Chapter 19: Functionalism — The Wiring Diagram

Why the right organization is necessary for a mind, and why it is not enough.

What this chapter does. You now live among machines that behave, in narrow stretches, exactly as though someone were home — and a tempting thought follows behind, so natural most people never notice it is a philosophy: get the organization right and the mind comes along for free, no matter what the system is built from. This chapter takes that thought, gives it its proper name and its full strength, and shows where it stops one step short. The thought is functionalism, and the chapter grants it everything it is owed before saying a word against it: it freed the mind from any one kind of stuff, and it was right to. Then the case against its central claim — that organization suffices — runs through Ned Block’s homunculus-nation, a system wired exactly like a brain in pain and built out of a billion people with radios, where the wiring is perfect and the lights seem to be off. The diagnosis is the chapter’s load-bearing move: what the perfectly wired system lacks is not an inner glow a soul would supply but states that are about anything — and that missing aboutness is fixed by a body’s causal traffic with a world, not by tightening the diagram. The verdict is close but incomplete: keep everything functionalism got right, add the world it left out. The chapter runs in six moves — functionalism at full strength (§I); why the machine question comes from substrate independence, not dualism (§II); Block’s nation and the diagnosis (§III); the three hardest replies, ending with Chalmers’s organizational invariance, the strongest in the literature (§IV); what the verdict does and does not commit us to (§V); and the hand-off to the Chinese Room and the rest of Part Four (§VI).

We now live among machines that behave, in narrow stretches, exactly as though someone were home. They answer the question, finish the sentence, console the griever at two in the morning when no human is awake to do it. And a tempting thought follows hard behind, so natural that most people who have it never notice they are having a philosophy: if a system is organized the right way — the same inputs producing the same outputs, the same inner states feeding into one another in the same pattern — then it cannot matter what the system happens to be built from. Get the organization right and the mind comes along for free. The hardware is incidental. The wiring diagram is the whole story.

That thought is the subject of this chapter, and it deserves better than the quick dismissal it usually gets in books like this one. It has a respectable name — philosophers call it functionalism — and for several decades it was the closest thing the philosophy of mind had to an orthodoxy.¹ It earned that standing. Functionalism solved a real problem that had stumped better-known theories, and any account of the mind that simply ignores what it got right will be the poorer for it. So I will do something that may look like a tactical error: state the wiring-diagram picture at its full strength, grant it everything it is owed, and make it as attractive as its ablest defenders have made it — before saying a word against it. A view worth answering is worth answering at its best.

Here is the question the chapter turns on. Does getting the organization right suffice for a mind? Not is organization relevant — of course it is — but is it enough, all by itself, so that any system wired the right way would thereby think, perceive, and feel? The functionalist says yes. I am going to argue no: the right functional organization is necessary for a mind but not sufficient, and the thing it leaves out is not a ghost, a glow, or a spark of the non-physical. What it leaves out is a world. The strong functionalist thesis has strong defenders, though, and the best of them have anticipated the objection I am about to press; the chapter has to earn its verdict against the cleverest reply on offer, which we take head-on in §IV, where it very nearly works.

§I. What Functionalism Got Right

Start with the part of the view that is true — there is a great deal of it, and the objection later lands harder once we have conceded everything the functionalist deserves.

For most of the twentieth century the philosophy of mind hung between two unhappy options. The behaviorist said a mental state was just a disposition to behave — to believe it is raining is to be disposed to carry an umbrella — and this foundered on the plain fact that no behavior follows from a belief by itself, only from a belief teamed up with a whole cast of other mental states. The umbrella goes up only if you also want to stay dry, notice the clouds, and have not decided the walk is worth a soaking. The identity theorist said, instead, that a mental state simply is a brain state: pain is the firing of C-fibers, full stop. This was tidier, but it bought its tidiness at an embarrassing price. If pain just is C-fibers firing, then a creature without C-fibers — an octopus, a Martian, anything whose nervous system took a different evolutionary road — cannot, as a matter of definition, be in pain. And that is not a discovery about octopuses. It is a prejudice about plumbing.

