Subject:     		Q. on Penrose arg: why new *physics*?
From:         		[email protected] (Anders N Weinstein)
Date:         		1997/08/03
Organization:	University of Pittsburgh
Message:		5s0kn5$

Since some contributors have raised the Penrose Godelian argument again,
I am moved to express my own puzzlement about one feature. 

Suppose for the sake of argument I accept the claim that human
mathematicians can perform mental feats no computer can (insofar as it
is acting as a computer.)

Well, so what? In particular, why would the *physical* science of the
brain have to be revised to account for this? After all, we explain
things only relative to a description. But physical science explanation
does not employ descriptions like "believes that p" or "proved a
theorem".

Perhaps it could turn out that whenever we undertake to view a
mathematician through the lens of physical science concepts, we never
encounter any "swerves", i.e.  transitions inexplicable by the lights
of contemporary physical science. But when we view the mathematician
through the lens of cognitive concepts like "just proved such and such
a theorem", we find them doing things no machine can do.

In that case the world of objects of experience would always and
irremediably contain this sort of double aspect.  If we distinguish them,
we might even be moved to talk about two worlds. But physical science
itself would not have to change. For such concepts as belong to the
second world are basically outside the explanatory purview of physical
science. 

I think there is an assumption here that physical science really ought 
to explain the human mind, because it ought to explain everything. But that
would seem to be simply false.