(Note: this entry is still in rough draft form.)
In my own essay IV where I argued my position on why subjective experience can’t be given any physicalist analysis in principle, I avoided any mention of “zombie arguments” by name completely. There are enough confusions about what the purposes of zombie arguments are, and how it is that they’re supposed to achieve them that I thought I could make my own points—which are in the end the same points that the “zombie arguments” ultimately get at—more efficiently simply without even invoking them and then setting up the task for myself of cleaning out all of their baggage. As a supplementary note for any interested readers, however, I’d like to do so in this separate footnote. In part, this will be a clarification of a few things that “zombie arguments” are and are not supposed to do; in part, it will be a clarification of where exactly I see them fitting into my own argument.
To begin, I want to offer a quick overview of the major misconceptions. For now, you’ll just have to take my word for it that they are, in fact, misconceived—the reasons why will become clearer and clearer as my explanation proceeds into what they actually try to do; what I think they can do much more effectively if we make certain small tweaks to the way the argument is presented that can get us around some massively irritating, excessively abstract technical quagmires that the argument often ends up in; and as I explain the arguments which I think get down to the root of what the “zombie arguments” actually want to say without those problems. In short, I think that the zombie arguments are sound—but I think they are a terrible strategic choice for making the conclusion they aim to prove “make sense” to the person whose position they reject. The argument has become hopelessly abstract, muddled by sub–arguments about things like the relationship between logical and metaphysical “possibility”—things I’m willing to bet that no one on Earth actually wants to talk about, and things I’d wager probably no one—I’d wager not even a philosopher—can actually reach any clearly understood conclusions in their mind by thinking about. But I think the argument doesn’t need to be this abstract. With only a few minor tweaks and clarifications, the issue can very easily become a lot more straightforward—and easier to understand—while still doing everything that the standard zombie argument ordinarily tries to achieve.
If conceptual clarity—turning “knowledge” into “understanding,” in P. M. S. Hacker’s words—is the goal of philosophy, then philosophers have been failing their job atrociously. Philosophy does, of course, require the invention of new and sometimes complex terminology in order to clarify the language we use to talk about concepts; ordinary language is often muddled and imprecise, and it is sometimes the needed task of philosophy to recalibrate it with clearer distinctions than everyday language ordinarily has—but it should aim for as much simplicity and clarity as possible except where abstract terminology is actually needed to make a point. If philosophy can’t learn to communicate clearly with non–philosophers, then the fault is not all on non–philosophical disciplines or the laity if the latter end up ignoring philosophy’s insights. And in ordinary formulations of the zombie argument, technical complications—much like the walking dead themselves—well outlive their actual usefulness.
_______ ~.::[༒]::.~ _______
⌦ One of the most common misconceptions is that these arguments somehow either imply, try to prove, or rely on the assumption that zombies should actually be possible—taking the word “possible” to mean “actually possible in our particular world, under the peculiar laws of our particular world, whatever they are.” Not only is this misconception widely prevalent in countless “amateur” discussions I’ve seen firsthand (someone recently told me: “You also take as granted that zombies could exist, and behave exactly as if they were sentient, without being sentient. We have no experimental data to back this. … I don’t think that zombies can exist. But … that is an experimental issue.”), it also exists in the literature.
⌦ The most significant complication that results from the way the argument is formulated is that, at least as Chalmers presents it, it requires justifying a premise which Chalmers defines as so: “If it is conceivable that there be zombies, it is metaphysically possible that there be zombies.” This, in turn, invites room for a complicated debate over the relationship between prima facie conceivability, logical conceivability, logical possibility, metaphysically possibility, and any number of related concepts that may or may not collapse into others that muddles the entire debate in hopelessly arcane abstraction. I doubt that anyone—even a philosopher—can reach clear conclusions in their mind by reasoning through concepts like these, and I’m equally confident that no one actually holds any intrinsic interest in that sort of conversation in the first place—but I think the argument can be only slightly reformulated in a very simple way that entirely avoids any need for any premise of this sort.
⌦ One unnecessary complication that follows from the way that the argument is formulated is that it leads to the assumption that we have to imagine the possibility of zombies that behave exactly identical to the human beings that exist in our world. A corollary that follows from this confusion is the deeper confusion that the zombie argument could only succeed at establishing its point if epiphenomenalism were true—that is, the critic reasons that we could coherently imagine “zombies” who behave exactly as we do without having conscious experiences or intentions in the way required for the argument to go through only if conscious intention and experience played no role in our behavior.
