Re: Chalmers: Computational Foundation

From: HARNAD Stevan (
Date: Tue Mar 20 2001 - 19:41:55 GMT

On Sun, 4 Mar 2001, Wright Alistair wrote:

> Wright:
> Chalmers goes on to describe how a physical system can be interpreted to be
> implementing some computation

This actually has it backwards. A computer that is actually doing
computation can (if the computation is nontrivial) be systematically
interpreted as doing something meaningful. For example, it could be
simulating an airplane, flying. So the computing is interpretable as

You seem to be saying that the reverse is true too: That the flying is
interpretable as computing. This is probably just Turing Equivalence
and the Church/Turing Thesis: Just about every physical system, such as
an airplane, flying, can be SIMULATED by a computer.

But this is similar to saying that any picture can be described by a
bunch of sentences. Yes, the sentences can describe the picture as
minutely as we like. And yes, there is a systematic correspondence, an
"isomorphism" between features of the description and features of the
picture ("green" in the description corresponds to green in the picture,
"round" to round, etc.).

But it is the computing that is being interpreted as flying and the
description that is being interpreted as being about the picture, not
vice versa, in both cases.

A computer can implement a flight simulation; a plane, flying, is
flying. It is not implementing a flight simulation!

> > the relation between an implemented computation and an
> > implementing system is one of isomorphism between the formal structure of
> > the former and the causal structure of the latter. In this way, we can
> > see that as far as the theory of implementation is concerned, a
> > computation is simply an abstract specification of causal
> > organization.

For the discussion of causal organization, see:

A computer can simulate the causal structure of flying, but it can't
implement it, because surely part of the causal structure of flying (if
"cause" has any meaning here at all) is the ability to get off the
ground, which the computer lacks!

Thinking may be analogous to flying in this respect.

> Wright:
> Computation which implements a cognitive system should give rise to
> cognition, by definition, without requiring the presence of an
> external observer to nterpret any semantics.

By definition?

> > mental properties [can be divided] into two
> > varieties: psychological properties - those that are characterized by
> > their causal role, such as belief, learning, and perception - and
> > phenomenal properties, or those that are characterized by way [they
> > feel]

If belief, learning and perception did not FEEL like something, then
they would simply be behavioural capacities, and the mechanisms that
generate them. It is the fact that they feel like something that makes
them mental. The Turing Test would indeed be decisive if there were no
feelings. That would mean there was no other-minds problem; there would
be no need to BE the thinker in order to know whether or not it was
thinking. Thinking would just be Turing capacity.

> > the low-level laws of physics are computable. If so, then low-level
> > neurophysiological processes can be computationally simulated; it follows
> > that the function of the whole brain is computable too, as the brain
> > consists in a network of neurophysiological parts.

But what follows from that? Certainly not that a computer simulation of
the brain therefore thinks -- any more than a computer simulation of a
plane flies. It just means the Church/Turing Thesis is correct: Just
about everything in the universe can be simulated computationally --
EXACTLY as just about everything in the universe can be described
verbally. In neither case does it imply that everything in the world is
just computation, or description. Nor that the implemented computation
or the description has the physical or causal properties of the thing
being simulated or described. It is merely a symbol system that is
INTERPRETABLE as having those properties.

Stevan Harnad

This archive was generated by hypermail 2.1.4 : Tue Sep 24 2002 - 18:37:25 BST