Into this impasse stepped Hilary Putnam, a philosopher restless enough that he would later become functionalism’s most distinguished defector — he built the house and then, on reflection, moved out of it, which tells you something about both the house and the man.² Putnam’s founding move defined a mental state by its role, not its material. A pain is not a particular kind of stuff in the head. A pain is whatever state in a creature gets caused by bodily damage, produces wincing and avoidance and the nursing of the injured part, and interacts with that creature’s beliefs and desires in the way pain does — keeping it off the sprained ankle, teaching it to mind the stove. Define the state by the job it does, by its place in the causal traffic of inputs, outputs, and other inner states, and you have set the mind free of any particular substrate. The octopus can be in pain. So could the Martian. So — and here is where the whole question of this final Part of the book comes alive — could a machine, provided only that its internal organization plays the same causal role.

This is multiple realizability: the same mental state realized in many different physical systems, the way the same melody can be played on a piano, a cello, or a synthesizer, or the same program can run on machines with nothing physical in common.³ David Lewis — who made a career of regimenting common sense into theory precise enough to argue with, and who trusted our everyday talk about minds further than most philosophers dared — sharpened the picture into a method: gather up everything we ordinarily take to be true of a mental state, the whole web of its causes and effects, and define the state as whatever occupies that role, leaving the question of what occupies it open for science to fill in differently in different creatures.⁴ And the picture is almost certainly right in what it affirms. There is no good reason to think only carbon can mind. To insist that minds require neurons specifically, and could never be built any other way, mistakes the one example we happen to have for the only example there could be — the identity theorist’s chauvinism dressed up as hard-headedness. Functionalism earned its long dominance by saying this first and saying it cleanly, and nothing in this chapter takes it back. When I argue against functionalism, I am not arguing that mind needs a special stuff. That door, once functionalism opened it, stays open.

So the wiring-diagram picture does not start out looking naive. It starts out looking like the sophisticated, anti-superstitious, broadly scientific view — the natural theory of mind for anyone who wants minds to be part of the natural world without being chained to one kind of meat. That is exactly why it repays a careful answer rather than a quick one.

§II. Where the Machine Question Really Comes From

Almost everyone misplaces the source of the worry we are chasing.

The reflex is to make it Cartesian. Machines are mere matter; minds are something more; you cannot get the more out of the mere. The machine question, on this telling, pits materialists against dualists, and your answer to “could a computer be conscious?” follows straight from your answer to “is the mind physical?”

This is a mistake, and an instructive one. The thesis that licenses “a computer running the right program could be a mind” is not dualism. Dualism, if anything, makes machine minds harder, not easier — if thinking takes an immaterial soul, the engineers are out of luck no matter how good the wiring. The thesis that actually opens the door to the machine is functionalism’s own: substrate independence. If a mental state just is a functional role, and anything that does the causal job can fill that role, then no principled reason bars a silicon system from filling it. The path to “the machine could think” runs not through Descartes but through Putnam. It is the most resolutely physicalist, anti-dualist theory of mind on the menu that makes the strong machine claim live.⁵

This changes what a good answer has to look like. If you think the worry about machine minds is really a worry about dualism, you will try to answer it by being more of a physicalist — by insisting, against the dreamers, that the mind is just the brain doing its physical business. But that reply concedes the war while disputing a battle, because the strong-AI claim was never dualism’s child. It was physicalism’s. To meet it you cannot out-physicalist the functionalist; you have to show that organization, the physical thing the functionalist points to, leaves something out — and leaves it out without smuggling a soul back in to fill the gap. That is a narrower and harder task, and it is the one the rest of the chapter takes up. We are not asking whether minds are physical. We are asking whether being physically organized in the right way is the whole of what a mind is.

§III. The Machine That Breaks the Picture

Now the case against sufficiency. It is owed, like much of the trouble functionalism has had to weather, to Ned Block — a philosopher who has spent his career insisting, against every tidy reduction the field has offered, that something real always gets left out, and who builds his counterexamples with an engineer’s relish.⁶ Block’s machine is designed to grant functionalism everything and then ask the one question the theory cannot dodge.

