A Physicalist Challenge to Chalmers's Two-Dimensional Semantics — Epoche B2
Where the Zombie Argument Is Actually Vulnerable: A Review of Chalmers's The Conscious Mind The layout of the building David Chalmers's The Conscious Mind [1] : In Search of a Fundamental Theory (1996) is built in storeys, and a reader who wants to disagree with it needs to know which storey carries the weight. At ground level there is the phenomenon: conscious experience, and the distinction between explaining what a mind does and explaining why there is something it is like to be one. Above that sits an apparatus of supervenience and reductive explanation. Above that, the conceivability argument with its zombies. Above that, the two-dimensional semantic machinery that turns conceivability into a metaphysical conclusion. And on the top floor, the positive proposal: a naturalistic dualism in which experience is fundamental, together with a frank accounting of what it costs. This review argues that only the fourth storey is load-bearing against physicalism, and that the physicalist objections most often made are aimed at the third and pass beneath the building. The question is where a physicalist should press. My answer: at the claim that ordinary macroscopic truths are a priori entailed by the physical truths, which is the assumption Chalmers's bridge needs and which Ned Block and Robert Stalnaker denied in 1999 [2] . The claim is refutable, and the section on the Chalmers–Jackson reply says how [3] . Ground floor: what a zombie is missing Chalmers separates the "easy" problems of consciousness — explaining discrimination [4] , integration, reportability, the control of behaviour — from the hard problem: why any of that should be accompanied by experience. The distinction only works if there is something to be missing, and here the book's most useful neighbour is Ned Block's 'On a Confusion about a Function of Consciousness' (1995 [5] ). Block distinguishes access consciousness — a state's being poised for use in reasoning, report and the control of action — from phenomenal consciousness , the state's having a felt quality. Without that distinction a common objection looks decisive: if the creature discriminates, reacts, reports and remembers exactly as I do, what on earth is left out? With it, the answer is exact. The zombie is access-conscious in every respect and phenomenally conscious in none. Whether that combination is possible is the question; that it is a well-formed description is not seriously in doubt once the two notions are separated. Second floor: conceivability A philosophical zombie is a being physically identical to a conscious person, particle for particle, in a world with the same laws, which has no experience at all. Chalmers argues that such a being is conceivable: no amount of reflection on the physical description reveals a contradiction in the supposition, in the way that reflection reveals the contradiction in a round square or in a female bachelor. It is worth stressing that conceivability alone gets Chalmers nothing. Plenty of things are conceivable and impossible: that water is not H2O, that Hesperus is not Phosphorus. Since Kripke [6] , everyone accepts that there are necessities knowable only a posteriori , and every one of them is a case where something conceivable turns out impossible. So a physicalist can grant zombie conceivability outright and deny zombie possibility, on the ground that "consciousness is physical" is one more a posteriori necessity. That position is what Chalmers calls Type-B physicalism, and answering it is the job of the next storey. The load-bearing storey: two-dimensional semantics Chalmers's response is to ask why the standard cases of a posteriori necessity work, and to argue that the explanation is unavailable in the case of consciousness. Two-dimensional semantics assigns each concept two intensions. The primary intension of "water" picks out, in any world considered as actual , whatever plays the watery role there — the clear drinkable stuff in the lakes. The secondary intension picks out, in any world considered as counterfactual , the substance that actually plays that role here, namely H2O. That is why "water is H2O" is necessary and yet knowable only by investigation: the necessity comes from the secondary intension, the ignorance from our not knowing which substance satisfies the primary one. Chalmers's key claim is that phenomenal concepts have no such gap. There is no appearance/reality distinction for a feeling: the way experience seems is not a mere mode of presentation of something whose real nature might be otherwise, because with experience the seeming is the thing. So for phenomenal concepts the primary and secondary intensions coincide. And where they coincide, the mechanism that generates a posteriori necessity is absent, and conceivability does entail possibility. This is the argument's structural heart, and noticing that is the first thing a would-be objector must do. Two consequences follow at once. A physicalist who tries to answer the zombie argument by appealing to two-dimensional semantics — treating the physical–phenomenal identity as another Kripkean a posteriori identity — is reaching for the very tool Chalmers built to block that move, and will find it turned against him. And a physicalist who attacks conceivability is attacking a step Chalmers can afford to weaken, since what he needs is not that zombies be vividly imaginable but that no contradiction be derivable from their description. Why the functionalist reply misses The reply most often made runs: if the zombie is physically and functionally identical to me, then it instantiates every functional role I do; phenomenal properties just are functional roles, or are constituted by them; so the zombie must be phenomenally identical too, and the description is incoherent. This is Type-A physicalism, the position of Daniel Dennett's Consciousness Explained (1991) and of the analytic functionalists before him, and it is a serious philosophical position.