**From:** Salcedo Afonso (*afonso@mac.com*)

**Date:** Tue May 01 2001 - 18:17:35 BST

**Next message:**Yusuf Larry: "Re: Chalmers: Computational Foundation"**Previous message:**Godfrey Steve: "Re: Turing: Computing Machinery and Intelligence"**Maybe in reply to:**Henderson Ian: "Harnad: The Symbol Grounding Problem"**Next in thread:**Sparks Simon: "Re: Harnad: The Symbol Grounding Problem"**Messages sorted by:**[ date ] [ thread ] [ subject ] [ author ] [ attachment ]

*> HARNAD:
*

*> It is still a criterial, if not a definitional property of
*

*> computation that the symbol manipulations must be semantically
*

*> interpretable (...) globally.
*

Salcedo:

Does this mean that for some operation to be considered a computation it

has to make some sense to whoever is interpreting it? Harnad gives the

example of primitive algebra symbols such as '+' and '='. When students

learn algebra basics such as adding and subtracting, they are also

taught what patterns or rules should be used for the formulas to make

any sense. We know that "=+224" wouldn't make any sense in Algebra, but

"2+2=4" would. There is an intrinsic semantic importance to what is

syntactically expressed as being computation.

If formulas from Quantum Physics would be seen by anyone else not

related to the field, then they would still know that formula

represented some computation, although it didn't make any sense to them.

The thing is, they would only conclude it to be a computation if they

recognized any of the symbols they learned before, such as '+' or '-'.

Computation is therefore not computation if it doesn't make sense in

whole or in part.

*> HARNAD:
*

*> It would be trivial to say that every object , event and state of
*

*> affairs is computational because it can be systmatically interpreted as
*

*> being its own symbolic description. A cat on a mat (...) Why is this
*

*> not computation? Because the sapes of the symbols are not arbitrary in
*

*> relation to what they are interpretable as meaning, indeed they are
*

*> precisely what they are interpretable as meaning.
*

Salcedo:

I'm making some confusion with this. Imagining someone with its own

native language, completely different from english. If that person read

the sentence "cat on a mat", the symbols would not be directly

associated with its meanings... actually they could be meaning something

completely different. Would that be computation, as the symbols would

then be arbitrarily related to what they could mean?

What about metaphoric expressions such as "It's raining cats and dogs".

Why is this not computation? For anyone not familiar enough with English

this would mean a completely different thing than for an English person,

and the symbols would not be directly associated with what they truly

mean in the real world.

**Next message:**Yusuf Larry: "Re: Chalmers: Computational Foundation"**Previous message:**Godfrey Steve: "Re: Turing: Computing Machinery and Intelligence"**Maybe in reply to:**Henderson Ian: "Harnad: The Symbol Grounding Problem"**Next in thread:**Sparks Simon: "Re: Harnad: The Symbol Grounding Problem"**Messages sorted by:**[ date ] [ thread ] [ subject ] [ author ] [ attachment ]

*
This archive was generated by hypermail 2.1.4
: Tue Sep 24 2002 - 18:37:30 BST
*