The non-histor

of capitalism
Elen Meiksins Wood
Since historians frst began explaining the emergence of capitalism, there
has scarcely existed an explanation that did not begin by assuming the
very thing that needed to be explained. I Almost without exception,
accounts of the origin of capitalism have been fundamentally circular:
they have assumed the prior existence of capitalism in order to explain its
coming into being. My intention here is to sketch a kind of potted history
of these question - begging explanations and to consider their implica­
tions. I shall start with what has been called the 'commercialisation
model', which has its origins in classical political economy and
Enlightenment conceptions of progress. This model is arguably still the
dominant one, even among its harshest critics, including both the demo­
graphic explanations that claim to have displaced the commercialisation
model, and also most Marxist accounts, from the transition debate which
started in the '50s until today. But there does now exist an important
alternative account, and I shall end by considering that too.
I want to emphasise straight away that my whole argument assumes
the radical specificity of capitalism, the uniqueness of its principles and
laws of motion. My criticisms of many historical accounts depends on
that assumption and on the conviction that what makes these accounts
profoundly inadequate is not only their weakness as historical explana­
tions but their tendency to obscure the nature of capitalism as a specifc
social form. There is perhaps, a sense in which my argument is itself open
to the criticism that it begs the question precisely by assuming that there
is indeed something requiring explanation. But I shal simply assume that
readers are willing not only to grant that there is such a thing as capital­
ism but also that there is something sufciently distinctive about it that
the question about its origin is a reasonable one.
The market as opporunit
The easiest way to sum up the dominant conceptions of capitalism is to
say that for them, the capitalist market represents an opportunity rather
than an imperative. Even Marxists who seem very attuned to the coercive
force of the market in capitalism often treat the histor of capitalism as if
the issue were the liberation of an opportunity rather than the imposition
of an imperative. Virtually every defnition of the 'market' in the dictio­
nary implies an opportunity: as a concrete locale or institution, a market
is a place where opportunities exist to buy and sell; as an abstraction, a
market is the possibility of sale. Goods 'fnd a market', and we say there
is a market for a service or commodity when there is a demand for it,
Historical Materialism
which means it can and will be sold. Markets are opened to those who
want to sell. The market represents (according to The Concise Oxford
Dictionar) 'conditions as regards, opportunity for, buying and selling'.
The market implies ofering and choice. Now of course these definitions
do not preclude market forces, and, on the face of it, force implies coer­
cion. But in the conventional language of capitalist ideology the market
implies not compulsion but freedom. If the impersonal regulatory mecha­
nisms of the market are in any way coercive, it is simply in the sense that
they compel economic actors to act 'rationally' so as to maximise choice
and opportunity. This implies that capitalism, the ultimate 'market soci­
ety', is the optimal condition of opportunity and choice. More goods and
services are on offer, more people are freer to sell and proft from them,
and more people are freer to choose among and buy them.
Critics of capitalism, and Marxists in particular, do seem to start from
the premise that the most distinctive and dominant characteristic of the
capitalist market is not opportunity or choice but, on the contrary, com­
pulsion. Material life and social reproduction in capitalism are universally
mediated by the market, so that all individuals must in one way or anoth­
er enter into market relations just to gain access to the means of life. The
dictates of the capitalist market - its imperatives of competition, accumu-
proft-maximisation, and increasing labour-productivity - regulate
not only all economic transactions but social relations in general. All this
may seem familiar to Marxists, but I shall argue that this perception often
gets lost even in their historical accounts of capitalist development.
This, then, is where the main problem lies: the tendency to look upon
capitalism as an opportunity rather than an imperative. The fatal flaw in
conventional accounts of the origin of capitalism is that they fail to
acknowledge that the market became capitalist when it became compulso­
r. They tend to suggest that capitalism emerged when the market was
liberated from age-old constraints and when, for one reason or another,
opportunities for trade expanded. Capitalism here represents not so much
a radical qualitative break from earlier forms as a massive quantitative
increase, an expansion of markets and a growing commercialisation of
economic life.
The commercialisation model
There are many different versions of this history of capitalism, but most
of them are rooted in the classic commercialisation model, which goes
something like this: with or without a natural inclination to 'truck, barter
and exchange' (in Adam Smith's famous phrase), rationally self-interest­
ed individuals have been engaging in acts of exchange since the dawn of
history. These acts became increasingly specialised with an evolving divi­
sion of labour, which was also accompanied by technical improvements in
the instruments of production. Improvements in productivity, in many of
these explanations, may in fact have been the primary purpose of the
increasingly specialised division of labour. So there tends to be a close
connection between these accounts of commercial development and a
kind of technological determinism. At any rate, capitalism - or 'commer­
cial society' i.e. the highest stage of progress - represents a maturation of
age-old commercial practices and their liberation from political and cul­
tural constraints.
Non-histor of capltaUsm
But even if these tendencies are ascribed to something as universal as
human nature, it turns out that the constraints on them have been nearly
as universal and that they were decisively lifted only in the West. In the
ancient Mediterranean, commercial activity was already fairly well
advanced, but its further evolution was interrupted by an unnatural break
- the hiatus of the Middle Ages. For several 'dark' centuries, economic
life was again fettered by irrationalism and the political parasitism of
landlordly power. The classic explanation of this interruption has to do
with 'barbarian' invasions of the Roman Empire. But there is a later and
for a while very influential version of this model which is somewhat dif­
ferent. Henri Pirenne blamed the rupture of the Mediterranean commer­
cial civilisation on the Muslim invasion. He argued that this had the effect
of suppressi ng t he old commercial system by cl osi ng off the
Mediterranean trade routes between East and West. A growing 'economy
of exchange', led by a professional class of merchants, was replaced by an
'economy of consumption', the rentier economy of the feudal aristocracy.
In Pirenne's version, commerce revived when the trade routes were
reopened; and in all the versions of this story, commerce revived with the
growth of cities and the liberation of merchants. This time, however,
there emerged cities with a distinctive and unprecedented autonomy,
cities devoted to trade and dominated by an autonomous burgher
This class would free itself once and for all from the fetters of the old cul­
tural constraints and political parasitism. The urban economy was thus
liberated, and with it came the liberation of commercial activity and mer­
cantile rationality. Of course this process was accompanied by the
inevitable improvements in techniques of production, which seem,
according to these arguments, inevitably to follow from the emancipation
of trade. The liberation of these already existing impulses was apparently
enough to account for the rise of modern capitalism. The implication of
all this is that the specifcity of capitalism simply disappears. The age-old
practice of commercial profit-taking in the form of 'buying cheap and
selling dear' is not, in these models, fundamentally different from capital­
ist accumulation, proft created in production and through the appropria­
tion of surplus-value. Certainly there was, according to the classic model,
a major shift from feudalism to capitalism; but the transformation seems
not to have been in the nature of trade and markets themselves. The
change was rather in what happened to the forces and institutions - polit­
ical, legal, cultural, ideological as well as technological - that had imped­
ed the natural evolution of trade and the maturation of markets.
If anything, in these models it is feudalism that represents the real his­
toric rupture. The resumption of commer
ial development, beginning in
the interstices of feudalism and then breaking through its constraints, is
certainly treated as a major change in the history of Europe. But feudal­
ism itself seems to be something like a course-correction in a more or less
directional historical process, even if that process was temporarily
deflected-albeit drastically and for a rather long time. These assumptions
tend to have another important corollary, namely that towns and trade
were by nature antithetical to feudalism, so that their growth, however it
came about, undermined the foundations of the feudal system. But if feu­
dalism, according to these explanations, had derailed the progress of
commercial society, the intrinsic logic of the market never signifcantly
Historical Materialism
changed. From the beginning, it involved rationally self-interested indi­
viduals maximising their utilities by selling goods for a proft whenever
the opportunity presented itself. More particularly, it involved an increas­
ing division of labour and specialisation, requiring ever more elaborate
networks of trade, and, above all, ever-improving productive techniques,
to cut costs in order to enhance commercial profts. This logic could in
various ways be suppressed, as it has been in most times and places. But
in principle, the logic of the market remained always the same, always an
opportunity to be taken whenever possible, always conducive to economic
growth and the improvement of productive forces, always bound eventu­
ally to produce industrial capitalism if left free to work out its natural
What gets lost here are the imperatives specifc to capitalism, the spe­
cific ways in which the market operates in capitalism, its specfic laws of
motion which uniquely compel people to enter the market and compel
producers to produce 'efciently' by improving labour-productivity - the
laws of competition, profit-maximisation, and capital accumulation. I
seems to follow, then, that there is no real need to explain the specific
social property relations and the specifc mode of exploitation that deter­
mine these specific laws of motion. There is, in fact, no need really to
explain the emergence of capitalism at al, because it is assumed to exist,
at least in embryo, from the dawn of history, if not at the very core of
human nature and human rationality. In fact, history itself has proceeded
by the laws of capitalist development: a process of economic growth sus­
tained by developing productive forces. If the emergence of a mature cap­
italist economy requires explanation at all, all we really need is a expla­
nation of the barriers that have stood in the way of its development, and
of the process by which those barriers were lifted.
Afer the classic commercialisation model
That, then, is the basic commercialisation mode!. There have been vari­
ous refnements of that model, from Weber to Braude!. I should probably
hasten to explain that when I attribute this model to someone like Weber,
I certainly do not mean that Weber failed to see that a fully developed
capitalism emerged only in very specifc historical conditions and not in
othes. It is true that he was more than willing to see some kind of capi­
talism even in classical antiquity. But he did, afer al, set out to distin­
guish Europe from other parts of the world; and he did, of course,
emphasise the uniqueness of the Western city and European religion,
especially in order to explain the unique development of Western capital­
ism. The point, however, is that he always tended to talk about the factors
that impeded the development of capitalism in other places - their kinship
forms, their forms of domination, their religious traditions, and so on -
as if the natural, unimpeded growth of towns and trade and the liberation
or autonomisation of towns and burgher classes would by defnition give
you capitalism. Weber also, it should be added, shares with many others
the assumption that the development of capitalism was a trans-European
(or Western European) process - not only that certain general European
circumstances were necessary conditions for capitalism but that all of
(Western) Europe, for all its interal variations, followed essentially one
historical path.
Non-histor of cpltaUsm
More recently, there have been frontal attacks on the 'commercialisa­
tion model' in general and the Pirenne thesis in particular (which is now
generally out of favour). The most important of these has been the demo­
graphic model, which attributes European economic development to cer­
tain autonomous cycles of population growth and decline. But it is not
really clear that the underlying premises of the demographic explanation
are as far removed from the 'commercialisation model' as its exponents
claim. They certainly challenge the primacy of expanding markets as a
determinant in European economic development. But they continue to
assume that development was determined by the laws of supply and
demand. Those laws might be determined in more complicated ways than
the commercialisation model could account for, in response to the com­
plex cycles of population growth and de
line or Malthusian blockages.
But the nature of the market and its 'laws' is never really put in question.
The demographic model may not expli
itly defend the old convention. It
may not explicitly deny that the capitalist market is qualitatively different
from, and not just quantitatively larger and more inclusive than, markets
in non-capitalist societies; but neither does it represent a frontal challenge
to that convention, and it seems to take it effectively for granted.
In one way or another, then, whether by processes of urbanisation and
growing trade or by the cy
lical patterns of demographic growth, the
transition to capitalism in all these explanations is a response to the uni­
versal and trans historical laws of the market, the laws of supply and
demand. Needless to say, neo-classical economics has done nothing to
displace these assumptions - not least because it is generally uninterested
in history altogether. A for historians today, the ones who are interested
in the longue duree are likely to belong to the demographic school, unless
they are less interested in economic processes than in menta lites or dis­
course. Other historians, especially in the English-speaking world, are
generally suspicious of long-term processes of so-called social change
anyway and are more interested in very local or episodic histores and in
proximate causes. They do not actually challenge the existing theories of
long-term development so much as they merely dismiss or evade them.
The new wave of historical sociology is different. It is, of course, pri­
marily interested in long-term processes of social change. But even here
there is a tendency to beg the question in various ways. For instance, in
one of the most important recent works in this genre, Michael Mann
explicitly adopts what he himself calls a 'teleological bias', according to
h industrial capitalism is already prefigured in medieval European
social arrangements.2 Not surprisingly, for all its complexities his argu­
ment situates the driving force of European development in the 'accelera­
tion largely of intensive powers of economic praxis' and the 'growth of
the extensive power of commodity circuits" - in other words, technologi­
cal progress and commercial expansion. And, yet again, his explanation
depends on the absence of constraints - in particular, the autonomy of
various actors resulting from an essentially acephalous social organisation
(with the addition of the 'rationalism' and the normative order provided
by Christianity). Furthermore, private property was allowed to develop
into capitalist propert because no community or class organisation pos­
sessed monopoly powers. In short, not only the emergence of capitalism
but its eventual and apparently inevitable maturation into its industrial
Historical Materialism
form are explained above all by a series of absences.
It seems, then, that if only by default, the traditional 'commercialisa­
tion model' still prevails, whether on the surface or in more subterranean
form. There has been one major exception to this widespread pattern,
Karl Polanyi, though even he had certain afnities with the dominant
model: It is, in any case, striking how little his important book, The
Great Transformation, has managed to affect the dominant model, even
though there now seems to be a revival of interest in Polanyi. In general,
we are still where we were. Either the question of capitalism and its ori­
gins simply does not arise at all, or else, even when questions are raised
about how and why capitalism did emerge in some special case or cases,
they tend to be overtaken by another question: why capitalism did not
emerge in others. Take the idea of 'failed transitions' as a way of describ­
ing what happened - or failed to happen - in the commercial city-states
of Northern Italy, or in the Netherlands. That phrase 'failed transition'
says it all.
The Marist debate
The most complicated case is the Marxist debate on the transition from
feudalism to capitalism. Matters are not helped by the fact that there are
two different narratives in Marx's own work. One of them is very much
like the conventional model, where history is a succession of stages in the
division of labour, with a transhistorical process of technological
progress, and the leading role assigned to burgher classes who seem to
bring about capitalism just by being liberated from feudal chains. In fact,
capitalism already exists in feudalism, in a way. It exists, to use Marx's
own words, in the 'interstices of feudalism', and it emerges into the main­
stream of history when it 'bursts asunder' the fetters of feudalism. This is
basically the narrative of the German Ideolog and the Manifesto. And
this is the narrative at least implicit in traditional Marxist ideas of 'bour­
geois revolution'. But there is another story, or at least the foundation of
one, in the Grundrisse and Capital, which has more to do with changing
property relations, especially in the English countryside. Instead of
exploring Marx's own ideas, however, I want to look at more recent
Marxist histories. I shall leave out of account altogether the crudest kinds
of technological determinism, which have all too often passed as Marxist
theories of history, and shall concentrate instead on the most serious and
challenging Marxist accounts.
Some years ago, Robert Brenner accused some Marxists of being
'neo-Smithians'.5 He had in mind especially Paul Sweezy, Andre Gunder
Frank and Immanuel Wallerstein. Now he did make a powerful case
about the way some Marxists have effectively swallowed the assumptions
of the old commercialisation model, the tendency to treat the specific
dynamic of capitalism - and its need for increasing labour-productivity­
as an inevitable outcome of commercial expansion. But there was some­
thing more complex going on. So let me just look briefly at the so-called
Transition Debate which began in the '50s with an exchange between
Paul Sweezy and Maurice Dobb and which went on to draw in a whole
range of distinguished Marxist historians. The central question at issue
between Sweezy and Dobb was where to locate the so-called 'prime
mover' in the transition from feudalism to capitalism. The main debate
Non-histor of capltaUsm
came down - or appeared to come down - to this: Sweezy emphasised
commercial expansion, the opening of trade routes and the growth of
trade, explicitly following Pirenne. Dobb insisted that trade was not the
solvent of feudalism. In fact, trade and towns were not inherently inimical
to feudalism at all. Instead, feudalism was dissolved, and capitalism
brought about, by factors internal to the primary relations of feudalism
itself, the relations between lords and peasants.
On the face of it, Sweezy's argument seems, in its main outlines, com­
pletely consistent with the commercialisation model, as Brenner suggests.
By contrast, Dobb's account seems lie a frontal attack on the commer­
cialisation model - and, in fact, it is in many ways a devastating counter­
blast against that dominant model. And yet, on closer inspection, matters
are not so simple. Sweezy made one point that tends to get lost in consid­
erations of the transition debate. He certainly did ascribe the dissolution
of feudalism to the effects of commercial expansion and the growth of
towns. It seems to me that he was wrong to do so, and that in this respect
Dobb was right. But Sweezy insisted that the dissolution of feudalism was
not enough to account for the rise of capitalism, that these were actually
two distinct processes; and there is an interesting contrast here between
Sweezy and Dobb: Dobb seems more inclined than Sweezy to treat the
dissolution of feudalism as essentially the same process as the rise of cap­
Why is this signifcant? Consider the implications of such an arg­
ment: if the dissolution of feudalism is sufcient to explain the rise of
capitalism, are we not very close again to the assumptions of the com­
merciaisation model? We may now be in the countryside instead of in
the town, and we may be focussing on class struggle between lords and
peasants instead of on the expansion of trade. But one critical assumption
remains the same: capitalism emerges when the fetters of feudalism are
removed. Capitalism is somehow already present in the interstices of feu­
dalism, just waiting there to be released. Dobb actually makes such an
argument quite explicitly. He argues, for example, that, although class
struggle did not 'in any simple and direct way' give rise to feudalism, it
did serve to 'modify the dependence of the petty mode of production
upon feudal overlordship and eventually to shake loose the small produc­
er from feudal exploitation. It is then from the petty mode of production
(in the degree to which it secures independence of action, and social dif­
ferentiation in turn develops within it) that capitalism is born'. 6 In the
same debate, Rodney Hilton made a similar point. He argued that peas­
ant resistance to lordly pressures played a crucial role in the process of
transition to capitalism by the 'freeing of peasant and artisan economies
for the development of commodity production and eventually the emer­
gence of the capitalist entrepreneur'. 7
In a sense, then, even these historians tended to assume some of what
needed to be explained -perhaps, i some respects, more than Sweezy, for
all his attachment to the commercialisaton model. For them, the problem
seemed to be mainly to discover how direct producers were freed to avail
themselves of market opportunities, how small or 'middling' producers were
freed to become bigger ones, and how 'petty commodity production' was
liberated by class struggle and allowed to grow into capitalism. In other
words, these arguments reproduce some of the question-begging assump-
Historical Materialism
tions of the commercialisation model, even though this time middling farm­
ers or yeomen serve as a kind of functional, if rural, equivalent of the
burghers in the old model. In case it needs to be emphasised, the issue here
is not the transposition of capitalism from an urban to a rural setting. It is
still possible to beg the question in the manner of the conventional com­
mercialisation model, even when the action moves to the countryside, and
even when the 'prime mover' is class struggle between lords and peasants
rather than between aristocracy and bourgeoisie. Conversely, it would in
principle be possible (i not in accordance with available evidence) to situ­
ate the transition in the city without making the usual assumptions -
though I am not aware of any historian who does this. In any case, despite
the focus on production and exploitation rather than simply exchange,
despite the emphasis on class struggle, and despite the shift from city to
countryside, these Marxist explanations have in common with the commer­
cialisation model a tendency to treat capitalism as a 'shaking loose' or liber­
ation of an already existing economic logic and an enhancement of market
Perry Anderson on absolutism and capitalism
Now l et us turn to another influential Marxist, Perry Anderson.
Unfortunately, he has not (yet?) completed the trilogy in which he was
going to explain the rise of capitalism. So we have to deduce what we can
from the first two volumes, especially from Lineages of the Absolutist
State, and from various bits and pieces elsewhere. It may tell us some­
thing that the climax of this magisterial three-volume work was going to
be the bourgeois revolutions of Europe. Let me state my view baldly frst:
for all the sophisticated complexity of of Anderson's argument, it is a
refnement - fascinating and in many ways illuminating, but no less a
refinement - of the 'commercialisation model'. Here is the most recent
statement of his argument, in a review of Robert Brenner's book,
Merchants and Revolution. Anderson is commenting here on Brenner's
account of capitalism as, in the first instance, a specifcally English phe­
The idea of capitalism in one country, taken literally, is only a bit more plausi­
ble than that of socialism. For Marx, the different moments of the modern
biography of capital were distributed in cumulative sequence, from the Italian
cities to the towns of Flanders and Holland, to the empires of Portugal or
Spain and the ports of France, before being 'systematically combined in
England at the end of the 17th Century.' Historically, it makes better sense to
view the emergence of capitalism as a value-added process gaining in complex­
ity as it moved along a chain of inter-related sites. In this story, the role of
cities was always central. English landowners could never have started their
conversion to commercial agriculture without the market for wool in Flemish
towns -just as Dutch farming was by Stuart times i advance of English, not
least because it was conjoined to a richer urban society."
I should probably point out that Marx, in the passage cited by Anderson,
is explaining not the origins of capitalism but the ' genesis of the industrial
capitalist', not the emergence of specifcally capitalist 'laws of motion', or
specifically capitalist social relations, or the imperatives of self-sustaining
Non-histor of cpitalism
growth.9 What Marx is trying to explain is the accumulation of wealth,
which in the right conditions - that is, in already capitalist social condi­
tions (in England) - was converted from simply the unproductive profts
of usury and commerce into industrial capital. A for the origins of the
capitalist system, the 'so-called primitive accumulation' which gave rise to
specifcally capitalist social property relations and the specifc dynamic
associated with them, Marx situates it firmly in England and in the coun­
tryside. Here too the conditions emerged for the unprecedented kind of
interal market that Marx regarded as the sine qu non of industrial capi­
talism. Like Brenner, Marx acknowledges the need to explain the distinc­
tiveness of England's development. Not the least of England's specifci­
ties, as Brenner points out, is that while other centres of production even
in the medieval period had experienced export booms, early modern
England was unique in maintaining industrial growth even in the context
of declining overseas markets. In other words, albeit within a network of
interational trade, capitalism indeed in one country.
But there is no need to get distracted here by speculations about
Marx's views on the relation between agrarian and industrial capitalism
(or about the questions he left unanswered and, indeed, the inconsisten­
cies he lef unresolved). We might simply note that Anderson's observa­
tions here precisely beg the question. It is one thing to say, for example,
that English commercial agriculture presupposed the Flemish market for
wool. It is quite another to explain how 'commercial agriculture' became
capitalist agriculture, how the possibility of trade became not only the
actuality but the necessity of competitive production, how market oppor­
tunities became market imperatives, how this specifc kind of agriculture
set in train the development of a capitalist sstem. We cannot just assume
that commerce and capitalism are one and the same, or that one passed
into the other by a simple process of growth. Anderson is here assuming
the very thing that needs to be demonstrated, namely that commerce, or
indeed production for the market (a widespread practice throughout
much of recorded history), became capitalism by means of sheer expan­
sion, which at some point achieved a critical mass. His argument, in other
words, suffers from the very circularity that has always afflicted the com­
mercialisation model.
Earlier in his career Anderson laid the foundation for his projected
account of bourgeois revolutions by analysing the development of abso­
lutism. It is interesting to see how the absolutist state figures in his argu­
ment about the rise of capitalism. Absolutism, Anderson wrote, was 'a
redeployed and recharged apparatus of feudal domination, designed to
clamp the peasant masses back into their traditional social position'.10
This redeployment of feudal power was necessary, he argues, because the
old feudal bonds had been weakened by the commutation of feudal dues
into money rents, and, more particularly, by the growth of a commodity
economy. If the absolutist state was essentially feudal, according to
Anderson, what was its role in the transiton to capitalism? He starts with
a defnition of feudalism: feudalism was a mode of production defned by
'an organic unity of economy and polity', which took the form of a 'chain
of parcellised sovereignties'. In other words, feudal lordship represented a
unity of political and economic power, and this was accompanied by 'a
mechanism of surplus extraction', serfdom, in which 'economic exploita-
Historical Materialism
tion and politico-legal coercion were fused'. 1
But ' (w)ith the generalised commutation of dues into money rents', he
continues, 'the celular unity of political and economic oppression of the
peasantry was gravely weakened, and threatened to become dissociated.
The result was a displacement of politico-legal coercion upward towards
a centralised, militarised summit - the Absolutist State' .12 In other words,
feudal lords concentrated their formerly fragmented or parcellised coer­
ive powers into a new kind of monarchical state. Ironically, the effect of
this displacement upwards of feudal coercive power - at least its principal
contribution to the evolution of capitalism according to Anderson - was
to facture the unity of economy and polity, which had characterised feu­
dalism. On the one hand, political power was concentrated in the royal
state. On the other hand, the economy began to achieve a certain autono­
my. Meanwhile, in the interstices of the fragmented feudal system, in the
town, an economic sphere had emerged that was not controlled by the
aristocracy. At the same time, these towns became the sites of technical
innovations. Anderson concludes that, while 'the political order remained
feudal,.society became more and more bourgeois'.l3
That, then, is Anderson's conception of absolutism in broad outline.
And much of it is very illuminating too. His characterisation of the abso­
lutist state as essentially feudal is especially useful - but it also demands
closer scrutiny. Keep in mind what Aderson means by this. The abso­
lutist state was essentially feudal, he insists, because it represented the
displacement upward and the centralisation of the feudal lords' politico­
legal coercive powers, separating those powers from economic exploita­
tion. In other words, the absolutist state separates the two moments of
exploitation, the process of surplus extraction on the one hand and the
coercive power that sustains it on the other. The two then go on in sepa­
rate spheres. The feudal fusion of economy and polity starts giving way to
the separation characteristic of capitalism.
Now there is another way of looking at absolutism, which is to say that
it represents a centralisation of feudal power in a different sense: namely
that the state itself becomes a form of property, an instrument of appro­
priation, in ways analogous to feudal lordship. Economic and political
power are still fused - though the lord appropriates rents whie the state
and its officeholders appropriate peasant surpluses in the form of tax.
Sometimes Anderson does seem to think of absolutism in these terms, as
still a unit of economic and political spheres. But his whole argument
that absolutism plays a pivotal role in the transition to capitalism depends
on the essential function of the absolutist state in separating political and
economic spheres. He is at great pains to emphasise that what gets 'cen­
tralised upwards' in the absolutist state is not the feudal fion of political
and economic spheres but rather just the political-legal or coercive
moment of feudalism as distinct from the moment of economic exploita­
tion. The absolutist state simply represents for him the politico-legal
coercive power which enforces the economic exploitation that takes place
on a different plane.
In effect, then, the displacement upward of feudal political power
plays the same role in Anderson's argument as the removal of fetters does
in other versions of the old model. In fact, it seems that absolutism is a, if
not the, essential means by which the fetters of feudalism are removed
Non-histor of cpitaUsm
from the economy. Absolutism, in fact, seems to have been a necessary
transitional point between feudalism and capitalism. In any case, freed
from direct political bondage, commodity production was able to grow,
and the 'economy' could follow its own inclinations. Capitalism was the
result of liberating the economy and removing the dead hand of feudalism
from the natu
al bearers of economic rationality, the burghers. There are,
of course, certain serious problems in this treatment of absolutism as an
apparently essential phase in the transition from feudalism to capitalism -
not the least of which is the fact that English capitalism did not enjoy the
benefit of absolutism, while French absolutism did not encourage the
advent of capitalism. It seems more plausible to argue that absolutism
was not a transitional phase between feudalism and capitalism, that it
was, on the contrary, an alternative route out of feudalism; but this is not
the place to pursue that line of argument. My intention here is simply to
identify the ways in which Anderson reproduces the old commercialisa­
tion model.
Robert Brenner's alternative
There has been at least one Marxist account of the transition that really
has departed from the old model and has deliberately set out to explain
the emergence of capitalism without assuming its prior existence in order
to explain its coming into being.14 Robert Brenner has looked for a
dynamic internal to feudalism that does not itself presuppose an already
capitalist logic. Class struggle figures prominently in his argument as it
did in Dobb's and Hilton's, but here it is not - or not usually - a question
of liberating an impulse toward capitalism. Instead, it is a matter of lords
and peasants, in certain specific conditions peculiar to England, involun­
tarily setting in train a capitalist dynamic while they acted to reproduce
themselves as they were, as essentially non-capitalist classes, operating
according to a non -capitalist logic.
Brenner's argument, then, emphasises the differences among various
states in Europe, and not just the specificity of Europe in relation to other
cases. His explanation has to do with the very specific conditions of
English property relations, in which an exceptionally large proportion of
land was owned by landlords and worked by tenants. These conditions
may have produced the famous agrarian 'triad' of landlord, capitalist ten­
ant and wage-labourer; but it is important to keep in mind that, for
Brenner, distinctive economic imperatives began to operate even on ten­
ants who might themselves be the direct producers, without employing
wage labour. The conditions of tenure were such that growing numbers of
tenants were subjected to market imperatives - not the opportunity to pro­
duce for the market and grow from petty producers into capitalists but the
need to specialise for the market and to produce competitively - simply in
order to guarantee access to the means of subsistence. At the same time,
landlords in England were also in a special position. Although they con­
trolled a uniquely large proportion of the best land, they did not enjoy -
and did not really need - the kinds of extra-economic powers on which,
say, the French aristocracy depended for much of its wealth. The English
ruling class was distinctive in its growing dependence on the productivity
of their tenants, rather than on exerting coercive power to squeeze more
surplus out of them.
Historical MateriaUsm
In other words, these property relations had their own distinctive
'rules for reproduction', and these rules produced their own distinctive
laws of motion. The result was to set in train a new historical dynamic, an
unprecedented rupture with old Malthusian cycles, a process of self-sus­
taining growth, new competitive pressures which had their own effects on
the need to increase productivity, reconfguring landholding, its further
concentration, and so on. This new dynamic is what we might be allowed
to call 'agrarian capitalism', and it was specific to England. In other
words, the distinctive conditions which, for example, Michael Mann
attributes to Europe in general in the Middle Ages are not enough to
explain the development of capitalism, or the specifcity of the process of
self-sustaining growth that emerged in England.
There have been various criticisms of Brenner, and some of the local
disagreements about specific historical points are no doubt well taken,
but I wil just briefly outline some of the more general criticisms that have
implications for the larger issues in the debate on the transition. First,
there is a general criticism of the very idea that English agrarian relations
were distinctive enough - in the 17th or even the 18th century - to justify
calling them agrarian capitalism. There are two different kinds of argu­
ments against the idea of agrarian capitalism. One has to do with whether
English economic growth really was distinctive, whether, in particular,
English agriculture even in the 18th century was distinct, specifcally in
its drive to improve productivity. Why, for example, some historians have
asked, was French agricultural productivity in the 18th century roughly
equivalent to that of English agriculture? The second objection has to do
with wage-labour: since capitalism is defined above all by the exploitation
of wage labour, they say, is it not a decisive argument against the concept
of agrarian capitalism - or at least against its existence in the 17th centu­
ry - that England was not yet a predominantly wage-earning society, that
especially permanent and regular wage-labour was still very much in a
minority? What about the processes of expropriation and proletarianisa­
tion, the differentiation of the English peasantry into prosperous farmers
on the one hand and a propertyless class on the other? Do these process­
es not belong to the pre-history of capitalism?
These objections seem to me revealing - but they may reveal more about
the critics than about Brenner or the concept of agrarian capitalism. The
first objection - about agricultural productivit in France - fundamentally
misses the point. It turn out that what these critics mean is that French
agricultural production in the 18th century was roughly equivalent to
English agriculture in its total output. But consider the fact that it took a
much smaller rural population and a much smaller number of people
engaged i agricultural production to produce that output in England than
in France. This means that the so-called euivalence of French and English
'productivit' - it would be btter to call it production, not productivity - far
from challenging the distinctiveness of English property relations and agrari­
an capitaism, actually confrms it. And these same distinctive conditions
created both a potential non-agricultural labour-force and a potential mass
market for cheap consumer goods like food and textiles, which were neces­
sary conditions for the development of industral capitalism. The tendency to
miss this point about productivit is perhaps rooted in some basic precon­
ceptions about capit
lism, and particularly the old tendency to identify it
Non-histor of cpitalism
with towns and trade rather than with certain distinctive laws of motion and
rules for reproduction which are rooted in certain property relations,
whether urban or rural.
How, then, is Brenner's argument affected by the other question,
about the extent of wage labour? The problem here is not only an empiri­
cal one. We can agree that the extent of wage labour was limited. And we
can agree that the process of expropriation and proletarianisation was
certainly central to the story of capitalism. But here too there is a begging
of the question, and here again Brenner sets out to explain what others
have taken for granted. Brenner does not assume that a pre-existing divi­
sion between rich and poor peasants (such as has existed at other times
and places) would inevitably lead to their polarisation into rich farmers
and dispossessed labourers. For example, both England and France in the
later 15th century possessed a middle peasantry with relatively large hold­
ings. Yet from this common starting point, they diverged in substantially
different historical directions, the French toward increasing morcellisa­
tion of peasant holdings, the English toward the agrarian triad of land­
lord, capitalist tenant and wage labourer; the English toward agricultural
improvement, the French toward agricultural stagnation. Nor does he
assume that the English ruling class could simply have expropriated smal
farmers by brute force, or that they would have done so even if they
could, in the absence of very specifc economic conditions that made the
dispossession of small producers not only possible but profitable.
It still remains, then, to explain the differentiation of the English peas­
antry (the 'rise of the yeoman') which eventually ended in a polarisation
betwen capitalist farmers and propertyless labourers. Brenner's explana­
tion, again, has to do with the new economic logic that subjected English
farmers to the imperatives of competition in unprecedented ways and
degrees. This logic was imposed on farmers whether or not they consis­
tently employed wage labour. It applied even when the tenant was him­
self, or together with his family, the direct producer. This is a particulary
important point: Brenner makes it clear that direct producers could be
deprived of non-market access to the means of their own self-reproduc­
tion even while remaining in possession of the means of production, and
that such a condition subjected them to the imperatives of the market
(which, by the way, is a point worth keeping in mind in debates about
'market socialism' today). Again, peasants elsewhere and at other times
had availed themselves of market opportunities, but English farmers were
distinctive in their degree of subjection to market impertives.
Brenner set out to explain why and how this came to be so, how pro­
ducers were deprived of non-market access to the means of their self­
reproduction and even to land itself, how landlordly forms of exploitation
were transformed from 'extra-economic' surplus extraction to the appro­
priation of capitalist rents, how it came about that both landlords and
tenants were compelled (as well as enabled) to move in response to the
imperatives of competition, how new forms of appropriation established
new compulsions and how those compulsions conditioned the differentia­
tion, and in large part the dispossession, of the peasantry. This happened
both through purely 'economic' pressures of competition and through
more direct coercion by landlords with a new kind of economic interest
in large and concentrated holdings. A mass proletariat was the end not
Historical Materialism
the beginning of the process. I cannot be emphasised enough that for
Brenner, the market-dependence of economic actors was a cause not a
result of proletarianisation. Some might want to argue that the term 'cap­
italist' should be reserved for that end result. This may be so. But at the
very least, we need to acknowledge that the end of the process presup­
poses the process itself. It also needs to be acknowledged that this
process itself already represented a major historical rupture and new eco­
nomic 'laws of motion', without which a mature capitalism would not
have emerged. It makes sense to me to recognise this historical specifcity
by calling it agrarian capitalism. But in any case, the great strength of
Brenner's argument as against the others is that it recognises the speci­
fcity of that new historical process, that new economic logic, and makes
a convincing effort to explain how it came about.
To put it another way, many historians have claimed to be explaining
the transition from feudalism to capitalism, but Brenner is one of the very
few who actually do deal with a process of transition: the transformation
of one kind of society to another, one set of rules for reproduction to
another and even one historical dynamic to another. In their various
ways, most attempts to explain the process of transition tend to generalise
laws of motion specifc to capitalism and turn them into universal princi­
ples of historical movement. In other words, even when they acknowledge
the particularity of capitalism as a specific historical form, the emergence
of that historical form takes place by means of essentially capitalist
There is one fnal criticism of Brenner that I fnd especially revealing.
Perry Anderson, and others, have seized on Brenner's recent book,
Merchants and Revolution, which attributes an important role to certain
kinds of merchants in the English Revolution. 'Aha' they say, after insist­
ing that capitalism was born in the countryside, Brenner has had to read­
mit the bourgeoisie and bourgeois revolution afer all. There is, Anderson
says, a 'deep paradox' in Brenner's work, a fundamental contradiction
between his original thesis on the origin of capitalism and his latest work
on merchants. 'Here, if ever,' Anderson continues,
were revolutionary bourgeois. The species declared a fiction in France was bel
et bien a reality in England, a hundred years before the Convention. There is a
nice irony that it should be massive historical evidence, running against - not
with - a theoretical conviction which has brought a Marxist scholar to this con­
clusion. The detractor of the significance of merchant capital in principle has
been the first to establish, in spell-binding detail, its role as a demi-urge i
Brenner, it must be said first, has conceded nothing of the kind. But in
any case, by now it should be clear what my answer would be to
Anderson's 'deep paradox'. This criticism seems to me exquisitely to beg
the question. Mter all, we may be utterly convinced that, say, the French
Revolution was undoubtedly 'bourgeois', indeed much more so than the
English, without coming a flea-hop closer to determining whether it was
also capitalist. A long as we accept that there is no necessary identifica­
tion of 'bourgeois' (or burgher or city) with 'capitalist', the 'revolutionary
bourgeois' can be far from being a fiction, even - or especially - in
Non-histor o capltaUsm
France, where the model revolutionary bourgeois was not a capitalist or
even an old-fashioned merchant but a lawyer or offce-holder. At the
same time, if the 'revolutionary bourgeois' in England was inextricably
linked with capitalism, it is precisely because capitalist social property
relations had already been established in the English countryside.
There is, of course, much that Brenner does not do. Two or three
things in particular seem pressing to me: frst, the commercialisation
model may be fatally fawed, but that does not change the fact that capi­
talism emerged within a network of international trade and could not
have emerged without that network. So a great deal still needs to be said
about how England's particular insertion into the European trading sys­
tem determined the development of English capitalism. England arguably
transformed the nature of trade by creating a distinctive national market,
in fact the frst national market, and perhaps the frst truly competitive
market. Much still needs to be said about how this affected the nature of
interational trade. The other big issue has to do with the European
state-system and its contribution to the development of English capital­
ism. Both the system of trade and the state system operated as the con­
duit through which England was eventually able to transmit its competi­
tive pressures to other states and economies, so that, as I have suggested
elsewhere, non-capitalist states could become the engines of capitalist
development in response to these external pressures. We have hardly
begun to explore the mechanisms by which capitalism imposed its imper­
atives on other European states, and eventually on the whole world. A
systematic exploration of these historical questions might, among other
things, be a big help in dealing with the so-called process of 'globalisa­
tion' today.
At any rate, I have been arguing that the main problem in most stan­
dard histories of capitalism is that they start - or end - by conceptualis­
ing away the specifcity of capitalism. What we need is history that brings
this specificity into sharp relief. We need to acknowledge the difference
between commercial proft-taking and capitalist accumulation, between
the market as an opportunity and the market as an imperative, between
transhistorical processes of technological development and the specifc
capitalist drive to improve labour productivity; and so on. Most Marxists
would, no doubt, claim to be doing all or most of these things, but I have
tried to show that their accounts of history often fail to proceed on that
basis and tend to be complicit in disguising the specifcity of capitalism.
The question of the origin of capitalism may seem arcane, hardly interest­
ing even to historians these days. But some of the question-begging
assumptions I have been talking about are so deeply rooted in our culture
that they must take some responsibility for the widespread and dangerous
illusions about the so-called free market.
Historical Materialism
1. This paper is based on a tak given at the University of Sussex in May 1996. In
some parts of it, I have drawn upon articles published elsewhere: Wood 1994
and 1 996.
2. Mann 1986, p. 373.
3. Mann 1986, p. 374
4. Polanyi 1957; for a discussion of Polanyi, see Wood 1994.
5. Brenner 1977.
6. Dobb 1976, p. 59.
7. Hilton 1976, p. 27.
8. Anderson 1993, p. 1 7.
9. Marx 1976, pp. 914-26.
o. Anderson 1974, p. 18.
1 1 Anderson 1984, p. 1 9.
12.Anderson 1984, p. 19.
13. Anderson 1984, p. 23.
14.ln addition to the article cited in n.s, the most important discussions are
Brenner 1985a, 1985b, 1986, 1989 and 1993.
1 5.Anderson 1993 p. 17.
16.Wood 1991, pp. 159-60.
Anderson, Perry 1974, Lineages of the Absolutist State, London: NLB.
Anderson, Perry 1993, 'Maurice Thomson's War', London Review of
Books, 4 November: 13-17.
Brenner, Robert 1977, 'The Origins of Capitalist Development: A
Critique of Neo-Smithian Marxism', New Left Review, 104: 25-92.
Brenner, Robert 1985a, 'Agrarian Class- Structure and Economic
Development in Pre-Industrial Europe', in The Brenner Debate:
Agrarian Class Structure and Economic Development in Pre-Industrial
Europe, edited by T.H. Aston and C.H.E. Philpin, Cambridge:
Cambridge University Press, pp. 10-63.
Brenner, Robert 1985b, 'The Agrarian Roots of European Capitalism', in
The Brenner Debate: Agrarian Class Structure and Economic
Development in Pre-Industrial Europe, edited by T.H. Aston and
C.H.E. Philpin, Cambridge: Cambridge University Press, pp.
Brenner, Robert 1986, 'The Social Basis of Economic Development', in
Analytical Marxism, edited by John Roemer, Cambridge: Cambridge
University Press, pp. 23-53.
Brenner, Robert 1989, 'Bourgeois Revolution and Transition to
Capitalism', i The First Moder Society, edited by A.L. Beier et aI.,
Cambridge: Cambridge University Press, pp. 271-304.
Brenner, Robert 1993, Merchants and Revolution: Commercial Change,
Political Confict, and London's Oerseas Traders, 1 550-1653,
Cambridge: Cambridge University Press.
Dobb, Maurice 1 976, 'A Reply', in The Transition fom Feudalism to
Capitalism, edited by Rodney Hilton, London: NLB, pp. 57-67.
Hilton, Rodney 1 976, 'Introduction', in The Transition from Feudalism to
Capitalism, edited by Rodney Hilton, London: NLB, pp. 9-29.
Non-histor of capitaUsm
Mann, Michael 1986, The Sources of Social Power, Vol. 1: A Histor of
Power from the Beginning to A.D. 1 760, Cambridge: Cambridge
University Press.
Marx, Karl 1976, Capital: A Critique of Political Economy, Vol. 1, trans­
lated by Ben Fowkes, London: Penguin/NLR.
Polanyi, Karl 1957, The Great Transformation: The Political and
Economic Origins of Our Time, Boston: Beacon Press.
Wood, Ellen Meiksins 1991, The Pristine Culture of Capitalism: A
Historical Essay on Old Regimes and Moder States, London: Verso.
Wood, Ellen Meiksins 1994, 'From Opportunity to Imperative: The
History of the Market', Monthly Review, 46: 3: 14-40.
Wood, Ellen Meiksins 1996, 'Capitalism, Merchants and Bourgeois
Revolution: Refections on the Brenner Debate and its Sequel',
Interational Riew of Social History, 41: 209-32.
