That is entirely true, unfortunately. We could (in theory) link every thinking being in this imaginary future world together, and they would all experience the same qualia - but that does not prove that they would experience these sensations separately. If we linked an arbitrary number of entities together, then split them up again afterwards, we wouldn’t know if they continue to experience anything after the split, or if it was actually only one of the linked entities that ever experienced anything, and all the rest were briefly contaminated by the link-up to believe that they were experiencing sensations.
At the end of the day we could never be sure that anyone else is conscious, unless we were directly linked to them in order to create a new being. That does seem like a bit of an absurd position, but from a position of extreme skepticism it is true.
But this argument can also be applied to rocks. If we can’t disprove that rocks are conscious, despite showing no evidence of being conscious, then we should accept panpsychism as a plausible hypothesis. The p-zombie argument is an evidence-free way of arguing for dualism without the possibility of disproof. Can you think of a way to disprove the p-zombie experiment?
Sure, but that would just mean that it’s imaginable that a rock might not have any conscious experience. Which, well, it actually doesn’t.
The zombie argument doesn’t mean that you can just make up conscious experience, or the lack of it, willy-nilly. It means that since all of the physical causality going on in a human brain can be conceived of as occurring without any conscious experience, and since that activity fully serves to explain everything about the actions of a human, it is possible for a being physically identical to a human, yet without conscious experience, to exist. Consequently, the physical facts don’t fix the experiential facts.
Sure, I even published it:
Basically, the notions of predictability and possibility don’t exactly align; we can’t derive phenomenal facts from the physical theory, but physical facts fully fix phenomenal facts. (In some sense, this is analogous to the case in mathematics: the natural numbers have certain properties that can’t be derived from any given axiomatized theory of the natural numbers, due to the incompleteness theorem. But to the extent they exist, the natural numbers still have those properties, it’s just that theories are intrinsically bounded in their ability to predict them.)
You’re positing that physical properties alone won’t produce conscious experience, and then trying to use that to prove that physical properties alone can’t produce conscious experience.
The fact that you can imagine creatures with the same physical properties who aren’t conscious is meaningless. To use a classic example, I can imagine there’s a pink unicorn in my kitchen. No amount of imagining will allow me to prove anything about the existence of unicorns from that imagination.
I can imagine quite easily that a rock does have conscious experience. Again, my imagination doesn’t mean that it does.
How is that different from making up the lack of conscious experience willy-nilly?
You have to be careful to look at the augment details.
Chalmers is using the p-zombie premise to argue that physicalism - the idea everything is physical/material and that there is no spiritual/metaphysical existence - is false. He uses crooked logic to justify his claim. He is proposing human existence - consciousness - is not physical in nature.
The general premise doesn’t rely on Chalmers’ absolutist position. The p- zombie premise allows room for the internal, subjective experience to be different, but all outward behavior is identical.
Assuming such a case could exist, if we ever managed to run that mental hookup, then it could conceivably demonstrate if there is source experience. It would all depend on the mechanism of data exchange, but in the proposed case, the signal data would be blank coming from the p-zombie.
The problem isn’t that there isn’t a conceivable way to tell them apart, it’s that the conception relies on science fiction and conjecture. We can’t possibly begin to conceive of how to link up consciousnesses if we don’t know anything about the hardware or software that makes consciousness in the first place.
One response to the p-zombie argument is to argue they are impossible. If the starting premise is physicality, i.e. materialism, then whatever our experience is is an outcome of the nature of our existence.
It has been stated we can make models of human behavior that don’t have experience. However, these models are all simplistic to the point of being a game of Sims.
We are nowhere close to being able to make a complete model of the entirety of human behavior. We can’t even do a dog. Hell, it’s questionable if we could replicate an insect.
I would argue the very nature of life is a sensory processor and behavior modifier based upon that data and synthesis and evaluation of all accessible data. The nature of animal capability to respond to the environment in real time and change outcomes is itself the result of what experience, what qualia, actually are.
That doesn’t explain how they actually occur, but does provide a meaningful answer to why they arise. Qualia are the functioning of the sensory data in the complexity of the brain.
We don’t know yet if it can be tested or disproved, because we don’t have a good model of what consciousness is.
It would be like saying that the transporter problem is silly because we have no idea how we could prove either* position. OK but…there’s nothing in principle that would prevent us constructing some kind of transporter, and “that’s silly” doesn’t give us any guidance on whether it would be sensible to use such a device.
