TWILIGHT OF THE ZOMBIES
Department of Philosophy
Hartford, CT 06106
860 297 2528
A philosophical zombie is a being indistinguishable from an ordinary human in every observable respect, but lacking subjective consciousness. Zombiehood implies *linguistic indiscriminability*, the zombie tendency to talk and even do philosophy of mind in language indiscriminable from ordinary discourse. Zombies thus speak *Zombish*, indistinguishable from English but radically distinct in reference for mental terms. The fate of zombies ultimately depends on whether Zombish can be consistently interpreted. If it can be interpreted consistently, then zombies remain possible, but no test could ever reveal whether anyone (oneself included) is speaking Zombish. Any materialist theory of consciousness is therefore already a theory in Zombish, and is equally confirmable in its human language edition (applicable to humans) and its zombie-language edition (applicable to zombies). On the other hand, if Zombish cannot be consistently interpreted, then the zombies described in Zombish are logically impossible. Either way, the search for a materialistic theory of consciousness should be untroubled by the (possible) zombies among us.
Key words: consciousness,
qualia, phenomenology, materialism, "zombie problem"
TWILIGHT OF THE ZOMBIES
I. I, Zombie
Not too long ago a certain Silicon Valley
concern was poised to shake up the cognitive neuroscience world
with a revolutionary new brain scanning machine. Existing brain
imaging technology, particularly PET or functional MRI, detects
metabolic changes in the brain, and thus offers indirect evidence
of neural function. The new device, in contrast, took two giant
steps forward from these indirect methods. First, owing to
techniques I'm not at liberty to discuss, this new scanner imaged
neural function directly. Moreover, and most important, the
device could be adjusted to directly detect *qualia*, the
heretofore unobservable subjective and feeling component of all
conscious experiences! The machine, dubbed the "Phenomenal
Energy Estimator with Linked Imagery Extraction," or
PHEELIE, was astonishingly small, in shape and heft much like a
blow dryer, and was to have been priced at under a thousand
As a philosopher of neuroscience, I'm often asked to be a "beta tester" of various brain-related products, and as a result I found myself with a prototype PHEELIE, mine to examine for several weeks. Naturally, I was quick to put my new toy into use. I had long believed that qualia were nothing other than patterns of neural activation (see Lloyd 1989, for example); finally I could look into the issue directly. My family and friends will tell you that I was relentless. In a few weeks I had not only frayed their patience but replicated a fair piece of brain imagery research, all the while with the qualia setting on. Though my subject population was very small, I found a nice convergence between the specific areas of activation reported in the literature as correlated with various cognitive tasks, and their qualitative dimensions. As a materialist, I was very pleased.
It was inevitable that I would ultimately turn the device on myself. One morning I turned on the PHEELIE, gave myself a good pinch on the forearm, and waited eagerly while the scanner processed the image. The result clearly showed the outline of my brain in three dimensions, but within the neural boundary -- nothing. No qualia at all. At other settings, I found I had all the metabolic activity appropriate to a human with a brain, but nothing qualitative. The scanner still found qualia among family and friends, and the PHEELIE technicians could find no mechanical defect. It was true. I had discovered that I was a living, breathing example of a favorite philosophical thought experiment: I was a grade A bona fide zombie! I was a being beyond the boundaries of B-movie imagination, beyond the legendary human zombies of Voudoun. I was a *philosophical* zombie, as described by several recent writers (e.g. Tye 1995: 22). I had all the right neurons, the right function, the right behavioral repertoire -- including dispositions to talk about my qualitative states -- but none of the REAL feelings. Apart from the PHEELIE scan, there could be no way to distinguish me from an ordinary human.
You can't imagine -- you really can't -- what it felt like to discover this about myself! My first reaction was philosophical, a complete reassessment of my thinking about the "zombie problem." Many philosophers have found the mere possibility of zombies to be a crippling problem for identification of consciousness with anything material. As Mchael Tye summarises the problem, "[I]f a person who is microphyscially identical with me, located in an identical environment, can lack *any* phenomenal experience, then facts pertaining to experience and feeling, facts about what it is like, are not necessarily fixed or determined by the objective microphysical facts. This the physicalist cannot allow.... The physicalist, whatever her stripe, must believe at least that the microphysical facts determine all the facts, that any world that was exactly like ours in *all* microphysical respects ... would have to be like our world in all respects (having identical mountains, lakes, glaciers, trees, rocks, sentient creatures, cities, and so on)." (Tye 1995: 23. The same argument can be found in Chalmers 1996: 97.)
