(OPE-L) Re: how simple is too simple?

From: Gerald A. Levy (Gerald_A_Levy@MSN.COM)
Date: Sat Sep 04 2004 - 11:38:43 EDT

Hi Ian.

Thanks for the reply.  I agree that this is a difficult question to
answer in the abstract. Yet,  it is an enormously important
methodological and history of thought issue.  I believe that the
most persuasive critiques of marginalism are _methodological_
critiques which focus on whether the assumptions made in the
basic theory prevent an  adequate comprehension of the subject matter.
E.g. Joan Robinson's critique of the Hecksher-Ohlin-Samuelson
theory of international trade ("The Need for a Reconsideration of the
Theory of International Trade").  This form of critique, whereby
the assumptions are challenged based on whether they are adequate
for grasping the subject matter as a historical subject, has been a
major tool employed in the struggle over ideas with bourgeois
theory: it seeks to expose how the assumptions made eternalize,
fetishize, and misrepresent the character of capitalism.    Such
critiques, which often have a historical and empirical component,
have probably represented the primary form of critique of bourgeois
thought since Marx, including critiques by Hilferding, Rubin, Linder
and many others.   It seems to me that we can not on the one hand
advance methodological critiques of bourgeois thought which skewer
and ridicule the assumptions made in those mainstream theories
and on the other hand not subject _our own_ theories to a self-questioning
critique of the assumptions that we make.  We also can not forget
that capitalism is a historically- and socially-constituted subject
which defies overly-general and trans-historical explanations.

> There are two rules of thumb I generally follow: (i) make theories as
> simple as possible, and (ii) make them explain as much as
> possible. Clearly these requirements are contradictory.
> A justification for (i) and (ii) can be found in information theory,
> via a modern form of Occam's razor. Basically there is a smaller
> probability that a simpler model will fool us into thinking we have a
> good model for the phenomena than a more complex model.

I do not think you were on the list yet when we had an extended
discussion of whether/how Occam's razor could be used for
evaluating alternative Marxian theories.  An issue I posed -- that
was discussed in that context -- was whether Occam's razor can
be used as a partial explanation for the superiority of value-form
theory over labor-embodied theory,  surplus approach theory
over other theories,  single-system theories over dualist theories,
etc.  Perhaps you read some of that discussion in the archives.
The contradictory requirements are to, as you say, make the
theory as simple as possible (and legitimate) and at the same time
one that is not overly-general.  E.g.  a theory without money
(or, more accurately, where money is treated as a numeraire)
is simpler than a theory with money, but if one is attempting to
comprehend the subject matter that 'advantage' of simplicity
can inhibit the comprehension of the subject.  It's simpler
to have static, classless, money-less, equilibrium, distribution-only
theories -- but that does not  represent an advantage for such

It is in this context that I questioned whether means of production
can be assumed to not exist in a model which purports to
explain production -- whether simple commodity or capitalist.

More broadly, I think we have to insist upon a minimum specification
of the model/theory that allows us to grasp the subject.  Some
of those minimum institutional specifications should include:

-- commodity-production (and all that entails, including money and

-- the main classes (wage-workers and capitalists) associated
with capitalism (this requires, btw, that v be greater than -0-);

-- surplus value and capital.

Now, let's suppose that a simple (e.g. trans-historical) model
is developed without these minimum specifications.  What would
it mean for how we interpret the meaning of that model?  It
would mean, I believe, that _nothing_ that was derived from that
model could be _assumed_ to also be the case for a model of
_capitalism_.  I.e. whether the conclusions arrived at in the simpler
model could be held to also be the case  for a model of capitalism
can not be determined until we have a 'complete' model
of capitalism in which all of the essential (which include complex)
determinations are grasped.

> For this overly-complex model we can infer that
> anything can happen in the aggregate, in other words we cannot infer
> much at all. As Duncan Foley says somewhere, "it is "methodologically
> too ambitious"; it attempts to model too many things in too much
> complex detail.

Yes, that's an issue as well.   The perspective that allows us to avoid
either simplistic or overly-complex theory is to _always_ keep the subject
matter that we are attempting to comprehend in sight.  (I believe
Marx did this starting with the very first sentence of _Capital_, Volume I).
This means that in the development of theory we be aware of the need to
layer our analysis in a systematic fashion so that we can grasp the subject
matter -- both in terms of its essential nature and in terms of its more
complex unfolding and determinations.  It is precisely one of the failures
of bourgeois thought that they typically lose sight of -- or purposely
obscure -- the subject matter of capitalism.  For Marxists as well, losing
sight of the subject as a totality can trap us in the simplistic world of
elementary theory.  We need to break out of that trap.  Whether one
has a labour-embodied, value-form, surplus-approach (or _whatever_)
interpretation, there is a need to, imho, advance those theories by
systematically examining more complex determinations which, for instance,
grasp the state, international trade, and the world market.   If all of
those (and other) heterodox streams of thought could develop such
enriched theories then we could be able to comparatively evaluate those
theories in a richer way.  This doesn't mean that disagreements over
'simple'  theory should be avoided or will go away afterwards: what it
suggests instead is that all Marxian theories need much further internal
development before we can have the most meaningful and complete
cross-paradigm evaluation of the subject.  The above comments are
directed at _all_ Marxian interpretations rather than just your own.

In solidarity, Jerry

This archive was generated by hypermail 2.1.5 : Sun Sep 05 2004 - 00:00:00 EDT