⌦ Finally, the most basic objection is that the argument merely begs the question, because if physicalism is true, then zombies aren’t ‘metaphysically possible’ after all. I want to set the tone for that discussion with a quote from Dmitry Sepety: “Typically, begging the question is described as an “informal fallacy where the conclusion that one is attempting to prove is included in the initial premise of an argument, often in an indirect way that conceals this fact”. However, such a definition creates the problem of interpretation: how are we to understand the statement what in an argument, the conclusion is included in the initial premise in an indirect way that conceals this fact? The problem arises because in a sense, any valid argument (an argument where the conclusion logically follows from the premises) contains the conclusion in the conjunction of its premises; otherwise, it would be impossible to (validly) draw the conclusion. Logic is not a hocus-pocus: you cannot draw a “rabbit” from a “logical box”, if it is not there. That is why some logicians say that all valid arguments beg the question. However, if this is the case, then, obviously, begging the question cannot be a fallacy. On the other hand, if begging the question is a fallacy, then for an argument to beg the question, it is not sufficient that its conclusion is contained in the conjunction of its premises—some further conditions are needed. What are these conditions?”
_______ ~.::[༒]::.~ _______
To start with, arguments from the conceivability of zombies have nothing to do with any notion that zombies should actually be possible in our actual world, given whatever particular laws or kinds of entities it actually has. This particular kind of “possibility” is what Chalmers refers to as either “natural possibility” or “nomological possibility” (the term “nomological” refers to the laws of nature). In Consciousness and Its Place In Nature, Chalmers plainly states: “Zombies are probably not naturally possible: they probably cannot exist in our world, with its laws of nature.” It couldn’t get any more clear than that.
“But the argument holds that zombies could have existed”—and pay very close attention to the phrase he follows this up with: “perhaps in a very different sort of universe.” The argument does not make any assumptions either about what kind of universe we are actually in, or about what kind of properties the stuff we think of as “physical” could perhaps surprisingly turn out to possess. It just says we can imagine worlds with physical properties like ours without being thereby logically compelled to assume that the entities that exist in that world have conscious experiences. And we most certainly can do this—even if it should turn out that all we’re really doing when we do this is abstracting away from the material entities our world actually has whatever properties it is in virtue of which they actually do end up producing consciousness.
In contrast, consider the relationship between the macro–behavior of water and the micro–behavior of molecules of H2O: you can’t conceive of a world that has H2O that behaves like the H2O in our world without thereby imagining a world that has water that behaves like the water in our world. If you imagine a world that has H2O that behaves exactly like the H2O in our world, you are logically compelled to imagine that the water in that world behaves exactly like the water in our world—because the fact that molecules of H2O form loose bonds (for example) just is demonstrably identical to the fact that solid objects ‘sink’ when placed in water—by slipping through the gaps in those molecular bonds. To spell this out in the most simplified possible terms, I can bring you with me to a giant chalkboard and draw a giant close–up view of the molecular interactions between molecules of H2O and molecules of some other substance slipping through the gaps between them. Then, I can bring you some twenty yards or so back away from the chalkboard, and you will see that what you have just drawn is literally “something sinking in water”—and you will see that there can be no other way.
On the other hand, I can draw all the neuronal interactions defined in external, objective physical terms I like, and no matter how detailed that drawing gets, there is no distance I can stand from that drawing at which I will suddenly see a subjective, privately experienced qualitative representation of something other than those interactions themselves “inside of” those interactions. No one has even the faintest hint of a clue how to so much as even begin hypothesizing about how such a thing might be the result of inert causal interactions of any kind whatsoever—not the slightest pitiful speck of progress has been made on that question since Leibniz posed it in 1714 when he wrote that: “ … perception, and that which depends on it, are inexplicable by mechanical causes, that is, by figures and motions, And, supposing that there were a mechanism so constructed as to think, feel and have perception, we might enter it as into a mill. And this granted, we should only find on visiting it, pieces which push one against another, but never anything by which to explain a perception.”
As I wrote in my essay (IV) — The Case of the Lunatic Fish, “We aren’t merely failing to see how an explanation from tools like these could be possible; we can positively see that an explanation of a phenomena like this with tools like this cannot be possible—in just the same way that we can see that a two–dimensional canvas is not capable in principle of allowing us to draw a three–dimensional object on its flat surface. Picture all the blind physical entities you like moving in any inert causal pattern you wish—at no point are you just going to be literally looking at a subjective private conscious experience. You don’t have to sit and contemplate the entire near–infinite combination of ways to picture blind physical forces moving through space in order to see why” — any more than you need to spend centuries drawing lines every which conceivable way on the flat canvas to justify the determined conclusion that getting any three–dimensional figure onto its surface is going to be, in principle, utterly impossible — “It’s right there contained within the very concepts themselves.”
