Alex Carruth (Durham) discusses that question in a post at the OUP Blog, building on a recent paper in The Philosophical Quarterly, which — for now — is freely available. A few key paragraphs:
… we can form a picture a human being in our heads, and say to ourselves ‘and it doesn’t have any phenomenal experience’. But this is far from conceiving of an exact physical duplicate of a human which lacks phenomenal experience.
Here’s an analogy, think of a mechanical clock, indeed, an exact duplicate of a mechanical clock you’re acquainted with. Can you conceive of the duplicate’s hands running anti-clockwise, rather than clockwise, or not running at all? You certainly could form a mental picture of the clock and say ‘and the hands run backwards’. But under close inspection, it’s not clear one could maintain this picture under scrutiny without making some change to the clock — say by rearranging the gears, or changing the direction of the motion imparted by the motor. …
What about the zombie case? Well, it’s clear we can conceive of the notion of ‘an exact duplicate of a human being’, and, separately, of the notion of ‘lacking phenomenal experience’. Conjoining the two doesn’t lead to an obvious contradiction … But it is far from obvious, given our relative lack of knowledge of the relationship between the mind and the body, whether or not a contradiction lies waiting to be unearthed in the notion of zombie.
The full post is available here.
The post makes a nice point but it perpetuates the myth that the zombie conceivability argument is valid. Whether the conclusion follows from the premises depends on assumptions about accessibility between possible worlds that beg the question at issue. See G. Piccinini, “Access Denied to Zombies,” Topoi, 2015.
A fun article. I take it that the zombie infestation was taken care of pretty well by a cadre of people like Gualtiero and Frankish. Since I positively think about conscious experiences as brain states, the zombie arguments hold little sway with me.
However, there is something lurking through Chalmers’ (and most of the dualists’ work since Huxely), that is at least *inductive* evidence against materialism. The problem is when folks take this inductive evidence and try to cook it into a deductive argument. Then things won’t work and people find loopholes like phenomenal concepts.
The fact that many materialists feel the need to invoke phenomenal concepts, which are unprecedented in the natural sciences, actually points to the problem. People pushing for phenomenal concepts accept that no matter how much natural scientific study of system X (e.g., a cat) you do, you will not come to conceive of X in phenomenal subjective terms. This to me seems the thread in Chalmers (and Huxely, and pretty much everyone with sympathies to dualism) that is evidence against materialism (in the Bayesian sense). The materialist who takes the phenomenal concept strategy should admit this. It’s basically saying “Let’s make a special concept that you can only have this one special way, that happens to save materialism against the classical objection to materialism that has been known since forever. Just a coincidence. Wink.”
I think materialists should just accept there is this weakness in their position right now. Even though the zombie argument falls flat, there is the larger inductive picture to consider, and that seems as solid as ever. Attacking the zombie argument seems almost like a straw man at this point, or at least kicking an undead man when he’s down.