In order to zombify the tough question we need to restructure it so that it makes reference to zombies. The zombies we have in mind are philosophers' zombies, not those creatures who shuffle about half-dead in the movies.11 Philosophers' zombies, to use Stevan Harnad's [Harnad, 1995] felicitous phrase, are bodies with ``nobody home" inside. Such zombies are characters in a variation arising from a gedanken-experiment lifted directly out of the toolbox most philosophers of mind, today, carry with them on the job: Your brain starts to deteriorate and the doctors replace it, piecemeal, with silicon chip workalikes which flawlessly preserve the ``information flow" within it, until there is only silicon inside your refurbished cranium.12 John Searle [Searle, 1992] claims that at least three distinct variations arise from this thought-experiment:
Scenario V2 seems to us to be clearly logically possible (a proposition written, using the possibility operator from modal logic, as V2); that is, V2 seems to us to be a scenario free from contradiction, perfectly coherent and conceivable. After all, Searle could, at the drop of a hat, provide a luxurious novel-length account of the scenario in question (or he could hire someone with the talents of a Kafka to do the job for him).13
Not everyone sees things the way we do. Daniel Dennett has registered perhaps the loudest and most articulate dissent. In fact, Dennett has produced an argument (based, by the way, on the Rosenthalian definition of M-consciousness discussed above) for V2 in his recent Consciousness Explained (= CE; [Dennett, 1991], 304-313).14 One of us (Bringsjord) has formalized Dennett's argument [Bringsjord & Zenzen, 1997], and found it wanting, but there isn't space here to recapitulate the argument. We don't ask that you regard this attempted refutation to be sound, sight unseen. We do ask that, for the sake of argument, you join the many prominent thinkers who affirm the likes of V2 (e.g., Dretske [Dretske, 1996], Block [Block, 1995], Chalmers [Chalmers, 1996], Flanagan, [Flanagan & Polger, 1995], and Harnad [Harnad, 1995]). Moreover, we ask that you grant that V2 is physically possible, that is, that V2, though no doubt monstrously improbable, could come to pass without violating any laws of nature in our world. This seems to us to be a reasonable request to make of you. After all, why couldn't a neuroscience-schooled Kafka write us a detailed, compelling account of V2, replete with wonderfully fine-grained revelations about brain surgery and ``neurochips"? Then we have only to change the modal operator to its physics correlate -- to .15 Each and every inch of the thought-experiment in question is to be devised to preserve consistency with neuroscience and neurosurgery specifically, and biology and physics generally. Our approach here is no different than the approach taken to establish that more mundane states of affairs are physically possible. For example, consider a story designed to establish that brain transplantation is physically possible (and not merely that it's logically possible that it's physically possible). Such a story might fix a protagonist whose spinal cord is deteriorating, and would proceed to include a step-by-step description of the surgery involved, each step described to avoid any inconsistency with neuroscience, neurosurgery, etc. It should be easy enough to convince someone, via such a story, that brain transplantation is physically possible.16
This last assertion will no doubt be challenged; we hear some readers saying: ``Surely the two of you must be joking. To concede that such neural implantation is physically possible is (debatable) one thing, but to concede that and that the result would be a V2-style zombie is absurd. In any case, if it is `perfectly reasonable' to allow V2 as a physical possibility, then anything extra about logical possibility is superfluous, since the former entails the latter."
The part of this objection which consists in observing that
Let us make it clear that we can easily do more than express our confidence in Kafka: We can provide an argument for V2 given that Kafka is suitably armed. There are two main components to this argument. The first is a slight modification of a point made recently by David Chalmers [Chalmers, 1996], namely, when some state of affairs seems, by all accounts, to be perfectly coherent, the burden of proof is on those who would resist the claim that is logically possible.17 Specifically, those who would resist need to expose some contradiction or incoherence in . We think most philosophers are inclined to agree with Chalmers here. But then the same principle would presumably hold with respect to physical possibility: that is, if by all accounts seems physically possible, then the burden of proof is on those who would resist affirming to indicate where physical laws are contravened.
The second component in our argument comes courtesy of the fact that V2 can be modfied to yield V2 , where the superscript `NN' indicates that the new situation appeals to artificial neural networks, which are said to correspond to actual flesh-and-blood brains.18 So what we have in mind for V2 is this: Kafka really knows his stuff: he knows not only about natural neural nets, but also about artificial ones, and he tells us the sad story of Smith -- who has his neurons and dendrites gradually replaced with artificial correlates in flawless, painstaking fashion, so that information flow in the biological substrate is perfectly preserved in the artificial substrate and yet, as in V2, Smith's P-consciousness withers away to zero while observable behavior runs smoothly on. Now it certainly seems that V2 ; and hence by the principle we isolated above with Chalmers' help, the onus is on those who would resist V2 . This would seem to be a very heavy burden. What physical laws are violated in the new story of Smith?
We are now in position to ``zombify" Q1 :