A Quarterly Review of Social and Political Theory and
Criticism, from Buenos Aires
Papers, interviews, book reviews, conference reports on
Sociology, Histor, Philosophy, Economy, Politics, Labour
Studies, Cultural Studies (in Spanish)
Issue No 16. Sng - Summer 1996-7: S C and Globalisation: Is the debate over?
Burm, Stat and Market in Interational Political Economy: a Marxian critique; Rosencrace,
The rise of the virtual state: when territory becomes obsolete; Negri, Crisis of the political space;
Dinerstein, Global capital, labour and unions: on forms and contents. Plus more.
Issue No 17. Autmn The discret c of citizenship: Subjects. Diferenc and
Belucci, From tyranny of singularity to city identities; Rapisardi, The constuction of homosexual
identity i the "Gays and Lesbian Studies" and the "Queer theory"; Wortman, The aesthetic of
marginality: young people and video clips; Fachi, Sexual mutilations in women. Multiculturalism
and the law; Bowen, the myth of global ethnic confict.
Annual Subscription Rates for 3 issue: 16, 17, & 18 (includes postage)
Individual: £ 24 Libraries and Institutions: £ 4
Back issues (includes postage)
Single issues: £ 6 Bak issues: nos. 13-16: £ 5
Potal Orders or cheque t: Tesis 1 1 . Grupo Editor SRL
Some reflections on two books by
Ellen Wood
Coln Barker
Some time ago, the editors of Monthly Reie invited me to submit a
short review of two recent books by Ellen Wood: The Pristine Culture of
Capitalism, and Democrac Against Capitalism.! I found myself, in the
course of re-reading these books, filled with admiration for most of what
the author said, and indeed, for the manner in which she presented her
case. At various points, however, I found myself not fully satisfed. But a
short review was not the place to develop my concers.
Ellen Wood represents a needed voice too little heard today, that of a
rational and militant Marxism, concerned to re-examine the very founda­
tions of Marxist thought in order to make them relevant to the contempo­
rary world. What follows should be understood not so much as a 'cri­
tique' but as a friendly and partly disputatious conversation.
In the first part, I attempt to situate the significance of Wood's work,
and to explore a number of points of fundamental agreement. In the sec­
ond part, I seek to indicate some issues requiring further examination
and development.
Part One. Situating Wood's work
1. The point of departure
World capitalism is in a long drawn-out crisis of stagnation. It is marked
in the most advanced countries by the highest levels of persistent unem­
ployment since the 1930s. Some former 'communist' countries have
experienced extraordinary collapses in production levels and living stan­
dards. In large parts of the Third World, post-war trends towards a rising
quality of life have been completely reversed, and the plagues of debt,
famine and war are resurgent. There is an urgent need for new develop­
ments in the critique of capitalism as a system of society.
Yet there is a paradox. A large part of the Left has turned away from this
project, to develop a set of arguments and associated practices which
devalue just such a critique. Symptoms of this shift in left thinking and
practice include a rejection of any focus on 'capitalism' and 'class strug­
gle' and a turning away from the working class as the key agency of
social transformation. Ofen this shif has been associated with an effec­
tive abandonment of socialism as an aspiration; an acceptance of the lim­
its to possibility set by a pluralist liberalism; and a widespread pessimism
of intellect and will together.
In part, these developments can be seen as the product of a specifc
historical conjuncture. The thinking of wide sections of the Lef has been
Two book b Elen Wo
shaped - not necessarily consciously - by a period of working-class
defeats and retreats effected by the decline of the last great international
wave of popular struggle associated with 'the sixties'. What Tony Cliff
identified as a 'downturn' in the level of the class struggle proved to be of
more than simply British significance. Radicals must adapt to changed
circumstances, but how they do so is crucial. They can maintain their
aspirations to fundamental change, on condition that they analyse the
roots of their own defeats and reorient accordingly. Or they can give up
those aspirations and invent reasons for their irrelevance. Much of the
Left tended to follow the latter path. Much of the apparently radical 'new'
thinking of the 1 980s and early 1 990s embodied signifcant retreats to
more conservative positions and accommodations to a mood of political
misery.' The Left has, perforce, had to engage in a defence of its posi­
tions against a series of critics who have departed its ranks while declar­
ing themselves the 'True Radicals'"
Marxism is the only theory with the ambition to capture simultaneous­
ly the nature of global society as a single capitalist system and the practi­
cal prospect of humanity's self-emancipation from that system. However,
if Marxist ideas are to play a central part in future struggles, a necessary
part of the labour of the Left is to survey the very foundations of its own
thinking and practice.
The critique of the ills of capitalist society requires more than simply a
set of separate objections to various of its immediate features, but an
understanding of capitalism as a historical social form. This requires that
we apply to the study of capitalism two principles of analysis identified
clearly by an earlier generation of Marxists: what Lukacs termed the prin­
ciple of 'totality', 5 and what Korsch called 'the principle of historical
specificity' .6 Capitalism must be understood, in one of Marx's phrases, as
'an organic system', as a whole with its own characteristic particular fea­
tures which mark it off, sharply, from other histo
ical forms of society.
To grasp the character and historical speifcity of capitalism requires
the development of categories capable of registering the particularities of
capitalism as a system, of differentiating it adequately from other histori­
cal or imaginable forms of society, and of identifying its specifc patterns
and dynamics of development. Potentially, at least, Marxist theory pro­
vides a kind of self-reflective understanding capable of developing ade­
quate instruments for the critique of other, weaker, bourgeois concep­
tions of the nature and problems of contemporary society, and of offering
an unprecedented clarity about both how to understand and how to
transform the contemporary world.
We can assert that, but then immediately we face a difculty. We must
first determine what is to count as Marxism. For, not only are there pow­
erful alternative bourgeois paradigms which attempt to explain the rise of
and character of modern society, and which enjoy a partial hegemony
within contemporary thought. But also, that partial hegemony extends
within varieties of Marxism itself. Indeed, some forms of twentieth centu­
ry Marxism themselves fall back into the very assumptions of bourgeois
Several elements help fog our vision about the founding assumptions
of Marxism. Ellen Wood, drawing on Robert Brenner's work, suggests
one cause lies within Marx's own intellectual development. Marx himself
Historical Materialism
did not immediately arrive at a complete answer to the problems posed by
the development of an adequate historical materialism. His writings of the
1 840s, she suggests, are still marked by acceptance of a conception of
history derived from the great classical political economists and philoso­
phers of the European Enlightenment. It was only in the 1 850s that he
began to fully break clear of their framework of assumptions, and to open
the way to a more complete and radical conception of the historical speci­
ficity of capitalism.
But there are deeper problems. Above all in the twentieth century, te
terminology of Marxism became associated with a set of practices which
ran completely counter to those assumed by its founders. The most obvi­
ous problems here concern on the one hand the development of Social
Democracy, and on the other hand of Stalinism. Both used the language
of Marxism, but denied its key premise: the banner of working-class self­
Marxism today is thus faced with the task, frstly, of cleaning out from
its own premises an accumulated historical debris of confusion, misrepre­
sentation and direct contradiction of its own fundamental principles.
However, that i s by no means the only diffculty. At least three additional
sets of problems need to be signalled.
First, Marx did not bequeath us a completed 'system', even in his own
terms. Capital itself, his monumental contribution to the comprehension
and critique of the essential structures of contemporary life, is itself -
according to his own original plans for the work - anything but fnished.
Indeed, as a number of signifcant recent Marxist works have suggested,
that 'incompleteness' has had serious effects on the contemporary recep­
tion and (mis)understanding of the work and its assumptions.7 The very
methodological foundations of Marx's own work require re-inspection
and development.
Second, Marx wrote over a century ago. From the standpoint of today,
the development of capitalism and its specifc forms were, if not in infancy, in
their relative youth. A Marxism relevant to today must account for the enor­
mous transformations wrought wthin the character and institutional forms of
capitalism itself over the past century and a quarter. And, in important
respects, this reqires that Marxists reshap and develop their own analytical
instruments, in order to deal wit these matters. O this agenda we fmd such
signifcant matters as the development of corporate (national and multna­
tional) and state property forms, combined and uneven development of capi­
talism across the globe, immense development of state stuctures and inter­
ventions, global military conlicts and so forth.
Third, since Marx's time there has occurred a tremendous development
both in the extent of popular struggles within and against the confning
structures of capitalism, but also in the issues posed by those struggles and
in the forms they have taken. Let me just mention a few: the emergence of
mass workers' organisations; huge waves of popular revolt on an intera­
tional scale; a globally signifcant but ultimately defeated attempt, begun in
Russia, at the practical revolutionisation of world society; the massive
defeat of the German working-class movement by the Nazis, along with
other appalling tragedies and defeats suffered by other working-class move­
ments; the emergence of large-scale movements challenging aspects of con­
temporary life such as 'racial' and gender divisions, oppressive defmitions
Two bok b Elen Wo
of sexuality, ecological degradation and the threat and reality of war.
Marxism is a theory centred not merely on the analysis of the nature and
limits of capitalism as a mode of production, but on a global struggle for
human emancipation from those limits. Historically, the revivals and recon­
structions, as well as the perversions of Marxist thought that have occured
during the past century, have been intimately energised and shaped by the
various fates of these movements. Their critical assessment must be central
to any Marxist agenda of reconstruction.
Ellen Wood's work needs to be situated in the light of this complex
agenda. She does not - as it's doubtful that any single Marxist thinker
could - offer a set of answers to al the problems and issues signalled here.
However, she does focus on suficient central problems to make these
books important interventions in contemporary processes of sociallst re­
2. Methodological assumptions
Let me begin with a point of fundamental agreement with Ellen Wood's
methodological starting point. Marx, she writes in DAC, presented the
world in its 'political aspect', even in his most technical writings. Part of
his charge against the classical political economists was that they
obscured the political face of the social relations which tey were the frst
to begin to analyse seriously. It is necessary for Marxism to recover, fully,
that political critique of capitalism.
Above all, it is necessary to insist on the political nature of those social
relationships which are commonly termed 'economic' relations. The
essential semantic apparatus of Marx's critique of the political economy
of capitalism involves such terms as 'commodity', 'value' , 'exchange',
'capital', 'surplus-value', 'exploitation', 'appropriation' and the like. These
concepts, in Marx's hands, refer always to historically constituted social
relations between human beings, and not simply to 'economic' relations
devoid of or abstracted from any social considerations. One example
must sufce. Where Max Weber (to whose critique Wood quite rightly
devotes a major chapter of DAC) defnes 'capital' as a universal, purely
economic category, which refers to a stock of money measured in a spe­
cifc way by accountancy practices, Marx insists that capital is a 'social
Marxism's logical starting point - to which Marx won through via a
major conceptual effort during the 1 850s - is a view of history as a
sequence of 'modes of production', each characterised at root by a shared
set of fundamental social relations of production. These social relations of
production are not simply 'economic' but equally involve 'political',
'juridical' and 'ideological' assumptions as part of their very constitution.
In a famous passage in Capital III, Marx proposed that the key to under­
standing the social relations at the heart of the various social forms that
class societies have taken lies in grasping their 'innermost secret'.
Decoding that secret consists in identifying the specifc form in which
surplus labour is pumped out of the direct producers.8
There may be particular arguments among historians about how many
such 'forms of pumping' need to be identified and differentiated in histo­
ry, about how each should b understood in its own right and about the
specific forms of development and conflict each form has generated.
Histoical MateriaUsm
Such arguments need not especially concern us here. What must be
stressed, however, is that grasping the 'form of pumping' predominant
within a given epoch of human history involves understanding these
social relations of production as possessing simultaneously both econom­
ic and political-juridical aspects. If we consider them in terms of a slightly
different language, that of 'property forms', the same argument obtains ­
and rather more obviously, for 'property' is a term with a natural 'juridi­
cal' register. In whatever specifc historical form, the pumping of surplus
labour must, as part of its very constitution, involve not just a fow of
resources from the direct producers to an appropriating class, but equally
a specific configuration of power capable of sustaining that flow, and
indeed a predominant set of moral and even aesthetic rules expressing
everyday assumptions about reproduction
ithin that social form.
The argument perhaps needs to be taken even further than Wood
overtly pushes it. The very process of ongoing exploitation - and this is
merely a different term for that same 'pumping of surplus labour' - is an
inherently conflictual social process. Exploitation directly generates resis­
tance on the part of direct producers, in all manner of forms, and these in
response generate among those who appropriate surplus labour an
immense variety of strategies and institutional forms concerned with
beating back, controlling, managing and containing such resistance.
Marx's analytical starting point - a specifc form of pumping out surplus
labour - immediately assumes a process of class struggle, whose particu-
1ar forms must themselves be explored. 'Class struggle' is therefore
embedded as an active principle of the ' social relations of production'
right from the beginning.
It is a central presupposition of Marxist thinking, as Wood argues,
that the core processes wich defne a given mode of production have
immense determinative effects on all aspects of social life. On the one
hand, it requires a serious work of analysis to uncover and develop an
adequate account of the underlying social relations of production in a
given epoch of human history, in order to arrive at an adequate 'abstract'
understanding of the principles shaping the organisation and develop­
ment of society in that period. On the other hand, it is an essential
assumption that these 'abstract' and 'general' principles of social repro­
duction are themselves powerfully generative. If they are the most
'abstract' forms, they are also the most 're
l'. A Wood remarks, the very
point of Marxist critique is political and practical. Its purpose is to illumi­
nate the nature of the terrain on which people act and struggle, where
indeed their most fundamental social relations confront them with press­
ing and recurring problems which affect their everyday reproduction.
Part of the signicance of Wood's work - especially in the frst half of
DAC - lies in the work she does towards recovering these notions, and
using them to clarify a series of crucial issues at the heart of social and
political thought.
First, the absolutely proper stress on the constitutively 'political'
nature of the social relations of production leads her to a critique of those
forms of thought which mark off and enclose a separate realm of social
relations denoted as 'economic'. Such forms of thought are characteristic
of the deep structures of bourgeois thinking. They re-appear in contem­
porary historical sociology (for example in the writings of Anthony
Two bo k by EUen Wo
Giddens, Theda Skocpol and Michael Mann) as well as in their forebear,
Max Weber, but they also make their appearance in some varieties of
contemporary Marxism itself. Here the 'economic' is treated as a realm,
sphere or level, by itself, enclosed and separated from other constitutive
social relations. The regular effect of this way of thinking is to de-histori­
cise, de-socialise and de-politicise the understanding of social production
relations. In other words 'economic' relations, far from being a problem
requiring historical analysis and critique, come to be treated as 'natural',
thus replicating the exact mistake of which Marx accused the classical
political economists. His charge was very precise:
Political economy has indeed analysed value and its magnitude, however
incompletely, and has uncovered the content concealed within these forms. But
it has never once asked the question why this content has assumed this particu­
lar form, that is to say, why labour is expressed in value, and why the measure­
ment of labour by its duration is expressed in the magnitude of the value of the
To explore such questions as those Marx poses requires a sense of the
historical specificity of the social relations within which such peculiar
practices occur. We pay a severe price if these questions are not posed.
For while we may gain some understanding of social relations, our
understanding is limited by a framework of thinking which takes them to
be natural and eternal, and thus also removes them from the sphere of
political critique and of practical political action.
Second, if a mode of production cannot be understood in 'economic'
terms in isolation from its political, juridical, ideological and moral
aspects, then the theoretical language in which we discuss these matters
itself requires careful inspection. In this connection, Wood examines the
enormously troublesome metaphor with which, seemingly, Marxism has
been saddled: tat of 'base and superstructure'. One of the dificulties of
this metaphor is that it can provoke or reinforce a pattern of thinking
which does, indeed, divide the world of human experience up into distinct
'spheres', labelled variously the 'economic', the 'political and juridical'
and the 'ideological'. These are, fIst, considered separately, each in its
own right, and then, second, related to each other by links of determina­
tion and causality. The 'economic' is held to have some kind of 'primacy'
in social determination, by comparison with the 'merely political' or
'merely ideological' . The metaphor of 'base and superstructure' - and it
is sometimes forgotten that it is only a metaphor - tends to encourage
various kinds of 'mechanical' thinking. On the one hand, we have the
deeply 'passive' conceptions of socialist development within classical
social democracy; on the other, Stainism adopted the 'base-superstruc­
ture' notion as a philosophical underpinning to its inherently anti -Marxist
project of developing the material forces of production at the expense of
working-class control over society. More recently, the Althusserian school
- as Wood argues in a very cogent section of DAC - attempted to 'save'
the metaphor at the cost of dissociating 'theory' more or less entirely
from actual history.
In its own way Wood suggests, the metaphor, by encouraging a view
that there is a distinct set of economic and political-juridical 'spheres',
Historical Matrialism
replicates - as an abstract model - the working assumptions of a so
form which is itself peculiarly the product of capitalism. That is, it reflects
in its own way an institutional 'separation of the economic and the politi­
cal' to which Wood pays considerable attention, and about which we shall
have more to say later.
Wood argues, following the lead offered by both Edward Thompson
and Raymond Williams, that the 'base-superstructure' metaphor is proba­
bly best junked altogether from Marxism's conceptual system. Much
more promising is the alterative generalising metaphor (or sequence of
metaphors) Marx offered in the introduction to the Grundrisse:
In all forms of society there is one specific kind of production which predomi­
nates over the rest, whose relations thus assign rank and influence to the oth­
ers. It is a general illumination which bathes al the other colours and modifes
their particularity. It is a particular ether which determines the specific gravity
of every being which has materialised within it.IO
Or, to adopt yet another metaphor, from Thompson, there is a 'kernel'
of social relations at the heart of a given mode of production, which
shapes and conditions all aspects of social life.
This approach locates, at the root of Marxism, a specific argument
about historical development and its possibilities in the future. The argu­
ment entails several propositions. First, the term 'society' which domi­
nates a good deal of sociological thinking is a false abstraction, obscuring
the fact that human social development has occurred within a number of
different forms, or modes of production, each with its own organising
characteristics and patterns of development. Second, within any given
mode of production we need to identify and analyse the 'core' set of
social relations which characterise that form of social production and
provides its innermost drives. Third, that 'core' must be understood in
terms of its simultaneous and inter-related economic, political-juridical
and ideological-moral aspects. These aspects are to be grasped, not as
constituting a hierarchy of causality among themselves - and most decid­
edly not as a temporal sequence of 'cause and effect' - but as mutually
In the case of capitalism, its characteristic property relations involve a
whole social complex with an inherent logic which is, itself, possessed of
a 'driving force' that shapes every aspect of social life. This certainly
includes the 'economic' allocation and distribution of material resources.
However, it also includes, equally, the organisation of people in determi­
nate places in a system of power in relation to each other, i.e. a system of
super- and sub-ordination. Also, it includes a characteristic pattern of
'behaviours' including a pursuit of proft at the expense of the needs and
the lives of direct producers, institutionalised 'selfishness' and 'greed', the
requirement that human energies be directed towards mutually competi­
tive ends, and the calculation of costs and benefts in terms of competi­
tively defned ends. All of this assumes a set of 'political' and 'moral',
quite as much as 'economic', imperatives and logics which shape every­
day human action within capitalism and provide the system's dynamic
processes, its 'laws of motion' .
Those who would retain the 'base-superstructure' metaphor need to
Two book b EHen Wo
reconfgure its meaning. It is not a matter of an 'economic base' deter­
mining a distinct 'political and ideological superstructure', but rather of a
'base' that consists of the whole core complex constituting the social rela­
tions of production in its manifold aspects, and which determines a
'superstructure' consisting of the particular specifics of different histori­
cal moments. The movement is not from 'economic' to 'political and ide­
ological', but from 'general to specifc' or 'abstract to concrete' . The pat­
tern of 'determination' is understood to be one in which an underlying
common 'logic' drives and shapes every aspect of social life. A crucial
part of the work of historical and contemporary investigation consists
precisely in exploring how this process of determination, of driving and
shaping, actuaIly occurs in specific settings.
Drawing again on some remarks by Thompson, Wood indicates a fur­
ther weakness apparent in many readings of the 'base-superstructure'
metaphor. The conventional model is better at explaining how the practi­
cal and ideological activities of some classes are determined than it is in
explaining others. That is, it may work quite weIl to explain why the rul­
ing-class ideologies take their characteristic forms. But it is less clear that
the model can account for the various ideologies and impulses towards
resistance and opposition among subordinate classes. To me this suggests
an argument that is not fuIly explicit in Wood, and which bears repetition.
Whichever metaphor we adopt - 'base', 'form of production that pre­
dominates over the rest', 'illumination', 'ether', 'kernel' or 'core' - it must
from the beginning be understood as including something centrally
important. In every class-divided mode of production the underlying
social relations have an inherently conflictual character. The analytical
starting point identifed by Marx in Capital, that of the ' specific form in
which surplus is pumped out of the direct producer', must be compre­
hended, from the start, as the specifc form of a structured process of
class struggle. It is that 'struggle over pumping' (in and with all its eco­
nomic, political and ideological aspects and implications) which shapes
not only the specifc ideological and practical responses of rulers and
exploiters but also those of the ruled and exploited. Explanations of ide­
ologies and practices of resistance, quite as much as of domination,
require reference to the shaping power of underlying relations of produc­
tion, themselves grasped as relations of confict. In this light, the value of
the notion of 'determination' also appears more strongly. For the argu­
ment is that practices and ideologies, among all classes in their interac­
tions with each other and with the material world, are shaped by the fact
of them being responses to general characteristics of the world they share
and that they struggle within.
3. Two bourgeois paradigms
a. Technological determinism
Marxism is commonly held to involve a 'technological determinism'. Here
the underlying motor of development is the historical advance of the pro­
ductive forces, which in turn drives alterations in the mode of production.
A Wood argues, such ideas have no serious place in Marxist theory.
Even if some thinkers using an apparently Marxist terminology (for
example, G.A. Cohen or John Roemer) have sought to import them into
Historical Materialism
the core of historical materialist thinking, these ideas are more character­
istic of bourgeois thought.
Capitalism, as Marx argued in the Manifesto, is marked historically by
a powerful tendency to revolutionise the forces of production. Indeed, by
comparison with preceding societal forms, this appears to be one of its
distinguishing characteristics. However, as Marx went on to argue, most
notably in Capital, this is not, as might appear, capitalism's principal
characteristic. Rather, the immense development of the productive forces
which has most certainly marked capitalist history is a by-product of its
essential feature: namely, capital's drive to expand the production of sur­
plus labour in its specifc form as surplus-value. One mark of capitalist
production from its inception was its tendency to take command of the
labour process and to subordinate it to its own inner compulsions. The
compulsive logic of competition compels all capitalists, on pain of extinc­
tion, to follow two inter-related patterns of behaviour. First, they must
seek to expand the amount of surplus-value extracted from the direct
producers falling under their sway, and second, they must not simply
consume the surplus-value thereby appropriated but re-invest it in means
to its continual enlargement. That drive to accumulate, fuelled by compe­
tition, is the social root of capitalism's expansion of the productive forces.
Technological development, by itself, separately from social relations,
explains little. It cannot - contrary to bourgeois theories of 'industrial
society' (or indeed 'post-industrial society') - account for the specific
patterns of productive development within capitalism. For one thing,
'technology' cannot explain why technical development is biased in par­
ticular directions. Why, for for example, is the immensely enhanced con­
trol over nature's forces and resources in the contemporary world not
directed towards solving endemic problems of human poverty and depri­
vation in the so-called Third World? Or, why are far less resources
directed towards medical and environmental research than to weapons
The development of the productive forces, as a general principle, is
incapable of explaining a key problem in historical analysis: the transition
from one mode of production to another. Theories of 'technological
determinism' direct attention away from the fundamental causes of his­
torical development, away that is from the social struggles around the
very kernel of human social relations in given epochs, and towards a
dubious and misconstrued universal principle. It may be that we can
descry, throughout human history, a weak tendency towards improved
human control over nature. But this cannot be deployed convincingly to
explain historical change. At best, what can be claimed at any point for
the productive forces is that they determine, in extremely broad terms,
'the ultimate conditions of the possible'. They can explain why it is
improbable that capitalism could emerge directly out of hunting and
gathering societies, or socialism out of feudalism. But they cannot explain
why and how feudalism arose in the ashes of the Roman Empire, nor how
and why the transition from feudalism to capitalism occurred. Neither,
we should add, can technological development account, in itself, for any
possibility of the progress of humanity towards socialism. Capitalism's
immense technological achievements, which Marx himself celebrated,
may provide new and more encouraging 'ultimate limits of the possible',
Two book by Elen Wo
but it will not be through these means that any transition to socialism will
be won.
Technological determinist theories assume, in some sense, that the
nature of society can be read off from the nature of the material-technical
means available to and regularly employed by society at a given time. Its
proponents also assume, commonly, that the path of social development
is therefore in some sense pre-determined. Here, indeed, humankind
does appear as the slave of its own machinery, subordinated to its own
technical inventions. By contrast, a Marxism whose starting point is the
social relations of production, and the patterns of struggle embodied
within them, is inherently 'open-ended' in its assumptions about histori­
cal development. In its essentially non-determinist view, development is
always contingent on the outcome of actual social struggles between
human beings in defnite historical settings, and thus necessarily re
a place in any adequate explanation for the consciousness, social o
sation and the like, of those human beings.
Technological determinism - too often wrongly associated, both by its
defenders and its detractors, with Marxism - embodies a version of 'tele­
ology' in which historical development is taken to be the product of
something outside or preceding actual social struggles. It involves a mis­
reading of the nature of capitalist development itself, in which an effect of
capitalist social relations (rapid technological development of a specific
kind) is firstly understood as a fundamental cause, and then read back
into human history as the prime motor of social development. It is, in that
sense, a specifcally 'bourgeois' theory.
b. A capitalist teleolog
However, as Wood suggests, if technological determinist theories are
ultimately rooted in a bourgeois teleology, they do not by any means
represent the only such bourgeois teleology. Interwoven sometimes with
technological determinist premises is another 'bourgeois paradigm'
which Wood spends a good deal of proftable energy analysing and sub­
jecting to critique in both PCC and DAC. (See also Wood's contribu­
tion to this issue of Historical Materialism.)
The originators of this paradigm were the great political economists
of the eighteenth century. For them, the emergence of capitalism -
Adam Smith's 'commercial society' - involved no rupture in social
development. Rather, the principles of capitalist society were already
present in pre-capitalism: in medieval feudal society, and indeed in the
classical Greek and Roman world. Only, there they were 'fettered' by
political relationships which prevented their coming into full flower.
Capitalism's development involved doing away with these historical fet­
ters, liberating a rational society from its former irrational limits. Thus,
the bourgeois paradigm identifies, within the development of medieval
towns and their burgher classes, within the expansion of trade and mar­
kets, the lineaments of capitalism forming within the 'womb' or the
'interstices' of feudalism and finally breaking free in the modern era.
Explicitly in Adam Smith, but implicitly in other theorists, is an in-built
assumption; namely, that the fnal breakthrough of capitalism from its
previous confinement is in conformity with the principles of human
society itself, and of human nature. Long-existing basic human features
Historical MateriaUsm
- in Smith, the predisposition to 'truck, barter, and exchange one thing
with another' - finally found a social form where they could gain ade­
quate expression. In modern capitalist society, eternal principles of
human sociation have fnally reached their terminus point and full reve­
Wood suggests that Marx himself, in his earlier writings, took over
from Smith, Ferguson and others an account of social history modelled
on these lines. As she suggests, a similar line of argument is pursued by
Alan Macfarlane, who interprets the whole history of England from the
Dark Ages and the Germanic forests as the working out of an immemo­
rial principle of individual private property. Th�re has been no real
'transition' from one form of society to another, for society (in England
anyway) was always organised around the same broad, transhistorical
principle. The ' Rational Choice Marxist' John Roemer, too, operates
with a similarly modelled account, in which 'capitalism always existed'.
Involved in these kinds of works is an essentially teleological account of
historical transformation, in which capitalism is read backwards from
the present, and the principles of present-day society are sought in their
presumed origins in the mists of time.
This 'bourgeois teleology' which treats capitalism as the presumed
end and goal of historical development, is found powerfully in bour­
geois sociology's favourite anti-Marxist, Max Weber. At the heart of
Weber's ideas, Wood suggests, lies a conceptual framework which fil­
ters the whole of history through a mesh of assumptions based on mod­
ern capitalist economy. Weber is famous, not least, for his theses about
the role of the capitalist work ethic in releasing the medieval world from
its pre-capitalist values . But, Wood suggests, on closer inspection
Weber's very terminology is shot through with quite vulgar bourgeois
If, for example, we ask what the term 'work' means in Weber's soci­
ology, we find that it is characteristicaly something performed, not by
direct producers, but by capitalists. Likewise, what counts for Weber as
'economic action' in
cludes neither the act of labour within actual pro­
duction, nor the act of appropriation of the fruits of surplus labour by
any exploiting
lass, but something altogether more narrowly focused:
the act of market exchange alone. What counts as 'rational economic
action' by workers is the moment when they sell their labour-power in
the market, but not the moments of their actual labour-time. 'Capital'
for Weber is explicitly a stock of money, but not (as it was for Marx) a
' social relation' . Indeed Weber insisted that 'social factors' must be
excluded from his definition.
Furthermore, Wood charges, because Weber based himself on
entirely 'bourgeois' assumptions, he offered a fundamental misreading
of pre-capitalist history. For him, labour was 'degraded' in the ancient
world, and explicitly within Athenian democracy, but much less degrad­
ed within the medieval European city. A Wood points out, what Weber
actually means is that the bourgeoisie had less space to develop in
Athens than in medieval Hamburg. Weber links the rise of capitalism in
Europe to the rise of burgher-dominated towns and to the rise of the
merchants. Against him, Wood suggests that it was not in those regions
of Europe where such forms were especially predominant that capital-
Two book b Elen Wo
ism first established itself as a functioning system, but in England,
where autonomous towns and burghers were least prominent.
The same bourgeois paradigm, Wood suggests, affects the argu­
ments of Perry Anderson and Tom Nairn, the Ne Left Review writers
who sought to explain the apparent backwardness of Britain in terms of
its failure to have a proper bourgeois revolution. The transition from
feudalism to capitalism in England, Anderson argued, was never prop­
erly accomplished, with the result that capitalist development there was
always perverted by its archaic political survivals. I I The political impli­
cation was that what Britain needed was a second, radical bourgeois
revolution to sweep away all its aristocratic survivals (including the
effete figure of Lord Hume, the Tory Prime Minister at the time that
Anderson and Nairn originally developed their theses) .
For Wood, arguments about England go to the core of the question.
English history is the centre of her attention in PCC, and she draws on
the same arguments again at various points in DAC. England, far from
being 'backward', was the one site where a genuine and autonomous
transition from feudalism to capitalism occurred. Drawing on Brenner,
she suggests that in England the outcome of the feudal struggle
between lords and peasants took a novel form. English farmers were
converted into small capitalists, compelled to enter the market for the
means of production (land and labour power) by an aristocracy that
converted its relations with producers from essentially feudal to capital­
ist ones. A Marx argued in the last part of Capital volume I, the key
process in the transition from feudalism to capitalism in England was
not the rise of an autonomous urban bourgeoisie, but the forced pro­
duction, within the English countryside, of a new class - a propertyless
proletariat. A new and specifically capitalist class configuration
appeared clearly in sixteenth-century England: a 'triad' of commercial­
ising landlords, capitalist tenant farmers, and propertyless wage work­
ers. Only in England did this happen. Elsewhere in Europe, the out­
come of landlord-peasant struggles within feudalism was not the devel­
opment of capitalist social relations but rather 'absolutism', a new shape
for pre-capitalist society in which burgeoning tax-exacting states com­
peted with rent-collecting landlords for peasant surpluses. The history
of early modern England, far from being marked by 'backwardness',
was rather 'peculiarly capitalist'.
The theoretical signifcance of this argument is considerable. Instead
of reading medieval and early modern European history as a story in
which an already developing and nascent capitalism burst finally from
its feudal fetters, achieving what had long been implicit in its potential,
Wood (like Brenner) sees the emergence of capitalsm as a contingent
product of a process of class struggle, within feudalism, between pre­
capitalist landlords and pre-capitalist peasants. Elsewhere - notably in
France, but also across the whole of Europe - that same struggle, which
intensified in the feudal crisis of the late middle ages, produced differ­
ent and decidedly non-capitalist outcomes. The medieval urban com­
mune was not, as so many historians working within the bourgeois par­
adigm have argued, the site of developing capitalism; it remained
locked within the assumptions of a pre-capitalist mode of production.
Nor were the growth of medieval European trade and merchants the
Historcal Materialism
key to capitalist emergence, for - as Marx argued in Capital volume III
- both trade and merchants played their own parts within various forms
of pre-capitalist society without possessing any socially transforming
Only in England were the social relations of production decisively
transformed, and only in England, therefore, was a whole new impulse
to economic and political transformation initially felt. The rise of
England to world predominance in the eighteenth and nineteenth cen­
turies rested on this early start. And it was English power, rooted in its
developing capitalist relations of production, which compelled other
states to emulate its transformation. Only with this difference: in
England, the causes of the transformation were autonomous and
domestic; but elsewhere they were externally induced by the pressures
of English economic and military success, itself founded on the advan­
tages it gained from its early advance. States elsewhere nowhere
achieved a transition from feudalism to capitalism, but from some other
form of pre-capitalism (for example, absolutism in France, the tributary
state formation in Tokugawa Japan, etc.) .
Contrary to the bourgeois paradigm, with its tendency always to 'nat­
uralise' and 'eternise' capitalist social relations by reading them back
into the past, the Brenner-Wood approach to the emergence of capital­
ism stresses the degree to which it involved a qualitative break in devel­
opment, the creation of something utterly new with its own assumptions,
social relations, and its own laws of motion, all quite distinct from any­
thing that had gone before. The surface appearance of backwardness in
English development - its peculiar constitution, the role of unsystema­
tised common law, the weak republican tradition - become, in Wood's
analysis, explicable features of the very advance that England made
towards capitalist predominance. On the surface, it might seem that
English history is marked by continuity. Before, during and after the
transition to capitalism, England's richest and most powerful men con­
tinue to bear aristocratic titles, for example, and to dominate high poli­
tics and culture. Indeed, that continued dominance is what leads the
Tory historian, Jonathan Clark, to deny that anything much happened by
way of change in England. A Wood remarks, for Clark 'aristocracy and
monarchy are by defnition ancient. It seems not to matter what they do
or how they do it' . 1 2 The very things that mark the peculiar development
of English aristocratic landowners - that they became 'improving' capi­
talist landlords, involved in entirely new social relations with their ten­
ants and their agricultural labourers, and shaping a new form of state
power in England - are quite irrelevant to Clark, who explains nothing.
Keeping his eyes frmly shut, he sees nothing to explain.
At the centre of capitalist development stands the fgure who is side­
lined in all variants of the 'bourgeois paradigm' : the dispossessed wage­
worker, compelled to offer labour-power to a capitalist employer, the
secret of whose proft-making lies - as Marx argued - in his workplace
extraction of surplus-value.
What Wood achieves through her whole discussion of the weakness­
es of the 'bourgeois paradigm' is a complex, and to my mind very satis­
fying, inter-weaving of historical and theoretical argumentation. The
treatment of history illuminates the central theoretical points she makes
Two book b Elen Wo
for the necessity of a Marxist understanding, and Marxism is returned
to its proper place as a theory grounded in serious historical research
and understanding.
4. The foundations of capitalist politics
A characteristic of much contemporary writing and thinking on the Left
is a strong tendency to turn away from Marxism in two especially signifi­
cant senses. One is a denial of the centalit of the capitalist nature of the
modern world, and in particular of its 'totalising logic', and the other -
crucially interconnected with this - is a downplaying of the signifcance
of class struggle in politics. Ellen Wood is, by training and inclination,
frst and foremost a political theorist, and her work is informed above all
by the argument that the Lef, in turning away from Marxism in the above
ways, weakens its own sense of the nature of the problems it faces, and
misrecognises its own requirements.
Wood identifies the nature and limits of 'bourgeois democracy' as a
central issue which needs to be confronted directly. In the western liberal
tradition, two aspects of capitalism are held to offer, together, an
immense justifcation for the status quo. The frst is the growth of free­
dom and of legal equality of all persons before the law, itself a direct,
entailed function of the growth of market principles of contract, free
choice and opportunity. The second is democratic government. I combi­
nation, these form the lineaments of 'liberal democracy', the specific
achievement of western capitalism. In the aftermath of the collapse of the
'communist' regimes of Stalinist Russia and its satellites in Easter
Europe, liberal democracy appears especially triumphant. For the sole
apparent alternative to itself has imploded. Indeed, at the very moment
that the East European regimes were finally crumbling, Francis
Fukuyama asserted that History had now, effectively ended. Liberal
democratic capitalism was now revealed as the sole viable alternative form
of society. Indeed, he suggested in his 1 989 article, nothing of major sig­
nificance would ever happen again, for the essential principles of the
modern world were now fixed. Mankind could expect, from now on, a
prospect of 'centuries of boredom' .
Fukuyama's thesis is remarkable not for its utterance, but its recep­
tion. Much of the Left treated it as expressing a basically incontrovertible
truth. The market was indeed, as liberalism asserted, a sphere of opportu­
nity and freedom, contrasting exceptionally favourably with the state­
direction characteristic of Stalinist societies. And the formal democratic
rights enjoyed by the citizens of most modern industrialised countries
were, indeed, an imperishable achievement that should not be threatened
or undermined. There did still remain a space for the Left, but it was now
much more constrained than Marxists had previously assumed. Since the
principles of the market and the liberal democratic state were now con­
ceded, what remained for the Left was an expansion of democratic and
egalitarian principles within a more delimited sphere, that of 'civil soci­
Now, it was also argued, the central role that Marxism had allocated
to the working class' struggle against capitalism was proved to be mis­
placed. The variety of issues posed by 'new social movements' emerging
out of the struggles of the 1 960s all raised questions which could not be
Hisorical Materialism
subsumed within Marxism, and which indeed Marxism was ill-equipped
to handle. In place of a socialist politics centred on the unity of '
struggle', what was now needed was a 'rainbow' politics centred on diver­
sity and difference, and on the politics of the various 'identities' and
'oppressions' still powerfully current within contemporary society.
Ellen Wood frst entered into the fray on these matters with an earlier
book, The Retreat Frm Class ( 1 986) . In DAC, especially, she returns to
these questions. As in 1 986, her concern is once again to assert the cen­
trality of a classical Marxist critique of this rightward moving left agenda.
There are several themes in her discussion. Underpinning them all is
the argument developed in the opening chapter of DAC.
There is a puzzle about capitalism, which can be addressed in various
ways. One of those ways was that adopted by Marx, who asked a simple
question. Let us assume, he said, that the world of market relations is as
its defenders assume. That is, it is indeed marked by the principles of
'freedom, equality, property and Bentham'. Within the market, everyone
exchanges in conditions of free choice, everyone enters into contractual
relations with others, nobody has ownership rights in the bodies and per­
sons of others, and everyone pursues their own self-interest. Assume that
this account is correct, and a major problem remains. Where on earth
does surplus-value come from? It seems impossible that it could be gen­
erated in a formally free and equal world. To solve the problem, Marx
argued, it is necessary to recognise that there is more to capitalism than
this world of market relations alone. Alongside, and interwoven with the
market, lies a quite differently constructed world, that of the capitalist
enterprise. Ad it is here that the secret of exploitation is located.
In Marx's briIIiant account, the buyer and seller of labour-power meet
together, as formally free contracting agents, in the market. They transact
their business under conditions of legal equality, with each of them enti­
tled to refuse to enter into a contract with the other, and each entitled to
bargain over the terms of any contract they do make. But, once the bar­
gain is struck, they move off into a quite different sphere of social rela­
tions, within the workplace. Here the buyer of labour-power demands the
use of the property he has just hired, and sets the seller to work, as a
worker. Now the relationships between them need to be described in a
quite different language. The formal freedom and equality of the market­
place now disappear: Marx turns to the political language deployed by the
analysts of slavery and ancient empires, the language of 'despotism'.
Within the workplace, the buyer of labour-power has become the 'mas­
ter', the 'boss', while the seller has transmuted into the 'hand', the 'work­
er', the 'wage slave'. Here the capitalist can, under the terms of the con­
tract, compel the worker to produce a 'surplus-value' which is embodied
in the capitalist's own property, and which he is then free to realise back
in the market. At the end of the whole circuit when they both return again
to the market-place, the worker remains no better off than he started, but
the capitalist has been enriched by the labouring activity, the expenditure
of toil and effort of the worker.
The real functioning of capitalism thus rests upon the intersection of
two distinct spheres of human interaction, the market and the enterprise.
Each is organised upon distinct principles. A Marx makes very clear,
those distinct principles inherently involve quite different political rela-
Two bo k b Ele Wo
tions obtaining among the participants: contractual freedom and autono­
my of the individual subject in one sphere, and practical subordination in
the other. Capital expands by moving constantly from one sphere to the
other and back again, in an ever-repeated cycle of expanded self-repro­
duction, shifting its form as its moves through the cycle from money to
means of production to commodities. The whole movement, of course,
rests upon a prior condition: the propertylessness of the workers, who are
compelled by need to enter into this unfavourable bargain, to hire out
their most vital and defning human powers, and thus to put their very
capacity for creative work under the domination and direction of another.
It is on the basis of this brilliant analysis of Marx's that Wood builds.
A she points out, at the heart of Marx's analysis is a 'political' and not
simply an 'economic' account of the core, constitutive social relations of
capitalism. Marx's intention, in his critique of political economy, was pre­
cisely to reveal 'the political face of the economy which had been
obscured by classical political economists'. Marx's analysis embodied,
one might add, the sharpest possible attack on liberalism, for he showed
that at the very heart of capitalism - the very system upheld by liberalism
- there was a necessary sphere of unfreedom and human subordination
which completely undermined liberal claims about the pervasiveness of
'freedom' and 'equality'.
Any account of capitalism which obscures Marx's fundamental argu­
ment involves a retur, backwards, to an uncritical rendering of classical
political economy's themes. There is not, as we noted before, a sphere of
life termed the 'economic' which is devoid of both political and legal pre­
suppositions and practical relations. When Marx defines capital as a
'social relation', he assumes that this relation is itself, inherently, a politi­
cal one.
However, this is not how capitalism is normally presented. Here
Wood's specifc argument begins. Whereas, within pre-capitalist forms of
class society, relations of political power are clearly and openly embodied
as necessary elements in processes of exploitation, matters appear other­
wise within capitalism. Capitalist property can be the basis of a process of
pumping out of surplus-value, without needing any 'political and social
embellishments and associations'. A capitalist need not be a 'lord' or
enjoy any state-backed privileges at all, he need not possess any special
direct access to means of legal and political coercion. If all previous forms
of exploitation rested on 'extra-economic coer
cion', within capitalism this
is no longer necessary. What Marx called 'the dul compulsion of eco­
nomic relations' - the simple need to find an employer to pay them wages
- is sufcient to drive workers every day to submit to capitalist exploita­
tion and domination. Workers are impelled, by the everyday neediness
that fows from their lack of anything to sell in the market but their own
labour-power, to hunt for 'work'. And 'work', under these conditions, is
necessarily a direct process of giving up surplus labour.
