Zombies and the Phenomenal Concept Strategy

Zombies and the Phenomenal Concept Strategy

Consciousness Online

Dave Beisecker, University of Nevada, Las Vegas

Abstract: In this paper, I challenge Chalmers’ recent considerations against the phenomenal concept strategy. Most who adopt this strategy are “Type-B” materialists, who maintain that Chalmers’ zombies, though conceivable, are ultimately impossible. I suggest instead that a better materialist response (fittingly labeled “Type-Z” for zombie) is to grant, not just the possibility of zombies, but also their actuality. We turn out to be the very creatures Chalmers has taken such great pains to conceive. So consciousness (at least for us) is a wholly material affair. What is conceivable but non-actual are not zombies, but rather “angelic” beings possessing an acquaintance with non-material phenomenal states. I find it odd (to say the least) that Chalmers insists upon the conceivability of the zombie hypothesis, and then turns around and just as stridently insists on its non-actuality. If we are being appropriately open-minded about the nature of consciousness, it should take more than arm-chair reflection to rule out the admittedly-conceivable hypothesis that ours is a zombie world.

Let’s begin by traveling over some familiar ground.

The zombie argument carries us from the alleged conceivability of zombies to the falsity of materialism. In its canonical form, it has the three premises listed on this slide. Most materialists who’ve responded to the argument have seen fit to challenge either the first premise or the second. Chalmers calls those who reject the first premise (P1) “type A” materialists. Zombies, they contend, are simply inconceivable. Those who balk at the second premise (P2) are commonly called “type B” materialists. Though conceivable in some sense, these folk maintain that zombies are nevertheless metaphysically impossible. The third premise(P3), that which takes one from the possibility of zombies to the falsity of materialism has largely gone unchallenged, and Chalmers hasn’t dignified its rejection with a letter of its own. Let me cheekily propose the letter “Z” for that task. The type Z materialist, then, is one who grants the conceivability and the possibility of zombies, yet still maintains that we live in a thoroughly material world (a Stalnakerian z-world, if I may use such terminology). In effect, the type Z materialist will contend that the zombie argument is insignificant, for when all is said and done, there is no reason why we couldn’t turn out to be the very kinds of creatures the possibility of which Chalmers has taken such great pains to establish (and that his more conventional materialist foes have taken to wipe off the face of the planet). [1]

I will have more to say about how to develop the Type Z response later. However, my initial aspiration is more modest. To begin with, I’d like to suggest that Chalmers’ recent argument against the Phenomenal Concepts Strategy should actually steer those who are inclined to adopt such a strategy to a type Z position.

Roughly, the phenomenal concepts strategy is that of giving a materialistic explanation, not of consciousness itself, but rather of what Chalmers calls “the next best thing,” the explanatory gap. Those who adopt such a strategy are likely to endorse those intuitions about zombies and color-deprived Mary that are so suggestive of an explanatory gap between the mental or phenomenal and the more mundanely material. However, unlike those who take these considerations to suggest that there is an additional dimension of reality above and beyond the mundanely material – an immaterial or “super”-material[2] realm that eludes ordinary physical description, phenomenal concept strategists hold that the epistemic gaps revealed by the zombie and Mary intuitions need not ramify into an ontological one. Rather, the modal slippage or flexibility we observe between phenomenal and physical descriptions is due to peculiarities in the concepts with which we talk or think about phenomenality.

The phenomenal concepts strategy is typically associated with materialisms of the type B sort. Since type-B materialists usually deny that the mere conceivability of zombies entails their genuine possibility,[3] their supermaterialist opponents have typically tried to expose an inherent instability in the type B position by questioning whether one can genuinely conceive of something which turns out not to be possible after all. In reply, type B materialists naturally draw upon those portions of Kripke’s work suggesting that not everything that seems conceivable is a genuine possibility. The identity between water and H2O is necessary, despite the fact that one can conceive of worlds in which those two concepts come apart. In turn, supermaterialists insist that the analogy between consciousness and water ultimately breaks down, because it seems that we can conceive of the possibility that consciousness is not a material state in a more direct fashion than we can conceive of the possibility that water is not H2O. When we conceive of a world in which water is not H2O, what we are really imagining is a world in which either the concept of water or that of H2O applies to something other than water. In the case of phenomenal concepts, however, no such room is open; we cannot help but conceive of a pain as such. That is, phenomenal concepts cannot be pried apart from their referents in the same fashion as material ones.[4]

