5

Related Posts that I read:

Context

Just to make sure we talk about the same type of Zombie, here is the SEP definition:

Zombies in philosophy are imaginary creatures designed to illuminate problems about consciousness and its relation to the physical world. Unlike the ones in films or witchcraft, they are exactly like us in all physical respects but without conscious experiences: by definition there is ‘nothing it is like’ to be a zombie. Yet zombies behave just like us, and some even spend a lot of time discussing consciousness.

From my interpretation of the descriptions of individuals with specific types of brain damage (see the example in the next paragraph), it seems that in some cases part of the movie of consciousness could be missing.

In the case of Hemispatial Neglect (but also many others), and related paper there may not be conscious visual experience but the remaining processing may occur, completely or partially.

Claim

A subject may not have some conscious experiences, but in all cases this is correlated* to spongy brain modules and interconnections to other modules. Missing modules can even correlate specifically to the visual experience / qualia.

If experience is dependent upon these visual modules in the brain, how could a zombie exist ?

There may still be zombies unlike us: individuals that may not have a subjective experience, at least with respect to some specific modality.

At the same time, this would contradict (to my understanding at least) Delusionism.

Question

  • Is then the concept of a Zombie a useful thought experiment, or actually misleading ? Why or why not ?

*: there is an explanatory gap for a materialistic explanation

Minsky
  • 249
  • 11
  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on [meta], or in [chat]. Comments continuing discussion may be removed. – Philip Klöcking Jan 16 '24 at 19:20

6 Answers6

4

Chalmer's definition of a P-zombie has a hidden feature that is not often discussed explicitly:

  • If matter is causal on consciousness, OR consciousness is causal on matter, then a human who does the functions of mentality that consciousness does but is not itself conscious, CANNOT be physically identical to us regular humans.
  • For causal consciousness to be a reality, our bodies must be configured for consciousness to act on it.
  • For bodies to be causal on consciousness, then there has to be psycho-physical laws (emergence, or whatever) which bodies make use of
  • Chalmer's definition forces him into an argument for epiphenomenal dualism.
  • It is not just reductive physicalists who must object. Any interactive dualist, or emergent non-reductive physicalist, cannot accept that Chalmers' P-zombies exist either.

Chalmers' conception of P-Zombies -- that one cannot tell the difference, also makes them pretty useless for experimentation.

However, this does not make zombies useless. One just needs to start asking different questions, and be open to variants on what a zombie may be.

The first different question one should ask, is whether it is plausible that, if a zombie exists, that it would describe its qualia when asked about them? Chalmers says they will, by definition. But then we can ask: if zombies existed, why would they develop qualia language? The answer, of course, is they would not. We can extend this question to Chalmers' epiphenomenal dualism. And the same answer is arrived at. If there is no causal effect from our consciousness, then our having epiphenomenal conscious experiences would not appear in our language. We would not answer "yes" to experience/qualia questions. These conclusions are based on applying evolutionary principles to our language use.

These points lead to realization that we CAN test for the causal nature of consciousness, using evolution as the test bed. Evolution's natural selection can only apply to causal phenomenon. And if our consciousness, the structure of our consciousness, or our language about our consciousness, is at all tuned by evolution, then consciousness is causal, and Chalmers' version of P-zombies cannot exist in this world.

We can then also extend evolutionary tests to other mind-body speculations, and also potentially refute some of them. William James first applied evolutionary test cases to epiphenomenalism, showing it is falsified by them. Karl Poper extended these evaluations to all Identity Theories, showing them to be falsified as well.

A key part of these arguments is a different form of Zombie. A human very much like us, but with sufficiently differing neurology, that it can do mental functions without consciousness. Interestingly, WE are such zombies, for we are unconscious of 99.9% of our mental processing. And we do many functions sometimes consciously, and sometimes unconsciously.

