Re: Use of the word "logicism" in AI

[z007400b@bcfreenet.seflin.lib.fl.us (Ralph Silverman)]

David Longley (David@longley.demon.co.uk) wrote:
: In article <829285594snz@campion.demon.co.uk>
:            rbj@campion.demon.co.uk "Roger Bishop Jones" writes:

: > In the introduction to "The Philosophy of Artificial Intelligence", (OUP 1990)
: > Margaret Boden uses the word "logicism" for the view that:
: > 
: >    Our basic common-sense knowledge can be formally axiomatised.
: > 
: > Can anyone tell me whether this usage is generally accepted in the AI
: > community and if so where the usage originated?
: > 
: > Roger Jones    http://www.rbjones.com/
: > at home        rbj@campion.demon.co.uk
: > 

: There's an obvious problem here. There are no catalogues of "common-sense"
: to formalize, despite the promises of CYC and the manifesto of a few bold
: GOFAI researchers. 

: Similarly, apart from the relational notion of IQ, there's no workable
: definition of "intelligence", which is why, I thought, Turing came up
: with the "Turing Test" (which is, nevertheless, still a relational measure
: not a property.

: It is not just laymen who are immersed in implicit 'folk psychologism', and
: my criticism that the whole concept of AI is a "category mistake" in no way
: vitiates any of the excellent work which is undertaken my very able 
: researchers, in the name of "AI", all over the world. 

: But I think we would all do well to seriously consider the thesis that the
: notion of "Artificial Intelligence", like the concept of "mind", is indeed
: a category mistake, and that in reality it is no more or less than what we
: have in our ever evolving repertoire of extensionally based science and 
: technology. 

: Beware of Cognitive "Scientists" bearing promises of deep insights <g>...
: -- 
: David Longley

--
*************begin r.s. response**************

	common sense,
	as ordinarily understood,
	is,  largely
		learned
		^^^^^^^
	.
	since artificial systems can
		learn
	;
	the path to
		common sense in ai
	need not pass through
		axiomatization of this
	...	

*************end r.s. response****************
Ralph Silverman
z007400b@bcfreenet.seflin.lib.fl.us




--

Ralph Silverman
z007400b@bcfreenet.seflin.lib.fl.us