The Qualia Trap
Why Eliminativism undermines itself
The following is a novel critique of eliminativism, that I believe is quite robust and offers a new approach.
Eliminativism is a school of thought in philosophy of mind claiming there is nothing worth saying about consciousness besides what can be discovered via neuroscience or other physical sciences. While it remains fringe in academia, it has found significant popularity online as a way of circumventing the hard problem of consciousness.
The core tenet of this approach is that experiential concepts used in philosophy of mind, such as “qualia” or “phenomenal consciousness,” are crucially flawed. The eliminativist argues that we should discard these terms entirely. By doing so, we avoid the challenge of untangling the “hard problem” of consciousness, which is the problem of accounting for experience and subjectivity in a purely physical reality. Unfortunately, this position necessarily undermines itself.
The eliminativist is not merely like someone shouting “Don’t speak English in this room”; that would be a simple pragmatic contradiction. It is worse. The position is analogous to someone shouting, “Don’t use the English language in this room, because it is not sufficiently sensical here.”
If I shout “Don’t speak English” in English, I have disobeyed my own command, but my command might still be understood. But if I shout “Don’t speak English because it is nonsensical,” my very ability to convey that instruction using English falsifies the premise of the instruction. To articulate the ban, the eliminativist must use the very resource they declare too confused or nonsensical. This essay will demonstrate why this contradiction is inescapable.
The dynamic in question
To understand the trap, we must first understand what the eliminativist is actually policing. Eliminativists rarely claim that ordinary language references to experiences should be jettisoned entirely. They don’t believe a woman giving birth who screams “I’m in so much pain!” is speaking nonsense. They accept that references to pain, sadness, and taste are sensical within ordinary language usage.
The problem arises when we try to refer to these affairs within theoretical language games. Daniel Dennett, the most prominent proponent of this view, evidently wished to make this distinction. He did not deny the validity of ordinary talk about experience, but argued that experience-talk is founded on too much confusion to be in adequate shape for serious theory.
As Dennett asserted:
My claim, then, is not just that the various technical or theoretical concepts of qualia are vague or equivocal, but that the source concept, the ‘pretheoretical’ notion of which the former are presumed to be refinements, is so thoroughly confused that… it would be tactically obtuse—not to say Pickwickian—to cling to the term. [1]
He further sharpened this by stating:
Qualia’ is a philosophers’ term which fosters nothing but confusion, and refers in the end to no properties or features at all. [2]
Yet, crucially, Dennett distinguished this theoretical ban from ordinary usage. When he notoriously claimed, “Nobody is conscious” [3], he immediately added a footnote warning against taking this out of context, protecting himself from the charge of denying the meaningfulness of experience-talk in ordinary practice.
We can therefore map the eliminativist dynamic by distinguishing between two functions of experience-talk:
P1: Ordinary first-person experience-talk as a public practice and treated as sensical (e.g., “Ouch, that hurts”).
P2: Theoretical or philosophical uses of experience-talk treated as sensical (e.g., “This quale picks out the feeling of pain”).
The eliminativist policy is clear: P1 is acceptable, but P2 is spurious and must be excluded from serious discourse because it lacks sufficient clarity and meaning. The trap is set. The eliminativist wants to enforce a boundary between P1 and P2 based on meaningfulness. However, to enforce this boundary, one must be able to define it.
The formal argument
We can represent this trap formally to see precisely where the logic collapses. Although the trap springs whenever experience-talk is considered to be in inadequate conceptual shape for serious theory, for brevity and simplicity I will refer to this inadequacy as nonsense. On another note, know that any reliance on determinacy here is implicit in the eliminativist’s own assumption that a boundary can be set.
Let us define our terms:
O: Ordinary language as a public practice.
L: Serious theory.
E: Experience-talk treated as sensical.
P1: E in O.
P2: E in L.
B (Boundary Policy): Allow P1; ban P2.
J (Justification): E in L is nonsensical.
HL (Hinge Lemma): To apply B inside L, participants must be able to identify tokens as unacceptable.
The deduction:
Aim: To enforce B, with J as the stated motivation.
Requirement: By HL, applying B requires delineating instances that show what counts as P2. To distinguish P2, one must implicitly treat P1 as the “acceptable,” sensical kind of experience-talk that P2 operates in contrast to.
The Collapse: Applying B is a form of L. Because this application requires the implication of a sensical P1 to define the target, we create a new entity: P1L (Ordinary experience concepts treated as sensical and utilized within a theoretical context).
The Violation: Any E in L is by definition P2. Therefore, P1L is P2 (P1L ∈ P2). Hence B is breached by its application.
The Paradox: By J, P2 is nonsensical in L. Therefore, the application of B, which relies on P1L, hinges on nonsense according to J, given that P1L is P2.
By implicitly relying on an acceptable P1, the explanation constitutes experience-talk treated as sensical in serious theory, thus falling under the very category the rule polices against on the premise it is nonsensical. If the explanation is indeed nonsensical, the rule cannot be understood any better than the experience-talk in serious theory (P2) it is intended to police. If the explanation is sensical, then experience-talk can function sensically within theory, proving the justification false. The eliminativist is caught shouting in a language they claim doesn’t work.
