I must confess that I feel rather like a hillbilly at Ascot in this discussion.
I am just a simple modeller using simulation techniques to describe how people
see their worlds or to find plausible ways of replicating key features of
observed social phenomena or their statistical signatures. This is somewhere
between history and journalism but using a more formal language that gives me
less ambiguity and more power than normally achieved by historians and
journalists but at the expense of expressiveness.
The discussion on this list over the last few days seems analogous to a post
modernist critique of verbal history or journalism. For the post modernist, only
the text matters. But for the historian or journalist the text is meant to
capture real events and it is the events rather than the text that matters. I
presume that the historian or journalist would be amused (or mayby pixilated) by
the post modernist critique but would not be influenced by it.
My approach is always to specify individual agents with some representation of
cognition that reflects or is inspired by independently validated results from
relevant disciplines (mainly cognitive science and social psychology) and then to
investigate the consquences of their behaviour when they interact with other
agents. The range of knowledge, skills, motives, etc. that individual agents
might have is determined by the (usually empirical or system design) problem I am
addressing in that model.
I guess this is what Keith calls 'the "trial and error" of successive programming
of simulations, observing what fails and what works, and expanding the
methodology when it seems necessary.' That seems an appropriate approach to me
for two reasons:
Good theory -- indeed good science -- seems to come from good observation and
representational social simulation that is validated against evidence and domain
expertise is one kind of good observation (along with careful history and
journalism). If our trial and error approach develops means of representing
social processes that are systematically well validated, it will doubtless be
appropriate either to call these representational means a theory (thus giving
rise to a long discussion on a variety of email lists about whether it really is
a theory) or to identify more abstract and formal means of determining when some
representational means are appropriate and when they are not (again giving rise
to a long discussion on a variety of email lists about whether that is a theory).
Whether any of this makes me a nominalist or a methodological individualist or a
social realist and so says something about my ontology, epistemology and probably
my proctology doesn't seem at all important to me. But then I am probably
missing out on a lot of fun.
--
Professor Scott Moss
Director
Centre for Policy Modelling
Manchester Metropolitan University
Aytoun Building
Manchester M1 3GH
UNITED KINGDOM
telephone: +44 (0)161 247 3886
fax: +44 (0)161 247 6802
http://www.cpm.mmu.ac.uk/~scott
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|