What this allows us to speculate on, is a zombie who is neurologically slightly different, such that they are a zombie for 100% of their mental functioning, rather than 99.9%. This is not a Chalmers' zombie, but just a functional zombie. Could such a zombie exist? Neither reductive physicalists, emergent physicalists, emergent dualists, nor spiritual dualists need to deny it could. This subtlety different zombie forces a different question:

  • If brains could have gotten by without consciousness, why are we conscious?

THIS is the real "hard problem of consciousness".

Dualism has an answer -- evolution happened upon using consciousness for "System 2 functions" (under Kahneman's 2-systems model from Thinking Fast and Slow). Whether this was through an accidental emergence event, or an ensoulment event, causal consciousness was useful, and was tuned by evolution.

Physicalism cannot provide an answer to this question, as physicalism is committed to consciousness not being causal.

Chalmers' epiphenomenal dualism, and Goff's epiphenomenal pan-psychism, are likewise unable to answer the hard problem, as they too are both committed to consciousness not being causal.

Dcleve
  • 13,610
  • 1
  • 14
  • 54
  • That first bullet point is well stated and centrally important, +1. – TKoL Jan 15 '24 at 09:16
  • @TKoL -- though that point is technically true, it doesn't cover the case when the counterfeit does a decent job at imitating the original. ChatGPT is one example to the point. And this imitation does not have to be artificial either. Some suggest that most of our behaviour is automatic/habitual/unconscious -- which means that all humans possess a separate cognitive faculty that can act as an autopilot, and whose sole purpose is to imitate consciousness. – Yuri Zavorotny Jan 16 '24 at 06:07
  • 2
    @YuriZavorotny "and whose sole purpose is to imitate consciousness" I don't know that this is accurate. Just because it happens to do an okay job at briefly, from a distance and without much probing, convincing people it's behaving in a conscious-like behavior doesn't mean that's what it's for. – TKoL Jan 16 '24 at 09:32
  • @TKoL — don’t you think that the purpose of an autopilot is to imitate the pilot? Of course the autopilot’s functionality is relatively limited, but that’s the nature of any imitation — it is never as good as the original. – Yuri Zavorotny Jan 16 '24 at 14:43
  • 1
    @YuriZavorotny no, when the conversation is about human consciousness I don't believe the purpose of the thing you're referring to as 'autopilot' is to imitate consciousness. Just because you've made an analogy to autopilot doesn't mean the analogy holds at every level - don't get lost in your own metaphor. – TKoL Jan 16 '24 at 14:53
  • "Karl Poper extended these evaluations to all Identity Theories, showing them to be falsified as well." Would you have a reference for that? Thanks. – Olivier5 Jan 17 '24 at 12:59
  • @Olivier5 this was in The Self and It’s Brain. I summarized the reasoning above. – Dcleve Jan 17 '24 at 14:28
  • @Dcleve Thanks. – Olivier5 Jan 18 '24 at 14:14
1

IMO the lessons learned from the Zombie idea are: It is not enough to restrict the discussion about consciousness on the physical components of the organism. One has to consider also the aspect of information processing in the organism. Information processing is not a static structure but a dynamic process.

The investigation of the processing of stimuli in the visual field V1 and subsequent cortical fields, i.e., the corresponding cortical wiring diagram, is a main task in the study of the neural correlates of consciousness (NCC).

I do not know whether the concept of a Zombie as being “exactly like us in all physical respects but without conscious experiences” is helpful to stimulate further research about NCC.

Possibly the concept is already too coarse for the present level of research in neuroscience.

