Honoring the dualist intuition at full strength — and showing where the inference to a non-physical mind fails
What this chapter does. The sense that consciousness could not be merely physical is not a confusion to be scolded out of people — it is a response to something real, felt most sharply by the most careful thinkers. This chapter takes that feeling seriously enough to state it at full strength before showing where the inference from it goes wrong. Wave the feeling away and you have not earned your conclusion; sit with it honestly and you can see exactly where the step from I can’t yet explain this to therefore it isn’t physical quietly fails. It diagnoses what makes the explanatory gap feel unbridgeable, and shows that the feeling tracks a feature of how we describe experience rather than a second realm of facts. By the end you should be able to feel the pull of dualism and see exactly where the inference from that pull to a non-physical conclusion goes wrong — and what genuine residue survives once the bad inference is removed.
The chapter’s main argument moves in five steps:
- The Hard Problem at full strength (§I). Nagel’s bat and Chalmers’s gap, stated so they bite.
- The diagnosis (§II). What makes the gap feel unbridgeable — the difference between a missing fact and a missing way of grasping a fact.
- Zombies and the limits of conceivability (§III). Why “I can conceive of it” does not deliver “it is possible,” and what the conceivability claim quietly assumes.
- What dualism gets right (§IV). The real datum the dualist has hold of, preserved.
- Living with the residue (§V). What remains after the bad inference is removed, and why it isn’t a second substance.
Two load-bearing arguments run to full length in the back matter, for readers who want them: the conceivability trap (Appendix B) — the explanatory gap and the zombie argument anatomized together as one inference that builds its own bridge; and Inverted Earth (Appendix A) — the hardest version of the challenge, which tries to pry felt character apart from representational content, and the price the identity claim pays to answer it. The knowledge argument (Mary’s Room) has its own chapter immediately before this one — Chapter 9 — and is not re-argued here. The thesis defended here is the one argued in Chapter 7; this chapter shows why it keeps meeting resistance.
Consider the bat.
Thomas Nagel asks us to imagine what it would be like to be a bat — navigating by echolocation, building a sonic picture of the world from the echoes of its own high-pitched cries.1 We can imagine a lot about bats. We know they have nervous systems, that they process acoustic information at remarkable speed, that their brains map spatial relationships from returning sound waves in ways neuroscience has made increasingly precise. We can imagine, in some thin sense, hanging upside down in a dark cave. What we cannot imagine, Nagel says, is what it feels like — from the inside, from the bat’s own point of view — to experience the world as a moving, three-dimensional sound-map. There is a fact about what that experience is like. We simply cannot access it from the outside.
This is the point Nagel wants to make, and it cuts deep. The problem isn’t that we lack data about bats. It’s that all the data in the world — every neuron mapped, every acoustic pathway charted, every behavior explained — describes the bat from a third-person perspective, as an object in the world. But experience has a first-person character. It presents itself from somewhere, to someone. And that somewhere, that to-whom, seems to fall clean outside anything science’s objective vocabulary can reach.
Notice that Nagel isn’t trafficking in mysticism. He isn’t claiming that bats have souls, or that experience floats free of the physical world in some ghostly parallel realm. His claim is more modest and, for that reason, more difficult to dismiss: there is something it is like to be a bat, and no third-person description fully captures what that something is. The subjective character of experience — the what-it’s-likeness — seems to resist reduction to the terms physics and neuroscience use to describe everything else.
Most people, on reflection, find this compelling. And they should. Something genuinely important is in view here. The mistake — and it does eventually become a mistake — comes later, when we try to explain what we’ve noticed.
§I. The Hard Problem at Full Strength
David Chalmers gave the phenomenon its most influential contemporary name. In a 1995 paper that launched a thousand dissertations and made life considerably harder for physicalists everywhere, he distinguished the easy problems of consciousness from the hard problem. The easy problems — explaining how the brain integrates information, produces attention, controls behavior, discriminates stimuli and reacts to them appropriately — turn out to be not quite easy, but tractable. Neuroscience and cognitive science can, in principle, crack them. The hard problem is different: even if we solved every easy problem completely, we still seem to face a further question. Why does any of this feel like anything at all? Why doesn’t all that neural processing happen in the dark — efficiently, functionally, without a single flicker of subjective experience? That it doesn’t happen in the dark seems obvious. Why it doesn’t seems utterly mysterious.
This is the dualist intuition at full strength — felt, before it gets diagnosed.