The power of compelling the production of surplus and appropriating
it can, under these conditions, be 'privatised', in the sense that it can now
be dissociated from direct political power. Direct coercion is no longer an
immediate necessity of exploitation, needing only to remain as a back­
ground condition, protecting property rights in general.
Thus capitalism can appear as a system with distinct 'spheres'. On the
Historical Materialism
one side is an 'economic' sphere, a sphere of 'private life', whose own
juridical and political forms are themselves purely 'economic', concerned
only with maintaining the rules of exchange, the rights of private proper­
ty. On the other side is a specialised public and political sphere, that of
the state, which enjoys a monopoly of legal coercive force. Within capital­
ism, uniquely, the moment of exploitation and the moment of coercion
can be institutionally separated from each other. Those who appropriate
surplus-value need not, in principle, concern themselves with the sphere
of the state, nor need they perform any public or social functions. The
fnctioning of capitalist enterprise can occur in a distinct 'private' sphere
of power, quite apart from the world of the state and 'politics'. This sepa­
ration does not really involve the emergence of a 'non-political' sphere of
'eonomic life', but actually the differentiation of political power into dis­
tinct spheres. In one sphere, that of the state, it is openly acknowledged
as a sphere of coercion, or 'legitimate force'; in the other its political
character is concealed as 'the rights of property'.
Historically, Wood suggests, feudalism played a signifcant part in this
development. It involved the 'privatisation of political power' as the lords
moved to integrate private appropriation with the authoritative organisa­
tion of production. The emergence of capitalism perfected this privatisa­
tion by its complete expropriation of direct producers and its establish­
ment of absolute private property. The necessary correlate of this devel­
opment was a new and stronger form of centralised public power, a state
that stood apart from the exercise of power in 'economic' life while pro­
viding it wth general protection. Those who managed the process of
exploitation were purified of public, social functions, which were no
longer any part of their responsibility.
Thus, behind the surface appearance of a 'non-political' realm of eco­
nomic relations a quite different reality is to be descried: a division of
political powers as between (public) states and (private) capitals, in which
the exercise of differentiated powers in each sphere is actually a condition
for the other. It is precisely the function of Marxist theory to uncover that
political - and contingent, historical - aspect of the economic, as against
all those tendencies within bourgeois theory, from the classical political
economists to modern sociology, which tend to make the specific rela­
tionships of capitalism appear eternal and natural. Bourgeois theories
achieve this mystification through two characteristic and obscurantist
errors. First, they treat the division of modern society into distinct
spheres as a feature of all history, rather than understanding it as a spe­
cifc feature of capitalism. Second, they treat the social relations of the
capitalist 'economic sphere', once they have denuded them of their politi­
cal and historical features, as 'natural' attributes of human social life, pre­
sent throughout all history.
This separation of the political and the economic, which appears only
within capitalism, provides a necessary starting point for considering
modern politics. In particular, Wood focuses attention, in the second half
of DAC, on two specifc issues. The frst of these is modern democracy,
and the second the kinds of claims currently offered on the Lef about
'new social movements'. In both areas, Wood seeks to rescue and develop
classical Marxist principles of analysis.
Two book b Ellen Wo
5. Democracy
Ellen Wood is also the author of a fne study of ancient Greek democra­
cy, Peasant-Citizen and Slave. She draws on that study to illuminate the
meaning, and the limits, of contemporary 'democracy', and to trace its
connection to the changing contours of class struggle in different modes
of production.
The term ' democracy' originally had an unambiguous meaning:
direct power in the hands of the common people, the demos. Indeed, its
original meaning was not that far from modern political notions of 'the
dictatorship of the proletariat'. It was democracy in this form which was
regularly attacked by aristocratic thinkers from Plato onwards.
Democracy, as the rule of the lowly, stood necessarily opposed to the
power of any exploiting minority. Indeed, that opposition was precisely
what defined democracy's emergence in Athens two and a half millennia
ago. The key characteristic of Greek city-state democracy was not, first
and foremost, slavery, but the citizenship powers won through class
struggle by working peasants and artisans. In ancient Greece, 'free
labour' was first invented as a practical political relationship, inextricably
linked to the limitation of landlord powers of debt bondage, serfdom and
taxation. There, where no separate 'economic' realm of non-political
domination and exploitation existed, democracy had a direct and sub­
stantive content. To win democracy in such a social setting, where pro­
ducers were not subject to the 'private' control of employers nor subject­
ed to the compulsions of the market as a regulator of production, was to
defeat exploitation. Indeed, Greek democracy was, in practice, a far
more complete form of democracy than any enjoyed today within
Western 'liberal democracies' .
What we know as democracy today has quite different presupposi­
tions. The very term democracy has been tamed so far as its implications
for the mass of the population are concerned. It has been diluted of its
social content, confined to a special sphere, and developed as a mecha­
nism for excluding the modern citizen from the direct determination of
public policy. The possibility of this dilution can only be understood with
reference to the formal separation of the political and the economic
which, as Wood has argued, arises within the framework of capitalism
alone. For capitalist society permits the combination of formal citizen­
ship and democracy both with the practical despotism of the workplace,
and wth the general subordination of its members to the market-driven
imperatives of capital accumulation. Where, within ancient Greek soci­
ety, the winning of citizenship and democracy signalled an attack on
exploitation, within modern capitalism the expansion of modern democ­
ratic rights leaves exploitation and alienation completely intact.
Within conventional histories of modern democracy, the great turn­
ing points which are regularly celebrated are not those moments when
the dispossessed struggled to overturn the power of their rulers - not the
peasants' revolt of the fourteenth century, not the Levellers' or Diggers'
movements of the seventeenth-century revolution, nor the Chartists of
the mid-nineteenth century - but a quite different sequence. Magna
Carta and the 'Glorious Revolution' of 1 688 are taken to be founding
moments. Yet what was their social content? It was not the 'people' win­
ning citizenship, but lords and masters asserting their power and their
Historical Matrialism
freedom vis-a-vis both the monarchy and the majority of the population.
What orthodox history celebrates is the freedom of property, the devel­
opment of an institutional framework of liberalism, that is, of the limita­
tion of the state's power over the propertied and their private exploita­
tion of direct producers. Those rights of property were established and
made secure before ever a breath of 'democracy' was permitted, in the
shape of an extension and completion of the franchise. Indeed, what is
remarkable about democracy is how late it developed in the history of .
capitalist politics. If democracy means 'one (adult) person one vote',
then - as Goran Therborn pointed out - Britain only became a democ­
racy in 1 928, and in the USA the condition was only achieved in the
mid- 1 960s, when black voters in the South fnally won the right to regis­
ter. "
The process of extension of democratic citizenship rights within the
framework of capitalist society also involved its dilution by comparison
with the ancient Greek model. The eighteenth-century struggle over the
American constitution, Wood suggests, represents a key turning point in
Western capitalism. The participation of working artisans and farmers in
the processes of the American revolution made it impossible for the
Founding Fathers to exclude them directly from any political participa­
tion. Those who designed the new American constitution invented, and
secured, a watered-down version of popular self-government termed
'representative democracy'. Here popular power is 'filtered' through
'representative' institutions which are alienated in practice from the elec­
tors. Politics, in such circumstances, becomes necessarily a duplicitous
process, a matter of evasions and misrepresentation. The term 'politi­
cian' itself accumulates associations with charlatanism and bad faith.
'Politics' becomes, in any case, limited in its scope to a mere 'sphere' of
social life, in which the continuity and expansion of the power of capital­
ist property is assumed, and the whole of political life is subordinated to
an impersonal form of regulation through the workings of the capitalist
market. States themselves, and thus public policy, are subordinated to
the completely undemocratic impulses of a global imperative of capital
In short, a socialist critique of modern democracy must insist on its
essentially limited character. Private power, the domination of the few
over the many, is the dominant principle of capitalist society, and con­
fnes the scope of such democracy as the citizenry have won.
Yet this is not how a large part of the Left currently thinks about the
question. Far from offering a fundamental critique of the restricted
meaning of democracy within contemporary capitalism, much of the Left
tends to celebrate is foundations. At the centre of its position is a mud­
dle, which Wood's fnal essays in DAC address. This muddle centres on
a concept - 'civil society' - which has recently enjoyed something of a
6. 'Civil society' and 'new social movements'
'Civil society' is half of a pair of concepts whose other term is 'state'. Its
moder meaning was formed in the liberal Enlightenment discourse of
the eighteenth century, to delineate that sphere of social life which was
neither the purely private sphere of the family-household, nor the public
Two bok b Ellen Wo
sphere of the state. At its heart were the institutions of the capitalist mar­
ket and private property, the sphere of life anatomised by the classical
political economists, and the key focus of attention for Marx in his cri­
tique of political economy. In liberal hands, civil society denotes a sphere
of social activity marked by freedom and voluntary action, untrammelled
by the inherently coercive relations of the state sphere. The 'civil soci­
ety' /'state' distinction thus replicates, in conceptual terms, the formal
separation of the economic and the political which is a form of social
organisation peculiar - in its full extension - to capitalist society.
The concept of civil society enjoyed a renaissance on the Left during
the 1 980s. One reason was that it appeared to oppositional thinkers in
Eastern Europe to clarify the nature of their struggle against the mono­
lithic pretensions of their Stalinist state and party apparatuses. Another
reason was that, in the West, it seemed perhaps to promise a way of
thinking about the emergence of various emancipatory movements
which were seemingly dissociated from 'class struggle' - struggles
against racist and gender oppression, against ecological destruction and
militarism, against limitations on sexual identity, and so forth. Its attrac­
tion was thus rooted in two signifcant matters. First, it focused - as did
neither Stalinist nor social democratic practice and thought - on the
dangers of state aggrandisement and oppression, insisting powerfully on
the importance of freedoms of speech, association, tavel, and all man­
ner of forms of free and voluntary social and cultural activity. Second, it
stressed the significance of a variety of forms of oppression, and resis­
tance, which could not be simply 'reduced' in status or understanding to
mere 'expressions' of class. 'Civil society', considered in terms of the
positive value it embodied, was thus potentially important as part of the
process by which the Left might clear out from its theoretical attics the
practical and conceptual lumber it had accumulated during the larger
part of the twentieth century, when most socialists saw themselves as
having to choose between two narrow and constricting forms of left poli­
tics - Stalinism and social democracy - which themselves proved to be
deeply conservative.
But any virtues the term 'civil society' might possess were outweighed
by the confusions which came with it. Civil society, as Wood remarks, is
a 'conceptual portmanteau', which bundles together under a single head­
ing all manner of disparate forms of social activity and organisation. In
the hands of those who currently celebrate it, 'civil society' includes the
organisation of households, all manner of voluntary association, and the
institutions of the capitalist system, all of them lumped together into a
single category on the ground that all involve freedom from direct con­
trol by the state. The effect is both to confuse the real issues facing the
Left, not least by disguising the essentially coercive character of the
social relations at the very kernel of capitalism. Knowingly or not, con­
temporary theorists of 'civil society' demand that socialist thought
retreat back to the plane of liberalism. I.
What disappears from view in their analyses is capitalism itself and
what Wood terms its 'totalising logic', that is, capitalism's tendency to
subsume all social relations under its characteristic inner compulsion to
competitive accumulation and heightened exploitation. If all that is
recorded about capitalist property is that it is 'free' of state control, and
Historical Materialism
if freedom from state control is taken as the sole significant badge of
emancipation, then both the domination of capital within the world of
everyday production and the inherently coercive nature of market rela­
tions slide quietly out of sight. The Lef begins to celebrate what it
should oppose, and to join hands with its former enemies.
The explanatory power of much contemporary left theory, and thus
its capacity for strategic thinking, is accordingly weakened. A parallel
and intimately related error appears in the ways that much of the con­
temporary Left thinks about such issues as movements against women's
oppression, racism, war and ecological degradation. On essentially
moralistic grounds, all forms of oppression and exploitation are declared
to be of equal significance, none (and in practice especially 'class'
oppression) being allowed a 'privileged' status above any other. Al cats
are grey, all suffering is equally valid, all resistance practices and move­
ments demand equal respect and uncritical support. Such moralism con­
ceals from view vital theoretical and practical questions about both
causal determination and strategy. How are different kinds of oppres­
sion and exploitation internally related with each other and with the
encompassing logic of capitalist production relations? To what degree
can resistance to them be logically compatible with the continued domi­
nance of capitalist property and market regulation?
In posing such questions, Ellen Wood opens up a systematic discus­
sion, badly needed on the Left, both about the theoretical and strategic
possibilities and signifcance of the so-called 'new social movements' and
about the ways in which their participants and members think about these
matters. If there is a single theme uniting her various discussions, it is
that the Left must get beyond the uncritical celebration of 'diversity' and
'difference', to an understanding of the practical, if differentiated, unity
of these matters imposed upon them by the totalising logic of capitalist
social relations. It is the workings of capitalism which allocate to each
separate struggle and each separate issue their particular signifcance.
In the final chapter of DAC, Wood picks up and criticises yet another
way in which the contemporary Lef reveals itself to be sliding back,
uncritically, towards liberal positions. In precisely a period when the
capitalist market's devastating impact on the world is ever more appar­
ent, whole sections of the Left seem intent on finding reasons to cele­
brate it rather than to anatomise and oppose its workings and assump­
tions. The Left requires, she urges, a conception of democracy which
does not trim and diminish itself to acceptance of market competition as
a regulator of the fundaments of social productive life; it needs to reach
and struggle for an immense extension of democratic principles into the
regulation of economic life itself. Socialist theory and practice need to
focus on the problems signalled in its original and highest aspiration:
that of subordinating production, not to the imperatives of competitive
capitalist accumulation, but to human need. The pursuit of such a goal
must involve the struggle to remake and democratise not merely rela­
tions within workplaces but also those between them, to oppose the spe­
cifc domination of every separate capitalist over labour and the general
domination of the capitalist market over the whole of society. That
struggle needs to be brought back to the heart of a left-wing agenda
from which it is almost disappearing.
Tw book b Elen Wo
7. Class struggle
It is not suffcient, however, to criticise contemporary left currents for
their rightward drift, unless, in parallel, there is an ongoing recovery and
development of authentic Marxist ways of looking at and acting in the
world. This is an especially necessary task, given the accretion within
Marxism itself of all manner of misconstruals of some of its key con­
cepts. We have already seen how this effort at rediscovery forms an
important part of Ellen Wood's overall agenda, but one element within
her reconstruction remains to be emphasised. This is her discussion of
the term 'class', a matter of great importance for any serious effort to
counter the way most of the Lef has tended to turn its attention away
from matters to do with 'class struggle'.
Wood's account of this issue is focused around a reconstruction of
the logic embodied in Edward Thompson's historical writings. At various
points, Thompson set a distance between his own and sociologists'
understanding of 'class'. Within orthodox (bourgeois) sociology, class
appears as a 'structural location' , a place in a system of distribution. If
we were to represent the typical sociological conception in pictorial
form, class would appear as a 'scale' (or a ladder) of positions within
society, each point on that scale being associated with a greater or small­
er quantity of some uniform function. That function might be, as in Max
Weber, market-derived 'life-chances', or - as in the British decennial
census - 'occupational status' . Such notions are imported into Marxist
discourse by the self-styled 'rational choice Marxists', who treat class as
a distribution of 'assets' and 'endowments' - the very conception that
Rosa Luxemburg attacked within the 'revisionist' Bernstein.15 Here, peo­
ple belong to a 'class' if they share a similar position on the scale of dis­
tribution, or a similar share of the particular quality that is being mea­
sured. To speak of 'class' within this framework is to discuss the degree
to which the distribution of 'goods' and 'chances' etc. is unequal. What
is immediately apparent about such a notion of class is that it does not,
in itself, directly imply anything about social relationships. To speak,
within such a conception, of 'class struggle' or of the formation of 'class
consciousness' is usually to speak of an inherent improbability.
Marxism distinguishes its conception of class from that of bourgeois
sociology, not through identifying some other quality to be scaled difer­
ently, but by giving a quite distinct meaning to the term in the first place.
A Wood expresses the matter, for Marx and for Thompson alike, class
meant not 'structural location' but 'process and relationship'. The start­
ing point of a Marxist conception is not a given distribution of money,
education, occupational status or other 'life chances', but 'class strug­
gle'. The core defining feature of a given mode of production, Marx
argued, is the form in which surplus is pumped out of direct producers,
and that core feature -which defnes the key relationships and processes
of an epoch of social production - is precisely the root of class struggle.
The social relations of production distribute people into class positions
which embody, necessarily, antagonisms of interests and thus generate
conditions of confict. These antagonisms and conficts concern not sim­
ply the distribution of various means to consumption and status, as in
the bourgeois sociological conception, but equally concern the relations
of domination and subordination inherent within the relations of pro-
Historical Materialism
duction, and the moral and value assumptions tied to them.
Far from 'class' being simply - as Marxism's critics allege - an ideo­
logically motivated theoretical construct which is imposed on historical
evidence from without, class is understood as a real generator of a vari­
ety of actual struggles and developmental processes within a given form
of society. That is, class is seen as having real and powerful determina­
tive effects on everyday social relations and development.
I is from this 'basic material', as it were, that people may develop a
conception of themselves as members of 'classes'. It is from this starting
point in practical struggle that there may occur that transformation
which Marx summarised as a shift from 'class in itself' to 'class for
itself' . Or, in short, it is via people's experience of underlying social
antagonisms that they learn ways of organising themselves and concep­
tualising their situation. I, on the one hand, Marxism insists that there
are 'objective determinations', these do not impose themselves on some
undifferentiated, blank raw material, but always on active and conscious
historical beings, shaped in part by their own prevous history of struggle
and development. To the degree that class formations emerge, i.e. that
people come to identify their situation in 'class' terms, they do so as they
live their productive relations and the determinations to which they are
subject. There is no automatic or mechanical process by which partici­
pants within a situation come to define it in any given way. Thus a dis­
tinction needs to be drawn between two distinct sets of questions. First,
what social relations define the specific class antagonisms of a particular
historical epoch; and, second, how and in what forms do people engage
in 'class formation'. It is the latter issue which poses the more pressing
questions for historical analysis. And, one might add, it is the latter issue
which poses the key questions for any active Marxist politics.
What this view of class enabled a Marxist historian like Thompson to
do was to reveal, in his studies of eighteenth-century England, a set of
processes and relationships which he termed 'class struggle without
class'. That is, Thompson identifed, withi
the collective and customary
practices of the eighteenth-century poor, forms of resistance to their
rulers and exploiters which rested on the antagonisms inherent in devel­
oping capitalism in that period, but which the poor had not yet learned
to conceptualise in 'class' terms. Further, his famous study of the early
nineteenth-century working class was focused on precisely the processes
through which workers in Britain did indeed develop what might be
termed a 'class sense' of their situation and of the general antagonism
between themselves and the British bourgeoisie and the state. The prac­
tice of class struggle, one might say, led workers to form themselves into
a consciously developing 'class' , and it was in this sense that, in
Thompson's phrase, 'the working class was present at its ow making'.
Thus, the 'experience' of class struggle, whose forms of expression
may be enormously variable, is a crucial 'middle term', conceptually,
between the notions of ' social being' and ' social consciousness' . Far
from dissociating consciousness from being, this way of thinking about
class focuses on the ways that 'objective structures' of social relations do
actually impinge on people's lives, and come to be understood by them
(variably) in 'class' ways. To explore such questions, and to be sensitised
to them, is -for the historian - a question requiring empirical study. For
Two bo k b Elln Wo
active socialists, moreover, it demands a constant attention to the very
concrete forms of understanding, organisation and struggle through
which people confront the real problems that life within capitalism poses
to them. What the Marxist conception of class invites is not simply, or
principally, some continuous and abstract argument about the general
shape of the 'class structure', but a constant active and empirical investi­
gation of the processes of social and political struggle, and also - a point
less developed within Wood's writings - forms of practical intervention
into those processes in the shape of socialist politics. Underpinning this
conception of class is a notion which is central to Marxist politics: that
the working class is a key actor in processes of historical transformation,
and not least, in its own self-making and self-definition. That idea
underpins the core defning element of Marxist politics - the principle of
the 'self-emancipation of the working class' - and stands ineluctably
opposed to all conceptions of politics and of historical process which
treats the working class as merely 'acted upon' by forces it cannot
understand, and for whose practical solution it must turn to enlightened
intellectuals who alone can comprehend the whole. A democratic revolu­
tionary socialist practice is assumed within this conception of class,
which is entirely missing from conventional and decidedly bourgeois
sociological thinking.
There are matters requiring extensive theoretical development here,
concerning socialist politics, which are beyond Wood's selected remit.
Let me just note that Wood's general argument could be taken very
much further, and could indeed extend to the demonstration of impor­
tant linkages - which some might find very surprising - between
Thompson's historical arguments about 'class' and the Marxist theory
and practice of a revolutionary like Lenin.
Most of the Left today does not understand 'class' in this way. It
relies more on the orthodox sociological conception of class, which has
the effect ¯inter ala ¯ of narrowing the scope of the term to simply one
element of social differentiation. If we take '
lass' to refer to nothing
more than 'economic inequality', and 'class struggle' to refer to nothing
more than conflict between workers and employers over 'wages and
working conditions', then of course the ground is already well-prepared
for the argument that, beside class, and on a morally and indeed strate­
gically equivalent plane, there are other forms of inequality and distinc­
tion which must be given equal weighting. Indeed, the ground is also
prepared for conceptualising popular struggles against racism, sexism,
homophobia, militarism and ecological destruction, along with other
forms of injustice, discrimination and oppression, as essentially distinct
and separate conflicts, each with its own specifc 'issues' and its own
specifc 'movements' . Nothing intrinsic unites them, and the best that
may be hoped for is a provisional 'coalitional' politics founded on a lib­
eral 'respect for difference'. 'Class' becomes merely one among a host of
issues, of no greater significance than any other. Those who assert the
especial importance of 'class struggle' do, within this conception, seem
to engage in the 'economic reductionism' of which Marxism's critics
regularly accuse them. Why 'privilege' one form of 'stratifcation' and
'differentiation' over all others?
But if the issue is re-cast in the terms implied within both Marx's and
Historical Materialism
Thompson's (or indeed, Lenin's) understanding of the terms '
struggle' and 'class formation', then matters appear very differently. The
class relations constituting capitalism, as Wood suggests, refer not sim­
ply to personal identities, a principle of stratifcation, or a specifc system
of power relations, but also to a distinctive and encompassing social
process which shapes and limits all social relations in a given epoch:
namely, the accumulation and self-expansion of capital. And that encom­
passing social process, it needs to be stressed yet again, is to be under­
stood always as a process which necessariy generates manifold forms of
both partial and more general resistance. In this light, 'class struggle'
refers to one face of what Wood terms the 'totalising logic' of capitalism.
If one were, as much of the contemporary Left assumes, to set out
seriously to undermine the fundamental Marxist argument about the
determinative power of class struggle, one would have to demonstrate a
very difficult case. One would need to show that the various specific
issues and contested identities with which different particular move­
ments grapple do not come i any signifcant way within the determina­
tive force of capitalism, its property system, its expansionary and com­
petitive drives, and its inherently socially divisive tendencies. One would,
for example, need to demonstrate the historically implausible case that
racism has an existence and history quite independent of capitalism; that
struggles to defend the environment have nothing whatever to do with
resistance to the destructive logic of capital accumulation, and so on
through all the various spheres and movements.
Uncritical celebrations of both 'civil society' and 'new social move­
ments' in practice fail to demonstrate this, but instead rely on two unsat­
isfactory theoretical procedures. First, as already argued, a restricted
and impoverished conception of class as 'stratifcation'; and second, a
moralistic prescription about the dangers of 'devaluing' other spheres of
human experience than 'class'.
The implications of the account of class within Thompson and Marx
point in a quite different direction. 'Class struggle' is in no sense limited
to questions of 'economic distribution', but is generated by and calls into
question the very fundamental drives of capitalist society, in all its facets.
The interests of the working class lie, not simply in the re-division of the
economic cake, in higher wages, shorter working hours, or greater secu­
rity of employment, but in the re-casting of the whole social system. The
issue is not just 'the slice of the cake' but the bakery itself; not improved
conditions of employment, but abolition of the employment relation
itself; not higher wages alone, but the abolition of the wages system.
A central issue i the 'class struggle' is the process of c
lass formation
itself; that is, the inherently political and ideological battle to constitute a
practical movement which can challenge and transcend the very limits of
capitalist society. There is no inherent divorce between 'class struggle'
on one side, and the struggle against environmental destruction on the
other, for the ecological movement is itself precisely a specific manifesta­
tion of the class struggle. Struggles against racism, ethnic differentia­
tion, sexism and homophobia are not distinct from and opposed to the
ongoing political battle to 'make' a working
lass with a sufciently uni­
fed sense of itself as the agency capable of general human emancipation.
Those theorists and practitioners of 'social movements' who restrict the
To bo k b Elen Wo
meaning and scope of their own struggles to particular issues fall under
the same rubric as those the classical Marxists criticised for 'economism'
and 'mere trade unionism' . They contribute, in their own way, to repli­
cating and consolidating the separation of spheres characteristic of capi­
talist society. They turn attention from and obscure the very totalising
logic of class relations that impels and constrains their own activity.
They strategically disarm themselves and their audiences alike. They
claim that their political project, which takes in everything and values
everything on the same theoretical plane, is more inclusive than the
struggle for socialism; but in practice it excludes a struggle for socialism,
by denying the 'primacy' (i a determinative, not moralistic sense) of
class struggle as the key focus in a real struggle for general human
emancipation. For, apart from anything else, some recognition of rights
to diversity and difference, and thus some forms of struggle for equality,
are actually perfectly compatible with the continued functioning of capi­
talist competition and exploitation. There is, for example, no intrinsic
incompatibility between the recognition and positive valuation of differ­
ences in sexual preference and the maintenance of capitalist production
relations. On the other hand, the abolition of class exploitation and com­
petition is absolutely ruled out by the functioning rules of capitalist pro­
duction. By making no theoretical distinctions between different aspects
of the class struggle, contemporary left theory too often collapses back
into what turns out, in practice, to be an accommodative liberalism
dressed up in radical rhetoric.
This is in no sense an argument against struggles to combat racism,
sexism or homophobia, or in favour of acceding to the destructive
impulses of capitalism as witnessed in war and environmental pollution.
But it is an argument in favour of the development of a socialist analysis
of each and every distinct form and aspect of oppression, with three
concerns. First, we need to develop explanations of the relationships
between these oppressions and the overall functioning of capitalism,
without assuming in advance that each explanation will be of the same
order. Second, we need to introduce a socialist argument into move­
ments confronting these various aspects of contemporary capitalist soci­
ety, with a view to challenging their bourgeois leaderships and connect­
ing these specific struggles back into the general contest to transform
the deep structures of modern society. Third, we have to bring the spe­
cific issues posed by these distinct movements into wider circles of
opposition, for example by carrying the fight against racism into the
trade-union movement or the gay and women's movements.
Ultimately, the socialist argument is two-pronged. Not only will each
movement, without a sense of an overall anti-capitalist strategy, itself fail
in its specifc object, but it is in the interests of the movement as a whole
that each separate struggle should be successful. The interests of the
working class cannot be sensibly counter-posed to the demands of
female liberation, of anti-war movements, of anti-racism or of the fght
to make the planet habitable. To the degree that these issues are not
themselves the living tissue of the working class's struggle for its own
liberation, that movement will be divided against itself and incapable of
providing the leading force in the battle for societal transformation. The
value of the Marxist argument is its theoretical stress on totality and its
Historical Materialism
practical stress on unity.
In all the above, I have attempted to represent key arguments within
Ellen Wood's two books, and to explain why I think they deserve the
most serious attention. In the remainder of the article, I pick up several
themes where further exploration and development seems needed.
Part Two. Criticisms and developments
1. 'Unfnished Capital'
I suggested in the introduction to this review that one of the problems
Marxism has to deal with is the 'unfinished' character of Marx's own
master-work, Capital. We do not have even the barest outlines of the vol­
umes he proposed to produce on 'Wage - Labour' , ' The State' ,
'International Trade', or 'The World Market and Crises'. There i s a good
case to be made for the proposition that the 'missing volumes' do need to
be developed by Marxists.
One reason is that, without a sense of the incompleteness of the
planned whole work, we shall misunderstand what we do have. All man­
ner of false readings of Marx can flourish. For example, Marx can be
seen as merely producing a alternative 'economics'.17 Or Marx can be
seen as taking classical political economy so seriously that he came to
adopt its standpoint.
Ellen Wood, as already indicated, draws on various discussions by
Edward Thompson to tease out the meaning and re-assert the centrality
of 'class struggle' within Marxist thought. For Thompson, that emphasis
was part of a larger and utterly appropriate effort to restate the 'human­
ism' at the core of Marxist theory and practice. However, in one respect
at least, Thompson came to the conclusion that Marx's theoretical mas­
ter-work actually involved a kind of falling back from the apparent
humanism of his earlier writings. The issues involved are signifcant, for
they concern the sense we are to give to one of Marx's key concepts: the
social relations of production.
In his famous attack on Althusserianism in The Povert of Theor,
Thompson expressed serious reservations about the Marx of the
Grundrisse and Capital. Marx made the mistake, he suggested, of becom­
ing so involved with the critique of political economy that he tended to
make its problems his own, and to accept himself the limits to under­
standing imposed on thought by its problems and categories. Against a
structuralism which he suggests Marx tended to fall into, and which
embodies an 'idea of capital unfolding itself, he upholds a vision, to be
found and developed from other insights elsewhere in Marx's writings, of
a historical materialism which has to do with 'a real historical process' .
Thompson does identify, in his discussion of the Grundrisse and
Capital, a real problem in Marx's greatest writings, but I am not con­
vinced that Thompson took proper measure of the nature of the difcul­
t. It is certainly the case that, at numerous points in Capital, for exam­
ple, Marx writes i a very peculiar way. The grammatical subject of many
of his sentences, the active element in his theorising, seems to be Capital
itself, not any human beings. A 'structural relation', rather than human
agency, appears to provide the motive force of historical development as
Marx accounts for it in his critique of political economy.
Two bo k b Elen Wo
Wood, as we have seen, very correctly insists that the very core of
Marx's own procedure was to present the 'political aspect' of the world.
She recovers the stress on 'class struggle' in Marxism. She draws on
Thompson himself to suggest that the generative 'kernel' at the heart of
every class-based mode of production is a struggle between exploited
direct labour and its ruling class. What, then, are we to make of
Thompson's criticism of the 'scientific' Marx?
We can, it seems to me, adopt to quite different attitudes to this. The
first is Thompson's. His instinct is to turn away from the Marx of Capital
and its apparent implicit structuralism. There is, however, a second possi­
bility, whose implications have begun to be drawn out in some recent
writing by Marxists like Michael Lebowitz and Felton Shortall. Their crit­
ical strategy is to highlight the inherently incomplete nature of Marx's
project, considered in its own terms. They point to a whole series of indi­
cations, frstly, that Marx's own intentions were to take the critique of
political economy far beyond the limits of the work we have inherited,
and secondly, that within the work there al manner of 'provisional clo­
sures' (Shortall), imposed by Marx on his own argument. These mark,
not a falling back into structuralism, but a 'setting aside' of numerous
problems for later attention. That 'setting aside' is an inherently necessary
element in any theoretical presentation. However, the issues thus set
aside do still require to be brought into play. The problem is, Marx never
did get back to them.
In Lebowitz's phrase, Marx's Capital as we have it is not simply
'unfnished' but radically 'one-sided'. And it is that one-sided incomplete­
ness that has permitted all manner of structuralist and objectivist readings
of a work whose overall intentions are decidedly different. Lebowitz and
Shortall both suggest that what is 'missing' from Capital, making it 'one­
sided' in its revelation of the character and inner workings of capitalist
society, is working-class self-activit, in its contest with capital. Lebowitz,
indeed, attempts a fascinating provisional reconstruction of the kinds of
theoretical categories which would need to underpin the missing 'Book
on Wage Labour' which Marx - in his original plans for the work -
announced he intended to produce. My own view is that both Lebowitz
and Shortall are essentially correct in their view of the status of Capital. It
is a radicaly incomplete work, in which Marx never got beyond delineat­
ing the heart of capitalism 'from one side'. Thus the work cannot be
properly understood unless - in some imaginative sense - we make the
effort to 'complete' it. Lenin concluded during the First World War that
the reading of Hegel was essential to understanding Capital (and hence
that for fifty years no one had understood i) . Now, it seems, the even
more challenging implication is entailed, that the writing of the rest of
Capital is essential to its understanding!
The ramifcations of such an argument extend far beyond matters that
can be discussed here. One implication, though, does demand attention.
Marx's Capital represents the most systematic effort ever made to defne
the social relations of production of capitalism. If the effort was never
completed, the incompleteness consists, in essence, in the absence from
the very heart of his conception of capitalism (as presented in Capital) of
a developed sense of class struggle as a vital principle at the very kernel of
the system. The further development of the critique of political economy,
Historical MateriaUsm
which requires placing the class struggle back in its heart, perhaps offers
the only way to overcome the apparent disjunction between ' Marxist
political economy' and the Marxist historian's view, in which agents are
more than 'bearers' of structural relations, but are
qually active and self­
transforming contestants in an open-ended, developing process of strug­
gle to transform society.
What bearing has all this on our reading of Ellen Wood? First, it sug­
gests a need to stress, even more strongly than she does, that all the cate­
gories of Marxist theory - including 'base' or 'kernel', 'social relations of
production', 'commodity' and 'value', 'exploitation' and 'accumulation'
and the like - refer to inherently contradictory and contested forms of
social practice. Second, if Marx's work was massively unfnished, then a
complex work which further develops his approach is required if we are
to use it effectively to make sense of the world since his death. A whole
series of problems - historical and theoretical, and ultimately practical -
then present themselves. In what follows I try to mark out a few of these
as they are posed in Wood's two books.
2. Competition and the 'market'
There is, it seems to me, an underdeveloped theme in Ellen Wood's
account of capitalism as a mode of production. She suggests, with
increasing emphasis through the essays in DAC, that the 'market' is itself
a system of coercive relations, which compels its participants into pat­
terns of action. Yet at no point does she analyse this social form. There is
a curious silence on this matter, for the social relations that underpin the
market offer opportunities and challenges to a Marxist political theorist,
and especially one with Ellen Wood's theoretical and political acuity.
Indeed, I shall suggest that her failure to explore this question leads her
into potential weaknesses in her understanding of some features of capi­
talist development.
Marx famously began Capital, not with an immediate leap into consid­
eration of the exploitation of wage-labour, but at a more fundamental
point: with the commodity-form. The concepts he frst discusses are
'value', 'money' and 'exchange'. One might wish that he had taken the
development of his own argument around these matters further than in
fact he did - especially into a specifc discussion of 'competition' as a cat­
egory. Yet the materials Marx does provide make it sufciently clear that
in this area, as in all others, he was indeed following the protocols that
Ellen Wood identifes as lying at the heart of his critique of political econ­
omy. That is, here as elsewhere, he is concered to 'present the world in
political terms' . Instead of treating commodity production as a 'natural'
form of human association, he offers a perspective on it which goes far
beyond anything the classical political economists attempted. First, he
insists on the historical character of the social relations underpinning
commodity production, drawing a sharp contrast between the social rela­
tions in that form of society where commodity production is predominant
and the social production relations obtaining both in past societies and
within a future communist society. Second, he poses a problem which
none of his bourgeois forebears had even thought to ask. Why is it that
the goods which people produce in contemporary society take the form of
commodities, i.e. why is there a 'form of value' at all?
Two book by Ellen Wo
To answer that question, we need to investigate a side of capitalism's
social production relations which, to date, remains unexplored in Ellen
Wood's writing. Any such investigation, however, would need to stress
just the features that, she insists, characterise Marx's account of produc­
tion relations generally. That is, the social relations of commodity pro­
duction, of value, exchange, market competition and the like are anything
but simply 'economic' relations, but also have political and juridical
assumptions built into them along with accompanying ideological and
moral prerequisites. What seem, on the surface, to involve purely 'eco­
nomic' matters are actually historically specific and politically very signif­
cant social relations. Furthermore, just as that side of the social relations
of production which we understand as 'class' can only be understood if
we grasp, from the outset, that they embody relations of social confict, so
too with the 'commodity aspect' of those same social relations of produc­
tion. Liberal theory itself recognises this, in a restricted way, for it posits
a welter of conflicts involved in the individual pursuit of self-interest and
competition. Out of this competitive struggle it then declares that there
emerges a kind of accidental harmony of interests. From 'the war of all
against all' which Hobbes proposed as the foundation of social life, the
liberal political economists, at least from Mandeville onwards, set out to
show that a common interest was nonetheless generated through the self­
ish and divided nature of social life. Private vices produced, they argued,
public benefits.
Part of the whole socialist case rests on a rejection of this position.
Indeed, in the fnal chapter of DAC, Wood herself presents an argument
which shares the classic socialist assumptions. But the theorisation is
What does constitute the fundamental 'kernel' of social relations
which Marxists take to characterise capitalism? Is it a set of class rela­
tions resting on the dispossession of the direct producers alone, or is it
something more? If we return to Marx's famous if abbreviated remarks in
Capital volume III as a starting point, we can begin to outline an answer.
Marx suggests that the innermost secret of every mode of production lies
in the form i which surplus is pumped out of the direct producers. What
is that form within capitalism? It is surely the production of 'surplus­
value'. Now, 'surplus-value' is a double-sided concept. If we focus solely
on the 'surplus' aspect, we stress the as-it-were 'vertical' or class division
within capitalist society, but at the risk of a one-sided and impoverished
conception. For, in summary terms, the 'value' aspect of the same con­
cept points us towards an equally characteristic and constitutive 'horizon­
tal' division of society, embodied within capitalism's social division of
labour, capitalism's social division of the means of production, and the
whole antagonistic nexus of relations summed up in the terms 'private
property' and 'competition'. To put the matter slightly differently, the
social relations constituting capitalism involve a dualit of fundamental
conflicts. On the one side, capitalism involves a class antagonism between
a propertyless proletariat and those who own and control the means of
production. On the other side, it involves a competitive antagonism
between all property-owners and -controllers, irrespective of the precise
form of their property. Competition and class struggle are equally foun­
dational, and it is the combination of these two social antagonisms which
Historical Materialism
provides capitalism with its characteristic 'laws of motion'.
When Wood explores the very foundations of capitalist production
itself, her eye is always on one of these aspects, and less clearly on the
other. At one point she suggests, absolutely correctly, that the relations
'between' and 'within' classes are political and juridical in character, and
not simply economic.18 But her development of the argument focuses only
on the relations between classes. What is missing is, in one sense, an
exploration of Marx's argument in Capital volume III, with respect to the
capitalist class, that it is simultaneously united against the working class
but divided against itself, forming as he puts it a band of 'hostile broth­
ers'. That observation is pregnant with all manner of possibilities for
political analysis. To that one might add that the competitive struggle
between workers, individually and sectionally, also contains important
political implications, not least for a socialist movement which seeks to
overcome divisions within working-class ranks and to constitute a politi­
caly united revolutionary force.
It is not, I should stress, that Ellen Wood is wrong about these mat­
ters. At a whole number of moments in the development of her argument
within the essays making up DAC, she makes it abundantly clear that
Marxism's fundamental stance towards the market - the most immediate
and apparent embodiment of the principles of competitive antagonism -
is one of opposition. Not only that, but she also insists that the social
relations of the market are inherently political relations. My comments
have nothing, therefore, to do with teaching grandmother to suck eggs.
Rather, my suggestion is that Wood's own analyses are incomplete unless
she also explores this whole side of the social relations of production,
extending her own analysis along the same methodological lines that she
herself outlines and develops. Indeed, were she to do so, she would fnd
that a tangle of quite fundamental - and fascinating - problems surface
to view which a Marxist political theorist must address.
What are the political and juridical presuppositions of commodity pro­
duction and of the 'market regulation' of social life? At a minimum, they
involve a set of practices and assumptions which people in other forms of
society would regard as extraordinary, and indeed deeply immoral and
offensive. It is not only that direct producers are excluded from posses­
sion of the means of production, and compelled to produce surpluses for
their exploiters and rulers under conditions of strict subordination within
production. But also, the whole world of needed things is parceled up
and divided, with property fences set about it. All society's members - of
whatever class - are systematically cut off from direct access to the very
means of satisfying their most basic needs in both production and con­
sumption alike. Passage across capitalism's constitutive property fences is
permitted only to those who can adequately address the 'selfshness' of
the property-owners by offering an adequate 'exchange'. The starting
point of the 'market' is not 'mutual exchange' but mutual exclusion.
This system of mutual exclusion, i.e. of private property, involves as
one of its inherent presuppositions a jurisprudential notion of 'right' .
'Right' is the legal correlate of the economic concept, 'value'. The dialec­
tical elaboration of 'right' into concepts of 'contract', 'law' and 'state' is
as much needed, for a systematic analysis of the foundations of capitalist
production relations, as the elaboration of the economic concepts of
Two bok b Elen Wo
'exchange', 'money' and 'capital' out of 'value' .
Furthermore, there is inherent in the social relations of commodity
production a further, political element, of 'coercion' or 'force'. For the
'privatisation' of property - itself as much a historical process as the dis­
possession of the direct producers - involves a situation in which human
need and human right stand directly opposed. The property-owner, as
Adam Smith remarked, can never sleep secure in his bed for fear that
others will invade his property, pursuing their own needs. The constitu­
tion of a system of private property requires, as a necessary correlate, the
constitution of a system of 'defence' or coercive power capable of resist­
ing ever-repeated threats to its integrity. Every expenditure of 'productive
labour' must be matched by a parallel expenditure of 'unproductive
labour' deoted to maintaining the secur
ity of property. The analysis of
the very 'base' of capitalist society, of its kernel of social productive rela­
tions, requires not simply an 'economic' theory but equally a theory of
jurisprudence, a theory of politics, and a theory of war. The study of
'competition' - a sadly neglected topic in Marxist theory - must be at
once a study of the economic, political and ideological aspects of these
social relations and their determinative power.
The previous paragraphs merely indicate a whole area of inquiry
which a Marxist theory of politics needs to encompass. A fuller consider­
ation of these matters would suggest the need to distinguish between and
further analyse different forms of competitive struggle. At a minimum, we
should distinguish within the field of 'economic competition' between, on
one hand, the division and struggles between buyers and sellers in mar­
kets and, on the other hand, competition between producers of similar
commodities. But we should also explore the distinctions between purely
'economic' and 'extra-economic' forms of competition. And considera­
tion of such matters, in turn, might cause us to re-shape some of the
ways in which we - consciously or unconsciously - 'think' the nature of
our society and the struggle to reconstruct it on new social foundations.
Marxist political thery stil needs to address more adequately than it
has done, to date, the foundations and implications of competition, and
the political consequences of 'value' relations. In this respect, Wood's
work seems to me symptomatic of a wider weakness within our own tra­
dition as a whole. The underdeveloped character of this whole area in
turn, I think, underpins the other problems I find when reading her
3. Combined and uneven development
Ellen Wood dedicates the frst of these two books to 'the British Marxist
historians', and with good reason, for their collective achievement in
reconstituting and developing the framework and content of Marxist his­
torical thinking has been considerable. But the limits of their achievement
must also be noted.