Picture it. We are going to build a system whose functional organization is, by stipulation, an exact duplicate of a human brain in pain — every input, every output, every internal state-transition matched. Only we will build it out of people. Take the population of a large nation, a billion or so, hand each citizen a two-way radio and a simple rule about whom to call and what to say when called, and have them collectively implement, for one hour, the precise functional profile that your brain instantiates when you stub your toe.⁷ One person stands in for each neuron; the radio calls stand in for the synaptic signaling. Set it running. The inputs are right. The outputs are right. Every internal state-transition is right. By functionalism’s own official criterion — a mental state just is the occupant of a functional role — the system as a whole is in pain.

Now look at the thing whole, the entire chattering continent of it, and ask the plainest question in philosophy: is anyone home? Does the nation hurt? Not “is some citizen uncomfortable,” but does there exist, over and above the billion busy people, a further subject who is feeling the toe?

The intuition that there is not is very hard to shake. And that — stated with care — is the trouble. Functionalism, by its own definition, must say the nation feels pain; most of us find that nearly impossible to believe. Block called these absent qualia cases: the functional wiring is perfect and the lights appear to be off.⁸ Notice what the case does and does not claim to show. It does not show that functionalism is obviously false, or that the bullet cannot be bitten — we will hear from the bullet-biter in a moment. It shows something more modest and more useful. Functionalism specified the form of the organization, exhaustively, and said nothing whatever about what, if anything, fills it. It handed us the grammar of a mind and forgot that grammar is not yet meaning.

Here the diagnosis matters more than the intuition, because there are two ways to react to the dark nation and only one of them is any good. The first — the reflex — is to conclude that what the wiring lacks is a non-physical extra: a glow, a spark, an inner light that physical organization can never capture and that only a soul could supply. Resist that. The homunculus-nation gives the dualist no comfort, and reading it as a proof of the soul throws away its real lesson. What the nation actually lacks is not an inner glow but an outer grip. Its states are not about anything.⁹ A real pain represents a particular damaged body — this knee, mine, in trouble now. A real perception of red represents a feature of a surveyed world. The nation’s frantic radio traffic represents nothing at all; it is a pattern of signaling with no answering object, a syntax with no semantics, a role played in a vacuum. The citizens are passing notes whose “meaning” lives entirely in the stipulation that set the exercise up — which is to say, nowhere in the system itself.

And meaning of that kind is not something you can install by tightening the diagram, because what a state means was never fixed inside the system to begin with. The system’s history of commerce with a world fixes it — this state reliably caused by that worldly feature, in a body that could be harmed and a creature with something at stake in how the encounter came out.¹⁰ This is why the repair for functionalism is not a retreat into dualism but an advance into the world. Add genuine, world-directed content to the functional story — let the states actually track damage, actually register red, through the right embodied causal engagement with the things they are about — and the absent-qualia worry loosens its grip. A system whose states really are about the world, in the way an embodied animal’s states are about the world, is no longer a nation passing meaningless notes; it is a creature in touch with its surroundings. The missing ingredient was never a ghost in the machine. It was the machine’s missing world.

I should mark the seam plainly rather than paper over it, since the honest version of this argument wears its debts on its sleeve. The move from “the nation’s states are about nothing” to “the nation feels nothing” needs a bridging premise, and the premise is the identity claim this book defended in Part Two: the felt character of an experience is not a further thing laid on top of its representational content but consists in that content — content of the right embodied kind.¹¹ Grant that, and to settle what a state is about is already to settle whether there is anything it is like to be in it. The two halves of the case then weld together: the nation’s states are about nothing, so there is nothing they are like, so no one is home. And what makes content the right kind — what turns mere correlation into genuine aboutness — is the business of Part Three, which this Part has been quietly drawing on; functionalism’s gap and meaning’s grounding turn out to be the same topic seen from two sides.

§IV. Three Replies, Taken in Order

We take the three strongest objections in turn, weakest to strongest. The third in particular is the most serious challenge in the literature to everything §III just argued.

