Thanks Ben for reviewing and sharing the post (and to Michael for your resources) I did find it very interesting. But after reading it, giving it some thought, and reading the Aeon essay you recommended I came away quite intellectually frustrated. I think whenever some claims to have answered the hard problem, the more likely explanation is that they havenât understood it.
(Full disclosure for me confusion, Iâve all but become a strong dualist about consciousness after many many years of being or identitfying as a physicalist. This was a result of reading the philosophical literature, but also a lot of what was a personal reflection on my own experiences through meditation practice, especially informed by Douglas Hardingâs Headless Way. I basically canât grok illusionist perspectives any more try as I might[1], so readers might better read this comment as coming from a dualist loyal opposition, rather than a balanced assessment)
Something I found particularly puzzling was in the 3rd section of the essay (beginning âOver the past 50 years...â), Humphrey seems to confuse mind-brain identity theory with panyschism, which doesnât for me bode well my expectation of his âsolvingâ of the hard problem. M-B-I to me is a strongly physicalist (if not eliminativist) position, while pansychism is by necessity dualist. [2]
The D.B. case is also an interesting oneâI donât see why it isnât plausible to imagine that the operation (and similarly split brain cases as documented by Nagel) might lead to a second vestige of consciousness as cut off from you as that of your family, friends, and coworkers. Except this fragment doesnât have control of motor or speach functions, how horrible! It can only pass information onto the âdominantâ one.
The case of whether Humphrey is right about Octopi not playing is outside of my domain of expertise but initially I am sceptical. But again just because we know that we (humans) are conscious in a certain way, why ought we to imagine that consciousness must only exist in this way? Iâm not sure it follows, and there seem to be gaps like this in his arguments that actually cover up important parts of his case.
Two examples of the above to end off my comment:
In the Aeon article Humphrey states that sensations (i think he means qualia) are ideas. I agree! And these ideas exist. It feels very hard to explain how they exist in a physicalist or reductionist story, and attempts to explain often fail to the âMooreanâ argument for phenomenal realism.[3]
In Figure 2 he shows a simplified, 4-step diagram of how a brain might create an âattractor stateâ of phenomenalisation. This is a clear story but seems to me to lead to eliminativist illusionism. The hard question is this, why is this attractor state experiencing phenomenality, and not just a p-Zombie? And it seems that, as all physicalist theories go, Humphrey has no answer to this.
And so help me I have tried! Iâve directly read Dennett, Frankish, and Kammerer! But they all just seem so obviously wrong when they make clear claims; or annoyingly slippery and vague when they donât. Imo Chalmers blows them all away.
Kammerer reviews this argument here and states it fails. But I just found it a convincing refutation of his own illusionism! I suppose one manâs modus ponens really is anotherâs modus tollens
I tend to think that questions about which organisms or systems are conscious mostly depend on identifying the physical correlates of consciousness, and understanding how they work as a system, and that questions about panpsychism, illusionism, eliminativism, or even Calmerâs Hard Problem donât bear on this question very much. I think thereâs probably still a place for that philosophical debate because (1) there might be implications about where to look for the physical systems and (2) as I said to Michael earlier, illusionism might change our perspective on whether we assign special moral value to conscious experience at all. But I think (1) is marginal, and (2) is sort of a long shot.
In contrast, I think empirical and scientific investigation can help us understand a lot about which systems are conscious, and about what sort of conscious experiences they have, so I think most morally cruxy questions of consciousness are scientific and empirical.
Consequently, I wasnât too bothered by Humphrey side-stepping this issue, although I basically agree he did, because he offered solid theory and empirical investigation that suggests further empirical tests that might help us make progress on understanding consciousness in animals and other systems.
The D.B. case is also an interesting oneâI donât see why it isnât plausible to imagine that the operation (and similarly split brain cases as documented by Nagel) might lead to a second vestige of consciousness as cut off from you as that of your family, friends, and coworkers. Except this fragment doesnât have control of motor or speach functions, how horrible! It can only pass information onto the âdominantâ one.
That was my reaction when I first read about split-brain patients. I now doubt itâs all that horrible. First, thereâs been plenty of research of split-brain patients and I donât think anyone has discovered signs of distress from split-halfs that are cut off from speech expression; those halves do have other ways of communicating, e.g., through signs. Second, in humans, much of distress is governed physiologically, and so (1) we would be able to detect physiological signs of stress, but more importantly (2) even if thereâs a conscious half of a split-brain which canât express itself, its mood-state might be normal because it shares a body with the other half, and so the two jointly set mood, and the system overall might not be in distress. Finally, even if consciousness isnât illusory, conscious will often is; much more of our decisions are determined subconsciously than we think, and if the illusion still holds, the loss of conscious control might not even be perceived.
In the Aeon article Humphrey states that sensations (i think he means qualia) are ideas. I agree! And these ideas exist. It feels very hard to explain how they exist in a physicalist or reductionist story, and attempts to explain often fail to the âMooreanâ argument for phenomenal realism.
Couldnât they just be beliefs? There are various (physicalist+reductionst-compatible) accounts of belief here. (Iâm not sure if thatâs what Humphrey intended, though.)
The hard question is this, why is this attractor state experiencing phenomenality, and not just a p-Zombie? And it seems that, as all physicalist theories go, Humphrey has no answer to this.
I think Humphrey is an illusionist and so would dny that itâs experiencing phenomenality and that the hard problem needs to be solved (instead, illusionists dissolve it). However, a complete illusionist theory should explain how beliefs of phenomenality arise (e.g. how these attractor states cause these beliefs); thatâs the âhard problemâ for illusionists. From what Iâm reading in this post and comments, it seems he doesnât explain that. So, whether interpreted as realist or illusionist, it seems thereâs still an important gap.