This is not the place for an overall evaluation of a complex and bril­
liant collective oeuvre, even were I competent to provide it. However, the
British Marxist historians to whom she refers shared a negative charac­
teristic. Their history was not 'Trotskyist'. I do not mean by this that
none of them ever joined parties or groups, deriving at some remove,
from the Left Opposition. Rather, one characteristic and necessary
Historcal Materialism
emphasis within Trotsky's particular contribution to Marxism tended to
be missing from their conceptual mind-set: namely, what Trotsky termed
the law of combined and uneven development. The elaboration of this
'law' gave Trotsky an unequalled capacity to theorise about the processes
of development of world capitalism, and of the class struggles arising
within it. The development and application of an understanding based on
this 'law' set the thought of Trotsky apart from the other major strands in
twentieth-century Marxist thinking, and specifically among those who
formed the post - 1 91 7 'communist' tradition. Whether they stood in the
tradition of Bukharin or of Stalin, most Marxists saw the development of
the world in terms of single countries, taken one at a time. Only on this
basis could they even contemplate asserting a doctrine like that of 'social­
ism i one country'. And while it was to the enormous political and theo­
retical credit of most of the British Marxist historians (Hill, Saville,
Thompson and others) that they came to deny the 'socialist' claims of the
Stalinist regimes in the name of the original Marxist vision of working­
class self-emancipation, I think it is less clear that they entirely overcame
the 'national' way of thinking that became predominant within most
Marxism from the mid- 1 920s. That limit to their thinking did not, of
course, prevent them from active solidarity with the struggles of the
oppressed and exploited across the globe, and not least within the USSR
and its satellites. But it did provide a conceptual limit to their thinking
about some questions to do with capitalist development.
It might be premature for me to suggest this, but I sense a develop­
mental shift in Ellen Wood's own thinking about these matters, rather on
the margins of her major concerns, but hopefully pointing in a more ade­
quate direction. There are specifc ways of formulating problems in PCC
which, frstly, strike me as unsatisfactory but, secondly, seem not to reap­
pear within DAC. They revolve around the way in which she situates what
Thompson termed 'the peculiarities of the English' and their relation to
capitalist development elsewhere.
Let me briefy identify some of these, initially in a rather pedantic way.
England, Wood states at the beginning of PCC, was 'the world's frst cap­
italist society' . I have no quarrel with the main thrust of her argument
here, indeed it is vitally important to the reconstruction of Marxist theory
that the signifcance of her key point is grasped. Yet can one treat a coun­
try as equivalent to a 'society' ? Particularly when she also refers, in pass­
ing to 'other capitalist economies' as if these were discrete entities. Why
pick up on such a quibbling point? We all use shorthand expressions for
which others can make us suffer, if they probe hard enough. Is there any
more significant problem here?
I think there may be, because of the way in which - above all in PCC ¯
Wood develops part of her argument. The matters involved are, as I have
already remarked, tangential to the main line of her argument. But they
take on an increasing signifcance if we ask how her theorisations help or
potentially block our understanding of capitalist development in the twen­
tieth century.
Part of the problem arises, perhaps, from the way issues are posed by
those she criticises. In the early 1 960s, Anderson and Nairn offered a
version of English history in which England's problem was that it wasn't
suficiently capitalist, having failed to have a proper and thorough-going
To book b Ellen Wo
bourgeois revolution. Or, in the second version of Anderson's thesis,
Britain had failed to have a second bourgeois revolution to bring itself up
to speed with the more advanced capitalsms.19 The political implication
of either version of the position seemed to be - rather extraordinarily for
a socialist - that the Left's project should be concerned with the condi­
tions for winning Britain a more up-to-date capitalism. There was a
national-capitalist conception underpinning the whole argument, and the
whole project. Discrete national 'capitalisms' were compared with each
other, and the conditions of relative advance and backwardness became
the main criterion of judgement. It's not clear that Ellen Wood, at least in
the pages of PCC, ever broke entirely clear of this way of posing ques­
Capitalism was, as Wood argues very effectively, born in the English
countryside, and its development from that originating source gave
England a head-start in world development. However, the history of
world deelopment thereafter was crucially conditioned by that advance
made in England. The impulse towards capitalst transformation was felt,
through a variety of economic, political and military mechanisms, across
the rest of the globe. Ruling classes elsewhere had to learn to adapt, to
reshape their own institutions in the direction of specifcally capitalist
development, or suffer the consequences to their own positions. One
important method of adaptation consisted in their converting existing
institutions to new purposes, and in that process transforming their inner
character away from pre-capitalist and towards capitalist patterns of
social reproduction. Within a seeming surface continuity of ruling institu­
tions and cultures, massive changes were in fact wrought.
Wood identifes exactly such a pattern of change within seeming con­
tinuity in her critique of Jonathan Clark in PCC. For Clark, the mere con­
tinuation of aristocratic dominance in culture and politics is evidence of
nothing much happening. This position involves his ignoring the enor­
mous changes wrought i those aristocrats' relations with other classes
and with the state, in fact their conversion from a 'feudal' to a 'capitalist'
aristocracy. Wood's critique of Clark on this point is very clearly and well
done. She applies a similar argument in her critique of Aan Macfarlane,
who presents a formal continuity in the existence of 'private property'
from before the medieval period, aU the way through to contemporary
times, and therefore argues that nothing signifcant altered. Like Clark on
the aristocracy, Macfarlane fails to identify massive shifs in the social
meaning and implications of 'private property', and in particular its sub­
ordination to the imperatives of competitive accumulation.
Wood's principle of critique here is, surely, absolutely correct. But it's
far from clear that, when she then turns to consider the further impact of
English capitalism on the rest of the world, she always recognises the
same essential principle of substantive transformation within formal con­
tinuity. Discussing the way that 'post-absolutist states responded to the
competitive challenge and the example of English capitalism', she sug­
gests that here 'a dynamic capitalism could develop prematurely, in
advance of fully ripe indigenous conditions and even adapting pre-capi­
talist relics to the needs of capitalist development' . 20 The idea here of
'prematurity' seems to me dubious, especially as it is tied to a notion of
'fully ripe indigenous conditions'. Once capitalism developed in England,
Historical Materialism
one might say, there never were again simply 'indigenous conditions'
which could be considered in isolation. For ' England' in a sense 'entered'
the rest of the world, and became part of all 'indigenous conditions' . And
as capitalism spread, so too the inter-penetration became more and more
developed. Within the spread of capitalism, ' combination' as well as
'unevenness' became a key principle of development to a qualitatively new
Where capitalism frst developed, the key actors shaping the transfor­
mation of social production relations were 'private' . They gave capitalism
in England its 'peculiar constitution'. Where Wood explores this ques­
tion, she seems to me to be basically correct in her arguments. But she
risks then treating the form of capitalism that developed in England as a
model of what capitalism is in general, a model of 'pure' or 'mature' capi­
talism, in which 'civil society' is clearly demarcated from and has a
supremacy over the state.
A she remarks, the nineteenth-century French state was transformed
from 'a parasitic growth fed largely by peasant-produced taxes into a cat­
alyst of economic development'. From then on, 'the processes of state­
integration and economic - that is to say capitalist - development went
hand in hand'. A major impulse to this development was British competi­
tion. The same forces can be seen at work elsewhere: 'When later in the
nineteenth century German unity was finally effected, the process was
still imbued wth a pre-capitalist logic, driven by the external pressures of
geo-political competition and war' . Here, it seems to me, we reach closer
to the nub of the argument. Were 'external pressures of geo-political
competition and war'2I expressions of an essentially 'pre-capitalist logic'?
There seems to be a decidedly normative cast to her theorising here, for
she adds: 'Just as state-centralisation was achieved by imposition from
above and in response to external impulses, so too in Germany capitalism
was driven beyond its own organic level of development by motivating
forces from without and above.'22
Let me identify the problems as I see them. 'Capitalism in Germany'
seems to be treated as if it were an organic unity, capable of possessing
something termed 'an organic level of development' . By itself, if it had
existed on the moon, production relations in Germany would not have
been rapidly transformed as they were along capitalist lines. No doubt
that is true, but Germany never did exist 'on its own', nor could its level
of development ever be considered in isolation. Because, by comparison
with its rivals, Germany was 'backward', different mechanisms and actors
played a crucial part in its capitalist development from those central in
English history. The institutional shape of German politics and economic
development in the later nineteenth century was, indubitably, different
from that obtaining in England. Later in PCC Wood returns to the ques­
tion of Germany, to suggest that, like Japan, i t represents an example of
industrialisation associated with the 'persistence of pre-capitalist forces'.
[S]ome of the most dynamic later capitalisms have developed under the aus­
pices of archaic institutions, notably in Germany and Japan. Indeed, in such
cases a 'mature' industrialisation could occur in response to imperectly capi­
talist impulses, as external competitive geo-political and economic pressures
compensated for the weaker imperatives of domestic social relations.23
Two book b Elen Wo
the continental states could promote capitalist development even while in many
respects following a pre-capitalist logic, as they were forced to compete with
the predominance of English economic power. 24
And again:
The role of the state, and indeed state enterprises, in shaping the course of
industry in Germany, and the particular character of the industrial giants (such
as Krupp) , suggest that industry was responding not only to the demands of
capitalist competition but also to different economic needs - needs closely
associated with the demands of the state and political-economic or military
interests with nothing distinctively capitalist about them. One need only consid­
er the importance of war as a motivating force in the advancement of European
industrial production.25
The issue is this: should we understand the competitive needs of states,
especially those concerned with preparations for or conduct of war, as
following a 'pre-capitalist logic'? Is the intervention into, and indeed the
direction of capitalist development by nation-states to be understood as a
sign always of backwardness and imperfect capitalism? If we adopt such
a position, we pay a potential price. For one implication might seem to be
that the world has become less rather than more capitalist in the past cen­
tury and more.26
There is, however, an alternative. That is to apply precisely the logic
that Wood brings to bear against Clark and Macfa
lane, and to look at
how, beneath a seeming continuity of institutions and practices, the sub­
stantive meaning of state control and direction has in fact been trans­
formed. We should then be driven to explore how, concretely, capitalist
development took distinct forms in different national settings, and how
often what appears to be 'archaism' in social and poltical institutions
indeed reveals on closer inspection the expansion of capitalist social rela­
tions and bourgeois dominance. A look at, say, Geoff Eley and David
Blackbour's essays on German society in the nineteenth century sug­
gests just such a pattern.27 Also, Eley and Blackbour direct us away from
a normative account of capitalsm conducted in comparative terms, where
one national capitalism is declared more 'complete' or 'pure' than anoth­
er. Using a model of world capitalist development, centred on an under­
standing of development as precisely 'combined and uneven', permits a
more fexible but realistic account of the ways in which the essential prin­
ciples of capitalism, competition and accumulation, have worked them­
selves out through historical development.
This in turn, I think, requires us to return again to re-examine the way
in which we think about some aspects of capitalist social relations, and in
particular the notion of 'competition'. Capitalist competition does not
always take an 'economic' form, but can and does equally take an 'extra­
economic' or military form in certain historical circumstances. This idea
- frst developed, I think, by Nikolai Bukharin in his attempt to explain
the capitalist roots of the First World War - has a whole series of impli­
cations for the way that we understand the fundamental social relations of
Historical Materalism
capitalism, and specifically that complex of concepts associated with
'commodity' and 'value', which I cannot develop further here.28 But
unless we grasp the nettle of reconsidering the foundations of Marxist
concepts by reference to such ideas, we open the way to just the kinds of
positions that, in other settings, Wood urges us to criticise. Unless
Marxism adopts and develops the essential ideas embodied in Trotsky's
'law', especially as applied to 'world economy', it risks being pushed back
into the position where modern society is understood as dominated not
by a single totalising logic, that of capitalist production relations, but as
dominated by more than one 'logic', each with its own specific patterns
and appropriate concepts.
Indeed, this is the path that neo-Weberian theorists like Michael Mann
and Theda Skocpol would recommend to us. For them, there is on one
side a purely 'economic' sphere of life, essentialy understood in terms of
market relations, and beside it a second and distinct 'geo-political' sphere
of relations between states, which is subject to a quite distinct logic of
inter-state competition through political, diplomatic and military means.
The two logics are separate and intrinsically disconnected.
If this argument is correct, then capitalism can be dissociated from
any especial connection with modern wars. Bukharin and Lenin were
mistaken in their efforts to explain modern imperialist war in terms of
capitalist development. And the Marxist case, that a necessary pre-requi­
site for the removal of military danger from the world is the destruction
of capitalism, is quite undermined. Ellen Wood herself is resistant to any
such notion, as she makes admirably clear in her brief discussion of the
matter in DAC. But a full development of the Marxist case requires a crit­
ical re-examination of some elements of its own conceptual apparatus
going beyond what she herself suggests.
I should perhaps be noted, in this connection, that Edward
Thompson - for whom she rightly expresses considerable admiration -
was himself prone to accept such a separation of 'logics' . In a famous
essay which he wrote as a peace campaigner, he suggested a sharp con­
ceptual distinction between the logic of capitalism and a new and differ­
ent logic arising from the Cold War, which he termed a 'logic of exter­
minism', and which he saw as utterly dissociated from any capitalist
determination. Thompson drew a political conclusion, which ran quite
counter to the implications of Wood's own brief remarks on the question,
namely, that the struggle for peace had, essentially, nothing to do with the
struggle against capitalism, and indeed that the peace struggle possessed
an urgent primacy over the class struggle.29
4. State
Interconnected with this question is another, where I sense an uncharac­
teristic tentativeness in Wood's thinking. In the course of the frst essay
in DAC, Wood discusses the history of the state form. The very existence
of the state and its performance of social functions implies, as she notes,
'a social division of labour and the appropriation by some social groups of
surplus produced by others' . It may even be, she hazards, that historically
the state was 'the first systematic means of surplus appropriation and
perhaps even the first systematic organiser of surplus production' . It is
not necessary to argue (as did Engels, for example, in Origins . . µ ) that
Two bo k b Ellen Wo
'private appropriation is a necessary pre-condition to the emergence of
' such an authority' .30 Propositions about the relations between 'class and
state' need cautious formulation, and it need not be assumed that 'class
precedes state'. What can be said is that the existence of a state has
always implied the existence of classes. With all of this, I fnd myself in
agreement, yet I sense that Ellen Wood holds back. Rather in the way that
Robert Brenner, theorising absolutism, referred to the 'class-like' nature
of the state, I suspect there is a nettle she prefers not to grasp.
I will express the matter very simply. The state is a social relation, and
the nature of that relation is itself directly a class relation. Every state, in
every social formation, appropriates surplus, directly or indirectly, from
direct producers. Just as it is essential to insist on the 'political' dimension
of seemingly economic relations, so too we need to explore the 'econom­
ic' face of apparently political phenomena like states.
Discussion of the matter is seriously underdeveloped within the
Marxist tradition, and not least within Capital. In volume III, for exam­
ple, Marx explores the different 'surface forms' taken by surplus-value as
it is distributed among those who make various claims for a share, and
the bases of that distribution. Surplus-value is distributed between differ­
ent kinds of capital as 'industrial proft' and 'commercial profit', but also
to fnanciers as 'interest' and to landlords as 'rent'. There is a notable
absence in Marx's analysis here. There is no discussion of the distribution
of surplus-value to states, i the form of 'tax', nor of the nature of the
mechanisms by which states too dip their paws in the surplus-value hon­
eypot. In much the same way that 'absolute rent', ultimately, rests on a
monopoly of land ownership, so 'absolute tax' might be hypothesised as
resting on states' monopolies of armed coercive force within given terri­
tories. There is a great deal more that might be said about this question,
but I suspect that real progress in Marxist thinking about state forms
demands a more resolute recognition of the direct 'class' character of
states, in and of themselves. Is Ellen Wood's own slightly irresolute dis­
cussion of this matter perhaps connected with her seeming disinclination
to come down on one side or another in the analysis of ' Soviet-type
states'? She marks the existence of the problem in a footnote,31 but does
not attempt to resolve it.
5. I nter-state competition and state propert
The problem is not simply one of developing adequate categories to dis­
cuss 'the state', as if it existed in the singular." For one of the characteris­
tics of the development of capitalism across the globe has been its accom­
plishment within a 'system of states', whose theorisation as yet, Marxism
has not satisfactorily accomplished.33 I can do no more here than indicate
a problem which requires theoretical development - and whose resolution
will require, I strongly suspect, an acceptance but also a frther develop­
ment of the methodological principles discussed earlier, which Wood has
herself outlined.
The whole question of states and war is important for the way we
understand capitalism. Not least because it provides additional grounds
for rejecting, as Wood herself does, an understanding of the modern
epoch in terms simply of a drive to develop the productive forces. True,
historically capitalism has developed the human forces of production to a
Historical Materialism
degree that previous generations could not even have guessed possible.
But that development has been a by-product of its real central drive,
which is the accumulation of capital. From that actual central drive
appear tendencies quite opposite to the expansion of the forces of pro­
It is not just a matter of capital quite regularly engaging in world-wide
potlatches of developed means of production through the ravages of its
own characteristic economic crises. For, frst, capitalism is also marked
historically by an unprecedented development of the forces of warlike
destruction, themselves based on the same socio-technical foundations as
'civilian' production. Second, capitalism's history is one of the direct
actual use of its means of destruction on a scale never previously paral­
leled, to smash up already constituted productive forces. I suggested ear­
lier that the practical deployment of 'unproductive labour', devoted to the
protection and expansion of existing capitalist property against rivals, is a
necessity that can and needs to be identifed at the very kernel of capital­
ist social relations. This deployment of unproductive 'defence' labour has
been quite as characteristic of capitalist development as any expansion of
'productive labour' devoted to the expanded reproduction of capital by
means of direct exploitation. Or, to put the matter slightly differently, we
would seriously misrepresent the concrete history of capitalist society
across the globe if we assumed that forms of competition based on 'plun­
der' and the like were only to be understood as 'pre-capitalist' survivals.
If the forms taken by the competitive struggle between capitals are
more than simply purely 'economic', and if the relative balance between
the different forms of competition has altered at different times in the his­
tory of world capitalism, the same must be said for the institutional forms
taken by capital itself. The assumption (present, for example in
Macfarlane) that 'private property' underpins the capitalist system is false
- or, to be more precise, false in the sense in which Macfarlane treats the
issue. For central to his treatment of private property is an assumption,
that the property relation is always between an individual human being or
a family and a specifically demarcated bundle of objects: land, goods,
equipment, etc. Yet one of the characteristics of developed capitalism, as
Marx already began to note in Capital volume III, is the dissociation of
individuals from real ownership of capital. Capital has developed in new,
de-individualised property forms, where the practical link between an
individual 'owner' and functioning capitalist property is more and more
tenuous. Forms of 'corporate' capital now predominate in the world of
'private' capitalism, the 'ownership rights' in many cases being ultimately
traceable back to workers, as in the massive investments in large compa­
nies undertaken by the fund-managers of private pension funds and
insurance trusts.
Furthermore, beside corporate capital there also grew up, beginning in
the nineteenth century but on a much extended scale in the twentieth
century, forms of direct 'state capital', in which ownership rights were
invested in state bodies and thus, purely nominally, in their citizens.
There have been, of course, large numbers of socialists and communists
in the twentieth century who assumed - quite mistakenly -that these new
forms of capital (especially state capital) represented something non-cap­
italist or even socialist, simply because individual ownership had been dis-
Two book by Ellen Wood
sociated from the practical functioning of enterprise. Both Engels and
Marx made some sharp remarks about such ideas in the 1 870s. What
many socialists and communists ignored was the elementary fact that,
regardless of formal ownership, all such enterprises were still compelled
by competition with other capitals to exploit wage-labour, to accumulate
surplus-value and to re-invest it continuously in means to expanded
Such forms gained immense significance, and posed all manner of
problems for many Marxist theorists, during the twentieth century, above
all because of their tremendous extension in the USSR under Stalin and
his successors. For here was a whole region of world society whose rulers
were driven by the imperatives of competition into some of the most sav­
age forms of exploitation of wage-labour and some of the highest levels of
accumulation, yet where the chief competitive impetus derived from the
military, inter-state sphere, where 'private property' in the wester capi­
talist sense was reduced to a purely marginal role, and where - although
a labour market functioned, albeit with massive state-imposed limitations
- the distinction between state and civil society was almost effaced. Yet
the different elements that marked its distinctiveness were all of them
found, in different degrees, within the functioning of western capitalism
in the twentieth century as well: in the war economies of the Great
Powers in both world wars, in the Nazi economy, in the nationalised sec­
tors that developed in the interwar and postwar years (although already
prefigured in nineteenth-century France and Japan alike) and in sO1e of
the labour practices of apartheid South Africa. True, the status of these
specific forms of social production within Marxist political-economic cat­
egories has been disputed for most of three quarters of a century, but it
hardly seems helpful to confne our understanding of 'capital' to forms
predominant only in the nineteenth century and to consign all these ques­
tions to something 'non-capitalist' or somehow 'pre-capitalist'.
6. The separation ofthe political and economic
If these war economies and the like represent forms of capitalism, as I
believe we have to argue, then Wood's case about the 'separation of the
economic and the political' within capitalism has to be reconsidered. It is
not, I think, that she is wrong in identifying that 'separation' with capital­
ism, for it is indeed a characteristic of capitalist social relations of pro­
duction. But then, as she also suggests, so too is the elaboration of for­
mal, representative democracy.
Yet no one would claim that representative democracy is a necessary
and universal characteristic of capitalism. Nor, I suggest, is the formal
separation of the political and the economic. Both are forms of which
capitalism is capable. Indeed one might argue, with caution, that the gen­
eral tendency of advanced capitalist society has been to extend them
under working-
lass political pressure. (The caution, of course, arises
from the fact that we know very well from the history of advanced capi­
talism that any such tendencies are capable of being counter-revolu­
tionised. Witness only the Nazi experience, or military threats to 'democ­
racy' in a whole series of states.)
I am not, therefore, persuaded that the separation of the economic
ical MateriaUsm
and the political is, as it were, 'absolutely constitutive' of capitalism as a
system of social production relations. It is not immediately, if you like, of
the 'kerel', in the way that both the dispossession of the direct producer
or the imperative of competitive accumulation is 'absolutely constitutive'
and 'of the kernel'.
This might seem to be mere semantic hair-splitting, or the product of
some sectarian spirit. But I think the issues are of some signifcance, if we
are to make progress in comprehending both the history of our own cen­
tury and indeed the meaning of socialism as the transcendence of capital­
ist society. I would express the hope that as acute a Marxist intellect as
Ellen Wood displays in these books will make a contribution to the solu­
tion of some of the problems indicated above.
1. Ellen Meiksins Wood, The Pristine Culture of Capitalism: A Historical Essay on
Old Regimes and Moder States, London: Verso, 1991; Democrac Against
Capitalism: Renewing Hstorical Materialism, Cambridge: Cambridge
University Press, 1995. These are referred to in the text hereafter as PCC and
2. Cliff 1 979.
3. See for example the ways in which Alex Callinicos 1989, or Terry Eagleton
1995, situate the whole 'postmodernist' phenomenon in academic thought.
4. Some of them were deftly anatomised in Ellen Wood 1986.
5. Lukacs 1971.
6. Korsch 1938.
7. See for example Michael Lebowitz in this issue of Historical Materialism.
8. Marx 1959, p. 791 .
9. Marx 1976, pp. 173-4.
1. Marx 1973, pp. 106- 7.
11 . Anderson 1964.
12. Wood 1991 , p. 1 30.
13. Therborn 1977.
1 4. The work of John Keane is representative of this tendency.
1 5. Luxemburg 1966.
16. For some indications of this, cf. Shandro 1995
1 7. That view is subjected to critique in Meikle 1994.
1 8. Wood 1995, p. 27.
1 9. Anderson 1987.
20. Wood 1991, p. 16, emphasis in original.
2 1. Wood 1991, p. 26
22. Wood 1991, p. 26, my emphasis.
23. Wood 1991, p. 102
24. Wood 1991, p. 104, my emphasis.
25. Wood 1991 , p. 1 05, my emphasis.
26. A version of this position has indeed been explicitly argued by Hillel Ticktin
1991 and others -and very well answered by Geoffrey Kay 1991 - in the
pages of Critique.
27. Eley and Blackbourn 1984.
28. Bukharin 1972.
Two book b Ellen Wo
29. Chris Harman 1980, 1984 developed a serious Marxist critique of Thompson's
position, and went on later to develop the outlines of a Marxist theory of war
and 'war economy', which Thompson's position badly needed.
30. Wood 1995, p. 32
3 1. Wood 1995, p. 33
32. von Braunmuhl 1978, Barker 1978.
33. Recently, there has been a welcome tendency, within what we might term 'crit­
ical international political economy', to explore the implications and possibili­
ties of a Marxist understanding of modern inter-state relations, but I am not
yet convinced that the authors I have read have yet arrived at satisfactory solu­
tions to the analytical-historical problems involved.
Anderson, Perry 1964, 'Origins of the Present Crisis', New Left Review,
Anderson, Perry 1 987, 'The Figures of Descent', New Left Review, 1 61 .
Barker, Colin 1 978, 'A Note on the Theory of Capitalist States', Capital
and Class, 4
Barker, Colin 1 978, 'The State as Capital', Interational Socialism, sec­
ond series, 1
von Braunmiihl, Claudia 1 978, 'On the Analysis of the Bourgeois Nation
State within the World Market Context', i State and Capital: A
Marxist Debate, edited by John Holloway and Sol Picciotto, London:
Edward Arnold.
Brenner, Robert 1 985, 'Agrarian Class Structure and Economic
Development in Pre-Industrial Europe' and 'The Agrarian Roots of
European Capitalism' in The Brenner Debate: Agrarian Class Structure
and Economic Development in Pre-Industrial Europe, edited by T.H.
Aston and C.H.E. Philpin, Cambridge: Cambridge University Press.
Brenner, Robert 1989, 'Bourgeois Revolution and Transition to
Capitalism', in, The First Modern Society: Essays in English Histor in
Honour of Lawrence Stone, edited by AL. Beier, David Cannadine
and James M. Rosenheim, Cambridge: Cambridge University Press.
Bukharin, Nikolai [ 1 91 6] 1 972, Imperialism and World Economy,
London: Merlin.
Callinicos, Alex 1 989, Against Postmoderism, London: Polity.
Clark, J.C.D. 1 985, English Society 1 688- 1832, Cambridge: Cambridge
University Press.
Cliff, Tony 1 979, 'The Balance of Class Forces in Recent Years',
Interational Socialism, second series, no 6.
Cohen, G.A. 1 978, Karl Mar's Theory of Hstory: A Defence, Oxford:
Oxford University Press.
Eagleton, Terry 1 995, 'Where Do Postmodernists Come From?' Monthly
Review, 47:3.
Eley, Geoff and Blackbourn, David 1 984, The Peculiarities of German
Histor: Bourgeois Society and Politics in Nineteenth Centur
rmany, Oxford: Oxford University Press.
Engels, Friedrich 1 948, Origins of the Family, Private Propert and the
State, Moscow: Progress Publishers.
Engels, Friedrich 1 959, Anti-Duhring, Moscow: Foreign Languages
Historical Materialism
Publishing House.
Fukuyama, Francis 1 989, 'The End of History?', The National Interest.
Giddens, Anthony 1 981 , A Contemporary Critique of Historical
Materialism, London: Macmillan.
Harman, Chris 1 980, 'Marxism and the missiles', Socialist Review,
Harman, Chris 1 984, Explaining the Crisis. A Marxist Re-Appraisal,
London: Bookmarks.
Kay, Geoffrey 1 991 , 'Notes on Moshe Machover's "The Nature of this
Epoch"', Critique, 23.
Keane, John 1 988, Democrac and Civil Societ, London: Verso.
Korsch, Karl 1 938, Karl Marx, London: Chapman and Hall.
Lebowitz, Michael 1 992, Beond Capital: Marx's Political Economy of the
Working Class, London: Macmillan.
Lukacs, Georg 1 971 , History and Class Consciousness: Studies in Marxist
Dialectics, London: Merlin.
Luxemburg, Rosa [ 1 899] 1 966, Social Reform or Revolution, Ceylon:
Young Socialist.
Macfarlane, Alan 1 978, The Origins of English Individualism, Oxford:
Mandeville, Bernard 1970, The Fable of the Bees, Harmondsworth:
Mann, Michael 1 986, The Sources of Social Power, volume 1 ,
Cambridge: Cambridge University Press.
Marx, Karl 1 959. Capital, vol.ume II, Moscow: Progress Publishers.
Marx, Karl 1 973, Grundrisse, London: Penguin.
Marx, Karl 1976, Capital, volume I, London: Penguin.
Marx, Karl 1 974, 'Notes on Adolph Wagner' in Texts on Method, edited
by Terrell Carver, Oxford: Oxford University Press.
Marx, Karl and Engels, Friedrich 1 964, The German Ideolog, Moscow:
Progress Publishers.
Marx, Karl and Engels, Friedrich 1 952 Manifesto of the Communist
Part, Moscow: Progress Publishers.
Meikle, Scott 1 994, 'Was Marx an economist?' in Contemporary Political
Studies, edited by Patrick Dunleavy and Jeffrey Stanyer, Swansea:
Political Studies Association, pp. 923-33.
Nairn, Tom 1 964, 'The Anatomy of the Labour Party', New Left Review,
27 and 28.
Roemer, John 1 988, Free to Lose, London.
Shanin, Teodor 1 983, Late Marx and the Russian Road: Mar and the
Peripheries of Capitalism, London: Routledge.
Shandro, Alan 1 995, ' ''Consciousness from Without": Marxism, Lenin
and the Proletariat', Science and Societ, 59. 3.
Shortall, Felton 1 995, The Incomplete Marx, Aldershot: Avebury.
Skocpol, Theda 1 979, States and Social Revolutions, Cambridge:
Cambridge University Press.
Smith, Adam 1 976, The Wealth of Nations, Oxford: Clarendon Press.
Therborn, G6ran 1 977, 'The Rule of Capital and the Rise of Democracy' ,
New Left Review, 1 03.
Thompson, E. P. 1 963, The Making of the English Working Class,
London: Gollancz.
Two bok b Ellen Wo
Thompson, E.P. 1 978, 'Folklore, Anthropology and Social History' ,
Indian Historical Review, 32.
Thompson, E.P. 1 978, The Poverty of Theor, London: Merlin.
Thompson, E.P. 1 980, 'Notes on Exterminism: The Last Stage of
Ciilisation', New Left Review, 1 21 .
Thompson, E.P. 1 991 , Customs in Common, London:- Merlin.
Ticktin, Hillel 1 991 , 'The Decline of Capitalism', Critique, 23
Trotsky, Leon 1 962, The Permanent Revolution and Results and
Prospects, London: New Park.
Trotsky, Leon 1 965, A Histor of the Russian Revolution, London:
Weber, Max 1 978, Economy and Society, Berkeley: University of
California Press.
Weber, Max 1 988, The Agrarian Sociology of Ancient Civilisations,
London: Verso.
Wood, Ellen Meiksins 1 986, The Retreat fom Class. A New True
Socialism?, London: Verso.
Wood, Ellen Meiksins 1 988, Peasant-Citizen and Slave. The Foundations
of Athenian Democrac, London: Verso.
Wood, Ellen Meiksins 1 991 , The Pistine Culture of Capitalism: A
Historical Essay on Old Regimes and Modem States, London: Verso.
Wood, Ellen Meiksins 1 995, Democrac Against Capitalism: Renewing
Historical Materialism, Cambridge: Cambridge University Press.
Postgraduate degrees in politics and sociology
by part-time or fll-time study - teaching takes place at evening seminars.
MSc Politics and Sociology
MSc Politics and Administration
MSc Public Policy and Management
MSc European Politics
PhD degrees
These postgrduate courses are recognised for the receipt ofESRC awards.
Further inforation is available from: The Depament of Politics ad Sociology,
Birkbeck College, Malet Street, London WClE 7H. Tel: (0171) 631-6780
On making-up and
breaking-up: woman and ware,
craving and corse in Walter
Benjamin's Arcades Project
Esther Leslie
The preamble of the flaneur
Walter Benjamin's writings on the Paris shopping arcades and nine­
teenth-century urban industrial culture are frequently referenced in con­
temporary examinations of 'modernity' . In current cultural studies
Benjamin's investigation of the aesthetics of merchandise and his
insights into the social fact of mass consumerism are repeatedly invoked.
Indeed these investigations may be alluded to even more frequently than
reference is made to Benjamin's once much reproduced essay 'The
Work of Art in the Age of Mechanical Reproduction' . I A decade and a
half ago Benjamin's 'Artwork' essay (1 935-9) was one of the most fre­
quently cited essays in new art history and cultural studies academic
textbooks. To put it crudely, a turnaround has occurred. In the 1 970s
academic (and non-academic) attention spotlit Benjamin's materialist
history of artistic production, distribution and reception as presented in
the 'Artwork' essay and in 'The Author as Producer' ( 1 934). The politi­
cal events of 1 968 had made Benjamin extremely readable.2 His thoughts
discharged after some years delay. Most alluring to the German 68ers
were the statements on political art and Benjamin's dissections of fas­
cism. Also entrancing were Benjamin's analyses of experience. Benjamin
wrote extensively, and from early on, about ways of expanding the con­
ceptualisation of experience: sometimes philosophically - by means of
Kant-critique, sometimes aesthetically - by a probing of surrealism and
psychoanalysis, and sometimes practically - through experiments with
hashish, which were later written up as protocols. John Berger's Ways of
Seeing ( 1 972) represented an original attempt to introduce Benjamin to
an English audience, va the appropriately mass mediation of television.
Benjamin was adopted as a leftist mascot, and a materialist who could
recommend directions for art interpretation and more importantly, cul­
tural practice. The approaches of the 1 980s and 1 990s, inflected by the
priorities of feminist and postmodernist scholarship as they have loomed
in cultural studies, art history and sociology, increasingly turned to those
aspects of Benjamin's work that appear to illuminate a burgeoning inter­
est in urbanism and consumerism.3 Interest has shifted away from cul­
tural production and critique towards consumption and characteriza­
tion. These days, Benjamin is regularly served up as one of the theorists
who can vindicate a feel-good consumerism, lending a glamourizing and
theoretical loftiness to the activity of shopping. Indeed, far from blasting
the chimeras of commodity fetishism, Benjamin becomes the commodi-
BenJamin's Arcades Poje
ty's high-priest. 4
In an essay on the place of Benjamin in the discipline of cultural
studies Angela McRobbie argues that the resurgence of interest in
Benjamin in the 1 980s was due to the fact that Benjamin offered a cri­
tique that was not formulated around the Marxist 'fetishes' of the
1 970s. She lists these 'fetishes' - the idea of the working class as an
emancipatory force, the notion of history moving inexorably towards
socialism, the belief in social progress.' To what extent this characteri­
zation of ' 70's Marxism' is accurate, and which version of Marxism is
being so described, are matters for debate (and this is not a debate that
McRobbie entertains) . 6 But the rub here is that this s ketch of
Benjamin's work i s not only inaccurate,1 it also seeks to make Benjamin
amenable to postmodernism' s affirmations of actuality. 8 Of late,
Berger's mate
ialist and pedagogic version of Benjamin has been over­
written and replaced by a consumerist, postmodernist, superficial
Benjamin. This Benjamin is peeled from his materialist body, fragile
and unsubstantial, always collapsing.9 Attempts are made to yoke
Benjamin's researches to assaults on Marxist historical materialism, and
to postmodernism's fascination with consumption. In the process, the
activity of production slips out of the frame. This is peculiar, for
Benjamin's curiosity, certainly in the last fifteen years of his life, was
directed at questions of production, from questioning determinants on
the production of art to the aesthetic demands raised by production art
to modes of production and the experience of labour.
Much of the recent Benjaminiana that foregrounds consumerism
and the cityscape fixes on the figure of the ffneur, a type who was the
epitome of fashionability in late nineteenth-century France, and is now
back in vogue. The flaneur, an ephemeral figure, has been made solid,
and is now embodied in postmodern (wo)man. The flaneur is a positive
role-model for the postmoders. This situation is a reversal of the femi­
nist condemnation of the flaneur in the 1 980s. The postmodern take
celebrates the flaneur's submission to the pleasures of consumption,
and so reverses the feminist condemnation of his patriarchal, control­
ling gaze across public space. These two mirrored positions both mis­
understand the make-up of Benjamin's flaneur, and the mainspring of
Benjamin's 3600 critique. Both bind him to currently fashionable ideas.
In this ding-dong of positionings, Benjamin is abstracted from his time,
and as a consequence, we also lose any sense of ours.
Benjamin had examined the fgure of the flaneur as he appeared in
social fact and in Charles Baudelaire's writings. Baudelaire was one of
the first to delineate the faneur, in his essay 'The Painter of Modern
Life' . Baudelaire depicts the flaneur as a quintessential modern specta­
tor. The flaneur loves to wander through the streams of urban masses,
anonymous in the throng, and observing the spectacle of modern life.
Sometimes faneurs simply observe, sometimes they record their experi­
ences in writing or in art. Flaneurs imagine that by looking they are able
to master a visual field, and maintain autonomy from it. They commend
the virtue of being cut off. Flanerie, for Baudelaire, is a kind of dallying,
an assertion of the primacy of leisure and spectating. Baudelaire aligns
the flaneur with the modern artist, whose task it is to extract the gait,
the fleeting glances, and the brisk gestures visible in modernity, and
Historical Matrialism
express these evanescent impressions, for eternity, in art. The faneuris­
tic artist chronicles the spaces of leisure of a middle class at play, but
tries to remain solitary, on the margins or lost in the crowd. This
becomes more difficult in an era of surveillance and reaction. The
flaneur is threatened by the advancement of modernization - the
increasing number of controls, an elaborate web of registrations (house
numbers, police photography, and the like) that catch traces and bring
life ever more tightly into an administrable grid. For Baudelaire, to be a
faneur is to possess a certain attitude, perhaps defined by a paradox of
an ardour for modernity - and a suspicion of modernization.
In the researches that he undertook into the Parisian arcades of the
nineteenth century, Benjamin summons up Baudelaire's depiction of
the faneur as city stroller, window-shopper and ponderer of modernity.
But he does not just give the faneur a walk-on part through the scintil­
lating verdure of modernity. He disabuses the character's illusions.
Benjamin is keenly attuned to the economic context of the faneur. He
evokes him, and fixes him historically and socially. For example, he
As fneurs, the intelligentsia came into the market-place. As they thought, to
observe it - but in reality it was already to fnd a buyer.
In actuality, claims Benjamin, flanerie outflows in journalism - the
writer as word-hack, the artist as illustrator. The musings on the crowd
and the reflections on urban seductions have to be translated into hard
cash. This was Baudelaire's disappointment: the impossibility of detach­
ment, the want of a free lunch. Flaneurs have to retail their wares by the
column inch or box. What was to be art for eternity turns into copy for
tomorrow's issue. Of the hero of modernity, the flaneur, Benjamin
writes: 'He parades the concept of purchasability itself. 11 The faneur
watches and is watched, buys into the spectacle and is the commodity
for sale, as intellectual labour-power. Economically fixing the faneur
permits Benjamin to accent an analogy between flaneurs, prostitutes
and labourers : the mind on sale matches the body for sale. For
Benjamin, it is clear that flaneurs and artists, like manual and mental
labourers too, are ensnared in the contradictions of capital. A vendors
of their efforts they are precariously poised, always subject to market
rejection. And the rebellious milieu from which they emerge and in
which they hatch their misgivings about modernization and their fan­
tasies of social and economic autonomy, is a strife-flled home. It hous­
es an array of competing individuals, outbidding each other in self­
abasements before the market. Benjamin indicates that the shelf-life of
the flaneur is limited: in time, he will have to scrape a living, and scrap
his illusions.
But, postmodernised flaneurs - and by extension postmodernised
renderings of Benjamin - are not on the way out. Postmodernised
Benjamin and his flaneuring friends are the dandies who perennially
hang out in malls, providing legitmation for theorists' relish in adopting
fops' frilly blouses, proselytizing for shopping, and raining on the
parade of those still passionate about the critique of capital. '2
Benlamln'sArtdes PJe
Sex, shopping and 'feminine' modernit
{The schema of empathy/The commodity empathizes with the customer/Virt­
uoso of this empathy: t 1l the whore/The customer empathizes with the
commodity/The empathy with the commodity is empathy with exchange
value/But that means empathy with the price/Aotheosis of this empathy: love
of the whore} "
Benjamin's fneur - a fgure derived from, but also modelled on, the
poet Charles Baudelaire - has become an object of fascination and arche­
type of selfhood in modernity and postmodernity. In her essay on
Benjamin and cultural studies, McRobbie argues that Benjamin matches a
fashionable intellectual fixation on:
the integrated experience of everyday life including the urban environment,
archit�cture, consumer culture, and the 'passage' of the individual at whatever
precise historical moment in time through these forms, whether he or she, for
example, is the flaneur of urban modernity, or the insulated walkman of post­
McRobbie emphasises Benjamin as scholar of metropolitan consumer
culture, and nominates the fHneur (or his modern-day incarnation as
walkman) as candidate for analysis and identifcation. Susan Buck-Morss
echoes this: 'In the flaneur, concretely, we recognise our own con­
sumerist mode of being-in-the-world' . ' 5 For these writers, and others, the
faneur seems to stand for the consumer in us. And it appears, at frst
glance, as if this consumer is understood truly to be the fgure who is sold
to us; leisured, master of the market, possessor of choice, pleasured by
capitalism's spectaCUlar displays. End of story.
A number of those who write now of flaneurs, consumerism, and
modernity's ocularity assert unproblematically a double-gendering, and
so mask the extent to which this fgure who strolled the public space of
the arcades had become, in the 1 980s, the focus for an argument between
feminists about the position of women in modernity, in modernism, and
in 'male' accounts of the modern. Arguments revolved around whether
the concept of the flaneur offers a subject position for women. '6 The
debates, oscillating between discussions of fctional representations and
historical personages, range across themes: the ideology or actuality of
women's position in the late nineteenth and early twentieth century; the
condemnation of a patriarchal regime of ocularity; the celebration or den­
igration of a flaneurist mode of being in the world per se. When the
debates turned their attention to Benjamin, the majority veered towards
condemnation of Benjamin before the feminist tribunal convened to judge
on moral conduct. These debates might simply be of archival interest,
except that they do leave an aftertaste, and are on occasion remobilised,
as part of the sum-total bouillon of current common-sense attitudes to
Janet Wolff's article 'The Invisible Flaneuse: Women and the
Literature of Modernity' opens with the following assertion: 'The litera­
ture of modernity describes the experience of men.'17 Wolff insists that
'the literature of modernity' details a public world that excludes women
Historical Materialism
or perceives them solely as signs of their husbands' powers of consump­
tion. This results from the fact that women are confined to the private
realm. Women, asserts Wolff, are socially denied literary modernity's
'way of seeing'. This 'way of seeing' is the voyeuristic gaze of the flaneur,
at home strolling the streets, supreme in the public realm.!8 When, in
male accounts of modernity, women step out of their private realm -
most often, notes Wolff, in the guise of whores, murder-victims and wid­
ows - they are subjected to men's lecherous looks, powerless to express
their point of view. Though sometimes visible in the public arena, the ide­
ology of appropriate spaces denies them the opportunity to formulate
their own accounts of their experience. They are ignored or spoken for.