Objection 1: Bite the Bullet

The orthodox functionalist does not flinch. The nation feels pain — yes, however bizarre that sounds — and your refusal to believe it is not an argument but a prejudice. Our intuitions about who can feel were trained on creatures with faces, fur, and eyes; they are parochial witnesses, hopeless judges of radio networks and continental populations. We once found it obvious that the sun went round the earth. The strangeness of the homunculus-nation is a fact about the narrowness of our imaginations, not about the distribution of consciousness. Bite down, and the case loses its teeth.¹²

Reply to Objection 1. There is something to this, and I want to grant it before I deny it: intuition-pumping over exotic systems really does prove little on its own, and a reader who found the whole case resting on “but it just seems weird” would be right to walk away. If absent qualia were the entire indictment of functionalism, the bullet-biter could chew his way out and we would have to let him. But the bare reply has a hole in the middle of it, and the hole is exactly where an explanation should be. The functionalist owes us an account of why organization brings experience along — why running the right pattern should kindle a felt life rather than merely shuffle states. Biting the bullet supplies no such account. It restates the sufficiency thesis in a defiant tone of voice and dares us to deny it. That is not an answer to the case; it is a refusal to feel its force. We are owed the mechanism, and “obviously there must be one” is not the mechanism. The intuition the bullet-biter dismisses was pointing at a real and unpaid debt: organization specified to the last detail still leaves the question of aboutness wide open, and so leaves the question of experience open too.

Objection 2: You Have Only Renamed the Mystery

A subtler critic grants that the nation’s states are not about anything and presses a different worry. Fine — but “world-involving content” is just a new label on the old gap. You say the wiring lacks aboutness; I say aboutness is precisely what we do not understand, and you have explained the obscure by the more obscure. Worse: you claim the missing ingredient is not a non-physical extra, yet “genuine meaning” looks suspiciously like the very thing the dualist was gesturing at, now wearing a naturalist’s coat. Either grounding is more functional organization — in which case you are still a functionalist and owe the nation its pain — or it is something extra, in which case welcome back to dualism.

Reply to Objection 2. This dilemma is the sharpest thing in the neighborhood, and the answer to it carries the chapter, so I will be careful. Take the second horn first. Is world-involving content “something extra” in the dualist’s sense? No — and the difference is not a dodge. The dualist’s extra is intrinsic: an inner quality, a private glow, something that could in principle be present or absent with every physical and relational fact held fixed. Grounding is the opposite kind of thing. It is relational and wholly physical: a matter of which worldly features have, over a causal history, fixed which of the creature’s states, in a body with stakes. Nothing immaterial is added; what is added is a world, and the creature’s tracked, embodied traffic with it. You can be the most thoroughgoing physicalist alive and insist on grounding, because grounding is made entirely of physical relations — just relations the functionalist declined to count.

Now the first horn, which is the one with real bite. Isn’t grounding just more function — long-armed function, reaching out to include the causal links to the world? And here I concede the words while denying the victory, because the concession is fatal to the thesis under examination, not to mine. Yes: you may, if you like, call world-involving content a species of “wide” or “long-arm” functional role, the functional story enlarged until it takes in the states’ causal-historical bonds to the very things they are about.¹³ Call it that. But notice what has happened. Machine functionalism — the thesis this chapter set out to test, the thesis that makes the strong-AI claim live — individuates mental states by their internal role: inputs, outputs, and relations among inner states, the whole specification readable off the wiring diagram with the world switched off. That thesis is the one Block’s nation satisfies to the letter, and it is the one that fails. The moment you reach past the diagram and let the world in to fix content — the moment two internally identical systems can differ in what their states mean because they are plugged into different environments with different histories — you have conceded the precise point at issue: organization alone, the wiring with the world bracketed, does not suffice. Whether we then file the enriched view under “wide functionalism” or under “world-involving representationalism” is a question of shelving, not of substance. The wiring diagram, read as the functionalist reads it, is no longer the whole story. That is all the chapter needs, and the relabeling hands it over.

Objection 3: Organizational Invariance (Chalmers)

The strongest reply does not come from the functionalist who bites the bullet or quarrels over labels. It comes from a philosopher who is not a functionalist at all, and who grants from the start that the dark nation feels perfectly possible — then argues, with real ingenuity, that it cannot be. David Chalmers is a property dualist; he thinks consciousness floats free of the physical facts, which is very nearly the opposite of the view I am defending.¹⁴ That is exactly what makes his argument the one to beat. He has every motive to say organization leaves experience out — and on independent grounds he says it does not.