Chalmers formalized the intuition through the conceivability of philosophical zombies. A zombie, in this technical sense, is a physical duplicate of a human being — molecule for molecule identical — but with no subjective experience. The lights are off inside. All the neural processing happens; all the behavior results; but nothing feels like anything. The zombie walks, talks, reports having experiences, and acts in every way indistinguishable from a conscious being. It just isn’t conscious.
Is a zombie conceivable? To most people, yes — with effort, it seems possible to imagine the physical duplicate without the inner experience. And Chalmers argues that conceivability provides evidence for possibility: if zombies are coherently imaginable, they are metaphysically possible. And if zombies are metaphysically possible — if a physical duplicate could lack consciousness — then consciousness cannot be identical to any physical state. It constitutes something over and above the physical. Dualism of some kind follows.
The argument is clean. It is also less secure than it appears — and its insecurity traces directly to the picture diagnosed in the previous chapters.
§II. The Diagnosis: What Makes the Gap Feel Unbridgeable
Joseph Levine coined the phrase explanatory gap in 1983 to describe a phenomenon distinct from, and less dramatic than, Chalmers’ Hard Problem.2 Grant for the moment that consciousness involves nothing beyond the physical. Grant even that pain amounts to some neural process — C-fiber stimulation, in the textbook caricature, or some more complex story neuroscience eventually delivers. A puzzle still seems to linger. We can explain why that neural process produces avoidance behavior, why it recruits attention, why it integrates with memory. What we seem unable to explain is why it hurts. The physical story, however complete, leaves the felt character of the experience hanging in the air, unaccounted for. Levine’s gap lives in our understanding rather than in the world. It troubles even committed physicalists.
Chalmers took Levine’s gap and hardened it into an ontological claim. The gap, on his telling, does not merely register a shortfall in our concepts; it tracks a real division in the furniture of the world. Consciousness, on this reading, falls outside the physical altogether. The Hard Problem then becomes the problem of explaining why physical processes give rise to any experience whatsoever, given that the two appear to belong to fundamentally different categories.3
The Hard Problem feels hard — the gap feels unbridgeable — because both sides of the gap have been misdescribed. On one side stand physical processes, described in entirely third-person, functional, structural vocabulary. On the other side stands phenomenal experience, described as inner, private, intrinsic, self-contained: as qualia in the philosopher’s sense, as the contents of the inner theater diagnosed in Chapter 3.
Once experience gets described that way — as inner phenomenal items whose intrinsic properties could in principle obtain independently of any representational relation to the world — no physical process could plausibly produce it. Of course a gap opens. You have stipulated that the two sides occupy entirely different categories. The mystery follows from the description, not from the phenomenon.4
But strong representationalism — developed across Chapters 5 to 6 from the transparency of experience — contests precisely that description. Phenomenal character is not an inner self-standing item at all. It is a way a system represents its environment — the representational content of perceptual and bodily experience, made manifest from the inside. Experience involves the world in its very constitution. It directs itself, by its nature, outward. The physical processes that realize experience are the processes by which an organism represents its environment, and those representations, lived from the inside, simply are the experience of that environment. There exists no further inner item to account for, because the supposed inner item never figured in any honest description of experience to begin with.
Once that description shifts, the gap narrows. It does not vanish — I will not pretend the Hard Problem evaporates entirely, and §V honors what survives — but it narrows considerably. What looked like an unbridgeable chasm between physical processes and inner phenomenal items turns out to depend on a picture the previous chapters have already dismantled. Strip the picture away and the chasm shrinks to a hard scientific question: how does a physical system come to represent its environment in the perspectivally rich way that conscious organisms do? Hard, certainly. A reason to multiply ontologies, no.
§III. Zombies and the Limits of Conceivability
Return to the zombie argument. Its key premise is that zombies are conceivable. But conceivability arguments carry a hidden assumption: that what seems conceivable from the armchair reliably tracks what is genuinely metaphysically possible.
This assumption should make us nervous. Conceivability and possibility come apart — water’s seeming free to be something other than H₂O, heat’s to be something other than molecular motion, the morning star’s to be other than the evening star, each dissolving the moment the science arrived. What the armchair can conceive turns out to be no guide to what the world can contain.
