Alan:
Yes, good question. My feeling is this. In the big
scheme of things, a theory can be viewed as a condensed,
abstract model. So I see the theory as the form of the
model in its mathematically pure essence. Newton's laws,
or the Laws of Thermodynamics, for example are theories
as well as things like Petri net theory or the theory
of Discrete Event Systems. Model-based agents can have many
theories as well, and many of you may have some.
Most theories tend to be mathematical
in form, but there appears to be no ultimate reason for
this; it is just that our state of knowledge seems to
lend itself to the centuries of thought that have coalesced
into sub-branches of mathematics such as logic, topology,
and algebra. To a theory, we may apply oft quoted
heuristic from William of Okkham: the theory is good if
it has no fatty substances.
Models, apart from the models-in-the-limit called theories, serve a greater purpose: they link our human sensory
needs, aesthetics, and analogy to bridge the gap between
the minimalist theory and the phenomenon, which undergoes
experimentation. The models are mental and physical tools
for us to come to grips with the reality of a thing, and
we do not throw them away once the theory is distilled,
since they allow us to understand, explain, enjoy, and
reason about the theory and the phenomenon. The model
is the link between theory and reality, whatever that is.
The view that I'm taking here is debatable, and follows
a long chain of philosophy. There are those who wish to
relegate the senses and other human capacities to the
dustbin, and those who revel in these possibilities. So,
it ultiimately comes down to the side of the fence you
stand on. The discussions that I find interesting are
ones, for example, stemming from Pierre Duhem's classic
work "The Aim and Structure of Physical Theory". In it,
Duhem finds himself unable to come to grips with the
"dirty, unorganized English mind" versus the European
mind that thrives upon logic and formalism. Duhem does not
really intend to promote xenophobism, but refers to the
British model makers of the 19th century. It is also
unclear whether this dichotomy is real, but Duhem makes
some interesting points. On the other side of the fence
from Duhem is Campbell (Foundations of Science), Nancy
Cartwright, Mary Hesse, Max Black, etc, who find great
uses for models and they refine why they are useful and
when. I don't know if any of you like this sort of stuff,
but for modelers, it appears to be grist for our mill.
Thoughts?
-paul
-------------------------------------
Paul Fishwick, University of Florida
CISE Department, Bldg. CSE 301
PO Box 116120
Gainesville, FL 32611
web: http://www.cise.ufl.edu/~fishwick
email: [log in to unmask]
phone + fax: (352) 392-1414
=> -----Original Message-----
=> From: [log in to unmask]
=> [mailto:[log in to unmask]]On Behalf Of Alan Penn
=> Sent: Thursday, September 28, 2000 12:55 PM
=> To: simsoc
=> Subject: Re: on modeling and social science
=>
=>
=>
=> ----- Original Message -----
=> From: Paul Fishwick <[log in to unmask]>
=> To: Alan Penn <[log in to unmask]>; simsoc <[log in to unmask]>
=> Sent: Thursday, September 28, 2000 3:42 PM
=> Subject: on modeling and social science
=>
=> I thought I was going to say that I completely agree with Paul
=> and that all
=> I need to say was that what he calls a model is what I call a
=> theory, until
=> I got to this bit of his response:
=>
=> > Ah..this is ripe material. Modeling is at the core of science,
=> > situated between theory and experiment.
=>
=> Paul, would you care to expand on the distinction in your terms between
=> 'model' and 'theory'.
=>
=>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|