Re: on money substance and abstract labor

From: Ian Wright (iwright@GMAIL.COM)
Date: Fri May 28 2004 - 18:17:57 EDT


Hi Howard

>The form of value is an organizing principle that constitutes the
>product of labor as a commodity.  It is a form, an organizing principle of
>social labor.  So what does it organize?  It can't be a way of organizing
>tailoring or brickmaking as such because it also has to be a form capable of
>organizing lumbering and carpentry and opera singing.  So it is an
>organizing principle set over against labor in general.  There is the
>materiality of labor activity on the one hand and the organizing principle
>of a given social form of labor on the other.  You put the two together in a
>product of labor and the product has value.

This is excellent and also very clear. I've nothing substantive to add
to your post, but I thought it might be interesting to note that there
are other ways of talking about the "form of value" or "organising
principle", which I find more natural.

In the case of a simple commodity economy it is particularly clear
that the "form of value" is an emergent control system, in the full
engineering sense of the term, which allocates social labour according
to monetary control signals. The control system is implemented in the
actions of the economic agents, who may or may not be conscious of its
existence. A simple commodity economy has interesting parallels with
reinforcement learning algorithms and neural networks, both of which
use a quantitive
representation of value to adaptively allocate internal resources.

The two, maybe unfamiliar, elements of this point of view are: (i)
the notion of the unplanned emergence of novel ontologies, and (ii)
the notion of a control system being implemented in the distributed
activity of human actors.

Taking case (ii): we're used to the idea of algorithms being multiply
realised in different material substrates -- e.g., the algorithm for
adding two numbers together has been multiply realised in lots of
human minds, mechanical devices, such as an abbacus or Babbage's
difference engine, and modern computers. We're less used to the idea
that an algorithm may be implemented on human components. But it can
be, although the robustness of the implementation is quite different,
due to the complexity and unpredictability of the human components,
compared to components explicitly designed for robustness, such as
transistors, logic circuits etc. A control system can be thought of as
a type of embodied algorithm.

This is what I understand the "organising principle" to be -- an
emergent, economic control system.

>Also, I don't need the prior existence of capital here or the reduction of
>skilled labor to labor anybody can do.  On the other hand, as I argued to
>Paul, I may very well need some such reduction in order theoretically to be
>able to conceive of a causal organizing principle unspecific with regard to
>weaving, house building, etc.  But obviously the causal potency of the commo
>dity form doesn't depend on my being able to conceive it.

I agree -- whether we have an adequate understanding of the process
and current state of the reduction it nonetheless occurs in reality.
As Krause ("Money and Abstract Labour") clearly explains, the exchange
of money for commodities induces an equivalence relation on the
different concrete labours. The concept of abstract labour is
maintained by the economic control system and is emphatically not an
everyday mental generalisation performed by the economic actors,
although this is a concomitant phenomenon. Abstract labour is
represented on a social scale, not an individual one, and the
reduction coefficients of one type of concrete labour to another are
continuously calculated via a distributed computation implemented in
market exchanges.

That all labour must be reduced to simple or unskilled labour, that is
one kind of concrete labour, is a mistaken idea that derives, I think,
from ignoring the unique
representational role of money.

Nothing new here -- just different language perhaps.

ATB,

-Ian.


This archive was generated by hypermail 2.1.5 : Sat May 29 2004 - 00:00:02 EDT