst, and misinformation of influential people in the party that they can't be regarded as epistemic peers.

Republicans regard world opinion as badly as it regards them. You can see it even in their relationship with mainstream American opinion, where they've constructed an alternative news infrastructure in Fox News and talk radio that they regard as free from the distortions of the mainstream media. While Democrats have their own preferred blogs and websites, they haven't built full-fledged Fox-News-style alternative versions of mainstream news institutions. Globally, this becomes even stronger. Republicans' relation to respected international institutions like the UN (on the political side) and the BBC (on the news side) has long been hostile. When international weapons inspectors claimed that Iraq didn't have weapons of mass destruction, Republicans ridiculed them. Thinking the Wikipedia editors of the world are biased against them, Republicans created Conservapedia. If you don't regard educated people throughout the rest of the world as your epistemic peers, this is what you do, and maybe you start ordering freedom fries. Of course, this leaves you in a situation where the rest of the world isn't going to think you're an epistemic peer of theirs.

So where does this leave Americans who aren't Republicans? I don't think it's possible for us to treat both Republicans and educated people throughout the world as epistemic peers. This would involve having some level of trust each group's deeply held belief that the opposite group has gone totally off the rails. This leaves you suspecting that two different groups of people are deluded on the say-so of people who you suspect are deluded about issues like who is deluded. That's a pretty bad position, and not one we can stay in very long. We could also just regard global affairs as a huge area of general confusion where nobody knows what is going on, and withdraw from politics. If we're going to continue doing politics, however, we need to decide which group we're going to treat as epistemic peers and whose opinions we're going to regard as tainted by bias and misinformation.

There's plenty to be said about how exactly we should make that decision. But I'm going to conclude this post by observing that the noble intentions of Democrats and independents to treat both Republicans and educated foreigners as epistemic peers about global affairs can't be satisfied in our unhappy world. If we're going to engage in politics, we have to be either xenophobes or partisans. There are no other options.


Justin said...

Philosophy bit: I wonder which version of peerhood you're working with, and which position on the epistemology of disagreement. I myself more or less incline towards the total evidence view, on which it doesn't obviously follow that you can't take yourself to be peers with S & T, where S & T do not regard themselves as peers.

Substantive political bit: exactly what has Obama done that justifies those changes in European opinion? Obviously his attitude is very different than Bush's, and climate change is a big and substantive issue where he looks to act differently, but I dare say a lot of objectionable Bush era policies are being maintained.

Also, whose peers? There is no country or political party whose median voter I view as my peer. I view some Republicans, and more Democrats, as my peers, but not the average voter of either party. And of course some members of either party are my epistemic superiors. So I'm not sure what level of generality you're aiming for.

Neil Sinhababu said...

I'd welcome more comments on the philosophy bit. I don't have a clear view of what positions are available here regarding peers -- I've only read a couple papers on the epistemology of disagreement.

But I think there's some kind of puzzle here. Suppose you regard S & T as peers on a topic like global politics that includes the question of who in the world your peers are. Each of S and T are convinced that the other is their epistemic inferior. If you antecedently regarded both as peers, this should incline you to take more seriously the possibility that one of them is your epistemic inferior. I'm thinking this will bring on a tipping-point situation where further evidence will more easily move towards thinking one of them is your inferior.

Substantive political bit:
-Nuclear nonproliferation agreement with Russia
-Iraq withdrawal over the next 2 years
-Climate change, like you said
-Lower likelihood of starting new unpredictable and ridiculous international conflicts

I agree that 'Republicans' and 'Europeans' are very broad categories. And I don't know if I can precisely describe the person who I'd regard as my Republican peer or my European peer. But I think there's some Republican and some European for whom the above analysis holds, even if we disagree about who exactly they are.

Angus said...

I wonder about this: "Each of S and T are convinced that the other is their epistemic inferior. If you antecedently regarded both as peers, this should incline you to take more seriously the possibility that one of them is your epistemic inferior."

It's certainly true that both S and T agree that one of S or T is an epistemic inferior. So you have two (apparent) epistemic peers agreeing that (P) "one of S v T is your epistemic inferior." Which sure looks like it should give you reason to believe (P) and, indeed, more reason to believe (P) than you would have if only one of S or T acknowledged it. But S derives (P) from:

(A) T, but not S, is epistemically inferior.

And T derives (P) from:

(B) S, but not T, is epistemically inferior.

So we know that (at least) one of S or T has inferred (P) from a falsehood. I'm not exactly clear on where to go from here, but it seems like this should play havoc with the strength of our reason to believe (P). Consider: S's believing (A) gives us a reason of weight W to believe (A) and ~(B); T's believing (B) gives us a reason of weight W to believe (B) and ~(A). So we have reasons of equal weight to believe that (A) / (B) are true / false. Likewise, then, we have reasons of equal weight to believe that deductions from (A) / (B) are true / false. So considerations of epistemic peerage give us reasons of equal weight to believe (P) and ~(P). Doesn't that leave belief in (P) precisely as justified as it was before we went through al this mess?

Angus said...

Whoops. I am wrong here: Likewise, then, we have reasons of equal weight to believe that deductions from (A) / (B) are true / false. So considerations of epistemic peerage give us reasons of equal weight to believe (P) and ~(P).

But not, I think, if we change "true / false" to "justified / unjustified," which seems to get us to the same place?

Neil Sinhababu said...

Thanks, Angus. The point shouldn't be "Both of my apparent epistemic peers agree that I have an inferior among them. I should trust the opinions of apparent epistemic peers, so I should believe that."

I think the idea is more like this: As you say, at least one of my apparent epistemic peers has a pair of false beliefs, either (A) or (B). The other of my apparent epistemic peers offers me a plausible explanation of why the first apparent peer has gone wrong. This explanation entails that the other apparent epistemic peer is an inferior. (That's why the first peer got it wrong.) Except that I don't know which peer is which, I have no good reason to reject it.

Under normal circumstances, if an epistemic peer offers me an explanation of this sort, I should accept it. The only trouble is that I have no idea who is who.

Justin said...

I disappeared for awhile, but here are some disjointed thoughts:

Reading your comment, I realized there's two factors at issue, only one of which was in view in what I said. First, should you take your peer's opinion as a reason to change your own opinion. In the case at hand, this means saying "Europeans are my peers, and they think the Republicans aren't my peers, therefore I should decrease my credence that Republicans are my peers." That's an example of the standard question in the literature on disagreement. So if you hold the equal weight view, the answer is that you should split the difference, if you hold the total weight view, you might reduce your credence that you're peers with Republicans, but you might not. It'll depend on what the underlying evidence supporting peerhood looks like.

Second, (and what I wasn't thinking about), there's the question about whether, once you find out about the Republicans and Europeans not viewing each other as peers, you should conclude that one or more of them is actually not your peer, on the grounds that they seem to have answered this question wrong.

That's obviously a different question, since you still have to answer it after you've already chosen between the EW and TE views and adjusted your credence based on them

That said, I think the effect should be pretty small. It's generally hard to say what evidence is relevant to peerhood, and judgments about peerhood are hard to make. Error about such a topic may not be that significant, at least so long as you're willing to endorse many substantive judgments made by each side.

Second, even if it is significant, it should be symmetrical. I guess I'm not sure how you're thinking of the tipping point phenomenon that it's supposed to rule that out.

Neil Sinhababu said...

On the first thing -- yeah, I suppose it could be the the broad degree of contradiction between their other judgments that's doing the work here. Maybe the epistemic peer judgment contradiction isn't that big a deal since there's so many other beliefs of theirs that are just as important.

On the second thing -- nothing so far gets us to asymmetry. That all comes from other stuff.