Jo Wehler
  • 30,912
  • 3
  • 29
  • 94
  • I'd say it's a structure that has a function -- in other words, it's a machine (even if some parts of it are virtual, a. k. a. software). – Yuri Zavorotny Jan 16 '24 at 01:23
  • @YuriZavorotny I now changed the wording at this point to clarify what I want to emphasize. – Jo Wehler Jan 16 '24 at 03:53
  • If consciousness is actually information processing, why would it feel like something? Are you proposing that the 'fundamental' description is not matter/energy but information ? (which, stated as I did, doesn't mean anything to me since you need substrates that interact to 'have' information.) – Minsky Jan 17 '24 at 11:03
  • 1
    @Minsky Please note, that to understand consciousness we need to consider both(!), the wiring of the physical components and their information processing. I did not play off one against the other. – Jo Wehler Jan 17 '24 at 14:34
  • @Minsky A focus on information and its processing merely picks out the organization of that substrate as being the issue that matters (pun sort-of intended), and abstracts the issue away from any particular substrate. – sdenham Mar 25 '24 at 20:01
1

The zombie argument is used to demonstrate on modal grounds that the identity theory of mind, claiming that mind-related states (events) are identical with brain-related states (events), only under a different description (two popular variants are anomalous monism and functionalism), is incorrect. It's similar in certain respects to Kripke's modal argument (from Identity and Necessity) and Frank Jackson's knowledge argument (the famous "Mary's Room" thought experiment).

Because it is intelligible that there's a creature, p-zombie, which is physically identical to a conscious individual but who doesn't have a mind, then in the world, in which such a creature exists, minds are distinct from bodies. It follows from that and the necessity of identity, a theorem of modal logic, that in every world mind and body are distinct.

There are, of course, answers to the argument. Many people consider the very notion of a mind-body dualism simply entirely unproductive on scientific grounds. Paul Churchland (in Philosophy of Mind meets Logical Theory) has, for example, argued that dualist eliminativism, which asserts that though mind and body are distinct, minds aren't actual (they are merely thinkable, but we won't find them in our world), is intelligible - and therefore considerations about philosophical zombies are vacuous. This would be equivalent to saying that everyone might be a p-zombie.

Richard Rorty has argued (ex. in Philosophy and the Mirror of Nature but also elsewhere) that nothing interesting can be said about the entities and properties whose distinctness from merely physical stuff the p-zombie argument is seeking to establish.

Others, like Willard Quine, believe that modal arguments for mind-body dualism are based on abusing modal logic and/or our commonsense intuitions (see his response to Kripke in Theories and Things).

David Lewis famously objected to Kripke's argument by challenging that terms referring to mental entities such as pains etc. are rigid designators. A p-zombie equivalent to this objection would be saying that p-zombies aren't actually intelligible.

  • We seem to be thinking in different terms, probably from the authors we have read. The p-zombie may be useful to show that identity theory is incorrect, but a zombie can not be proposed as being the same than a human, and not having internal experience (the links I included are a proof of exactly this.); this is, assuming from the start that we do have it. – Minsky Jan 14 '24 at 15:12
  • @Minsky If so, then p-zombies in the strict sense are impossible. A p-zombie is supposed to be truly indistinguishable from a conscious being. That's exactly the problem and that's why I find the idea very sketchy. –  Jan 15 '24 at 17:43
  • I see. So we agree, I didn't read that in your answer; actually, I don't understand the context of that sentence art the rest of the text. – Minsky Jan 15 '24 at 22:06
  • @abcga "claiming that the mind and the body are identical" - who made such a claim? Could you please give a reference. - Not even to adherents of monism I would ascribe such a position. – Jo Wehler Jan 15 '24 at 23:23
  • @JoWehler That's quite literally what monism is. By "identity" here I mean token identity, not type identity... mind is the same as body but under a different description (functionalism, anomalous monism). And by "body" I don't mean one's body as a whole but neurological states or events. May that have confused you? –  Jan 16 '24 at 14:09
  • 1
    @Minsky You asked about disproving the p-zombie argument. I wanted to show that it's, strictly speaking, impossible through empirical study unless we modify the original form of the argument. But I also wanted to give an understanding of how p-zombie argument could be refuted. I hope that clarifies my response. –  Jan 16 '24 at 14:14
  • 1
    @abcga i'll ponder your answer a few more times then, thanks for your reply – Minsky Jan 16 '24 at 15:04
  • @abcga To express the claim of “the token-identity of mental and physical events” as claiming “that the mind and the body are identical” seems a rather wide shortening. Is this really helpful, do other readers understand what is rejected in the first section of your answer? – Jo Wehler Jan 16 '24 at 15:05
  • @JoWehler I will edit my answer to make it more understandable. Thanks for your remarks. –  Jan 16 '24 at 15:06
1

To paraphrase Samuel Beckett, we are all born zombies. Some remains so.

