The discussion of different modes of exploitation raises some very
important points. In this case I think Marx's conceptual vocabulary is
actually quite precise and unambiguous. The important thing is to
distinguish the relation through which surplus labor is exploited from
the form the surplus labor takes. If a Roman usurer collects interest
from a spendthrift heir, the form the surplus value takes is interest,
and thus is part of the primitive circuit of capital that underlies
usury. But the ultimate source of the surplus value is the exploited
labor of the serfs working the latifundium which is the wealth of the estate.
I think John Roemer's intervention in this area, whatever other merits it
may have, is actually more confusing than enlightening. Roemer adopts a
different definition of exploitation from Marx, one based on individuals
rather than classes, and rooted in neoclassical general equilibrium
theory. I have written at some length on this in a review in Foley,
Duncan, 1989, Roemer on Marx on exploitation, Economics and
Politics, 1:2, (July) 187-199, so I won't repeat myself here. Some of
capitalist surplus value undoubtedly actually represents other forms of
exploitation that the wage-labor relation (for example, exploitation of
consumers through interest, or in the form of residential rents), but it
is possible in principle to sort this out, and the great bulk of the
surplus value of industrial capitalist societies clearly arises from wage
labor.
Duncan