Mike,
I think that the central point of language is that it can be treated as
consisting of general, abstract, "open-ended scripts" (the last being
another way of describing "concepts").
The value of language then is that I can tell you "Go to the movies for 2
hours" - and I do not have to tell you any of the vast details of what to
do - how to go - which transport to take, which movie to watch, which
cinema to go to, how to divided your time, and so on. Or I can tell you "Go
and buy me something nice for supper" and again I can leave the complex
details up to you. Or I can say "The cat sat on the mat" and I don't have to
draw you a detailed picture.
All words - all "scripts" - leave the individuals concerned - both speaker
and listener - immense latitude as to how to interpret them. (To the narrow
AI, convergent mentality this is terrible. To a broad AGI divergent
mentality it is a great virtue. It enables you, for example, to adapt to
dynamic environments - to change your route to, and choice of movies if
unforeseen obstacles arise - whereas a narrow AI program, that held your
hand every step of the way, would get you stuck).
The disadvantage of course of language's open-endedness is that it can leave
room for considerable misunderstanding as to what are and are not proper
details of a given script (or proper individual concretisations of those
general abstractions). I might get upset, for example, if you didn't leave
the house but watched movies on the house TV, (strictly a legitimate
interpretation of my command).
An additional advantage of language is, as you indicate at the end, that
different individuals can agree on certain basic interpretations of any
given set of words, and yet bring in addition their own rich associations -
fill in those scripts with different details. We may agree that "an
open-ended language is the sine qua non of AGI" and yet each have v.
different associations with language/AGI etc. - which can be mutually
enriching.
P.S. The only thing I disagree with you about is that I don't think language
is much use for analogies - I think they are derived primarily from
graphics/ schemas and images.
MD: MT>> An open-ended, ambiguous language is in fact the sine qua non of
AGI.
Thankyou for indirectly pointing that out to me.
Would you agree that an absolutely precise language with zero
ambiguity would be somewhat stifling for use in a "creative" mode?
It seems to me that new points are discovered when different observers
attempt to relate their positions relative to a third point of
discussion. The analogies, misunderstandings, reconciliation, and
meta-symbols that are required for even the simplest agreement often
generates more context about the other party in the conversation than
the point upon which they eventually agree.
you think?
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?&;
--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.503 / Virus Database: 269.16.9/1155 - Release Date:
11/27/2007 8:30 PM
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=70039594-936fe6
