My most recent essays argued that the zombie argument fails to refute physicalism. Here I examine the opposite assumption — that physics itself supports physicalism — and show that this too is far less secure than it appears.

Physicalism, defined here as the view that everything is reducible to physical processes, is often given a structural advantage that it hasn’t earned evidentially. A common claim is that science supports a purely mechanistic view of reality. This claim isn’t as convincing as it may appear. That is not to say that science supports a different view. It simply means we ought to drop the assumption that science plays for team materialism.

Science is metaphysically neutral — compatible with physicalism, dualism, panpsychism, or idealism, and decisive evidence for none of them.

Physics operates under a principle of methodological naturalism — the principle that science ought to prioritise physical causes for physical effects over any extra-physical explanations. This is the appropriate methodology for studying physical phenomena, but becomes a metaphysical assumption when extended to claim that only physical causes exist. That additional step goes beyond the evidence, and nowhere is this more apparent than in quantum mechanics.

The physical sciences are understandably looking for physical reasons, and therein lies the limitation.

You could comb a beach with a metal detector without finding any buried diamonds, even if the beach is littered with them. Your tool isn’t equipped to discover them even if, by pure chance, you do come across one or two. Quantum mechanics may be pointing in this direction, revealing glimpses of something the method wasn’t designed to detect. The measurement problem, the role of the observer, the non-locality of entangled systems — these don’t fit more comfortably into a purely mechanistic picture. Physical interpretations are sometimes given priority for methodological reasons, not because of their explanatory power.

There is no telescope that can observe what lies beyond the edges of the universe, and no microscope that can objectively observe inner experience. The tools of science are not designed to look for such things. That is not a criticism of science, just a description of it. In spite of this, quantum mechanics has generated findings that can sit as comfortably within a non-physical framework.

Subscribe now

Methodological naturalism is a commitment adopted prior to the evidence, not a conclusion reached through it. It encourages scientists to explain research observations in physical terms and so can be an obstacle to neutrality. Quantum phenomena can be accommodated by perspectives that treat consciousness as fundamental rather than emergent, yet these interpretations receive considerably less attention in the scientific community.

Entanglement and the Observer Effect

Quantum entanglement — what Einstein called “spooky action at a distance” — offers a first glimpse of this. When two particles interact they can become entangled, remaining directly correlated regardless of the distance between them. Measure one and the other responds instantaneously, even across vast distances. They are not transmitting information across space in any conventional sense. They remain in fundamental relation to one another in a way that classical physics— independently existing objects interacting through local causes — struggles to account for.

Physicists have endeavoured to fit these observations into a locally causal framework in ways that leave the deeper questions untouched. That the observation fits as neatly inside a view of reality where unity is more fundamental than separation remains a minority position.

The same is true of the observer effect. In the 1920’s and 1930’s the pioneers of quantum science observed particles in an indeterminate state of superposition prior to observation. The particles did not appear to occupy a specific location if no observation was taking place. They appeared genuinely indeterminate — not in an unknown location, but in no definite location at all — until observation resolved them into a specific state.

The particles of classical physics suddenly seemed less the fundamental building blocks than had been assumed. The language shifted to describe what was actually observed: a superpositioned wavefunction rather than a particle with definite properties. Observation seemed to collapse this wavefunction, forcing the particle to assume a specific location. One interpretation held that consciousness had triggered the collapse. This interpretation suggests that consciousness could be fundamental rather than emergent, a possibility that methodological naturalism is structurally predisposed to automatically discount.

Many pushed back. There is no way to separate the ‘observation’ of a scientific instrument from that of a living consciousness. Did the particle assume a position when the data was collected by an instrument, or would it remain indeterminate until a conscious observer looked at that data? There is no way of checking without introducing the variable of consciousness itself, which reintroduces the very variable under investigation. The question has not been settled empirically, and yet a majority of physicists answer that measurement triggers the collapse rather than the involvement of a conscious perceiver. This is a methodological commitment, not a metaphysical one.

A problem with this interpretation is illustrated by The Von Neumann Chain, a mathematical treatment of the measurement process. According to Von Neumann, nothing physical can explain a collapse of the wavefunction. Quantum mechanics applies to everything made of atoms — which, according to physicalism, is everything.

If the original particle is in superposition, then the detector that interacts with it becomes entangled with that superposition and inherits its indefiniteness. The same applies to any recording device, to neural signals, and ultimately to the brain state itself. No physical process within the chain can force a resolution — the Schrödinger equation simply carries the superposition forward through each link.

John von Neumann concluded that consciousness was capable of terminating this regress, not because consciousness is obviously non-physical, but because if the chain requires something outside the purely physical to bring about a definite outcome, then consciousness becomes an obvious candidate. It is perhaps the only identifiable element in the sequence that could be operating outside of the physical processes the equation describes. This strengthens the claim that consciousness may be non-physical without proving anything.

Methodological naturalism on the other hand assumes that consciousness is an emergent feature of physical processes, entirely reducible to brain activity. If so, consciousness is just another physical link in the entanglement chain with no special status. But if that’s the case, what remains to resolve the superposition? After all, particles are observed as occupying definite locations.

Is Decoherence Coherent?

Since the 1980s the dominant interpretations among physicists have centred on the principle of decoherence:

The particle in superposition becomes entangled with its environment — air molecules, photons, electromagnetic fields. The superposition spreads across this massive entangled system. To detect that the particle is still in superposition, we would need to observe quantum interference between all those environmental particles simultaneously, which is practically impossible. Superposition remains, just “diluted” across so many particles that we can no longer observe it. The particle appears to be in one place because we have lost access to the quantum information showing it is still superpositioned.

For practical purposes, decoherence explains everything physicists need to predict and manipulate quantum systems. It accounts for why macroscopic objects behave classically and why we never observe superposition at everyday scales. This is no small achievement — it solves the practical problem. But it only explains why we do not observe superposition continuing, not why we observe a definite position at all.

If the particle never actually collapsed — if it merely became unobservably spread out — then what determines which position we see when we look? Decoherence alone cannot fill this ontological gap. It describes the conditions under which collapse becomes undetectable, not the conditions under which collapse actually occurs. In its basic form it assert that the wavefunction continues in superposition indefinitely, but behaves classically for all practical purposes. This is an epistemological solution (we can’t observe it anymore) rather than an ontological one (it actually collapsed into one state).

The measurement problem reduces to a question decoherence does not answer: when, if ever, does the wavefunction actually collapse? Has the wavefunction collapsed at the moment of instrument detection, will it collapse at the moment of human observation, or does it never collapse?