In a later article, Wolff reiterates her claim that the female version of the
flaneur, the janeuse, does not exist in the literature of the modern, insist­
ing to boot, that, far from being a critical refection, 'Benjamin's images
collude with a patriarchal construction of modernity' . !9 The faneur, like
all men according to patriarchy-theory, is assumed to possess power.
Power, in such analysis, is a diffuse, unspecified force, possessed only by
the bad. Wolff gives no indication of how Benjamin's flaneur - as artist,
as journalist, as urban stroller - is presented as fundamentally endan­
gered through his encounter with capitalist modernity. He is presented
simply as sovereign. This is because she does not perceive the faneur as
producer and consumer, and as such subjected to the contradictions of
commodity society. Wolff overlooks the sense in which, for Benjamin, the
flaneur is a historical figure who traverses an economic space that
demands agency: the selling of wares, poetry, journalism, knowledge, in
the marketplace. A such, the faneur's subjectivity is allied not with men,
but with those who sell themselves. They are allied, but they are also in
Elizabeth Wilson takes issue with aspects of Wolff's account.2
Questioning assumptions around the apportioning of power to genders in
public and private spheres, she contends that, in the late nineteenth centu­
ry, women did participate in modernity. They participate, she argues, not
only passively as part of the spectacle of the modern, but also actively and
creatively.2! In Paris, in Baudelairean moderity, women gained certain
freedoms, though this occurs, claims Wilson, at the price of their oversex­
ualization. And, switching focus to representations, Wilson insists that, if
faneurs were, as Wolff declares, inhabitants of the streets, then modern
writers did envisage the female version: after all, the spaces of modernity
detailed by Zola, Proust, Dickens and others were frequently urban sites of
consumption populated by women.22 In art, then, as in life. In numerous
realist and naturalist novels the fi neuse appears as shopper, just like her
male counterpart. And like him, committed to making window-shopping
capitalism's favourite pastime, she too adheres to the edict: don't touch
until bought. But, if a beady-eyed presence in public space is qualifcation
enough for faneurdom, then the faneuse might also be that woman who
surveys the teeming crowds for 'tricks', and whose promise is that buying
permits touching - the prostitute. Wilson is willing to countenance the
synthesis of three principal figures in Benjamin's and Baudelaire's visions
of modern life: the poet parallels the prostitute parallels the faneur, all
engaged in a 'universal prostitution created by consumerism' . 23 This posi­
tion is also advanced by Buck-Morss in her studies of Benjamin's version
Beniamln's Ardes PJe
of modernity: prostitution is a female form of flanerie.24 Wilson rebuffs
Wolffs claim that the particular intense gaze of the flaneur is completely
unavailable to women. If the flaneur's gaze is a witnessing of the fervid but
unfulflled promises of urban pleasure, then the prostitute knows it as well
as any urban fend. A her final point, Wilson asserts that the faneur's
position was actually unstable, indicative of masculinity in crisis rather
than in control. He represents not the triumph of masculine power, but its
attenuation, to the point of invisibility.25 A such the faneur, in truth a
largely fantasy figure of an avant-garde's anxious imagination, communi­
cates a reflection on bourgeois sexual and social relations. Wilson adheres
to Lacanian psychoanalysis for her claim: the faneur is the Oedipal under
threat. The city is a castrating labyrinth that feminises all who enter it.26
To be sure, the instabiity of the flaneur's position in the late nine­
teenth century - at least as recollected in Benjamin's project - issues
from menaces that are contingent on the development of urban industrial
capitalism. His livelihood - for Benjamin's faneur gets by as a homme de
lettres or journalist, a poet of everyday life supplying the feuilleton press -
is permanently under threat. This is because of his subservient relation­
ship to the market, an anarchic and choosy mechanism. His life-style, in
as much as it relies on an intimacy with the cityscape, is further threat­
ened by the continuation of processes of urbanisation. Though spawned
by the assemblage of metropolitan modernity, modernity's processes
threaten to annihilate him. Haussmannisation, a modernization project
inaugurated by Emperor Napoleon III, was the name for the construction
of vast boulevards designed to confound barricade-building. I aimed to
flush out the hidden haunts of low-life where bohemia had once gathered
and barricaded themselves in. The modernizing city turns visibility back
on the strollers, making them subject to surveillance. It takes from them
the possibility of curiosity about the urban environment. The flaneur
becomes a suspect, and is subjected to authority's scrutiny, more than he
is able to use the power of his gaze to fathom modernity. (The prostitute
too would know those probing legal, medical, official looks.) Another
menace to the flaneur's lifestyle comes some time later in the guise of
Taylorism, with its factory system of mass production demanding univer­
sal speed-up and standardisation in all areas of life.21 Benjamin notes how
'Taylor popularised the watchword "Down with dawdling!", and this slo­
gan becomes part of a general cultural war against lassitude.'28 The
flaneur's sluggish pace comes to represent a protest against speed-ups
and the 'division of labour which makes people into specialists' . 29
Benjamin insists that the faneur possesses the artful Marxist knowledge
that value stems from socially necessary labour time - and that is why he
drags his heels.30 In as much as the production system of Taylorism wins
out, the flaneur is on the losing side of a class struggle over pace of life
and autonomy of action.3I Sadness accumulates around the flaneur,
because the social environment no longer supports the fantasy-life that
this figure desires. Benjamin observes:
With the £aneur, it could be said, the idler returns, just like the conversation
partner that Socrates picked up at the Athenian market. It's just there is no
Socrates any more and so no-one speaks to him. And the slave labour that
guaranteed his idleness has ceased too."
Historical Materialism
Benjamin notes that the gaze of the flaneur, the allegorist's gaze 'which
falls upon the city is rather the gaze of alienated man'. n This faneur leads
a way of life that is still just about able to light up a conciliatory gleam on
the approaching desolation of the city dweller.
Benjamin describes the flaneur's position in doubled terms: he is
caught between visibili
y and invisibility. On the one hand, the flaneur is
overly vsible, exposed, 'the very defnition of a suspect' and so subjected
to the mechanics of power and law." On the other, he is 'completely
undiscoverable, concealed', overlooked and ignored.35 He has to be seen
to sell his wares, and yet the battle to grab customers' attention is intense.
Strewn through the Arcades Project and studies of Baudelaire are subject
and subjected positions for the flaneur, bound up with social organization
and control and with the market and the process of commodity produc­
tion. Benjamin details the flaneur's point of view. The flaneur is an
observer - of the market.36 To that extent he is a consumer - as Wolff
and Buck-Morss suggest, for sometimes he decides to buy.37 But, impor­
tantly, he is also a supplier for the market. 38 The figure of the faneur
demonstrates, in his very being, that consumption and production are
two unseverable parts of a whole process.
The flaneur is a vision of every-person who sells his or her self, for he
is implicated in all facets of the universality of exchange in commodity
society. The flaneur, like the worker, is subjected to the penetration of
social relations by the market. The crisis for the faneur, as for everyone
who sells his or her labour power, is the marketing of his individuality. In
capitalism all workers sell their labour power as commodity. Submitting
to the process of commodity fetishism in a Marxist sense, each person is
transformed into a thing. Benjamin notes that:
Marx speaks of the fetish character of commodities. 'This fetish character of
the commodity world arises from the peculiar social character of labour that
produces commodities ... It is only the specifc social relationship of people that
assumes for them here the phantasmagoric form of a relation between things.'"
The animated allure of commodities is matched by the reifcation of rela­
tions between people. The insight that social relations in capitalism com­
pel people to regard each other as objects and to see things as spirited
spans Benjamin's critique of commodity capitalism.
Benjamin's Arcades Project is made of snippets from various commen­
tators' work. By investigating Baudelaire's oeuvre, scanning Marx's eco­
nomic analyses, and trawling through masses of archive material docu­
menting the technological developments of the mid- to late nineteenth
century, Benjamin initiated a materialist study of moderity. But it never
got beyond the fragments. It is impossible to know how Benjamin might
have worked through the masses of material for a finished book.
However, interpolated in the text are commentaries on and analyses of
the material gathered, and he was able to work elements of it into various
studies of Baudelaire, as well as other essays. A leitmotif of the project is
the interrogation of the commodity form. Various sections of the Arcades
Project - especially the sections titled 'Fashion', 'Marx', 'Arcades, Novelty
Shops, Banner (makers) ' , ' Methods of Display, Advertisement,
Grandville', 'The Collector', ' Stock Market, Economic History', The
Bnjamin's Armde PJe
Flaneur', 'Baudelaire' and 'Prostitution, Game' - theorise the commodity
and commodity production. The arcades are shown to provide a gateway
to a commodity-hell. Central to the study is a probing of the subjectivity
of those implicated in commodity relations. While, for Benjamin, the
faneur represents one instance of the person as consumer and commodi­
ty, the prostitute appears as another, possibly more significant, victim of
commodity culture - a commodity to be bought and a consumer of its
trappings - as well as its practical critic.
In debates on women, flanerie and the public sphere, feminist critics
note Benjamin's fascination with prostitution - and some bemoan the fact
that Benjamin was too caught up in his own desire for the buyable
woman to evaluate the prostitute's subjectivity critically. Wolff and Buck­
Morss complain that the prostitute is reduced to a sign:o Referring to
Benjamin's comment that, unlike Brecht, Baudelaire never wrote a poem
about a whore from the whore's point of view,4l Buck-Morss reports that:
he cites Baudelaire, who addresses prostitutes in his poems while the women
themselves remain mute: 'Baudelaire never wrote ... about whores from the
whore's viewpoint' writes Benjamin, and proceeds to do likewise."
Wilson agrees that 'Benjamin as much as Baudelaire objectifes the pros­
titute'. " Ad Wolff reiterates the point, asserting that, though Benjamin
recognised Baudelaire's failure to speak from the perspective of the
whore, for him too 'the prostitute stands for a male-defned set of possi­
biities and sexual and (economic) meanings'." McRobbie concurs:
Benjamin's relationship to the world of prostitution might be seen, after twenty
years of feminist debate and scholarship, as no different from that of any of his
middle-class counterparts of the period. The prostitute remains a shadowy,
anonymous fgure in his writing. A in the writing of Baudelaire [ µµµJ she is, at
most, a fellow deviant, another outsider. Benjamin notes that she is a commod­
ity, a 'mass article' but he takes this no further, preferring to remember the
more pleasurable Jnction of the prostitute and the even greater excitement of
seeking her out. 4
It is worth returning to Benjamin's panorama of the modern to see, in con­
trast to McRobbie's representation, how pivotal the fgure of the prostitute
is in Benjamin's critique of capitalist moderity, as also are two more
female figures: the fashionable mannequin and the female factory worker.
Benjamin does attempt to decode the subjective experience of women in
modernity, but he does this within social and historical limits, insisting that
subjects inhabiting capitalist modernity have become objects. They are
objectified, subjected to processes of commodifcation (of their labour­
power, of the culture they consume). Objectifcation occurs in the social
world, prior to representation of that social world. In the section on com­
modity fetishism in Capital Marx indicates that, in capitalism, the labour of
the individual is realised as an act of exchange between products, and so,
for the producers, the relations connecting the labour of one producer with
that of the rest appear, not as direct social relations between individuals at
work, but as they really are, as thing-like or material relations between per­
sons and social relations between things.46 Note Marx's insistence that the
Historical Materialism
fetishised experience of social relations is not incorrect, not an ideological
delusion, but is a correct experience of a defective actuality. Circumstances
appear to people 'as they really are'. Marxist criticism soberly faces the real
conditions of life. It acknowledges the existence of reifed relations and,
then, sets out to explain and overcome them. If the prostitute is a thing for
Benjamin, it is because she is a thing for capital. The dehumanising effects
of capital's organisation act precisely to turn workers into things.
Benjamin's investigation of thingliness invokes a double-action: it attempts
to explain and overcome it.
Consumption has active and passive connotations, designating con­
sumer and consumed, and it is women who - in actuality and in
Benjamin's allegorical system - smash against the full force of con­
sumption's duality: women as consumers - especially of fashion, and
women as consumed, as whores, as wives, as workers. The focus on
women in Benjamin's studies of capitalist modernity is extraordinary.
The female body is laid out as the landscape of hell and Benjamin pro­
vides a guided tour:7 Hell is a name for capitalist modernity. Women,
their prostration and objectification, saturates the critique. To focus on
Benjamin's interest in the deathly and negative aspects of socially and
historically consummated femaleness is to reiterate the often sub­
merged fact that Benjamin's studies of commodity-producing society
form a critique of capitalism and not a celebratory description.
Benjamin registers a historical shift in gender, sexual and social rela­
tions. The twin issues of gender in crisis and sexuality under re-evalu­
ation undergird many of Benjamin's conceptual moves. The challeng­
ing aspect of Benjamin's gesture lies in his construction of female sex­
uality as potentially politically disruptive: that is transgressive and
modernist. He selects the prostitute and the mannequin as model fig­
ures, emissaries of a whole system of exploitation, reification, alien­
ation. Like the flaneur, and yet more socialised, these women stand in
for every person in commodity-producing society. They signify our
vestigial natures.
Diamonds ar a gir's best fiend: women and fetish
Benjamin's shattered modernist historiography in the Arcades Project
investigates a metro-modernity. This modernity resonates with the stories
of the 'heroic' voiceless who nestle in its new zones of commerce.48
Benjamin hopes to make audible the fragmented speech of the un-dead
heroes and heroines of the tormented modern:' Mirrored in the endless
refections of shop windows, the crowd, claims Benjamin, transforms into
a spectacle. It sees itself walking and buying. Efforts ar
made to tame
and train the nineteenth-century mob, to turn it into a consumer crowd
that forgets its role in production. Acts of production are obscured, con­
cealed, in the commodities' cocky sovereignty. Mass society's visible body
is formed by the swelling ranks of customers, audiences and located at
the sites of product and culture consumption. 50 Particularly visible here
are ranks of women. Capitalist modernity, Benjamin argues, conjures up
a market of female types, spectacularly on view in the shopping arcades.
They are frst glimpsed inside the labyrinthine storehouses of products.
Benjamin refers to this array of types:
BenJamin's Ades Poje
The female fauna of the arcades: whores, grisettes, old witch-like sellers, junk­
shop women, glove-makers, demoiselles - the last one was the name given to
arsonists who dressed as women around 1830."
These female types include sex-objects, sex-sellers, commodity retailers
and a feminine moment of political action (albeit transvestite) . These;
together with numerous other women who populate Benjamin's phantas­
magoria of nineteenth-century Paris - widows, sex-murder victims,
fashion mannequins, lesbians, female factory workers - are specimens
who signify conventional femininities in crises. They are examples of
non-domesticated feminine bodies, with an excess, an absence or a 'rev­
olutionising' of sexuality and sexual trappings. The visibility of female
types dovetails with a certain commotion in the delimiting of gender
roles. Their dificult relationships to traditional sexuality coincide with a
generalised crisis of social relations - unleashed not least by capitalism's
revolutionising of the modes of production and consumption.'2
In various accounts of nineteenth-century and early twentieth-century
modernity the prostitute features as a threat to family stability.'3 The
attack on family values was not the prerogative of the prostitute alone.
Benjamin's archival burrowings register a cult of gender ambiguity and a
feminist critique of sexual relations. Both set in turmoil traditional female
and male roles. He collects quotations by and about the Utopian
Socialists, with their claims to androgyny, and he also records the bisexu­
alty of a sectarian tendency that he names 'anthropological material­
ism'.'4 This interest in sexual ambiguity is taken by Benjamin as a by­
product of a more extensive questioning of the construction of sexual
roles in industrial capitalism. He observes a change in sexual roles and
relations wrought by economic reorganisation. And he approves the
Utopian socialist critique of traditional sexual roles, in as much as they do
still exist. Chronicling this tension preoccupies him. The new status of
metropolitan women, subjected by work and urbanisation to industrial
violence and sexual mass uniformity, throws much sedimented custom
into chaos. Cultural, legal and politically policed divisions of gender dis­
integrate. Factory workers, particularly, are seen to smack up against
political culture - a process that is termed by Benjamin, and Baudelaire
before him, a 'masculinization of the woman'." Benjamin asserts:
The nineteenth century began to use women without reservation in the produc­
tion process outside the home. It did so primarily in a primitive fashion by
putting them in factories. Consequently in the course of time, masculine traits
were bound to manifest themselves in these women. These were caused pri­
marily by disfiguring factory work. Higher forms of production, as well as the
political struggle as such, were able to promote masculine features of a more
refined nature. The movement of the Vesuviennes can perhaps be understood
in such a way. It supplied the February Revolution with a corps composed of
women. 'We call ourselves Vesuviennes,' it says i the statutes, 'to indicate that
a revolutionary volcano is at work in every woman who belongs to our group'. 56
Where Wilson perceives the metropolis as a maze that feminises all who
enter it, Benjamin's proposes rather that the city, or better, industrial cap­
italism, actually masculinises those who come into contact with it. The
Historical Materialism
spaces opened up by mechanical labour invte or demand a new access of
women to modern life. And this access may result in political organisa­
tion, a revolutionary movement that further tramples the gender barriers
that have already been made shaky by social change. However out of step
with current progressive vocabulary, this depiction recognises a shift in
. the actuality of women's roles. This shift is, of course, classed. Engels
speaks similarly, using the phrase 'unsexed' to describe women workers,
in order to point to the historically changing form and ideological basis of
femininity and masculinity. Such observations about the mutability of
gender concur with Marx's assessment of the social-sexual changes ush­
ered in by capitalist modes of production. Marx reminds the reader of
Capital of the revolutionary potential interred within capitalism:
However terrible and disgusting the dissolution, under the capitalist system, of
the old family ties may appear, nevertheless, modern industry, by assigning as it
does an important part in the process of production, outside the domestic
sphere, to women, to young persons, and to children of both sexes, creates a
new economical foundation for a higher form of the family and of relations
between the sexes. 57
In Benjamin's Arcades Project, as in Marx, women are shown to be inject­
ed into commodity relations. The effects of such an insertion are multi­
ple, providing the basis for a revolutionary change in family relations, as
well as an involvement in an absolute and deathly exploitation. This is the
contradictory, at once revolutionary-futurist and reactionary-restorative
drive of capitalism. In Benjamin's work this double-movement is mani­
fested in the various positionings now available for the female - amongst
others, the whore, who sells her body, the worker who sells her
body/mind as labour power, and the mannequin, a super-consumer who
models on her body the constricting grip of the commodity. In a sense,
then, this becoming ungendered or thing-like is the admittance into
exchange, which is entry into equality, the strange equality of all who
stand before the labour market.
Far from being invisible, women - in the guise of prostitutes, workers,
and mannequins - are in the spot-light, mainlined, through their com­
modity-entwinements, into public zones, and into the theory and litera­
ture of modernity. The metropolis is a stage set for the new woman.
Benjamin registers that the illuminated shopfronts are eyes.58 For females
even the eyes of passers-by are hanging mirrors.59 Such visibility they
themselves cannot overlook. Commenting on Parisian street style,
Benjamin remarks that:
Women see themselves here more than elsewhere. The peculiar beauty of
Parisiennes stems from these mirrors. Before a man catches a glimpse of them,
they have already seen themselves refected ten times."O
In city streets women know they are surveilled, subjected to looks. Even
their own eyes peruse and judge their own appearance. They are objects
for themselves. When they catch a glimpse of themselves in mirrors, they
sometimes confuse themselves with the hard bodies of shop-window
mannequins. The mannequins' rigid but exquisite forms fuse with the
BnJamin's ArCe Pje
dream-egos of women, providing perfect, stiff role-models. And 'fashion
draws them further into the material world' of the inorganic-anatural.61
Fashion 'prostitutes the living body to the inorganic world', making
women pioneer-explorers in a new continent of artifice.62 This new conti­
nent is commodity-country, a historical, manufactured zone, far away
from the world of nature and the legitimating narratives that discourses
of nature make available.
Women, in Benjamin's analysis, whether prostitutes or fashion-vic­
tims, are related as absolute commodity and ultimate consumer. Always
on view, like wares, they view themselves on view, like buyers snared in a
circuit of self-consumption. Women and commodities, sister-seductress­
es, are seen to fabricate an appearance, in an attempt to mark them off
from other goods on offer. Both women and commodities are objects to
be looked at but not touched, until 'picked up' or bought. Benjamin
underlines the commodifcation of the female, by noting her involvement,
via fashion, in capitalism's endless production of accoutrements. The
fashionable woman is masked by cosmetics and clothes, emanations of
non-natural 'feminine fauna'. She is a consumer. But she is also to be
consumed. Recording the development of the woman as focal point in
commercial appeals, Benjamin argues that: 'modern advertising demon­
strates how much the enticements of women and commodities have
merged'.63 Benjamin records how, in capitalism, an economic system of
universalised exchange is socially broadcast using women as objects and
enticing them as complicit subjects.
Benjamin maps out a theory of fashion as the 'dialectical switching­
station' between 'Weib und Ware' (woman and commodity) .64 Prostitutes
rely on fashion, the commodity appeal of the revealing dress, to sell their
body as thing. The prostitute is an extreme example of a (mass-) con­
sumed woman. She is nonetheless a consumer of the trappings of cos­
metics and fashion to heighten her appeal or signal her status. In the
nineteenth century she slinked around, for convenience sake, in the
arcades - a natural habitat of the commodity too. The arcade is described
by Benjamin as ' a horny trading passageway, designed to arouse
desires'.65 That is why, he comments, it is not surprising that prostitutes
lurked in their shadows, before Haussmannisation shoved them out onto
city streets. Benjamin recognises an erotics of capitalism, that is, that cap­
italism is a system that banks on the production and maintenance of
desire. Benjamin is convinced that capitalism and commerce produce an
erotic charge, peculiarly aligned with the modern organization of sexuali­
Benjamin's Arcades Project and Paris investigations tender splinters of
a history of love in industrial capitalism. The city whore enstages the
coordinates of a new lust, an 'erotology of the damned' . 66 Benjamin notes
that, in high capitalism, earning money becomes a mad passion and love
a financial affair. Capitalism's mesh of want eroticises money - that
already dead, congealed residue of productive labour. The erotic ideal i
not the grisette who gives herself to men but the lorette who sells herself
to men.67 Fixation on the prostitute in Benjamin's writings cannot be
detached from his wider interest in criticizing and demolishing commodi­
ty-capitalism. Lust for a prostitute, claims Benjamin, is the apotheosis of
empathy with the commodity on the part of the man.68 Cash itself is
Historical Materialism
invested with an erotic charge. He writes:
Purchasability can itself become part of the sexual attraction. This attraction
increases where the plentiful supply of women underscores their commodity
character. Later on, the revue, with its girls in strict uniformity, introduced this
aspect into the sexual drive of the city-dweller."
Mass society's forms of entertainment are involved in the commodifca­
tion of desire. Economic fantasies are tansmuted to the erotic sphere,
reforged as the love of another who has been touched by the magic wand
of monetary value. The desire for multiples becomes part of the sexual
drive of the city inhabitant. This label - prostitution - is a name for an
approach to love in a mass age where everything and everyone is involved
in commodity relations and where the process of universal commodity
exchange converts all difference into identity. Sexuality is transformed by
commodifcation,1° Prostitution is, for Benjamin, not just a by-product of
city culture, it is the way of life for all, and moreover, it is a strategy for
survivng in city culture. That is to say, it recognises affairs 'as they are' . 71
Benjamin writes:
Prostitution seems to contain the possibility of surviving in a living space in
which the objects of our most intimate use are increasingly mass articles. In the
prostitution of the large cities the woman herself becomes a mass article.72
Under the domination of commodity fetishism, relations between social
subjects are masked. The other is objectifed, as is the self. A relationship
between thigs takes place:
It is said that 'money makes sensual' and this formula offers a rough guide to a
situation that extends far beyond prostitution. Under the domination of com­
modity fetishism the sex-appeal of each woman is tinged to a greater or lesser
degree with the appeal of the commodity. Not in vain has the relationship of
the pimp to his woman as a 'thing' to be sold on the market massively excited
the sexual fantasies of the bourgeoisie.73
The client, the private man, represses knowledge of material and human
relations in order to avoid confronting contradictions between his class
ideology and his activity. He clings to the object. He empathises with it.
Such a relationship typifes social relations in the moder. A such it is
significant. The woman is commodifed, as relationships revolve around
the fetish of money - who has it, who's worth it, who costs and how
much. Love relationships appear to mirror social relationships in the
workplace in which people regard each other as things.74 This is why, for
Benjamin, the lesbian may be modernity's heroine, her life unfolding as a
protest against social order, the family and the role of motherhood. Hers
is the 'heroic' attempt to escape commodifcation and to avoid slotting
into the compulsive logic of heterosexual attraction, that appears to
underpin commodity culture. 75 Other women are shown fending off
nature's role in ideological justifications of oppression. Fashion,
Benjamin insists, while locking the mannequin into the endlessness of
modish consumption, might also provide some sort of escape from the
Bn!amln'sAde P/et
despotism of the natural. For example, the lorette, notices Benjamin,
dresses masculinely, with cravats and suits and monocles, and also in
over-bright, screeching colours. '6 These are taken to be gestures of a
utopian attempt to escape the tyranny of the natural and the burden of
the biological.77 In that re
pect, Benjamin assumes that they are i soli­
darity with the nineteenth-century feminist Claire Demar, whose mani­
festo he quotes:
In the concluding section she writes: 'No more motherhood! No law of the
blood. I say: no more motherhood. Once a woman has been freed from men
who pay her the price of her body . . . she will owe her existence only to her own
creativity. To this end she must devote herself to a work and fulfil a function . . . .
So you wl have to decide to take a newborn child from the breast of its natural
mother and place it in the hands of a social mother, a nurse employed by the
state. In this way the child wlbe raised better . . . . Only then and not earlier will
men, women, and children b freed from the law of blood, the law of
mankind's self-exploitation.' 7
Benjamin is enticed by these moments where women, conventionally con­
signed to the realm of nature, trash the connection and lead an assault on
nature's laws. He takes it so far as to assume that all women in capitalism
participate in an attack on nature, via their intimacy with fashion and
make-up. This attack on nature is enabled by capitalist commodifcation
and its particular production of social-historical relations between people.
Benjamin is not caught up in a romantic nostalgia for a lost naturalness,
though he may be substituting his articulacy for a subjectivity commonly
denied voice.79 His aim in presenting the sociological fact of objectifica­
tion of women is to validate, out of the wreckage, the explicit shif of
women into the realm of history and culture, recognising the enormity of
its social and political implications. It is the revolutionary chance for sal­
vation. Benjamin is not a moralist, providing positive images, but his
images of negativity are dynamite.
The Arcades Project is rarely seen as forming a commentary on his
contemporary society, despite the fact that references to his own epoch
are scattered through the work. It can be argued that Benjamin's promo­
tion of an idea of sexuality as artefact specifically criticises fascism's
patronage of a swindle of the natural: biology as destiny, nature as fIity.
The revolutionary function of prostitution, argues Benjamin, is to release
erotic life from the chains of biological necessity, the 'odious laws of
nature' to which love submits.80 In some sense capitalist modernity's
smudging of the boundaries between the sexes is responded to by
nazism.8' Nazism propels an immense rearguard action to re-defne sexu­
al boundaries - with a focused attack on sexual liberational politics - as
part of an attempt to clamp down a new social order. The de-auraticised,
technological sexuality that Benjamin ascribes to modern women is part
of a political-critical attempt to force adjustment of romanticised percep­
tions of women. Women are no longer wedded to a place inside the pri­
vate and natural, but enter the realm of the social and historical.
In Benjamin's accounts, women are presented as potential purveyors
of the inorganic and taunters of death.82 Women lock onto the fetshistic,
the hard, unproductive and dead. Fashion's fibres and gemstones are
Historical Matrialism
dead. Women display the congealed substance of the commodity. The
female body is dissected, turned into a landscape (a place where aura had
once been experienced), cultivated in an imitation of bleak, inert nature.
This nature is no longer nature, but historical because commodified.83
The mass-produced mannequin-woman has sloughed off auratic feminin­
ity.84 She underlines her commodity-character, her thorough socalisation
and historicisation, through cosmetics. No-one does this more than the
whores who wear make-up like masks. There is no authentic self, just
presentations of selves for the needs of the job. Benjamin writes:
The form that prostitution has taken on in the large cities lets the woman
appear not only as a commodity, but in its most precise sense a a mass article.
This is indicated by the disguise of the individual expression, in order to
achieve a professional look, as is make-up's task. Later on this is underlined by
the uniform girls in the revues.8
Sex for money, consumer-sex, while indicative of the exchange-fixated
relations of commodity fetishism, is also sex denuded and disruptive of
the ideologically abused mythology of individualised love harmony.
Benjamin's descriptions of women and female effects conjures up the
idea of anti-nature. Fashion 'invents an artificial humanity'.86 The body,
clothed and unclothed, mimics the inorganic world. Benjamin develops a
dissectional aesthetic, using such diverse sources as Baroque poetry,
Baudelaire's writing and fashion discourse. The female body turs into an
ornament. The fragentation of female beauty into its most noteworthy
parts is like an autopsy. In such fetishistic fragmentation body parts are
popularly compared with alabaster, snow, jewels, minerals or other inor­
ganic forms.·7 While such a vision of femininity is transgressively mod­
ernist, because it refuses the fable of nature, it is also true, a chilling re­
presentation of social actuality, of relations in the modern age. Such
deadly inorganicism impersonates a more generalised social actuality,
wherein the body is a machine for work, the machine-pendant such as is
described by Marx.8•
Benjamin's woman is ruined, her fashionable body tabulating the cor­
rosion of human relations, her worn-out, but ceaselessly performing,
body exhibiting the consumption of mortal energy. This woman-con­
struct is a casualty of capitalist moderity, a sacrifce even. Murderous
traces mark themselves on the commodity-body. This commodity-body is
inorganic, consumed and humanly devalued. Her 'permanent efforts at
beauty' are like a never-ceasing punishment meted out in hell: a
Sisyphusean labour.89 For the man too, work has invaded the space of the
erotic. The 'love at last sight' spasm that grasps the lover on the street as
he spies a desirable woman is not a convulsion initiated by the raptures of
love.90 It is not sexual pleasure that charms him but a laborious obsession
with futureless reiteration, induced by both the massifcation of bodies on
the street and the endless, fruitless labour of mass reproduction.
Benjamin cites Baudelaire's description of lovers 'exhausted by their
labours' .91
Benjamin correlates death and decay with consumerism and entertain­
ment. For example, he mentions the catacombs on the Place Denfert
Rchereau, where he rented an apartment. With their skulls and bones
BnJamin's AI de Pjet
tossed out of their graves and arranged in geometric shapes, for a touris­
tic confrontation with millions of the dead, they represent an ironic com­
mentary on the delights of the arcades.92 Death and life, sex and death,
pleasure and pain forced together illuminate the deepest, most secret
empathies in the mock-intimacy, enforced by capitalism's coining of
Through his depictions of social relationships in the metropolitan
space, Benjamin illustrates Lukacs' idea of reification, the process of
becoming thing-like.93 This reifcation might be aligned with deadliness.
The prostitute is permeated by deadliness. Her whorish eyes do not see,
but instead display 'the look in which the magic of distance is extin­
guished'.94 Benjamin entwines 'Lust' and 'Leiche', desire and dead body.95
This fusion of sex and death, is, of course, a romantic trope, but for the
romantics too such fatal desire was historically precipitated, connected to
the industrial onslaught on nature. But Benjamin is not mournful. Better
to recognise the current basis of social relations, than to perpetrate the
cosy lie of humanism.
Benjamin's Freudian interests and Marxist concerns lead him to coag­
ulate sexual fetishism and commodity fetishism. The woman as commod­
ity fetish also embodies a sexual fetish. Benjamin conceives the sexual
fetish as that which is hardened, glossy, glamorous and a fatality.
Fetihism, which succumbs to the sex-appeal of the inorganic, is its vital nerve,
and the cult of the commodity recruits this to its service.96
Capitalism's lethal logic takes eroticism to the limit: the sadistic, fetishis­
tic eroticism of the fragmented body is socially and historically located.
Sadism and fetishism are intimately related. Both emanate from the store­
house of capital's deadly fantasies.
Sadism and fetishism are laid down in the fantasies that want to make all
organic life into a fund of the inorganic!'
The turning of organic life into the inorganic is intrinsic to capitalism's
mechanisms. It is apparent in the fgure of the commodity, begot of the
transmutation of living labour into congealed labour. Fetishism (sexual,
commodity) is connected to a certain deadliness, or, at least, to a cross­
ing of the border between death and life. The certainty of what is dead,
what is living, is confounded by fetishism's mechanisms. Sex, like the
commodity, straddles two realms - the realms of the living and of the
dead. It mirrors too the actuality of capital, the rule of abstract labour,
that is dead labour, over living labour. In fetishism sexuality dismantles
the barriers between the organic and the inorganic worlds. Clothing and
jewellery are its allies in this. The landscapes of the body traverse paths
that lead sexuality into the realm of the inorganic. And fashion itself is
just another means to tempt it to travel even more deeply into the materi­
al world.98 Fashion affirms the rights of the corpse over the living. Much
of Benjamin's discussion in the section devoted to fashion in the Arcades
Project concentrates on the connections between fashion and death. The
textile industry is central to Benjamin's 'poetics of Marxism'. Benjamin is
sure that the alterations in fashion beat out incessantly the tempo of
Historical Materialism
moderity. He writes: 'Fashion prescribed the ritual by which the fetish
Commodity wished to be worshipped' . It shows too how much the new is
indeed the eteral return of the ever-same.99 Fashion is, by nature, con­
demned to death. Marx, in Capital, provided a materialist core for
Benjamin's idea of fashion as intimate with death. Marx details how 'the
murderous, meaningless caprices of fashion' are linked to the anarchy of
production, where demand cannot be predicted and where gluts lead to
starvation. IOO The textile industry was central to the formation of the fac­
tory system of exploitation, in Marx's account. It was in the cotton mills
that women and children were employed en masse, 'cheaply' and
mechanically spinning materials harvested by growing numbers of slaves,
born to work and worked to death, in the U. S. slave states.'O' The con­
nections between products and death alert us to the fact that every thing
consumed has been produced under conditions that occasioned suffering.
This indicates one source for Benjamin's famous phrase: there is no doc­
ument of civilization that is not at the same time a record of barbarism. 102
Contradiction and critique
The construction of the woman-fgure as deadly, through her intimacy
wth fashion, cosmetics and artifice, unearths a meaning embedded in the
word consumption. Consumption is destruction, a wasting away, a
devouring. This meaning is obvious in the English medical term con­
sumption. Consumption eats up the fesh of the 'natural' female body.
Her body is attacked by this wasting-disease as she is drawn and draws
others into death; that is into relationships between things. And so she
mirrors the system's truth. Desire functions through the client-lover's
self-objectification. Fetishistically, the client identifies with the woman­
as-thing, with exchange-value. She also displays the system's potentially
revolutionary drive - to blast away nature and enter into the beginning of
human history.
The whore is a 'destructive character'. The destructive character, a
fgure drawn up by Benjamin in 1931 , is a type opposed to repression in
its political and psychic senses. The destructive character, motivated by
the wish to forestall nature, eliminates the traces of the age, cheerily, for
everything cleared away means to the destroyer a complete eradication of
her own condition.,03 The whore plays out the logic of the capitalist sys­
tem. This system is immanently destructive. She sees it through to its bit­
terest end. She exposes. She just shows, repeating the methodology of
the Arcades Project itself. Benjamin described the project in the following
Method of this work: literary montage. I have nothing to say. Only to show. I
wn purloin nothing of vaue and will appropriate no ingenious formulations.
But the refuse, the trash: I do not want to make an inventory of them, but to let
them gain their rights in the only way possible: by putting them to use. 10
In putting the whore to use Benjamin does transform her. The prostitute
is 'elevated' into theory. Benjamin turns her into allegory - allegory is a
tool for making meaning of matter. 105 Matter is transfgured by theoretical
activity. This transmutation of historical material is not part of a romanti­
cism that simply imposes on the actuality of the matter, but an attempt to
Bnjamin'sArdl P/e
enact critique by considering the objective position of the walking
wounded. Such activity is redemptive, because, in presenting and theoris­
ing, it makes graphic the crises inherent in the relationship between
objects in the world and the revolutionary possibilities ushered in by those
crises. In the case of the prostitute, she tenders an image of brute social
reality. The prostitute provides a charged instance of the real that thought
snatches in order to expose graphically, via its contemplation, the social
relations that haunt reality. Benjamin chronicles the crisis of social-sexual
ordering through examinations of the prostitutes' insurrection against
female sexuality as traditionally constructed in romantic bourgeois fanta­
sy. The de-auraticised love is truly modern, and smashes to pieces the
system's self-understanding. Benjamin is excited by the aura-denying ele­
ments of love-for-sale. This acts out and exposes - for therapeutic pur­
poses - the commodity basis of social relations and the solidarity of con­
seller and consumed. The traces of the wage-labourer who pro­
duces the commodity are obliterated in the commodi
y on display in the
temples to consumerism. In this respect, the seller of her own body, at
least as theoretical fgure, retains one advantage. A allegory, she reveals,
for the purposes of social analysis, the visibility of two moments, dissolv­
ing the screen between production and consumption erected by commod­
ity society. This theoretical holism is important, because it smashes
through the divisive illusions of life under capitalism. A dialectical image
the prostitute synthesises the form of the commodity as exchange-value
and the exploitative process of marketing, in that she is commodity and
salesgirl in one. '06 She is the woman working in the public realm, a realm
of alienated labour, and she is also the product to be consumed. She
forms an emblem of commodity society, riddled with strained contradic­
tions that comprise the social. This is why she so enticingly fulfils the
function of exposure for the purpose of cognition. The writing of the
whore in allegory tears apart the harmonious naturalised fa�ade of the
world.107 And her actual existence reminds us again of the contradictions
and hypocrisies that grease the social world. Allegory freeze-frames reali­
ty, presenting it as deadened ruin (on a theoretical level) . Prostitution, in
denying the function of sex as the reproduction of life, denies nature (on
a practical level) and militates against socialised gender roles. The politi­
cal twist of Benjamin's argument is to invest the radical negativity of the
commodity-woman with a politically positive charge. Prostitution and
denatured femininity are read as social symptoms, an negation that
destroys heart-warming ideologies, a move that might eventually lead to
the destruction of its conditions of existence, though this effect is not qui­
etistically guaranteed. Wat more can theory attempt than to expose, to
make this process visible, in order to counter the mystifications that
accrue around the fantastic utopian image of the whore as inexhaustible
and never-refusing mother or the reactionary myth of woman as nature,
outside history, outside the social?108
Capital's rule fractures and fragments the female body as a prostitut­
ed, dehumanised commodity. Capital's organisation murderously con­
sumes life through the actions of reification and commodity fetishism.
These forces are redeemed by Benjamin and endowed with a critical
charge. It is quite clear that Benjamin's critical methodology is ripped
through on the female body. The whore is emblem of the dismal spectacle
Historical Matrialism
of unlove at the chilled heart of moderity. The fashion mannequin is a
token from the realm of the dead. The woman worker has relinquished
her sex. A methodology that aims to encapsulate the destructive activity of
capitalism mimics its actions, repeating its violence in the text and on its
images, by turning the whore into allegory, the mannequin into effigy, the
worker into material. The question is whether the turning of the whore
into allegory or the mannequin into efgy or the worker into material
. reduces her to a sign, or whether, as Benjamin seems to claim, such a
move honestly appraises the brutality to which she is subjected and which
objectifies her, whilst equipping her, allegorically, as super-critic of the
system that lays waste to her. The ultimate consumer-commodity con­
sumes itself and is consumed, and in these processes lays waste to social
order, in theory at least. A 'fellow deviant, an outsider', 'shadowy' and
'anonymous', dead and deadly, a consumed consumer, a worker, a seller
of mental and manual labour: all hold the promise of subversion.
1 . Benjamin 1 992, pp. 2 1 1-44. An English translation of one version of the
'Artwork' essay has appeared in many books.
2. The specifically ' 1 960s' rediscovery of Benjamin by the cadres of social revolt
was evidenced in their picturing of him with a photocopier in one hand and
joint in the other.
3. For example: Bolz/Witte 1 984; Frisby 1985; Buci-Glucksmann 1 986; Weigel
1 988; Cohen 1 993 Marcus and Nead (eds) 1 993.
4. See, for one example, Nixon 'Technologies of looking: retailing and the visual'
in Hall 1 997, pp. 333-6. Rather than shopping around disceringly and
knowledgeably in Benjamin's oeuvre, Nixon betrays his status as that of the
hooligan-theorist on a smash and grab ram-raid by his mistaken belief that
Grandville is some type of department store. Grandville was in fact a caricatur­
ist, and indeed Benjamin's favourite satirist of commodity fetishism.
5. McRobbie 1 992, pp. 1 47-69.
6. It is also controversial whether a theory that is said to have no notion of social
progress can be a critique.
7. For example, while it is true that Benjamin is scathing about the
bourgeois/social democratic notions of social progress, he formulates these
criticisms in order to theorise the conditions and measure of actual social
progress. He is not sceptical about progress per se. It is also inaccurate to sug­
gest that Benjamin forswears the working class as agent of revolutionary
change. His argument is rather with the leadership of reformist and revolution­
ary parties.
8. Of course, the relationship of Benjamin' s speculations to Marxism has been
argued over, and for good reason. Benjamin does not sit easily with any one of
the various currents of Marxism in the first half of the twentieth century. And
he adhered to no party and no orthodoxy. His relationship to Marxism and his­
torical materialism was a probing one. He read Marx, Lenin, Trotsky,
Luxemburg, Bukharin and Korsch and Lukacs, along with other less illustrious
names from the communist movement, and the ideas that he drew from these
works -as well as from his discussions with other Marxists of various hues -
including Brecht and Bloch and Adorno -flowed into his attempts to think
Inlamln'sArodes Poje
technology dialectically, to imagine social change and its conditions, to under­
stand the fetishism of the commodity and the pressures of commodification on
cultural production.
9. Gerhard Wagner notes the emergence, in the 1 980s, of an image of Benjamin
as ambivalent, caught, immobilised like a rabbit, between the headlamps of
Marxism and theology. Wagner puts this down to a crisis of legitimation for
the Left in the 1 980s. See Wagner 1 992.
10. Benjamin 1 973, pp. 1 70-1 .
1 1 . Benjamin 1 991 , p. 562. All translations from Passagenwerk, hereafter Benjamin
1 991 , 'Zentralpark', hereafter Benjamin 1 991 b, and'Notes', hereafter
Benjamin 1 991 c, are mine.
12. The fneur is not the same as the dandy. Benjamin locates the dandy histori­
cally as 'a creation of the English who were leaders in world trade'. The
dandies developed out of the merchants on the stock exchange who were
expected to react to conficts and the ups and downs of global commerce
quickly, without betraying inner emotions by their facial expressions. This
detached attitude is characteristic of the dandy, who occupies a relationship to
the movements of capital fow that is unlike the flaneur's. It is somewhat more
controlled, and identified with the ruling class. See Benjamin 1 973, p. 96.
