Re: Response to Gonzalo

[Jim Devine <JNDF%LMUACAD.BITNET@xxxxxxxxxxxxxxxxx>]

I have a question, which I'm afraid has probably been asked before,
but I missed the answer:

why is the "generality" of a theory so important?

A general
theory is one that includes a lot of other theories as special
cases, right?  In the standard vision of Keynes as presenting
a general theory, which differs from Davidson's, Keynes
drops the classical full-employment assumption. However,
the full-employment case can be captured within Keynes'
theory. Einstein's "general relativity" theory includes
special relativity as a case, etc. (By the way, I
think that this is a less ambiguous definition of a
theory's "generality" than the number of its axioms,
since number of axioms can be lowered by combining
them, etc.  See also Barkley's comment to which this
is a reply.)

But a general theory might also be too abstract and not
have any implications about the empirical world which
we wish to understand. In order to say anything about
the world, we may have to deal with special theories
rather than general theories.  The general theory of
customs unions, for example, says that breaking down
trade barriers among countries might promote international
trade -- but that the negative effect on trade between
the countries that join the customs union and those
that don't might cause international trade to fall when
all is said and done. That's a nice, general, theory, but
it doesn't automatically say anything at all about
NAFTA.  So we need to move to a special theory of trade
within North America, which the general theory can
help us formulate. (To my mind, a model of that economy
with coefficients based on econometric estimation is
simply a special theory; it's not a complete description
of empirical reality, which is probably impossible.)

To me, this suggests that we shouldn't make a fetish
about a theory's generality (not to mention the number
of its axioms).  We should start with the question that
we are trying to answer. That helps us figure out
how to deal with the trade-off between generality
and empirical relevance.  The key is whether the
theory, general or special, helps us understand the
question at hand, perhaps allowing us to make conditional
predictions and guidelines for action.  Generality _per
se_ is secondary here, a possible means to an end.
We also care about the "special case" of the empirical
world.

sincerely,

Jim Devine
jndf@xxxxxxxxxxxxxx or jdevine@xxxxxxxxxxxxxxx
Econ. Dept., Loyola Marymount Univ., Los Angeles, CA 90045-2699 USA
310/338-2948 (daytime, during workweek); FAX: 310/338-1950