Over the years I had been dismissive of this line of reasoning. It had always seemed to me that the "zombie problem," like the problem of other minds or the possibility that I might be dreaming, raised a skeptical specter of no consequence to the progress of an empirical science of consciousness. I compared the zombie issue to discussions of "twin water," the wet, thirst-quenching, universal solvent that is made of XYZ rather than H2O. XYZ-based water is logically possible, at least by some lights, and yet no one is inclined thereby to doubt that the stuff on tap in the kitchen is anything other than H2O, and no one is inclined to doubt the adequacy of the theory that water is H2O, strictly on the basis of possibility that it may be something else.
The zombie threat seemed momentous, I realized, only because of the relatively undeveloped state of consciousness science. Theories of consciousness, not unlike zombies, are themselves mere possibilities at this point. They are infant theories, loose collections of hypotheses, advanced tentatively by their authors with a hopeful "maybe" attached. Or maybe not, suggest the zombophiles. The zombie threat would naturally recede when a scientific theory of consciousness achieved a consensus through its powers of explanation, prediction, and taxonomy. For example, consider that familiar souvenir of mind-brain identity theory, the equation of pain and "C-fiber stimulation." At present, it hardly matters what is substituted for "C-fibers," since in every case it is easy to imagine someone with the firing C-fibers or whatever, but lacking the attendant pain. But now let the decades roll, until the C-fiber hypothesis becomes thoroughly woven into a tapestry of science, as tightly knit into it as the water-H2O hypothesis is laced into present chemistry, physics, biology, etc. At that point the zombie hypothesis would be analogous to the claim that it is possible that H2O is not water. In some quarters the water-H2O hypothesis is considered a necessary truth, but even if the possibility that H2O is not water makes sense, it is not a possibility that rattles the H2O-water hypothesis, so entrenched is it in both science and common sense. <1>
But all these comforting thoughts had fled. The possibility of zombies had collapsed into the reality of zombies. And not just zombies among us -- I, zombie! My philosophical reflection ceased as I faced the devastating personal consequences: The crushing burden, the desolate feeling of isolation from my fellow, I mean, my formerly fellow humans. The ironic part is that my shock and depression weren't real shock and depression. These human feelings were utterly lacking in me. Of course, being a regular zombie, I reported these missing feelings with exactly the hand-wringing and sobs you'd expect from a regular human. The worst of it was that all my relations to others have been reduced to a sham. Any exchange of empathy or understanding I may have had with anyone, I suddenly realized was a profound miscommunication, total misunderstanding. You couldn't share my pain, because I didn't have any. Appearances notwithstanding, I discovered that I was just not a sensitive guy.
After a period of depressing non-depression, I decided to at least try to rectify the chronic failure of communication. I had, after all, been using a raft of qualitative terms for my whole life. I'd just been using them differently. I could mark my zombie mental vocabulary in some general way, so as to indicate to my human friends that my words, so similar to theirs, nonetheless meant something radically different. So I decided to use a time-honored philosophical custom of attaching asterisks to mark a special usage of a term. For conversational purposes, I made a sign to hold up whenever I used a mental term to refer to myself. With this adjustment to my life, I felt* better. I very quickly came to realize* that a major part of my vocabulary should be marked as Zombish.<2> Not only my introspective* and emotional* terms, but huge expanses of descriptive language -- colors*, tones*, smells*, and anything that might derive therefrom -- perhaps my entire language. Sign in hand, I was able to assert* that Zombish has a rightful place as a special dialect of English. I felt* empowered*, enfranchised*, legitimated* at last!* I was no longer a mere zombie, but instead a zombie* -- that is, a human*.
The exhilaration* was short-lived. I remained lonely* in my uniqueness, which I assumed* would never change. But then one day, as I was enjoying* lunch with a friend(*?), I looked(*?) across the cafe to see* a woman with a sign just like mine, flashing away in the rhythm of animated conversation. She saw* me, and naturally a friendship(*?) was born. My new friend(*?) Aphelia turned out to be a well-connected zombie. It was she who introduced* me to Zombies Anonymous. Imagine, if you will, my joy* at a roomful of star-signs, all flashing away. Better still, at ZA we all understand* each other perfectly, and we've recently taken to leaving our star-signs at the door. In the blessed precinct of ZA, our hopes* and fears* are normal. We can treat them like (regular, human) hopes and fears.