Analysis
This brings us back to the specific analogy: “Don’t use English because it is nonsensical here.”
Imagine you are an eliminativist policing concepts that are not in adequate shape to be treated as sensical in serious theory. Someone asks, “Why is P1 acceptable and not P2?” You must explain that P1 represents ordinary language concepts (which are fine), but P2 represents those concepts dragged into a context where they lose coherence.
In giving this explanation, you are engaging in theoretical discourse (L). To make your explanation intelligible, you must refer to P1; you must invoke and treat as sensical something like “the everyday concept of pain.” You are relying on your audience understanding the meaning of “everyday pain” to grasp why the theoretical version is confused.
But here is the contradiction: If “everyday pain” (P1) can be treated sensically within your theoretical explanation to distinguish it from “qualia,” then experience-talk is sufficiently sensical to function in theory as sensical. You have just used it successfully!
If you insist that experience-talk is in inadequate shape for serious theory to rely on and treat as sensical, then your explanation for the ban, being an example of serious theory, relies on and treats as sensical something that by its own lights is in inadequate shape, thus contradicting itself.
The phlogiston objection
A critic might object that this argument proves too much. They might say, “We eliminated the concept of ‘phlogiston’ from chemistry by talking about it. We mentioned phlogiston to debunk it. Does that mean the elimination of phlogiston was self-defeating? Why can’t we treat qualia just like phlogiston?”
When scientists eliminated phlogiston, they did not attempt to preserve a “folk-phlogiston” (P1) while banning a “theoretical-phlogiston” (P2). They argued that the term referred to no actuality; it was factually empty. The boundary was simple: “The term ‘phlogiston’ is useless; do not use it.”
However, the qualia eliminativist does want to preserve P1. Dennett was clear that he thought ordinary talk of pain is useful. If the eliminativist treated qualia exactly like phlogiston, they would have to deny P1 as well. They would have to say, “When a woman in labor says she feels pain, her words are false and useless.”
Few philosophers are willing to bite that bullet. Because they refuse to deny P1, they must distinguish P1 from P2. This need to distinguish is what triggers the trap. The phlogiston eliminativist points to an empty set; the qualia eliminativist points to a useful, everyday token and says, “Don’t theorize about that.” But pointing at “that,” identifying the experiential role, is itself a theoretical act that smuggles the meaning of the experience-talk into the discourse.
The unicorn objection
Someone might point out that plenty of terms have meaning and are sensical but don’t pick out a real reference, such as “unicorn.” Why can’t the eliminativist say “qualia” is like that?
They certainly can, but in doing so, they abandon eliminativism. If a philosopher treats “qualia” like “unicorn” (sensical but without a real target), they are admitting that the concept is sufficiently coherent, clear, and intelligible to be deployed in serious theory; even if only to conclude that the count of such entities is zero. Zoologists can have a rigorous, coherent debate about why unicorns don’t exist, because the concept of a horse-like creature with a single horn protruding from its forehead is in perfectly adequate shape for such discourse; it just happens to have no instance in the world.
If “qualia” and its ilk are like that, then experience-talk in serious theory (P2) is sensical. If P2 is sensical, it is admissible. If P2 is admissible, then we are no longer talking about eliminativism. Moreover, the notion of a unicorn is a combination of concepts that pick out actualities; horses and horns. If P2 also picks out actualities, then eliminativism is false.
The “meaning is use” objection
A sophisticated critic might object that this argument relies on a crude view of eliminativism. They might argue: “The problem is not that theoretical experience-talk is meaningless nonsense. The problem is that philosophers misconceive experience-talk. They think it points to a private, inner object (qualia), when in fact, as Wittgenstein showed, experience-talk derives its meaning solely from public use. The eliminativist is simply correcting this misconception, not banning the talk.”
To correct the misconception, the eliminativist must identify it. They must say, “You are wrong to believe that your pain has a private, qualitative nature that exists beyond your behavior.” To formulate this correction, the eliminativist must refer to the concept of “private, qualitative nature.”
But to distinguish such notions from mere gibberish or random noise, the eliminativist must appeal to it as something distinct. They must articulate how it seems to the user. In doing so, they are treating this seeming as a comprehensible token within their theoretical diagnosis.
This reveals that the concept is sufficiently sensical for theoretical use (specifically, for the theory of eliminativism). If the concept were truly just a nonsensical confusion with no semantic reference (like “the zorble of the number seven”), the eliminativist could not distinguish the “illusion of qualia” from any other gibberish. The fact that they can target the specific “seduction” of the hard problem proves that the concept has semantic content in serious theory, refuting their own premise.
The “it’s just mention, not use” objection
The eliminativist might argue, “I am not using the concept of pain in serious theory; I am merely referring to the word ‘pain’ or the concept ‘qualia’ to ban them from serious theory, much like saying ‘Don’t speak French’ is an English sentence.”