_______ ~.::[༒]::.~ _______
Notice, again, that Chalmers is not presupposing that the world is necessarily purely mechanistic in this way. What the argument aims to establish is simply that if it were purely mechanistic, then consciousness as we know it could not have appeared. Perhaps you might think this is true, but trivial—because no real materialist actually thinks the world is just made up of blind forces exerting causal influence over each other and nothing else.
Great! Chalmers doesn’t have to disagree, and the logic of the argument doesn’t imply that he should. The argument is not meant to settle the entire investigation; it merely aims to establish a sensible starting point: “If the world was like that, then consciousness as we know it could not exist. Alright, so what does the world have to be like, then? What—upon adding it in—would alter that picture in such a way as to render it intelligible that consciousness does in fact appear?”
For his part, Chalmers goes on to consider panpsychism—in other words, the possibility that the entire physical world might be full of experience all the way down to its most fundamental core—as a possible solution; and his ultimate answer is to suggest that we must simply posit a brute set of “psycho–physical” laws determining what experiences are had in our world alongside the ordinary “physical” laws determining what causal events take place. Both of these proposals work exactly by trying to find some way to suppose that the actual natural world we live in is not as the zombie argument itself has us imagine it might conceivably have been!
I happen to think that both of these solutions also fail in principle (panpsychism for reasons I explained in entry (VII) and Chalmers’ psychophysical law posit for reasons I haven’t discussed yet) and that the only plausible answer turns out to be that the thing we need to add to the picture is just simply consciousness itself—consciousness conceived as a basic phenomena in the same sense in which, say, the electromagnetic force is considered to be “basic” (see entry (III)) with a unique, basic defining nature of its own (represented in such concepts as subjectivity; intentionality—see entry (V); primitive identity—see entry (VI)—etc.) But either way, the point here is that as soon as Chalmers, I, or anyone else turns to consideration of these possible solutions or any others, we are moving beyond the zombie argument itself.
So the idea that zombie arguments beg the question against materialism in the sense that they make assumptions about the internal composition of the material entities in our world and therefore rule out the possibility of the “emergence” of consciousness from “matter” out of hand because of this assumption rests on an absolutely fundamental misconception about the nature of the argument. This is simply not what the argument is actually even trying to do.
The point is not that we can conceive of worlds with material entities which are necessarily in all possible respects exactly identical to ours which do not have conscious entities by consequences, and that this is supposed to prove that nothing about the composition of our own material world as it is could potentially explain consciousness (not, at least, so far as the zombie argument itself goes); Chalmers is always intent to clarify that his only premise is that zombies are conceivable “perhaps in a very different kind of world”—so it is simply false to say that it begs the question because materialism implies that the material entities in our world produce consciousness as a logical consequence of their intrinsic composition. The point is simply that we can conceive of worlds that have only the causal mechanistic properties that the material entities in our worlds have—leaving entirely aside whether or not these causal mechanistic properties are the only kinds of properties that the material entities in our world actually have—without it following as any logically entailed consequence that subjective conscious experiences therefore exist. Even if the material entities in our world do have properties other than the raw causal mechanistic properties of physics, we can conceive of a world where they have only these properties.
So the conclusion properly drawn from this is that it cannot be in virtue of those causal mechanistic properties that subjective conscious experience exists. Perhaps, so far as the zombie argument itself is concerned, there are some entirely different sort of properties which the material entities in our world possess in virtue of which the appearance of subjective conscious experience out of such ingredients can be rendered intelligible. If so, of course, we should be able to specify what they are, or might be—and the debate can move on from there. But the actual premise here does not beg the question against the particular premise by which materialism assumes that consciousness will appear by logical necessity in light of some facts about the material entities in our world. The materialist himself, of course, does not think that causal mechanical properties logically necessitate the appearance of subjective conscious experiences either—and this is given by the fact that he himself thinks that atoms (or whatever micro–physical entities you may wish to substitute) are capable of possessing causal mechanical properties without having subjective conscious experiences.
_______ ~.::[༒]::.~ _______
What would “moving forward from here” have to entail? First, I want to emphasize yet again that the moment I begin to discuss this question, I am moving on from the zombie argument itself. Still, I want to give the outline of my own answer.