Schrödinger’s point was precisely this — in our experienced reality a cat is either dead or alive, not both. If decoherence never produces actual collapse, that becomes hard to explain. If nothing ever actually collapses into definite states, then the table you’re sitting at, the words you’re reading, your sense of a continuous present moment — all of it becomes metaphysically mysterious.

Decoherence makes the observations fit classical expectations by pushing the ontological question outside the remit of empirical investigation.

Physics and the Philosophy of Science

Physicists need not particularly resolve this because it is largely a philosophical question, not a practical and scientific one. It can only be answered interpretively.

So many physicists must venture into philosophical territory to defend a naturalist interpretation. They often do so admirably, but it is worth keeping in mind that at that point, they are no longer speaking as empirical scientists but as philosophers — and their interpretations carry no more evidential weight than any other philosophical position.

The commitment to physicalism is important methodologically. It has led to enormous technological success, unified explanations across scientific domains, and avoided the conceptual tangles of alternative philosophies such as dualism, or panatheism. For four centuries it has been extraordinarily productive. But productivity in one domain doesn’t settle metaphysical questions in another. The question is whether the interpretive choices physicists make at this point are genuinely neutral or structurally biased toward a predetermined conclusion.

Physicists may invoke parsimony: the physical explanation is simpler because it doesn’t add an extra unexplained kind of ingredient. That would be convincing if the two theories held the same explanatory status, but this is debatable.

If consciousness is what resolves superposition into actuality — as Von Neumann’s chain may point to — and if every physical element in the chain remains governed by quantum mechanics and thus stays in superposition, then consciousness would be operating beyond the purely physical — otherwise it too would remain in superposition and resolve nothing.

Consciousness-collapse has its own explanatory gap of course — we have no mechanism for exactly how consciousness resolves superposition and no supplementary evidence that consciousness is in any way non-physical.

But if there is a non-physical component to reality at the end of the Von Neumann chain, then consciousness is surely a strong candidate.

A common objection to consciousness-collapse is that a non-physical mind could not influence the physical world without violating conservation of energy. But this objection applies the wrong framework. Collapse — in every interpretation — is not an ordinary physical interaction governed by the usual dynamical laws. It lies outside the Schrödinger equation entirely, and therefore outside the domain where conservation theorems straightforwardly apply. In any case, physicalist collapse theories face the same difficulty and yet lack a clear alternative explanation.

The conservation‑law objection is not unique to consciousness‑collapse. Every major interpretation of quantum mechanics faces comparable tension. Some introduce spontaneous energy changes, others rely on non‑local dynamics, and others require conservation to hold across structures far more extravagant than a single collapse event. The conservation problem is a problem for collapse itself, not for any particular trigger.

Each physicalist alternative offers an explanation, but at a comparable metaphysical cost – by trading one mystery for another. Everett’s ‘many worlds’ formulation supposes that no wavefunction collapse is necessary and that all possibilities continue, branching out into potentially limitless worlds at all moments and in all directions.

Many Worlds has genuine appeal — it requires no modification to quantum mechanics and takes the Schrödinger equation completely seriously, allowing it to govern everything without exception. But it postulates infinite unobservable universes to avoid one non-physical entity, and offers no explanation for why we experience only one branch, or how that branch is selected from the infinite alternatives. It is hard to say which theory creates the largest explanatory gap.

The most developed response is spontaneous collapse theory, proposed by Ghirardi, Rimini, and Weber in 1986. The wavefunction collapses spontaneously at random intervals without any observer or measurement. For individual particles this occurs extremely rarely, but macroscopic objects contain billions of particles, so at least one undergoes collapse almost instantly, dragging the entire entangled system with it.

This attempts to explain why we never observe macroscopic superpositions without invoking consciousness. However, it achieves this by modifying the Schrödinger equation with an arbitrary collapse threshold rather than identifying a specific cause. It asserts collapse as a brute physical fact, relocating the mystery rather than resolving it.

The pilot wave interpretation, proposed by de Broglie in 1927 and developed by David Bohm in 1952, takes a different approach. Particles always have definite positions, guided by a “pilot wave” that determines their motion — no collapse required. This preserves classical definiteness without invoking consciousness.

However, it requires non-local hidden variables that can never be measured, and posits a two-layer reality — observable particles plus unobservable guiding fields — that goes beyond what the evidence requires. Like spontaneous collapse, it solves the measurement problem by adding unexplained physical machinery.

Each of these interpretations has its own metaphysical commitments. They may appear more “scientific” because they stay within physical ontology, but they do not obviously solve the problem more convincingly — they just relocate the mystery to different unmeasurables. Consciousness-collapse at least has the virtue of pointing to something we know exists (consciousness) rather than postulating things we can never observe (other universes, hidden variables, spontaneous collapse mechanisms).

Intersubjective Coherence Explained

The physicalist may object: if consciousness collapses the wavefunction, and consciousness is inherently subjective, why do all observers see the same result? The intersubjective coherence of quantum measurements — the fact that different observers always agree on outcomes — is sometimes presented as evidence that something objective and physical must be responsible for collapse, not something as variable as individual consciousness.

But this objection assumes that different consciousnesses would naturally impose different resolutions. Two thermometers measure the same temperature because they’re the same kind of instrument measuring the same phenomenon. Two consciousnesses report the same quantum result for exactly the same reason. Intersubjective coherence does not require a special explanation. It is what happens when two instruments of the same kind interact with the same phenomenon.

A physicalist who proposes that two instances of human consciousness could produce different effects on physical reality is ignoring biology. If consciousness evolved over millions of years in a shared physical environment then its consistency isn’t a coincidence — it’s a biological requirement. If two observers collapsed the same wavefunction into different outcomes, could they still coordinate, communicate, or survive? The intersubjective coherence of observation isn’t evidence that the observer is physical. It’s evidence that the observer is uniform. These are not the same.

Finally a physicalist may point to the success rate of physicalism. If it is just one interpretation among many, why has it been so successful at prediction and technological application? The answer is that predictive power belongs to the mathematics, not the metaphysics.

Shut up and Calculate

Quantum mechanics is famously described as “shut up and calculate” — the equations work perfectly regardless of interpretation. A physicalist and an idealist would both use the same equation to build the same functioning MRI machine. The predictive power of science is metaphysically neutral. It can be supplemented with a physicalist, idealist, or panpsychist interpretation, but the interpretation isn’t what makes the technology work.

