Harmless Zombies

A zombie world is a world physically just like our world but in which there is no consciousness. Must a type-A materialist deny the conceivability of zombie worlds? No, not quite.

Compare the rather uncontroversial hypothesis that "the HI virus" denotes the (type of) virus responsible for most AIDS infections. Is it conceivable that a world could be biologically just like ours but not contain the HI virus? Yes, for it might turn out that scientists have been wrong all the time and no virus is involved in most AIDS infections. If it turned out this way, our own world would be a world biologically just like ours but not containing the HI virus.

Similarly, a type-A materialist can concede the conceivability of a zombie world. What a type-A materialist must not concede is the conceivability of "there is a zombie world and our world is not a zombie world".

Oddly, this gives the type-A materialist a reason to reject the step from conceivability to possibility in the zombie argument:

  1. It is conceivable that there is a zombie world.
  2. It is possible that there is a zombie world.
  3. Our world is not a zombie world.
  4. Therefore, materialism is false.

Again, compare the HIV case: It is conceivable that a world is biologically just like ours but does not contain the HI virus. (Because it is conceivable that there actually is no HI virus.) But from this it doesn't follow that there really is such a world. After all, if our world does contain the HI virus, then surely the HI virus is biological, and hence contained in any biological duplicate of our world. So only if our world does not contain the HI virus (which is conceivable) is a world biologically like ours but not containing the HI virus possible (and indeed actual).

So where is the fault in the two-dimensionalist argument from conceivability to possibility? It's in the assumption that the A-intension of "zombie world" coincides with its C-intension. Even if A- and C-intension coincide for all phenomenal and physical terms, it doesn't follow that they coincide for "zombie world". They don't if "zombie world" is defined, as usual, by means of the indexical "a world physically like ours but not containing consciousness": Let w be some world physically unlike ours and not containing consciousness. w is in the A-intension of "zombie-world": if our world turns out to be w, it turns out to be a zombie world. But it is not in the C-intension of "zombie-world", since necessarily, every zombie world is physically like ours, whereas by assumption w is not.


# on 04 August 2004, 20:25

Hi Wo,

Yes, this is why one should really run the conceivability argument with the conceivability of P&~Q (for the specific P that specifies the actual physical truths) as a premise, rather than the conceivability of "there are zombies". That's the way I always do it when formalizing the argument.

# on 05 August 2004, 17:54

Hi! Right, that seems to block the loophole.

(The situation is a little confusing. I think a type-A materialist should concede that the consciousness role could turn out to be unrealized. If it does, P&~Q is true, and conceivable and possible. Yet if the consciousness role is realized, P&~Q is not even conceivable. So could a type-A materialist say that while it's conceivable that P&~Q is conceivable, in fact P&~Q probably isn't conceivable? That seems at least somewhat misleading, since there isn't really any single sentence or proposition 'P&~Q' whose conceivability is merely conceivable: if the consciousness role is unrealized and P&~Q conceivable, then P is a different sentence than if the consciousness role is realized and P&~Q inconceivable.)

Don't want to bug you, Dave, but I'm really curious how you'd answer the scrutability question I asked a while ago (while you were in Italy): is "there are planets" *verified* at zombie world?

# on 05 August 2004, 21:29

Right, this is why one needs the truth of Q as an (explicit or tacit) premise, too. An eliminativist type-A materialist will say that P&~Q is actually true, so they can happily concede its conceivability and possibility.

On verification in zombie worlds:

From the claim that PQTI implies (apriori entails) all macrosopic truths, it doesn't follow that PTI implies no macroscopic truths. It could be e.g. that it implies spatial truths but not color truths, for example. But personally I'm inclined to think that Q plays a significant role in implying most macroscopic truths, even spatial truths, because of a response-dependent element in many or most macroscopic concepts. Taking the Rubber Ball World case, if told that there are entities with a certain microstructural composition at the actual world, but told nothing at all about how entities with this sort of composition (or any composition) typically look or feel, then I think we're not in a position to make any judgment about whether or not they're made of rubber. This means that the 1-intension of "There are rubber balls" isn't true at Rubber Ball world, but it doesn't mean that it's false there -- better to say that it's neither true nor false there. Something similar may well apply to the 1-intension of "there are planets" at zombie world, at least if we grant that appearances play an important reference-fixing role for "planets" (which I think is plausible though not 100% obvious).

# on 05 August 2004, 21:52

Somehow that got posted twice -- sorry. One other point re your original post on scrutability: of course the terms with the partially response-dependent element that I mentioned will have distinct 1-intension and 2-intension. On my view this goes even for spatiotemporal terms. But if one denies this, then spatial terms won't be response-dependent and their 1-intensions will apply normally at zombie worlds. On my view the most basic terms with identical 1-intension and 2-intension are mental terms, causal/nomic terms, and logical/mathematical/categorical terms, plus terms grounded in some fashion in these: so e.g. "friend", "philosopher", "action" plausibly have identical 1- and 2-intensions. Of course most of those won't apply at a zombie world if that world lacks mentality (setting aside issues about whether consciousness is required for mentality). So the main terms whose 1-intensions will apply to the zombie world just as they do to our world are causal/nomic/logical/mathematical/categorical terms (plus spatiotemporal terms if you think those go in too), and those grounded in these.

# on 06 August 2004, 16:49


# trackback from on 05 December 2006, 08:12

Does the conceivability of zombies threaten type-A materialism, the claim that all mental truths are a priori entailed by the physical truths? Can we imagine beings exactly like us in all physical respects, but lacking consciousness? If so,...

Add a comment

Please leave these fields blank (spam trap):

No HTML please.
You can edit this comment until 30 minutes after posting.