Chalmers’s thesis is the principle of organizational invariance: any two systems with the same fine-grained functional organization have qualitatively identical conscious experiences. Hold the organization fixed and you hold experience fixed, whatever the system is made of.¹⁵ His argument for it is a thought experiment as careful as any in the literature. Imagine replacing your neurons one at a time with functionally identical silicon chips, each chip doing exactly the signaling job of the cell it displaced, the rest of your brain none the wiser. Now ask what happens to your experience as the swap proceeds. One possibility: it fades by degrees — the red of the tomato drains toward gray, the pain dulls — while you, your functional profile untouched, go on sincerely reporting that everything looks and feels exactly as before. You would be radically, systematically wrong about your own present experience, your judgments and your qualia coming apart. The other possibility: experience holds steady and steady and then, at some single decisive chip, blinks out all at once — which seems arbitrary past belief. Chalmers finds both options absurd enough to reject, and concludes that experience must track organization smoothly all the way down. If it does, the homunculus-nation, sharing your organization, shares your pain. The lights are on.

Reply to Objection 3. This is a serious argument, and the one thing that will not answer it is to stamp one’s foot and re-insist that the nation is dark. The intuition Chalmers grants and then defeats cannot defeat him back. The argument has to be answered where its force actually lives — and its force, looked at squarely, comes from a feature of the thought experiment that the conclusion then quietly forgets. The neuron-replacement story preserves the subject’s entire causal embedding in her world. The silicon isomorph is still in a body. It is still tracking the same damage, still answering to the same surveyed environment, still carrying states with the same causal-historical content as the cells they replaced. Grounding is held fixed throughout; only the stuff is swapped.

Of course experience rides along through that replacement. I insist that it does. The content rides along — same world, same body, same tracked features — and on the identity claim the felt character consists in that content, so the felt character rides along too. Chalmers has demonstrated something true and considerably narrower than he takes himself to have shown: replace a creature’s parts while preserving its grip on its world, and you preserve its mind. With that I agree entirely. What he has not shown is that organization alone, severed from that grip, suffices — and the severed case is precisely Block’s nation, a functional duplicate with no causal-historical traffic with anything its states could be about. The invariance principle is safe exactly where grounding is held fixed and silent exactly where grounding is stripped away. Chalmers built his isomorph inside a living body in a real world and never took it out; Block built his in a vacuum. No wonder one seems lit and the other dark — they differ in the very thing that, on the view defended here, settles the matter. Pressed to its full strength, then, the deepest objection ends up confirming the diagnosis it was meant to refute: what the wiring needs is not an inner light but a world, and wherever the world is genuinely present, the wiring was never the whole story by itself.

§V. What the Chapter Commits Us To — and What It Does Not

It is worth saying flatly what has and has not been argued, because a thesis of this shape attracts three predictable misreadings.

The verdict on functionalism is close but incomplete, and the emphasis falls on both words. Close: functionalism is right that the mind is not tied to a particular substrate, right that mental states are individuated by more than their physical composition, right that multiple realizability holds and that the old identity-theory chauvinism about neurons is a prejudice. None of that is withdrawn. Incomplete: it specifies internal organization exhaustively and says nothing about the world-relations that fix what the organized states are about, and so leaves both aboutness and — by the identity claim — felt character undetermined. The repair is not subtraction but addition: keep everything functionalism got right and add the world it left out.

First misreading: this is substrate chauvinism after all — you are sneaking back the view that minds need a special kind of stuff. It is the reverse. The chapter keeps multiple realizability and uses it against the bullet-biter as freely as the functionalist ever did. What a mind requires is not a particular material but a particular relationship — embodiment, world-directedness, a causal history one is answerable to — and that relationship is as available, in principle, to silicon as to carbon. Nothing here says the machine cannot mind. It says what minding would take, and the answer names a relation to a world, never a roster of approved materials.¹⁶

Second misreading: you have admitted a non-physical ingredient. No. Grounding is built entirely from physical relations — causal commerce between a body and its environment over time. One world, not two; nothing reaches in from outside nature. The whole point of routing the repair through grounding rather than through an inner glow was to keep the account inside the physical world while still denying that the bracketed wiring diagram exhausts the mind.

