How, in layman's terms, should this Conifold argument against illusionism be interpreted?

Question

There's a discussion about philosophical zombies and illusionism going on in The Symposium, which is the main chat room for Philosophy on Stack Exchange and Conifold posted this a day or two ago (the Chat timestamp is Wed 14:02):

I am skeptical of illusionism (about consciousness), but not for the usual reasons that it is counterintuitive or “crazy”.

I just do not find the motivation minimally attractive. It is quite transparent that the only connotation of “illusion” of interest is that what is so called can be set aside. It is designed to solve a metaphysical problem for physicalism, which is seen as struggling to accommodate qualia. I submit that this perceived problem is itself an illusion.

Let me explain. There is no reason why objects should not have qualities, 1-place relations. On the other hand, physics, and science generally, are designed to represent only structural information (about 2 and more place relations) for the simple reason that only it can be represented (modeled) and communicated. Qualities must be lost in any structural representation by their definition.

So we have a ready made metaphysical shelf (qualities) and a phenomenon begging to fill it (qualia). Yes, technically, qualia are something “science cannot explain”. But that is the real illusion. It is so only for the trivial reason that “explaining” is meaningless here, asking for structural models of qualities (which is what is meant by “explanation”) is non-sensical.

And even if we do manage to write them off as “illusion”, we will still have to explain why our world was so handicapped as to have 2+-place relations but not 1-place ones.

Let me call structural physicalism a position that all structural (non-quality) information is describable by physics/science. It achieves the exact same thing that illusionists want from the “illusion” without all the pain, qualia can be set aside as qualities and forgotten about. Existence of qualities does not threaten the physicalist program at all.

If we can map neural correlates of “conscious experience” why should we care that the map does not contain the accompanying ‘feels’? We will even be able to induce the ‘feels’ themselves in individuals by stimulation, just not cross-compare them because, again, ‘comparing’ qualities is non-sensical. Only structure can be divorced from its original carrier, lifted and transferred, i.e. ‘compared’.

And the pain that structural physicalism avoids is great indeed. Until recently, illusionists ignored that their use of “illusion” does not fit into the paradigmatic examples of optical or cognitive illusions and did not explain what they mean beyond vague metaphors. To use “illusion” as normally understood a seeming is needed as a background, making the seeming itself into an illusion does not work. Just as making into an illusion existence of someone who wants to assert it does not work.

This is what Wittgenstein called “bewitchment of our intelligence by means of language”, pushing words beyond their usefulness. Kammerer recently acknowledged this as “illusion meta-problem”. His solution is that the “illusion” does not really mean what one might think it means, other than having the desired side effect of setting something aside.

Essentially, he has a “phenomenality operator” that does what qualities would do, “coloring experience”, only it is anchored in the subject rather than in objects, hence “illusion”. This strikes me as more Kantian than physicalist, and more a theory of qualia than of them being an illusion in any normal sense. And a very convoluted solution to a non-existent problem.

I tried to understand, but I couldn't, and Google seems not to have heard of the phrases I took from here and used as search strings.

But then by searching "place relations in philosophy", Google started to understand. I found this: https://plato.stanford.edu/entries/relations/ and this:https://en.wikipedia.org/wiki/Relations_(philosophy)

Unfortunately, I still don't understand Conifold's argument, mainly because I can't understand the SEP and Wikipedia articles.

In layman's terms, what is this argument saying?

Answer 1

Conifold is a philosophically sophisticated thinker who runs laps around me, to be sure. I would agree that the term 'illusion' is meaningless, and Conifold is taking to task illusionism since it seems fundamentally resolve around a false analogy. Let's see walk through:

I am skeptical of illusionism (about consciousness), but not for the usual reasons that it is counterintuitive or “crazy”. I just do not find the motivation minimally attractive. It is quite transparent that the only connotation of “illusion” of interest is that what is so called can be set aside. It is designed to solve a metaphysical problem for physicalism, which is seen as struggling to accommodate qualia. I submit that this perceived problem is itself an illusion.

So, physicalists tend to want to reduce or explain mentality and mind in terms of the "physical things" of science. This is reductionism in the moderate form because it posits that the mind is somehow related to the brain and body such that the mind doesn't exist but for the brain and body. In the extreme, eliminative materialists claim that consciousness, a purported property of minds to many, is an "illusion". Daniel Dennett is such a claimant of illusionism, probably the most famous. So, what exactly does it mean to call consciousness an "illusion"? It's not the literal sense of the term, such as a what a magician does. It's a ontologically loaded term that implies something unreal about consciousness. Does consciousness exist? To a nominalist, probably not, but to a realist yes. What 'exist' means, is a long-running conversation.