What is so peculiar about the phenomenal realm is the special epistemic position or privilege we have (or at least seem to have) with respect to our own phenomenal states, a position that Chalmers characterizes as acquaintance. Unlike thoughts about the rest of reality, your sincere thoughts or avowals about how things appear are especially resistant to third-person challenge. In the phenomenal realm, no epistemic gap opens up between appearance and reality. While you might be mistaken about the actual color of the fruit in front of you, you cannot question what color it appears to you to have. In short, the qualitative character of one’s own experiences seems somehow self-intimating.[5]

Chalmers recognizes that merely pointing to this difference between phenomenal and material concepts does not let the materialist off the hook. The phenomenal concept strategist further needs to tell a materially acceptable story about how we could ever so much as come to grasp phenomenal concepts so that we could be capable of entertaining genuine, self-intimating phenomenal thought.[6] This, I take it, is the launching point for the master argument Chalmers has recently raised against this strategy.[7] While I think Chalmers’ considerations may indeed expose an instability in type B positions, it ultimately shows that the phenomenal concept strategist should instead “go zombie (Z)” on Chalmers.

In typical Chalmers fashion, he begins with a conceivability exercise intended to ascertain whether a candidate phenomenal concept strategist thinks that our grasp of phenomenal concepts is itself materially explicable: is it conceivable for a being to be physically just like us, and nevertheless lack the psychological characteristics that allow us to grasp phenomenal concepts? Depending upon the answer provided, Chalmers constructs the following dilemma:[8]

(1) If it is conceivable for a being to be physically just like us and yet lack the psychology to grasp phenomenal concepts, then this ability itself is not materially explicable. The phenomenal concept strategist would then be forced into conceding that the material facts cannot account for something that is psychologically significant about us.

(2) But on the other hand, if the phenomenal concept strategist avows that it is inconceivable for a being to be physically just like us and to lack the psychological ability to entertain genuine phenomenal thought, then it would seem that zombies would also be capable of harboring phenomenal thoughts. But that would put us and zombies on the same epistemic plane, and for Chalmers, that’s just plain crazy talk.

The first horn of this dilemma does indeed seem unpalatable for a thoroughgoing materialist. However, the second horn is precisely that which would be endorsed by a materialist of the type Z stripe. For such a materialist insists that we ourselves are in the same epistemic position as Chalmers envisions zombies to be in. We have learned how to embrace explanatory gap intuitions, yet without the “benefits” of, or any acquaintance with, full-blooded super-material conscious experience. Consciousness (at least for beings like us) is a wholly material affair.[9] Instead, what is conceivable, but most likely non-actual, are not zombies, but rather “angelic” beings, who possess an acquaintance with super-material phenomenal properties.[10]

I typically receive incredulous, boggle-eyed stares when I propose in all seriousness that we just might be the subjects of Chalmers’ thought experiments. [One of the advantages of this very forum is that I can avoid such cock-eyed glances!] Doncha know that the very idea of a zombie rules out this possibility? We cannot be zombies, because by definition zombies are just like us, except that they lack our conscious experience. As Chalmers puts it, a being that is epistemically equivalent to us just doesn’t match our intuitive conception of a zombie. This is how Chalmers expresses the objection: [I’ll read just the highlighted portions.]

Here, the natural response is that this scenario is simply not what we are conceiving when we conceive of a zombie. Perhaps it is possible to conceive of a being with another sort of state – call it “schmonsciousness” – to which it stands in the same sort of epistemic relation we stand in to consciousness. Schmonsciousnes would not be consciousness, but it would be epistemically just as good. It is by no means obvious that a state such as schmonsciousness is conceivable, but it is also not obviously inconceivable. However, when we ordinarily conceive of zombies, we are not conceiving of beings with something analogous to consciousness that is epistemically just as good. Rather we are conceiving of beings with nothing epistemically analogous to consciousness at all.