13. Benjamin 1 991c, p. 1 1 59.
1 4. McRobbie 1 992, p. 1 49.
1 5. Buck-Morss 1 986, p. 105.
1 6. For key contributions see Wolff 1 989; Wilson 1 991 , 1 992; Pollock 1 988; Petro
1 989.
1 7. Wolff
989, p. 1 41 .
1 8. Wolff 1 989, p. 1 54.
1 9. Wolff 1 993, pp. 1 21 -2.
20. Wilson 1 991 , p. 56; 1 992, pp. 98-100.
21 . Wilson 1 991 , p. 56.
22. Wilson 1 992, p. 1 01 .
23. Wilson 1 991 , p. 55. There was, of course, a historical affinity between these
three fgures. All existing outside conventional morality and eshewing more
regular ways of making money, they inhabited the suspect zones of bohemia.
24. Buck-Morss 1986, p. 1 1 9. For Buck-Morss's development of this material, see
Buck-Morss 1 989. This book hopes to defuse a hardened contradiction
between Marxism and messianism, through her reconstruction of Benjamin's
fragmentary notes on the Parisian arcades. In the chapter titled 'Materialist
Pedagogy', Buck-Morss points out how Benjamin's view of the faneur and his
social function shift, as he turns from streetwise bohemian to salaried cop,
hack and political thug. See Buck-Morss 1 989, pp. 304-7.
25. Wilson 1 992, pp. 1 09-10.
26. Wilson 1 992, p. 1 10.
27. Benjamin 1 991 , p. 547; 1 973, p. 1 29.
28. Benjamin 1 991 , p. 1 29; 1 973, p. 54.
29. Benjamin 1 973, p. 54.
30. See Benjamin 1 973, p. 1 29 and 1 991 , p. 547.
31 . Benjamin 1 991 , pp. 426, 538.
32. Benjamin 1991b, p. 685.
33. Benjamin 1 973, p. 1 70.
34. Benjamin 1991, p. 529.
35. Benjamin 1991, p. 529.
Historical Materalism
36. Benjamin 1 991 , pp. 537-8.
37. Benjamin 1 991 , p. 562.
38. See Benjamin 1 991 , pp. 559-60.
39. Benjamin 1991 , p. 245.
40. See Wolff 1 993, p. 1 22; Buck-Morss 1 986, p. 1 22.
41. See Benjamin 1 991 b, p. 672.
42. Buck-Morss 1 986, p. 1 20.
43. Wilson 1 992, p. 106.
44. Wolff 1 993, p. 1 22.
45. McRobbie 1 992, p. 1 54. The charge, implict here and elsewhere, that Benjamin
visited prostitutes has also been used in evidence against hi, particularly in
that period when many feminists assumed that all [mention of acts of sex
meant advocating the rape of women. For what it is worth, George Steiner is
adamant that there is no proof that Be
jamin ever did pay for sex. Such insis­
tence might be but moralism from the other direction.
46. See Marx 1 906, p. 84; in this translation, as in the more recent Penguin
Capital, vo!' I , the German word sachlich is translated as 'material'. Sachlich
is an interesting word, for it means 'material', 'objective', 'businesslike', 'func­
tionaI'. In one word Marx combines the idea of people treating each other as
objects, while also getting across the extent to which such a relationship is
functional, apt, the necessary mode of intercourse in capitalism. This concrete
state of relations is, at the same time, however, abstract for it misses out, or
makes into an add-on the real core of human relations, the sensuousness of
substantial, human interaction. It may be that Marx has a antinomy in his head
that is suggested by the sound of words -sachlich versus sinnlich [sensuous J .
47. Benjamin 1 991 , p. 1 1 8.
48. See Benjamin 1973, pp. 74, 79.
49. See Benjamin 1 991 , pp. 1 010-1 1 .
50. Benjamin 1 991 , p. 1 08.
51 . Benjamin 1 991 , p. 61 7.
52. See Benjamin 1 991 , p. 637.
53. See Wilson 1 992, pp. 92-3.
54. See Benjamin 1 991 , pp. 971-81 ; 1 973, pp. 91 -4.
55. Benjamin 1 973, p. 94.
56. Benjamin 1 973, pp. 93-4.
57. Marx 1 906, p. 536.
58. Benjamin 1 991 , p. 447.
59. Benjamin 1991, p. 667.
60. Benjamin 1991, pp. 666-7.
61 . Benjamin 1991, p. 1 1 8.
62. Benjamin 1 973, p. 1 66.
63. Benjamin 1 991 , p. 436.
64. Benjamin 1 991 , p. 1 1 1 .
65. Benjamin 1 991 , p. 993.
66. Benjamin 1 991 , p. 438.
67. Benjamin 1 991 , p. 1 25.
68. Benjamin 1 991 , p. 475.
69. Benjamin 1 991 , p. 427.
70. Benjamin 1 991 , p. 436.
71 . There are connections to be made here between Benjamin's view of prostitu­
tion as the truth of social relations under capitalism and a sentiment in 'On the
Bnjamin's Arcde Pje
Jewish Question' [ 1 843] where Marx pinpoints social relations in the context
of commerce, whose legitimating philosophical core is labelled Judaic. Marx
writes: 'What is present in an abstract form in the Jewish religion-contempt for
theory, for art, for history, for man as an end in himself-is the actual and con­
scious standpoint, the virtue, of the man of money. The species-relation itself,
the relation between man and woman, etc., becomes a commercial object!
Woman is put on the market. ' Marx 1 975, p. 239.
72. Benjamin 1 991 b, p. 668.
73. Benjamin 1 991 , pp. 435-6.
74. Marx, who used as his gauge of human liberation the relationship between
man and woman, makes this observation, in the Paris economic and philo­
sophical manuscripts of 1 844: 'Prostitution is only a particular expression of
the universal prostitution of the worker, and since prosttution is a relationship
which includes not only the prostituted but also the prostitutor -whose infamy
is even greater - the capitalist is also included in this category.' Marx 1 975, p.
75. See Benjamin 1 991 , p. 400.
76. Benjamin 1 991 , p. 1 25.
7. See Benjamin 1 991 , p. 616.
78. Benjamin 1 973, p. 91 .
79. See Wilson 1 992, p. 1 06.
80. Benjamin 1 991 , p. 61 6.
81 . Benjamin's historical interest in the cult of sexual ambiguity may have been
stimulated by Weimar Berlin's famed sexual liberalism and experimentation.
82. Benjamin 1 991 , p. 1 1 1 .
83. Benjamin 1 991 , p. 1 1 8.
84. Benjamin 1 991 b, p. 668.
85. Benjamin 1 991 , p. 437.
86. Benjamin 1 991 , p. 1 31 .
87. See Benjamin 1 991 , p. 130. Marx's thoughts on minerals could be brought in
here to show how reflection on minerals can reveal the system's dynamic. In
the Paris economic and philosophical manuscripts of 1 844 Marx discusses
sensuous experience and its curbing under the rule of capital. The man who is
starving, says Marx, knows no human form of food, only its abstract form.
That is to say, he knows only its crude, restricted form. The man with worries
has no fine sense for drama's represented traumas and joys. And 'the dealer in
minerals sees only the commercial value, and not the beauty and peculiar
nature of the minerals; he lacks a mineralogical sense'. The dealer in minerals
sees only money, another hard, inhuman substance, when he sees his jewels.
But Marx insists that, in freedom, there wlbe a vision of jewels and minerals
that discerns their beauty and their specifcity truly and above all. The culinary
sense, the aesthetic sense, the mineralogical sense, have albeen restricted
under capital's dominion. And yet they hold out a promise; the very notion of
restriction implies a countervailing force that strives to realise itself. This
ambivalence, the stony monetary value of minerals coupled with their glorious
aesthetic value, resides in the object, and is struggled over in the struggle over
the form of social relations. Astract forms must be filled with sensuous, aes­
thetic, human meaning. See Marx 1 975, pp. 353-4.
88. See Benjamin 1 973, pp. 1 32-3.
89. Benjamin 1 991 , p. 1 23.
90. See Benjamin 1 973, p. 1 25.
Historical Mateialism
91 . See Benjamin 1 991 , p. 464.
92. Benjamin 1 991 , p. 1 002.
93. See Lukacs 1 983, pp. 83-1 1 0.
94. Benjamin 1 991 , p. 396; see also p. 41 7.
95. Benjami 1991, p. I l l .
96. Benjamin 1 973, p. 1 66.
97. Benjamin 1 991 , p. 448.
98. See Benjamin 1 991 , p. 1 18.
99. See Benjamin 1 973, p. 1 66.
1 00. Marx 1906, p. 525.
1 01 . Marx 1 906, pp. 484-5.
1 02. Benjamin 1 992, p. 248.
1 03. See Benjamin 1 979, pp. 1 57-9.
104. Benjamin 1991, p. 574.
1 05. See Benjamin 1991, p. 274.
1 06. See Benjamin 1 973, p. 1 71 . The translation is modified, for Harry Zohn uses
the word 'seller' to translate Verki uferin. This carries slightly different con
107. Benjamin 1 991 , p. 41 4.
1 08. See Benjamin 1 991 , p. 457.
Benjamin, Walter 1 973, Charles Baudelaire: A Lyric Poet in the Era of
High Capitalism, London: New Left Books.
Benjamin, Walter 1 979, One Way Street and Other Writings, London:
New Left Books.
Benjamin, Walter 1 991 , Passagenwerk, in Gesammelte Schrifen, volumes
V. l and V.2, Frankfurt am Main: Suhrkamp.
Benjamin, Walter 1 991 b, 'Zentralpark', in Gesammelte Schriften, volume
1.2, Frankfurt am Main: Suhrkamp.
Benjamin, Walter 1 991 c, 'Notes', in Gesammelte Schriften, volume 1.3,
Frankfurt am Main: Suhrkamp.
Benjamin, Walter 1 992, Illuminations, London: Fontana.
Berger, John 1 972, Way of Seeing, London: BBC/Harmondsworth:
Bolz, Norbert and Bernd Witte 1 984, Passagen, Munich: Wilhelm Fink.
Buci-Glucksmann, Christine 1 986, 'Catastrophic Utopia: The Feminine
as Allegory of the Modern', Representations, 1 2: 220-9, reprinted
from her L raison barque: de Baudelaire a Benjamin, Paris: Editions
Galilee, 1 984, also reprinted in her Baroque Reason: The Aesthetics of
Moderity, London: Sage, 1 994, pp. 97-105.
Buck-Morss, Susan 1 986, 'The Flaneur, the Sandwichman and the
Whore: The Politics of Loitering', Ne German Critique, 39: 99-140
(special issue on Benjamin) .
Buck- Morss, Susan 1 989, The Dialectics of Seeing: Walter Benjamin and
the Arcades Project, Massachusetts: MIT Press.
Cohen, Margaret 1 993, Profane Illumination: Walter Benjamin and the
Paris of Surrealist Revolution, Berkeley: University of California Press.
Frisby, David 1 985, Fragments of Moderit, Cambridge: Polity Press.
Hall, Stuart (ed.) 1 997, Representation: Cultural Representations and
Bnlamln's Arde. ProJe
Signifing Practices, London: Sage/Open University.
Hansen, Miriam 1 987, 'Benjamin, Cinema, Experience; The Blue Flower
in the Land of Technology', Ne German Critique, 40: 1 79-224.
Lukacs, Georg 1 983, Histor and Class Consciousness, London: Merlin
Marcus, Laura and Lynda Nead (eds) 1 993, 'The Actuality of Walter
Benjamin', Ne Formations, 20.
Marx, Karl 1 906, Capital, New York: The Modern Library.
Marx, Karl 1 975, Early Writings, London: Penguin/New Left Review.
MlcRobbie, Angela 1 992, 'The Passagenwerk and the Place of Walter
Benjamin in Cultural Studies; Benjamin, Cultural Studies, Marxist
Theories of A', Cultural Studies, vol. 6: 2: 1 47-69; reprinted in
McRobbie 1 994, Postmoderism and Popular Culture, London:
Petro, Patrice 1 989, Joyless Streets: Women and Melodramatic
Rpresentation in Weimar Germany, Princeton: Princeton University
Pollock, Griselda 1 988, 'Modernity and the Spaces of Femininity' in her
Vision and Difference; Femininity, Feminism and the Histories of Art,
London: Routledge.
Wagner, Gerhard 1 992, Benjamin BUder: Aspekte der Westeuropiischen
Rezeption Walter Benjamins von 1978 bis 1991, Hamburg: Bockel
Weigel, Sigrid 1 988, 'Traum - Stadt - Frau: Zur Weiblichkeit der Stidte
in der Schrift' in Die Unwirklichkeit der Stidte;
Grojstadtdarstellungen zwischen Modere und Postmodere, edited by
K.R. Scherpe, Reinbek: Rowohlt.
Wilson, Elizabeth 1 991 , The Sphinx in the City; Urban Life, the Control of
Disorder, and Women, London: Virago.
Wilson, Elisabeth 1 992, 'The Invisible Flaneur' in Ne Left Reie, 1 91 :
90-1 1 0.
Wolff, Janet [ 1 985] 1 989, 'The Invisible Flaneuse: Women and the
Literature of Modernity', in Problems of Moderity, edited by Andrew
Benjamin, London: Routledge, pp. 1 41 -56, reprinted from Theory,
Culture and Society, vol. 2: 3: 37-46; also reprinted in her Feminine
Sentences: Essays on Women and Culture, Cambridge: Polity Press,
1 990.
Wolff, Janet 1 993, 'Memoirs and Micrologies: Walter Benjamin,
Feminism and Cultural Analysis', Ne Formations, 20: 1 1 3-22.
Histrcal MatlaUsm
The law of value
and the analysis of
John Week
I ntroduction
Karl Marx entitled his frst major work on the theory of capitalism an
Introduction to the Critique of Political Economy, I not, it should be
stressed, An Introduction to . . . Political Economy. The inclusion of the
crucial 'the critique of provides the key to Marx's break with classical
political economy. A much as he respected the contribution of bourgeois
writers, especially Ricardo, he did not consider himself a radical member
of the political economy school. That the political economy school's most
outstanding members focused upon class relations did not save them
from an analysis that, in Marx's judgement, was 'vulgar', in that it
focused upon the appearance of phenomena rather than their underlying
causes. Political economy focused on relations of exchange, rather than
on class relations among human beings. A he wrote famously in an oft­
quoted letter,2 for at least a generation before him bourgeois writers had
recognised both class divisions in capitalism and that the basis of proft
was exploitation; were these the central elements of his work, his contri­
bution would have been trivial. Marx identifed what in his assessment
was the central failing of the political economy of Smith, Rcardo, et al:
Political Economy has indeed analysed. however incompletely, value and its
magnitude, and has discovered what lies beneath these forms. But it has never
once asked the question why labour is represented by the value of its product
and labour-time by the magnitude of that value.'
At frst reading this passage may seem obscure. It refers to failure of
the political economy school to understand that 'markets' are associated
with specific social relations in which production, distribution and
exchange are organised. Smith's 'invisible hand' purported to be a social
mechanism for all time and all places; what Smith conceived as a guiding
principle of self-sufcient individuals, Marx revealed as unique to a soci­
ety divided between proletarians and capitalists. Marx's formulation of
the labour theory of value reveals 1 ) the laws of reproduction of a capital­
ist society; 2) that capitalism ('market society') is a historically specifc
form of class society; 3) that changes in class relations explain the transi­
tion from pre-capitalist to capitalist society; and 4) why others explain
both the transition and the laws of capitalism in a alterative framework.
This four-fold character of the law of value constitutes its methodological
break with the political economy school. It is not merely one theory
Histrical Matrialism
among competitors, but a theory which subsumes its competitors within
it by demonstrating that they focus upon the appearance of social phe­
nomena rather than their essence.
Modern political economy ('neo-classical economics') rejected classi­
cal political economy's value theory; it shares with the latter its ahistorical
methodology. The ahistoricism of both the Classicals and the Neo-classi­
cals is not an oversight that might be remedied by the inclusion of an his­
torical analysis in the chronological sense. Especially Smith, but also
other classical political economists, made reference to the historical
development of capitalism. Their method was ahistorical. Similarly, neo­
classical political economy has produced its own economic historians, but
their 'history' is ahistorical in as far as it treats the processes of produc­
tion, circulation, and distribution of the social product. For the Classicals
and Neo-classicals, economic history is the study of relations of
exchange. As such, all periods are essentially the same; society has
chronology, but no history. Strictly speaking, it is not valid to criticise the
Classicals and Neo-classicals for lack of an historical perspective: given
their value theory, they should have none.
This paper frst develops the core of Marx's theory of value. The main
theme is given in the text, with elaboration in footnotes. Value theory is
used to reveal the ahistorical character of neo-classical political economy.
Then, it is possible to consider the role of value in the passage from pre­
capitalism to capitalism. Finally, the insights obtained are employed to
develop generalisations about the barriers to accumulation in societies i
which capitalism is not fully developed. In order to avoid idolatry, the
framework developed will not be called 'Marxist', though Marx was its
originator and most infuential exponent. Rather, we use the oft-maligned
term 'historical materialism' to mean that approach in which the process
of social reproduction derives from the social relations in which produc­
tion is organised: Within this school of thought, the analysis of capitalism
is that of a particular historical period, in which social reproduction
involves the circulation and distribution of the products of labour as com­
modities. It is within this period and this period only that the law of value
Mar's theor ofvalue
In capitalist society, the products of labour appear as 'an immense accu­
mulation of commodities'. 5 That commodities are the product of human
labour in itself implies no particular value theory; it is a statement of the
obvious. A commodity has a dual nature. For the seller it represents a
quantit of value, which when realised in generally-equivalent form can
be used to acquire another commodity through further exchange. This
quantity of value is the exchange-value of the commodity (what it fetches
in exchange) . For the buyer of the commodity, it represents a qualit
which is sought for a particular purposeful use. This quality constitutes
the use-value of the commodity. The distinction between the quantitative
and qualitative aspect of a commodity is obvious and descriptive. It is the
pursuit of this obvious and uncontroversial dichotomy that yields the
labour theory of value and the laws of capitalist development.
The exchange-value and use-value of a commodity are not at peace
with each other. While the former can vary due to immediate and longer-
Analysis of undereelopment
term infuences, the latter retains an intrinsic character; more specifcally,
improvements in the methods of production can reduce exchange-value
for a given use-value. On the basis of the tension caused by this real
dichotomy arises the need for money, which can now be defned as a
general equivalent commodity of exchange.6 At this early juncture in the
theoretical discussion, the analytical method should be noted. We did not
at the outset presume the existence of money; rather, its role emerged in
consequence of considering the nature of commodities.
Out of the 'unpacking' of the commodity arises the need for a further
concept. Since commodities do not exchange directly for each other, but
through the intermediary form of money, the possibility arises that the
exchange-value of a commodity can vary as conditions of exchange vary.
This raises the question of what determines the exchange-value of a com­
modity; i.e. the underlying determinant of exchange-value as the money a
commodity fetches fluctuates due to stochastic influences. All theories of
market prices posit the existence of an underlying determinant of
exchange-value which is hidden beneath the price form of exchange. In
neo-classical political economy the underlying determinant is the 'oppor­
tunity cost' of both producers and buyers; for Ricardian political econo­
my it is the technology of production; and for historical materialism it is
socially necessary labour, a concept we have yet to unfold. For all the
schools, there is a value of commodities which lies beneath the surface of
exchange. 'Value' has a straightforward and unambiguous meaning: that
which determines price, appearing in the form of quantities of money.
Like commodities, money has a contradictory nature. A the general
equivalent, it circulates with commodities, but, unlike other commodities,
it need not be sold to realise its exchange-value (it is exchange-value) .
With this characteristic it can serve as a general store and claim on value.
A a claim on value, it can initiate exchange for commodities, commodi­
ties which can, in turn, be sold for money again. This process, exchang­
ing money for commodities, then commodities for money, would only be
done if the second quantity of money exceeds the initial quantity. I is in
this way that money serves as capital, which is defmed as self-expanding
value; money which through circulation yields more money. To this point
an increasingly complex series of phenomenon has been unfolded: from
the commodity and its two-fold character, money (implying value) , to
capital. This unfolding as yet produces no theory of the determination of
value; indeed, it can be taken as descriptive of the process of exchange. It
allows one to note that exchange can be viewed in terms of two forms of
commodity circulation: commodities via money to other commodities,
and money via commodities to a greater quantity of money. The former is
simpler, selling in order to buy (commodities-money-commodities, C-M­
C) . This is simple, because it requires little theoretical explanation. It
involves disposing through exchange of a commodity whose use-value is
not desired, in order to obtain money, which can be employed to acquire
a desired commodity. 7
The second, buying in order to sell, is theoretically complex. It
demands an explanation of the source of the increased quantity of money.
Following Marx, we call the process of buying in order to sell the circuit
of capital. This increased amount of money that appears through buying
in order to sell Marx gave the straightforward name, 'surplus-value'. On
Historical MateriaUsm
the surface, surplus-value is a simple concept: it is the quantitative differ­
ence between the money at the end and beginning of a process of buying
i order to sell. The theory of value arises from the need to explain the
source of surplus-value." Here, again, we must pause and refect on the
implications of what has been, at least superficially, a descriptive discus­
sion. When money does not serve as capital, there is no surplus-value9 to
explain, thus no role for a theory of value. Value theory, whatever its logi­
cal basis, is historically specifc to the circuit of capital, though we have
yet to make that historical specification.
We can rule out surplus-value deriving in the aggregate from
exchange itself.lo It follows that capital (money) must exchange initially
for a commodity whose value increases between buying and selling. A
commodity's value increases by entering into a process of production.
That value expands in production, would be met with agreement by the
neo-classical political economy school, 1 1 though its view of production
would not conform to that of the historical materialist school. 12 Since an
increase in value in the aggregate arises from production, it follows that
capitalists must pass through production to obtain surplus-value. This
obvious point implies that the circuit of capital needs expanding to take
the form:
M-C ... P . . . C'-M'.
[Money > > Commodities, Production, Commodities > > Money]
The next step in the analytical unfolding is also obvious and non-con­
troversial: the first purchase by the capitalist (M-C), must involve the
acquisition of a commodity or commodities which, when used in produc­
tion, add value. By defnition, the elements of production can be divided
between workers - the human agency of production - and the material
elements of production. We shall call these the labour input and the
means of production, where the latter can be sub-divided between those
that are materially transformed or consumed during production (for
example, raw materials, electricity), and those which retain their material
form (for example, machines, buildings). Including these elements, the
full circuit of capital becomes:
M-C [L, MP] ... P ... C' -M'
[Money > > Elements of Production, Production, Commodities > > Money]
To this point the labour theory of value plays no role; it lies latent in
the analysis. The fundamental difference between the foregoing analysis
and that of neo-classical political economy is that the former has re-cast
the analysis of value as derivative from capital, while the latter treats value
as the outcome of the desire by individual human beings for consump­
tion. 1 The point now has come to declare one's value theory. The analy­
sis cannot progress beyond description without an explanation of the ori­
gin of the surplus-value arising from the circulation of capital. 14 Because
of the manner in which materialist analysis describes aggregate reproduc­
tion, the neo-classical theory of value cannot be utilised.
Even if one preferred to use the opportunity cost theory of value, our
framework, M-C . .. P . . . C' -M', precludes it. In the neo-classical theory of
Analysis o underdeelopment
value, commodities are not produced in the real-world sense. Stocks of
'primary inputs' ('labour' and 'capital') , when combined with a given
technology, generate a fow of new value. In this stock-fow description of
the economy an opportunity cost theory of value is consistent, albeit
under highly restrictive conditions. " The materialist description of aggre­
gate reproduction is not stock-fow, but involves the production of com­
modities by means of commodities, to use Sraffa's term. ' o In this descrip­
tion, the process of reproduction is considered in time periods. In
arbitrarily selected initial time period (in principle one could go back to
the Garden of Eden), there is produced a set of commodities which will
be the input in the next time period. In the next time period those com­
modities are transformed into different material objects, during which
value is added to them. ' 7 This view of production formally excludes mar­
ginal productivity analysis. It does so not because it allows no substitution
between inputs, '8 but because formally there is no difference between
material inputs that are consumed during one period of production and
those used-up over many time periods. Just as electricity is consumed in
production and passes on no more value than its own, machinery, build­
ings, etc., are exhausted of their value over many (though in principle a
definite number) time periods, passing on their value but no more.'9 The
commodity which can expand value is the labour input. This commodity
which capitalists buy is the capacity to work, or labour-power. However,
this analysis does not as yet provide a theory of value.
The production of commodities by means of commodities framework
requires either a labour-based theory value or the Ricardian technology­
based theory. We consider only the former. There is little controversy
over whether units of labour time can be employed to measure value.20
Measurement is essentially a trivial exercise for which there are several
possibilities. For example, if we consider only material commodities
(excluding services), one could aggregate by weight. However useful this
might be for certain purposes, such as planning the transport of com­
modities, it makes no sense as a theory of value. Similarly, labour time
can be used as the unit of measure; the debate is over its signifcance for
understanding aggregate reproduction. Marx's argument proceeds from
the tautology that each commodity is the product of human labour. When
commodities exchange, they are rendered equal in practice. By definition,
the labour that produced them is rendered equal through exchange: the
concrete labour expended in production is converted to abstract labour in
exchange (i.e. into money) .
This purely formal conversion of concrete labour into its opposite
becomes a real conversion through the process of competition.
Competition among producers of a particular commodity establishes a
standard input requirement, which Marx called socially necessary labour.
This is rendered abstract through exchange, becoming abstract socially
necessary labour. The labour theor of value is explanation of the process
by which abstract necessar labour is established through a social process
(how the assumption achieves credibility) . This breaks with the Ricardian
framework, which explicitly or implicitly takes abstract socially necessary
labour as given: all labour is treated as homogeneous (or can be rendered
so), and all producers of a commodity use the same technology with the
same efciency (or can be treated as doing so). The Ricardian approach
Historical Materialism
does not explain the special historical conditions under which a norm in
production is established (i.e. why the functioning of society requires it) .
Unfolding the nature of commodities has provided the explanation: com­
petition results from the general production of commodities when labour­
power is also a commodity; a common norm for the production of each
commodity arises from the exchange of inputs and outputs. At each stage
in the input-output process capitalists encounter the discipline of
exchange.21 Marx referred to the disciplining effect of exchange when he
wrote, 'a commodity is, in the frst place, an object outside US.'22
We can now summarise the development of the materialist theory of
value or law of value: commodities are the products of human labour
which are produced within the discipline of capitalist exchange, both for
the output and the inputs that are used to create the output. Production is
formally private, but essentially social.23 Every producer participates in
social interaction in which his/her commodity is but one part of an
organic whole. This system of social reproduction arises from labour­
power being a commodity. The commodity status of labour-power results
from the separation of producers from the means by which production
can be carried out.24 Producers (workers) are re-united with the means of
production through the agency of capital. Having re-united workers and
means of production through exchange, capitalists must transform com­
modities back into money.
Theory of proft
The surplus-value arises from the extraction of surplus labour by capital­
ists in the process of production. This surplus labour is created as a result
of the difference between the value of the commodity labour-power and
the value which labour-power creates during a production period. We
have yet to explain why these two should differ. In as far as labour-power
is a commodity, its exchange-value is determined by its cost of produc­
tion. Of all the elements of Marx's theory of value, the analysis of wages
is perhaps the subtlest, and only the basic argument is presented here. A
a superfcial level, it can be treated as a subsistence theory of wages.
Without specifying relations of production, society's aggregate labour can
be divided between the labour necessary to reproduce the working popu­
lation, and the labour expended over and above that reproductive mini­
mum. Marx called the former necessar labour and the latter surplus
labour. This division implicitly assumes the division of society into class­
es, so that there is a dominant group which appropriates the surplus
In a capitalist society this appropriation occurs through the interaction
of exchange and production. Capitalists enter into a transaction with
workers, in which money (variable capital) is exchanged for labour­
power. The price at which this exchange occurs (the wage) is determined
by the exchange-value of the collection of commodities which workers
require to reproduce their labour-power. In return, capitalists receive
control over the productive potential of workers for a prescribed length of
time. Capitalists use the labour-power to produce commodities which
they exchange for money. A Kaldor said, workers spend what they get;
capitalists get what they spend (plus some, he might have added).26
The distinction, between the exchange-value of the commodity labour-
Analysis of underevelopment
power and the subsequent exchange-value of the commodities workers
produce, is not the materialist theory of profit as such. Profit as such
results from surplus labour. The existence of surplus labour follows from
the analysis of production by means of products: labour is the input to
production which can expand value in this framework. The distinction
between the exchange-value of labour-power and the exchange-value of
what workers produce provides the analysis of accumulation; i. e. the
explanation of how proft can be increased.
Given the exchange-value of the commodities required for the repro­
duction of the labour force (the value of labour-power) , surplus labour
can be increased in two ways. It can be raised absolutely by the extension
of the working day or an increase in the intensity of work. This method of
increasing surplus labour has natural limits, but also limits in terms of the
norms of 'fairness' in society. Marx called this the raising of surplus-value
absolutely (or, the production of absolute surplus-value) ; extracting more
effort from workers without compensation. He associated this with the
early stage of capitalist development, when capitalists faced social rela­
tions that limited their ability to introduce technical changes.27 A the
fnancial system develops, allowing capital to be redistributed towards
more eficient producers, and the struggle of the working class sets limits
the working day, the raising of surplus-value absolutely becomes sec­
ondary to the raising of surplus-value relatively. surplus-value is raised
relatively by the reduction of the necessary labour component in society's
aggregate working time. 2
While individual capitalists can raise surplus-value absolutely through
their own efforts, it is raised relatively through a social process. Consider
a capitalist that produces a commodity that serves as input into a com­
modity that is part of workers' means of subsistence. A technical change
which lowers the cost for one capitalist allows that capitalist to enjoy
greater than average profit for that sector of production. This is followed
by a process of competition which induces the other capitalists in the sec­
tor to adopt the same or similar cost-reducing technology. A all or most
capitalists in the sector come to enjoy above average profits, capital
migrates from other sectors into the now more-proftable sector. This
drives down the exchange-value of the commodity, which makes it cheap­
er for all capitalists that use it as input. In a second competitive round,
the commodity using the cheaper input falls in exchange-value. Since by
assumption this second commodity enters into workers' consumption,
the value of labour-power falls. I the standard of living of workers
remains the same, then the exchange-value of labour (the wage) falls.
This process of technical change and competition demonstrates the social
character of capitalist production: in as far as any capitalists enjoy a rela­
tive rise in surplus-value, all do.29
It was through his analysis of absolute and relative surplus-value that
Marx further demonstrated the historically specifc character of capitalist
production. Surplus labour, appropriated by a dominant class, charac­
terised societies for thousands of years. It was first under capitalism that
the increase of surplus labour through the social interaction of producers
became a dynamic force of acumulation. This dynamism arises from the
general production of commodities, in which society's surplus labour
takes the form of surplus-value (i.e. it must be realised in money form) .
Historical Matrialism
Value and social transformation
Since the elements of production in a commodity society manifest them­
selves as values in exchange, they are converted in the minds of producers
into elements of value.30 This thrusts upon society a set of exchange cate­
gories that define the nature of commodities in as far as they are values:
materials costs, wages, proft, interest, rent. These categories are real in
that they represent actual payments that capitalists must make. More
important, they represent the form by which the social character of pro­
duction is enforced. A independent producer, such as a farmer who
owns her/his own land, fnds these categories imposed upon her/his
activities. Marx makes this point with concise clarity:
To himself as wage-worker he pays wages, to himself as capitalist he gives the
profit, and to himself as landlord he pays rent. Assuming the capitalist mode of
production and the relations corresponding to it to be the general basis of soci­
ety, this subsumption is correct."
This imputation is the basis for the ahistoricism of neo-classical politi­
cal economy. As a result of the real subsumption of non-capitalist pro­
duction to the categories of capital 'the illusion is all the more strength­
ened tat capitalist relations are the natural relations of every mode of
production.'32 Independent producers are not capitalists (they do not hire
themselves in a market) ; they are not landlords (they do not rent their
land to themselves); and they are not fnancial rentiers (they do not lend
themselves money) . From the imposition of capitalist categories on all
social production in a capitalist society, neo-classical political economy
moves to the generalisation of the categories to all societies in all time
periods. Thus, the hunter-gatherer is interpreted as weighing the trade­
off between current consumption and the accumulation of capital;" the
sharecropper's behaviour is analysed in terms of transactions costs. A a
result, the social dynamics of all societies are treated as if they occur
under the rules of a capitalist order. This gives neo-classical political
economy its a- and anti-historical character, rendering it incapable of
analysing social change.
In contrast, the materialist theory of value provides both a general
(abstract) explanation and concrete analysis. Capitalism is the frst form
of social production in which class relations are sustained through the
general circulation of commodities. Beneath this generalised commodity
production lies the separation of labour from the means of production.
This separation results in competition among capitals, which is the source
of the dynamism of the capitalist mode of production. While in previous
societies the products of labour were exchanged against money to varying
degrees, this exchange was incidental or at most secondary to the domi­
nance of the appropriating classes over the direct producers. A a result,
there was little tendency for exchange-value to feed back upon produc­
tion and create a social norm of eficiency (abstract socially necessary
labour) . In other words, pre-capitalist production occurs in isolated units,
not socialy integrated through capitalist markets. In contrast, capitalist
production is formally private (based on private property) , but socially
integrated through commodity circulation.
Analysis of underdevelopment
The prices attached to pre-capitalist products are superficial; they do
not regulate the production process. In general, land is not a freely­
vendible commodity in pre-capitalist societies, so prices do not determine
its allocation. Similarly, labour-power is only marginally a commodity,
because the direct producers are not separated from the means of pro­
duction. A a result of the incompleteness of land and labour-power's
commodity status, exchange has a limited effect on the social order. To
take the most obvious example, it can drive independent producers out of
production for exchange, or to the margin of markets, but cannot dispos­
sess them through mark
t processes alone. For this reason Marx made
his famous comment that individual private property (petty-bourgeois
private property) is everywhere a barrier to the development of the capi­
talist mode of production.34 The law of value shows, on the one hand, the
source of capitalism's dynamism, the separation of labour from the means
of production and the rendering into commodities the natural environ­
ment. On the other hand, the law reveals through its absence or limited
applicability the historica specificity of its social existence.
In all class societies there are two major sources of conflict: 1 )
between the appropriating (ruling) class and the producing class over the
conditions of exploitation of the latter; and 2) among factions of the
appropriating class for control of the state. In pre-capitalist society, the
latter conflict manifests itself typically as armed conflict in order to
extend control over populations and territories. Under the rule of capital,
conflict among factions of the dominant class takes the form of competi­
tion among capitals (firms) . Over no other issue is the diference between
neo-classical and materialist theory more unmistakable than for competi­
tion. The former theorists laud it as the mechanism for harmony and
social welfare gains; in the analysis of the latter competition is the source
of instability and uneven development, which can provoke armed conflict
among capitalist states. In materialist theory, warfare is the continuation
of market competition by other means.
Conflict in the form of cheapening of commodities does not eliminate
armed conflict; quite the contrary." But it is progressive because it results
an unprecedented development what Marx called humankind's mastery
over nature. This development of the productive forces, inherent in com­
petition, is not necessarily a good thing. 'Progressive' is used i a descrip­
tive, not a judgmental sense, to focus on an essential characteristic of the
expansion of capital: inherent in it is the progressive development of the
power to produce.36 While technological innovation occurs in other social
forms of production, only in under capital is it the principal method of
struggle within the appropriating class. This makes the dynamics of social
change different in pre-capitalist and capitalist societies. Class struggle i
capitalist society is driven in great part by the necessity for capitalists to
raise surplus-value relatively; class struggle in pre-capitalist society
occurs in the context of the appropriating class largely restricted to rais­
ing surplus-value absolutely.
One can contrast a capitalist society with a pre-capitalist, because all
capitalist societies share a common characteristic: human labour-power
takes the form of a commodity (for that is what capitalism is) . However,
other than being divided into classes, pre-capitalist societies share no
analogously defning characteristic. That is, they share no universal mode
Historical Materialism
of surplus appropriation which generates a common social dynamic
among them. For this reason, a general and abstract discussion of transi­
tion from pre-capitalist to capitalist society is by its nature extremely lim­
ited in analytical power. Theory tells one the abstract outcome of the
transition (the appropriation of a surplus product through free wage
labour),37 but not from where the transition came. It is tempting to con­
struct a false, universal pre-capitalist society which, in effect, is the oppo­
site of an abstract, fully-developed capitalism. This treatment of pre-capi­
talist societies, as what they are not rather than what they are, de-empha­
sises to the point of insignificance changes in the social relations of pro­
duction. The transition to capitalism becomes the growth of exchange
relations, rather than a revolution in production and property relations.
The analytical unfolding of value theory reveals the superficiality of
using the spread of exchange to explain the transformation of pre-capital­
ist social relations into capitalist wage labour. Societies maintain their
coherence by the effectiveness of the appropriating class in controlling
the direct producers. The importance of exchange in a society is deriva­
tive from the form that control of the direct producers takes.38 Exchange,
even capitalist exchange, is a surface phenomenon, constructed upon the
prevailing relations of production. Marx called te analysis of exchange
'commodity fetishism', meaning that which treats social relations as rela­
tions of exchange:39
[The] money-form [price] of the world of commodities ... actually conceals,
instead of disclosing, the social character of private labour, and the social rela­
tions between individual producers.'
In a capitalist society, exchange plays a role unique unto that society:
it conceals the appropriation of surplus labour. All exchanges appear as
equal exchanges - the buyer receives the commodity and the seller its
exchange-value in a formally voluntary transaction. This equality in
exchange is then imposed downwards ideologically to production rela­
tions, so that each element of the valued added, wages, rent, interest and
profit, is imputed to a 'factor of production' as refecting equal exchange.
The interpretation of these components as flows of value arising from the
contribution of land, labour and capital to newly created value Marx sar­
donically called the 'Trinity Formula'. These income categories ('rev­
enues' was Marx's term) become relevant only under capitalism. They do
not refect the inherent nature of production, but rather the class divi­
sions in a capitalist society. To treat them as relevant to all historical peri­
ods is the essence of 'commodit fetishism', ignoring social relations to
focus on exchange.
Limits to accumulation in transitional societies
Capitalism develops within pre-capitalist society through the conversion
of labour-power and land into commodities!' This conversion is achieved
through a process of coercive social change, such as an armed insurrec­
tion or a confiscatory land reform. Fo a time, perhaps a considerable
time, the old social relations persist alongside the emerging relations of
capitalist wage-labour. Pre-capitalist relations tend to persist longest in
agriculture, because of the difculty of divorcing the peasantry from the
Analysis of underdeveloment
land, and, thus, making land and labour-power commodities. A long as
pre-capitalist relations continue in agriculture, this provides a break on
the process of accumulation. 'Structuralist' economists have given con­
siderable attention to this problem:2 In the Latin American structuralist
approach, the process of urbanisation generates an increasing demand
for food for the growing working class, and the expansion of capitalst
production increases the demand for agricultural raw materials .
Backwardness in agriculture resuls in an inelastic supply of foodstuffs
and inputs for industry. A a consequence, relative prices move against
the capitalist sector depressing the proft rate.43
While the description is correct, this analysis lacks a clear theoretical
foundation. The movement in relative prices is wholly dependent upon an
inadequate supply of products from the pre-capitalist sector. This inade­
quate supply results from the specific behaviour attibuted to the pre-capi­
talist landlords, who are assumed not to respond to market signals. The
materialist theory of value provides a more analytically general explanation,
whch is not dependent upon specifc institutional arrangements. Consider
again the circuit of capital, shown in Table 1 . Historically, capital emerged
frst as merchants' capital, which involved buying in order to sell, the form
of capital, M - C ¯ M' , without the essential characteristic of capital, wage
labour. In this perod, prior to the industrial revolution, European merchant
houses mediated in the exchange of products arising from pre-capitalist
social relations. Profit (M' > M) arose from the monopoly over trade, buy­
ing cheap and selling dear. In this process, the export of commodity capital
could reinforce pre-capitalist production relations, since the origin of the
products exchanged was of no concern to merchants' capital, as long as
their supply was assured. The trade in commodities neither required nor
necessarily brought about a transformation in social relations.
1 . Mom!nt of circulation 2. Moment of 1roduction 3. Moment of Circulation
Advance of capital Realisation of value
(Money capital into (Productive capital into (Commodity capital into
productive capital) commodity capital) money capital)
M-C (MP & LP) C (MP & LP)-C' C'-M'
[M ¯ C, no change in [C' > C, value expand] [C'
¯ M', no change in
value] value]
Financial capital Industrial capital Merchants' capital
InternatiQl al EX1ansion
of ca1ital:
Export of money capital Export of industrial Export of commodity
capital capital
[capital movement on [transformation of the [tra
e on the basis of
basis of existing social social relations in pre- existing social relations of
relations of production] capitalist societies] production]
Table 1: The life cycle of capital
Historical Matrialism
By the middle of the nineteenth century, the growth of merchant
houses and the emergence of capitalist relations in Western Europe and
the United States brought the rise of large fnancial institutions, whose
purpose was the vending of money as such, rather than commodities. In
underdeveloped regions, the export of money capital fnanced trade, and
also provided loans to governments, especially in Latin America, to fund
public works such as ports and railroads. A with the trade in commodi­
ties, financial houses could pursue their profts without the development
of capitalist wage labour.
The export of commodity capital and money capital had a contradicto­
ry tendency in underdeveloped regions, as implied by their position in the
circuit of capital. By enhancing the wealth and power of ruling classes in
the underdeveloped regions, pre-capitalist social relations could be ren­
dered more tenacious. At the same time, the import of capitalist com­
modities into underdeveloped regions acted to destroy local artisanal and
peasant production. Simultaneously, infrastructure projects fnanced by
the export of money capital created an emerging class of wage labour.
However, even by the early twentieth century internationally trade com­
modities from underdeveloped countries arose overwhelmingly from sys­
tems of forced labour, debt-bondage, and forms of patron-clientage.
Upon this infertile ground of unfree labour, the scope for the export of
productive (industrial) capital was extremely limited: the industrialisation
of underdeveloped regions required a prior process of the dissolution of
pre-capitalist relations. The materialist theory of capitalist development
provides few generalisations about the process by which wage labour
emerged in each country and region. No general theory of the rise of cap­
italism is possible, for in each case the failure or success of the relations
of capital to take root is dictated by the nature of the pre-capitalist soci­
ety. However, some general insights are possible. In those underdevel­
oped regions where the relations of capital took hold, accumulation was
limited by its essential incompatibility with pre-capitalist production.
From a superficial point of view, pre-capitalist relations appear beneficial,
since they could represent a source of cheap labour-power. However, this
remains a latent beneft to industrial capital until the pre-capit
list sector
unravels and sheds its labour. Even as this occurs, value theory reveals a
profound incompatibility that calls the concept of 'cheap labour' into
question; i.e. reveals it as an essentially vulgar concept at the level of
Because pre-capitalist production does not conform completely to the
discipline of the law of value, production techniques continue, primarily
in agriculture, which would be abandoned by capitalists as unproftable.