There is no reason why objects should not have qualities, 1-place relations. On the other hand, physics, and science generally, are designed to represent only structural information (about 2 and more place relations) for the simple reason that only it can be represented (modeled) and communicated. Qualities must be lost in any structural representation by their definition. So we have a ready made metaphysical shelf (qualities) and a phenomenon begging to fill it (qualia). Yes, technically, qualia are something “science cannot explain”. But that is the real illusion. It is so only for the trivial reason that “explaining” is meaningless here, asking for structural models of qualities (which is what is meant by “explanation”) is non-sensical.

So, qualia, qualities, and explanation are all philosophically challenging topics. What is being stated is that qualia qua 'units of experience' and qualities which are taken to be fundamental metaphysical statements about reality are related but not the same. To see red, and to say, "An apple has the property or quality of being red" are different matters, because one is purporting to be a conscious experience regarding red, and the other is purporting to be a conscious expression regarding red as seen as a distinction in use-mention. When one asks "What is the difference between seeing red and saying 'Apples are red'?", one is now engaged in an act of explanation that seeks to clarify the relationship between experiencing redness (qualia) and talking about entities that have redness (qualities). Philosophical explanation, taken simply, is that one can adequately explain the necessities or consequences of one term in terms of the other. In other words, how do they relate? Science is equipped to talk about qualities in physicalist terms (wavelengths, neurons, the visual cortex, etc.), but not so much about consciousness, which while some presume can be reduced to the physical, strictly speaking in metaphysics, can be presumed not. That is, in philosophy, one can admit phenomenalism.

That creates a tension between dualists who believe that consciousness is fundamental to the universe (think anthropic principle) and physicalists who believe that consciousness is derivative (or in the extreme just discount it entirely as an illusion). A proponent of illusionism might claim that qualia make no sense because there is a problem in explanation, but Conifold dissents and says that the problem is with the language of illusionism because it speciously presumes something about explanation of qualia in terms of qualities, and that presumption is a function of the problems of the language surrounding 'qualities', 'qualia', and 'explanation', and not the object of discourse itself.

And even if we do manage to write them off as “illusion”, we will still have to explain why our world was so handicapped as to have 2+-place relations but not 1-place ones. Let me call structural physicalism a position that all structural (non-quality) information is describable by physics/science. It achieves the exact same thing that illusionists want from the “illusion” without all the pain, qualia can be set aside as qualities and forgotten about. Existence of qualities does not threaten the physicalist program at all. If we can map neural correlates of “conscious experience” why should we care that the map does not contain the accompanying ‘feels’? We will even be able to induce the ‘feels’ themselves in individuals by stimulation, just not cross-compare them because, again, ‘comparing’ qualities is non-sensical.

Conifold then seeks to clarify why the original language that illusionism used is flawed; that it's better to relax the demand that explanation has to invoke an arity of predicates two or greater as opposed to having a single-place predicate, then it is to honor the demand and then have to declare the mental an 'illusion'. Why do we have to have to explain in terms of objective, scientific statements AND account for conscious description, because third-person and first-person descriptions of quality can't be compared since they're fundamentally in different categories and to do so would be a category mistake.

Only structure can be divorced from its original carrier, lifted and transferred, i.e. ‘compared’. And the pain that structural physicalism avoids is great indeed. Until recently, illusionists ignored that their use of “illusion” does not fit into the paradigmatic examples of optical or cognitive illusions and did not explain what they mean beyond vague metaphors. To use “illusion” as normally understood a seeming is needed as a background, making the seeming itself into an illusion does not work. Just as making into an illusion existence of someone who wants to assert it does not work. This is what Wittgenstein called “bewitchment of our intelligence by means of language”, pushing words beyond their usefulness. Kammerer recently acknowledged this as “illusion meta-problem”. His solution is that the “illusion” does not really mean what one might think it means, other than having the desired side effect of setting something aside. Essentially, he has a “phenomenality operator” that does what qualities would do, “coloring experience”, only it is anchored in the subject rather than in objects, hence “illusion”. This strikes me as more Kantian than physicalist, and more a theory of qualia than of them being an illusion in any normal sense. And a very convoluted solution to a non-existent problem."

The use of the term 'illusion', therefore, is fundamentally problemantic, because the sense in which it is applied is based on the category error, and that now, philosophers are debating what exactly it means to apply the term illusion to consciousness since it is a false analogy to the normal senses of illusion, and merely serves to discount the importance, prominence, or reality of the object to which it is applied.

