Why I don't like the title 'Marxian Empirical Research'

Alan Freeman (a.freeman@greenwich.ac.uk)
Fri, 16 Jan 1998 13:00:11 +0000


Thanks for the many responses my NIPA post generated and I'm just going
through them to respond. I couldn't do that earlier because I wanted to get
NIPA(2) out first since it saves repeating too much

This is a quickie on the subject heading. This seems to have changed to
'Marxian Empirical Research'. I'm not happy with that title and I don't
consider it to describe what I proposed. At the risk of seeming pedantic
I'd like to say why.

I prefer a less presumptions title that specifies precisely, but neutrally,
what is the proposed activity.

The title we settled on for the 1997 IWGVT (EEA) sessions was 'Quantifying
Values Using Published Data'. We chose this after a lot of discussion, and
I like it for the following reasons:

(1) I don't like making grandiose claims. I'd rather just say what we are

(2) I don't like tying our own research to the name of Marx where it
conveys no extra information and serves only as a claim to authority.

(3) If we do it in Marx's name and get it wrong, Marx will get the blame
and not ourselves.

(4) I think non-Marxists will want to do the same thing and using Marx's
name to describe it sounds like a kind of Masonic condition of
admission, as if only True Believers can join.

The generic title 'Quantifying Value' doesn't exclude anyone, it's
sufficiently clear what is being done to provide for collaboration, and it
makes no claims beyond what is actually being undertaken.

A further point is this: I agree with Fred (13 Jan) who says he doesn't
want 'to invert matrices to derive labor-time estimates'. Neither do I,
and though I don't want to stand in the way of those who do, I don't
want to oblige anyone to do it.

But the matrix-inverters have more or less cornered the tag 'Marxian
empirical research'. At the risk of rousing sleeping dogs I want to
re-iterate that I think it's neither Marxian nor empirical, though I'm
happy to call it research.

So, first off, I'm not happy about any title that introduces any confusion
about the relation between my proposal and this activity. I don't want to
exclude matrix-inverters but I think the new title would convey the
impression that what they tend to call themselves is the generic title for
the project I'm proposing, which it ain't.

More generally, calling anything 'empirical research' begs the prior
question of the its source. I don't think all work with data is empirical.
In earlier posts I argued, and still maintain, that widely-reported
correlations between 'value' and 'price' have little empirical content.
They are a 'logical fact' about the data, of the same order as the fact
that a matrix has as many columns as rows: necessarily true of all
reasonably possible data.

In working with data, we must always enquire into its theoretical basis
and question it to ascertain when so-called 'facts' are in reality purely
internal constrctions of the theory.