To begin with, I think we have to establish the following premise “ … the plainest thing in the world to see is that the question of whether something is an experience or not is absolutely binary: the answer is either “yes” or “no,” and there are absolutely no steps in–between the two. The question of when a pile of sand goes from being a “heap” of sand to becoming a “mountain,” for example, is one that has rough edges: at exactly which point in the process of removing singular grains of sand from a “mountain” has it devolved into a “heap?” At exactly which point in the process of adding singular grains of sand to a “heap” does it become a “mountain?” Reasonable people could disagree, and there is no objective way to determine the answer. Some questions are like this: the question of when a new “species” has evolved has rough edges, and evolution can address the transition from one species to another through the small, gradual steps that are involved without needing to bridge any fundamental gap of absolute difference between an original “species” and a second. But the question of conscious experience is not like this—the difference between something being a subjective experience and something not being a subjective experience is as absolute as absolute can get. There may be various degrees of complexity or sensitivity or detail between experiences, but either something is an experience or it isn’t. There is no middle ground between the two—but this also means there is no ground that can be covered in any gradual steps as a means of bridging the gaps between the two. And there is, therefore, no way to proceed gradually in steps from non–experience to experience.”
Therefore, the only kinds of properties we can even propose as a candidate for attributing to microphysical entities in virtue of which subjective conscious experiences of the kind which compose our existence could coherently be supposed to “emerge”—just turns out to be subjective conscious experience itself, as Thomas Nagel argued in his 1979 article, “Panpsychism.” (Thomas Nagel is far from the only figure to argue for panpsychism, but he was one of the first in modern times to propose this particular type of panpsychism in motivation from these specific kinds of reasons.) I think panpsychism is a respectable attempt to answer these problems—it just turns out to fail for other systematic reasons of its own: namely, by either entailing epiphenomenalism (which can be refuted for its own separate reasons) or else implying its own equally insoluble and incoherent version of the ordinary Hard Problem, depending on the details of how it is formulated. I conclude in that entry, Panpsychism: Panacea, or Flash in the Pan? that “ … panpsychism doesn’t foot the bill. In fact, all it does is create the illusion of doing so by turning the tab upside down so that we might not so easily recognize the numbers that are now upside–down and on top of the tab instead of in ordinary, face–up recognizable form down at the bottom where we expect to see them. …
… We might say that the deep, fundamental conceptual gaps between “physical properties” as we have defined them (“mathematically describable geometric structures and mathematical–functionally describable tendencies towards patterns of spatiotemporal motion”) and the subjective, qualitative, phenomenal, intentionalistic (etc.) aspects of experiential consciousness are rather like the Grand Canyon. If the conceptual gaps are the Grand Canyon, then the intractable problems that appear on the ordinary materialist views which say that everything that makes up the human mind are at root ultimately ‘physical’ are the “jumping across the Grand Canyon from East to West” problem.
_______ ~.::[༒]::.~ _______
So really, the only thing the zombie argument is supposed to prove straightforwardly is that any simplistic “identity theory” between physical and mental properties is false—from there, we’re free to try to find any alternative solution to it that we like. Does the premise that zombies are conceivable “beg the question” against the literal identity theory?
I’m going to be lazy and just quote further from Dmitry Sepety’s excellent article, “The Zombie Argument against Materialism Without the Conceivability–to–Possibility Inference” again: “The zombie argument can start directly with the contention that phenomenal zombies are logically possible: there is no a priori contradiction in the idea of a zombie. Let us designate this contention as the Zombie Possibility Thesis.
Why the thesis is charged with begging the question against materialism (the identity theory)? There seems to be no other reason except that if the identity theory is true (that is, if mental states are identical with some brain states), then (it would follow that) the phenomenal zombies are logically impossible. However, it is arguable that this objection puts things on their heads.
Of course, it is really the case that if the identity theory is true, then the Zombie Possibility Thesis should be false―just like if the Zombie Possibility Thesis is true, then the identity theory should be false. The Zombie Possibility Thesis contradicts the identity theory; thus, they cannot both be true, and thus at least one of them is false.
We are to notice that this situation is not specific for the relationship between the Zombie Possibility Thesis and the identity theory―it is the situation that is common for all arguments. How otherwise can you argue against any theory? Any such argument is a contention (which may be a conjunction of simpler contentions) that we think to be true and that contradicts the theory at issue, from which we conclude that the theory is false. If the Zombie Possibility Thesis begs the question against the identity theory, then any argument against any theory begs the question, and if begging the question is a fallacy (as textbooks of logics usually tell us), then no nonfallacious argument against any theory is possible. If so, nonfallacious arguments for a theory are impossible too: any argument for a theory is an argument against all alternative theories and, therefore, begs the question against them, and, therefore, is fallacious. Thus, no nonfallacious argument is possible at all! On the other hand, if there are nonfallacious (not question-begging) arguments at all, then the fact that the Zombie Possibility Thesis contradicts the identity theory does not mean that the Zombie Possibility Thesis begs the question against the identity theory and that, therefore, an argument based on the Zombie Possibility Thesis is fallacious.