Third misreading: the felt-character claim is a separate, optional add-on. It is the hinge, and the chapter leans on it openly. Without the identity claim of Part Two, the slide from “the nation’s states are about nothing” to “the nation feels nothing” would be a non sequitur, and the whole argument would stall at aboutness without reaching experience. With it, the two come as one: fix what the states represent and you have fixed whether there is anything it is like to be the system. The chapter thus stands on Part Two’s result and reaches forward to Part Three’s — what makes content the right kind — which the next chapters redeem by name. I flag the debt rather than hide it; an argument that announces where it is leaning is sturdier than one that pretends to stand on its own in midair.

§VI. The Door to the Room

Functionalism was the right place to begin this final Part, because it is the view that makes the whole machine question live. State it at full strength, as we have, and the question “could a computer have a mind?” stops being idle and starts being urgent: if organization suffices and organization is substrate-neutral, then a sufficiently well-wired machine simply is a mind, and the only remaining work is engineering. The burden of this chapter has been to show that the “if” fails — that organization, the wiring with the world switched off, is necessary but not sufficient, and that what it omits is not a soul but a world.

That verdict sets the terms for everything that follows in this Part. The chapters ahead are, each of them, a different assault on the same sufficiency thesis we have just denied. The Chinese Room, next, attacks it at the level of meaning itself — arguing that running a program, manipulating symbols by their shapes, never adds up to understanding them, that syntax is never sufficient for semantics — which is the homunculus-nation’s lesson restated for the symbol-processing machine and pressed right to the question of meaning.¹⁷ After that we ask whether computation is even a determinate fact about a physical system or something an observer reads onto it; whether simulating a mind is the same as realizing one; and whether a system trained on the mere form of language ever touches the world its words are about. Different machines, different arguments, one target throughout — and in the Part’s closing chapter, the constructive question this chapter has been deferring: not what a mind lacks when it is only well-wired, but what, concretely, a machine would have to have — a body, a world, a history it answers to — to cross the line into genuine understanding. The wiring diagram leaves out a world. The rest of the Part is about what it would take to put one in.

Chapter Summary

Functionalism freed the mind from any one kind of stuff, and it was right to — but Block’s homunculus-nation shows that the right organization is necessary and not sufficient: wire a billion people exactly like a brain in pain and the lights still seem to be off. What the perfect diagram leaves out is not a glow a soul would supply but states that are about a world, and that aboutness consists in a body’s causal traffic with its surroundings rather than in anything internal to the wiring. The verdict is close but incomplete, and it hands the argument to the Chinese Room.