Yeah Iâm not quiet sure what Humphrey means by belief here (thanks for the link!). But then I donât really know what I mean by âbeliefâ to be honest! Iâm not sure I can define it without evoking my own phenomenal perspective, whether directly (a flash of inspiration) or via thought and introspection (trying to update my current beliefs with new evidence) - and I think thatâd already put me at odds with physicalists/âillusionists
I do think Humphrey is an illusionist (even if he doensât like the term) and view him as somewhat adjacent to Frankish. I think that the âmeta-problem of consciousnessâ isnât quite what Iâm hinting at (though it is a problem) - Iâm taking my phenomenal experiences as true. Dualists (like myself) need to try and get them to accord with our understanding of the physical world, but I think illusionists need to explain why Iâm experiencing anything at all rather than just reporting I am. We probably have very different intuitions on this, but part of the reason Iâve become more âdualistâ over time is that I found that I never had a good response to this criticism when I was a materialist/âphysicalist, so in the end I accepted it as a worthy criticism that disproved my original ideas.
Finally, Iâll note this isnât the first time weâve had a Forum discussion about consciousness[1] - maybe itâs something we could explore in a dialogue if itâs something you think would be a valuable use of our time and potentially useful for those reading on the Forum? It definitely touches on a number of EA cause areas.
I think illusionists need to explain why Iâm experiencing anything at all rather than just reporting I am
We need to first decide what we mean by âexperienceâ. I think there are two broad approaches (interpretations) of illusionism, which I described here and which could give us two different broad characterizations of âexperienceâ:
In the first, beliefs (illusions) of phenomenality/âmysteriousness/ânonphysical essence/âetc. themselves could be what distinguishes whatâs experienced from whatâs not experienced. These beliefs need not be verbalized (whether in inner speech or reported) and could be of a more intuitive kind, like Grazianoâs attention schema or Humphreyâs ipsundrum are meant to capture.[1] They might just be representations or models âdepictingâ phenomenal properties. See also my footnote here on Grazianoâs Attention Schema Theory. So, these beliefs would explain why youâre experiencing anything at all.
In the second, the physical properties that dispose us to have such beliefs could be what distinguishes experiences. This could be a kind of placeholder, but I suspect Frankish and Dennett would say that any reactive patterns and discriminations count, at least to some minimal degree.[2] So, thermometers and bacteria could be minimally experiencing things, too, and that youâre reacting and making any discriminations at all would explain why youâre experiencing anything at all. Those with blindsight could still have visual experiences, but in a way thatâs not accessible for standard verbal report and possibly of a more simple/âminimal kind.
I suspect thereâs no real fact of the matter which approach is ârightâ, but Iâm more inclined towards 1.
If you have something else in mind by âexperienceâ, I could try to respond to that.
Finally, Iâll note this isnât the first time weâve had a Forum discussion about consciousness[1] - maybe itâs something we could explore in a dialogue if itâs something you think would be a valuable use of our time and potentially useful for those reading on the Forum? It definitely touches on a number of EA cause areas.
I might be interested in having a (recorded) call, and then we can release it, the (edited) transcripts and/âor notes. I spend way too long writing comments (including this one, and others on this post), so I think I shouldnât commit to a text-based discussion.
That being said, Iâm not sure how useful this would be for other people, compared to them just reading writing by or listening to Graziano or Frankish. It was Grazianoâs papers (2021, 2022, some clarifications in 2020) that made illusionism click for me,[3] and I suspect I couldnât do a better job in explaining illusionism than to just linkpost or quote him, as well as Kammerer, 2022 (or just the short summary in Shabasson, 2021, section 9), which helps illustrate how the illusion could be so persistent.
I think the basic argument that convinced me roughly goes like this, based on Graziano (2021, 2022), and from a draft I wrote but never posted:
Our claims of conscious experience result from the depiction/ârepresentation of information processed in our brains as having properties we believe as common to our conscious experiences, like phenomenality, subjectivity, qualitativeness or a nonphysical essence. There must be information in our brains depicting these properties, because without such information, we wouldnât consistently talk about these properties in the first place. Of course, maybe the information processing appears to have these properties precisely because it actually has these properties, and thatâs a realist position. However, the depiction itself and access to it by systems necessary for belief formation would be enough, and thatâs the illusionist position. Thereâs no need to posit the actual existence of these properties, and in my view, thereâs currently no plausible explanation for the actual existence of these properties.
However, some things may make me unusually likely to accept illusionism:
I suspect my direct intuitions about physical phenomena and consciousness are relatively weak, and Iâm unusually inclined towards abstraction, so Iâve found little to count against illusionism for me. That consciousness just seems phenomenal, and red seems to be qualitative just doesnât count very strongly to me.
I have a very strong presumption in favour of physicalism,[4] and every non-illusionist physicalist theory doesnât seem to me to offer a serious attempt to solve the hard problem, so the best option seems to be to dissolve it, hence illusionism. It sounds like you went the other way towards dualism through your dissatisfaction with physicalist theories, and Iâd guess Chalmers did, too.
Dogs presumably do not think there is something it is like to be them, even if there is. It is not that a dog thinks there isnât anything it is like to be a dog; the dog is not a theorist at all, and hence does not suffer from the theoristsâ illusion. The hard problem and meta-problem are only problems for us humans, and mainly just for those of us humans who are particularly reflective. In other words, dogs arenât bothered or botherable by problem intuitions. Dogs â and, for that matter, clams and ticks and bacteria â do enjoy (or at any rate benefit from) a sort of user illusion: they are equipped to discriminate and track only some of the properties in their environment.
They were also the first explanations of illusionism Iâd read. I havenât settled on Grazianoâs AST in particular, but it seems like a promising direction.
But also, other than panpsychism, where could we possibly draw a line for the presence and absence of the extra nonphysical property/âproperties? I canât imagine there being any plausible responses.
Or, if panpsychist, how could these properties possibly combine in ways that correspond to what our brains are doing and our specific judgements? Maybe some kind of property dualism?
I also canât imagine there being any plausible account of how the nonphysical affects the physical (or else we would already have identified it and adopted it into our physical ontology), so Iâd be stuck with epiphenomenalism.