One topic that we struggle* with at ZA is particularly relevant here: "zombie denial." The personal(*?) discovery* that one is a zombie has always been a matter of sheer chance, as it was when I happened to point the brain-scanner at myself. Had that never happened, I would still imagine* myself(*?) to be human. We suspected* that there are uncounted legions of zombies who simply don't know* themselves. Accordingly, several of us in the local chapter urged* the PHEELIE developers to create a wide-scan, long-distance version of the original device. They have done so, and as a result at this very moment a massive zombie census is being tabulated by the super-computers out in Palo Alto. Readers of this journal may be particularly interested in the current totals for those of you who are reading or have read this very article -- to this end, a Java Applet has been automatically and discretely downloading itself onto your work station. The applet signals the superPHEELIE dish to swing your way, and add you, dear reader, to the current totals. The applet contacts the database and in turn updates the text of this paper every ten minutes.
Here are the most recent numbers. Of readers of
this journal, so far the PHEELIE has detected
Wow*!* In the entire readership of this paper there is exactly one human! Just one of you experiencing an inner life, conscious of the colors and sounds and textures as they subjectively present to you -- I use all these consciousness terms here without stars, in your sense, even though I have no idea what it would be like to be you. I guess* you think you know who you are. As to the rest of you, welcome, my brothers and sisters. You'll get over* the shock*, I promise* you. This is a glorious* day of reunion.
Far from being a remote possibility, zombies turn out to be nearly ubiquitous in philosophical circles! We are the majority culture here. But it occurs* to me, why should we be the ones stigmatized with signs? You lonesome human, whoever you are, maintain that we lack all experiences. But we talk about our experiences* all the time. And we do psychology* and philosophy* of mind* about them. Indeed, as I consider* this distinguished readership, I have to conclude* that a lot of what has been recently published on the subject of consciousness, having been written by accomplished zombies, is really about consciousness*. Our experiences* are something*, which it will be the task of zombie science to elucidate. You say we lack experiences, which is true, but we say back to you that you lack experiences*, the whatever-it-is that underlies zombie phenomenology (or phenomenology* for short). Given our numbers, who should bear the asterisk? Who is the zombie here?
II. Don't Worry*, Be Happy*
We zombies, like so many other denizens of bad science fiction, trace our beginnings to an experiment -- in this case, a thought experiment. Let us imagine a being, said the experimenters, that is physically identical to an ordinary human, but utterly devoid of conscious experience. To the horrified delight of the experimenters, the thought experiment was not incoherent, and we zombies were thereby hatched into the realm of logically possibility. Being physically identical to regular folks, it followed that any materialist theory of consciousness, being based on discernible physical states of brains or behavior, would not be able to tell us apart from ordinary humanity. This is just what the experimenters wanted, but they have overlooked the other lesson of science fiction: When one trifles with the elemental forces of nature, there are always consequences beyond those one intends.
In our case, there is more to the story. We are indiscernible from humans by any observed criterion, and this entails that our behavior is indiscernible from typical human behavior, and this in turn entails that we will talk about ourselves with exactly the same introspective and experiential language that humans use. In short, as I described in the previous section, we have our own psychology, and it sounds so much like human psychology that you can't tell the two apart without the stars to guide you. But there really are two distinct psychologies. Zombie psychological predicates only sound like human psychological predicates. Our experiences* are not your experiences (which, after all, we do not enjoy).
From this point a constructive dilemma unfolds. Let us suppose that you humans set out to *understand* me (rather than merely unleashing me against certain theories of consciousness). The search for zombie psychology could be our joint undertaking, because we share a common ignorance -- you can't tell me from you, but neither can I! (Physical indiscernability, after all, is reflexive.) With no conceivable experimental strategy to follow, discovering zombie psychology is a task of pure logical analysis. I lack experiences, but I talk all the time about them -- just like you -- and so we slap a * on whatever it is I'm talking about. From there, as I discovered*, zombie inflections seep into wide swaths of language. Human psychology-talk will rapidly devolve toward a massively starred zombie psychology-talk, as we mark every term that must mean something different to me in my phenomenal vacuum. We will surely enjoy* the ensuing debates on the meaning of meaning and a maze of other concepts.