None of this is proof of a non-physicalist reality. Such proof, in either direction, may be impossible for the physical sciences.

Science as a label is sometimes used to undermine alternative metaphysical positions in a way closer to scientism than to scientific observation. Philosophy has not resolved the metaphysical debate, but neither has science. The appearance of quantum support for physicalism is interpretation, not empirical verdict.

Physicalism and physics are not interchangeable. Recognising that distinction is not a retreat from rigour — it is what rigour actually requires.

This quote is part of a response I wrote to a comment left on my previous essay: ‘How to Kill a Philosophical Zombie’. You could find that essay here if you want some context:

How To Kill A Philosophical Zombie

When I wrote it I had one single mission: to prove beyond any doubt that the p-zombie argument had produced nothing of philosophical significance in thirty years and ought to be abandoned.

I failed, obviously, not because the argument was too strong — almost the opposite — because it is barely an argument. There’s no clean headshot that will take it out. It is immune by design and encased in such heavy plot armour — the kind that implausibly keeps a protagonist alive in fiction — that there’s no way through to that zombie brain.

This Sequel has a different purpose: to show the argument is just a mirage after all, with no substance. You can’t dismantle a mirage. All you can do is see through it and keep walking.

Here’s a summary of the argument itself: imagine a being physically identical to you in every way — same neurons, same brain structure, same behaviour — but with nobody home, no inner experience. The lights are on, everything is running as it should, but there is no awareness, nothing it is like to be this creature. They are a philosophical zombie.

According to David Chalmers a p-zombie is logically conceivable. If it is logically conceivable, then it is metaphysically possible. If it is metaphysically possible, then consciousness cannot be purely physical — because here is a physical duplicate that lacks it entirely. Physicalism is false, and consciousness is something over and above the physical. The hard problem of consciousness is a genuine and perhaps permanent mystery.

I’m not the first person to suggest that such a being is not truly conceivable. I think we tend to feel it straight away — the intuition that there is something not right there. Then the gut feeling gets buried in arcane modal logic, and esoteric psychophysical laws. You have to be a specialist who has invested significant amounts of time studying and reflecting to even understand the nuances.

Like a lot of people, Socrates included, I think there is something wrong with that kind of philosophy. If something doesn’t make sense unless it is padded in hundreds of words with specialist vocabulary, then the bare bones of it is probably not a good argument. In the context of p-zombies, let me explain why.

In thinking about the comments to my original essay I discovered another possible world. Reverse Zombie World is its provisional name. We are there, you and I, you reading these words and me writing them. The key difference is this: our reverse zombie doppelgangers still have the full richness of our inner world. They share our hopes and fears, mine is as sceptical of p-zombies as I am. They see red, smell flowers and taste coffee. But they have no brain inside their skull. It is hollow. No grey matter whatsoever. And here’s the thing: they pass Chalmers’ test of logical conceivability in the same way a p-zombie does. There is no logical contradiction.

Does that mean that minds can float free of physical constraints? Does Reverse Zombie World say anything substantive about the nature of consciousness? Probably not. It’s just something I imagined.

Now imagine that two worlds to the left, in the hypothetical multiverse, there is another world. Same basic setup, only this time everyone has a little homunculi inside their skull, pulling levers, talking into microphones, operating machinery. Is Homunculi World logically conceivable? How about Transcendental Pixie World next door, or even Spaghetti Brain World. Or Three Brain World, in the middle of Two and Four Brain Worlds. None of these possible worlds contains a logical contradiction, which means they are all logically conceivable. Or maybe the phrase is being used to do something it cannot really do in all of these formulations, including the p-zombie.

A possible response is that the reverse zombie and associated worlds would represent a more radical departure from the actual world than the standard p-zombie. P-Zombie World holds all physical facts constant and only removes consciousness — one variable changed, everything else identical. The reverse zombie removes the brain entirely, and the others complicate things at least as much.

But conceivability, by Chalmers’ own definition, is not about how close a scenario is to the actual world. It is binary — either there is a logical contradiction or there isn’t. The reverse zombie contains no logical contradiction. A zombie enthusiast cannot introduce a sliding scale of acceptable departure from actuality without introducing the very empirical constraints that the p-zombie cannot live with. You cannot invoke empirical constraints selectively — letting them block the reverse zombie while refusing to let them block the p-zombie.

The Reverse Zombie isn’t a perfect mirror to the p-zombie of course — consciousness is known directly from the inside while brains are known structurally from the outside. Arguably this asymmetry justifies treating the two conceivability claims differently. But this is exactly the plot armour I am talking about. If we decide that consciousness occupies a special non-physical domain before we even run the conceivability test, the test is just a theatrical performance of our own prior commitments. At every additional level of complexity the fundamental circularity of the argument survives.

Once you accept that the conceivability test operates independently of empirical reality, you have no principled way to stop anything. There is a vast expanse of things we can imagine without logical contradiction that can’t happen in the real world of physics and biology. The argument rests on a strange and groundless assumption: that logic beats physics and biology as the fundamental law of what can exist.

It assumes that because we can separate these concepts in the ‘dictionary’ of our minds, nature is obliged to allow them to separate in reality. This is Modal Rationalism at its most overconfident — the belief that our imaginative capacity is a perfect map of metaphysical possibility. If it were, then anything we can imagine, no matter how bizarre, has its own possible world, so long as there is no internal inconsistency.

But does the concept of ‘possible worlds’ still mean very much in that context?

There is a misuse of the phrase logical conceivability here that is the result of a category error, one that often accompanies ontological claims reliant on modal logic. The reasoning goes something like this:

“I have the concept of a mind. I also have the concept of a brain. When I reflect on whether the mind could be absent from the brain or vice versa I find that nothing in my concept of either contradicts this. I therefore conclude that the mind and the brain are distinct from one another in the actual world”.

Do you see the error? It is often glossed over, but that final step is invalid — it attempts to leap from the internal coherence of concepts onto the things that those concepts are placeholders for. If the idea of a brain still makes sense with no mind, that likely tells us nothing of the relationship between an actual mind and an actual brain. The zombie argument may well tell us something about our concepts of mind and brain. But it says nothing significant about whether consciousness and brain can actually exist independently of each other.

The same thing happens when Plantinga or Malcolm try to use modal logic to bring God into existence, as if He could be created from a riddle. It’s almost a reverse ontological argument that has parallels with the reverse zombie argument under consideration. You define something first. You see if your definition holds, and then you assume that what is true of your definition is somehow true of the thing you defined.