The way I see it, this is the concept otherwise known as tabula rasa -- meaning we are born blank slate with no knowledge and, therefore, no consciousness. It is, then, up to the individual (and the supportive environment) to gradually develop their understanding of themselves, the world, and their place in it. To stitch together a map of the Reality and to use it to act consciously, seeing where they are, where they want to be, and how to get there. Or not -- and stay on autopilot, sleepwalking through their life, never waking up from "dogmatic slumber" in Kant's words.1

So this is the choice. Unfortunately, again, we are not born with the awareness of it. Perhaps even more unfortunate is that no one2 tells us that attaining this knowledge, this mental clarity is not only possible, but is something that every person should strive for. Unaware of this choice and this possibility, most of us never go for it -- we "remain so".

1 In reality, this a spectrum -- how much of the "map" an individual manages to complete, how mindful they are of their actions, etc.

2 No one, of course, except Heraclitus, Buddha, Socrates, Jesus, Spinoza, Nietzsche, and so many others -- but not our our parents, or older siblings, and not before we decide that we already know everything and slip into that "dogmatic slumber".

P.S. This comics by Daniel Dennett offers a model of a "zombie" mind (and, incidentally, of how ChatGPT and other LLMs work): enter image description here

Yuri Zavorotny
  • 1,701
  • 4
  • 12
  • https://www.youtube.com/watch?v=uIAdmBJ22GQ – Minsky Jan 15 '24 at 19:00
  • 1
    Thank you for the video. I like how he paraphrased Descartes -- "I'm conscious, therefore I exist." Though less specific, I bet it resonates with more people that the original "I think..." – Yuri Zavorotny Jan 15 '24 at 21:14
  • You are welcome. Yes, I agree. I just don't understand IIT yet. Could be good if you explain the comic. I'm not familiar with many things. – Minsky Jan 15 '24 at 22:09
  • 1
    It seems that you are saying that anyone who is not self-realized / nondual is asleep / a zombie. Which is an old idea, yes. I once heard someone say, "Most people are firewood." +1 for dogmatic slumber! – Scott Rowe Jan 15 '24 at 22:16
  • @Minsky -- sure, I can try! The "Chinese Room" is a thought experiment proposed by John Searle back in 1980. It shows how the present-day ChatGPT works: it sifted through human-written texts over Internet to compile a "book of rules" suggesting how a human would reply to different questions. Daniel Dennett further proposed that we, humans, have a similar faculty working in our subconsciousness, learning the rules that let us reply w/o understanding. Indeed, as Stephen Covey put it, "Most people do not listen with the intent to understand; they listen with the intent to reply." – Yuri Zavorotny Jan 15 '24 at 23:54
  • 3
    @ScottRowe -- yes, zombie is a metaphor, a parable that we thought up to describe this particular side of human nature, our capacity for automatic/unconscious behaviour. There is huge swath of poetry and fiction dedicated to the subject of us over-relying on this capacity, us "dying without truly living", etc. Or think Plato's cave. Or White Noise, a 2022 movie. And again, it's not that we're lazy or incapable of doing better. But this is a complex issue, and we've been peeling layers trying to get to the bottom of it -- but we're not there yet. – Yuri Zavorotny Jan 16 '24 at 00:57
  • I see. This is yet another definition of a zombie. But if we include 'identical' but with X property missing, it seems to me a falsifiable claim. (See SEP Def in OP) Of course, one can create other definitions. – Minsky Jan 16 '24 at 08:31
0

Answer

It depends on underlying assumptions:

  1. Assume that consciousness is intrinsic

Then a materialist view would consider, as OP indicates, that either they are both conscious or both aren't.

