Jerry Fodor Caused This Post

"Weird" is right. In an exchange with Simon Blackburn et al., Jerry Fodor casts a routine feature of causal explanation as a knockdown objection to the theory of natural selection:

[Blackburn et al. contend that] '[w]hite polar bears . . . more camouflaged than their brown confrères[] were better at sneaking up on seals, were better fed and left more offspring.’ I don’t know whether this story is true (neither, I imagine, do they), but let’s suppose it is. [Nonetheless,] they’ve somehow left out the Darwin bit. To get it back in, you have to add that the white bears were selected ‘because of’ their improved camouflage, and that the white bears were ‘selected for’ their improved camouflage: i.e. that the improved camouflage ‘explains’ why the white bears survived and flourished. But...the theory of natural selection entails none of these. In fact, the theory of natural selection leaves it wide open what (if anything) the white bears were selected for. Here’s the argument. Consider any trait X that was locally coextensive with being white in the polar bear’s evolutionary ecology. Selection theory is indifferent between ‘the bears were selected for being white’ and ‘the bears were selected for being X.’ What’s ‘incoherent’ is to admit that the theory of natural selection can’t distinguish among locally coextensive properties while continuing to claim that natural selection explains why polar bears are white.

As if controlling for plausible third variables were a challenge peculiar to evolutionary theory. As if controlling for plausible third variables were a fatal conceptual problem for theories in general.

Simply amazing.

Still, fun to set to dialogue:

Ben: Jerry, did you know there were 43,000 deaths caused by car accidents in the U.S. last year?
Jerry: That's only if you ignore other states of affairs that covaried locally with these accidents.
Ben: Huh?
Jerry: Well, consider any state of affairs X that was locally coextensive with being killed while driving or riding in, or being struck by, a car.
Ben: [Pause.] Okay...
Jerry: Thus, car crash analysis is indifferent between 'the deaths were caused by the car accident' and 'the deaths were caused by X.'
Ben: [Silently waxing aporetic.]
Jerry: But then it is incoherent to admit that car crash analysis can’t distinguish among locally coextensive properties while continuing to claim that car crash analysis explains how "crash victims" were killed.
Ben: [Now staring incredulously.]
Jerry: So we really don't know what killed these people.
Ben: Look, what the hell are you on about, mate? 

Comments

[Jonathan]

Tee hee.

I agree and disagree. Nothing Fodor says makes natural selection any different to a regular everyday causal explanation, and all his "objections" would apply to any causal explanation.

*However*, I think Fodor is coming at this with the view that when scientists talk about "natural selection" explaining, they think they mean more than that they can give the causal history behind trait X. They think they are citing a law or force.

What I think Fodor is getting at (through some really terrible writing) is that natural selection can't be a force or law. Because we don't how it pushes. We can't analyse how the so-called law is instantiated through "selection for".

[Q the Enchanter]

Hi Jonathan, well, if that were what scientists meant, Fodor would have a point. Unfortunately, it seems unlikely that this is what scientists mean.

[Jonathan]

Frankly all those blogs suggest there's considerable debate about this. Many posts agree natural selection (NS) isn't a 'law', but few say in any detail what it *is*!

Many say it's a theory. In this context 'theory' just seems to mean 'socially accepted way of interpreting the facts'. I'd say NS is a name for a form of causal history, much like 'car accidents'. It doesn't subsume events under a law or common cause.

If this is so, an ID proponent would gleefully say: 'But my theory gives causal histories too!'

[Q the Enchanter]

Well, the posts show at least that the notion that biologists talking about natural selection categorically "think they are citing a law or force" doesn't quite hold water. Right?

More can be said about what makes a theory scientific, but I don't take Fodor to be raising this issue. Sure, an ID proponent might offer her own "causal history," but so what? If an Intelligent Driver ("ID") theorist offered her causal history of car accidents ("Jesus was not at the wheel!"), I'm sure it would be "socially accepted" by her peers, but I don't expect it would put any failure analysts out of work.

[Jonathan]

Presumably a lot of ID theorists *do* think all phenomena, including car accidents, are part of some great design! One doesn't like to pre-empt them. But while we may have no reason to convert to their way of seeing things, it's hard to say what makes it 'unscientific' without resorting to name-calling. I think one thing the Fodor controversy shows is how far we are from a justifiable demarcation criterion.

[Q the Enchanter]

"[I]t's hard to say what makes [intelligent design] 'unscientific.'"

? Parsimony. Falsifiability. Predictive utility. Productivity. Consistency with other sciences. Corrigibility. Provisionality. Progressivity.

Yep, tough call.