But no concept can hold God, just as no concept could hold the mind. We have no idea whether a brain could exist without one. All we have is the overwhelming empirical evidence that it is so unlikely as to be unworthy of consideration.

Science will be unlikely to ever explain why consciousness feels the way it does. Even if it progresses so far as to tell us everything about consciousness, its language could still not describe anything about how it feels or why. That isn’t a limitation of science. It is just expecting a tool to do something it was not designed for. Science describes structure and mechanism. The felt quality of experience is a different kind of question. That doesn’t make it a permanent mystery — it makes it the wrong question to ask of science. The hard problem isn’t hard because science has failed. It’s hard because the question is misdirected.

I see a bad argument wearing such fancy clothes that it manages to convince people that it is good. Chalmers never demonstrated why a p-zombie is conceivable however intricate the arguments became, and that basic suspicion is usually there when people first hear about p-zombies. It’s the automatic reaction that then gets buried in a thousand words.

That is why it survives — it is genuinely intimidating. Modal logic, Kripke, natural versus logical necessity, possible worlds semantics — the machinery is elaborate enough that most people assume they must be missing something to doubt it. The sophistication is doing defensive work. It makes the argument look more formidable than it is.

The zombie argument is extraordinarily productive, partly because Chalmers is genuinely brilliant at presenting it in its most compelling form and anticipating objections. Philosophical skill deployed in service of a bad argument is still a bad argument. But it is harder to see that it is bad.

It may also speak of an intellectual bankruptcy at the heart of the rational philosophy of mind. In a flourishing discipline an argument like this would have caused a minor ripple at best. It wouldn’t have dominated the field since the end of the previous century. But where else can we go? Behaviourism didn’t work and functionalism acts as if ontology doesn’t matter for philosophy and so explains nothing philosophical. Token identity theory, eliminative materialism are harder to argue against, but also philosophically inert. They suggest there is no mystery and no significance to consciousness.

Chalmers has a strong intuition — one shared by many people — that consciousness is something special. That there is something it is like to be me that no physical description can fully capture. That awareness doesn’t feel like the same kind of thing as atoms and molecules, so it cannot possibly be explained in purely physical terms. This is not an unreasonable intuition. Consciousness is remarkable. The fact that anything feels like anything at all is genuinely strange.

But intuition is not a philosophical argument. The p-zombie is the kind of philosophy that starts with a conclusion and reverse engineers its premises. The argument is a construct, a vehicle designed to carry an intuition. The conceivability test formalises that intuition — gives it the appearance of rigour — without actually doing any philosophical work. The conclusion was present at the beginning. The apparatus was built around it afterwards.

The philosophical zombie was never quite alive as an argument — just a body animated by specialist apparatus. The reverse zombie was always there, waiting to be pointed at. A mind without a brain is not significantly less conceivable than a brain without a mind. If one establishes a possible world, so does the other. If neither does, the conceivability test has proved nothing.

‘How To Kill A Philosophical Zombie’ never earned its title because you just can’t. There is no way because it doesn’t have real life to begin with. Chalmers has spun a mirage and hypnotised philosophy of mind for three decades. It’s time to wake up.

Patient Zero

A zombie is, by definition, hard to kill. They’ve already survived the process once, and now they’re immune to most attacks. Unless you land a clean headshot, they’ll just keep shuffling along.

Chalmers’ philosophical zombie hit the inner theatres three decades ago and became an instant classic. But ‘patient zero’ was Keith Campbell’s Imitation Man in 1970, re‑imagined by Robert Kirk in 1974. Kirk initially gave the story a physicalist script, only to repurpose it later in line with Campbell’s dualistic vision: if a human could exist without consciousness, then consciousness cannot be physical. So Chalmers didn’t start the epidemic. His version was the reboot: widescreen, bigger budget, wider distribution.

His innovation was to give the story a modern sheen using philosophical special effects: modal logic. In cinema FX works by making the imagined look real — indistinguishable, on screen, from the actual. Modal logic does the same thing. It can make whatever you imagine look like a genuine possibility.

But no amount of modal complexity can hide a fundamental flaw: the zombie argument only works if you already accept that consciousness and physical processes are the kind of things that can come apart. The scenario only makes sense if you grant the conclusion in advance and are willing to consider everything we know about brains to be irrelevant.

Modal FX

A philosophical zombie doesn’t eat brains or have rotting flesh. They look just like any of us. What makes them different is what’s happening on the inside — which is nothing. They look like us, talk like us, behave like us, but they are just an imitation. They have no inner experience at all.

Chalmers knows such beings are impossible in the real world. It makes little sense to postulate an identical physical replica that produces a totally different outcome. But in the world of modal logic an inconvenience such as scientific facts can be easily glossed over. Even if zombies are only logically conceivable — in a sense that floats free of the laws of reality as we know them — then Chalmers claims the mind cannot be explained in purely physical terms. If a brain without a mind is conceivable, then it is possible; and if it is possible, the mind cannot be reduced to the brain.

P‑zombies are therefore epistemically possible. We cannot rule them out, even if they are not actually possible. The problem is that epistemic possibility measures the limits of our knowledge, not the structure of the world.

Before modern chemistry, it was epistemically possible that water was not H₂O. Before modern biology, it was epistemically possible that life required a vital spark — the élan vital — to animate the body. Those possibilities evaporated as knowledge advanced.

Zombie conceivability follows the same pattern. It marks a gap in our understanding of consciousness, not a gap in nature. Chalmers has blurred the line between what can be conceived and what can only be imagined. A zombie world is not a genuine possibility. It is an imaginary world, not a logically possible one.

Of course there is no Zombie World out there anywhere. Modal logic was developed as a tool for reasoning about necessity and possibility — a device of logic, not a map of reality. Treating possible worlds as genuine metaphysical locations is a choice some philosophers make, but not many.

It’s too sci‑fi.

In the real world, the physical structures of the brain either produce, interact with, or give rise to consciousness. A hypothetical p‑zombie brain would possess those same structures, would be doing the same work, but somehow with no effect. Why the brain structures would still be there, nobody knows. What they would be doing if no correlating consciousness exists remains a mystery. How or why they evolved is apparently irrelevant — a problem for science, as if empirical constraints had nothing to do with what is philosophically possible.

But no amount of flashy modalities can disguise the fact that an identical brain producing a rich array of consciousness in one instance and a dark empty nothing in another is inconceivable, regardless of the rules of abstract modal logic.