Likewise saying p-zombies are silly doesn’t answer the question of whether all animals that behave as though they are in pain, are actually in pain, and how that pain might compare to mine or yours.
* Actually I think there are 3 positions when it comes to the transporter:
Your consciousness is transported; it’s rational to use the device
You are killed and the person at the destination is not you. If you want to survive, it’s not rational to use the device
There is never any continuity of consciousness in ordinary life anyway. The “you” now is a wholly new entity that happens to have memories in common with the “you” of 10 seconds ago, and has the illusion of continuing as the same entity. It makes no difference at all whether this (fleeting) entity uses a transporter.
That’s not how the zombie argument works. It invites you to imagine simple causation: say, a billiard ball hitting another. No matter whether there is any conscious experience associated to that, we can surely agree that just the physical parameters of the impact determine the further course of the two balls—we need not take into account how either of the balls feels at that given moment. So we can imagine the same process taking place without there being any associated conscious experience, because it plays no role in determining the outcome, and isn’t necessitated by anything about the physics of the setup.
Now consider a great number of such billiard ball collisions. Still, even if the calculations become intractable, there is no reason to assume that the physical parameters of each of the individual collisions don’t determine the overall pattern in exactly the same way—none of the individual balls upon making contact ‘know’ they’re part of some larger pattern: it’s just the prior situation over and over again.
Such a system can be used to perform arbitrary computations. Anything a computer can do, can be done with a large enough assembly of interacting billiard balls. Thus, in particular, we could imagine using a titanic billiard ball brain to control a robot (say, remotely, so we don’t have problems of size; or we just imagine shrinking the balls down, ignoring issues of quantum mechanics for the moment). This robot could do anything you could imagine a robot doing, but as stipulated, we can imagine it doing so—it is logically consistent for it to do so—without any conscious experience: everything is ultimately fully determined by collisions of two billiard balls, which is fully determined by the energy and momentum they have at impact. No conscious experience needed anywhere.
Then, if any behavior of a human being—all the walking and all the talking—can in principle be replicated by a robot, we have what’s called a ‘behavioral zombie’: a being behaviorally, but not physically, identical to a human, without any conscious experience (or which at least can be conceived of as having no conscious experience). But there’s nothing special about the physics of billiard balls: for any physical system, it will be true that the outcome of each interaction is fully determined by the parameters of that interaction (and perhaps random chance, depending on your preferred interpretation of quantum mechanics—although even this can be eliminated by appealing to Bohmian mechanics, for instance).
But all a human being ever does ultimately boils down to such physical interactions. From any stimulus to any reaction, there is an unbroken chain of physical causality, where every link and its behavior is fully determined by the physical parameters of the interaction. There is no need to appeal to experience at any point. I should stress that this does not entail that there necessarily is no conscious experience: even the two billiard balls colliding may think ‘ow, that hurt’ upon collision; but if they do so, then this has no bearing on the collision’s outcome, and hence, it can be done without. It’s the same with the entire complicated sequence of interactions that makes up an entire human being: conscious experience may be associated with that buzzing and bustling activity, but it isn’t necessary for all that activity to occur in exactly the same way. Hence, we can conceive of that same exact activity without any sort of conscious experience at all: the philosophical zombie.
But now if all of that physical activity can consistently occur with or without conscious experience associated to it, then it follows that it can’t be that physical activity that determines whether there is any conscious experience: there must be something else, some further fact not reducible to the physical facts.
So no, the zombie argument isn’t circular. All that’s needed really is the thesis of the causal closure of the physical, and then the fact that we apparently can do without any conscious experience is derived from there. And that thesis of course has some very good support: in physics, we never find we have to appeal to non-physical properties to find the outcome of our calculations. Furthermore, rejecting this thesis already amounts to accepting the conclusion of the zombie argument: namely, that there must be something else beyond the physical facts.
See, I don’t believe you can, not anymore than you can imagine a square circle. You might be able to think of a square, and a circle, and both sort of at the same time, but not of something that really is both together. Likewise, you might be able to think of conscious experience, and of the stone, and of that experience sort of associated to the stone, but you can’t imagine what it would be like to be the stone having that conscious experience, simply because you aren’t the stone. You are you imagining a rock as having conscious experience, but the only conscious experience in that act is that of you imagining, not of a rock having any experience at all.