Restrictions on the alienability of land and the immobility of unfree labour
restrict the ability of capital to transform agrarian relations. However, it is
the agricultural sector which provides a large portion of the means of
subsistence of the workers that capitalists hire. In an abstract, fully capi­
talist so
iety, the process of competition reduces the vaue of commodi­
ties, which feeds back through the system to reduce the value of labour­
power. Such is not the case for a society in which the means of worker
subsistence are produced in pre-capitalist relations.
Other things being equal, a worker with a lower standard of living is
'cheaper' to a capitalist than one with a higher standard of living."
Analysis of underdevelopment
However, this refers to increasing surplus-value absolutely, and cannot be
the mechanism for the progressive cheapening of commodities. No mat­
ter how low the standard of living of workes, subsequent cheapening of
commodities requires that surplus-value be raised relatively in the capital­
ist sector. A we have seen, surplus-value is raised relatively through the
reduction of the value of labour-power. In as far as workers consume
commodities produced in the capitalist sector, the process of competition,
by cheapening these commodities, reduces the value of labour-power.
However, the component of workers' subsistence which arises from the
pre-capitalist sector is not cheapened by the process of capital accumula­
tion. The fall in the value of commodities produced by capital is not
matched by an equal fall in the value of labour-power, resulting in a
decline in surplus-value per worker, even if the standard of living of
workers remains the same. Thus, the interaction of the capitalist and pre­
capitalist sectors results in a fall in the rate of proft in the former sector.45
In simple terms, competition in the capitalist sector results in a transfer of
value to the pre-capitalist sector via a movement in relative prices against
the capitalist sector.
The fall in the rate of proft does not require that machinery or other
means of production are substituted for labour inputs. A autonomous
increase in labour productivity46 is suffcient to bring down the rate of
profit. At the initial set of exchange-values of outputs and inputs, any
autonomous productivity increase appears to the capitalist as cost­
decreasing, though it subsequently results in a decline in unit profts.
Even if capitalists could foresee the fall in profts due to the productivity
increase, they would be forced to adopt by the pressure of competition: if
some capitalists did not, others would, i order to reduce their costs and
seizing a larger market share. The special conditions of underdevelop­
ment force capitalists to adopt productivity-raising techniques that pro­
voke a tendency for the rate of profit to fall.
The process of accumulation in the context of a pre-capitalist sector
producing the means of consumption for the workers in the capitalist
sector can be summarised as follows. Competition in the capitalist sector
results in the introduction of cost-reducing technical changes. By raising
the productivity of labour, these technical changes increase the use of
materials per unit of labour, and even more if accompanied by an
increase in machinery per worker (not considered in discussion) . Were
both sectors capitalist, the result of technical change in either sector
would be to raise the average rate of profit as exchange-values of the
means of workers' subsistence fall. Values, then, via competition, prices,
would fall. The general cheapening of commodities raises the rate of
profit in a purely capitalist society.
When the means of consumption are produced under pre-capitalist
relations, the result is dramatically different. A before, competition in the
capitalist sector results in the introduction of cost-reducing technical
changes. This brings down the price of capitalist commodities. In as far
as the capitalist sector uses its own outputs as input, the tendency for
proft to decline is counter-acted. However, this does not affect the distri­
bution of the working day between necessary and surplus labour time. In
order that surplus-value rise, it is necessary for the value of labour-power
to decline. This will occur if and only if there is a decline in the value of
Historical Materialism
the commodities workers consume, or a decline in the value of the inputs
to those commodities. Since pre-capitalist sector may use capitalist inputs
(as in the model), the price of its output may fall, and thus the value of
labour-power may fall. But this will be insuficient to keep unit proft
from declining in the capitalist sector.47
There are two major tendencies that can counteract the tendency of
the average rate of proft to decline in the context of pre-capitalist rela­
tions in agriculture. First, in as far as the pre-capitalist sector uses capi­
talist inputs, technical change in the capitalist sector will lower the value
of the means of consumption; i.e. as the pre-capitalist sector comes under
the discipline of exchange in its use of inputs. Second, the tendency is
counteracted as worker's consumption incorporates commodities pro­
duced in the capitalist sector. The specific historical conditions and social
relations in each country will determine the strength of these counteract­
ing tendencies. Most fundamentally, removing the limits to accumulation
requires a capitalist revolution in agriculture. The inability of the law of
value to bring about the progressive reduction of necessary labour rela­
tively to surplus labour refects the contractions arising from two systems
of social relations which are inconsistent in the long run.
In this context, one can identify a major source of the successful capi­
talist development of the East and Southeast Asian countries: many of
these countries benefited from a fundamental reform of agricultural and
tenure prior to their rapid accumulation process. In Latin America, land
reforms did not occur at all (for example, Guatemala) or were not com­
plemented by support services necessary for capitalist transformation
(Bolivia, Mexico and Peru) . The most successful case of capitalist devel­
opment, Chile, indicates the importance of land reform in laying the basis
for accumulation.48
The limits to the progressiveness of capitalism
A important implication of the theory developed above is that underde­
velopment is the incomplete development of capitalism in a society.49 This
would seem to imply that capitalism is the solution to the problem of
underdevelopment, a conclusion argued with considerable force by
numerous authors who considered themselves Marxists.50 It is hardly sur­
prising that many on the Left have drawn back from this conclusion,
implying as it does some rather unpalatable political practice. While the
logic of the law of value implies capitalism is progressive, it is relevant
here to recall Oscar Wilde's aphorism that madness is carrying any argu­
ment to its logical conclusion . .
The overall progressive nature of the capitalist mode of production is
integrally linked to the uneven development which capitalism generates,
which is inherent in capitalist competition. From the days of Adam
Smith, the political economy schools, both classical and neo-classical,
treated competition as a force that produces harmony among capitals
('firms') . This analytical outcome results from presuming that all capitals
are identical (the 'typical firm'), entering into the competitive process on
equal footing. This assumption presupposes the outcome which the com­
petitive process generates: the tendency for competition to establish a
norm of abstract necessary labour in production. By presupposing the
outcome, the neo-classical political economy school precludes the possi-
1 1
Analysis o underdevelopment
bility of winners and losers in competition. In this framework, competi­
tion processes no outcome at all, but rather a harmonious equilibrium
among rivals all of whom were identical at the outset and remain so to the
This approach, in which capitalist rivals are equals, involves no analy­
sis of competition itself. Indeed, it precludes the very mechanisms by
which the competitive struggle is fought. If al 'firms' are 'price-takers',
there is no role for price competition, product differentiation, advertising,
and the other tactics by which capitalists seek to gain advantage over one
another. 51 To understand the uneven development which capitalism gen­
erates, it is necessary to reconstruct the analysis of competition on the
basis of the law of value. This involves three basic principles: first, that
competition be defined as the movement of capital; second, the integra­
tion of technical change with the movement of capital; and, third, recog­
nition that within sectors of industry the efficiency of production is
unevenly developed. For the analysis of underdevelopment, to these must
be added the contradictory impact of the relations of capital on pre-capi­
talist formations.
A suggested in the discussion of Table 1, the impact of the expansion
of capital on pre-capitalist relations is determined by the interaction of
the form capital takes (commodity, productive, money) , and the nature of
pre-capitalist social relations. The expansion of capital in commodity and
money forms need not generate wage relations which are the basis of
capitalism's dynamism. On the contrary, capitalist trade and fnance can
reinforce the power of pre-capitalist elites, blocking the development of
industrial capital, both by domestic agents and foreign ones. Further, by
expanding international markets for commodities produced in pre-capi­
talist relations, the expansion of commodity and financial capital can
intensify and strengthen systems of unfree labour. In circumstances in
which pre-capitalist relations are transformed to wage relations, the
process is rarely smooth or harmonious. When labour-power is not com­
pletely separated from the land, exchange is a blunt instrument to achieve
reallocation. In this circumstance, force may be required, typically exe­
cuted by the capitalist state, to render pre-capitalist relations vulnerable
to capitalist penetration. The 'civilised' warfare of cheapening commodi­
ties is constructed upon a violent process of dispossessing peasants and
artisans from their means of production. 52
Even in an abstracted society of purely capitalist relations, the expan­
sion of capital is simultaneously destructive and creative. If, in general,
production units within industries vary in unit costs, then it follows that
the movement of capital does not reproduce the average production con­
ditions in an industry, but typically seeks to emulate or surpass the most
efcient operator. Far from establishing a harmonious equilibrium, capi­
talst competition disrupts, eliminates the weak and challenges the strong,
to force upon industry a new standard of efciency and cost. The move­
ment of capital to equalise profts across industries is the process of gen­
erating uneven development: equilibration in exchange (a single price in a
market) hides the generation of uneven development in production. In a
capitalist system, regulated but by capital itself, the frontier between the
'civilised' forms of cheapening commodities, on the one side, and bandit­
ry, fraud, and violence is easily and frequently crossed.
Historical Materialism
While capitalist social relations are progressive in the strict sense of
laying the basis for a revolutionary development of the productive forces,
it does not follow that unregulated capitalism generates this outcome in
all circumstances and regions. For example, the current globalisation of
commodity and financial markets has been associated growth 'miracles'
in East and Southeast Asia, but has reinforced underdeveloped in Africa
south of the Sahara. The latter i as much a part of the dynamism of cap­
italism as the former. Inherent in the progressiveness of capitalism on a
world scale is the simultaneous destructive impact of capitalism in partic­
ular regions.
Further, part of the progressiveness of capitalism arises from the class
struggle, out of which restrictions are placed upon the accumulation of
capita\. The distinction between the absolute and relative extraction of
surplus-value is at one level of analysis purely formal and defnitional. At
the more concrete level, it is a critique of unregulated capitalism. In the
absence of restrictions on capital, the more primitive absolute extraction
is forced upon capitalists by the pressure of competition. Precisely
because the raising of surplus-value relatively is a social process, it must
in part be imposed upon capitalists by limiting their power to raise sur­
plus-value absolutely. Historically, limits were imposed through the legal
restriction of the length of the working day, regulations on working con­
ditions, and prohibition of child labour. The struggle of workers for
workplace rights transforms capitalism from its primitive, repressive stage
of absolute extraction, and forces it to realise its more progressive charac­
ter in which society's surplus labour is increased through the dynamism
of technical change.
Technical appendix
In the section, 'Limits to Accumulation in Transitional Societies', it was argued
that the persistence of pre-capitalist relations tends to limit the accumulation
process. This annex provides a formal proof. At the outset we make a number of
simplifying assumptions. Consider a closed economy in which there are two sec­
tors, one capitalist that produces means of production, and one pre-capitalist that
produces the means of consumption. The capitalist sector does not use the com­
modity of the pre-capitalist sector as a direct input to production. The exchange­
values for each sector can be defned as follows:
+ P2Wllt
+ q)
) (
) ¯
Where a
, a
are the units of commodity
required to produce one unit of
and 2 (greater than zero and less than unity), respectively; w is the
amount of commodity 2 consumed by a worker in a day (greater than zero and
less than unity); n
, n
are the units of labour required to produce one unit of
each commodity (greater than zero); and
is a productivity index (equal to or
greater than unity) . There is no productivity index for the pre-capitalist sector
because it is assumed technologically stagnant relatively to the capitalist sector.
The p's are the exchange-values and the r's the profit rates. Exchange-values and
rates of return are measured in units of labour time.
While the pre-capitalist sector pays wages, the labour there is not 'free' wage
labour, but tied to the dominant rural class through relations of clientage. Land is
Analysis of underevelopment
not a freely vendible commodity, in part because the rural labour is tied to the
land. A a result, capital cannot move from the capitalist sector to the pre-capital­
ist, so there is no tendency for the rates of return to equalise. Each commodity's
exchange-value tends to equal its value. Because of the tendency for exchange-val­
ues to equal values, the following holds (price in each sector equals the labour time
objectified in means of production plus the new or current labour added in pro­
duction) :
l = PI
= P2
PI =
( 1 -a
) and (5)
P2 =
( 1 - a
) ]
If we substitute (5) and (6) into ( 1 ), we obtain (after some manipulation),
1 -
l +
l + (
- a
] ) (7)
l +
l +
(1 - a
The profit rate in the capitalist sector reduces to a function of several parame­
ters: the input and labour coefcients for each sector, the standard of living (w),
and the productivity index of the capitalist sector (
) . Inspection shows that a pro­
ductivity increase results in a decline in the rate of profit, by decreasing the numer­
ator and increasing the denominator. This occurs even though unit costs in the
capitalist sector decline." This decline results from the reduction in the living
labour content of capitalist commodities (the increase of b), in the absence of a
tendency for profit rates to equalise across the two sectors. Were both sectors cap­
italist, the relative surplus-value mechanism explained in the text would ensure that
the equalised rate of proft for both sectors would be higher than prior to the pro­
ductivity increase. 54 In the absence of this equalisation, the exchange-value of the
means of consumption may fall," but less than would be the case with equalisation,
and value is in effect transferred from the capitalist to the pre-capitalist sector.
1 . Indeed, this is the sub-title of volumes 2 and 3 of Capital (,A Critique of
Political Economy'), though not of the frst volume (sub-titled 'A Critical
Analysis of Capitalist Production').
2. See Marx and Engels 1 965, p. 1 92.
3. The passage appears in Capital, volume 1 , chapter 1, in the famous section,
'The Fetishism of Commodities and the Secret Thereof, Marx 1 970, pp. 84-5.
3. 'Marxism' or historical materialism does not argue that social dynamics can be
reduced to economic causes; rather, it argues that social dynamics derive from
the social relations in which the collective reproduction of society is organised.
5. Marx 1974, p. 43.
6. It is beyond the scope of this chapter to explain why value theory implies that
the general equivalent is a commodity (Le. money is produced and has value).
See Weeks 1 981 , chapter IV.
7. Marx called this process 'simple commodity circulation'.
8. In its role as a theory of prices, the theory of value is a theory of prices under
capitalism, when prices include surplus-value, part of which is capitalist profit.
Historical Materialism
The discussion of the division of surplus-value into its phenomenal forms (as it
appears in exchange), profit, interest, rent, and unproductive payments (for
example, salaries of priests and university professors), lies beyond the scope of
this paper.
9. 'Surplus-value' is at this point used in the purely descriptive sense of the differ­
ence between the money that initiates the circuit of capital and the money that
comes at the end.
t. Consider a two-commodity closed economy. The rise in the exchange-value of
one commodity implies a decline for the other. Thus, exchange redistributes
rather than creates surplus-value. A Neo-classical would not contest this theo­
retical argument, but would insist that individua and social welfare is
increased by exchange. In turn, one following the Marxian method would not
contest the neo-classical point, but would judge it to be of little theoretical
1 1 . The neo-classical disagreement would come on two issues: 1 ) the process by
which value is added in production (marginal productivity theory) ; and 2)
what constitutes 'production' (rejecting the distinction between productive and
unproductive labour) .
1 2. In neo-classical political economy, market prices oscillate around long run gen­
eral equilibrium prices. If short run and long run average cost curves have a
unique minimum point, and no 'above-normal' profts are earned in any sector
(perfect competition), then the general equilibrium set of relative prices is
independent of demand. Demand determines the composition of output in
general equilibrium, but not relative prices.
1 3. It is this re-casting that eliminates from the materialist analysis the concept of
1 4. Even as pure description the foregoing is a considerable improvement upon
neo-classical analysis, which treats a capitalist society as the exchange between
individual agents, or even the Keynesian 'circular flow' . For a critique of the
latter, see Weeks 1 989, chapter 1 .
15. If production involves more than one output the value theory is consistent only
in general equilibrium. See Weeks 1989, chapter 1 0 and Fine 1 980, chapter 3.
1 6. It would be more correct to write, 'production of products by means of prod­
ucts' . This indicates the generality of the framework; i.e. it is not limited to a
system of commodity (capitalist) production. When viewed in this way, pro­
duction cannot logically be treated in terms of value-added categories alone,
but must be analysed in terms of the total social product (value added plus
intermediate production) . The Keynesian categories of consumption, invest­
ment, etc. have their analogues in materialist theory, but are not the relevant
categories of analysis. See Weeks 1 983.
1 7. The essence of the production process is the material transformation of objects.
The addition of value is a historicaly specific outcome. For example, a subsis­
tence farmer who plants maize seed does not added value in production;
helshe engages in specific and concrete labour which, if successful, results in
more maize than was planted as seed.
1 8. Neo-classical marginal productivit theory is consistent with fixed coefficients
of production, as demonstrated decades ago in Dorfman, Samuelson and
Solow 1958, chapter 3.
1 9. At a more concrete level of analysis, the distinction between materials of pro­
duction (circulating means of production) and tools of production (fxed
means of production) is crucial. See Weeks 1 981 , chapters 7 and 8. Unfolding
Analysis of underdevelopment
this argument is beyond the scope of this paper.
20. There is debate over the proper method to the aggregate different quantities of
labour. This is discussed below.
21 . Weeks 1 990.
22. Marx 1 974, p. 43. Pursuing the implications of this quotation takes one
through the theory of alienation. A commodity is, among other things, the
product of purposeful human activity. Because it must be exchanged, it pre­
sents itself to human beings as something external to them, created by a
process beyond their control ('the market'). Thus, competition is a process of
alienating people from their labour. This is eloquently explained by Marx 1 971 ,
chapter 50 ('Illusions Created by Competition' ).
23. 1t is essentially social for two reasons. First, within units of production it
involves co-operation among human beings, a social process. Second, no com­
modity is created within one production unit. Every commodity is the result of
the production of many commodities, which serve directly and indirectly as
inputs to it. Thus, the essence of production is production in the aggregate,
wth each individual commodity a constituent part.
24. If producers control their means of production and labour-power is not a com­
modity, then production is essentially isolated, not social. This point is elabo­
rated below. It results in one of the fundamental ironies of a capitalist society:
while the production of commodities alienates workers form their labour, by
the same process it integrates those workers into a social matrix of production.
25.A society without classes is a society without a surplus; i.e. a surplus product is
not basically a technological phenomenon, but emerges from technical possi­
bility to realisation when a class develops to approp
iate it from the direct pro­
26. There is nothing specially Marists about the insight that workers are involved
in simple commodity circulation and capitalist in the circuit of capital. Indeed,
Walt Disney (rather, one of his employees) made the point with succinct clarity
in cartoon strip of Un
cle Scrooge (the richest duck in the world) . Donald
Duck (nephew of Scrooge) receives his paycheque from McDuck Enterprises.
He then goes to a McDuck petrol station to filhis car, to a McDuck super­
market to do his shopping, and pays his rent to McDuck Estate Agents.
Subsequently, Donald goes to his uncle to ask for an advance on his next
week's salary. His rich uncle berates him for lack of thrift and tells him, 'I have
no trouble keeping my money'. Had this run in a left-wing periodical, it would
have been dismissed as communist propaganda.
27. Weeks 1985.
28. Marx used the distinction between absolute and relative surplus-value to divide
capitalism into two great epochs - the age of manufacture and the age of mod­
er industry(see Weeks 1 985). It may or may not be that Marx believed that
extracting surplus-value absolutely became of little importance in the second
epoch. However, at the end of the twentieth century it is clear that capitalists
are constrained in doing so by the strength of the working class, not the logic
of accumulation. The neoliberal ideology of 'flexible' labour markets can be
seen as a theoretical justification to re-introduce the extraction of absolute sur­
plus-value in the advanced capitalist count
29. For surplus-value to rise relatively, technical change must affect either the
inputs into the commodities that workers consume or those commodities
themselves. Technical changes that reduce the labour time in commodities not
consumed by workers ('luxuries') do not affect the value of labour-power. In
Historical Materialism
neo-Ricardian terminology, non-luxuries make up 'basic' commodities.
30. Excessive quotation from Marx can prove tedious, but on this point it is difi­
cult to improve on his clarity: 'This division of a product into a useful thing
and a value becomes practically important only when exchange has acquired
such an extension that useful articles are produced for the purpose of being
exchanged, and their character as values has therefore to be taken into
account, beforehand, during production.' Marx 1 970, p. 78.
3 1. Marx 1971, p. 875.
32. Marx 1 971 , p. 876.
33. The hunter-and-gatherer must assess and give priority to his/her use of time
(Marx called this 'the economy of time') . However, the tool he/she might
choose to produce in place of immediate consumption is not capital.
34. Marx 1 974, p. 670.
35. Quite the contrary, the civilised competition in markets produces irresolvable
tensions which develop into open warfare. Both World War I and World War
II began as conflicts between capitalist countries. The former remained so
throughout, while the character of the second became more complicated when
Nazi Germany invaded the Soviet Union. The political instability that results
from competition in markets was pursued by Marxists within the theory of
imperialism, which until after World War II focused primarily on intra-capital­
ist rivalries. The relationship between advanced capitalist and underdeveloped
countries involves the analysis of the interaction of modes of production, dis­
cussed below.
36. There is a second, better known, sense in which Marx considered capitalism
progressive. Especially in his early writings (see the Communist Manifesto) he
argued that capitalism is progressive because it creates the proletariat, which
will be the historical vehicle for the overthrow of capitalism and exploitation in
general. Consideration of this argument lies beyond the scope of this paper.
37. The term 'free' wage labour refers to the specific character of social labour
under capitalism: 'Free labourers, in the double sense that neither they them­
selves form part and parcel of the means of production, as in the case of slaves
. . . nor do the means of production belong to them . . . they are, therefore, free
from ... any means of production of their own.' Marx 1 974, p. 668.
38.A quite interesting development of this point, with concrete examples, is found
in Chapter 20 of volume III of Capital, called 'Facts about Merchant's Capital'.
For example, ' . . . Of course, commerce will have more or less of a [disintegrat­
ing] effect on the communities between which it is carried on. It will subordi­
nate production more and more to exchange-value ... thereby it dissolves the
old relationships ... Nevertheless this disintegrating effect depends very much
on the nature of the producing community.' Marx 1 971 , p. 330.
39. He quotes a writer named Ferando Galiani as asserting 'value is a relation
between persons', and comments, 'he ought to have added: a relation between
persons expressed as a relation between things', Marx 1 974, p. 79, footnote.
40. Marx 1 974, p. 80.
41 . The conversion of labour-power and the natural environment into capitalist
commodities Marx called 'primitive accumulation' . See Marx 1 974, chapter
26, 'The Secret of Primitive Accumulation' .
42. Summarised in Kay 1 989, chapter 2
43. A similar effect occurs in the famous Lewis model of 'economic development
with unlimited supplies of labour'. In that model, the absence of productivity
change in the pre-capitalist sector results in a relative, then absolute, decline in
Analysis of underdeelopment
food production as 'under-employed' labour is transferred to the capitalist sec­
44. On the condition, among other things, that the difference in the standard of liv­
ing is not cancelled by an equal are larger difference in productivity.
45. This process is demonstrated in a formal model in the appendix.
46. 'Autonomous' i the sense that more output is achieved from a given input of
labour-power and means of production.
47. Formally, this is equivalent to the oft -demonstrated result that technical change
which raises the ratio of means of production to labour-power (what Marx
called the technical composition of capital) will result in a decline in the aver­
age rate of profit i the value of labour-power remains unchanged. In current
discussion, the decline does not result from assuming a constant value of
labour-power, but from the relatively lower rate of productivity growth in the
pre-capitalist sector.
48. Land reform in Chile was initiated in the early 1960s, then pursued with vigour
by the Christian Democratic government (1964-1970) and the
Socialist/Communist government ( 1970-1 973). After the military coup in
1 973, the right-wing government continued agrarian modernisation through
capitalist farming.
49. This conclusion is in direct contrast to that of dependency theory. The latter
concludes that countries are underdeveloped because of capitalist penetration,
while materialist theory concludes that underdevelopment reflects the incom­
plete dominance of capitalist relations over pre-capitalist ones.
50. See Warren 1 980.
51 . A more detailed critique of neo-classical competition can be found in Weeks
1 994.
52. Marx termed this process so-called primitive accumulation: 'The so-called
primitive accumulation, therefore, is nothing else than the historical process of
divorcing the producer form the means of production ... [T]he history of prim­
itive accumulation . . . [are] those moments when great masses of men [sic!
people] are suddenly and forcibly torn from their means of subsistence, and
hurled as free and 'unattached' proletarians on the labour-market.' Marx 1 974,
pp. 668, 669.
53.Assume that initially b equals unity. Then, cost per unit of output ('cost price'
was Marx's term) is, pial
p2wnl ; for b > I, unit cost is [plat
(p2wnl/b)] < [pIal
p2wnl] . The proft rate falls even though unit costs
are lower.
54. This is the 'Okishio Theorem': any technical change which reduces the cost
(cost price) of a commodity at prevailing prices wl result in a rise in the sys­
tem-wide average rate of profit after all prices have adjusted to the new values
generated by the new technology matrix. See Okishio 1 961 .
55. This is because pl a2 falls when pi declines.
Dorfman, R., P. Samuelson and R. Solow 1 958, Linear Programming
and Economic Analysis, New York: McGraw-HilL
Fine, Ben 1 980, Economic Theor and Ideology, London: Edward
Ky, Cristobal 1 989, Latin American Theories of Underdevelopment,
London: Routledge.
Historical Materialism
Marx, Karl 1970, A Contribution to the Critique of Political Economy,
Moscow: Progress.
Marx, Karl 1 971 , Capital, volume III, London: Lawrence & Wishart.
Marx, Karl 1 974, Capital, volume I, London: Lawrence & Wishart.
Marx, Karl and Friedrich Engels 1 965, Selected Correspondence,
Moscow: Progress.
Okishio, Nobuo 1 961 , 'Technical Change and the Rate of Proft', Kobe
University Economic Review, 7.
Warren, Bil 1 980, Imperialism: Pioneer of Capitalism, London: Verso.
Weeks, John 1 981 , Capital and Exploitation, Princeton: Princeton
University Press.
Weeks, John 1 983, 'On the Issue of Capitalist Circulation and the
Concepts Appropriate to its Analysis', Science and Societ, XVII: 2.
Weeks, John 1 985, 'Epochs of Capitalism and the Progressiveness of
Capital's Expansion', Science and Society, XI.
Weeks, John 1 990, 'Abstract Labour and Commodity Production',
Research in Political Economy, volume 1 2, edited by Paul Zarembka,
Greenwich, Conn: JAl Press.
Weeks, John 1 994, 'Fallacies of Competition: Myths and Maladjustment
in the "Third World", An Inaugural Lecture delivered on 1 3 October
1 993, London: SOAS.
The neoclassical and Marian
theories of technolog:
a comparison and
critical assessment
Tony Smith
Neoclassical economics remains the leading theoretical alternative to
Marxian economics. In this article I shall contrast the accounts of techni­
cal change in capitalism proposed by both theories. I shall introduce five
criteria relevant to a comparison of competing social theories, and argue
that the Marxian perspective on technical change in capitalism [ is superi­
or on all five counts.
A Technolog and neoclassical economics
The first task of the neoclassical theory of technology is to explain the
selection of a specific technique from among the variety of techniques
available on a given level of technology. The production function provides
the basic framework for this explanation. [
In the simplest neoclassical model a firm produces a single homoge­
neous output. To produce this output it makes use of two sorts of inputs,
two 'factors of production': capital ('K') and labour ('L') . A given quanti­
ty of output ('0') can be produced with a variety of different techniques.
In some techniques relatively more capital is employed and less labour,
while in others the reverse holds. This state of affairs can be formulated
mathematically by the production function O=f(K,L), depicted graphi­
cally by an isoquant curve:
A technology can be defned as the set of all the techniques represent­
ed by points on this curve.
Why, then, is one technique selected over all the others in a given pro-
Historical Materialism
duction function? To answer this question neoclassical economists intro­
duce the concept of the isocost line, derived as follows. The costs a firm
must pay (C) equals the amount of labour used (L) multiplied by the
price of labour (pI), plus the amount of capital used (K) multiplied by the
price of capital (pk) : C =Lpl+Kpk. This equation can be rewritten as
K=(-p1!pk) L+C/pk, a straight line. With the price of labour and capital
given, we can vary the total cost to the frm by varying the quantity of
labour and capital purchased. This is depicted graphically by a family of
parallel lines, for example:
Graph 2
Let us assume that a frm wishes to produce Q units of some product,
estimating that this is the amount the particular market can absorb. It
obviously will not purchase inputs at the levels depicted in C 1 , for then it
will not be able to produce the desired output. This output can be pro­
duced if capital and labour are purchased in the amounts depicted by
points a and b on line C3, representing a capital-intensive technique and
a labour-intensive technique, respectively. But it would not be rational for
the frm to purchase the amount of capital and labour required by these
techniques. The frm will be able to produce the same level of output at
lower cost if it selects the technique depicted by the point of tangency
between the isoquant and isocost line C2. In the given situation this tech­
nique is superior to all other possible choices as well.
The slope of the production function expresses the amount of one fac­
tor of production that must be added to compensate for the loss of a unit
of the other input, keeping the quantity of output fxed. More precisely, it
is defned as the negative of the ratio of the marginal productivity of
labour to the marginal productivity of capital, -Mp1!Mpk. At the point of
tangency this ratio will be equal to the slope of the isocost line, -p1!pk.
Transforming the equation Mp1!Mpk=p1!pk to Mp1!pl=Mpk/pk pro­
vides further insight into why the technique represented by this point of
tangency will be selected by the rational frm. Mp1!pl can be interpreted
as the increase in quantity of output r�sulting from the purchase of an
additional unit of labour at price pI. Suppose this is less than the increase
Teres of tehnoios
in output resulting from the purchase of an additional unit of capital at
price pk. In this case it would be rational for the firm to adopt a more
capital-intensive technique, for it would then be able to produce the same
output at a lower cost. By symmetrical reasoning, it would be rational for
the frm to shif to a more labour-intensive technique if Mpl/pl exceeded
Mpk/pk. The eventual result of these shifts will be the selection of the
technique represented by the point where the two ratios are equal, and
this is the point of tangency.
Of course, if the ratio of factor prices changes, the point of tangency
will shift as well. When this occurs a rational agent will select a new tech­
represented by the new point of tangency. This new technique will
either be more labour-intensive or more capital-intensive than its prede­
cessor, reflecting the phenomenon referred to in neoclassical theory as
'factor substitution' .
If the first task of the neoclassical theory of technology is to explain
the selection of a particular technique, the second is to account for tech­
nical change, that is, the shift from one level of technology to another,
depicted by a new production function:
Rtional agents will choose to move to a new technology only if it
allows more effcient production. In other words, the techniques repre­
sented by the new production function must produce the same quantity
of output as the techniques of the initial production function while
employing fewer inputs. This is captured graphically by drawing PF2
closer to the origin than the original production function, PF1 .
At this point the concept of a 'bias' in technological change must be
introduced. Starting from a given production function, technological
advances can take a number of different directions, as illustated i graph 4
If we move from the initial production function to PF3, capital and
labour are used more efficiently, but in the same proportions. This is
called neutral technological change. If we move to PF2 or PF4 the pro-
Historical Materialism
portions shift. We have a more capital-intensive technology and a more
labour-intensive technology, respectively.
What determines whether technological change is neutral, biased in a
capital-saving direction, or biased in a labour-saving direction? The stan­
dard answer to this question was provided by John Hicks: if the state of
scientific-technical knowledge advances to the point where neutral,
labour-saving, or capital-saving technological change are all possible, the
firm's choice will depend on existing trends in factor prices. If the prices
of both inputs are fairly stable, or if these prices are both falling or rising
in sync, then neutral technological change will occur. If the cost of capital
is rising significantly relatively to the cost of labour, then the firm will
select a capital-saving technology. If the opposite holds, a labour-saving
technology will be chosen. In Hicks' s view this last case has been the
dominant tendency:
The real reason for the predominance of labour-saving inventions is surely that
which was hinted at in our discussion of substitution. A change in the relative
prices of the factors of production is itself a spur to invention, and to invention
of a particular kind - directed to economising the use of a factor which has
become relatively more expensive. The general tendency to a more rapid
increase of capital than labour which has marked European history during the
last few centuries has naturally provided a stimulus to labour-saving invention.'
In other words, the direction of technological change is explained by a
change in the ratio of factor prices, the same phenomenon that explains
factor substitution on a given level of technology.
There are good reasons to believe that this standard account of techno­
logical change is incompatible with the logic of neoclassical economics. It
is true that within this framework an increase in factor costs may induce a
search for a new technology that will lower costs. But rational neoclassical
agents will be concerned only with the sum total of costs. Any new tech­
nology that signifcantly lowers total costs will be welcome, even one that
1 1161
Tere of tehnol
employs relatively more of the factor whose price is increasing:
If [Hick's] theory implies that dearer labour stimulates the search for new
knowledge aimed specifically at saving labour, then it is open to serious objec­
tions. The entrepreneur is interested in reducing costs in total, not particular
costs such as labour costs or capitals costs. When labour costs rise any advance
that reduces total cost is welcome, and whether this is achieved by saving
labour or capital is irrelevant. There is no reason to assume that attention
should b concentrated on labour-saving techniques, unless, because of some
inherent characteristic of technology, labour-saving knowledge is easier to
acquire than capital-saving knowledge.'
The neoclassical approach can be elaborated much further, resulting
in vast edifices of mathematical sophistication. The above account, how­
ever, is suficient for our purposes.
B. Technolog and Marian economics
Marx's theory in Capital has a quite different logic
l structure from neo­
classical economics. Neoclassical economists use the precepts of formal
logic to deduce conclusions from given axioms. Marx, in contrast,
attempted to reconstruct the capitalist mode of production in thought by
moving step-by-step from simple and abstract economic categories to
ever more complex and concrete categories. This is not the place to
explore the methodology of this unique type of theory, termed 'systemat­
ic dialectics'" A brief sketch of certain parts of the theory of importance
to the question of technical change must sufce.
In the beginning of volume I, Marx examined the 'cell form' of the
system of generalised commodity production and exchange. He distin­
guished 'concrete labour', privately undertaken labour that may or may
not prove to be socially necessary, from 'abstract labour', defined as pri­
vately undertaken labour that proves its social necessity through produc­
ing commodities that are successfully exchanged for money. Marx then
noted that commodity production and exchange is generalised only
when labour- power has itself become a commodity that can be pur­
chased by capital. The purchase of labour-power, and the settng of it to
work producing commodities with economic value, thus occur within
the context of the capital circuit. This circuit begins with the initial
money capital (M) invested in commodity inputs (C) , these inputs falling
into the two categories of means of production and labour-power (pur­
chased with constant capital and variable capital, respectively) . The
result of the production process (P) is then a set of commodity outputs
(C), which the owners and controllers of capital hope can be sold for a
profit, that is, a sum of money (M ') exceeding the initial M invested.
When this occurs the valorisation process is complete, and capital may
be accumulated.
The central question of volume I can be posed as follows: if all com­
modities exchange at their value, that is, if there are no 'rip-offs' i the
process of exchange - or if 'rip-offs' cancel each other out, the gain of
one party being matched by the loss of another - how can we explain the
existence of net profts in the capitalist system? Marx's answer is that
the ultimate source of profits is the exploitation of wage-labour; while
I s
Historical Materialism
labourers may receive wages that correspond to the value of their
labour-power, they produce more economic value (,surplus-value') than
they receive back in the form of wages.
This theory is to be applied on the level of the total social capital. It
does not necessarily hold for any given unit of capital. A given unit of
capital may go bankrupt and not receive any profits, no matter how
much labour has been undertaken by its workers. Also, as we discover in
volume III, next to no proftable units enjoy profts directly proportional
to the surplus-value produced by their own workers. But in the system
as a whole the profts distributed among the various factions of the capi­
talist class are produced by the surplus labour of the working class. The
most basic social relation in capitalism is the capital/wage-labour rela­
tion, and this relation is necessarily antagonistic, being based on
This is the very heart of Marx's theory. From the standpoint of capi­
tal, capital is the only 'subject' of the valorisation process; wage-labour­
ers are simply one specifc form capital takes in that process, the form of
variable capital. According to Marx's theory of exploitation, however, it
is the working class that is the true subject of economic life. 'Capital' is
nothing but a form taken by the surplus-value produced by the collective
working class, a form which has insanely become an alien power sub­
suming real human subjects under its imperatives so that ' [t]he produc­
tive forces of social labour appear as inherent characteristics of capital."
Marx derived a necessary tendency to technical change in capitalism
from this theory of exploitation. The owners and controllers of capital
necessarily tend to introduce innovations that decrease the amount of
time workers engage in necessary labour (the labour necessary to pro­
duce an amount of economic value equivalent to the wages they receive),
while increasing the amount of time they spend in surplus labour, which
produces the surplus-value appropriated in the form of profits. There is
thus an inherent tendency in capitalism to introduce machinery with the
potential to increase labour productivity. There is also a tendency to
introduce technologies that allow a less skilled - and thus a less expen­
sive - workforce to be employed. Further, there is a necessary tendency
to seek innovations that restructure the labour process so as to lessen the
'pores' in the working day. This increases the intensity of labour, so that
more surplus-value can be produced in a given period of time even if
there have been no advances in labour productivity. Finally, there is a
tendency to seek technologies that enhance capitai's control over the
production process in order to reduce waste, lessen the opportunities for
sabotage, and so on.
How these various tendencies interact in given socio-historical con­
texts is a complex and contingent matter. Technologies that raise the
level of labour productivity may sometimes require a higher overall level
of skill in the workforce for an extended period in certain regions and
sectors. If so, the drive for productivity advances may dominate the ten­
dency to seek deskilling technologies in some circumstances; in others, it
may be dominated by that other tendency. To take another example, in
certain contexts innovations decreasing the pores in the working day
might overlap with innovations that increase the control over the labour
process enjoyed by the representatives of capital. In other contexts, how-
Teres of tenolg
ever, the pores might best be reduced if these representatives relin­
quished a certain degree of controJ.6 Whatever permutations may occur,
however, one thing remains constant: in all these cases technology is
employed as a powerful weapon in capital's class struggle against labour.
Labour, however, is hardly passive in the process of technical change.
Technological change demands more than technical blueprints; the blue­
prints must actually be implemented in the concrete material conditions
of the shopfloor or ofce. This inevitably requires the active participation
of the labour force,7 thereby creating a space within which it is possible
to renegotiate the terms of technical change. Depending upon the
strength of labour organisations and the external economic and political
environment, such renegotiations may shift the social dynamic of techni­
cal change in favour of the interests of workers to a greater or lesser
degree. In the limit case workers may constitute themselves as revolu­
tionary subjects within this space, radically challenging the workings of
society as a whole, and not just this or that workplace practice."
However central class struggle at the point of production may be to
Marxian theory, it remains just one moment in the circuit of capital,9 A
main thesis of the second volume of Capital is that while the production
of surplus-value is the ultimate source of capital, it is not the only factor
affecting capital accumulation. The more time capital is tied up in the
various stages of the capital circuit, the less capital can be accumulated
in a given period of time, everything else remaining equal. 10 It follows
directly that there necessarily is a tendency in capitalism for technical
innovations to be introduced that compress the turnover time of capital.
In the first two volumes of his masterwork Marx abstracted from fea­
tures that distinguish one sector of industrial capital from another. In
the beginning of volume III Marx moved from this level of 'capital in
general' to the more concrete and complex level of 'many capitals'. In his
account of the first category on this level, 'cost prices', Marx noted that
different sectors have different technical compositions; some sectors
require more means of production for a given number of workers than
others. These differences are generally correlated with differences in the
organic composition of capital, that is, in the ratio of constant capital
invested in the purchase of means of production to variable capital used
to purchase labour-power. Marx noted further that profts are not just a
function of the variable capital invested and the surplus-value produced
in a given sector. Profts are also a function of constant capital expendi­
tures. It follows that there is, in the capitalist mode of production, a nec­
essary tendency to seek technical innovations that reduce the costs of
constant capital. This conclusion is reinforced later in volume III in the
discussion of counter-tendencies to the falling rate of profit, where inno­
vations lowering constant capital costs are considered alongside innova­
tions increasing surplus-value.
With respect to industrial capital, the transition from 'capital in gen­
eral' to 'many capitals' is completed with the discussion of stratification
within sectors in chapter 10 of volume III. Here Marx discussed how the
most productive individual units of capital within a sector are able to win
surplus profts. Surplus profits can also result from product innovations
that meet a new want or need, or that satisfy an old want or need in a
new way. In Marx's view, surplus profts involve a redistribution of sur-
Historical MateriaUsm
plus-value from firms with relatively low levels of productivity and out­
moded products to more eficient firms with ' state of the art' products.
Intercapital competition, which necessarily tends to generate a drive to
appropriate surplus profits, thus necessarily tends to result in technical
change as well.
In the remainder of volume III Marx introduced the categories of
merchant capital, financial capital, and rent. Marx also planned to write
subsequent volumes on the state, foreign trade, and the world market.
Although these topics are all relevant to the Marxian theory of technical
change in capitalism, I shall conclude my presentation here. I believe
that the above account, compressed as it is, is suficient to allow a criti­
cal comparison of the Marxian and the neoclassical perspectives on the
issue at hand.
C. Critical assessment of the competing theories
The neoclassical and Marxian theories of technology could hardly be
more different. And yet both claim to provide a framework for explain­
ing technology in capitalism. Is it possible to assess these competing
It has been fashionable in recent years to say that in cases of this sort
the theories are incommensurable and cannot be compared rationally.
All we can do is opt for one 'entry point' rather than another," or pick
one or another view to defend rhetorically.12 I believe that there is no
good reason to rule out a priori the possibility that rational argumenta­
tion can establish that one theory is more sound than another, at least in
some cases. Of course, there are no guarantees this will occur. But the
radical scepticism of postmodernism and rhetoricism dogmatically rule
this out, and therefore should be avoided, at least initially.
And so the question arises once again: how should an assessment of
these sorts of theories proceed? Much more is involved than simply veri­
fying or falsifying theories separately. The comparative component is
essential: how well does each theory measure up to its main competi­
tors? A corroborated account should still be rejected in the face of a
stronger theoretical viewpoint; a perspective that has been falsified in
certain respects can still be maintained if there are no satisfactory alter­
Building upon Lakatos' s work on the methodology of scientific
research programs, I believe that there are fve criteria relevant to a com­
parison of competing theories: i) their explanatory scope; ii) the internal
consistency of their foundational categories; iii) the status of the abstrac­
tions they employ; iv) their compatibility with social practice; and v)
their empirical accuracy. I3 One of the most vexing questions in the phi­
losophy of the social sciences is how to weigh these different criteria in
cases where they lead to conflicting assessments. I shall postpone that
issue for now, and examine the neoclassical and the Marxian theories of
technology with reference to each of these yardsticks.
i) Scope of phenomena explained
A frst criterion that can be used in comparisons of competing theories
has to do with the range of empirical phenomena brought within the
scope of the relevant theories. This is a distinct matter from a considera-
Teres of tehnolol
tion of the extent to which the competing theories have been confirmed
or falsified by empirical evidence, a topic that will be taken u
The neoclassical approach to technology concentrates on the drive to
keep the cost of factor inputs as low as possible for a given level of out­
put. This underlies the selection of the technique represented by the
point of tangency between the production function and an isocost line,
rather than some other point of intersection between the production
function and an isocost line. And neoclassical theory emphasises the
drive to produce a given output more efciently, which accounts for the
shift from a given production function to one closer to the origin.
Further, neoclassical theory explicitly acknowledges that both processes
make use of scientific knowledge as a free good.