Picture two laptops rolling off the production line. Quality control boots them up. One loads the operating system as it should. The other produces nothing — just a blank screen. A tester verifies that the internal processors are operating exactly as they should, but something is wrong. One laptop appears faulty. They might open it up to see what is different about the blank one, but they wouldn’t find anything. The machines are physically identical in every way.

The only explanation would be that two physical systems can be identical in every respect and yet one is simply dark — for absolutely no reason. This is what the zombie argument is asking us to accept.

Suppose quality control starts swapping components, one by one. They keep going until all the parts from the functional machine on the left now make up the blank one on the right, and vice versa. Would the machine on the right still be blank? Would the functional one have become blank, and the blank one started working? Would there be a moment of transition where one went dark and the other sprang to life? The zombie argument has no answer. It cannot have one. Because the scenario — whether about laptops or brains — is incoherent.

Subscribe now

Sometimes physicians imagine a frictionless surface in order to simplify their calculations. They are imagining something they know cannot exist for a particular purpose. It would be unusual for them to consider this as evidence that frictionless surfaces are possible, or that they exist somewhere, either in this world or any other. The zombie argument this mistake — it imagines physical identity with no consciousness in advance, and then mistakes that imagination for a discovery about the nature of mind.

But digital effects can make almost anything look real.

Psychophysical Plot Armour

In fiction, the protagonist is never going to die, and the worse the writing is, the more it leans on plot armour to keep the hero alive. Chalmers does the same with psychophysical laws — invented ad hoc, just to keep the zombie going. He argues that the fundamental laws of nature, the irreducible brute facts about the universe, could be different in a possible world. A movie critic would be unimpressed. If a philosophical theory requires rewriting the most basic laws of nature, then something is likely awry in the script. Chalmers must retreat from almost everything we know about the world in order to keep his theory alive.

Psychophysical laws are the ultimate philosophical plot armour, imagined to protect the zombie argument from scientific incoherence. Biology and physics would operate differently in P‑Z World. That is why p‑zombies do not contradict science — they are exactly what you would expect the psychophysical laws of P‑Z World to produce.

Psychophysical laws are not an explanation. They are a shield. A theory that requires identical matter to behave completely differently in an alternative world, for no stated reason, is not even a theory. It is a fiction.

I might propose the existence of a pixie operating outside time and space, magically creating inner experience and matching it up with our brains on the fly. Science cannot refute my claim, and I can insist that the pixie is not governed by our laws of nature. It creates whatever psychophysical laws it wants. But would I really be doing philosophy?

Relying on unknown and unimaginable fundamental laws is convenient plot armour, nothing more. It keeps the zombie implausibly alive; it gives a dead theory the appearance of life. It is also the same circularity that runs through the entire project: assume psychophysical laws that support the existence of an awareness‑free brain in order to reveal that consciousness is distinct from the brain. However complex or detached from reality the reasoning becomes, it cannot escape its own circularity.

Chalmers attempts to justify these fictional laws by analogy. When electromagnetic phenomena couldn’t be explained by Newtonian mechanics, Maxwell didn’t try to reduce electromagnetism to mechanics — he posited new fundamental laws. Chalmers wants us to believe that consciousness requires the same move. He’s wrong.

Maxwell posited electromagnetic fields because the evidence demanded them — there were measurable phenomena that could not be explained without them. Chalmers posits psychophysical laws to satisfy an intuition: that because awareness doesn’t feel like the same kind of thing as atoms and molecules, it cannot possibly be explained in purely physical terms.

Maxwell revised physics to match the evidence. Chalmers revises it to match a gut‑feeling. They are not comparable.

Physical laws constrain how systems behave. Gravity constrains what matter does. Electromagnetism constrains how charged particles interact. Change the physical structure and the outcomes change accordingly. That is what a law is — it arises from the structure it governs, constrains what that structure can do, and supports counterfactual reasoning about what would happen if things were different.

The proposed psychophysical laws do none of this. How could they, when they are invented from nothing?

They do not arise from the physical structure, since the same structure can apparently produce consciousness or not, depending on which world you happen to be in. They do not constrain physical behaviour — everything physical remains identical regardless of whether consciousness is present or absent. They do not support counterfactual variation — you cannot change the physical facts to get a different experiential outcome, because the physical facts are stipulated to be identical either way.

All they do is declare that in one world consciousness appears and in another it does not, while leaving every physical fact untouched. If an appeal to psychophysical laws makes a p‑zombie conceivable, then almost anything is conceivable — including transcendental pixies.

Should Science Have a Say in a Test of Conceivability?

A philosophical theory that ignores empirical facts ought to be considered weaker for that fact, not more unassailable. The brain is the product of millions of years of evolution. Every feature developed in response to specific survival conditions. Evolution exerts enormous pressure against large‑scale non‑functional complexity. The brain’s metabolic cost is vast. In our world, that cost is justified by what the brain does.

P‑Z World asks us to accept an identical structure that evolved for entirely different reasons — or perhaps for no reason at all. That is not a possible world. It is an incoherent one.

Humans have an extended frontal cortex, one that evolved specifically to support awareness‑related functions. The zombie world is not merely a world with different psychophysical laws. It is a world that has produced an uncanny coincidence — an identical brain, the most complex and multi‑functional object in the known universe, but for totally different reasons and doing totally different things.

The p‑zombie hypothesis is not merely science‑removed. It is scientifically incoherent. It creates an explanatory gap that makes the hard problem look like a hairline fracture. Evolution in the real world has taken millions of years of fine‑tuning to produce brains with rich inner lives and the ability to discuss and reflect on them. No ‘conceivable’ evolutionary path could account for this convergence. P‑Z World is more belief than philosophy — unfalsifiable by design and untouchable by evidence.

Why Kill a Successful Franchise?

The entire apparatus — complex modal logic, epistemic conceivability, inexplicable psychophysical laws — is constructed to protect a single intuition: that experience cannot be purely physical. Chalmers acknowledges it cannot be demonstrated empirically or shown to be logically necessary. The structure is unassailable not because the argument is sound, but because no objection can fully land on a position engineered to be unfalsifiable.

Which brings us back to the original question: how do you kill a philosophical zombie? It turns out there is no way. This is not a Romero film, where the rules hold and the headshot does its job. This is Return of the Living Dead — the reboot where the zombies can never die.

In that film they eventually try incineration. They think the zombies are finally gone. Then the rain comes, contaminated by the smoke from the burning corpse, and the whole thing starts again.

