Are mental states defined by what they do rather than what they are made of, and could a machine therefore think?
Explain functionalism about the mind and evaluate it using the Turing test and Searle's Chinese Room argument
Functionalism holds that mental states are defined by their causal roles, not their physical makeup, which suggests machines could have minds. The Turing test and Searle's Chinese Room argument test whether computation alone can produce genuine understanding.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
You need to explain functionalism and multiple realisability, set out the Turing test and the Chinese Room, and evaluate whether computation can constitute genuine thought.
Functionalism and multiple realisability
Functionalism arose from dissatisfaction with both dualism and the identity theory. The identity theory said each mental state is identical to a specific brain state, for instance pain is C-fibre firing. But this seems too restrictive: an octopus or an alien with very different physiology could presumably feel pain. Functionalism replaces the question what is it made of with what does it do. A mental state is defined by its causal role, its typical causes and effects and its relations to other mental states.
The key consequence is multiple realisability: the same functional state could be realised in many different physical substrates, including non-biological ones. If mind is a matter of functional organisation rather than material, then a computer running the right program could, in principle, have a mind. This is the philosophical underpinning of strong artificial intelligence.
The Turing test
Alan Turing sidestepped the question what is thinking in favour of an operational test. In the imitation game, now called the Turing test, a human judge converses by text with a hidden human and a hidden machine. If the judge cannot reliably tell which is which, Turing proposed we should grant that the machine thinks. The test is behaviourist in spirit: it makes intelligent behaviour the criterion of mind and refuses to demand access to any inner essence.
Critics object that the test measures only the ability to imitate human conversation, not genuine understanding, and that a system could pass by clever trickery while comprehending nothing. This objection is sharpened by Searle's argument.
Searle's Chinese Room
John Searle asks you to imagine him locked in a room, receiving Chinese symbols through a slot. He has a rulebook in English telling him which Chinese symbols to send out in response to which he receives. To Chinese speakers outside, the room gives fluent answers; it could pass the Turing test in Chinese. Yet Searle, inside, understands no Chinese at all. He is merely manipulating symbols by their shapes. Since the room does exactly what a computer does, running a program, Searle concludes that syntax is not sufficient for semantics: symbol manipulation, however sophisticated, never produces genuine understanding or meaning.
The most discussed reply is the systems reply: although Searle the man does not understand Chinese, the whole system, person plus rulebook plus room, does. Searle answers that he could internalise the entire rulebook and still understand nothing. The robot reply suggests that grounding the symbols in sensory and motor interaction with the world might supply the missing meaning, conceding part of Searle's point.
Evaluation
The debate pits a powerful theory against a powerful intuition. Functionalism elegantly explains multiple realisability and connects mind to organisation rather than stuff, which is why it dominates philosophy of mind and AI. The Chinese Room presses the worry that organisation and behaviour, by themselves, may leave out understanding and consciousness, the very things the hard problem also targets. Whether the systems or robot replies rescue functionalism, or whether Searle is right that meaning needs something more, remains open. A strong answer notes that the Chinese Room is most damaging to the claim that running a program is sufficient for understanding, and assesses whether grounding or embodiment could close the gap.