[Note: the part of Sepety’s argument which I fully concur with without my own qualification stops here.
I want to continue quoting him anyway. Just note that our positions don’t completely align past here.]
Now, let us consider things as they really stand on their “logical feet.” To begin with, there are several alternative theories about the mind–body relationship: several varieties of materialism (including the identity theory), dualism, idealism, and panpsychism. We look for arguments for and against these theories. Any such argument has premises. For an argument to be convincing, its premises should be plausible. Of course, their plausibility should be evaluated independently of the theories at issue. It would be really fallacious begging the question if we judge the plausibility of the premises of the proposed arguments by their consistency with our pet theory. On the contrary, we should (try as hard as we can to) begin with a neutral (with respect to the theories at issue) standpoint. To begin with, we do not know whether the identity theory, or interactionist dualism, or epiphenomenalism is true. Now, without assuming the truth of any of these theories, let us consider the question: are phenomenal zombies logically possible?
It seems that they are: imagine an exact atom-to-atom (or quark-to-quark, if you like) copy of your body, so that each atom of your zombie-twin is located and moves relative to its body exactly as the corresponding atom of your body relative to yours. There are all just the same physical interactions between atoms, and all just the same physical fields. Because neither atoms (quarks) nor physical fields experience anything, it is logically possible for all those processes to occur without any experiences.
If someone thinks otherwise, it is incumbent upon him/her to explain how those movements and interactions of atoms and physical fields can logically necessitate subjective experience. (A mere postulation of the identity of mental states with certain brain states does not count as an explanation.) If no such explanation is available, we should admit that there is no such logical necessity; that is, phenomenal zombies are logically possible. From this, we should proceed to what logically follows as to the truth/falsity of the theories at issue.
Unlike the Zombie Possibility Thesis, the identity postulate is not prima facie plausible at all. On the one hand, there are physical structures and processes―microparticles with certain spatial locations relative to one another that move (change their spatial locations) in certain ways and influence one another’s movements (interact according to the laws of physics) and physical fields (which are spatially distributed and changing with time in a law-abiding manner dispositions of influencing the movements of physical bodies―in particular, of microparticles). In all this, nothing implies (logically necessitates) subjective experiences. On the other hand, there are subjective experiences―what it is like, how it feels for a person to have a certain experience. The postulate that subjective experiences are identical with some physical structures and processes, as a mere postulate, without a substantial explanation of how subjective experiences can be identical with some physical structures and processes, is not merely prima facie implausible, but downright unintelligible.”
I found this article after writing my own entry (IV) — The Case of the Lunatic Fish. But in that entry, I expressed the point Sepetry concurs with in my own way: “ … circularity and ‘begging the question’ are not fallacies of thought, but fallacies of argument. An argument is circular and will ‘beg the question’ if it contains premises which will be seen with equal skepticism by someone who is skeptical of the conclusion of that argument for the same reasons they are skeptical of the conclusion. This is classified as a fallacy because the goal of an argument is to prove that skepticism wrong to the satisfaction of the skeptic—so an argument that begs the question fails at this task because it merely repeats implicitly the conclusion the skeptic doesn’t want to accept as one of its assumptions. The fact that the question-begging argument fails to objectively disprove the skeptic doesn’t mean, however, that any train of thought that is circular is either false or irrational for an individual to accept. The real question worth asking is: “Is this circle making contact with reality?”
And there very well may be true statements which we absolutely cannot, in principle, support in any way without at some point “begging the question.” To return to a previous example, … solipsism …. … You absolutely know without a shred of doubt that he is absolutely wrong—and yet, you just as absolutely have no conceivable way of “proving” it to him with any sound, non–circular argument. Appropriately, the example of solipsism deals (in different ways) with the same subject matter addressed in philosophy of mind: private subjective experience. The solipsist denies its existence anywhere but in the one case he experiences immediately and directly—his own. For this, the solipsist is universally considered absurd. Yet the eliminative materialist goes on to not only do that, but to deny it in even the one case he actually experiences indisputably, immediately and directly for himself—and for this he’s respected enough to publish in prestigious philosophy journals.”
