by Tim Sommers
Or, rather, the very real threat to Physicalism posed by Philosophical Zombies.
On the one hand, there are the Night of the Living Dead, Dawn of the Dead, Shaun of the Dead, Walking Dead, zombies, which are rotting, but animated, corpses that devour human flesh and can only be stopped by destroying their brains. They pose a physical threat.
On the other hand, there are philosophical zombies that pose a much deeper, more unsettling threat to physicalism itself. Philosophical zombies look and act just as we do, but they have no internal, phenomenal mental life. Like celebrity influencers.
Physicalism is, roughly, the view that everything real is either physical or reducible to something that is physical. Arguably, it’s the philosophical view that undergirds modern science. The physicalist slogan that I grew up with was “all concrete particulars are physical.”
That is, there are abstract objects (like numbers or sets) as well as generalizations about concrete objects (say, about horses having four legs). And these are in some attenuated sense real. But any given example of a nonabstract object will be physical. When it comes to the human mind, physicalism does not have to mean that “redness” or “pain” is some particular unitary brain state, but rather that every time you perceive something as red your perception is caused, and constituted, by you being in some particular brain state. That’s called token physicalism or nonreductive materialism.
There are plenty of objections to this view. In a recent 3 Quarks Daily article Katalin Balog discussed the problem of causality. Since most philosophers accept the “causal closure of the physical” – every physical event has a physical cause, so nothing nonphysical can cause something in the physical world – mental events like pain or redness are, at best, epiphenomenal*. Even if they exist, in other words, they are caused by physical events or processes, but they can’t cause anything. Yet, the causal objection goes, isn’t it the fact that I see a light exemplifying the phenomenal property of redness (as philosophers used to say) that causes me to stop?
I want to discuss a different argument, however. The one about zombies, which Balog also mentions in passing. Rather, than being causal this is an objection to physicalism from phenomenal consciousness. Phenomenal consciousness is the subjective, qualitative, “what it feels like” experience of consciousness. “Pain” and “redness” are paradigmatic examples. Nagel famously fleshes it out like this. There must be something that it is like to be a bat, specifically, something that it feels like to be a bat. But no amount of scientific study of physiology or bat behavior will ever allow you to feel what it would be like to have echolocation. Frank Jackson famously uses the example of someone never allowed to see the color red. When they finally see red, do they know something they didn’t know before? They know, phenomenally, the anti-physicalist says, what red looks like. Here’s how the zombie argument uses phenomenal consciousness against physicalism.
A philosophic zombie is physically identical to a normal human in every way – body, brain, and behavior – but they lack phenomenal consciousness. The lights are on, but no one is home. There’s nothing that if feels like to be them. It helps me to picture a normal person who becomes a zombie. They can still pick out a red apple or tell when their pain receptors are activating, but they don’t see or feel anything anymore. The change is undetectable. And yet they are very different. We don’t know if there are any philosophical zombies. But many have argued that the mere fact that they are conceivable and not metaphysically impossible refutes physicalism. Here’s the argument. **
(1) If physicalism is true, then phenomenal consciousness is physical.
(2) It is conceivable that there could be a being physically identical to us, but without having phenomenal consciousness.
(3) If it’s conceivable, then it’s metaphysically possible.
(4) Therefore, consciousness is not identical to physical properties.
(5) So, physicalism is false.
Obviously, all the action is at (2) and (3).
So, (2) first. Are philosophical zombies conceivable? Chalmers has said “‘it certainly seems that a coherent situation is described; I can discern no contradiction in the description.” He is not saying zombies are possible, or physically possible, at this stage, just that they are conceivable. You can coherently imagine one.
Some, including Dennett, have argued that you only think you can conceive of philosophical zombies. You might also think that you can conceive, in some sense, of a squared circle, or more easily, a perpetual motion machine. But you think that either because you overestimate your power to conceive, or because you don’t fully understand the thing you imagine you are conceiving, or both. Not everything imaginable under some description or another is actually conceivable in the relevant sense.
So, (3) let’s say philosophical zombies are conceivable, why does this make them metaphysically possible? Metaphysical possibilities refer to what could exist or occur without any logical contradiction, regardless of the actual laws of nature. It means that a scenario is coherent in principle when we imagine all possible ways the world might have been and it is that way in one of them. For instance, a world where zombies exist (beings physically identical to us but lacking consciousness) is metaphysically possible if it doesn’t involve any logical inconsistency.
This is a huge, complicated debate which I cannot hope to do justice to. I’ll just point to one line of criticism. Conceivability is an epistemic notion. It’s about what we know or what we can’t know. Possibility is a metaphysical notion. There’s no reason to think that whether or not philosophical zombies are possible can be settled a priori, as philosophers like to say, that is through prior reasoning ahead of a factual investigation (a posteriori). Indeed, this daredevil leap from conceivability to possibility may simply be mistaken.
On the other hand, we have an example now that only came into existence recently that seems disturbingly apt. Don’t large language models often behave (verbally, at least) in a way that is indistinguishable from humans? And yet we don’t think they have phenomenal consciousness. Do we? In fact, take your list of ways in which you think LLMs don’t act like humans and ask yourself this. Is it conceivable that in the future there will be AIs that fit all the slots on your list? Will they necessarily then have phenomenal consciousness? Or not? Maybe, anything that can exactly imitate the full range of human capabilities just will have phenomenal consciousness. Honestly, I for one, was more confident of that before I met ChatGPT.
________________________________________________________________________
*To be clear, Balog’s aim, as I understand it, is not to defend the causal argument, but to question the significance of the whole mind-body debate. I’ll just mention one relevant disagreement I have with her piece. She writes, “There is a strange dialectic to these views. Both materialism and dualism can defend themselves from attacks from the other position and view the other position as question-begging. Neither seems to offer an overall better theory, and they don’t have empirical consequences that could distinguish between them. It is impossible to know, either on empirical or philosophical grounds, which is true, but strangely, the stakes are very low.” I am not sure I agree with all of that, but evens so I feel that, for better or worse, those characteristics are what make the mind-body a typical, rather than different, from most philosophical problem. Balog ws kind enough to respond to a comment I made in the comment thread. She denied that the “philosophy of mind” in general has low-stakes, “Talking strictly about the mind-body problem!” I don’t know what that separation between the philosophy of mind and the mind-body looks like, though. That one takes the study of the brain very seriously, but not the question of whether everything going on in the brain constitutes consciousness? Personally, I suspect that the impression that the debate has low-stakes comes from the recognition that, despite questions remaining, for most people physicalism won the argument long ago.
** David Chalmers (The Conscious Mind: In Search of a Fundamental Theory, 1996) usually gets the bulk of the credit for this argument, but, as important as his contribution has been, variations of the argument are all over the place in philosophy. Saul Kripke, for example, back 1972 in Naming and Necessity suggested a global version. When God made the whole physical world, was God done? Or did God have to do more work to bring consciousness into existence?