Suppose that we can agree on the range of starred terms entailed by zombiehood. Then zombie psychology will just turn out to be human psychology with the Zombish terms starred appropriately. Because Zombish terms carry meanings radically unlike their human counterparts, the two psychologies describe entities (minds in your case; minds* in mine) that are radically unlike, ontologically as different as night and day. But it is a strange sort of difference, because, as always, the words in which we express the two psychologies will be the same. We might say that our two psychologies are identical on the surface.<3> I don't need to know whether you are human or not, because any claim I make of you will be the same surface claim, and carry with it the same surface consequences, whether you are human or zombie. It turns out, therefore, that I can abandon my star-sign altogether. If you are human, you will hear my claim automatically translated into your language, even though I utter it in Zombish. And if you are a zombie, then you will hear my claim in Zombish, just as I meant it. We can work side-by-side to develop our theories of consciousness, based on the best empirical research conducted on either humans or their physically indiscernible zombie allies, and reach full agreement. We will be in surface agreement only, but each of us will understand the theory in the way appropriate to our nature, and our collaboration can proceed indefinitely on that basis.
And this is only the beginning. It may be, indeed it is likely, that there are many, perhaps an infinite number, of *zombie types*, one for each logically consistent assignment of stars to the predicates in a language. For example, perhaps there are color zombies, beings who have no color experiences whatsoever, but are otherwise human. These beings are not exactly color blind, since they are, once again, indiscernible from humans and other types of zombies, which in turn entails that they will point to red patches when asked, avoid lime-green pants, and hold an informed opinion on the merits of fauvism. Regular color-blind humans can't do these things, and their condition is accordingly easily detected. No zombies they.
Furthermore, without the PHEELIE to guide you, you can't be sure whether you are a zombie or not. Of course, you are eager to claim the reality of your experiences. But so was I. And you are ready to argue forcefully that you are human. So was I, and so will any zombie. After all, with physical indiscerniblility comes every other kind of resemblance. The thought experiment unleashes a plague of zombies, and a vast army of zombie types, at the same time afflicting all with the skeptical worry that our surface psychology may turn out to be one of a number of psychologies*. But on reflection the antidote emerges. Without the PHEELIE, my zombie claims to experience are as legitimate as human claims, because the word "experience" carries a different meaning for me. In the end, there are many different types of experience, as many types (plus one, the human) as there are types of zombies. But this parade of differences makes no difference at all, since one and the same surface psychology describes all the possibilities.
At this point it would be a kind of arrogance to maintain that humans have the only kind of consciousness that really deserves the name. Look at it this way. Suppose I was not a zombie but an extraterrestrial, still without experiences of the sort detected by the PHEELIE. Yet imagine that your best efforts to translate my language suggest that I am as capable of talking about experiences as any human. I have terms that make sense as counterparts to "pain," "red," and so forth. Would it not be pure chauvinism to let the scanner have the last word and assert that only ordinary human experiences count, and that I therefore have no experiences? In this case, the coherence of my alien self-descriptions would surely be decisive in favor of my mentality, however different that mentality might be from the ordinary human case. We zombies merely ask for the same consideration.<4>
Finally, a kind of vertigo overtakes us all. Are you...? Am I...? Who is...? Earlier, I mentioned a constructive dilemma. We've now toured lemma one: Zombie psychology can be consistently constructed by marking terms within human psychologies. But perhaps this strategy will not work. Perhaps the zombie hypothesis entails that a specific term must be starred, but other considerations militate against it. For example, can a zombie like me discover a new element? (I owe this example to Hauser 1995.) Discovery is an epistemic achievement, and as such arguably requires that the discoverer *notice* his, her, or its discovery. Zombies don't notice anything, and so discovery* is the best they can hope for. But if I call a press conference and announce the discovery of Zombesium, it would seem obtuse to claim that as a mere zombie I had not discovered anything. If I didn't discover the element I'm describing, then who did? The concept of discovery is unsteady in a zombie infested world, as are many other concepts. Perhaps, as in lemma one, the ambiguities do not lead to contradictions. But perhaps they do. Perhaps there are terms that cannot be consistently assigned to the human or zombie camp. If so, then by virtue of incoherence, we zombies cannot exist after all.
