The brain in a vat

What it tests How can you know you are not a brain in a vat, and if you cannot, can you know anything about the world outside your mind?

The case

Suppose that while you slept a scientist removed your brain, placed it in a vat of nutrients, and connected it to a computer that sends it the very signals your eyes, ears and skin would have sent. You wake, as it seems to you, in your own bed. You see the ceiling, feel the sheets and read the news. Nothing about your experience is different. But there is no bed, no ceiling and no news: only a brain, a vat and a computer.

The sceptical argument

1. If you know you have hands, you know you are not a handless brain in a vat.
2. You do not know you are not a brain in a vat, since your experience would be exactly the same if you were.
3. So you do not know you have hands.

The same pattern threatens almost everything you believe about the world. This is the modern form of the doubt René Descartes raised in 1641 with his evil demon, a deceiver of great power who might be feeding him every experience he has. Descartes’s way out ran through the certainty of his own thinking and the existence of a God who is no deceiver; few now accept that route, which is why the problem is still alive.

The replies

Moore’s reply. G. E. Moore held up his hands and said that he knew he had them. He was more certain of that than of any premise in a sceptical argument, so the reasonable response is to run the argument backwards: since I know I have hands, I know I am not a brain in a vat.

Putnam’s semantic reply. Hilary Putnam, who made the vat scenario famous in Reason, Truth and History (1981), argued that it undermines itself. Words refer to the things they are causally connected to. A brain that has always been in a vat has never interacted with real brains or vats, so its words “brain” and “vat” refer to images or computer states. When it thinks “I am a brain in a vat”, it is thinking something false. So whoever can truly think the thought is not a brain in a vat. Critics note that the argument works only for a lifelong envatted brain and depends on a contested theory of reference.

Contextualism. Some epistemologists hold that “know” shifts with context. In everyday contexts you know you have hands; in the philosophy seminar, where vat scenarios are raised, the standard rises and you do not. Both claims can be true in their contexts.

Externalism about justification. If knowledge requires only that your beliefs be produced by a reliable process, then in the actual world, where your senses work, you know you have hands, even though you cannot prove from the inside that you are not envatted.

Why it still matters

The vat is a philosopher’s tool for isolating a question: what connects your experience to the world, and what entitles you to trust that connection? The same question arises whenever information reaches you through a channel you cannot inspect. An answer from a language model is a modest version of it: you receive fluent content, and its relation to the world is hidden from you. What you can know through such a channel, and what you have to check yourself, is taken up in Should you trust what ChatGPT tells you? The probabilistic cousin of the vat, the simulation argument, asks whether we might actually be in one.