The software unification of art and science
Jorn Barger
"Empathy, Language and Evolution" and sci.lang was to be
included in the xpost.]
In the same way that Whitehead saw Western philosophy as
footnotes to Plato, I foresee future software design as
footnotes to Will Wright, author of "The Sims"...
Someday your computer will allow you to create a Sims-
style model of everything you've ever done or planned to
do, everywhere you've been, everyone you've met,
everything you own, etc.
So when you use your word processor, it should link the
Sims 'avatars' of whoever you're writing to, and model
the effect you hope to achieve with them, and 'know' how
many copies you'll need to print, and what level of
archiving the document will need.
If you create a sign, for example, it should model the
wall where you plan to post the sign, and guess how long
it will be needed there, so that reprinting it in the
future will require only clicking on the representation
of the sign on its corresponding wall, and choosing
Replace (or Revise or Remove) from a menu of options.
In order to implement this grandiose vision, we'll first
need an exhaustive classification system for all the
_stuff_ in the world. This has been a dream since
Aristotle, with the Semantic Web movement as its most
ambitious current advocates.
And so long as the _stuff_ they're talking about is the
stuff of corporate bean-counters-- grommets and resumes
and tax shelters, etc-- they have a reasonable chance
of success.
But "The Sims" needed a very different startingpoint--
hunger and romance and boredom and sleep and going to
the bathroom!
And when you try to enumerate human motives as part of
your inventory of world-stuff, you open a can of worms
that includes all the last five millennia of vagaries in
the _arts_.
My 500k timeline of knowledge-representation--
http://www.robotwisdom.com/ai/timeline/0000.html
--tries to illustrate this paradox by including a
sampling of humanistic 'knowledge' among the scientific--
Homer and Aeschylus, Dante and Aquinas, Blake and Joyce
and Wallace Stevens-- especially emphasizing their
attempts to analyse human behavior into categories.
I'm sure this problem must have been alluded to from time
to time, by philosophers or linguists or A.I.
professionals... but I'm not sure I've ever seen such an
allusion, and I'm pretty sure I've never seen it directly
addressed.
What I want to propose is that the _language_ of the
humanities and social sciences may be fundamentally
different from the language of the natural sciences, and
so can't be formalised by the same 'scientific method'.
For millions of years before the invention of language,
primates communicated their internal states by a nonverbal
_empathic_ language of shrieks and growls, built on
patterns of pitch and rhythm that were more analog than
'digital'.
This was the language of survival, sex and society, and the
relationships it embodied are the ones the Sims need most.
The classical language of the humanities has evolved to
effectively tap into this intricate intuitive foundation,
but without remotely _articulating_ the underlying system
itself.
The subtle effects of poetry and music must also derive from
this nonverbal language, so a universal classification scheme
that includes humanistic knowledge will someday have to
classify poetry and music as well. (Imagine clicking on your
Sims-avatar when it's in an emotional situation, and hearing
music or reading poetry that captures that exact emotion.)
So the next step for Will Wright should be to include more
of (eg) Georges Polti's "36 Dramatic Situations" (still after
more than a century the most complete survey of dramatic
plotlines).
Such literary surveys always have to be built on a subjective
value-system that ranks some plots as more important than
others.
One way to unify the deepest layer of plotlines might be to
merge them into the lifecycle of a single Everyman, suffering
and triumphing and being supplanted by a new generation.
But it can't be coincidence that this is exactly what Joyce
claimed to be attempting in "Finnegans Wake"...
http://www.robotwisdom.com/jaj/fwake/
Jim Ward
> In order to implement this grandiose vision, we'll first
> need an exhaustive classification system for all the
> _stuff_ in the world. This has been a dream since
> Aristotle, with the Semantic Web movement as its most
> ambitious current advocates.
"The Celestial Emporium of Benevolent Knowledge" divides animals into:
those that belong to the Emperor,
embalmed ones,
those that are trained,
suckling pigs,
mermaids,
fabulous ones,
stray dogs,
those included in the present classification,
those that tremble as if they were mad,
innumerable ones,
those drawn with a very fine camelhair brush,
others,
those that have just broken a flower vase,
those that from a long way off look like flies.
Jorn Barger
http://groups.google.com/groups?scoring=d&q=author:jo...@enteract.com
I want to open with a wild surmise, which I'll then try to
justify:
Computer adventure games often include (lame, imho)
combinatoric puzzles in which you have to find the right
sequence for a small set of elements, to 'unlock' the next
section of the game.
But I think there's a very similar combinatoric puzzle for
unlocking the Meaning of Life (42, idaho), that requires
finding the right sequence for Mortimer Adler's 102 Great
Ideas: http://www.robotwisdom.com/ai/syntopicon.html
Adler offers them alphabetically, which I'd normally scorn
(cf: http://www.robotwisdom.com/web/infodesign.html ) but
I think this is a rare case where his discretion was the
better part of valor.