The question for zombies is whether the conceivability of a physical duplicate without experience reflects a genuine metaphysical possibility or reflects a limitation in how we conceptualize physical processes and experience. If experience amounts to representational content — if it directs itself constitutively at the world — then a physical duplicate would, by virtue of having the same physical processes, have the same representational relations to the environment, and therefore the same experiential character. The zombie isn’t really conceivable once the nature of experience is correctly understood: what seemed conceivable was a consequence of thinking about experience the wrong way — as inner phenomenal items that could in principle be present or absent independently of any physical story.5
I want to be careful here. This is not a knockdown refutation of the zombie argument. A committed dualist will say that even after all the representational content is in place, it remains conceivable that the subjective character is absent. And I cannot simply assert otherwise without begging the question. What I can say is that the conceivability of zombies depends heavily on accepting a particular picture of what phenomenal character is — the inner-theater picture — and that once that picture is questioned, the conceivability becomes less clear. The zombie argument inherits the assumption that phenomenal character could float free of physical processing because it presupposes that phenomenal character involves something beyond representational content. That presupposition is exactly what is at issue.
§IV. What Dualism Gets Right
Dualism persists not because philosophers are confused but because it captures something real. Let me be honest about what it gets right before explaining where it goes wrong.
It gets right that experience has a first-person perspective. Experience happens for someone, to someone. The bat’s echolocation presents the world to the bat. The bat occupies a point of view. This first-person character cannot be reduced away without loss, and any adequate account of consciousness must preserve it.
It gets right that objective description leaves something out — not something mysterious over and above the physical, but something perspectival. The third-person vocabulary of physics and neuroscience describes the world from no particular point of view. Experience has a point of view. These are different modes of description, and switching between them involves a genuine shift, not just a change of vocabulary. Nagel was right about this.
It gets right that consciousness matters. That the world contains subjective experience — that there is something it is like to be a bat, or a human being, or presumably many other organisms — is not a trivial addendum to the physical story. It is one of the most remarkable features of the universe as we know it, and any honest physicalism must take it seriously rather than explaining it away.
Where dualism goes wrong is in its explanation of these genuine observations. From the fact that experience has a first-person character, dualism infers that experience comes equipped with properties belonging to a separate ontological realm. From the fact that objective description leaves the perspectival out, it infers that the perspectival must be non-physical. From the fact that consciousness matters, it infers that consciousness must be something beyond the physical processes that realize it.
These inferences don’t follow. The first-person character of experience is real, but it is the first-person character of a physical organism representing its environment from its particular embodied position in the world. The perspectival is left out of objective description not because it belongs to a separate realm but because objective description is designed to abstract away from particularity — that is its function, its virtue, and its limitation. Consciousness matters because experience is genuinely a form of world-engagement, not because it is made of ghostly stuff.
§V. Living With the Residue
Even after all this, something remains. Not a Hard Problem in Chalmers’ full sense, but a genuine puzzle about why the physical universe contains anything like a first-person perspective at all. Why should any arrangement of matter represent the world to itself? Why should there be any inside view?
I find this question genuinely hard. I don’t think anyone has answered it fully, and I’m suspicious of those who claim to have done so. But it looks very different once the inner-theater picture is dismantled. The Hard Problem in its most dramatic form — why physical processes give rise to inner phenomenal items with self-standing intrinsic properties — dissolves, because the description of experience it depends on is mistaken. What remains is a more tractable question: why certain physical systems have a perspectival relationship to their environments, why representing the world from the inside feels like anything at all.
That question may ultimately find a scientific answer, or it may represent a permanent limit on third-person explanation. I lean toward the latter, not because consciousness is non-physical but because perspective is inherently first-personal and third-person description is inherently impersonal. There may be no third-person explanation of why there is a first-person perspective not because the first-person perspective is made of different stuff but because the question asks one mode of description to capture something that is, by its nature, available only from the other mode.
This is not defeat. It is the correct shape of the problem — stripped of the misleading pictures that made it look worse than it is, and stripped of the misleading pictures that made it look easier than it is. Dualism points at something real. Physicalism, handled carefully, can honor what it points at without following it into a second ontological realm.
Something it is like to be you, reading this: real, physical, directed at the page, and thoroughly of this world. The bat is welcome to its own version.
Chapter Summary
This chapter honored the dualist intuition at full strength — Nagel’s bat, Chalmers’s Hard Problem, the zombie — then diagnosed why it keeps winning: both sides of the explanatory gap get misdescribed, with inner phenomenal items bearing self-standing intrinsic properties on one side. Once phenomenal character consists in representational content, the gap narrows and the zombie loses the picture it depends on; dualism is right that experience has first-person character, wrong to infer a separate ontological realm. The book turns next to the sophisticated rival that grants all of physics and relocates consciousness into the intrinsic natures physics leaves blank.