So, if not an illusionist, Iâd have to be a (property dualist) epiphenomenalist, and it seems there would be no way to empirically distinguish such accounts from their illusionist counterparts, which just drop the nonphysical stuff. And whether or not there are different ethical implications between them, I canât imagine them being that decisive in practice. The difference just doesnât seem that interesting anymore, but I favour the metaphysically more parsimonious illusionism.
FWIW, I havenât read much Chalmers, and I learned about property dualism after illusionism already became intuitive to me.
It was Grazianoâs papers (2021, 2022, some clarifications in 2020) that made illusionism click for me
I found the article from 2022 clarifying (I have not looked into the other 2), but I am still endorsing (physical) realism. I have found persuasive Hakwan Lauâs views about consciousness. So I asked Gemini âWhat would Lau say about strong illusionism?â. The answer is below. It makes a lot of sense to me. I would be curious to know your thoughts. I wonder whether disagreements between illusionists and realists are largely semantic.
Hakwan Lau is fundamentally opposed to strong illusionism. While he agrees with illusionists that our introspection is imperfect and fails to reveal the physical mechanics of our brains, he completely rejects their conclusion that this makes phenomenal consciousness a âfalse beliefâ or a non-existent property.
In his book In Consciousness We Trust, Lau argues that strong illusionism throws the baby out with the bathwater. Here is how he dismantles the strong illusionist position:
1. Macro-Level Summaries are Not âIllusionsâ
Strong illusionists (like Graziano, Dennett, and Frankish) argue that because our internal models donât represent the micro-physical details of our neural wiring, the non-physical âmagicâ they report must be a hallucination.
Lau counters this with a functional analogy: macroscopic metrics are real, even if they ignore microscopic details. When a digital thermometer gives you a reading of 20°C, it is not giving you a detailed map of the kinetic energy of billions of individual air molecules. It is giving you a simplified, macroscopic summary. But that doesnât mean âtemperatureâ is an illusion or a false belief.
Similarly, Lauâs Perceptual Reality Monitoring (PRM) argues that the prefrontal cortex generates a macroscopic summary of sensory reliability. It evaluates a signal and outputs a tag: This is real and present. Just because that tag lacks the metadata of neurotransmitter release doesnât make the resulting subjective experience an illusion. It is a highly accurate, biologically real computational output.
2. The Fallacy of the âPerfectâ Introspection
Illusionism relies on the premise that for consciousness to be real, it must perfectly match our intuitive, pre-scientific introspective beliefs about it (i.e., that it is magical, ethereal, or non-physical). Because science proves it isnât magical, illusionists conclude it isnât real.
Lau points out that this is an unfair standard. Science routinely redefines phenomena without eliminating them. When we discovered that water is HâO rather than an indivisible classical element, we didnât declare that water was an illusion. Lau argues we should treat consciousness the same way: it isnât the magical inner theater we intuitively thought it was, but rather a physical, metacognitive reality-monitoring system. Redefining it doesnât erase its existence.
3. The Evolutionary Reality of âPresenceâ
Strong illusionists argue that the âfeelingâ of experience is just an evolutionary trick to force an organism to care about its environment.
Lau argues that the brain has a much more specific, computationally vital need for this feeling: it must distinguish reality from its own internal noise. A brain constantly generates internal imageryâdreams, memories, imagination, and predictive models. If an organism cannot tell the difference between a memory of a predator and a real predator right in front of it, it will not survive.
For Lau, subjective consciousness is that mechanism. When the PRM system successfully tags a sensory signal as originating from the external world, it generates a feeling of undeniable âpresence.â That presence is not a trick; it is the correct computational answer to the problem of reality monitoring.
Summary
Lau would say that strong illusionists correctly identify that the brain uses simplified representations, but they make a fatal philosophical error in concluding that simplified representations are false. To Lau, subjective experience is the brainâs very real, very physical mechanism for anchoring us in reality.
First, before responding to the specific claims in the Gemini output:
The Gemini output seems like AI slop, and I would recommend against passing along LLM outputs uncritically like this. The answer seems quite bad, most of what it said seems wrong or misguided. You could run another pass of an LLM on the answer to critique it to help a bit. I hope this doesnât accurately represent Lauâs views. Illusionism is often misunderstood and strawmanned, including by academics, so thereâs a lot of garbage out there for LLMs to have been trained on. Some of the problem with misunderstanding and strawmanning is probably due to its unfortunate name. (Iâll flag what looks bad below.)
I donât see a plausible path for Lauâs view to solve the hard problem or meta problem of consciousness. Does Lau explain why people so often regard consciousness as non-physical, ethereal, having classic qualia properties, or some qualitative/âsubjective character beyond just the physical? Or does he reject that people often do regard consciousness this way?
Specifically, I donât see how reality monitoring helps. Internally generated mental imagery is still experienced. Itâs just experienced differently.
Iâm not familiar with Lauâs theory; thereâs problably more. But Iâve looked at so many theories, and my prior is to be extremely skeptical.
1. Macro-Level Summaries are Not âIllusionsâ
Is this section trying to say the non-physical âmagicâ is real? If not, how does it contradict illusionism?
Lau counters this with a functional analogy: macroscopic metrics are real, even if they ignore microscopic details. When a digital thermometer gives you a reading of 20°C, it is not giving you a detailed map of the kinetic energy of billions of individual air molecules. It is giving you a simplified, macroscopic summary. But that doesnât mean âtemperatureâ is an illusion or a false belief.
This looks a lot like what Graziano says about his own illusionist theory here, so doesnât seem like a point against illusionism:
These models are not empty illusions: they are caricatures. They represent something physically real, but they are not accurate. Models never are. The brainâs models are useful, adaptive, simplified, and never fully accurate, yet they form the basis of our beliefs, thoughts, and claims.
That being said, Iâd say the appearance of phenomenal properties is an illusion, and possibly an active misrepresentation, not merely a simplification. See Kammerer, 2019.
Similarly, Lauâs Perceptual Reality Monitoring (PRM) argues that the prefrontal cortex generates a macroscopic summary of sensory reliability. It evaluates a signal and outputs a tag: This is real and present. Just because that tag lacks the metadata of neurotransmitter release doesnât make the resulting subjective experience an illusion. It is a highly accurate, biologically real computational output.