III. To conclude...
The zombie hypothesis begins with the
conjunction of physical indiscernability and the absence of all
conscious experience. Physical indiscernability entails
indiscernability of utterances, a subset of which is zombie
psychology (indiscernible from human psychology). If the terms of
zombie psychology can be reinterpreted in a way consistent with
the absence of experience, then there are one or more logically
possible types of zombie (one for each consistent
interpretation). (Note that the surface similarity is essential
to zombies. Two psychologies which are not surface identical
enable us to tell zombies from humans.) However, given
1. The surface similar psychologies will be equally true in general, if either is true. Any predictions, etc. of the one are the same in the other.
2. How you interpret claims made of others, not of your type, will not matter, since the claims will be surface identical, and all entailments or explications of the claims will be surface identical.
3. How you interpret claims made of you will not matter, because you will understand them in the way appropriate to your type.
4. You will not be able to tell whether your interpretation is human or zombie.
5. You will not be able to tell whether you are human or zombie.
6. You will claim you are human.
7. You will believe you are human (although this statement has different meanings according to your type).
On the other hand, if the terms in zombie psychology, indiscernible from human psychology, cannot be interpreted in a way consistent with the absence of experience, zombies are logically impossible.
In the end, then, I am saved not by my stars or by the support of Zombies Anonymous, but by the logical grace of pure philosophy -- just as it should be. The zombies' tale is not one story, but many possible stories, reflecting all the configurations of zombie types, or -- on the other hand -- collapsing into fundamental incoherence. But whatever the story, it is unimpaired by zombies, leaving materialist philosophers of mind to pursue their theories of consciousness, happily ever after.
1. Another popular tactic is to challenge the initial premise of the zombie arguments, namely, that zombies are imaginable and thus conceptually or metaphysically possible (Dennett 1991, 1995; Tye 1995). But this escape was suddenly closed to me -- I was more than just imaginable. I was a real zombie.
2. As I began to look* into the literature of my condition, I found* that Todd Moody (1994) had also adopted the strategy of marking terms in Zombie-English to distinguish them from ordinary English. Moody's fantasy ran toward "Zombie earth," a planet populated entirely by Zombies. Would they develop or even recognize the psychological vocabulary of humans? He argues that they would not, and thus that Zombie culture, at least, might be distinguished from human culture. While this is an interesting variant on the Zombie thought-experiment, the initial problem for materialism remains untouched as long as one Zombie might be passing as human. (A similar point against Moody is made in Bringsjord 1995.) Even the minimal case -- the condition I seem to exemplify! -- is enough counterexample to challenge materialist theories of mind.
3. The psychologies must be surface-identical just because, where zombies are concerned, every aspect must be indiscerible from its human counterpart (a point also emphasized in Dennett 1995). If this condition were violated, the exception would be the "mark of zombiehood." In that case, one could sort zombies from humans, and the "zombie problem" would evaporate as a philosophical challenge to materialism. One would solve the problem by proposing one materialist theory of consciousness for humans and another (a theory of consciousness*) for zombies.
4. Dennett (1991) argues from indiscernability and the threat of chauvinism to a stronger conclusion, namely, that we should regard zombies and humans as conscious to the same degree. (This is amplified in Dennett 1995.) In contrast, I've granted the zombie premise of "real" difference, but argue that in any reasoning, including ethical reasoning, any conclusion reached about a human will also be reached about a zombie.
Bringsjord, Selmer (1995). In Defence of Impenetrable Zombies. *Journal of Consciousness Studies*, 2(4): 348-351.
Chalmers, David (1996). *The Conscious Mind*. (New York: Oxford University Press)
Dennett, Daniel (1991). *Consciousness Explained*. (Boston: Little, Brown)
Dennett, Daniel (1995). The Unimagined Preposterousness of Zombies. *Journal of Consciousness Studies* 2(4):322-326.
Hauser, Larry (1995). Revenge of the Zombies. Paper presented at the Eastern Division of the American Philosophical Association.
Lloyd, Dan (1989). *Simple Minds*. (Cambridge: MIT Press.)
Moody, Todd (1994). Conversations with Zombies. *Journal of Consciousness Studies* 1(2):196-200.
Tye, Michael (1995). *Ten Problems of Consciousness: A Representational Theory of the Phenomenal Mind*. (Cambridge: MIT Press.)