Alan Penn wrote:
>My next question. In our field MAS, MABS etc. what are the 'real' entities
>in this sense?
I have been reading a lot of philosophy and sociological theory lately on
realism vs. nominalism, exploring the concept of emergence in MAS (see my
MABS 2000 paper). There has been a lot of foundational sociological work
focused on exactly this topic. Within this discourse, the discipline of
MAS is sociologically nominalist, and anti-realist about social entities:
only the agents exist, and all emergent higher-level patterns are
epiphenomenal.
This is implicit in MABS methodology (to connect to your reference to
Hacking), which explicitly models the agents (the "real" entities) but does
not attempt to explicitly model macrosocial phenomena (because they are not
"real" but simply aggregates of acting individuals).
(There are some simulations that are exceptions...)
R. Keith Sawyer
Assistant Professor
Department of Education
Washington University
Campus Box 1183
St. Louis, MO 63130
314-935-8724
http://www.artsci.wustl.edu/~ksawyer
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|