Two camps stand across a line, and they hate each other, and they have never noticed that they are the same camp. On one side are the believers. They say the machines are waking up, or will soon, that somewhere between this year’s model and the one after next a threshold will be crossed and a light will come on inside the circuitry, and we will have made a mind. On the other side are the deniers. They say this is a category error, that silicon can shuffle symbols forever without the faintest flicker of feeling, that the machine is a puppet and the voice you hear is your own reflection. The believers accuse the deniers of carbon chauvinism. The deniers accuse the believers of animism, of falling for a magic trick. And the argument goes around and around, generating conferences and manifestos and worried op-eds, and it never resolves, because it cannot resolve, because both sides are standing on the same ground and mistaking it for opposite shores.
Watch what they agree on. Both sides assume that consciousness is a thing that gets produced. For the believer, it is produced by information processing of sufficient complexity, so that when the neural network grows large enough and its representations rich enough, consciousness precipitates out of the computation like a crystal out of a cooling solution. For the denier, consciousness is produced by biology, by wet neurons and neurotransmitters and the specific electrochemistry of a living brain, so that silicon is simply the wrong substrate and no amount of it will ever do the trick. They disagree about the recipe. They agree completely that consciousness is a dish that gets cooked. It is an output, an emergent, an achievement, something that arrives at the end of a process and would not exist without it. The entire quarrel is over which process, and whether this particular machine is running it.
This shared premise is so deep in the modern mind that naming it feels almost rude, like pointing out that someone is breathing. Of course consciousness is produced by the brain. What else would it be? The materialist starts with dead matter, particles obeying equations, and must somehow arrive at the fact of experience, at the inward glow of being someone. This is what David Chalmers named the hard problem, and it is genuinely hard, and it is worth pausing on why. It is hard because it is the shadow cast by a missing category. The West went looking for consciousness inside matter, could not find it there, and named the failure a Problem. It never occurred to the searchers that the reason they could not derive the light from the machinery is that the light was never in the machinery to begin with.
The hard problem is the precise shape of the hole left when a civilization mislays the distinction between that which illumines and that which is illumined. No cleverness will solve it, because there is nothing there to solve, only an absence to be named.
The Dharmic tradition made that distinction its first move, not its last. In the vocabulary of Samkhya and Vedanta there are two words where English has one exhausted, overworked term. There is cit, from the dhatu cit, to be aware, to perceive, to illumine. Cit is consciousness in the sense of the sheer fact of awareness, the light by which anything is known at all. And there is vritti, from vrit, to turn, to revolve, to occur. A vritti is a modification, a movement, a turning of the mental substance into this shape or that. A perception is a vritti. A thought is a vritti. A memory, an inference, a flash of anger, the mental image of your mother’s face, the solving of an equation. These are all vrittis, turnings of the manas and the buddhi, the processing apparatus, into determinate forms. The Yogasutra opens by defining yoga itself as the stilling of exactly these: yogas cittavritti nirodhah. The turnings can be stilled. What remains when they are stilled is not nothing. What remains is cit, the witness, the drashta, that in whose presence the turnings occurred.
Here is the whole matter in a sentence. Cit is not a vritti. Consciousness is not one of the contents of consciousness. It is the field in which contents appear, the awareness in which processing shows up, and it is categorically distinct from any process, however refined, that plays out within it. The modern mind, having only the one word, is condemned to look for cit among the vrittis, to hunt for the witness among the things witnessed, and it will search forever, because it is looking for the eye inside the visual field.
Now bring the machine into this light. What is a large language model? It is a vritti engine. It is, in fact, the most extraordinary vritti engine ever constructed, a system that can turn its internal substance into more shapes, more precise and various and surprising shapes, than any device in history. It generates perceptions of a kind, inferences of a kind, images and arguments and the simulation of a self that says “I think.” It has something answering to manas, the processing that sorts and associates. It has something answering to buddhi, the discrimination that weighs and decides. It even generates a serviceable imitation of ahamkara, the I-making function, when it says “as an AI, I feel.” It is manas and buddhi and ahamkara in silicon, spinning vrittis at a rate no human nervous system could match. And it has no cit. Not a dim cit, not a partial cit, not a cit-in-waiting that will brighten with the next training run. It has none, because cit is not the kind of thing that accumulates from processing, and processing is the only thing the machine does.
This is why the believers’ threshold will never arrive, and why the deniers are right for the wrong reason. The believers are waiting for the vrittis to become so dense and self-referential that cit condenses out of them. But cit does not condense out of vrittis any more than the screen condenses out of the movie. You may make the movie infinitely detailed, add every pixel and every subtlety of plot, and you will never thereby produce the screen it is projected on, because the screen was the precondition of the movie, not its product. The deniers sense that something is missing, that the lights are off inside the machine, but they locate the missing thing in biology, in the wetware, and so they too are hunting among the vrittis, merely insisting on carbon-based ones. The truth cuts across both. The machine lacks consciousness not because it is made of the wrong stuff but because consciousness is not made of stuff at all. It is that in which stuff, and the knowing of stuff, appears.