Notes

Block opens “Troubles with Functionalism” by observing that functionalism “may now be dominant,” then immediately notes that the label covers several distinct projects — reformulations of behaviorism, mind-machine analogies, applications of empirical psychology, and arguments about mind-brain identity. The version pressed and answered here is the strong one: each mental state is identical to a functional state, a state defined by its causal relations to inputs, outputs, and other inner states. Nothing in the argument turns on the machine-table formalism specifically; it turns on the claim that functional role is sufficient for mentality. Ned Block, “Troubles with Functionalism,” in Perception and Cognition: Issues in the Foundations of Psychology, ed. C. Wade Savage, Minnesota Studies in the Philosophy of Science, vol. 9 (Minneapolis: University of Minnesota Press, 1978), 261–325. ↩
Hilary Putnam, “Psychological Predicates,” in Art, Mind, and Religion, ed. W. H. Capitan and D. D. Merrill (Pittsburgh: University of Pittsburgh Press, 1967), reprinted as “The Nature of Mental States.” The functional-state hypothesis individuates mental states by their place in a causal network — sensory input, behavioral output, relations to other states — rather than by physical composition; this is the sense in which a mind, on the strong reading, is like a program, specifiable independently of the hardware that runs it. Putnam later abandoned functionalism, in part on the externalist grounds congenial to the diagnosis offered here: see Representation and Reality (Cambridge, MA: MIT Press, 1988). ↩
Multiple realizability is the part of functionalism this chapter keeps in full. The objection is not that mind requires a special substrate — that would trade functionalism’s error for the identity theory’s chauvinism, or for a biological essentialism about neurons. The claim is narrower: functional organization is necessary but not sufficient, and what must be added is not a substrate but a relation to a world. Substrate independence survives; organization-as-sufficient does not. ↩
David Lewis, “Psychophysical and Theoretical Identifications,” Australasian Journal of Philosophy 50, no. 3 (1972): 249–258, and “Mad Pain and Martian Pain,” in Readings in Philosophy of Psychology, vol. 1, ed. Ned Block (Cambridge, MA: Harvard University Press, 1980). Lewis’s Ramsey-sentence method defines mental terms as whatever occupies the causal roles specified by the platitudes of common-sense psychology, allowing the occupant to differ across creatures (the “mad pain” and “Martian pain” cases dramatize the two ways role and realizer can come apart). The device makes precise the sense in which functionalism keeps the role fixed while leaving the realizer open. ↩
The point that substrate independence, not dualism, is what underwrites the strong-AI claim is worth pressing because it is so often missed: the popular framing pits “machines are just matter” against “minds are more than matter,” when in fact it is the thoroughly physicalist functionalist, not the dualist, who holds that the right-running machine would think. The chapter’s strategy follows from this — one cannot answer functionalism by being more of a physicalist, only by showing that organization, the physical thing it points to, leaves out the world-relations that fix content. ↩
Block has pressed the same conviction — that reductive theories of mind keep omitting something real — across absent qualia, the distinction between phenomenal and access consciousness, and the “mental paint” objection to representationalism. The homunculus-nation is the version aimed squarely at functionalism’s sufficiency thesis. Block, “Troubles with Functionalism” (1978). ↩
Block’s “homunculi-head” and “Chinese nation” (or “China-body system”) cases, both from “Troubles with Functionalism” (1978). The scenarios are built so that the system’s functional organization is, by stipulation, identical to a human’s, isolating the single question of whether that organization suffices for phenomenal consciousness. The detail that one person stands in for each neuron, and radio signals for synaptic transmission, is doing real work: it makes vivid that every functional fact can be satisfied while the materials and their arrangement are nothing like a brain’s. ↩
Block distinguishes absent qualia (the system satisfies the functional specification but has no phenomenal experience at all) from inverted qualia (two systems functionally identical but with systematically swapped experience). A parallel pressure comes from inversion, but I leave it aside here: turning the inverted spectrum against a representational theory of consciousness, rather than against functionalism, opens a separate and harder debate, taken up in Part Two’s treatment of the identity claim. Against functionalism’s sufficiency thesis specifically, the absent-qualia case is enough. For the standard taxonomy see the Stanford Encyclopedia of Philosophy, “Qualia,” §3. ↩
This is the load-bearing move, and it cuts against the dualist as sharply as against the functionalist. The dualist and the bare functionalist share a premise — that the only thing the wiring could be missing is an intrinsic felt quality — and disagree only about whether the wiring supplies it. The diagnosis here rejects the shared premise: what the homunculus-nation lacks is not an inner glow but intentional content, states that are genuinely about a world. This is the project’s general anti-reification line applied to the machine case — “consciousness” and “qualia” name relational, world-involving achievements, not inner substances that organization either secretes or fails to. ↩
This is semantic externalism brought to bear on the theory of consciousness: what a state represents is not settled by anything internal to the system at a time but by its causal-historical relations to the environment, so two internally identical systems can differ in what their states are about because they are embedded in different worlds. This is precisely the resource functionalism lacks — it specifies internal role exhaustively and is silent about the world-relations that fix content. The positive account is developed in Part Three; see especially the chapters on teleosemantics and on semantic externalism. ↩