  1. Assume consciousness is an addition to the physics of the system

Then this remarks the extra option that one could be a zombie even though the physics are the same.

So the Zombie concept is actually a way to detect the pre-conceptions of the philosopher using it.

Thought Experiment

For fun, we could also ponder a redefinition of Zombie that leads to some interesting questions.

Imagine a Zombie that is almost like us but it lacks functionality that would – hypothetically – give rise to experience.

If Zombies operate without the brain structures (and connectivity) needed for consciousness to happen, then:

  • Why do we have subjective experiences? (we could still operate in the world without these.)

In this way, the Zombie hypothesis seem to be illustrating a question. The question, as said before, now in other terms, is:

  • Why do we have structures that give rise to consciousness when we don't need them?
  • But then, Don't they have a function ?
  • Are they just a cool side-effect of evolved and selected structure ?
Minsky
  • 249
  • 11
  • 1
    This is an interesting way to use this site! The assumption would be that they have a function, even if we don't know what it is yet. But that assumption could be false, in which case it might be an accidental by-product of something else, or a survival of some function that we have dispensed with (i.e. epiphenomenalism). – Ludwig V Jan 14 '24 at 18:35
  • @LudwigV -- accidental (non-causal) byproducts do not survive evolutionary variance unless they are logically entailed. Consciousness is not logically entailed by mental function, and this is readily demonstrated by how little consciousness we exhibit. The "spandrel" out against the evolutionary test case is not available for consciousness. – Dcleve Jan 15 '24 at 06:48
  • 1
    "accidental (non-causal) byproducts do not survive evolutionary variance unless they are logically entailed." I'm not familiar with that law. But even if it is true, evolution will take time to eliminate them. In the mean time, they survive. But whatever the truth, the assumption that everything has an evolved function is not safe. You yourself seem happy to consider that possibility. – Ludwig V Jan 15 '24 at 08:40
  • 1
    Your supposition is that we may find that some people seem to lack certain brain structures, but nonetheless function pretty much like us. How would you independently prove that they also lack subjective experience, unless some of their behaviour is different from us? And what differences would specifically prove that they lack subjective experience? – Ludwig V Jan 15 '24 at 08:44
  • @Dcleve I think consciousness could be a by-product of the needed interconnection between brain modules. – Minsky Jan 15 '24 at 10:35
  • @LudwigV I agree with your last comment. If they lack subjective experience, it will be correlated with some malfunctioning of the brain i.e zombie can not be identical to the 'original'. – Minsky Jan 15 '24 at 10:38
  • @Minsky We regular humans have the modules, with the interconnects, and use them for multiple functions that are not conscious. The recursive neural nets we use, and their interconnects, are all multipurpose. The problem with the NCC project is that it has basically only established NCUc not NCC. – Dcleve Jan 15 '24 at 15:29
  • 1
    That said, your revision of the zombie thought problem IS fruitful, and is very close to the one I offered. There are somewhat different questions we can ask with somewhat different zombie hypotheses. I believe these questions show that consciousness is CAUSAL and physicalists dogma that it is not — is false. Most physicalists today are emergent physicalists, and there is a very small gap between emergent physicalism and Popperian style emergent dualism, which I expect that more and more of the emergentists to cross. – Dcleve Jan 15 '24 at 15:48
  • I've read -half way- your second review for Musolino's. I wonder if there is a misinterpretation (didn't read this book, but may guess the point.) I'm just trying to understand what you think. Anecdotal data are indicated by these scientists as useless not because they are first person necessarily but because they depend too much on the observer. For example wanted we to know objectively the shape of an object under occlusion only by asking questions to make a theory of the real object, that could probably lead to bad theories. – Minsky Jan 15 '24 at 15:57