In any case, I do think we can go one extremely important step further from here—I wanted to let Sepety speak first, because simply dismissing the argument as “question begging” is problematic in its own right and in a way well worth addressing on its own terms; because even if it was, that actually still wouldn’t be a reason to cast it aside as trivial or uninteresting. If all that we actually had here in the end was something like: [(intuition A) ⇆ (valid logic) ⇆ (intuition B)], then this can still be a valuable way to set the tone for a discussion over whether (intuition A) or (intuition B) is more plausible—for asking: which do we have more reason to accept, given that accepting it would mean rejecting the other? In the worst case scenario, if we couldn’t strictly prove either premise true, we still would have established that these are the two premises one has to choose between—and even if we couldn’t find any objective grounds by which to establish the truth of one or the other, then we would still have established that it comes down to a matter of intuition. The materialist wouldn’t be able to say against the zombie arguer that the zombie arguer is wrong unless he could prove his own choice of intuition true—he could only say “we begin from different starting points, because my gut instincts lead me to favor intuition B over intuition A, and yours lead you to start from intuition A. But who knows which of us is right? Is there any way we can find some neutral territory on which to settle the question, or is it really just so up in the air?” If the intuition that zombies are logically conceivable is, by default, to be considered untrue until proven true, then the same standards would have to go likewise for the so–called “identity theory.” Otherwise, we’re just rigging the courts by treating one suspect innocent until proven guilty and the other guilty until proven innocent. And that wouldn’t be any fair trial.
But Sepety doesn’t make the point nearly strong enough: the “identity theorist” materialist does not actually think that causal mechanical properties and instances of subjective conscious experience really are just literally identical. Not even by his own lights! Consider: the materialist himself thinks that atoms (or substitute whatever microphysical entities you like) do—or at least could! which is the only premise we actually need in order to insist on logical conceivability of their conceptual separability—possess causal mechanical properties without having subjective conscious experiences. So this premise really doesn’t even beg the question in the weak sense against the identity theory, because the “identity theorist” himself is in fact caught in an internal contradiction, whether he recognizes it or not. The only option that the materialist actually has here is to say that it is some other property of the material entities in our world besides their purely physical geometric structural properties and blind physical dispositions towards various inert patterns of motion through physical space in virtue of which our subjective conscious experiences can be intelligibly supposed to appear.
First, this—again—simply is not contrary to what the zombie argument as such actually aims to establish. If the materialist has acknowledged the need to provide an account of this sort at all, the zombie argument has already succeeded at its quite modest aims and we’ve already moved beyond that argument itself. But second, my evaluation of this attempt to solve the question is as so: because the question of whether something is or is not a subjective conscious experience (distinct from the question of how complex or robust or detailed a given experience is) is an absolute binary, there is no way in principle to move in “steps” from something that is not an experience into something that is. So, the only possible candidate for what this “other property” could even conceivably be is in fact conscious experience itself—therefore, unless the “identity theorist” wants to take the alternative route of eliminating consciousness as we know it and convert his statement into something like “the brain (conceived of as a composition of entities with purely physical properties, as defined above) is identical to the brain” while leaving the mind out of it entirely, his only alternative option is to try to adopt panpsychism (which may or may not run into equally decisive problems of its own). Otherwise, the “identity theorist” himself is caught in an absolute self–contradiction given the fact that he himself fully (even if only implicitly) acknowledges that “physical properties” as defined here can in fact exist without subjective conscious experiences existing with them any time he supposes that microphysical entities can in fact possess these kinds of physical properties without therefore having subjective conscious experiences by logical necessity. (Note, however, that part of my own argument against panpsychism proceeds from the realization that the physical properties and the properties of subjective experience and intentionality still wouldn’t be “identical” even on panpsychism—as given by the fact that the panpsychist can still logically conceive of “zombie atoms!” Even universal concomitance wouldn’t constitute an “identity” claim. In a world with different chemical properties, it could have been the case that every molecule of (other–world) oxygen would universally come bound to two molecules of (other–world) hydrogen and never chemically decompose, and it could be naturally impossible for anyone to break them down—perhaps because there was nothing in this world other than (other–world) H2O; we could, perhaps, even imagine a world containing nothing besides one giant ocean—but this still wouldn’t make (other–world) hydrogen and (other–world) oxygen identical.)
Our ordinary concept of consciousness—derived from our immediate, first–hand acquaintance with it “from within” our experience as conscious centers of experience—simply contains ingredients which our concept of physical causation manifestly does not. An “identity theory” that simply declares the two “identical” is just literally incoherent—it literally does not even rise to the status of a position on philosophy of mind—it is “not even wrong.” (See the following section for further elaboration.) The only way to even try to begin to formulate it into a position at all is either by adding something fundamental to our ordinary concept of the “physical,” or by taking something fundamental away from our ordinary concept of “consciousness.” Simply declaring the two “identical” by fiat doesn’t even begin to attempt anything like this.