Answer 2

Conifold's main argument could be framed as follows:

1. Illusion means that something seems to be the case but actually is not ontological reality.
2. Illusionism claims that mind, free will, and generally all qualia are illusions
3. Qualia literally are "how things seem to us", ie. the "seeming" is what we call qualia
4. This means that if Illusionism claims that qualia are illusions, it says that it seems like there are qualia but actually there are no qualia
5. This is a contradictio in adjecto since without qualia, nothing can "seem to be the case" in the first place, since if something "seems to be the case", it means that there are qualia
6. Therefore, illusionism about qualia is conceptually incoherent

That is why we should let go of the whole idea of "illusionism" and rather stick to other wordings that achieve the same but without the incoherence of presupposing something that we say does not exist, e.g. "structural physicalism".

The rest is about why structural physicalism does not need to be able to explain qualia (committing to a category error) as long as there is a possible mapping between the multi-perspective default position of the theory and the uni-perspective qualitative experience of the individual, e.g. through the ability to systematically induce qualia through physical manipulation and qualia repeatedly being linked to distinct physical patterns. Mapping is not the same as explaining, it is the ability to link entities from one category to entities from another category, ideally in a bidirectional way.

Answer 3

Conifold posted this:

[I]llusionism (about consciousness) ... is designed to solve a metaphysical problem for physicalism, which is seen as struggling to accomodate qualia.

[T]echnically, qualia are something “science cannot explain”. But that is the real illusion. It is so only for the trivial reason that “explaining” is meaningless here, asking for structural models of qualities (which is what is meant by “explanation”) is non-sensical.

• Qualia are something "science cannot explain".

• "Explaining" qualia is meaningless in the scientific/physicalism context.

• Illusionism is designed to solve a metaphysical problem for physicalism, that is, to explain what cannot be explained in the scientific/physicalism context.

---

Daniel Bonevac - Philosophy in One Lecture:

https://youtu.be/AycTgPJtBP0?t=101

[Philosophy] It's really all about the connection between the mind and the world. It's really all about the connection between our mind and reality.

Suppose you have a thought. I don't know how to draw a thought ... but imagine it is something very simple, for example, a triangle. This thought has a content, "That's a triangle." How do I know that I am actually having thoughts that correspond to the world? Or that the content in my mind corresponds to the content in the minds of others?

Helmholtz gives a cognitive model to answer this question when he argues that our depth perception and perception of moving objects in the world is generated by a cognitive process called an unconscious inference.

If I automatically think you have a conscious mind with reality in it then this thought too seems to arise either from the mind as idealism or if the mind links this to a structure in the body and brain then it seems to arise as an unconscious inference.

Dennet thinks life could operate without impairment of function on the principle of unconscious inference without ever becoming conscious and therefore consciousness and qualia in the mind are epiphenomenal by-products of the mechanistic cognitive models of the nervous system. But this is just a selective choice of what is necessary and sufficient biological function in his (addled) mind and brain. It is like the Buddha saying, "There is no self" which evokes my concept of self!

Answer 4

"And just as about a hundred years ago most scientists believed in a life force, in about a hundred years time few to no scientists will believe in qualia" (OP).

Rather than consciousness being another scientific paradigm or concept to overturn like life force, C is saying qualia were never in that domain. As long as something like life force has a structural description, it is susceptible to science. Qualia have yet to have a structural description, and life force never had a non-structural one.

Structural means knowledge of relations and structure. Even if there are nondivisible elements of structure, they can only be known by their relations (eg causal). Science is typically said to adopt some form of structuralism. Any experiment, thought or concrete, uses structural knowledge.

Not all knowledge is structural. Does A being A rely on any structure? I think not. Knowledge of a toothache does not depend on any knowing of brains and chemistry. But a diamond’s hardness does require experiment.

It doesnt matter that brains underlie quailia and brains are physical. The knowledge of quailia sensations depends on no other knowledge.

That is not how science investigates. It has no tools for non-structural knowledge.

Answer 5

I consider Conifold to be a gem who blesses this forum with thought-provoking posts. I often learn a great deal from them, as he brings forward relevant aspects of the some of the arcana of academic philosophic debates that I would never have found on my own. This is the case with his reference to structural physicalism, and to Ontic Structural Realism https://plato.stanford.edu/entries/structural-realism/#OntiStruRealOSR.

As that SEP article notes, Structural Realism is an effort to characterize the nature of reality as revealed by our scientific enterprise. The premise it uses is there is a physical world -- W1 -- which we experience as observations. Then we make theories around those observations, about what the physical world is like, so we infer what W1 is like indirectly, through experiences, and inference. However, our inferences are regularly wrong and we have repeatedly thrown out major portions of our physics. And we can also expect our current models of W1 to be wrong based on this history. So what is W1, if we are constantly wrong about it?

This leaves physicalism, the belief that W1 is all there is, on a very suspect foundation. And Illusionism, the claim that the experiences and the inferential process we use is a self-delusion -- makes even less sense.

In contrast, Structural Realism argues that the abstract structure of our theories is generally maintained, therefore we can treat that abstract structure as what is real, while the specific items of our theories, would NOT be real.

