[OPE-L:290] Re: An odd fact brough to light by Duncan Foley's post

Paul Cockshott (wpc@clyder.gn.apc.org)
Thu, 19 Oct 1995 14:45:16 -0700

[ show plain text ]

Some thoughts on the conceptualisation of
money in historical materialism.

Paul Cockshott

I had promised Jerry that I would send some comments
in on topics that I thought were open questions in
Capital I. I am in effect being hurried along by
Alan's last posting:


I make one point which might clarify things a bit: In
Marx and still, I think, in the real world, there is a
process which converts money (proper money) into
loanable capital, there is a process which converts
loans back into money, and there is a distinction
between the two.

Suppose for simplicity I keep money in a bedsock in the
form of a gold hoard. If, as a result of a high interest
rate, I take it out of the sock and loan it to someone,
then I have converted money into loanable capital. If, as
a result of a low interest rate, I call in my due loans
and issue no new ones, then I have converted loanable
capital into money. This is what, perhaps erroneously, I
thought Paul was referring to.

I had been intending to post something saying that
in my opinion one of the things that needs updating
is Marx's treatment of money, which, in the modern
world is anachronistic. This has been brought home
by Alan's post, which I think, shows the restrictions
we labour under if we try to understand modern money
in terms of a theory of a more archaic social form.

Whilst talking of holding bedsocks full of coins may
have a certain couthiness about it, its relation to
the modern world is slight. In practice, we know that
today money takes the form of digital records in
the computers of the banks. Gold has long ceased
to have a monetary role, and state money in the form
of notes and coin plays a diminishing role in circulation.

Rather than 'let the past hang like a dead weight on
the brains of the living', we should recognise that
this development - rather than indicating that modern
money is unreal or a swindle - tells us something about
what it was about gold money that allowed it to
function as money.

We have to recognise that money has always been
just information, and that its value was incidental
to its information processing function.
Money is the information structure that encodes
social relations of debt. It encodes the fact that
a given juridical person, who may be a human or
a company etc, is entitled to a certain portion of
the aggregate social labour product.

Consider what a stock control computer in a supermarket
does. It models how many packets of cornflakes there
are on the shelves by storing a number in its memory
which represents the packets on the shelf. Every
time a packet goes over a scanner at the till, this
number is decremented. For such a modeling process
to work two principles have to be observed:

1. Correspondance of state: to each distinct state
of the modeling system there must exist a distinct
partition of the states of the system being modeled.
This can be approximated by saying that there should
be a one to one correspondance between the states
of the two systems.
2. Correspondance of evolution: to each allowed state transition
of the modeling system there must be a corresponding
possible transition of the real system between

Now a stock control system is a system of economic
accounting in terms of use values, and it need not
be performed using a computer. The earliest record
keeping systems such as those of the proto Elamite
societies of the Tigris Euphrates basin, take the
form of records of stocks of grain sheep and cattle.
Prior to the formation of written number systems
these took the form of clay tokens representing
animals etc. These evolved into stylised tokens
representing quantities of things, with particular
shapes of tokens representing 1s 10s, 60s etc.
In due course these evolved into written numeral systems.
The social function is infered by archaeologists to
be the recording of movements of animals and grain
between temples.

A good account of this is given in Ifrah 'From one
to Zero' and in Polanyi 'Trade and markets in
early empires'. What appears to have occured is
that an agent of the temple is dispatched with
accompanying herd to another temple property, carrying
a sealed clay envelope holding the tokens that
represent the number of sheep he is taking. At the
other end the envelope is opened and the tokens
compared with the sheep to verify the transfer.
At a subseqent stage in the evolution the envelope
is replaced by a clay tablet with impressions of the
tokens pushed into it. These in turn developed into
a written numeral system and from there into written

What this indicates is that the need to model economic
activities, initially as physical models, then as
symbolic models, produces numeral systems and then
writing. Initially this modeling is done in terms of
physical quantities of objects, but, many centuries
later we find that the records include a unit of
account - the sheckle, which was not a coin but a
standarised weight of barley. The sheckle becomes
a unit of account to express transfers of mixtures
of goods between temple and palace complexes.

We can analyse the formal / algebraic requirements of
such a modeling system, I will refrain from doing this
here since I hope to circulate a paper through the
value group on this shortly, and because it takes
up space. But it is easy to see that there is a
homeomorphism between the algebras of value, sheckles,
gold coins and modern credit money. Any one of the
latter can act as a model for relations of debt, which are
claims on the former, ie, claims on social labour.

What then is a bank doing when it credits my account
with 40 pounds?

It is recording a ternary social relation of the form
<bank {oe} ockshott, 40>
This social relation will probably exist in a relational
database in its computer, but 50 years ago the same
relation would have existed in ledgers written in
India ink.
What this indicates that the formal and semantic
properties of a modelling system are independent of
the material means by which it is achieved.

It is easy to see what the supermarkets computer is
modeling when it records packets of Kellogs, but
what is the bank modeling?

Formally it is modeling a promise to pay me 40 pounds,
the 'real money' that Alan is so concerned with.
But if I go along to my bank Adam and Co and ask for redemption,
they will ( this being Scotland ) issue me with
banknotes issued by one of the clearing banks say
the Royal Bank of Scotland. These elegant sheets of
paper have written on them that the directors of the
Royal Bank of Scotland promise to pay the bearer
on demand the sum of Five Pounds.

If I then proceed to the headquarters of the Royal Bank
of Scotland and ask for my Five Pounds, I will recieve
a Bank of England Five pound note. This has again
written on it the same motto.

I make my way to Threadneedle street, and what does
the Bank of England do - refuses to let me in as
it does not do business with people like me - but if
we suspend disbelief, it will hand over 5 English
pound coins - metal tokens inscribed with the motto
'Decus et Tutamen', a thing of worth and beauty.

Whilst these are quite attractive, and are even designed
to look like gold sovereigns, they are made of a
copper nickle alloy whose scrap value is modest.

We seem to be in a hall of mirrors, a recursion whose
fixed point hides itself. Ironically what we terminate
with are updated versions of the proto-elamite counting
tokens, disks whose shape and size denotes their
numeric value. Strictly speaking what we have is
a composition of one to one mappings, which, like all
such compositions preserves the original structure.
But what is being modeled now, has the very peculiar
property that it would not exist but for the ability
to model it.

Packets of Kellogs exist indepenently of their computer
records, but bank accounts do not exist outside of
the records that the banks keep. In general debts
only exist in the records of debt. Destroy the system
of records, and the system of debt goes with it.
Hence the demand by the indebted peasants of Attica
to smash the mortgage marker stones, or the burning
of the money and ledgers of the Pnom Phen banks when
it was taken by the peasant armies of Pol Pot.

Gold coins are just one more method of recording
social relations. The information is real, and so
long as the information persists, the social relation
is too, but to search for 'real money' behind
the current techniques of recording these relations
is mistake a recording technology for the records.
Gold coinage was a recording technology that addressed
2 problems that any such system of recording social
relations encounters;

1. Durability of the records.
2. Prevention of falsification.

Gold lasts, and is so expensive to produce that, with
suitable choice of coin weights, forgery can be made

Nowadays we trust to replicated computer systems,
good backup tapes, and with the internet, public
key cryptosystems to achieve the same ends. But in
the end, all of these technolgies are devoted to
recording the same thing - entitlement to labour.