Date: Mon, 26 Sep 1994 09:57:44 -0400 (EDT)
From: "Paul W. Cockshott" <cockshpw-AT-wfu.edu>
Subject: Re: Positive side to anti-LTV case


On Mon, 26 Sep 1994 Steve.Keen-AT-unsw.EDU.AU wrote:

> The questions he posed were (a) whether I could think of social relations
> which would make uv and ev commensurable, (b) whether I thought that
> the value of lp was socially determined, (c) a test to distinguish my
> interpretation of Marx from the LTV.
> 
> (a) Yes, pre-capitalist and pre-imperialist exchange, when exotics
> were swapped between cultures which had no idea of the means by which
> the commodities were produced (see Hilferding on this in his
> reply to BB). Also, in my original post on this I tried to indicate
> ways in which this dialectic "folds back into itself", like any 
> dynamic system will do: workers will demand wages that exceed value,
> but obviously there's going to be pressure in the other direction;
> and money/assets are priced at their use-value.
> 
> Also, my use of commensurable is "non-standard". I use it to mean
> not causally related, as well as different.
Are you sure that this is what you mean?
This sentence implies that for two things to be commensurable they
must not be causually related. There are certainly instances where
things that are commensurable are causaly related. Suppose that
I have a dc to ac converter that takes the form of a dc motor driving
an ac generator. I may legitimately compare the power input  to the
power put out event though the power put in is the cause of the
output.
 
> Generally, uv and ev are
> incommensurable because one is qualitative and the other quantitative;
> in production, the use to which commodities are put is quantitative,
> therefore incommensurability there translates as quantitative
> difference.
>
This is not quite right. Both can be quantitative. A cake mix
specifies  a set of use values 2 eggs, 200 grams flour, 100 grams
sugar etc. These are all quantities, but dimensionally distinct.
To perform arithmetic on them we must express them in units on
a single dimension - as mass, as calorific value, as money.
Exchange values in their general form of prices, are type coercion
operators not simply numbers. Thus we have prices P such
that P_a:(commodity_a -> money)
     P_b:(commodity_b -> money)
etc.
Since exchange values are typed operators and use values are
typed scalars one can not construct a coherent logic
that allows commensuration between them - commensuration 
requires that they at least form a lattice, preferably a totally
ordered set.

It is perhaps unfortunate that few Marxists seem to have paid much
attention to the development of the theory of types in modern
logic. One would almost think that logic had seen no developments
since Hegel.

> (b) Yes, I do think that the value of LP is socially determined--but
> this basis reflects the level of technology as well (the level of
> training, health, etc., is affected by ther level of technology).
> The level of payments above (or below) value is also socially
> determined. I think there's "value" in separating the two (which was
> Sraffa's original intention, BTW), but that it is probably too
> difficult to do in practice (the decision Sraffa also reached).
>
What systematic procedure exists by which one can determine if
the level of wages is above or below the value created by labour
in your theory?
In Marxian theory it is clear, they are above value created when
profits and rent fall to zero.
 
> (c) I made a reply on this earlier--your own work, Paul. From memory
> you said that there was "little to choose" empirically between a
> Sraffian prices basis and the LTV; that is support for one as much
> as for the other. But also, the transformation problem is poerhapsd
> the strongest assertion of the LTV in a testable sense. If one could
> ever control for the differences in pay rates between labor intensive
> and capital intensive industries, and some how statistically
> separate values out of price data, then perhaps you could test the
> assertion that profit rates are higher in labor-intensive industries
> (and I'm being colloqiual here) than in capital-intensive.
> 
I will do the test on the British data.


     ------------------