Conifold then inferred something he referred to as structural physicalism from this structural realism, that the physical is real as inferred from our theory structures. Searching for this concept, the only paper I uncovered was a critical discussion of the reasoning of a philosopher who has proposed structural physicalism. Here is the link: https://www.uni-muenster.de/imperia/md/content/wissenschaftstheorie/pdf-dateien/comment_on_lyre.pdf

As far as I can parse Conifold's post, he is arguing that Illusionism, by basically denying experience, cannot support any justification of W1's realism, nor our revisions of what we think W1 consists of. What he then argues for, is that one can construct a W1 theory of physicalism based solely on the structure of our theories. This W1 structural physicalism can accept that we have experiences as an aspect of W1 events.

But then he asserts that experiences can never be mapped structurally, so the qualia of experience would never appear in any W1 model. I think he is further assuming that "not structural" is the same as "not functional", which would then lead to qualia and experience being non-causal (by definition from non-functional), and this leads to epiphenomenalism relative to qualia. What I think he then arrives as is the "Something near Enough" to physicalism that Kim argues for. Physicalists under this conception can happily accept qualia, but then ignore them from that point on, as they are causally irrelevant.

Conifold would have to comment on the accuracy of this interpretation of his post.

I noted that I have a great deal of respect for Conifold, but I do not agree with what I THINK he is saying here.

To start with, I consider structural realism itself to be highly suspect. From the SEP article, the tidal locking of the Moon's face is cited as a structural commonality between Newtonian Gravity and Genral Relativity. This is a common CONSEQUENCE, but the intrinsic structure of gravity is radically different between Newton and GR. And neither has a coherent emergence model to even describe what the Moon is, or tides. And neither has a logic system to account for tiers of structure and therefore reality, nor how these tiers would interact. Scientific realism, which assumed absolutist reductionism, actually held that all such "higher level structures" were just shorthand approximations to actual reality, and were not real themselves! What this structural physicalism seems to imply is that "moon, tides, tidal locking" which are cited as the common structural elements are real, but both GR and Newtonian gravity are NOT real. I considered putting in "attractive force" into the list of reals, BUT in GR there isn't one, there is instead a distortion of the dimensions of space, no attractive force at all.

I am not convinced of several aspects of this approach. I suspect that the "structural similarities" do not fully capture the knowns that carry over between our theories. And I consider the "structural physicalism" that is inferred by this severely vitiated structural realism to look nothing like what almost any physicalist considers physicalism to be.

Further, I consider, if one wanted to make an ontic categorization of structural realism as a monism, that it would best be described as an IDEALIST ontology, not a physicalist one. Structural commonality is a purely logic-abased abstraction, and if you want to call that the real, it sure looks like it fits better into W3 of Popper/Frege's three world ontology. For a reference on three worlds, see: https://m.moam.info/three-worlds-the-tanner-lectures-on-human-values-pdf_647a088f098a9e8f5f8b45d3.html?utm_source=slidelegend

As a further critique, Popper had no difficulty outlining a structural process in his three world ontology, where W1 generates phenomena, we detect them subjectively in W2, then postulate thru w2 processes a W3 explanation for the phenomena, from which we infer what W1 consists of. This is a functional process that runs thru experience, and experience, qualia, are explicitly causal in it. I THINK that Conifold is intrinsically assuming this Popperian process, as are the thinkers behind Structural Realism, hence for any of them to hold that experience is not functional/causal/structural (the three terms all seem to be being used similarly here by Conifold, per my understanding) appears to be as self-contradictory as Conifold is arguing that Illusionism is.

Accordingly, the ontology that drops out from our scientific process is that which Popper argues for -- three worlds. Ontic triplism. Note also that ALL our data is subjective, and generated via qualia within world 2, so science HAS to deal with qualia from the outset. And there is no intrinsic problem with science developing theories about qualia, nor for qualia to carry out a functional role (which they certainty appear to do), nor for constructing a structure which includes qualia.

As an aside, a major motivation that appears to be behind structural realism is to find a way to justify that science is advancing in the accuracy of our models. Popper set out to prove this as well, and came up with a metric of verisimilitude of our theories. But critics showed that for any definition of verisimilitude, we cannot show that our newer theories are always more "true". Popper's student Lakatos also tried to come up with a method to calculate the progressivity/regressivity of his version of science, Research Programmes, and this methodology was likewise shown to be logically invalid. There is a lesson here. Empiricism CANNOT satisfy closed form logical criteria. It is intrinsically a pragmatic activity, not a logical one. I strongly suspect that the efforts to reduce science solely to structure -- will suffer from the same sort of failings as Popper and Lakatos' analytical methodologies.