Another great series of postings! I think I will have one more brief go,
and then retire back to my writing. I am in the middle of various
sociological articles on these issues, and once they become available I
will notify the mailing list. All of this is controversial in sociological
theory as well; even so, I believe those controversies can be instructive
for MABS practice.
Although I am tempted by the philosophical discussion, I've decided to take
this "off list" and have responded directly to several of the postings.
Regarding the issue of modelling the environment (Alan): Yes, many MABS
explicitly model the environment, and it's true that such a model includes
things other than individuals. In sociological terms, however, that model
is individualist because it does not model any social groups or entities.
(Unlike some of Carley's work, which explicitly models organizational
structures, for example.) Then there could follow an interesting debate
about how some environments--like the architecture of a building--are
fundamentally social, and whether or not that sort of social constraint is
different from other sorts of institutional constraints, but that's not
appropriate for this list. In any case, in most MABS, the environment is
not "social" in the way that architecture is.
Regarding the term "emergence" and Rosaria's discussions of emergent social
phenomena: These are crucial issues and I don't have an answer, I'm only
raising questions. Emergents do not exist and are epiphenomenal, under
most philosophical accounts of emergence; the concept has been widely
discussed in recent philosophy of mind.* Yes, MABS are interested in how
macro patterns emerge from the actions of individual agents, and some MABS
researchers talk as if those macro patterns are "real" (when I said they
were not real at MABS 2000, there was some heated discussion!) But in MABS
accounts of emergence, the macro pattern is epiphenomenal--its causal power
attains only in virtue of the underlying causal powers of the composing
agents.**
I think there is a way to successfully argue that such emergents can
participate in causal laws, even if they are not real, but that is a
difficult philosophical argument (in preparation). For practicing MABS
researchers, I think the message is (1) there is a lot of support in both
philosophy and sociology for proceeding as if social entities actually
exist; (2) social simulation does not have to proceed from strict
individualist assumptions. I am glad to see that several of those posting
agree that MAS simulations can incorporate modelling of social entities,
and I find it interesting that no one has argued that MABS should not do
this, especially considering that only a few have.
*Kim, J. Supervenience and mind. 1993, Cambridge.
*Beckermann, Flohr, Kim. Emergence or reduction? 1992, de Gruyter.
**Sawyer, R. K. (1999a). The emergence of creativity. Philosophical
Psychology, 12(4), 447-469.
R. Keith Sawyer
Assistant Professor
Program in Social Thought and Analysis
Washington University
Campus Box 1183
St. Louis, MO 63130
314-935-8724
http://www.artsci.wustl.edu/~ksawyer
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|