i find myself in a variety of awkward positions. i will now try to
disentagle my body parts...
1) i think i have been mistakenly attributed an expertise which i did not
claim. allow me to reiterate the phrase 'interested observer and dilettante.'
i am not an alifer. i am not a pcter. i am a modeller in general, and do not
willingly adhere to any single paradigm, outside of the personal conviction
that cybernetic thinking is right thinking. i do not have, ready to hand,
a grab-bag of arguments for any occasion. i am essentially a tinkerer,
seeking 'the pattern which connects.' i am drawn here to this group because
i find in pct an explanation which seems to work, which is my second
criterion for interest in anything. i am, for the same reasons, drawn to
studies in neural networks, alife, ecology, semiotics, etc. but i have had
too many experiences (in the lab, in the studio, in my kitchen) of just
being wrong, regardless of the strength of my convictions on any given
topic. all of these things, then, including pct, remain as models for me.
i am in the middle of thinking about these patterns, and am simply not in
a position to declare any truth with any authority.
2) despite my initial irritation at his suggestion that alife research is
pursuing false goals, i find myself agreeing with the balance of mr nevin's
subsequent comments. he has clearly articulated his answer to my own
question ('what does he think alife goals are?'), and in so doing has
nicely defined the the current distinctions between alife and pct. i would
particularly like to express my thanks for his interpolation of my
'thought vs. action' phrase... i was interrupted several times during the
writing of that post, and did not make some connections as clearly as i
might have done.
3) i think that this distinction is also indicated by mr andrews' comments,
pointing out that alife is not a single study, say of robotics. what appears
to me to unify the field is the question 'how can organization be seen as
a phenomenon?' put differently, how can order arise from disordered elements,
without explicitly planning for it? certainly, at this stage in the game there
are a lot of simplifying assumptions being made, not the least of which is the
lack of certain kinds of disturbance in the medium in which order is to
manifest itself. and for certain directions of research, these will continue
to be valid simplifications... as in robotics, where much of the activity is
oriented toward obtaining particular features of behavior, regardless of the
mechanism. this, truly, is simply output modeling. but in other areas i think
mr nevin is right... it will be interesting, at some point, to introduce the
pct perspective to the alife toolkit, and see if the systems which order
themselves exhibit the property of 'control'. however, control is not, in my
view at least, synonymous with organization, nor is it the only property of
life, however fundamental. hence my contention that alife and pct are getting
along quite nicely without each other at the moment. (to the extent that they
are in fact getting along without each other. their distinctions can be
overdrawn, i think.) when the processes of organization are well enough
understood that they can be adequately predicted and interpreted, then
further modification to deal with problems of 'consistency in disturbance-
prone environments' will be appropriate. but let's play with one variable
at a time, shall we?
4) thinking about this has just brought a question to mind, the answer to which
might significantly increase my understanding of control theory, and close
the circle on this argument...
if that which is perceived (the effect of action on the environment) is
perceived without distortion (no noise, no disturbance, no delay), and if
the environmental response to action is at the very least deterministic
(possibly linear, certainly consistent, maybe even just transparent),
then isn't the perception of the outcome equivalent to the perception
of the action? it occurs to me that this is the central simplifying
assumption that leads to the criticism that alife is not in fact modeling
living systems (living systems, by pct definition, controlling outcome
and not action, as oft repeated). if so, i hold to my position that the
question most important to alife ('how can anything organize') needs to be
thoroughly addressed before questions of a different order ('how can
anything remain stable') are put forward.
5) so i will not claim that living things are not control systems.
nor will i claim that alife is engaged in producing things that are
fundamentally like living things. neither will i claim, in fact, that
any model which focuses on a given phenomenon can produce anything which
is fundamentally like living things. different features of interest require,
for the moment, different experimental approaches. and if the bug walks,
or the swarm gathers, or the predator kills, or the eye tracks, or the
perceiver controls... then the model is informative.
-----------< Cognitive Dissonance is a 20th Century Art Form >-----------
Eric Harnden (Ronin)
<HARNDEN@AUVM.BITNET> or <HARNDEN@AMERICAN.EDU>
The American University Physics Dept.
4400 Mass. Ave. NW, Washington, DC, 20016-8058
(202) 885-2748
---------------------< Join the Cognitive Dissidents >-------------------