Scott wrote:
> Two kinds of constraints are involved: social constraints arising from
> webs or networks and individual constraints such as the impossibility of
> consuming more than you have or buying more than you can pay for. So
> when Bruce says that it is almost a truism that constraining individual
> behaviour makes that behaviour more predictable, I imagine that he is
> referring to individual constraints whereas, if I understand them
> correctly, Kathleen and Rosaria are referring to social constraints.
OK, in my terminology the type of constraint is determined by what it
acts upon rather where it comes from. Thus a social constraint acts
upon society (e.g. the society can not consume more than it produces).
An individual can consume more than she produces, so society attempts to
propogate the constraint down to individuals via rules (e.g. money).
Thus money is a constraint upon individuals. Money is less of a
constraint upon societies (it can print more or even change the rules
governing it).
Here is a clear example (as long as one avoids the idiocy of assuming
that the presence of such constraints mean that individuals will
optimise action with respect to these constraints only).
In a recession the monetary constraints will be tighter on individuals
than in good times. Is there any evidence that the economy is more
predictable during recessions than periods of growth? None that I have
seen.
This is also a good example for distinguishing social embeddedness from
merely being part of a social system. If (as some economists would have
it) people traded *as if* they only took into account global information
such as price, then individuals could model this and make economic
decisions in a manner that essentially ignored the presence of other
individuals in the system. Here the fact that the constraints
originated from the economic web would be insufficient evidence to say
the agent was socially embedded as that agent could be acting entirely
asocially.
On the other hand, if people acted so as to take into account
specifically social information (e.g. "Fred likes Globo coffee, so I'll
try it"), so that the effects of these social interchanges did not
cancel out and cause emergent phenomena at the social level, then agents
would have a motivation to take this sort of information into account.
It has far more operational content to say that an agent in such a
position is socially embedded, because that agent needs to deals with
the causal effects of individuals.
Regards.
--------------------------------------------------
Bruce Edmonds,
Centre for Policy Modelling,
Manchester Metropolitan University, Aytoun Bldg.,
Aytoun St., Manchester, M1 3GH. UK.
Tel: +44 161 247 6479 Fax: +44 161 247 6802
http://www.cpm.mmu.ac.uk/~bruce
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|