Just a *quick* note on emergence as time is limited -- one study in which i
have recently become "immersed" has obviated my participation in the
discussion but has provided what seems, at times, an incredible opportunity,
and at other times, the frightening reality of power struggles in the
workplace. Anyway this situation has also obviated my moving my own modeling
inquiries from the MIT software to the Sante Fe institute software, which i
hope will give me a better understanding of what I am actually modleing.
The physics "emergent" modelers are, to the best of my knowledge" not really
"bottom up" -- it is a false claim. Per Bak's work, for example, uses the
conservation of energy (as do others inexplicitly) and in physics this is a
principle that spans the scales of individual element to collective
description - so it is perhaps simultaneously top down/bottom up -- but
arguable neither. Similarly, some early work on modeling phase transitions
in solid state physics used scale invariance as a means of searching for
critical points, i.e. the Kadanoff block transformations (I recall a 25-30
year old Sci Am article that might provide a good overview). Scale
invarience is the idea that certain kinds of order look the same at all
scales -- but only at these critical points. Here the modeling strategy is
to vary the system and do the search across all scales for the same order --
if one finds such a configuration, one is at a "critical point".
But something like conservation of energy, which in physics modeling could
arguably be "top down" perhaps throws back the curtain and reveals top/down
as red herrings. Except for temporary quantum violations, this is a physics
principle that holds at all scales of description -- "scales" here meaning
quantitative differences in distance, and time, and perhaps other dimensions.
In contrast, Giddens argues that large scale social structures ("structures"
being how we describe the observed patterns at these scales) are enacted by
individuals -- and to that extent the large scale processes "exist", it is in
the day to day actions that each person takes, in social and material
interaction with other people. So when i read text at the policy level about
how policies "interact" i immediately become suspicious, because the
theoretical description that spans from individual to the societal is just
not well formed at all. The contrast of policies, which are meant for
collective regulation, and the theoretical description of "top down"
processes are something i'd like to explore more.
The comparable case in physics is statistical mechanics, which enables a
movement in scale from the particle level to the ensemble level -- its
relation to the macroscopic description of thermodynamics might be
interesting to contrast to the problems of up/down debates. For example, one
difference is that a physicist is unlikely to say that the conservation of
energy is "enacted" at the particle level. Hmmm.. even there the level of
one particle is not a good unit of analysis, because the principle is
formulated for one of interaction between particles -- a supra-particle unit
of analysis, i.e. If one sums up all the energies before some particles
interact, and then lets them interact, and then sums up the energies after,
the numbers are the same.
But i really have to go -- duties call. Just hope i confused things a bit.
bb
This archive was generated by hypermail 2b29 : Mon Jul 01 2002 - 01:00:07 PDT