Notes
- Thomas Nagel, “What Is It Like to Be a Bat?”, Philosophical Review 83 (1974): 435–450. Nagel’s paper presents not an argument for dualism but a case for the irreducibility of subjective facts to objective description. His conclusion remains agnostic about the metaphysics: he does not claim that consciousness lies outside the physical, only that our current conceptual resources cannot explain the relationship. The paper sometimes gets recruited for anti-physicalist purposes it was never designed to serve. Nagel develops a more nuanced position in The View from Nowhere (New York: Oxford University Press, 1986), where the perspectival character of experience comes into sharper focus as a genuine feature of reality compatible with physicalism. The point about the first-person perspective as a real feature of consciousness that objective description cannot fully capture remains correct and important — the question, taken up below, concerns whether it supports dualism or merely reveals a limitation intrinsic to third-person description. Strong representationalism, properly understood, answers Nagel’s concern: experience is constitutively perspectival (it presents the world to an organism), and that perspectival character obtains within nature without requiring any non-physical ingredient. ↩
- Joseph Levine, “Materialism and Qualia: The Explanatory Gap,” Pacific Philosophical Quarterly 64 (1983): 354–361. Levine introduces the explanatory gap as a gap in our understanding rather than in the world, leaving open the possibility that physicalism remains true even though we cannot see why a given physical process gives rise to a given phenomenal character. The distinction between Levine’s epistemic gap and Chalmers’ ontological Hard Problem (footnote 3) often gets elided in subsequent literature; keeping it clear matters for evaluating which arguments succeed. ↩
- David Chalmers, “Facing Up to the Problem of Consciousness,” Journal of Consciousness Studies 2 (1995): 200–219; and The Conscious Mind: In Search of a Fundamental Theory (New York: Oxford University Press, 1996). Chalmers hardens Levine’s explanatory gap into an ontological claim: physical processes and phenomenal properties belong to genuinely different categories of fact, with consciousness falling outside the physical. The Hard Problem, on this reading, no longer registers a shortfall in our concepts but tracks a real division in the furniture of the world. The chapter resists this hardening; the diagnostic move appears in footnote 4. ↩
- The diagnostic move here — that the Hard Problem inherits its force from a particular description of phenomenal character rather than from phenomenal character itself — owes its shape to the strong representationalist tradition. Michael Tye, Ten Problems of Consciousness: A Representational Theory of the Phenomenal Mind (Cambridge, MA: MIT Press, 1995), and Fred Dretske, Naturalizing the Mind (Cambridge, MA: MIT Press, 1995), develop the identity version of intentionalism on which phenomenal character is (a specific kind of) representational content — not merely supervenes on it or is necessitated by it, but is numerically identical to it. This book endorses that identity claim. On the strong intentionalist view, the apparent gap between physical processes and phenomenal properties dissolves into a more tractable gap between physical processes and world-directed representational content — the latter itself a physical, naturalizable phenomenon. The identity is precisely what closes the zombie argument: if phenomenal character just is representational content of the right embodied, world-directed kind, there exists no further phenomenal ingredient a physical duplicate could lack. A zombie with the same physical processes, and therefore the same representational relations to its environment, would have the same phenomenal character — not as a contingent matter but as a consequence of what phenomenal character is. Angela Mendelovici notes that strong intentionalism of this stripe sits awkwardly with phenomenal-concept strategies but answers conceivability arguments at a deeper level, by contesting the description of phenomenal character those arguments presuppose; see The Phenomenal Basis of Intentionality (New York: Oxford University Press, 2018). ↩
- The conceivability-to-possibility inference receives critical examination in Brian Loar, “Phenomenal States,” Philosophical Perspectives 4 (1990): 81–108, where Loar argues that phenomenal concepts form a special class of recognitional concepts that directly pick out physical states, explaining the conceivability of zombies as an epistemic phenomenon without ontological significance. See also Christopher Hill, Sensations: A Defense of Type Physicalism (Cambridge: Cambridge University Press, 1991). Chalmers has replied to these responses at length; see “Does Conceivability Entail Possibility?”, in Conceivability and Possibility, ed. Tamar Szabó Gendler and John Hawthorne (Oxford: Oxford University Press, 2002), 145–200. My own response does not rely on phenomenal concepts but on questioning the picture of phenomenal character that makes the conceivability claim seem secure in the first place — a strategy that converges with phenomenal-concept replies on the conclusion (zombies are not genuinely possible) while diverging on the route (their conceivability fails because their description of phenomenal character was wrong from the start, not because phenomenal concepts have a special structure). ↩