Relation to the world· Mind and consciousnessnamed middle

Not by program alone

Axis: Can a machine think — Could a machine ever truly think?

Running a program will never be thinking: simulating understanding is not understanding. But nothing rules out that an artifact might think one day, if it reproduces what matters: , or a body's engagement with a world.

Who holds this position

Philosophers 1

Searleexplicit

The : shut in a room, Searle manipulates Chinese symbols following a rulebook, and his answers pass for a speaker's, without his understanding a word. A program is only a syntax, a set of rules bearing on shapes; understanding further requires a semantics, a relation to the world that formal manipulation never supplies. Yet he does not conclude that only the living think: the brain is a machine, and an artefact reproducing its causal powers would think. What he refuses is therefore not the thinking machine, but the idea that running the right program suffices.

"Minds, Brains, and Programs" (1980)

Arguments and objections

ArgumentThe syntax-and-meaning argument — Running a program will never be understanding; but what the machine lacks is not life, it is the causal powers of the brain.

A program is defined entirely by syntax, that is, by the shape of its symbols: one runs it correctly by following rules that speak only of the figure and place of signs, never of what they designate. That is exactly what lets one and the same program run on any machine.
But understanding is not handling signs, it is knowing what one is talking about, reaching the semantics and not the syntax alone: the signs must relate to something other than themselves, which is called . A system that has only form at its disposal has, of itself, no way of reaching meaning, and adding rules changes nothing, since they are still rules about forms.
It will be objected that a simulation, if perfect, comes to the same thing. But simulating is not reproducing: the simulation of a storm wets no one, that of a digestion feeds no one. There is no reason to make an exception for understanding.
The brain, however, genuinely does produce thought: it therefore does not merely run a program, it has causal powers of its own, belonging to what it is and not only to what it computes. Nothing forbids an artifact from possessing them one day, but it will have to reproduce them, not simulate them.

Therefore So running a program will never be thinking, without thought being the privilege of the living: what our machines lack is not life, it is the causal powers of the brain, or a body's engagement with a world.

ObjectionAttacks the reasoningThe systems reply — The middle position denies to organization what it grants to the brain, without saying what the brain has that it lacks.

The argument concludes from no part's understanding that the whole does not understand. It is a step refused everywhere else: none of your cells reads these lines, none of your neurons speaks English.
Understanding is therefore never in a component, but in the whole system, in the organization that links them: to look for it in whoever runs the rules is to help oneself to the answer in advance.
There remain the causal powers of the living brain, invoked to mark the difference. But so long as one does not say which ones, nor why another material would lack them, this is no explanation: it is a name given to the gap one wishes to preserve.
And the criterion turns around: our brain seems to us to understand because we know it from within. Seen from outside, it is only an assembly of components not one of which understands anything.

Therefore The middle position holds only by granting cerebral matter a privilege it does not define: either organization suffices, or one must name what is missing.