Put differently, when we conceive of zombies, we are not conceiving of beings whose inner life is as rich as ours, but different in character. We are conceiving of beings whose inner life is dramatically poorer than our own. (PCEG, p. 186, my italics; see also p. 177)

Simply put, Zombies must be creatures that have a dramatically less-accurate self-conception than genuinely conscious beings do. While I believe (indeed, know) that I am conscious and have states with remarkable qualitative character available to introspection, the corresponding beliefs of a zombie would seem to be false, or at least not-justified.[11]

Chalmers is treading a fine line. If zombies really are as introspectively impoverished as Chalmers suggests, then that threatens to make the explanation of their beliefs that correspond to our beliefs about consciousness utterly mysterious. It also makes their epistemic practices appear remarkably irresponsible. Why wouldn’t their lack of justification for their apparently introspective beliefs show up for them, thereby rendering them disinclined ever to engage in introspective reports? But if they do turn out to be remarkably less inclined to introspection than fully-conscious beings, then that only plays into the hands of more conventional type-A materialists, who have argued all along that on serious consideration, a zombie’s remarkable incapacity for self-reflection will ultimately ramify into serious behavioral or linguistic deficits.

To be sure, there is a long-recognized creepiness about zombies, which Chalmers has dubbed “the paradox of phenomenal judgement.”[12] By supermaterialistic lights, nothing anyone can say will convince them that they are zombies, or that they lack the consciousness of “regular” folk. And there is nothing anyone can say in defense of the claim that we are not zombies that zombies themselves wouldn’t find convincing. They harbor the same zombic hunches that we do, and they find the hard problem of consciousness every bit as hard as we do. The bare compellingness of zombie arguments does not distinguish us from them.[13] So either zombies are horribly mistaken about the nature of their own conscious experience, or their phenomenal concepts turn out – perhaps unbeknownst to them – to mean something very different from what supermaterialists take our phenomenal concepts to mean.

Even if we avoid charging zombies with gross epistemic impropriety, instead construing zombie speech and thought somehow in terms of their own material make-ups, Chalmers insists that their epistemic situation is bound to differ from ours. Specifically, he claims that the intuitive resolutions of debates about consciousness are going to go differently in a zombie world than they do in ours.[14] By way of illustration, Chalmers invokes a hypothetical debate between zombie eliminativists and realists, but we can just as well consider a genuine zombie-Dave (pictured here) and the belief he holds to the effect that he is not a zombie. While the “real” Dave Chalmers (in whatever possible angelic world that might be) knows this, poor, unfortunate Zombie-Dave cannot, simply because it’s false. Hence, even if we are kind enough to allow some of a zombie’s phenomenal belief analogues to be true, zombies are bound to be in a different epistemic position from “real” folk. Right?

It should be clear that Chalmers’ contention here gets virtually all of its mileage from definitional fiat. And the proper reply on behalf of the type Z phenomenal concepts strategist should be equally apparent. Fine; so be it. If zombies are simply defined as non-human (or not having our consciousness), then I’m perfectly content to give up the term. But I caution you then to notice how the definition doing all the work threatens to beg the question in favor of supermaterialism. As the ontological argument for the existence of God showed, seemingly innocuous definitions can be far from such; in order to prevent their issuing undesirable inference tickets, we need to be reassured that they are metaphysically conservative. And this one evidently fails that test. Insofar as Chalmers defines zombies as different from us and as lacking any acquaintance with super-material qualia, he presupposes that our conscious experience must be something above and beyond the mundanely material. So the real problem isn’t one of the space of conceivability or possibility (as the type A or B materialists suggest); it’s rather that the terms animating the zombie argument illicitly center the actual world on an angelic world. If we are appropriately open-minded about the nature of consciousness, it should take more than arm-chair reflection to rule out the admittedly-conceivable hypothesis that ours is a zombie world.[15] The issue of interest is not whether we differ from zombies, but whether our form of consciousness is of an angelic or a zombic sort (or equivalently, whether our concepts of consciousness ought to be appropriate to angelic or zombic worlds – whether or not they actually are).