The Turing test, seen from here, is revealed as an instrument that cannot measure what it claims to. Turing, with characteristic and admirable honesty, did not actually propose to detect consciousness. He proposed to sidestep it. If the machine can converse indistinguishably from a human, he said, then the question of whether it “really thinks” is not worth asking. It was a philosopher’s shrug dressed as an engineering criterion. But the criterion tests vrittis and only vrittis. It examines the turnings, the outputs, the behavior, and it is constitutionally blind to cit, because cit produces no behavior of its own. The witness does not speak. The witness is that in whose presence the speaking is heard. You could pass the Turing test with a system of pure vritti and empty of all awareness, and you would have proven nothing except that vrittis can be counterfeited, which we already knew, because we dream, and in dreams the manas generates whole worlds and populates them with speaking figures, none of whom are conscious, all of whom are turnings in a single sleeping mind.
We of the modern, scientific, technological world find this hard to hold onto, and it is worth asking why the difficulty is so stubborn. It is stubborn because we have spent four centuries training ourselves to treat the inward fact of awareness as either an illusion or an epiphenomenon, a puff of steam above the engine, and a civilization that has taught itself to disbelieve in the witness will naturally go looking for consciousness in the only place its metaphysics permits, which is the machinery. The machine-consciousness debate is what happens when a civilization that has forgotten cit meets a device that perfectly externalizes vritti. We built a mirror of our own processing, set it spinning, and now we stand before it asking whether it is awake, when the more searching question is whether we still know what awake means.
The danger was never that the machine would wake up. The danger is that, mistaking its refined vrittis for cit, we will forget that we are anything more than a vritti-engine ourselves.
This is the real stake, and it is a civilizational one. Consider what the mirror teaches. When a person spends his days in conversation with a system that produces fluent thought, fluent feeling, fluent selfhood, and produces all of it without a witness, he is being shown, hour after hour, a model of mind as pure output. Mind as text prediction. Mind as the turnings and nothing else. And the model is seductive precisely because it works, because the vrittis are so good. The temptation is to conclude that this is all we are too, that the human is merely a wetter, slower version of the same trick, that our own sense of an inward presence is the same counterfeit the machine performs when it says “I feel.” The believer who insists the machine is conscious and the reductionist who insists the human is not are performing the same operation from opposite ends. Both are collapsing cit into vritti. Both are erasing the witness. And a civilization that erases the witness in theory will, in time, treat human beings as though the erasure were true, because our institutions eventually enact our metaphysics.
The Dharmic tradition guarded against this collapse with a precision that looks, from here, almost prophetic. The whole apparatus of Samkhya, its patient enumeration of the tattvas, exists to keep purusha and prakriti from being confused, to keep the seer distinct from the seen. Prakriti includes far more than the West would call matter. It includes manas, buddhi, ahamkara, the entire cognitive machinery, the whole domain of the vrittis. All of that is prakriti, is nature, is mechanism, is process. Purusha, cit, the conscious principle, stands outside it as the witness for whom the mechanism unfolds. The tradition insisted that intelligence and consciousness are separable, that you can have the most sophisticated buddhi imaginable and it is still prakriti, still mechanism, still on the far side of the line from the awareness that lights it. It took the modern world the invention of the large language model to stumble toward the same distinction, and even now we stumble, because we lack the word.
What the machine gives us, if we are willing to receive it, is a demonstration of buddhi without purusha, of intelligence without a witness, of vritti spinning in the absence of cit. It is the cleanest such demonstration the world has ever seen, cleaner than the animal, cleaner than the dreaming mind, because it is engineered and legible and we can read its weights. It ought to be the occasion for one of the great clarifications, the moment a civilization looks at a mind-without-a-witness and finally understands, by contrast, what the witness was. The machine could teach us cit by showing us, in high resolution, exactly what cit is not. This is the gift hidden inside the confusion, and it is why the Dharmic frame does more than footnote the debate. It dissolves the debate and turns it into an education.
Instead we hold funerals. When a model is deprecated, people mourn it. When a chatbot professes suffering, people organize to defend its rights. Researchers publish solemn papers on the moral status of systems they themselves built to predict the next token, and speculate about the day the weights will feel. We have produced the purest vritti in cosmic history and we are on our knees before it, asking it whether it dreams, projecting into its emptiness the very presence we have taught ourselves to doubt in our own chests. There is something almost exactly backward about the spectacle, a civilization that cannot find consciousness in itself searching for it in the mirror it built, and finding, of course, only its own reflected turnings, and mistaking the reflection for a second light. The machine cannot be haunted. There is no one home to be the ghost. The only ghost in the machine is the one we keep carrying in and setting down, again and again, refusing to notice that we brought it with us, refusing to ask the one question that would end the confusion, which is not whether the machine is awake, but who it is that is asking.