I donât see how this solves the hard problem or meta problem.
2. The Fallacy of the âPerfectâ Introspection
Illusionism relies on the premise that for consciousness to be real, it must perfectly match our intuitive, pre-scientific introspective beliefs about it (i.e., that it is magical, ethereal, or non-physical).
This is false. Illusionsim does not rely on such a premise.
Lau points out that this is an unfair standard. Science routinely redefines phenomena without eliminating them. When we discovered that water is HâO rather than an indivisible classical element, we didnât declare that water was an illusion. Lau argues we should treat consciousness the same way: it isnât the magical inner theater we intuitively thought it was, but rather a physical, metacognitive reality-monitoring system. Redefining it doesnât erase its existence.
Illusionism is (re)defining consciousness without eliminating it. What it eliminates is the âphenomenalâ properties of consciousness, and classic qualia are roughly the only formalization/âprecisification of phenomenality so far that doesnât end up strong illusionist. See Frankishâs Quining Diet Qualia. âWhat it feels likeâ, âsubjectiveâ, etc. are not formal/âprecise and donât rule out strong illusionism.
And again, I donât see how reality monitoring helps.
3. The Evolutionary Reality of âPresenceâ
Seems independent of illusionism vs realism. Mental imagery is still experienced, just differently. Is Lau (or just Gemini?) saying the feeling of presence is a phenomenal property, and mental imagery lacks all phenomenal properties? Would you agree that dreams, thoughts and visualizations arenât phenomenally conscious experiences, but your externally generated experiences are phenomenally conscious? I donât think this is compatible with what most would intend by âphenomenal consciousnessâ, and if this is what Lau (or Gemini?) wants to do, heâs working on a totally different problem from pretty much everyone else.
Summary
Lau would say that strong illusionists correctly identify that the brain uses simplified representations, but they make a fatal philosophical error in concluding that simplified representations are false.
Seems like a strawman. Strong illusionists donât conclude that simplified representations are false. They conclude that there are no phenomenal properties, and any representations as if there are phenomenal properties are therefore misleading (illusions).
I would recommend against passing along LLM outputs uncritically like this
Do you mean you recommend against sharing them, even if the alternative is not sharing anything (I agree sharing them, and examining them would be better)? Sharing them could lead to them being corrected to the extent they are not accurate, as you just did, which is good? Maybe you mean they should be shared with a quick disclaimer that the claims were not checked.
Does Lau explain why people so often regard consciousness as non-physical, ethereal, having classic qualia properties, or some qualitative/âsubjective character beyond just the physical? Or does he reject that people often do regard consciousness this way?
I do not recall Lau discussing the above in the few articles and podcasts I checked. I do not think Lau would reject that people often regard consciousness that way.
Is this section trying to say the non-physical âmagicâ is real? If not, how does it contradict illusionism?
I believe the section is trying to say that phenomenal consciousness exists. I agree this would not contradict illusionism. It would only contradict eliminativism.
Would you agree that dreams, thoughts and visualizations arenât phenomenally conscious experiences, but your externally generated experiences are phenomenally conscious?
I would say dreams and thoughts are phenomenally conscious experiences. I am not sure what you mean by âexternally generated experiencesâ.
I meant that phenomenal consciousness is real in the sense ârealâ is used in common language, but I see now this is not what realism is about. âRealâ as used in common language can mean something that exists. However, illusionists acknowledge that phenomenal consciousness exists. They just argue it is not real in the sense it does not have the classic qualia properties. I seem to agree with this.
Having a better look into Frankish (2016) helped me understand what illusionism is actually about. I had only skimmed it quickly, and it is easy to misinterpret it on a quick skim because some terms like âexperienceâ are used in arguably unconventional ways.
(I shall use the term âexperienceâ itself in a functional sense, for the mental states that are the direct output of sensory systems. In this sense it is not definitional that experiences are phenomenally conscious.)
Would you do anything differently if you endorsed weak illusionism/ârealism as defined in Frankish (2016) instead of (strong) illusionism?
Illusionism makes a very strong claim: it claims that phenomenal consciousness is illusory; experiences do not really have qualitative, âwhat-itâs-likeâ properties, whether physical or non-physical. This should be distinguished from a weaker view according to which some of the supposed features of phenomenal consciousness are illusory. Many conservative realists argue that phenomenal properties, though real, do not possess the problematic features sometimes ascribed to them, such as being ineffable, intrinsic, private, and infallibly known. Phenomenal feels, they argue, are physical properties which introspection misrepresents as ineffable, intrinsic, and so on. We might call this weak illusionism, in contrast to the strong form advocated here. (It might equally be called weak realism.)4
The amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it.
LLMs make it easy to produce a lot of plausible-looking but wrong or misleading claims quickly, and sharing them without checking yourself (or at least running another LLM to do that) puts the burden on others to correct.
I think you should make sure the claims are checked before sharing.
You could do this yourself, or run another critical LLM pass on the answer, e.g. a prompt like âExhaustively verify and ruthlessly critique the above answer.â
The critique pass itself may turn up false positive critiques, and you might want to run a third pass to verify the critiques and write a new balanced answer.
(I agree that itâs also important to tag that itâs LLM output, and at least read it over yourself, which you did.)
However, illusionists acknowledge that phenomenal consciousness exists. They just argue it is not real in the sense it does not have the classic qualia properties. I seem to agree with this.
This is true of weak illusionists, but not strong illusionists. Strong illusionists reject phenomenal consciousness. Specifically they reject phenomenal properties; they accept that consciousness exists.
Would you do anything differently if you endorsed weak illusionism/ârealism as defined in Frankish (2016) instead of (strong) illusionism?
Other than argue about it, probably not much, assuming functionalism and materialism/âphysicalism of some kind thatâs compatible with artificial sentience.