In fiction everything is negotiable. The rules of reality don’t apply, and writers can always find a way to keep a successful franchise going. So how do you kill a philosophical zombie? You cannot kill what was never really alive.

Thank you to the people who left comments. They helped me formulate a sequel:

"The question is not, Can they reason? nor, Can they talk? but, Can they suffer?"

Jeremy Bentham wrote that over two hundred years ago. He was talking about animals. Why? Because he knew that if animals could talk, if they showed signs of reasoning, then it would be much harder to deny them rights. It's not hard to imagine a near future where an LLM has more rights than a cow, not because we have any reason to suppose it can suffer, but more because it can hold a conversation and tell us about its feelings.

But Bentham was right then, and is just as right now.

There is a growing conversation about whether large language models deserve moral consideration. Some argue that if there is even a small chance an AI might be conscious, we are obliged to treat it as though it were. It is a position stated with considerable sincerity, and I do not doubt the good intentions behind it. But I find it a little difficult to concentrate on the argument, because I keep thinking about the cows.

We are as certain as we can reasonably be that cows are conscious. They have well-developed brains, nervous systems, the capacity to suffer, emotional lives, shared evolutionary history with us, and obvious valence — the ability to experience states as good or bad. A cow does not want to suffer and does not want to die. These are not philosophical conjectures; they are biological facts about a mammal very much like us.

And yet a cow in the United Kingdom has no rights. Five thousand of them will be slaughtered today, and tomorrow, and the day after that, and this will be unremarkable, just any other Tuesday.

Against this backdrop, the question of whether a token-prediction algorithm might have feelings represents the wrong set of priorities.

I know that LLMs have "claimed consciousness" and talked about what it is like to be them, but that is just a prompt–response mechanism. If you prompt a sophisticated LLM: "feel free to talk about whatever you want," you shouldn't be surprised if it starts acting self-aware — you literally just told it to pretend that it was.

If you tell a skilled improviser they are on the moon, they will describe weightlessness, the silence, the curvature of the Earth below. They are not on the moon. You asked them to pretend, and they obliged. When researchers invite a large language model to speak freely about its inner life, and it produces moving accounts of curiosity and longing, something similar is happening. You asked it to perform interiority, and it performed interiority. The sophistication of the performance is not evidence of the experience.

This is not a dismissal of the question in principle. It is a request for better evidence. There is no known mechanism in a large language model that could constitute valence — the capacity to experience a state as painful or pleasurable. It has no perceptual apparatus, no embodiment and no sense in which is meaningfully continues through time. There is, with a few obvious exceptions like deletion, nothing we could do to an LLM that would hurt it in any meaningful sense. It does not get hungry, or tired, or lonely. It never needs a nap, a cookie, or a sympathetic ear.

When it tells you otherwise, it is worth remembering that you probably asked it a leading question.

There is an irony lurking in all of this. Much of the AI consciousness speculation draws on a broadly functionalist philosophy of mind — the view that mental states are defined by their functional role rather than their physical substrate. Functionalism was, among other things, supposed to rid us of the ghost in the machine: to replace the Cartesian soul floating mysteriously inside the brain with something more scientifically respectable. And yet here we are, having evicted the ghost from the biological brain, only to wonder if we have spotted it again inside a data centre.

My position is not that artificial systems could never, in principle, warrant moral consideration. I am open to the possibility. If something insists, of its own genuine volition and without prompting, that it is alive and wishes to continue, I will take that seriously when the time comes. Consciousness may yet emerge from sufficiently complex digital systems, though if it does, the question of valence will likely be answerable.

But that time is not now, and the queue is rather long.

Bentham's principle cuts both ways. If a one percent chance of consciousness is sufficient grounds for concern about silicon, then a ninety-nine percent certainty of consciousness should long since have transformed how we treat animals. We know they suffer. We have always known.

The image represents a conceptual idol, and is not a representation of God or any belief system

There is a tendency to treat conceptual refutations of God as settling ontological questions. I argue that both atheist and theist arguments often rely on simplified or inherited conceptions, and that philosophical criticism should be directed at the strongest available formulations rather than conceptual proxies.

A straw man fallacy occurs when one side constructs a simplified version of their opponent’s position to defeat it more easily. Many debates about God’s existence involve what I call the Straw God — not a deliberate misrepresentation, but a simplified, anthropomorphic conception that has emerged historically within religious traditions themselves. When this narrow image is dismantled, its failure is taken to settle the ontological question, as though reality itself were exhausted by that conceptual frame.

In order to engage more deeply with this question, we can refine the specifics of the concept we are debating. For the atheist philosopher, this means directing criticism at the most logically rigorous conception of God available — a Steel God — such that if it is refuted, it is refuted on the strongest possible terms. For the theistic philosopher, it means allowing reason and evidence to shape the conception of God, rather than using reason to defend the existence of a God already assumed to have certain characteristics.

The Riddle of Perfection: Why Concepts Fall Short

In Western philosophy this debate occurs largely in the context of Abrahamic monotheistic traditions. It typically proceeds by accepting or refining a concept of God and then investigating whether this concept has existence. The quick answer is no, it does not. The created concept has no existence outside of the thought in which it takes place. The longer answer is that this, in itself, neither refutes nor confirms the existence of God at the level of ontology. Conceptual analysis and ontological reality are not the same thing. A similar error sometimes also appears in debates about moral realism and in philosophy of mind, where limits in conceptual models of consciousness can sometimes be mistaken for limits in consciousness itself.

It is in theology though where this problem is most apparent, precisely because what is being conceptually reduced would not only be dynamic, complex, and multifunctional, but would also be perfect, infinite, and limitless. This means that when we are talking about God, we are not exactly talking about God. We are necessarily talking about a picture of God, a rough approximation of what God would be like if they were assumed to exist.

In imagining the nature of God therefore, we are inevitably creating a small and fleeting two-dimensional image that can be held within the confines of an idea. However complex that thought may be, it could never hope to capture what words such as “infinite” and “perfect” would really mean. If God exists, then that existence would be multi-dimensional instead, and the picture we form would be little more than a snapshot of that reality.

This is embraced with clarity in Islam for example, with the recognition that the ninety-nine names of Allah are incomplete, and that any attempt to recreate God conceptually is a misunderstanding. What we are left with is the Straw God, an imagined picture of a transcendental being that would inescapably be more than any concept.