  • " The recursive neural nets we use, and their interconnects, are all multipurpose " Fair point. For example, we could learn to hear with the area normally considered for vision, had we grow up with that architecture (read as: neurons and connections). But I don't know how deep that goes. Some areas could be "fixed" though. I do not know what you mean by NCU, or by " I believe these questions show that consciousness is CAUSAL and physicalists dogma that it is not — is false. " I think one can very reliably defend that consciousness is causal in physical terms. (Still needs the brain & net.) – Minsky Jan 15 '24 at 16:03
  • @LudwigV -- Our exploration of the details of evolution have revealed some exceptions to 100% evolutionarily selected functionality. These include accidental phenomena that have not yet been weeded out, coupled phenomena that are actually detrimental but happen to be died to even more advantageous properties, and genetic drift. None of these apply to consciousness, which is a complex feature, with submodules, well integrated to function together, but with a peculiar/inefficient architecture. These features are characteristic of an evolutionarily tuned feature, not any of the alternatives. – Dcleve Jan 15 '24 at 17:00
  • @LudwigV -- Here are some expansions of the "spandrel" option for consciousness. https://en.wikipedia.org/wiki/Spandrel_(biology) https://www.academia.edu/2795753/Is_Consciousness_a_Spandrel Note that spandrels ARE logical necessary side effects of another structure, and will remain no matter the variations that natural selection generates. Paul Churchland basically asserts a "spandrel" identity theory between consciousness and recursive neural nets in this book: https://www.amazon.com/gp/customer-reviews/R5048CH7VMV78?ref=pf_ov_at_pdctrvw_srp My review shows how variance refutes this. – Dcleve Jan 15 '24 at 17:32
  • @DCleve From your Wikipedia link "Gould and Lewontin proposed an alternative hypothesis: that due to adaptation and natural selection, byproducts are also formed. Because these byproducts of adaptations that had no real relative advantage to survival, they were termed spandrels." I don't have a problem with that. I didn't mean to assert that consciousness (whatever we decide that to be) is not functional in evolutionary terms - just to point out that it's not necessarily so. Spandrels are an interesting half-way house. – Ludwig V Jan 15 '24 at 17:58
  • @Minsky "zombie can not be identical to the 'original'." Quite so. That makes all the difference. – Ludwig V Jan 15 '24 at 18:04
  • @LudwigV -- yes spandrels are interesting. They are among the "logical necessity" alternatives that I was referring to. They appear to exist, in some cases, in our collection of observed evolutionary traits. However, they are a rare phenomenon, and they don't have the characteristics that consciousness exhibits, so the argument that they are the mechanism by which consciousness is not beneficial is -- not valid. – Dcleve Jan 15 '24 at 21:55
  • @DCleve Well, I'm not asserting that spandrels are the answer, partly because I don't have a definitive list of the characteristics of consciousness. So I won't argue with you. – Ludwig V Jan 16 '24 at 05:04
  • @LudwigV -- Do you need a "definitive list" to know that it is a complex and surprising architecture, with integrated and tuned modules? Evo-Devo has established this to be the case, as has Thinking Fast and Slow. – Dcleve Jan 16 '24 at 15:44
  • There are three lists in your review. They barely overlap. Unlike you, I don't know enough to decide between them. It's a vast and complex literature which I'm not familiar with. You are always worth reading, but I don't know enough to engage with you yet. Sorry. I do think that the phenomena we group under the label "consciousness" are probably evolved, but don't know enough to be certain. – Ludwig V Jan 16 '24 at 20:01
0

Zombies are a thought experiment. Thought experiments need a basis in reality. So, no square circles...or zombies, if you think it through. They are just like us but do not actually feel pain when they scream with pain. What are they screaming for then? The thought experiment is invalid. It also conflicts with the reasonable inference of other minds - leading to solipsism. Zombies are not useful.

Meanach
  • 2,337
  • 7
  • 36