FleshFactor FleshFactor


FleshFactor: macho Lagado?

A E C  F O R U M - "F L E S H F A C T O R"

Some remarks on nosology --

When Richard Brown talks about "the expressions we are making via the medium
of machines ... a mirror of our times and an indication, a litmus of our
possible futures", and detects "cultural reflections" in the popularity of
Jim Whiting's "violent robotic monster machines" in Switzerland and Austria,
or of Tamagotchi in Japan -- isn't he invoking a 'model' (a sort of 'mental
machine') expressing (or at least, inviting) some sort of causal (if not
totally deterministic, but rather, statistical and probabilistic) inference
connecting 'signs' associated with people belonging to national groups
(e.g., the things they like) and their intrinsic 'character' or 'essential
nature', i.e. what these people 'are like'?

Might he be more specific? Maybe if he could identify the traits typical of
the various nationalities?? For example -- English are well-mannered,
warm-hearted, worldly-wise, effeminate, voluptuous, consumptive,
self-effacing; the French are frivolous, garrulous, deceitful, childish,
Machiavellian, warlike; Muscovites are malicious, cruel, sanguine, inhuman,
asthmatic, tyrant-loving, and without skills; Spaniards are haughty,
credulous, masculine, susceptible to flattery, constipated, fertile; the
Poles are pleasant, timid, disdainful, braggarts and brawlers; Turks or
Greeks are fickle, superficially clever, lazy, self-loving, anemic. My
sincerest apologies to anyone whose nationality wasn't included; the same
again to anyone whose nationality was. (The list is digested from a table
found in an Austrian painting of the early 1700s, as reproduced in Jack
Goody's "The Domestication of the Savage Mind" ... those damned Austrians, eh?)

It's interesting that this sort of loose, spontaneous, from-the-hip
association and categorizing -- indulged in by everyone, a pleasant enough
pastime -- should be acceptable in casual discourse, and may even pass for
critical thinking, but becomes highly suspect when attempts are made to
quantify or substantiate such judgements, as for example in some stripes of
sociology or anthropology or, for that matter, marketing research. What's
the fear here? What's being threatened? Are we perhaps afraid we'd have to
give up some cherished opinions if it became fashionable to start backing up
beliefs with evidence? I can see how this might infringe on intuition's
territory. But no fear, it's an idea that will not soon catch on, got a
pretty long fuse on it I think.

Ahh -- but then Richard inquires after the why and wherefore of our
infatuation with thinking machines -- I hope I'm not putting words into his
mouth if I read this as asking about the 'agenda' behind the desire to build
an autonomous mind independent of any individual's predication, subjectivity
or prejudice. The old dream of objective knowledge, absolute truth, pure
impartial reason -- Leibniz' vision of a 'rational calculus' or 'universal
characteristic' for mechanically grinding out the truth -- so that any time
a dispute arises, it can be quickly settled by, in effect, doing sums. Lest
you think Leibniz' idea is as absurd as Swift did -- cf. the academicians of
Lagado in "Gulliver's Travels" -- there is a growing movement in
jurisprudence to instruct all jurors, and judges too, in the calculation of
Bayesian probabilities, although no-one's suggesting that we dispense with
unreliable human juries and replace them by machines. Not yet.


Example: A casual search of the web for "jurisprudence AND Bayesian" turned
up this partial list of required courses for a Doctor of Jurisprudence and
Master of Public Affairs degree, from the School of Public and Environmental
Affairs, Indiana and Purdue Universities, Indianapolis. 

::  Statistical Analysis for Public Affairs; 
::  Data Analysis and Modeling for Public Affairs; 
::  Environmental Policy Analysis; 
::  Public Management Economics ... 

(I rest my case.)


What it is, we've been carrying around the wrong idea of 'machine'. We think
of machines, we're seeing clockwork armatures, these great whirling
fly-wheels and gear-trains, spewing out smoke and filth, making a terrifying
racket, a clutter and confusion of moving parts going too fast and too many
different directions to comprehend. But the Futurists'
sado-mechano-eroticism is not, no matter that we may still mistake it for,
an accurate image of machine-nature. No, a machine is a far subtler piece of

A machine is an attempt, any attempt, to articulate the 'how' of something.
The operative word here is 'articulate'. Machines are articulate nature.
Language, in a word -- as in the lovely line quoted in Sandy Stone's book
(who said it?), "The first technology was other people." Or as Jay Michael
Bolter said (in his very wonderful 'Writing Space'): "As writing is deferred
speech, so AI is just deferred writing."

The operator of Japan's Bullet Train, he just sits there and watches the
train executing its program, he can no more intervene than he could fly to
the moon. Same again with the pilot of one of those new-generation Lockheed
jet-fighters, the function is purely symbolic. Anything goes wrong, all he
can do is watch it happen, an expensive emulation of a kamikaze. (The same
again for the Queen of England, come to think of it. She ever tries to
exercise her executive powers, she'd quickly find herself out of a job.
Leviathan manages just fine without a head.)

So now we come to the nub of the pancake, the luminous nose of rationality.
If our rationality is in fact, rational, it is reliable and replicable. It
is, in other words, a machine (or as near as makes no difference). Once
we've figured out the 'program' or 'algorithm' of rationality, well I reckon
we can all just go home. Fade back into the forest edge, gentle Eloi, all of
us be artists and poets and story-tellers, wizard computer animators,
dream-drummers, and really really inventive chefs.

Personally speaking, I wouldn't have it any other way.


Derek Robinson  <drdee@interlog.com>



to (un)subscribe  the Forum just mail to
fleshfactor-request@aec.at (message text 'subscribe'/'unsubscribe')
send messages to fleshfactor@aec.at

[FleshFactor] [subscribe]