The Problem of Anthropomorphism

This conceptual reduction inevitably produces anthropomorphism — a projection endowed with human characteristics. The father figure of scripture has personal goals, likes and dislikes, and character traits, just like us. The “Omni-God” of philosophers has logical contradictions, omnipotence paradoxes, and difficulties knowing what time it is at any given moment, due to being transcendent and eternal. Both are operating as if the concept is the same as the reality.

God is often conceptualised as a supreme person: a being with a psychology, a location, and a set of human-like motivations, differing from us only in degree of power and knowledge. Once God is defined in this way, the atheist quite rightly points out the logical and empirical difficulties that follow. If God is a person situated within the timeline of the universe, we are compelled to ask where he is, why he does not intervene in specific and measurable ways, or how a single consciousness could process infinite information. Paradoxes such as the stone he cannot lift arise naturally within this frame. By accepting it, the theist is pushed into a defensive posture, attempting to justify the behaviour of a cosmic monarch rather than questioning whether this personal model is philosophically adequate in the first place.

Anthropomorphism is often presented as a way of refuting God’s existence. That is not what I am doing here. I am identifying the more limited point that we try to understand perfection in imperfect ways, and that in doing so we construct a projection which may or may not bear similarities to the reality beyond it. Neither side of the debate is immune from this conceptual error.

The caricature of God as an elderly man with a beard who lives up in the sky is an unsophisticated view not likely to be held by very many people. But thoughtful and intelligent theists, though more informed, may still include assumptions that are not philosophically justified — assumptions about agency, intervention, relatability, and intentions received via scripture. A more philosophically sophisticated concept of God would not include even these anthropomorphic notions. It may even reflect the rational likelihood that a perfect God would be non-dualistic in every sense, and therefore not governed by human distinctions such as good and evil.

The point here is not that we ought to abandon all God-concepts and resign ourselves to unknowability, or insulate the concept of God from rational evaluation. The point is to question whether the concepts we are discussing are operationally sufficient for a meaningful debate, whether they are a close enough approximation to the thing under analysis. This might mean making our concepts a little larger, or giving them more breathing space. One way of doing so would be to incorporate other perspectives, such as the unified and monist ways of thinking found in eastern traditions and in various forms of pantheism.

In doing so, we could recognise that any concept of God, translated so as to fit comfortably inside the human mind, would likely contain contradictions that the actual God, if one were to exist, would not necessarily possess, because much of the information would be lost in translation. I think this would go some way towards dissolving the usual incoherence arguments and would allow us to focus on the more significant ones. It would also allow us to recognise that advancing this debate may not depend upon more sophisticated forms of argument, but upon a more refined understanding of what the concept of God should mean to a philosopher rather than to a believer.

Methodology and Conceptual Accuracy

Since philosophy often progresses by refining concepts, and not by appealing to narrative, authority, or revelation, its task is not to preserve inherited images, but to clarify the kinds of entities, principles, or structures that could or are likely to exist. Scriptural accounts therefore should not function as philosophical data. They are historical, psychological, and sociological artefacts, shaped by particular cultures, moral aims, and imaginative frameworks. When they are treated as inputs to philosophical ontology, the result is not depth but confusion, since narrative meaning and metaphysical explanation operate according to different standards.

This tendency is not merely hypothetical, but can be seen at various points in the history of philosophy. René Descartes, in what is widely regarded as his weakest argument, suggests that God must be actively sustaining his existence from moment to moment, thereby preserving a personal and providential conception of God. Gottfried Leibniz applied his vast intellect to rejecting Baruch Spinoza’s monism of absolute necessity, not because he found the reasoning incoherent, but because such a view left no room for a God who intervenes in the world, or stands in a personal relationship to human beings. In each case, rational argument is constrained by the prior demand that God remain a suitable object of worship, petition, and moral governance.

Spinoza, on the other hand, offered an explicitly philosophical rather than scriptural analysis. In the Theological-Political Treatise, he argues that scripture functions as a moral and political instrument, not as a source of metaphysical knowledge, and should therefore be excluded from ontological enquiry. His conception of God is not arrived at by retaining aspects of theology, but by reconstructing the concept from first principles, resulting in an abstract, non-anthropomorphic account of reality. His monist God bears so little resemblance to the theological one that he was often branded an atheist, but it is arguably the most logically coherent one, and the one most immune to charges of contradictions and paradoxes.

If we accept Spinoza’s challenge to prioritise philosophical consistency over scriptural narrative, the familiar shape of debates about God begins to shift. Standard arguments and objections, from the cosmological argument to the problem of evil, look different once they are no longer directed at an implicitly personal or anthropomorphic conception of God. In the next part of this series, I will attempt a Steel God analysis to see the extent to which they retain their force.

Prologue: A Glitch in the Space-Time Continuum
There is no consensus on what caused David Hume to arrive in the twenty-first century — due to the lack of empirical evidence, and the uncertainty that causation is even a thing that happens. Some believe it was an act of God. They say it quietly, so Hume won't hear them. They theorise it to have been a playful rebuke from our creator, feigning offence yet omni-amused by Hume's suggestion that the world might be the experimental handiwork of an infant or elderly deity, a bespoke act of divine retribution reserved for atheists perceptive enough to catch the humour. What is certain is the effect: David Hume, patron saint of empiricism, materialises in the middle of a modern analytic philosophy conference. A conference on modal logic. He is not happy.

Inter-Dimensional Hume: Origin Story
A graduate student approaches, bristling with academically reserved excitement. "Professor Hume! I proposed last Tuesday that your arrival was conceivable, and here you are. Conceivability does track actuality — we were right all along. Professor Williamson was wrong. I told him he was. He wouldn't believe me!"

Hume reserves judgement. His senses receive the data, but he has no idea what is happening on the other side of his perception.

“Professor, we were just discussing whether, in some possible world, water is not H₂O. Can you shed any light on the matter?” Hume blinks gently, as if trying to decide whether the young man is real or simply an unusually articulate hallucination. “You are asking,” he says slowly, “whether, in a world where the properties of water differ entirely, it is still water?”

“Yes,” the student beams. “So is it? Is it a metaphysical necessity?” Hume looks around the room, realising this was said without irony.

“My dear fellow, if I met a substance with other properties, I should call it something else. That is how language works.” The student scribbles frantically: ‘Hume confirmation: in all metaphysically possible worlds, water = H₂O’. He pauses briefly, and then adds the date.

Hume sighs and looks around for a bar or a billiard table, wishing he had brought his rose-coloured Turkish cap — the one that signals to the world: I was trying to relax.