Adler's list is unique in many ways, but ultimately it's a
sort of meta-ontology, synthesizing many of the greatest
ontologists in history. And he seems to have based his
final selection on the _noble sound_ of the words-- an
instance of what I was referring to in message
news:<16e613ec.04030...@posting.google.com>...
> The classical language of the humanities has evolved to
> effectively tap into this intricate intuitive foundation [...]
I'm sure he must have spent some time looking for a
natural sorting of the 102, comparable perhaps to Roget's
original 1000, but if you try it yourself you'll find it's
very tricky. (An earlier attempt of mine:
http://www.robotwisdom.com/ai/adler.html )
Since Adler's day (1952-- pre-Double-Helix), advances in
biology have made more practical what I call an 'etymogeny'
(cf etymology), a sorting of concepts in the order they
evolved in nature: http://www.robotwisdom.com/ai/etymogeny.html
Careful studies of chimp behavior should allow us to trim
those Great Ideas chimps entirely lack, and then moving
backwards from lemurs to shrews to lungfish, etc we might
reach the first nervous systems, and the behavior of
single-celled microbes.
This oldest level will be the trickiest, sorting out the
origins of _motives_-- especially sex, which predates
neurons.
Adler lumped sex into 'love', presumably because the word
'sex' would have sounded comparatively disruptive. He
omits food and hunger (and hunting), which have perhaps
been given short shrift by philosophers.
The distinction between self and other that forms the
biological basis of the immune system is probably lumped
in with 'soul'.
But here's yet another pre-etymogenic sorting of Adler's
102, in which I finesse the origin of motives by breaking
out three separate 'threads'-- (1) cognition without
emotion, (2)emotions without society, and (3) social
emotions:
1) change sense mind idea form same-other
language sign-symbol [the language of neurons]
element quality quantity one-many relation
space matter time world
memory-imagination experience habit
prophecy [neurons predicting futures]
universal-particular induction hypothesis
knowledge truth logic principle
opposition dialectic reasoning opinion
cause chance necessity-contingency
mechanics physics
definition science mathematics infinity
animal evolution nature man history astronomy
philosophy metaphysics being eternity
2) medicine [as a subset of the missing 'food']
will labor progress
pleasure-pain beauty emotion liberty
desire happiness
courage
wealth prudence temperance
fate art poetry
3) life-death
family love
religion god
sin good-evil virtue-vice judgment justice
punishment wisdom
custom-convention education rhetoric duty honor
war-peace aristocracy tyranny monarchy oligarchy
governmant state revolution law constitution citizen
slavery
soul immortality
ian glendinning
You said ....[Quote]
In order to implement this grandiose vision, we'll first
need an exhaustive classification system for all the
_stuff_ in the world. This has been a dream since
Aristotle, with the Semantic Web movement as its most
ambitious current advocates.[Unquote]
The key here is an exhaustive classification <i>system</i>,
(as opposed to an exhaustive list or set of classifications, or a once
and for all ontology).
I support the view that life, the unverse and everything has "evolved"
and that evolution forms the best "system" for classification.
Interesting that both (platonic / aristolean / scientific) Adler and
(socratic / pragmatic / artistic) Pirsig should both end up with
evolutionary frameworks.
I think you (we) are onto somthing here.
Glad to see you make the Douglas Adams connection too - I think his
links with Dawkins were very telling and would have led to a much more
widely understood view of this vision, had he not been taken from us
prematurely.)
(BTW I'm struggling - but still trying - to fit James Joyce into this
epistemology.)
Regards
Ian Glendinning
www.psybertron.org
bobbyhaqq
ancestor, don't seems to have any ideas, only behaviours. And it is
impossible to determine most concepts.
But I like the animal Idea. I have been doing a good deal of treking
in farm land and their is certainly concepts we share with mamials,
but they don't seem to be of the nature of ideas. We see higher
animals play, they form strong mother child relationships, they grow
familiar with others and come to feel secure with the same places and
animals, they find change difficult. He conserve energy, rest, and
are generally lazy but will exert energy for fun in play, especially
as children. They have curiousity tempered by fear but they quickly
form solid routine notions of the world. They grow up to be
untrustworthy, routine ruled and passive.
The best place to see people is probably a farm, where higher order
animals are placed under the government of society, and we see in
their responses the general themes of our own lives.
But what makes them animals is they have not had the great ideas we
humans have had. They don't really seem to evolve as much as just
explode: love of freedom, awareness of mortality, love beyond group,
tolerance, adventure. Certainly these ideas are grounded in the young
sheeps curiosity or the sheep dogs ability to learn, but about 5,000
years ago they seem to have exploded.
My approach, if anyone cares, to get at the root concepts lately has
been more artistic than observational, though it remains just a though
experiment. We should try to re-invent the late Ice Age humans key
inventions, to get at the key themes of that rapid march to
civiliztion and ideas.