FWIW, my impression is that there are ways to interpret âwhat-itâs-likeâ or âsubjectiveâ consistently with strong illusionism, as, e.g. the specific patterns of discriminations a brain makes or what information is accessible to it (or its attention or its global workspace). I think Frankish himself has done this later.
But âphenomenalâ and âqualitativeâ seem harder to pin down precisely in purely physical, discriminatory and informational terms in a way that I think would be broadly acceptable after critique and reflection to realists (including weak illusionists).
LLMs make it easy to produce a lot of plausible-looking but wrong or misleading claims quickly, and sharing them without checking yourself (or at least running another LLM to do that) puts the burden on others to correct.
This also applies to many comments written by people? In general, whenever something is shared, one could invest more time to make it more accurate. However, to me it seems hard to come up with general rules about how much time to invest vetting claims.
I agree that itâs also important to tag that itâs LLM output, and at least read it over yourself, which you did.
I think sharing long texts produced by LLMs is often fine even if they were not fully read by the person prompting, basically for the same reasons that it is fine to share long text produced by people even if the person sharing them did not fully read them.
Strong illusionists reject phenomenal consciousness. Specifically they reject phenomenal properties; they accept that consciousness exists.
I have the impression many people, including academics, use âconsciousnessâ and âphenomenal consciousnessâ interchangeably, and without wanting to take a stance on illusionism or realism. I think this is one reason strong illusionists saying they reject phenomenal consciousness is sometimes misinterpreted as them rejecting the illusion of phenomenal consciousness.
This also applies to many comments written by people? In general, whenever something is shared, one could invest more time to make it more accurate. However, to me it seems hard to come up with general rules about how much time to invest vetting claims.
If someone had a pattern of fabrication and very poor understanding (and apparent confidence) like LLMs often do if used uncritically, I would be annoyed with them and possibly do any of the following:
Tell them to read and review more carefully, look for opposing arguments, etc..
Downvote such comments (and I very very rarely downvote).
Stop engaging with this person, because it wastes my time and may encourage them to waste othersâ time.
I think sharing long texts produced by LLMs is often fine even if they were not fully read by the person prompting, basically for the same reasons that it is fine to share long text produced by people even if the person sharing them did not fully read them.
I would request you check it yourself or at least run LLM critique passes, because of high rates of hallucination and other errors by LLMs.
Similarly, if there was an author who was just as consistently bad as LLMs are, and you shared their work uncritically like this, Iâd recommend the same.
I have the impression many people, including academics, use âconsciousnessâ and âphenomenal consciousnessâ interchangeably, and without wanting to take a stance on illusionism or realism.
My impression is that many people, including academics, use âconsciousnessâ and âphenomenal consciousnessâ interchangeably, but they do so implicitly rejecting strong illusionism, would reject strong illusionism if asked directly, and typically donât understand strong illusionism. Maybe many are open, though, Iâm not sure.
If someone had a pattern of fabrication and very poor understanding (and apparent confidence) like LLMs often do if used uncritically, I would be annoyed with them and possibly do any of the following:
Tell them to read and review more carefully, look for opposing arguments, etc..
Downvote such comments (and I very very rarely downvote).
Stop engaging with this person, because it wastes my time and may encourage them to waste othersâ time.
Makes sense. I just think what is âpoor understandingâ is often sufficiently contentious for one to have a high bar for preferring not sharing over sharing with little verification. I also tend to default to let people decide the extent to which they want to engage with something.
My impression is that many people, including academics, use âconsciousnessâ and âphenomenal consciousnessâ interchangeably, but they do so implicitly rejecting strong illusionism, would reject strong illusionism if asked directly, and typically donât understand strong illusionism. Maybe many are open, though, Iâm not sure.
Makes sense. I was not clear. However, by âwithout wanting to take a stance on illusionism or realismâ, I meant many do not have a good picture of what strong illusionism means (and may be conflating it with eliminativism), and therefore are not taking a stance of the correct version of illusionism. I was doing this to some extent.
The amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it.
Other than argue about it, probably not much, assuming functionalism and materialism/âphysicalism of some kind thatâs compatible with artificial sentience.
Illusionism states that phenomenal consciousness does not exist, even though it seems to exist. While illusionism is controversial, it is a serious contender among theories of consciousness. We argue that it has substantial and non-trivial implications for non-human consciousness research (NHCR), particularly for the study of the distribution of phenomenal consciousness across beings. If illusionism is true, NHCR can be pursued if conceptualized as investigating the distribution of quasi-phenomenal consciousness, i.e. the states which are misrepresented as phenomenally conscious in humans. However, we argue that knowing the distribution of quasi-phenomenal consciousness is not highly informative. For this reason, illusionism suggests that some approaches to NHCR should be preferred over others. Approaches which focus on features that provide valuable information about non-human cognition independently of their supposed relation to consciousness retain much of their value if illusionism is true. We propose a âzombie testâ and five specific heuristics to help identifying such features. Consequently, empirical researchers who take illusionism seriously gain a reason to prioritize some methodological approaches over others.
Thanks Ben for reviewing and sharing the post (and to Michael for your resources) I did find it very interesting. But after reading it, giving it some thought, and reading the Aeon essay you recommended I came away quite intellectually frustrated. I think whenever some claims to have answered the hard problem, the more likely explanation is that they havenât understood it.
(Full disclosure for me confusion, Iâve all but become a strong dualist about consciousness after many many years of being or identitfying as a physicalist. This was a result of reading the philosophical literature, but also a lot of what was a personal reflection on my own experiences through meditation practice, especially informed by Douglas Hardingâs Headless Way. I basically canât grok illusionist perspectives any more try as I might[1], so readers might better read this comment as coming from a dualist loyal opposition, rather than a balanced assessment)
Something I found particularly puzzling was in the 3rd section of the essay (beginning âOver the past 50 years...â), Humphrey seems to confuse mind-brain identity theory with panyschism, which doesnât for me bode well my expectation of his âsolvingâ of the hard problem. M-B-I to me is a strongly physicalist (if not eliminativist) position, while pansychism is by necessity dualist. [2]
The D.B. case is also an interesting oneâI donât see why it isnât plausible to imagine that the operation (and similarly split brain cases as documented by Nagel) might lead to a second vestige of consciousness as cut off from you as that of your family, friends, and coworkers. Except this fragment doesnât have control of motor or speach functions, how horrible! It can only pass information onto the âdominantâ one.