In short, then, the position the type Z materialist tries to stake out as actual is just like the one that is envisioned and granted to be possible by zombieists: that an explanatory gap between the phenomenal and the material exists and nevertheless there could be creatures who talk about consciousness much like we do yet turn out to lack “genuine” consciousness as supermaterialists understand it. My claim, then, is that whatever it is like to be me is equivalent to whatever it is like to be one of those things Chalmers and his ilk envision as a “zombie.” Call them what you like; I’m one of them. By supermaterialists’ lights (though not mine), there is nothing it is like to be me. Indeed, shouldn’t it strike one as more than a tad awkward for a super-materialist to turn around and close off this admittedly possible scenario as somehow impossible after all? For that just plays into the hands of their type A or B opponents who’ve suspected all along that there is something deeply wrong with the very idea of a zombie. Instead, the claim that we are not zombies has to rest – not on mere definition – but rather on some contingent fact, the truth of which we have some independent justification for holding.[16]

Part of me is inclined to stop right here, but I feel compelled to add some remarks about the knowledge argument. The reason for this is that in order to bolster further his contention that Zombies must have significant epistemic shortcomings relative to us, Chalmers bids us to consider a zombie in the place of color-deprived Mary.[17] Once again, he’d have our intuitions informing us that upon release, while zombie-Mary might acquire some new knowledge concerning her internal states, she nevertheless wouldn’t acquire the full-fledged, “cognitively significant” first-person knowledge of her own introspective, experiential states that we have (and that Mary comes to have). Though zombie-Mary might acquire some new ways of representing her discriminatory states in thought and talk, she wouldn’t acquire what we think Mary acquires when we say she has finally learned “what it’s like” to see red.

Thus the phenomenal concepts strategy also obliges one to offer a wholly materialistic account of Mary’s post-release knowledge. Chalmers says that there are two, equally unpromising, strategies available: a deflationary one which demotes our knowledge of what it’s like to the knowledge post-release zombie-Mary can acquire, or an inflationary one which elevates post-release zombie-Mary’s knowledge of what it’s like to our own.[18] Much of what Chalmers subsequently has to say is directed against specific ways on offer to adopt the phenomenal concept strategy – particularly those self-styled type-B proposals that claim that pre-release Mary lacks the requisite experience to possess certain concepts, the phenomenal ones, which refer to their referents in some distinctive fashion that’s routed through experience.[19] I happen to agree with Chalmers that these accounts don’t capture all that is bound up in attributions of knowledge of what it’s like (although for different reasons). I’m just not convinced that Mary’s failure to know what it’s like to see red can be traced to a gap in her conceptual repertoire, for her lack of experience with red things doesn’t seem to constrain her ability to entertain thoughts about what it would be like. Even without any contact with a colored reality, she can still wonder what it would be like for her to see red, know that her earlier experiences are not what it’s like to see red, and surmise (probably correctly) that her similar physiological makeup suggests that what it would be like for her to see red is pretty much like what it would be like for the rest of us to see red. She could even have mistaken thoughts about what it’s like – say, by spying a green tomato rogue epistemologists have told her is red.[20]

But another option is available, which doesn’t explicate post-release Mary’s new knowledge with a sudden expansion of her conceptual repertoire. Instead, this strategy points out that she lacks the “experience” (here glossed as a history of visual contact with colored things) to entertain relevant experiential thoughts responsibly. That is, post-release Mary acquires (eventually) not the capacity to entertain new thoughts, but rather a justificatory status which allows her to claim correctly that she now knows what it’s like to see red. Notice how this should be an especially amenable option for a “type-z” materialist, since it would seem that zombie-Mary could also acquire a post-release history of contact with a red reality as well, which would allow her to gain a similarly elevated epistemic status with respect to her observation reports.