_______ ~.::[༒]::.~ _______
Of course, in my own writing up to here, I have made an explicit argument against the so–called “identity theory”—and the prevalence of well–ingrained misconceptions about the zombie argument is exactly why I made the strategic choice to simply go straight to those supporting arguments themselves, instead of presenting them explicitly in light of the zombie argument in particular: “ … the conceptual ingredients involved in efficient physical causation and the conceptual ingredients involved in subjective, qualitative, phenomenal, intentionalistic thought and experience simply are not identical. And providing an account which “identifies” them would require a conceptual unification of a sort that takes some third kind of phenomena and explains in those terms exactly how the concepts of subjective experience and physical causation are unified through it. To reiterate the analogy once again: to claim that the man who delivers my mail in the morning is identical to the man who drinks at the bar on Friday night is to take two spatiotemporally conceived events and then provide spatiotemporal terms that perform the actual substantive work required to actually link them in space and time—namely, it requires a story like this: “when the man who delivers my mail on mornings goes home, he changes clothes and heads out to the bar—and that is how the man who delivers my mail turns out to be the man who drinks at the bar on Friday night—discovering this additional fact is how I know it turns out to be the same man at all.” If I don’t have an account like this, then I am simply not justified to declare that the two are the “same man.” And if I can actually see the two standing side by side at the same moment (as I can for my physical brain and my subjective stream of conscious experiences) and see that they very well don’t even look alike, no less, then the statement is just literally incoherent unless and until it gets a whole lot of justifying explanation.
In a proper account where two things that weren’t obviously identical at first later empirically turn out to be, a bridging spatiotemporal event links two other spatiotemporal events together in space and time; two events composed of the same basic category of ingredients are linked by an account which actually bridges them in the clearly explicable terms of that same exact ingredient. But without an actual bridge to actually connect these two things in common terms, calling them “identical” would simply be incoherent. I can potentially provide an account which “identifies” the man who delivers my mail in the morning with the man who drinks at the bar on Friday night, but I cannot even potentially provide an account which “identifies” the man who delivers my mail in the morning with the year 1977—the very terms involved in the two different concepts are simply different. And the notion of “identifying” subjective first–person qualitative experience with physical structure and causal process is a conceptual confusion more on par with the latter example than with the former, not merely because the two concepts are not prima facie the same, but because they are composed of such different basic conceptual ingredients that there are simply no common terms that could possibly perform the actual substantive function of actually bridging them. And it is clear on looking at them that no supporter of any so–called “identity theory” has ever actually attempted to pull off the required task. “Identity theories” do not in practice amount to surprising discoveries overturning ordinary intuition, but rather to basic conceptual confusions that come nowhere close to actually doing what they claim to do.”
Again, I think we can see that subjective conscious experience cannot be produced by any combination of blind inert causal interactions, in and of themselves, without something extra (whether that’s some extra mysterious properties possessed by the objects involved in these blind, inert causal interactions themselves, à la panpsychism; an additional set of types of laws, à la Chalmers’ proposal; or, as I argue, quite simply the fundamental phenomena of consciousness itself—a phenomena which does not “have” the properties of subjectivity, intentionality, etc., but rather quite simply “is” the un–quantifiable, qualitative phenomena of subjectivity and intentionality (etc.) extended across time—“I” am this temporally extended stream of subjective conscious experiences and intentionalistic thoughts, and this extended stream of experiences and thoughts is not “identical to” or “reducible to” anything other than itself. That, at least, is my position, and what you’re signing up to watch me defend in rebellion against the current zeitgeist if you should choose to follow me).
_______ ~.::[༒]::.~ _______
One partial problem with the way the argument is formulated is that it leads some readers to think that zombies might only be “possible” in the relevant sense if epiphenomenalism is true and the consciousness we experience plays no causal role in our behavior. Otherwise, so the reasoning goes, if we took consciousness out of the picture, then we wouldn’t have a behavioral duplicate of the human beings in our world—therefore, zombie duplicates of the people in our world aren’t really conceivable, and the argument from the conceivability of zombies fails.
I’ve argued that epiphenomenalism is not just implausible in the way that many think, but as decisively and conclusively refutable as anything could possibly be (see my essay (IV)). It is also quite obvious that I think the zombie argument is demonstrating something. What is going on here? Chalmers’ own most direct response to this peculiar point doesn’t help us much either: “…the possibility of zombies does not obviously entail epiphenomenalism. To see this, note that an interactionist dualist can accept the possibility of zombies, by accepting the possibility of physically identical worlds in which physical causal gaps (those filled in the actual world by mental processes) go unfilled, or are filled by something other than mental processes. The first possibility would have many unexplained physical events, but there is nothing metaphysically impossible about unexplained physical events.”