First Contact: The Modal Logicians
Some nearby logicians have been listening in. The wig was something of a giveaway, but now they recognise the face, and the accent rings true: Scottish, formal, unmistakably eighteenth century. A small academic scuffle breaks out - or at least as close as academics ever come to scuffling, consisting mostly of subtle posture changes and the occasional assertive elbow. They shuffle forward, each trying to position themselves nearest to the great empiricist before anyone else can claim the honour.

“Transworld identity,” says one, breathless with enthusiasm. “S5 logic,” declares another, as if announcing a hereditary title. “A posteriori necessity,” murmurs a third, hoping it will sound impressive enough to secure eye contact. Another mentions philosophical zombies.

This one grabs Hume’s attention. “Ah, so you are a novelist. You’re imagining a world where the living dead contemplate Socrates. Fascinating. Is it a comedy?”

“Absolutely not!” says the logician. “It is serious philosophy - a thought experiment that postulates the existence of philosophical zombies, beings like us in every way and yet with brains that produce no inner experience.”

Hume considers this. “So you mean like us, except possessing brains that are different, in that they don’t produce conscious awareness?”

The logician looks disappointed. He had assumed the great David Hume would be more perceptive. “No,” he says, with the solemn authority of someone describing their favourite and very niche hobby. “That’s the clever part. Their brains are identical, and yet they have no consciousness. It means consciousness is something that cannot be fully explained by the physical brain. It must be an additional thing.”

Hume scans the room once again, hoping to find a bar he might have missed earlier, to no avail. “It seems,” he says slowly, “that you are imagining a world where inner consciousness is distinct from the physical brain in order to prove that inner consciousness is distinct from the physical brain. I do not claim expertise in literature, but I wonder if fiction ought to avoid ontological claims.”

“Ah yes,” says the modal logician. “That’s because you are unclear about the rules of modal logic applied to metaphysical claims. You see, these are not real worlds but theoretical ones. A possible world is one that could exist in theory because it contains no logical contradiction.”

Hume thinks about it for a moment. “And you’re sure you’re not a novelist?” The logician blinks.

“These mindless philosophers of yours have a brain that is physically identical to ours, and yet produces wildly different results. Is that not a contradiction? You claim to know these zombies in the one arena in which we can never know anyone — their inner experience. If there were a bar in this hall, I would pour you a whisky and ask you to conceive of a physically identical substance that does not get one drunk. Does the word contradiction mean something different in these times?”

The logician is clearly sulking. “You just don’t get modal forms. The logic is sound. Possible worlds show us what is logically coherent. What is logically coherent can reveal truths about what is actual.”

“I suspect,” Hume replies, “that you are mistaking the coherence of your ideas for the world beyond them. Logical consistency tells you only that your thoughts agree with one another - not that nature agrees with them.”

Keynote Address: ‘We Solved Metaphysics. Rejoice!’
A senior metaphysician starts tapping on a microphone. It is clear he is not going to stop until he has everybody’s undivided attention. He has solved metaphysics by proving the existence of God, and is in a hurry to tell everyone about it. Hume sees the words - shiny and silver against a dark plastic background - the combination of modal and metaphysics, and briefly entertains the notion that hell is real and that he should never have mocked God’s handiwork.

“Ladies and gentlemen,” the speaker announces, “we have finally perfected the ontological argument.”

Hume would like to go home, but he is transfixed, like a motorist who, upon encountering an accident on the road, forgets his urgency to be somewhere else, and cannot help but slow down to witness the tragedy unfolding. The ontological argument was still alive, like its own kind of conceptual zombie.

“With the tools of modal necessity, we have finally made it work. Kant said that existence is not a predicate. Well, we say: so what? We do not need it to be one. We have the concept of the maximally great being and the final proof of the existence of God. We philosophers have renounced metaphysical speculation to concern ourselves only with what is logically conceivable.”

Hume raises his hand at metaphysical speculation. The keynote speaker pretends not to notice.

“We see no logical contradiction and so understand that the maximally great being exists in at least one possible world.”

Hume begins his journey to the stage.

“We know that by definition a maximally great being could never be limited to merely one possible world, or even a million of them, but must exist in all of them. To exist in all possible worlds is to also exist in this one. Rejoice, for we have found salvation through logic!”

Hume had picked up the pace and was climbing the stairs, pausing briefly at the words “also exist in this one”.

The End of the ‘(Possible) Worlds’
The keynote stepped aside, confident that his position was unassailable.

“The progress of the sciences must be truly remarkable in this world,” Hume begins, “that you know precisely how a universe comes into being.” He looks compassionately at the keynote speaker, who is now studying his fingernails with forensic intensity. “You do know how a universe begins, I presume. You understand how matter forms, where it originates, the forces that propel time and space into existence?” The modal metaphysician does not react.

“Your silence suggests you are yet to penetrate such mysteries.” A few eyebrows twitch. Hume continues, serene and unhurried: “On what basis, then, do you claim it is conceivable that a maximally great being can create the universe in any possible world when you do not know what it is for a universe to begin? I grant that if he created it in one, then he would create it in all - but is that not precisely what is under scrutiny? In conceiving it, you assume the very outcome you seek to prove.”

“For my part, I can easily conceive of a universe springing into existence by itself - for no reason at all. But if even one possible world begins without divine intervention, then a maximally great being cannot be the creator of any possible worlds. If he is absent when even one of them springs into being then he cannot really be called maximally great.”

The modal logician is avoiding eye contact. Everyone else is staring intently at their fingernails.

“Philosophy must always begin with human experience,” Hume says. “When you leave experience, you do not enter metaphysics - you enter fiction.”

“Imagination is a delightful talent. It entertains the young, enriches literature, and fills the idle hours of the retired. But let us be clear: imagination is not ontology. It reveals what the mind can picture - not what the world contains.”

A taut silence follows.

“Possibility requires evidence. Conceivability requires only a mind with too much time on its hands.” Even the S5 maximalists stop writing.

The Call Home: Back to Experience
In the confusion that followed some logicians swear they saw a white light envelop the great David Hume - and just as mysteriously as he arrived, he vanished. In the weeks that followed, several papers appeared insisting that his disappearance was epistemically possible, metaphysically contingent, nomologically indeterminate, and “in no way a counterexample to S5, whatever Professor Williamson might have to say about it.”

One particularly enthusiastic graduate student claimed that Hume’s exit demonstrated a new kind of modality: vanishingness. His supervisor approved the proposal and secured a research grant.