The identity claim defended in Part Two, in the Tye–Dretske lineage: phenomenal character consists in representational content of the right embodied, world-directed kind, rather than being a further qualitative residue laid on top of content. Michael Tye, Ten Problems of Consciousness: A Representational Theory of the Phenomenal Mind (Cambridge, MA: MIT Press, 1995); Fred Dretske, Naturalizing the Mind (Cambridge, MA: MIT Press, 1995). It is this claim that licenses the inference from about nothing to nothing it is like, and so welds the chapter’s two halves together. The claim is doing acknowledged work here; the chapter inherits it rather than re-arguing it. (In the scholarly register the relation is the numerical identity of phenomenal character with content of the right kind; the main text renders it as “consists in.”) ↩
The orthodox bullet-biting reply: the nation does feel pain, however strange, and the contrary intuition is parochialism — our judgments were trained on creatures with faces and are unreliable witnesses about radio networks and populations. There is something to this; intuition-mongering over exotic systems proves little on its own, and were absent qualia the whole case against functionalism the bullet-biter might chew his way out. But the reply supplies no account of why organization should bring experience along; it stipulates the sufficiency it was asked to defend. For the bite-the-bullet line see the Stanford Encyclopedia of Philosophy, “Qualia,” on functionalist responses to the China-body system. ↩
“Long-arm” or “wide” functional role — functional individuation extended to include the system’s causal relations to items in its environment — comes in more than one form: Gilbert Harman’s “long-arm” conceptual-role semantics lets the relevant roles reach all the way out to worldly referents, while William Lycan’s wide homuncular functionalism builds the environmental relations into the functional characterization itself. Both hold, in their different ways, that a sufficiently externalist functionalism can absorb the content the narrow version misses. The concession in the text is deliberate: one may classify world-involving content as wide functional role if one likes. The point is that doing so abandons exactly the thesis under test — machine functionalism, which reads mental states off the internal organization with the world bracketed, and which Block’s nation satisfies. Once content depends on the environment, internally identical systems can differ mentally, and organization-alone is conceded insufficient. The dispute then becomes terminological. Gilbert Harman, “(Nonsolipsistic) Conceptual Role Semantics,” in New Directions in Semantics, ed. Ernest LePore (London: Academic Press, 1987); William Lycan, Consciousness (Cambridge, MA: MIT Press, 1987). ↩
David J. Chalmers, The Conscious Mind: In Search of a Fundamental Theory (New York: Oxford University Press, 1996). The dialectical importance of the argument lies in Chalmers’s position: he is a property dualist who holds that consciousness does not logically supervene on the physical, yet he defends organizational invariance on independent grounds. The chapter engages it as the strongest pro-invariance case precisely because it is not a functionalist house argument — its proponent has every incentive to deny that organization carries experience and nonetheless affirms it. ↩
Chalmers, The Conscious Mind, ch. 8, “Absent Qualia, Fading Qualia, Dancing Qualia.” The principle of organizational invariance holds that “any two systems with the same fine-grained functional organization will have qualitatively identical experiences.” The fading- and dancing-qualia arguments support it via gradual neural-replacement scenarios: a subject whose qualia faded or danced while functional organization held constant would be radically and systematically mistaken about her own present experience, which Chalmers takes to be incoherent. ↩
The reply turns on a scope distinction the invariance argument elides. Chalmers’s neural-replacement scenarios preserve the subject’s entire causal embedding in her environment — the silicon isomorph remains in the same body, tracking the same worldly features, with the same causal-historical content — and so preserve exactly what fixes intentional content on the externalist view of note 10. Invariance therefore secures phenomenal constancy across substrate change with world-relations held fixed, which the present view affirms; it does not secure the stronger functionalist thesis that organization alone, abstracted from those relations, suffices, and Block’s nation is built to instantiate precisely that abstracted case. Cf. Michael Tye, Consciousness and Persons: Unity and Identity (Cambridge, MA: MIT Press, 2003), ch. 6, arguing that nothing in Chalmers’s argument rules out a system with the right organization but the wrong (or absent) grounding for consciousness. The invariance principle and the grounding diagnosis are thus consistent: the former is true where grounding is held fixed, silent where it is stripped out. ↩
John Searle, “Minds, Brains, and Programs,” Behavioral and Brain Sciences 3, no. 3 (1980): 417–457, where the slogan that syntax is not sufficient for semantics gets its canonical thought experiment, the Chinese Room. The deeper point that syntax is not even intrinsic to a physical system but assigned by an interpreter — which sets up the observer-relativity argument of the next chapter but one — is Searle’s in “Is the Brain a Digital Computer?,” Proceedings and Addresses of the American Philosophical Association 64, no. 3 (1990): 21–37. The present Part borrows Searle’s diagnosis of the gap while declining his further moral that the missing ingredient is the brain’s specific biological causal powers; the view defended here holds that it is grounding — the right causal-historical engagement with an environment — which is substrate-neutral. One can keep Searle’s gap without his biology. ↩

← Chapter 18: The Assembled Self

Contents

Chapter 20: The Chinese Room →