Dear All
Despite some excellent simulation work (as mentioned by Leigh, or
others included in the proceedings of the social simulation events of
the last ten years, from SimSoc to ICCS&SS, from ABS to MABS, etc.),
the question whether simulation is a good tool for explanation
(whether sociological or not is another point, which I will return on
later in this message) is an important question to pose.
One of the contributors to this discussion (I am sorry I don't
remember who) said that simulation does not explain per se, but at
most (I think s/he used exactly these words) confirms preexisting
theories.
Well, I think this point deserves some further consideration. Put it
simply and clearly, simulation is not and should not be taken as
equal to theory! Simulation is a generator of phenomena. Personally,
I believe this to be extremely important for building experimental
and innovative theory, but it is not the theory itself. To produce a
scientific theory of a given phenomenon, it is not enough to generate
it. Otherwise, social beings would have a scientific understanding of
their society just because they produce it!
In my view, a scientific theory requires
- an explicit representation of both the essential requirements of a
given phenomenon and
- of the mechanisms and process by which these requirements bring it about.
This latter point allows me to go back to Thomas initial question
about sociological explanation. I am not a sociologist and therefore
I am perhaps not entitled to say words about this specific quesiton.
I will make a more general point.
I think that only when both the aspects just mentioned are included,
a theory can be said to be a scientific theory of any given
phenomenon at the appropriate level of analysis.
To see why consider tha classic question raised by the evolutionary
models of social action: what is minimally required for, say,
altruism or cooperation to evolve? Well, this is not a sufficient
question to ask. Even if answered adequately, this question would not
return us a theory of social phenomena at the appropriate level of
explanation. Ultimately, what is minimally required, indeed, other
than individuals or, still more minimally, particles moving in the
space?
In order to provide an adequate scientific theory of altruism,
cooperation etc. we need to know both what are the ingredients
involved at the foundational level and how they bring about the
effect we are interested in. We need an explicit theory of the what
and of the how.
Simulation per se cannot replace the theory, but it can help us
constuct it, by accomplishing both tasks.
Rosaria Conte
National Research Council, Institute of Cognitive Science and
Technology, V.LE Marx 15, 00137 Roma.
LABSS (Laboratory of Agent Based Social Simulation)
& University of Siena - Communication Sciences - "Social Psychology"
voice:+39+06+86090210;
fax:+39+06+824737;
cell. 3355354626
email: [log in to unmask] - http://www.istc.cnr.it/lss/
|