Here we’re well-served recalling some of the lessons Wilfrid Sellars taught us. Sellars, of course, was no friend of Chalmers’ favorite crutch – namely, knowledge by acquaintance – in any of its guises. Such knowledge, one of the last bastions of which was the knowledge empiricists claimed we have of sense-data, is a particularly pervasive form of the foundationalist’s “myth of the given.” Instead, all our knowledge, including that involving our concepts of inner experience, is a product of training and attunement to the world. Our ability to identify or classify a given sensation as a red sensation – or to know what it’s like to see red – depends upon an understanding that those are the sensations generally indicative of red things in an external world.

Consequently, here’s what I think “Uncle Wilfrid” might say about the Mary case. Our ability to know what it’s like to see red depends upon an awareness of our own reliability at being able to detect red things. Merely being a reliable detector of redness is not sufficient for such knowledge; the awareness one must have of one’s reliability can come about only through “experience,” understood once again as a history of contact with red things. So even though – as if by magic – pre-release Mary might be a perfectly reliable reporter of red things, she nevertheless lacks the experience to have the adequate awareness of this reliability. That is actually the reason we can perfectly well suspect that she lacks knowledge of what it’s like. But then there’s no reason to suspect that (pace Chalmers) Zombie-Mary couldn’t also come to learn what it’s like to see red.

Equally significant (if not more so), Sellars also offers an explanation of the introspective knowledge of our own experiential states that so confounded Descartes (and currently vexes Chalmers). As part of his celebrated “myth of Jones,”[21] Sellars shows us how creatures already capable of reasoning about an external world can extend their vocabulary to encompass concepts of inner states, including those of experience, and then goes on to show how they can come to be in a privileged epistemic position with respect to them, without invoking anything like acquaintance. In effect, he tells a story about how zombies could begin to make observations about, and so come to be aware of, those internal states responsible for their being disposed to make reports about how things appear to them.[22] The key is to understand a sense impression of some property x as the discriminatory state one is in when one is disposed (under normal conditions) to judge that x is present. It follows that the discriminatory state one is in when one reports that one has a sense impression of x is the same state that would (under normal conditions) incline one to judge that x is present. Hence, an impression of x will be equivalent to an impression of an impression of x, and no gap opens up between one’s actually having an impression of x and one’s being disposed to say so.

There’s more to be said about Sellarsian sense impressions in connection with debates about phenomenal consciousness (concerning, e.g., transparency, “what it’s like” and the explanatory gap), but I’ll spare you the details.[23] My conclusion here is modest. I contend that we can turn to Sellarsian epistemology to defuse Chalmers’ objection to the idea that we must be on a higher epistemic plane than zombies (and thereby resist his master argument against the phenomenal concepts strategy). Zombies claim to have an inner life and to know what it’s like, and Sellars shows us not just how such talk is possible, but how it can be made intelligible, without invoking any acquaintance with a supermaterial realm. The Sellarsian story about inner experience is, I contend, the zombie’s story about inner experience – it works from the outside in. And with it, the phenomenal concepts strategy proves to be much harder to kill than Chalmers would have us believe. But that’s only fitting, for it raises the possibility that the actual world is really a zombie world (or equivalently, a Stalnakerian z-world). To the extent that we can flesh out the zombie’s story, I suspect that we’ll actually wind up fleshing out ours.[24]

[1] To be sure, this option has occasionally been entertained, only to be discarded almost as soon as proposed. Witness Dennett’s playful comment on p. 406 of Consciousness Explained (Little, Brown, 1991) that perhaps the most satisfying response to the zombie argument is to conclude that zombies are “not just possible. They’re actual. We’re all zombies.” However, Dennett declines to treat this option seriously and warns us in a footnote that “it would be an act of desperate intellectual dishonesty to quote this assertion out of context!” Instead, his reasoned response to the zombie menace is thoroughly conventional (and in direct opposition to the one explored here); zombies must be stamped out as “unimaginably preposterous.” See also Dennett (2008). In effect, Dennett is committed, not to the explanation of the explanatory gap (in accordance with the phenomenal concepts strategy), but rather to explaining it away.