The zombie argument is worse than it seems. I imagine a scenario where a person is a p-zombie only on Tuesdays, Thursdays and Saturdays; their memories from those days would be physically identical in quality to those on Mondays, Wednesdays, Fridays and Sundays, so they would not be able to tell the difference.
Indeed a person might be switching between a non-zombie and a p-zombie state every second, like a set of lights on a Christmas tree; the person’s experience in such a situation would be indistinguishable from one second to the next. A difference that makes no difference is not a difference.
All you’re saying is that you don’t like the conclusion of the argument. Which, fine. But that’s not the same as having grounds to dispel the argument: for that, you’d have to show where the argument goes wrong, or else, accept that there must be something that’s not reducible to physical facts that determines conscious experience.
I’m saying that the argument is incoherent. A person that flashes between a zombie and non-zombie state is absurd, yet that is entirely consistent with the premise.
That’s not an incoherence, that’s just your judgment. You could just as well have a person flash into existence with all their memories intact, as in a Boltzmann brain, or you could have them do so either second—that may be counterintuitive, but that just says that your intuition might not be the most reliable guide to metaphysical possibilities. Which is, of course, why we formulate detailed chains of reasoning such as the zombie argument to probe those intuitions.
That is the point; consciousness is a technical issue, and we won’t solve it just by using philosophy; we can only investigate it using advanced neurology and technology, otherwise we will never know what the details are.
Here is the connectome of a fly; it has 1~30,000 neurons, and ~50 million synapses.
Just mapping such a thing is not sufficient - we need to replicate it in action. That is a task that will take much longer in practice, and yet would be child’s play compared to replicating a human connectome.
This is not a problem for philosophers, but for engineers. But I anticipate we will get there in the end.
Mind you, I didn’t say anything about the metaphysical possibility of p-zombies; they are possible on a metaphysical level, but so is almost anything else, no matter how absurd.
I think you’re using the term ‘metaphysical’ in an unfamiliar way. ‘Possible on a metaphysical level’ just means ‘possible’, because metaphysics is just the study of what is possible, what exists (ontology), and how we can know about that (epistemology). The reason it’s called ‘metaphysics’ isn’t because there’s anything extraphysical associated with it, or any other spooky stuff, it’s just that Aristotle’s books on the subjects were studied in the classical curriculum after the one on physics—‘metaphysics’ just means ‘after the Physics’, meaning Aristotle’s book.
So if zombies are metaphysically possible, they’re just possible, and the metaphysical position that everything boils down to physics (physicalism) would be false.
If something is metaphysically possible but not physically possible, that puts it in the realm of the imaginary, to me. But plenty of useful things are imaginary, especially in thought experiments and theology. I love inventing speculative religions involving metaphysical concepts with no reflection in the real; this conversation is grist to the mill. But it is always better to withhold actual belief in the non-physical side of things, since they are not compatible with proof.
Incidentally I’m reading your paper with interest - I feel another religion coming on.
Aristotle was a great man, but his philosophy of science was very basic; rather than doing experiments, he’d just consider the possibilities and choose one that he thought fitted the evidence best. This works sometimes, but not always.
You’re all over the place here. The zombie thought experiment argues for a metaphysical possibility, which is that the physical facts don’t entail the facts about conscious experience—thus, to get a world that includes conscious experience, we have to admit something else than physical properties—perhaps irreducible mental properties, as in panpsychism or property dualism, perhaps properties that are neither physical nor experiential, but which can give rise to either, as in neutral monism, or something of that sort. Without a counter to the zombie argument, just pushing anything non-physical into the realm of the imaginary is merely sticking your fingers in your ears and going ‘lalala, I can’t hear you’ as a response.
Even if you think you have good a priori grounds for only admitting physical properties into your metaphysics, that doesn’t license you to simply ignore the zombie argument (and others like it)—sure, that may be where the fault lies, but you could also just be wrong about your prior commitment.
Just to reinforce what @Half_Man_Half_Wit is saying, but in simpler words (because I don’t have the intellect to do otherwise).
We don’t know whether it is physically possible to make a p-zombie, with complex responses to external stimuli but no inner experience. We would like to know, and not just for philosophical interest, but for many practical reasons.
e.g. We’d like to know which animals suffer physical pain and to what extent. We’d like to understand human pain better. As we hurtle towards AGI, we would like to know the point at which it is having an inner experience and perhaps is capable of suffering.
Saying, essentially, “well, p-zombies seem silly to me” tells us precisely nothing.
