[ show plain text ]
Alfredo (and others who agree with his position) has proven a bit too much.
All empirical work is "messy." Are we then to abandon empirical research?
There is no empirical work that will ever perfectly conform to theoretic
concepts. Even something as socially obvious as "race," a phenomena that
all Americans either acknowledge, accept, or in some way act upon, has
10,000 thousand empirical problems. Indeed, race (like value) has a great
many theoretical problems. Nevertheless, it is not futile to do empirical
research where we employ the category "race," even though there is no
scientific definition of the concept!
Value is a scientific concept. Attempts to statistically measure value
categories simply represent the most concrete form of theoretical analysis.
The issue then isn't whether the statistics are flawless - they never are.
The issue is whether empirical initiatives help us to better understand the
relative importance of theoretical categories, help us to better understand
the empirical significance of particular problems and issues, and whether
the expected theoretical relationships have merit. Whether a given
empirical is "satisfactory" depends on the issue under discussion.
Again, no empirical procedure is flawless. If value is visible only through
price, but we can't get any information about values from prices, then
value theory is a waste of time. Under these conditions, value theory is a
needless metaphysical (and probably idealistic) exercise.
Even something as "simple" as one's "wage" or "income" involves an enormous
series of assumptions. These assumptions must be considered as more or less
problematic within the context of the problem at hand. The same is true of
the relationship between value theory and value estimation.
A claim of Marxian economics is that it provides a superior understanding
of social reality, in comparison to the supposedly obsfuscating tendencies
of neoclassical economics. If this claim is true, let's go to the data and
peace, patrick l mason
At 01:38 PM 5/24/00 EDT, you wrote:
>I agree with Andrew *and* Michael and, hopefully, with Paul and Allin.
>Only concrete labour time and market prices are observable magnitudes. Even
>the calculation of prices of production involves strong assumptions,
>especially averaging out market prices over long periods *and* somehow
>calculating the impact of technical progress and the changes in the quality
>of the goods. This exercise necessarily involves some degree of
>arbitrariness, as Michael explains very clearly in his papers.
>Within production, average techniques of production can be inferred if the
>actual techniques are averaged out, but the problems are similar to those
>It is even worse with labour; the reduction of skilled labour to uniform
>abstract labour is terribly difficult (for reasons brilliantly explained by
>Philip Harvey in his 1985 RRPE paper). Averaging out concrete labours, as in
>I-O tables, simply will not do because the skills will remain different - as
>in hours of ditch-digging and computer programming, which create different
>quantities of value.
>In my opinion, these difficulties *prevent* the calculation of prices of
>production from the technologies of production (however we may conceptualise
>the wage rate). Yet *this is not a problem at all*, because there is no
>reason why one should entertain this possibility or have any interest in
>exercise, for the reasons that Andy explains - namely, value is a social
>relation which expresses the mode of production under capitalism, and price
>is its form of appearance. There is no other form of apperance of value but
>The attempt to calculate production prices from I-O tables seems to me to
>reflect a certain dissatisfaction with this fact (ie, that price is the form
>of value, and value is social labour), as if we wanted calculate the price
>vector just to make sure that our concept of value is, somehow, 'right'. The
>point is that value is not hidden within commodity prices, and does not need
>to be 'found'. Value is visible, but only - I stress the word only - through
This archive was generated by hypermail 2b29 : Wed May 31 2000 - 00:00:11 EDT