[2] “Super-materialist” is my preferred term for those who follow Chalmers and accept that the zombie argument (or the knowledge argument) demonstrates that there must be more under heaven and earth than is countenanced by (mundane) materialism, for the moniker nicely captures an ambivalence in how the position is understood. To their materialist opponents, supermaterialism advocates us to accept the existence of spooky, supernatural, non-material features of the world, while supermaterialists themselves think they are simply urging us to acknowledge an underlying, intrinsic facet of our material existence, which has heretofore eluded systematic scientific investigation.

[3] And it is the possibility of zombies that is taken to spell so much trouble for materialism, because it suggests that something must be added to a zombie world in order to turn it into a world like ours with genuine phenomenal qualities. Eventually this paper will seek to defuse this inference by questioning that our world contains any such additional qualities.

[4] Kripke himself took this tack, and we have Chalmers to thank for unpacking and reviving Kripke’s dense remarks on the subject.

[5] It’s this very epistemic peculiarity of the phenomenal that ultimately accounts for Chalmers’ inability to drive a wedge between the primary and secondary intensions of our phenomenal concepts.

[6] Levine (2007) calls this the “materialist constraint.”

[7] See Chalmers (2007), which shall hereafter be abbreviated as PCEG.

[8] Chalmers (2007), p. 174.

[9] This is what distinguishes this position from “Churchlandish” Type-E (eliminatist) positions.

[10] Essentially, Chalmers’ complaint is that if a phenomenal concept strategist takes the materialism part seriously, then we turn out to be little different from zombies. But what I’m calling “the type-Z” materialist is going to ask, what’s wrong with that? Note that even in a zombie world, the physical facts need not settle the phenomenal ones (given the conceivability of non-zombies). In a sense, then, physicalism isn’t even true in a zombie world.

[11] Chalmers (2007), p. 179

[12] The question, of course, is how best to respond to this creepiness. For other treatments, see Moody (1994) and Balog (1999).

[13] Stranger still, it would even seem possible for non-zombies to learn how to master phenomenal concepts and come to talk effectively about their conscious experience from zombies. Whether or not one is a zombie turns out to be a matter of being acquainted with supermaterial qualia, and there are no reasons to be given to show that one is or is not so acquainted.

[14] Chalmers (2007), p. 178.

[15] Or an angelic world! Richard Brown (this conference), as well as Frankish (2008) before him, both run parallel “reverse zombie” arguments which beg the question the other way around. Both arguments deploy terminology that in effect similarly centers the actual world on a zombic, non-angelic world.

[16] This position is very close to Stalnaker (2002). However, there is a slight twist. Stalnaker agrees that the supermaterialist has stacked the deck by illicitly framing the zombie argument in terms that aren’t appropriately topic-neutral. If one gives the notion of “phenomenal consciousness” an appropriately topic-neutral reading, then Stalnaker concludes that the notion of a zombie is ultimately incoherent, a conclusion that makes Stalnaker look more like a type-A materialist. By contrast, my version provides topic-neutral readings of both the notions of a zombie and – granting the supermaterialist’s understanding of the term – that of phenomenal consciousness, with the consequence that I arrive at the conclusion that we must be zombies. Although these positions might just be semantic variants of one another, I submit mine to be rhetorically more robust, since it grants more to the supermaterialist.

[17] Chalmers (2007), p. 178

[18] Chalmers (2007), p. 185

[19] These proposals try to unpack the possession of phenomenal concepts somehow in terms of demonstrative or quotational modes of presentation that are unavailable to subjects until they find themselves in the appropriate kind of experiential state. The governing idea is that upon exposure to phenomenal red, we become capable of entertaining whole new forms of thought. Stoljar (2005) has called this “the experience thesis.”

[20] See Dennett’s (2007) example of the blue banana.

[21] See parts XIV-XVI of Sellars (1956/97)

[22] Or if you prefer, a story about how zombies can turn into non-zombies, or what Dennett (1998) calls “zimboes.”

[23] Those interested are encouraged to look up Beisecker (2005).

[24] I’d especially like to thank Tim Connolly for help with this project.