Re: [OPE-L] Ajit's Paper on Sraffa and Late Wittgenstein

From: Ian Wright (wrighti@ACM.ORG)
Date: Sat May 27 2006 - 15:26:11 EDT

Hi Ajit

> Hi, Ian! Thanks for reading and commenting on the
> paper. I find your above statement about Sraffa's
> mistake quite perplexing, since the question of
> capitalists' consumption or non-consumption does not
> come up anywhere in Sraffa's scheme including the
> dated labor approach.

Indeed. It is omitted.

> The dated labor approach only
> shows you how the price can be resolved into wages and
> profits only. The power to (1+r) according to "dates"
> does not imply any historical dates--it represents the
> unending chane of production at a given point in time
> only.

Yes I know. I did not claim that the "dated" representation refers to
historical time.

The error in Sraffa's dated labour representation is that the price of
money-capital, r, is not reduced to its labour cost. It's a real-cost
accounting error to not reduce it.

It necessarily follows from Sraffa's surplus equations that
money-capital is a commodity and r is its price, once the physical
distribution of the surplus is specified. Section 4 of the paper
deduces this result.

There's a simple low-dimensional numerical example of a corn economy
at the end of the paper that presents the Sraffian real-cost
accounting error. The results hold for arbitrary numbers of

>Sraffa's system does not allow movement of time,
> so the question of what happens to the commodities
> produced at a given point of time has no implications
> for his results. I think you are clearly making a
> mistake in interpreting Sraffa here.

For precision and clarity, we should first answer this question: Do we
interpret Sraffa's surplus equations to represent a state of
self-replacing equilibrium or not?

If the answer is "no", then real costs in general are undetermined,
including labour-value. Nothing much can be said about real costs as
we're talking about a novel distributional event with an undistributed

If the answer is "yes" then real costs in general are determined,
including labour-value. But in this case Sraffa's real-cost accounting
is faulty, as manifested in the dated labour representation.

> Not really! All you have to do to get over this
> confusion is to think of Sraffa's system with a given
> rate of profits, as Sraffa does, and not given wages.
> Your confusion will simply melt away. Cheers, ajit

Section 3 of the paper goes into detail.

Best wishes,


This archive was generated by hypermail 2.1.5 : Wed May 31 2006 - 00:00:03 EDT