The case of whether Humphrey is right about Octopi not playing is outside of my domain of expertise but initially I am sceptical. But again just because we know that we (humans) are conscious in a certain way, why ought we to imagine that consciousness must only exist in this way? Iâm not sure it follows, and there seem to be gaps like this in his arguments that actually cover up important parts of his case.
Two examples of the above to end off my comment:
In the Aeon article Humphrey states that sensations (i think he means qualia) are ideas. I agree! And these ideas exist. It feels very hard to explain how they exist in a physicalist or reductionist story, and attempts to explain often fail to the âMooreanâ argument for phenomenal realism.[3]
In Figure 2 he shows a simplified, 4-step diagram of how a brain might create an âattractor stateâ of phenomenalisation. This is a clear story but seems to me to lead to eliminativist illusionism. The hard question is this, why is this attractor state experiencing phenomenality, and not just a p-Zombie? And it seems that, as all physicalist theories go, Humphrey has no answer to this.
And so help me I have tried! Iâve directly read Dennett, Frankish, and Kammerer! But they all just seem so obviously wrong when they make clear claims; or annoyingly slippery and vague when they donât. Imo Chalmers blows them all away.
Galen Strawson actually has an interesting argument that physicalism must collapse into pansychism.
Kammerer reviews this argument here and states it fails. But I just found it a convincing refutation of his own illusionism! I suppose one manâs modus ponens really is anotherâs modus tollens
I tend to think that questions about which organisms or systems are conscious mostly depend on identifying the physical correlates of consciousness, and understanding how they work as a system, and that questions about panpsychism, illusionism, eliminativism, or even Calmerâs Hard Problem donât bear on this question very much. I think thereâs probably still a place for that philosophical debate because (1) there might be implications about where to look for the physical systems and (2) as I said to Michael earlier, illusionism might change our perspective on whether we assign special moral value to conscious experience at all. But I think (1) is marginal, and (2) is sort of a long shot.
In contrast, I think empirical and scientific investigation can help us understand a lot about which systems are conscious, and about what sort of conscious experiences they have, so I think most morally cruxy questions of consciousness are scientific and empirical.
Consequently, I wasnât too bothered by Humphrey side-stepping this issue, although I basically agree he did, because he offered solid theory and empirical investigation that suggests further empirical tests that might help us make progress on understanding consciousness in animals and other systems.
That was my reaction when I first read about split-brain patients. I now doubt itâs all that horrible. First, thereâs been plenty of research of split-brain patients and I donât think anyone has discovered signs of distress from split-halfs that are cut off from speech expression; those halves do have other ways of communicating, e.g., through signs. Second, in humans, much of distress is governed physiologically, and so (1) we would be able to detect physiological signs of stress, but more importantly (2) even if thereâs a conscious half of a split-brain which canât express itself, its mood-state might be normal because it shares a body with the other half, and so the two jointly set mood, and the system overall might not be in distress. Finally, even if consciousness isnât illusory, conscious will often is; much more of our decisions are determined subconsciously than we think, and if the illusion still holds, the loss of conscious control might not even be perceived.
Couldnât they just be beliefs? There are various (physicalist+reductionst-compatible) accounts of belief here. (Iâm not sure if thatâs what Humphrey intended, though.)
I think Humphrey is an illusionist and so would dny that itâs experiencing phenomenality and that the hard problem needs to be solved (instead, illusionists dissolve it). However, a complete illusionist theory should explain how beliefs of phenomenality arise (e.g. how these attractor states cause these beliefs); thatâs the âhard problemâ for illusionists. From what Iâm reading in this post and comments, it seems he doesnât explain that. So, whether interpreted as realist or illusionist, it seems thereâs still an important gap.
Yeah Iâm not quiet sure what Humphrey means by belief here (thanks for the link!). But then I donât really know what I mean by âbeliefâ to be honest! Iâm not sure I can define it without evoking my own phenomenal perspective, whether directly (a flash of inspiration) or via thought and introspection (trying to update my current beliefs with new evidence) - and I think thatâd already put me at odds with physicalists/âillusionists
I do think Humphrey is an illusionist (even if he doensât like the term) and view him as somewhat adjacent to Frankish. I think that the âmeta-problem of consciousnessâ isnât quite what Iâm hinting at (though it is a problem) - Iâm taking my phenomenal experiences as true. Dualists (like myself) need to try and get them to accord with our understanding of the physical world, but I think illusionists need to explain why Iâm experiencing anything at all rather than just reporting I am. We probably have very different intuitions on this, but part of the reason Iâve become more âdualistâ over time is that I found that I never had a good response to this criticism when I was a materialist/âphysicalist, so in the end I accepted it as a worthy criticism that disproved my original ideas.
Finally, Iâll note this isnât the first time weâve had a Forum discussion about consciousness[1] - maybe itâs something we could explore in a dialogue if itâs something you think would be a valuable use of our time and potentially useful for those reading on the Forum? It definitely touches on a number of EA cause areas.
And Iâve very much enjoyed learning from your perspective :)
We need to first decide what we mean by âexperienceâ. I think there are two broad approaches (interpretations) of illusionism, which I described here and which could give us two different broad characterizations of âexperienceâ:
In the first, beliefs (illusions) of phenomenality/âmysteriousness/ânonphysical essence/âetc. themselves could be what distinguishes whatâs experienced from whatâs not experienced. These beliefs need not be verbalized (whether in inner speech or reported) and could be of a more intuitive kind, like Grazianoâs attention schema or Humphreyâs ipsundrum are meant to capture.[1] They might just be representations or models âdepictingâ phenomenal properties. See also my footnote here on Grazianoâs Attention Schema Theory. So, these beliefs would explain why youâre experiencing anything at all.
