I normally like John Hawthorne’s papers, but I think I disagree with just about every word of his paper in the latest No{u^}s, “Why Humeans are Out Of Their Minds”:http://www.sims.berkeley.edu/~kahern/hawthorne/humeans.html. (That link is to a prepublication version that doesn’t include the footnotes, but I think it should be intelligible.) John wants to defend the following argument.

An intrinsic duplicate of any region wholly containing me will contain a being with my conscious life.
There are causal requirements on my conscious life.
Therefore, Humeanism is false.

My feeling is that both premises are probably false and the argument is invalid. Very briefly, here’s why.

The first premise is false if there are any holistic phenomenal properties – phenomenal properties that one has in virtue of one’s overall conscious experience being a certain way. It seems to me very likely that there are such properties. For instance, it seems there are phenomenal properties one has (or would have) in virtue of having a wholly symmetric visual field. Assume I have that property. A region containing a duplicate of me plus some extra consciousness-constituting stuff that, as it turns out, constitutes extra visual phenomena, might not contain any conscious being with this nice symmetry of the visual field property. So conscious properties are not intrinsic to regions.

John sorta acknowledges this point when he tries to stipulate that a conscious being with _more_ phenomenal experiences than I, say by living longer than me, should still count as having all my phenomenal properties. But this won’t work if there are holistic phenomenal properties, and I don’t think they can be stipulated away.

The second premise is trickier, but I’m inclined to think the disjunction of the following claims is true. (I won’t argue for this here, though if I ever write something up on this paper I might.)

* Phenomenal perceptual states are partially constituted by what they are perceptions of, as Dretske, Lycan and others say.
* It would make no difference to my phenomenal life if there were no causal connections between successive states, and God just made me have one experience after another.

John needs to deny both of these. He thinks perceiving a desk and having a desky illusion make for the same phenomenal state. But he thinks that attending to a headache is phenomenally different from having an experience as of attending to a headache while actually having a headache. (And this is all important to the defence of premise 2.) Veridical hallucinations of external objects are phenomenally like perceptions of those objects, but veridical hallucinations of qualitative states are phenomenally different from perceptions of those states. I can see why you might believe one or other of these claims, but can’t imagine why one would believe both.

(Note for the record I think there _might_ be a way in which premise 2 is true here. But only if the Dretske/Lycan line about phenomenal externalism is true, and if so premise 1 is as false as false can be.)

Onto validity. John runs together two things here – an object a having the extrinsic property F, and an object’s being extrinsically F. It is easy to find cases where these come apart. Assume, as many people do, that some valuable things are intrinsically valuable, and some are instrumentally valuable. Then _being valuable_ is extrinsic. But an intrinsically valuable thing does not have the property _being valuable_ extrinsically, it has it intrinsically. Similarly the Humean could say (and I think should say) that while the property _being a cause-effect pair_ is extrinsic, it could be possessed intrinsically by some pairs. Or, at the very least, it could be intrinsic to a region that it contain a cause-effect pair. (And that’s all the Humean needs here.) Perhaps some actual Humeans have thought that all cause-effect pairs have the property _being a cause-effect pair_ extrinsically. But that’s not essential to the Humean picture I think, although it is essential to John’s argument.

One big picture thing to note at the end. Although John takes himself to be arguing against Humeans, the only feature of Humeanism that he leans on is that causation is extrinsic. So for any -ism that entails that causation is extrinsic, the argument should go through just as well. Consider then causal extrinsicalism, the view that causation is extrinsic. If this argument shows Humeans are out of the their minds then, by parity of reasoning, it shows causal extrinsicalists are out of _their_ minds. But causal extrinsicalism is more or less conclusively proven to be true by the double prevention cases. So there has to be _something_ wrong with the argument, because there are no sound arguments against causal extrinsicalism.