If the sorting criterion is merely the word “qualia” (orthography), then the rule is just word-policing; I could simply invent a new word for the same thing. If the rule is substantial, it must target the meaning or role of the concept.
To identify that role, to say “don’t treat first-person reports as theoretical data,” requires recognizing the relevant experiential role. To identify that target, one must distinguish the nonsensical usage of the token (P2) that motivates the confusion. This distinguishing necessarily implies the acceptable usage (P1) and treats it as sensical.
Therefore, to apply the rule effectively, one must reference and treat as sensical the “acceptable” usage of the token, even if only in the form of an implication. However, this reference occurs within a theoretical context; the very context where such usage is declared unacceptable and nonsensical. By successfully referencing the token to explain the rule, the eliminativist proves that the concept can function sensically in theory, thus falsifying the justification for the ban. So, instead of it being like saying “Don’t speak French”, which merely issues a rule against one language by using another, it’s like saying “Don’t use English because it’s nonsensical here.” It breaks its own rule and falsifies its premise.
The meta-language objection
One can move the rule to a meta-language, or classify the ban as “metaphilosophy,” but a bridge back to practice either reinstates examples and role conditions or reduces to mere word-policing. One can recast the rule as therapy or house style, but then it loses philosophical relevance. With any of these approaches, the application of a boundary still either becomes vacuous or drags in the experiential referencing it forbids.
Contrast: eliminativism about physicalism
To verify that this argument is robust, consider “eliminativism about physicalism.” Could one argue that “physicalism” is a nonsensical term that should be banned?
Yes, but there is a key difference. The physicalism eliminativist can simply say there is never anything but nonsensical usages of “physicalism.” They do not need to save an “ordinary” version of physicalism. They can dismiss the term entirely.
For the physicalism eliminativist, there is no “acceptable” usage to collapse into an “unacceptable” one. The rule is fairly vacuous perhaps, but not self-defeating. Qualia eliminativism falls into the trap precisely because it tries to conserve the trees (ordinary experience-talk) while cutting down the forest (theoretical experience-talk).
Conclusion
The “Qualia Trap” reveals that eliminativism regarding experience is logically unstable. It is distinct from other forms of elimination (like phlogiston or zorble) because it attempts to bifurcate a concept (experience) into an acceptable ordinary version and an unacceptable theoretical version.
This bifurcation is impossible to maintain. To apply the distinction, one must invoke the meaning of the ordinary concept within a theoretical context and treat it as sensical. This action transforms the acceptable P1 into the forbidden P2, violating the very rule being established, and making its application nonsense (or at the very least in inadequate conceptual shape) according to its own premise. Or, it falsifies said premise by being sensical. The boundary is therefore either inapplicable or self-defeating. It is a paradox that arises because any attempt to escape experience naturally fails, like a raindrop trying to rinse off its wateriness.
I leave you with this final image: A woman giving birth shouts, “I am having a minimal common-sense P1 experience of pain, which can be used in philosophical theories as a token to build highly theoretical arguments about how overly abstract, theory-laden concepts of experience are nonsense!”
I am very grateful to Prof. Sam Coleman for his invaluable feedback regarding this argument.
First published by Essentia Foundation.
References:
Daniel C. Dennett, “Quining Qualia,” in Consciousness in Contemporary Science, ed. A. J. Marcel and E. Bisiach (Oxford: Oxford University Press, 1988), 42–77.
Ibid.
Daniel C. Dennett, Consciousness Explained (Boston, MA, Little, Brown and Company, 1991), 406.
I'm going to spend some more time on your essay but, But it sounds like you're on to something. There are physicists who are (a decreasing number), that say if you can't measure something, it doesn't exist. They are deterministic by nature and anything that can't be wrapped around by language or mathematics must be unreal. They have persistently tried to just disprove entanglement, complementarity and non-locality for 100 years, since the philosopher/physicists at the Solvay conference first proposed them. As you know, it's called the Copenhagen interpretation. It's as much a statement of philosophy of Mind in the end, as a model of quantum physics and reality.
I would love to see you post an article on "The Physicalism Trap," taking an eliminativist view.
This would be much more akin to the biologist pointing out there are zero examples of unicorns.
Simply, there has never been and can never be a single demonstration of ontologically self existent physical stuff.
There is no possible conceptual way to describe such stuff, unless (like the experientialists Dennett and Wittgenstein wrongly thought were confused about experience-talk) you have someone with a thought capacity so throughly damaged that they actually belief the abstract concepts (pointer readings, as Sir Arthur Eddington calls them) scientists derive from the study of sensory experience are real.
As a clinical psychologist who has conducted over 3000 psychological evaluations requiring more than 10,000 diagnoses, I can tell you there is a name for such a belief in the DSM (Diagnostic and Statistical Manual, the book used to describe mental illnesses).
On the other hand, it is rather poignant, I think, to observe that a psychiatrist once told religious scholar Huston Smith that from a purely technical view, a belief in physicalism qualifies for a diagnosis of Delusional Disorder, Grandiose Type.