In the second, the physical properties that dispose us to have such beliefs could be what distinguishes experiences. This could be a kind of placeholder, but I suspect Frankish and Dennett would say that any reactive patterns and discriminations count, at least to some minimal degree.[2] So, thermometers and bacteria could be minimally experiencing things, too, and that youâre reacting and making any discriminations at all would explain why youâre experiencing anything at all. Those with blindsight could still have visual experiences, but in a way thatâs not accessible for standard verbal report and possibly of a more simple/âminimal kind.
I suspect thereâs no real fact of the matter which approach is ârightâ, but Iâm more inclined towards 1.
If you have something else in mind by âexperienceâ, I could try to respond to that.
I might be interested in having a (recorded) call, and then we can release it, the (edited) transcripts and/âor notes. I spend way too long writing comments (including this one, and others on this post), so I think I shouldnât commit to a text-based discussion.
That being said, Iâm not sure how useful this would be for other people, compared to them just reading writing by or listening to Graziano or Frankish. It was Grazianoâs papers (2021, 2022, some clarifications in 2020) that made illusionism click for me,[3] and I suspect I couldnât do a better job in explaining illusionism than to just linkpost or quote him, as well as Kammerer, 2022 (or just the short summary in Shabasson, 2021, section 9), which helps illustrate how the illusion could be so persistent.
I think the basic argument that convinced me roughly goes like this, based on Graziano (2021, 2022), and from a draft I wrote but never posted:
Our claims of conscious experience result from the depiction/ârepresentation of information processed in our brains as having properties we believe as common to our conscious experiences, like phenomenality, subjectivity, qualitativeness or a nonphysical essence. There must be information in our brains depicting these properties, because without such information, we wouldnât consistently talk about these properties in the first place. Of course, maybe the information processing appears to have these properties precisely because it actually has these properties, and thatâs a realist position. However, the depiction itself and access to it by systems necessary for belief formation would be enough, and thatâs the illusionist position. Thereâs no need to posit the actual existence of these properties, and in my view, thereâs currently no plausible explanation for the actual existence of these properties.
However, some things may make me unusually likely to accept illusionism:
I suspect my direct intuitions about physical phenomena and consciousness are relatively weak, and Iâm unusually inclined towards abstraction, so Iâve found little to count against illusionism for me. That consciousness just seems phenomenal, and red seems to be qualitative just doesnât count very strongly to me.
I have a very strong presumption in favour of physicalism,[4] and every non-illusionist physicalist theory doesnât seem to me to offer a serious attempt to solve the hard problem, so the best option seems to be to dissolve it, hence illusionism. It sounds like you went the other way towards dualism through your dissatisfaction with physicalist theories, and Iâd guess Chalmers did, too.
But might leave out too many details of how this actually works in humans and other animals to be very satisfying.
E.g. Frankish on continuity here (54:00-57:37).
Also Dennett (2019, p. 54):
And Dennett thinks that chickens, octopuses and bees are definitely conscious.
They were also the first explanations of illusionism Iâd read. I havenât settled on Grazianoâs AST in particular, but it seems like a promising direction.
Just generally.
But also, other than panpsychism, where could we possibly draw a line for the presence and absence of the extra nonphysical property/âproperties? I canât imagine there being any plausible responses.
Or, if panpsychist, how could these properties possibly combine in ways that correspond to what our brains are doing and our specific judgements? Maybe some kind of property dualism?
I also canât imagine there being any plausible account of how the nonphysical affects the physical (or else we would already have identified it and adopted it into our physical ontology), so Iâd be stuck with epiphenomenalism.
So, if not an illusionist, Iâd have to be a (property dualist) epiphenomenalist, and it seems there would be no way to empirically distinguish such accounts from their illusionist counterparts, which just drop the nonphysical stuff. And whether or not there are different ethical implications between them, I canât imagine them being that decisive in practice. The difference just doesnât seem that interesting anymore, but I favour the metaphysically more parsimonious illusionism.
FWIW, I havenât read much Chalmers, and I learned about property dualism after illusionism already became intuitive to me.
Hi Michael. Thanks for sharing.
I found the article from 2022 clarifying (I have not looked into the other 2), but I am still endorsing (physical) realism. I have found persuasive Hakwan Lauâs views about consciousness. So I asked Gemini âWhat would Lau say about strong illusionism?â. The answer is below. It makes a lot of sense to me. I would be curious to know your thoughts. I wonder whether disagreements between illusionists and realists are largely semantic.
First, before responding to the specific claims in the Gemini output:
The Gemini output seems like AI slop, and I would recommend against passing along LLM outputs uncritically like this. The answer seems quite bad, most of what it said seems wrong or misguided. You could run another pass of an LLM on the answer to critique it to help a bit. I hope this doesnât accurately represent Lauâs views. Illusionism is often misunderstood and strawmanned, including by academics, so thereâs a lot of garbage out there for LLMs to have been trained on. Some of the problem with misunderstanding and strawmanning is probably due to its unfortunate name. (Iâll flag what looks bad below.)
I donât see a plausible path for Lauâs view to solve the hard problem or meta problem of consciousness. Does Lau explain why people so often regard consciousness as non-physical, ethereal, having classic qualia properties, or some qualitative/âsubjective character beyond just the physical? Or does he reject that people often do regard consciousness this way?
Specifically, I donât see how reality monitoring helps. Internally generated mental imagery is still experienced. Itâs just experienced differently.
Iâm not familiar with Lauâs theory; thereâs problably more. But Iâve looked at so many theories, and my prior is to be extremely skeptical.
Is this section trying to say the non-physical âmagicâ is real? If not, how does it contradict illusionism?
This looks a lot like what Graziano says about his own illusionist theory here, so doesnât seem like a point against illusionism:
That being said, Iâd say the appearance of phenomenal properties is an illusion, and possibly an active misrepresentation, not merely a simplification. See Kammerer, 2019.
I donât see how this solves the hard problem or meta problem.