We … could insist on that premise, sure. But at least intuitively, it very obviously weakens the strength of the conceptual clarification we’re trying to make. And the key point is that I think we can get around it entirely, with a very simple tweak—and it turns out to be the same tweak that gets us around the other major set of unnecessarily technical, complicated objections that try to defeat some step of inference from the “conceivability” to the “metaphysical possibility” of zombies.
Such arguments may, for example, take some condition which we seem to be able to conceive—and then point out that, despite first appearances, it turns out that things we (supposedly) thought we could conceive of being true actually are not logically possible after all. Is 7,741 a prime number? What about 7,742? We don’t know, therefore—the argument goes—we can conceive of either being prime or not prime. But it turns out that 7,741 is prime, while 7,742 is the product of 72 x 98—therefore it is logically impossible for 7,742 to be prime, and logically impossible for 7,741 not to be. And therefore, conceivability is a useless guide to logical possibility.
Now, I could follow along with this argument and counter by denying that I actually can “conceive of” 7,742 being a prime number or of 7,741 not being a prime number in any meaningful sense at all. I could make still yet some other fresh new irritating distinction between “epistemic conceivability,” defined as the mere “ability to imagine that it might turn out to be the case that…,” and “robust conceivability” defined as “the capacity to actually hold two concepts clearly in mind and actually imagine them all the way through with or without each other”—and then say that I can robustly conceive of the notion of a philosophical zombie, whereas it is only “epistemically conceivable” for me before discovering the answer that 7,742 might be prime or that 7,741 might be composite—and then I can deny that what goes for epistemic conceivability in cases like these goes for robust conceivability as it applies in the zombie case.
But again: why even go there? Once again, I think we can get around all of these kinds of objections and wipe a great deal of technical obfuscation out of the argument in one fell swoop. Instead of arguing that “zombie worlds are conceivable” and then trying to justify a modal premise to allow us to go from “conceivability” to “metaphysical possibility” (or whatever), I can simply say the following: “If the premises of materialism were true (that the world is, at root, a blind process of inert forces—whatever the details of their structural composition or how they might causally operate—evolving through sheer causal mechanism; in other words, if whatever the “bedrock” ingredients of reality are, they intrinsically lack subjective experience or intentionality), it would follow that consciousness of the sort that we experience immediately could not have appeared out of such ingredients. In other words, the premises of materialism predict the non–existence of the sort of consciousness we experience. But since we do have the sort of consciousness we know first–hand that we experience, then, the premises of materialism are falsified by its existence.”
Obviously, these premises would need to be justified. I think they can be—in exactly the ways I have summarized above, and elaborated on in extensive detail across this series. But the places where this argument would need support are just exactly the same places that the ordinary formulation already does; and notice that, for example, there is no basis anywhere in this presentation of the argument for anyone to even try to charge that it could only be sound if epiphenomenalism were true. On this formulation, if it follows that taking consciousness away would change our behavior, all this would mean is that materialism would be falsified both by the existence of the kind of consciousness we experience, and from the behaviors which we are capable of only because of its existence. This puts the dialectic back where it actually belongs (where it is, as I have explained, materialism which struggles to avoid either eliminating the mental or rendering it epiphenomenal as such)—and simply has no need to invoke any abstract, unintuitive modal premises whatsoever. This formulation obviously isn’t without its own need for complicated defense, but I think that defense can be provided—and as I see it, it takes away nothing that Chalmers’ formulation of the argument accomplishes, but it does save us a whole hell of a lot of wasted time by efficiently avoiding more than one redundant and exasperating detour that the argument as it is most commonly formulated so frequently ends up tied up in. We could address arguments like the epiphenomenalist critique and critiques of the step from “conceivability” to “possibility” on their own terms, and perhaps they do fail either way—but if we don’t need to, why even muddle the debate by adding them in? This reformulation focuses more clearly on the premises which actually form the crux of the debate: whether we can get subjective conscious experience and intentionality out of inert causal mechanism; and if not, what exactly we would need to add to the picture in order to be able to get it.
Bonus footnote: Philip Goff, Why Physicalists Have More to Fear from Ghosts than Zombies, which argues that Descartes’ original arguments for the conceivability of disembodied consciousness (the supposedly emergent macro–phenomena without its supposedly micro–reductive base) actually does have certain virtues in some instances, against a certain kind of physicalist position, over Zombie arguments for the conceivability of unconscious bodies (the supposedly micro–reductive base without the supposedly emergent macro–phenomena).