Re: uncertainty

["Alan G. Isaac" <aisaac@xxxxxxxxxxxx>]

As several people including myself have emphasized
in the past, we should not expect too much of real
word relevance to flow from ontological stances.
Recall that most mathematicians still cling to a
Platonic characterization of the objects of their
study, with no reduction in the quality of their
mathematics.  Related to this, it is not necessary
to invoke some kind of ontological uncertainty in
order to justify avoiding the silliness of rational
expectations models (as *descriptions* of the world).
To take an extreme case, natural limits on computational
abilities might grant rule-of-thumb behavior a permanent
place in the repertoire of rational human action even in
a fully deterministic world.  What is actually relevant
to PK economics is *epistemological* uncertainty,
whether or not it is rooted in ontological uncertainty.
For example, for human minds, complex deterministic behavior
imposes permanent limitations on our ability to understand
and predict, and it allows for large ``surprising''
changes in economic conditions.  See some of Barkley's posts.

Fundamentally I see no reason to suspect there is
are core PK ontology.
Ontological assertions are essentially religion;
espistemological stances can at least be explored.
PK economists have the advantage of stressing that
actual humans are subject to radical uncertainty about
their economic environments when making crucial
decisions.  References to ontological uncertainty are
just a kind of mystical proclamation that ``it will
always be so''. (Dramatic music, please!)

Alan Isaac



On Wed, 13 Dec 2000 Stephen.Dunn@xxxxxxxxxxxxxx wrote:
> Models based on an ontologically closed view of the world, such as rational
> expectations models (both New Classical and New Keynesian) make knowledge
> (epistemological) claims regarding the informational sets that agents are
> assumed to possess in order to obtain their short-run policy implications.
> Models that suggest that the future is not completely known, i.e. that all
> future state of the world cannot be specified, but can be learned, i.e.
> agents are subject to bounded rationality, invoke epistemological
> uncertainty.  In both these models the ontology of the system is fixed,
> i.e. there are governing immutable laws (stochastic or otherwise). Rather
> the epistemological uncertainty  results from finding out (discovering,
> learning) what these laws are.