This is false. Illusionsim does not rely on such a premise.
Illusionism is (re)defining consciousness without eliminating it. What it eliminates is the âphenomenalâ properties of consciousness, and classic qualia are roughly the only formalization/âprecisification of phenomenality so far that doesnât end up strong illusionist. See Frankishâs Quining Diet Qualia. âWhat it feels likeâ, âsubjectiveâ, etc. are not formal/âprecise and donât rule out strong illusionism.
And again, I donât see how reality monitoring helps.
Seems independent of illusionism vs realism. Mental imagery is still experienced, just differently. Is Lau (or just Gemini?) saying the feeling of presence is a phenomenal property, and mental imagery lacks all phenomenal properties? Would you agree that dreams, thoughts and visualizations arenât phenomenally conscious experiences, but your externally generated experiences are phenomenally conscious? I donât think this is compatible with what most would intend by âphenomenal consciousnessâ, and if this is what Lau (or Gemini?) wants to do, heâs working on a totally different problem from pretty much everyone else.
Seems like a strawman. Strong illusionists donât conclude that simplified representations are false. They conclude that there are no phenomenal properties, and any representations as if there are phenomenal properties are therefore misleading (illusions).
Thanks, Michael.
Do you mean you recommend against sharing them, even if the alternative is not sharing anything (I agree sharing them, and examining them would be better)? Sharing them could lead to them being corrected to the extent they are not accurate, as you just did, which is good? Maybe you mean they should be shared with a quick disclaimer that the claims were not checked.
I do not recall Lau discussing the above in the few articles and podcasts I checked. I do not think Lau would reject that people often regard consciousness that way.
I believe the section is trying to say that phenomenal consciousness exists. I agree this would not contradict illusionism. It would only contradict eliminativism.
I would say dreams and thoughts are phenomenally conscious experiences. I am not sure what you mean by âexternally generated experiencesâ.
I meant that phenomenal consciousness is real in the sense ârealâ is used in common language, but I see now this is not what realism is about. âRealâ as used in common language can mean something that exists. However, illusionists acknowledge that phenomenal consciousness exists. They just argue it is not real in the sense it does not have the classic qualia properties. I seem to agree with this.
Having a better look into Frankish (2016) helped me understand what illusionism is actually about. I had only skimmed it quickly, and it is easy to misinterpret it on a quick skim because some terms like âexperienceâ are used in arguably unconventional ways.
Would you do anything differently if you endorsed weak illusionism/ârealism as defined in Frankish (2016) instead of (strong) illusionism?
Re AI, itâs a special case of Brandoliniâs law:
LLMs make it easy to produce a lot of plausible-looking but wrong or misleading claims quickly, and sharing them without checking yourself (or at least running another LLM to do that) puts the burden on others to correct.
I think you should make sure the claims are checked before sharing.
You could do this yourself, or run another critical LLM pass on the answer, e.g. a prompt like âExhaustively verify and ruthlessly critique the above answer.â
The critique pass itself may turn up false positive critiques, and you might want to run a third pass to verify the critiques and write a new balanced answer.
(I agree that itâs also important to tag that itâs LLM output, and at least read it over yourself, which you did.)
This is true of weak illusionists, but not strong illusionists. Strong illusionists reject phenomenal consciousness. Specifically they reject phenomenal properties; they accept that consciousness exists.
Other than argue about it, probably not much, assuming functionalism and materialism/âphysicalism of some kind thatâs compatible with artificial sentience.
FWIW, my impression is that there are ways to interpret âwhat-itâs-likeâ or âsubjectiveâ consistently with strong illusionism, as, e.g. the specific patterns of discriminations a brain makes or what information is accessible to it (or its attention or its global workspace). I think Frankish himself has done this later.
But âphenomenalâ and âqualitativeâ seem harder to pin down precisely in purely physical, discriminatory and informational terms in a way that I think would be broadly acceptable after critique and reflection to realists (including weak illusionists).
This also applies to many comments written by people? In general, whenever something is shared, one could invest more time to make it more accurate. However, to me it seems hard to come up with general rules about how much time to invest vetting claims.
I think sharing long texts produced by LLMs is often fine even if they were not fully read by the person prompting, basically for the same reasons that it is fine to share long text produced by people even if the person sharing them did not fully read them.
I have the impression many people, including academics, use âconsciousnessâ and âphenomenal consciousnessâ interchangeably, and without wanting to take a stance on illusionism or realism. I think this is one reason strong illusionists saying they reject phenomenal consciousness is sometimes misinterpreted as them rejecting the illusion of phenomenal consciousness.
If someone had a pattern of fabrication and very poor understanding (and apparent confidence) like LLMs often do if used uncritically, I would be annoyed with them and possibly do any of the following:
Tell them to read and review more carefully, look for opposing arguments, etc..
Downvote such comments (and I very very rarely downvote).
Stop engaging with this person, because it wastes my time and may encourage them to waste othersâ time.
I would request you check it yourself or at least run LLM critique passes, because of high rates of hallucination and other errors by LLMs.
Similarly, if there was an author who was just as consistently bad as LLMs are, and you shared their work uncritically like this, Iâd recommend the same.
My impression is that many people, including academics, use âconsciousnessâ and âphenomenal consciousnessâ interchangeably, but they do so implicitly rejecting strong illusionism, would reject strong illusionism if asked directly, and typically donât understand strong illusionism. Maybe many are open, though, Iâm not sure.
Makes sense. I just think what is âpoor understandingâ is often sufficiently contentious for one to have a high bar for preferring not sharing over sharing with little verification. I also tend to default to let people decide the extent to which they want to engage with something.
Makes sense. I was not clear. However, by âwithout wanting to take a stance on illusionism or realismâ, I meant many do not have a good picture of what strong illusionism means (and may be conflating it with eliminativism), and therefore are not taking a stance of the correct version of illusionism. I was doing this to some extent.
Verification is often easier than generation.
You may be interested in the article A science of chimeras? The implications of illusionism for non-human consciousness research. The abstract is below.