The Marxian theory of technology provides explanations for the sorts
of things neoclassical theory addresses. We can derive from Marx's
account of the circuit of capital accumulation a necessary tendency for
innovations to be sought that lower wage costs and that lower capital
costs (including the costs of storage and waste) . We can also derive a
drive to introduce process innovations from the imperative to increase
the surplus-value produced by (and the surplus profts distributed to) an
enterprise. And Marx explicitly and repeatedly noted the importance of
science to the innovation process in modern capitalism. 1 4
The reverse does not hold. The neoclassical theory of technology
does not address the entire range of phenomena with which Marxian
theory is concerned. The most significant reason for this is that neoclas­
sical theory reduces all questions of technology to the rational response
of economic agents to demand, factor prices, and given scientifc-techni­
cal knowledge. For the Marxist, technological development in capitalism
involves much more than this. Specifically, there are four areas where
the scope of the Marxian theory of technology goes beyond that of neo­
classical economics.
For the neoclassical economist labour is considered solely as a 'fac­
tor' of production, an object that can be purchased. For the Marxist the
working class is a collective subject, albeit one treated as a commodity
under the capital form. Workers have the capacity to engage in 'learning
by doing', unlike other so-called factors of production. To quote a key
[Alll these economies [in the use of constant capital] arising from the concen­
tration of means of production and their employment on a massive scale, pre­
suppose . . . the social combination of labour . . . Even the constant improve­
ments that are possible and necessary arise solely from the social experiences
and observations that are made possible and promoted by the large-scale pro­
duction of the combined collective worker. 15
There are a number of reasons why learning by doing is signifcant. For
one thing, it calls into question one of the central assumptions of the
production function approach, the assumption that it is possible to con­
sider the marginal productivity of labour and the marginal productivity
of capital separately. The phenomenon of learning by doing implies that
the productivity of capital goods is a function of the capacities of the
workforce. Also, the incremental innovations arising from the learning
Histrical Materialism
by doing process are the most profound sources of productivity
advances in the long term. 1 6 Most importantly, wage-labourers alone
have the capacity to resist the social power of capital. Machines and raw
materials do not ask for wage increases, benefits, or breaks during the
working day; they do not band together into organisations, or engage in
slow-downs and strikes; and they do not possess the potential to call into
question the legitimation of a social order in which those making deci­
sions are not accountable to the vast majority of those affected by these
decisions. These phenomena affect the development of technology. The
owners and controllers of capital regularly seek innovations that keep
social relations on terms favourable to themselves, while labourers con­
tinually attempt to modify these innovations. A a result, the nature of
technical change in production cannot be concretely grasped in abstrac­
tion from the dynamic of class relations.1 7 This crucial feature of tech­
nology in capitalism does not come within the scope of neoclassical the­
Second, the production function approach emphasises process inno­
vations at the cost of neglecting product innovations. It do
s not address
the ceaseless development of new needs that is part of the dynamic of
the capital form. 1 8 In contrast, this is a central part of the Marxian story
of technology in capitalsm. For Marx, 'one precondition for the sale [of]
the commodity [is] that the commodity should have use-value, and thus
satisfy a social need' . 19 As capital accumulation expands, new social
needs arise, as well as new products to meet them: 'The discovery, cre­
ation and satisfaction of new needs arising from society itself . . . is like­
wise a condition of production founded on capital.'20
Third, neoclassical economics abstracts from time. This is obvious
with respect to static equilibrium theory. It holds for dynamic equilibri­
um theory as well; a theoretical comparison of one time-slce with
another time-slice is not a theory of the temporal process connecting
them. A a result of its atemporal nature, neoclassical theory neglects the
importance of reducing circulation time, a central issue in Marxian theo­
ry. And in so far as the question of circulation time is ineluctably con­
nected with spatial issues, neoclassical theory can be criticised for
neglecting space as well.21
Finally, in neoclassical theory there is no recognition that the intro­
duction of technologies is generally connected with negative externali­
ties, that is, burdens imposed on workers and communities. Social
movements and state regulations necessarily tend to arise in response to
capital's externalisation of the social costs of technical change. These
movements and regulations profoundly shape the subsequent direction
of technical change. Marxian theory includes these phenomena within its
explanatory scope,22 while neoclassical theory does not.
The scope of the Marxian theory of technical change in capitalism
incorporates the phenomena considered by neoclassical theory. Marxian
theory also considers matters that are beyond the scope of neoclassical
theory. Of course, this in itself does not resolve the comparison of the
competing theories. We must investigate further to discover whether
neoclassical theory might has sufcient advantages in other respects to
compensate for its defciencies here.
Teores of tehnolos
ii) I nternal consistenc of foundational categories
A second way to assess competing theories is to ask whether their foun­
dational categories can be stated in a consistent fashion. The production
function framework for explaining technology has a number of serious
difculties in this regard. Each point of this function is defned by a cap­
ital co-ordinate (K) and a labour co-ordinate ('L' ). What is this K? It is
a number representing the aggregate of all the non-labour inputs in pro­
duction measured in physical terms. But it is conceptually impossible to
add x tons of raw materials to y machines and z factories and come up
with a number that means anything.
Neoclassical economists have introduced the distinction between cap­
ital as clay and capital as putty in order to deal with this difficulty."
When capital takes on distinct forms that cannot be reduced to a com­
mon framework (raw materials, machines, factories), it is considered as
hardened 'clay'. When it is aggregated together in the production func­
tion it is considered as 'putty' that can take on any number of different
forms while remaining identical. Assuming that the same thing at the
same time has both the indefinite malleability of putty and the fixed form
of hardened clay does not remove the conceptual incoherence here. The
incoherence is merely institutionalised in terminology.24
In the Marxian framework this incoherence in the treatment of capital
goods is avoided. Raw materials, machinery, and so on, can be aggregat­
ed together under the money form as the total sum of money capital
invested in constant capital. But when so-called capital goods are con­
sidered in physical (use-value) terms they are disaggregated into differ­
ent means of production.
When we turn to the role of labour-power, here too the charge of
incoherence can be brought against the neoclassical framework. The 'L'
co-ordinate of a point on the production function assigns a single num­
ber to an aggregate of incommensurable physical and mental activities,
and this cannot be done in a meaningful way. This time, however, it
appears that Marxian theory suffers from the precisely the same sort of
problem. In passages devoted to the labour theory of value Marx wrote
of a reduction of complex labours to simple labours, and of the way in
which simple homogeneous labour provides a measure for the value
embodied in commodities. This too appears to assign a single number to
an aggregate of incommensurable physical and mental activities.
Despite this surface similarity I believe that the two theories are still
in a quite different position. Marx's theory of value shows that money is
the alpha and omega of the valorisation process.25 Money received after
the successful sale of some commodity, not some simple and homoge­
neous labour time pre-existing sale, provides the only socially objective
measure of abstract labour. While different concrete labours cannot be
made commensurate, the problem thus does not arise for abstract
labours, which are by defnition commensurable in money terms. Talk of
simple and homogeneous labour 'embodied' in commodities and mea­
suring their value is a residue of Ricardian theory that Marx never fully
overcame.26 Once this residue is overcome, the incoherence of treating
incommensurable labours as if they were commensurable disappears
from the foundations of Marxism.
The balance sheet must now be drawn. There is an irreducible inco-
Historical Materialism
herence at the foundation of the neoclassical perspective on technology.
In contrast, the incoherence at the foundations of Marxism can be
removed by purging Marxism of its Ricardian residues. We must con­
clude that Marxian theory is superior according to this second criterion
for evaluating competing theories.
iii) Status of abstracions employed
All theories make use of abstractions. Neoclassical theory and Marxian
theory are no exceptions. But there are different sorts of abstraction, and
some sorts are more appropriate for certain theoretical purposes than
others. If a theory employs a type of abstraction that is not suited to
attaining its objectives, this surely must count against it. We can begin
with an examination of three abstractions at the heart of neoclassical
a) Mathematically, there are an infnite number of points on the iso­
quant representing the production function. When the isoquant is used
to represent technological reality, the assumption is made that an infnite
number of possible techniques are accessible, that is, that an infinite
number of combinations of capital and labour can produce the desired
output at the given level of technology. This abstracts from the fact that
in reality the set of available techniques from which capitalist enterprises
must select is always restricted.
b) Neoclassical economists also assume that frms have perfect infor­
mation regarding all of the techniques represented by the production
function. But even in a world of static technology, no frm actually pos­
sesses anything approaching perfect information regarding all existing
techniques. The dynamism of technical change in capitalism makes the
assumption of perfect information even more untenable. The results of
research and development are inherently unpredictable. Firms must
make decisions regarding technical change in situations of fundamental
c) In neoclassical economics, proft maximisation motivates techno­
logical decisions. The technique corresponding to the point of tangency
is selected because it minimises costs and maximises profts. Once we
recognise that the assumption of perfect information abstracts from the
concrete reality, we must grant that the assumption of optimising behav­
iour does so as well. A given frm has reliable information regarding only
a subset of all possible techniques. Any attempt to expand that subset
involves costs, and it can never be known beforehand whether those
costs will be recouped. At some point the search must cease. But if the
firm's decision is thus based on only a subset of all possible techniques,
then it can never know whether its selection truly minimises costs and
maximises profts. It must be content to 'satisfice'.28
Neoclassical economists themselves readily admit that the assump­
tions of infnite techniques, perfect information, and maximising behav­
iour abstract from features of concrete technological activity in capitalist
markets. But they still insist that neoclassical theory generates results
that illuminate this concrete reality. Neoclassical models, they claim, are
'ideal-types' that capture the intelligibility of the world even if they are
too pure to provide literal descriptions of the world.
The most straightforward Marxian response is simply to deny the
claim. The three abstractions just considered are not abstractions of the
workings of capitalism; they are abstractions fom the mechanisms at
work in this mode of production. They distort, rather than illuminate,
the state of affairs under investigation.
A second response is a bit more involved. It refers to the tension
between the avowed theoretical purpose of the above abstractions and
their actual nature. These abstractions are supposed to contribute to the
scientifc explanation of technological activity. But the assumption of an
infinite number of techniques, perfect information, and maximising
behaviour are actually used by neoclassical economists to define what
they believe ideally rational agents would do in ideal situations. In other
words, these abstractions are used to defne a normative model of behav­
iour.2• I is in principle legitimate for social theorists to abstract from
concrete reality in order to construct a normative model. 30 But normative
assumptions do not provide an appropriate framework for the explana­
tion of empirical phenomena. And so the sorts of abstractions employed
by neoclassical economics are not suited to the theoretical purpose neo­
classical economists set for themselves.
The abstractions employed in Marxism are of a quite different sort.
In the beginning stages of Marx's theory, abstraction is made from cer­
tain complex social forms (fnance capital, the state, etc.) so that more
elementary social forms (for example, the commodity form, the money
form) can be considered in themselves. But these abstractions are 'real
abstractions' . Even the most abstract social forms in Marx's theory are
instituted in the material practices of capitalist society. And these forms
define tendencies that continue to operate in concrete instances of capi­
Marx' s ultimate theoretical purpose was to reconstruct the capitalist
mode of production in thought, starting from relatively simple social
forms and progressing step-by-step to ever-more complex forms. The
sorts of abstractions he employed are fuly congruent with his theoretical
purpose. The abstractions of the neoclassical economics, in contrast,
incoherently waver between normative idealisations and empirical
claims. This provides a third reason to consider Marxian theory superior
to its neoclassical competitor.
4. Compatibility with relevant social practices
In general, a theory'S soundness cannot be assessed in terms of its
capacity to orient practical activity. Quite mistaken views have guided
people's actions for extended periods of time. But if a theory proves
incompatible with the very type of social practice it claims to address,
this is quite signifcant. Neoclassical theory certainly informs the practi­
cal life of many on the political Right, legitimating their adherence to
capitalist market society. But it is astonishing how little practical rele­
vance the neoclassical theory of technology in capitalism has to techno­
logical activity in capitlism.
Let us consider the area of technological activity where we would
expect neoclassical theory to be most at home, the selection of tech­
niques by managers of capitalist enterprises. Neoclassical economists
assume that market constraints and the given level of technology sufce
Histrical Matrialism
to determine the selection of techniques. This view completely ignores
the role of strategic decision-making in a world of profound uncertainty
and dynamic competition.3
Examples of the sort of strategic decisions
regarding technology that must be made include: setting the size of the
Research and Development budget, allocating R&D resources among
the different divisions of the firm, establishing the general technical
objectives to be pursued by the firm, allocating resources to particular
projects, and deciding whether to continue or to conclude R&D projects
that are underway. Not a single one of these decisions follows automati­
cally from market demand, factor prices, and the given level of
technology. Among other considerations, these sorts of decisions depend
upon whether the management of a firm wishes to protect its existing
market, expand its share of an existing market, enter a market in which
it does not now participate, or create a new market that does not
presently exist. This in tr depends upon estimates of the future behav­
iour of labourers, consumers, state ofcials, competitors, etc., that are in
principle uncertain. Neoclassical theory is therefore almost completely
irrelevant to the technological activity of managers, and almost com­
pletely ignored by management theorists examining that activity.32
If we turn to the area of technological activity where we would expect
Marxian theory to be most at home, the response of workers to techno­
logical change, there is a clear contrast. A extended series of practical
maxims follows directly from Marxian theory, three examples of which
must sufce here: 'Do not accept uncritically utopian claims regarding
the social effects of new technologies'; 'Investigate how new technolo­
gies might shif the balance of power between capital and labour'; 'When
struggling to shif the direction of technological change so that it better
refects workers' interests, attempt to make as broad an alliance as possi­
ble among the different sectors of the working class.' Marxian theory, in
brief, emphasises the practical necessity for the working-class and other
social agents to formulate a strategic response to decisions regarding
technology made by the owners and controllers of capital.
Needless to say, this does not imply that every practical recommenda­
tion ever uttered by a Marxist has been successful in practice. Nor does
it mean that there have never been neoclassical economists capable of
giving sound pragmatic advice to the economic agents they choose to
address. But the above discussion does suggest that in principle neoclas­
sical theory is incompatible with the very technological practices most
relevant to it in a way that Marxian theory is not.
We still are not in a position to offer a decisive judgement between
the two competing theories. If adherents of neoclassical theory could
plaUSibly claim that their framework more accurately grasps the empiri­
cal state of affairs of technology in capitalism, this could in principle
compensate for any or al of the shortcomings considered thus far. And
so the ffth and final criterion for evaluations of competing theories may
be the most signifcant of all.
5. Empirical adequacy
The notion that empirical adequacy serves as an important criterion in
assessing theories seems rather straightforward. But things are more
complicated than they might appear at frst glance. In the early part of
Terie of thnolO
this century many philosophers of science thought that empirical facts
could provide a direct verifcation or falsification of scientifc theories. It
is, however, impossible to discuss empirical facts without employing the
categories of some theory or other. This leads to the following problem:
if there are no theory-free observations, how can observations provide an
independent test for the validity of theories?
Consider a familiar Marxian criticism of the neoclassical production
function, which presupposes that both investment capital and machinery
used in production can b considered apart fom labour. Marxists insist
that the ultimate source of investment capital is the exploitation of wage­
labour in production. Similarly, from this perspective machinery is sim­
ply embodied ('dead') labour, that is, the fruit of past labouring activi­
ty.33 Investment capital and capital goods are both objectifications of
labour, albeit objectifcations that take on forms alien to labour. For a
Marxist, this provides a compelling empirical refutation of the neoclassi­
cal framework. For the neoclassical economist, however, 'exploitation'
and 'alienation' are not at all neutral empirical facts. Describing states of
affairs in these terms presupposes the very theory whose validity is in
dispute; anyone rejecting Marxian theory would also reject the accuracy
of these alleged empirical descriptions.
Does this imply that references to empirical matters have no place in
comparisons of competing theories? It would do so only if there were no
relatively uncontroversial empirical facts to consider, i.e. facts that can­
not reasonably be disputed within either framework.'4 In the case at
hand there are a great number of such facts available. And so, in princi­
ple, we should be able to assess which position may claim greater empir­
ical adequacy, using these relatively uncontroversial empirical facts as
the measure. I shall argue that there are five central areas where the
Marxian theory of technology is superior to neoclassical theory accord­
ing to this criterion.
In the neoclassical approach, technology is treated as if it were exoge­
nous to economic activity. Firms simply accept as given the various tech­
niques depicted in a production function, and the technical advances
that permit a jump to a new production function. Yet capitalist enterpris­
es can be empirically observed to engage in searches for both techniques
and new technologies. Marxian theory can account for the indisputable
empirical fact that technical change is endogenous to capitalism; neo­
classical theory cannot. Marx showed in great detail how the logic of the
capital/wage-labour relation, capital's need to reduce circulation time
and constant capital costs, and inter-capital competition, all tend to lead
enterprises to introduce technical innovations. A a result,
The bourgeoisie, during its rule of scarcely one hundred years, has created
more massive and more colossal productive forces than have all preceding
generations together. Subjection of Nature's forces to man, machinery, appli­
cation of chemistry to industry and agriculture, steam-navigation, railways,
electric telegraphs, clearing of whole continents for cultivation, canalisation of
rivers, whole populations conjured out of the ground -what earlier century
had even a presentiment that such productive forces slumbered in the lap of
social labour? 35
Historical Materialism
To treat technical change as exogenous in the light of al this is surely to
be in tension with an indisputable feature of the capitalist world.
A closely related point to be made in this context concerns the eco­
nomic role of capitalst enterprises. In the standard neoclassical account,
firms simply respond to the quantity of output demanded, input prices,
and the given technology. In selecting a given technique they do not so
much initiate a state of affairs as merely note the point of tangency
between the isocost line and the production function. But it is a fairly
uncontroversial fact that capitalist enterprises do not passively respond
to price signals and given technologies. In capitalist markets innovative
firms are able to set prices sufciently high to provide surplus profits,
thanks to higher levels of productivity and/or growing markets. Marx's
theory of capital, in which the drive to accumulate is the spark setting off
technical change, accounts for the aggressive behaviour of capitalist
enterprises far better than the neoclassical framework. '6
The third issue concerns equilibrium states in the economy. The neo­
classical model is explicitly based on the supposition that there is a dom­
inant tendency for the economy to attain equiibrium. Once the tech­
nique represented by the point of tangency between the isocost line and
the production function has been selected, there is no internal dynamic
leading away from this technique. Of course neoclassical theory is not
limited to static models of this sort. There are also dynamic models that
incorporate economic growth. But these models too are based on the
supposition that the economy tends towards an equilibrium state, albeit
one of dynamic equilibrium in these cases.37
Relatively uncontroversial empirical evidence suggests that it is wrong
to see divergences from equilibrium as infrequent and temporary. A
economic environment in which frms are constantly introducing innova­
tions in the hopes of shifing the balance of class forces, winning surplus
profits, and increasing the rate of accumulation, is not conducive to
either static or dynamic equilibrium. Capitalism is not characterised by
the attainment of a stable response to already given conditions, but by
the generation of a ceaseless fux of new economic conditions.'s In a
world of ceaseless technical change disequilibrium is the norm, not the
exception.'· Marxian theory can account for this empirical state of
affairs, while neoclassical theory cannot.
Fourth, in the neoclassical model, any shift in the ratio of factor
prices leads to a shif from one technique to another. This implies that
any choice is reversible; a technique abandoned when the ratio of factor
prices changes can be reinstated if this ratio shifts back. This also does
not correspond to relatively uncontroversial empirical evidence. In the
course of employing a given technique the workforce develops skills spe­
cific to that technique. These competencies are not likely to be equally
applicable to all other techniques. This implies that a decision to operate
one technique cannot easily be reversed wt every shift of the ratio of
factor prices. History matters; technical choices made in the past restrict
the choices that can be made later. Marx's theory of labour-power and
its capacity to engage in learning by doing can account for this 'path
dependency', while production function models cannot.
Finally, the production function model assumes that there is always
and everywhere completely substitutability of labour and capital, i.e. that
Teores of tehnolog
the same output can be attained whether we use more capital-intensive
or more labour-intensive techniques. There are certainly cases where
this assumption makes some empirical sense. With sufficient extra
labour the same amount of firewood can be produced with a handsaw as
with a chainsaw. But it is a fairly straightforward empirical fact that
some tasks cannot be accomplished if the right machines are not avail­
able, no matter how many extra labourers might be hired to
compensate.40 Other tasks simply cannot be completed if labourers with
specifc skills are absent, no matter how many additional machines are
purchased.41 There is nothing in the Marxian framework that contradicts
these states of affairs.
In conclusion, the rational assessment of competing theories is a com­
plex matter. There are a number of quite different criteria that must be
employed. It is always possible that one of the competing theories may
appear more adequate when assessed by one criterion, while another
appears stronger when a different criterion is invoked. In such circum­
stances it may be unclear what weight each yardstick should be given, or
how the necessary trade-offs ought to be made. It would seem from the
above, however, that we are in the fortunate position of not having to
decide such vexing questions here. The relevant criteria for evaluating
the respective strengths of the neoclassical and the Marxist accounts of
technology in capitalism are explanatory scope, internal consistency of
foundational categories, the appropriateness of abstractions, compatibili­
ty with relevant social practices, and empirical adequacy. The Marxian
perspective appears to be far stronger on all fve counts. The conclusion
that Marxism provides a theoretically superior account of technical
change in capitalism is thus rationally warranted.
1 . Ferguson 1 969; Elster 1 983; Coombs et al. 1 987.
2. Hicks 1 932, p. 125
3. Salter 1 960, pp. 43-44; see Elster 1 983, pp. tol-2
4. See Smith 1 990, 1 993.
5. Marx 1 976, p. 756; see Ramtin 1 991 , p. 68
6. See Smith 1 994a, 1 994b, for a discussion of these matters in the context of the
technologies and forms of social organisation associated with lean production.
7. k Marx wrote regarding savings in the use of fxed capital, 'Finally, however, it
is only the experience of the combined worker that discovers and demonstrates
how inventions aready made can most simply be developed, how to overcome
the practical frictions that arise in putting the theory into practice - its applica­
tion to the production process, and so on.' Marx 1 981 , pp. 1 98-99.
8. Negri 1 989, p. 48.
9. These other moments, however, remain connected to the capital/wage-labour
relation. Their ultimate social significance lies in their contribution to the
reproduction of that relation.
to. 'During its circulation time, capital does not function as productive capital, and
therefore produces neither commodities nor surplus-value . . . The more that
the circulation metamorphoses of capital are only ideal, i.e. the closer the cir-
Historical Matrialism
culation time comes to zero, the more the capital functions, and the greater is
its productivity and self-valorisation' . Marx 1 978, p. 203; see also pp. 326,
388-9, 391-2.
1 1 . Wolff and Resnick 1 987.
1 2. McCloskey 1 983.
1 3. See Smith 1 997.
1 4. For example, 'It is only after a considerable development of the science of
mechanics, and an accumulation of practical experience, that the form of a
machine becomes settled entirely in accordance with mechanical principles,
and emancipated from the traditional form of the tool from which it has
emerged.' Marx 1 976, p. 505.
1 5. Marx 1 981 , p. 1 72.
16. Dertouzos et al. 1 991 .
1 7 . 'It took both time and experience before the workers learnt to distinguish
between machinery and its employment by capital, and therefore to transfer
their attacks from the material instruments of production to the form of society
which utilises those instruments.' Marx 1 976, pp. 55455; see also Noble
1 984, Shaiken 1 985.
1 8. This is quite ironic, given the emphasis on consumer sovereignty found in
other parts of the neoclassical paradigm.
1 9. Marx 1981, p. 283.
20. Marx 1973, p. 409.
21 . Storper and Walker, 1 989.
22. The classic example of how social movements and state regulations affect the
technical change process is found in volume I of Capital. Marx described how
overwork in factories led to social movements to reduce the working day. Once
regulations were passed limiting the length of the working day, large-scale
machinery was introduced in order to extract the same (or greater) amount of
surplus-value in the shorter working day.
23. Baumol 1977, pp. 641-42.
24. General equilibrium theory does avoid the problem of aggregating capital,
through the simple expedient of disaggregating the various capital inputs in the
production function: P " f(KI, K, K, ... K, L). But general equilibrium the­
ory avoids this problem only at the cost of even more unrealistic assumptions,
such as complete future markets for all goods and complete knowledge of
future technology. And so it is even further removed from a realistic theory of
technology than the standard production function approach.
25. 'A the dominant subject of this process [Le. the capital circuit], in which it
alternately assumes and loses the form of money and the form of commodities,
but preserves and expands itself through all these changes, value requires
above alan independent form by means of which its identity with itself may be
asserted. Only in the shape of money does it possess this form. Money there­
fore forms the starting-point and the conclusion of every valorisation process.'
Marx 1 976, p. 255.
26. Reuten 1 993.
27. If perfect information regarding the most efcient production function and the
given isocost line were available to al, there would be no diffusion process. Al
frms would simultaneously jump from one technique to another whenever
changes in technology or in the ratio of relative factor prices occurred. But dif­
fusion does extend in time; not all firms adopt innovations at once. Attempts to
explain why this is so solely in psychological terms - 'Some agents are more
Tere o tenolog
disposed to innovate than others' - are not plausible. Another crucial variable
in diffusion is knowledge about the innovation, which differs from frm to frm
at different points in time (Davies 1 979). Hence if we are to account for the
phenomenon of diffusion, we need a theory that does not presuppose perfect
information for all frms at all times. Also, there are many mechanisms avai­
able to frms to prevent perfect information, trade secrets being one obvious
28. Simon 1 954, p. to.
29. Of course, the normative principles embedded in this model (for example, the
ethical primacy of self-interested individual agents) are extremely dubious.
This does not affect the present point.
30. Smith 1 992, chapter I.
3 1 . Botwinick 1 993.
32. Szakonyi 1 992.
33. To say that the means of production and circulation are objectifcations of
labour should not be taken to imply that they are only objectifications of
labour. They have a material dimension that is irreducible to their social form,
as theorists attempting to synthesise Marxism and environmentalism have cor­
rectly stressed (Benton 1 989).
34. Sayer 1 984, chapter 2.
35. Marx 1 977, p. 225.
36. Botwinick 1 993.
37. Coombs et al. 1 987, chapter 6.
38. Storper and Walker 1 989.
39. This point is freely admitted by many non-Marxist economists: 'When the
inflow of major product innovations is high, as it clearly is in most of the world
today, giving birth to many new industries and rapidly shifting the demand in
old industries, industries tend to b out of equilibrium all the time. Equilibrium
conditions could then be expected to be at variance with the empirical evi­
dence, the latter mirroring the constantly transitory non-equilibrium situations
of industries.' Gomulka 1 990, pp. 1 60- 1 .
40. For example, how could the addition of more people allow space exploration in
the absence of rockets? The same point holds for producing polymers, slicing
genes, etc.
41 . General Motors found this out to its great cost when its attempts to build total­
ly automated factories in the early eighties failed (Hoerr et. al. 1 989, p. 363) .
Baumol, William 1 977, Economic Theory and Operations Analysis, .
fourth Edition, Englewoods Cliffs, NJ: Prentice-Hall.
Benton, Ted 1 989, 'Marxism and Natural Limits: A Ecological Critique
and Reconstruction', New Lef Review, 1 78.
Botwinick, Howard 1 993, Persistent Inequalities: Wage Disparity under
Capitalist Competition, Princeton: Princeton University Press.
Coombs, Rod, P. Saviotti, and V. Walsh 1 987, Economics and
Technological Change, London: MacMillan.
Davies, S. 1 979, The Difsion of Process Innovations, Cambridge:
Cambridge University Press.
Dertouzos, Michael, R. Lester, R. Solow, and the MIT Commission on
Industrial Productivi
y 1 991 , Made in America: Regaining the
I S3
Historical Materialism
Prductive Edge, Cambridge, M: MIT Press.
Elster, Jon 1 983, Explaining Technical Change, Cambridge: Cambridge
University Press.
Ferguson, F.E. 1 969, The Neoclassical Theor of Production and
Distribution, Cambridge: Cambridge University Press.
Forester, John, (ed.) 1 989, Computers and the Human Context,
Cambridge M: MIT Press.
Gomulka, Stanislaw 1 990, The Theor of Technological Change and
Economic Growth, New York: Routledge.
Hicks, John 1 932, The Theor of Wages, London: Macmillan.
Hoerr, John, M. Pollock, and D. Whiteside 1 989, 'Management
Discovers the Human Side of Automation', in Forester 1 989.
Marx, Karl 1973, Grundrisse, New York: Vintage Press.
Marx, Karl 1 976, Capital, volume I, New York: Penguin.
Marx, Karl 1 977, Krl Marx: Selected Writings, edited by David
McLellan, New York: Oxford University Press.
Marx, Karl 1 978, Capital, volume II, New York: Penguin.
Marx, Karl 1 981 Capital, volume II. New York: Penguin.
McCloskey, Donald 1 983 'The Rhetoric of Economics', Joural of
Economic Literature, X.
Moseley, Fred, ed. 1 992, Mar's Method in 'Capital', Atlantic Highlands,
NJ: Humanities Press.
Moseley, Fred, and M. Campbell, (eds.) 1 997, Ne Investigations of
Marxian Method, Atlantic Highlands, NJ: Humanities Press.
Negri, Antonio 1 989, The Politics of Subversion, Cambridge: Polity Press.
Noble, David 1 984, Forces of Production, New York: Knopf.
Ramtin, Ramin 1 991 , Capitalism and Automation, London: Pluto Press.
Reuten, Geert 1 993, 'The Difcult Labour of a Theory of Social Value:
Metaphors and Systematic Dialectics at the Beginning of Marx's
Capital', in Moseley 1 992.
Salter, W.G. 1 960, Productivit and Technical Change, Cambridge:
Cambridge University Press.
Sayer, Andrew 1 984, Method in Social Science: A Realist Approach,
London: Hutchinson.
Shaiken, Harley 1 985, Work Transformed: Automation and Labour in the
Computer Age, New York: Holt, Reinhart and Winston.
Simon, Herbert 1 978, 'On How to Decide What to Do', Bell Joural of
Economics, 9.
Smith, Tony 1 990, The Logic of Marx's 'Capital', Albany: State University
of New York Press.
Smith, Tony 1 992, The Role of Ethics in Social Theor, Albany: State
University of New York Press.
Smith, Tony 1 993, Dialectical Social Theor and Its Critics, Albany: State
University of New York Press.
Smith, Tony 1 994a, 'Flexible Production and the Capital/Wage Labour
Relation in Manufacturing', Capital and Class, 53.
Smith, Tony 1 994b, Lean Production: A Capitalist Utopia?, Amsterdam:
The International Institute for Research and Education.
Smith, Tony 1 997, 'Marx's Systematic Dialectic and Lakatos'
Methodology of Scientifc Research Programs', in Moseley and
Campbell 1 997.
Theories of technolog
Storper, Michael, and R. Walker 1 989, The Capitalist Imperative:
Territor, Technolog, and Industrial Growth, Oxford: Blackwell.
Szakonyi, Robert, (ed.) 1 992,Technolog Management: Case Studies in
Innovation, Boston: Auerbach Publications.
Wolff, Richard, and S. Resnick 1 987, Economics: Marian versus
Neoclassical, Baltimore: Johns Hopkins University Press.
The silences of Capital·
Michael A Lebowitz
Not too long ago, Michael Burawoy commented that 'two anomalies con­
front Marxism as its refutation: the durability of capitalism and the pas­
y of its working class'.2 So, has the time come, more than 125 years
after the publication of Capital, to admit that the 'facts' (which meant
something to Marx) simply do not support Marx's theory?'
It depends. It depends on what aspect of Marx's theory we have in
mind. What reason would we have on the basis of historical experience to
reject Marx's analysis of the nature of capital? Should we scuttle the idea
that capital rests upon the exploitation of workers, that it has an insatiable
appetite for surplus labour, that it accordingly searches constantly for
ways to extend and intensify the working day, to drive down real wages,
to increase productivity? What in the developments of world capitalism in
the last two centuries would lead us to think that capital is any different?
Do we think that, for example, Marx's statement that capital 'takes no
account of the health and the length of life of the worker, unless society
forces it to do so' no longer holds - and, indeed, that it does not apply as
well to capital's treatment of the natural environment?4 Was Marx wrong
in proposing that 'the entire spirit of capitalist production, which is ori­
ented towards the most immediate monetary proft' is contrary to 'the
whole gamut of permanent conditions of life required by the chain of
human generations', or that all progress in capitalist agriculture in
'increasing the fertility of the soil for a given time is a progress towards
ruining the more long-lasting sources of that fertility'? Does our modern
experience with chemical pesticides and fertilisers refute Marx's perspec­
tive on capitalism and nature, on what capitalist production does to 'the
original sources of all wealth - the soil and the worker'?'
Much has, of course, changed in the last two centuries - indeed, in
the last quarter of this century. But, is the nature of capital among them?
The apparent victory of capitalism over its putative alternative is not a
challenge to the theory of Capital. Modern celebrants of capital would
fmd in Marx an unsurpassed understanding of capital's dynamic, rooted
in the self-valorisation that serves as motive and purpose of capitalist pro­
duction. That capital drives beyond 'all traditional, confined, complacent,
encrusted satisfactions of present needs, and reproductions of old ways
of life', that it constantly revolu
ionises the process of production as well
as the old ways of life, 'tearing down all the barriers which hem in the
development of the forces of production, the expansion of needs, the all­
sided development of production, and the exploitation and exchange of
natural and mental forces' - all this was central to Marx's conception of
Silenc of CI p/tl
production founded upon capita1.6 Thus, if capital today compels nations
to adopt capitalist forms of production, creates a world after its own
image and indeed shows once again that all that is solid (including that
made by men of steel) melts into air, this in itself cannot be seen as a
refutation of Marx.7
Nor, finally, in these days of shutdowns, growing unemployment and
devaluation of capital, can we forget the contradictory character of capi­
talist reproduction that Marx stressed - his injunction that capital's ten­
dency towards the absolute development of productive forces occurs only
in 'the first act' and that the realisation of surplus-value produced
requires a 'second act' in which commodities must be sold 'within the
framework of antagonistic conditions of distribution' marked by capitalist
relations of production.8 In the signs of capitalist crisis about us, we have
yet another demonstration that the understanding of the nature and logic
of capital contained in Marx's Capital is as valid as ever.
And, yet, there is that so-obvious failure, that apparent anomaly. And,
that is that capital is still with us and shows no signs of taking its early
departure. For some on the Right (as well as adherents to the thesis of
the primacy of productive forces), this is simply proof that capitalist rela­
tions of production are not a fetter on the development of productive
forces and, indeed, that capitalism is 'optimal for the further development
of productive power'. 9
All this, of course, is despite Marx's assurance that capitalism was
doomed, that it would come to an end with 'the revolt of the working
class, a class constantly increasing in numbers, and trained, united and
organised by the very mechanism of the capitalist process of production' .
But, the 'knell' has not sounded for capitalism, and the expropriators
have not been expropriated. 1
We need to know why. What in Marx's Capital can explain it, what
should have prepared us for understanding this historic failure? The
answer, I suggest, is not what is in Capital but, rather, what is not.
I. The 'real silences'
What is missing in Capital was stated quite well by E. P. Thompson in his
Povert of Theor. Capital, he argued, is 'a study of the logic of capital,
not of capitalism, and the social and political dimensions of the history,
the wrath and the understanding of the class struggle arise from a region
independent of the closed system of economic logic' . " For Thompson,
the problems in Marx originated when he proceeded from political econ-
0my 'to capitalism . . . , that is, the whole society, conceived as an "organic
system.'" The flaw was that 'the whole society comprises many activities
and relations . . . which are not the concern of Political Economy, which
have been defned out of Political Economy, and for which it has no
terms' .12 Ad, the critical 'missing term' for Thompson is that of 'human
experience'. When we raise this point, he proposed, 'at once we enter into
the real silences of Marx' . 13
Who could deny that there is indeed this silence in Capital? There is
no place in Capital for living, changing, striving, enjoying, struggling and
developing human beings. People who produce themselves through their
own activities, who change their nature as they produce, beings of praxis,
are not the subjects of Capital. The idea of the 'rich human being' - 'the
1 1351
Historical Materialism
man in whom his own realisation exists as an inner necessi
y, as need' is
entirely foreign to Capital. " In its place, we have structures which domi­
nate human beings, a logic of capital that means that the individual con­
sumption of the worker 'remains an aspect of the production and repro­
duction of capital, just as the
leaning of machinery does' ; we have a
working class that 'is just as much an appendage of capital as the lifeless
instruments of labou are' . 15
Why, however, is there this silence? Thompson argues that it is the
result of the mature Marx's preoccupation with the critique of political
economy. That, in contrast to his early attack on the latter for not consid­
ering the worker 'when he is not working, as a human being', Marx fell
into a trap: 'the trap baited by "Political Economy. " Or, more accurately,
he had been sucked into a theoretical whirlpool' - one in which 'the pos­
tulates ceased to be the self-interest of man and became the logic and
forms of capital, to which men were subordinated'. For Thompson, the
problems of Marxism are the result of the 'system of closure' in which all
is subsumed within the circuits of capital, where capital posits itself as an
"organic system" .
And, yet, if we accept Marx's concept of an organic system as one in
h 'everything posited is thus also a presupposition' (i.e. in which all
premises are the results of the system itself, that claim cannot be con­
ceded. There is no organic system established in Capital.17 At the very
point of the discussion of simple reproduction, intended to consider capi­
talism as an organic system, we see there is an element which is not part
of capital, which is not produced and reproduced by capital - a point of
departure but not one of return in the circuit of capital, a premise which
is not a result of capital itself. And, it is one necessary for the reproduc­
tion of capital, required for the very existence of capital itself:
The maintenance and reproduction of the working class remains a necessary
condition for the reprduction of capital. But the capitalist may safely leave this
to the worker's drives for self-preservation and propagation."
Thirty-two words - and, then, theoretical silence. What is missing
from the circuit of capital is the second moment of production ('Moment
N') , the consideration of the production of the wage-labourer. And this
question, the subject matter for the projected book on Wage-Labour
which was to complete 'the inner totality', involves far more than concern
with physical reproduction or the household; it encompasses as well the
social reproduction of wage-labour. 19 Rather than a 'system of
Capital is only a moment in the development of an organic system.20
II. Situating the silence
Let us attempt to identify a few of the problems associated with the
absence of the book on Wage-Labour. A full discussion of these and
other issues can not be pursued here but is the subject of my book,
Beond Capital: Mar's Political Economy of the Working Class.
A. The struggle for higher wages
While Capital presents well the manner in which workers, indeed all
human beings, are means for capital in its drive for self-valorisation, it
1 1361
Silnc of cpitol
does not do the same for the side of the worker. Little is said about what
Marx identifed as the goal of the worker, about 'the worker's own need
for development', or how she strives to achieve that goa!.2
We under­
stand quite well why, for example, capital struggles to 'reduce wages to
their physical minimum and to extend the working day to its physical
maximum', but we do not know precisely why 'the working man con­
stantly presses in the opposite direction'.22 Further, there is no discussion
at all in Capital about the struggle for higher wages.
One aspect of Capital's silence is that it does not explore the manner
in which new needs are constantly created for workers. Marx consistently
stressed that the creation of 'new needs arising from society itself is 'a
condition of production founded on capital' and that the capitalist
searches for means to spur workers on to consumption, 'to give his wares
new charms, to inspire them with new needs by constant chatter, etc'.
Yet, although Marx emphasised that the growth of capitalist production
meant that the worker's 'subjective poverty, his need and dependence
grow larger in proportion', none of that plays a role in Capital." Even
though each new need becomes a new link in the golden chain which
secures workers to capital, even though Marx in the Grundrisse
announced that it is upon this creation of new needs for workers that 'the
contemporary power of capital rests', Capital is silent here.24
Thus, the underlying basis for the struggles of workers to secure high­
er wages is not present. But, then, what would be the point anyway?
Capital, after all, assumes the standard of necessity for workers to be a
'constant magnitude' for a given country at a given period; and, Marx did
this in order to avoid 'confounding everything' . 25 As he noted in the
Grundrisse, no matter how much the standard of necessity may change,
'to consider those changes themselves belongs altogether to the chapter
treating of wage labour' .2
Marx was very clear and consistent on this
point: changes in the needs of workers were not properly part of the sub­
ject matter of Capital. A he indicated in his 1 861 -3 draft notebooks for
Capital (known as Zur Kitik and newly translated into English) :
The problem of these movements in the level of the workers' needs, as also that
of the rise and fall of the market price of labour capacity above or below this
level, do not belong here, where the general capital-relation is to be developed,
but in the doctrine of the wages of labour ««s. Alquestions relating to it [the
level of workers' needs - ML] as not a given but a variable magnitude belong to
the investigation of wage labour in particular ...&27
The point was exactly the same in the material Marx drafted for volume I
of Capital in 1 864-5:
The level of the necessaries of life whose total value constitutes the value of
labour-power can itself rise or fall. The analysis of these variations, however,
belongs not here but in the theory of wages.28
Not only does this assumption of a constant standard of necessity (the
assumption in Capital which was to be removed in the book on Wage­
Labour) mean that there can be no examination of the implications of
changes in real wages, but it is not at all surprising that Marx had little to
Historical Mateialism
say in Capital about trade unions ('whose importance for the English
working class can scarcely be overestimated'29) . There is no discussion of
how the organised worker 'measures his demands against the capitalist's
proft and demands a certain share of the surplus-value created by him';
there is no consideration of how, despite capital's own tendency, workers
would not permit wages 'to be reduced to the absolute minimum; on the
contrary, they achieve a certain quantitative participation in the general
growth of wealth'. 30 As Engels commented, the great merit of the trade
unions is that 'they tend to keep up and to raise te standard of life' ."
But, all this is missing from Capital.
The point is that Capital does not have as its object the examination of
the movement when 'the workingman presses in the opposite direction' to
capital. Even in the case of the struggle over the working day (which
Marx did introduce into Capital) , rather than a theoretical exploration of
the inherent tendency of workers to struggle for a reduction of the work­
ing day because of their need for more time and energy for their own
process of production, he focuses upon the effort of workers to retain the
'normal' working day (Le. a defensive action) . In general, while we see
capital's tendency to increase the rate of surplus-value, there is no treat­
ment of wage-labour's tendency to reduce the rate of surplus-value. The
very tendencies of wage-labour which emerge from 'the worker's own
need for development' and which are the basis of the struggles of workers
for themselves are absent.32 There is no theoretical framework for dealing
with increases in the standard of necessity because Capital is meant to
explain the logic of capital but not the logic of wage-labour.
B. The inherent functionalism
Precisely because the worker as subject is absent from Capital, precisely
because the only subject is capital - and the only needs and goals those of
capital, there is an inherent functionalist cast to the argument which fows
from Capital. Characteristic of a one-sided Marxism that fails to recog­
nise that Capital presents only one side of capitalism is the presumption
that what happens occurs because it corresponds to capital's needs
(which are the only ones acknowledged) .
Thus, for one-sided Marxism, if the working day declines, it is
because capital needs workers to rest. If the real wage rises, it is because
capital needs to resolve the problem of realisation. If a public healthcare
system is introduced, it is because capital needs healthy workers and
needs to reduce its own costs; if a public school system is established, it is
because capital requires better-educated workers. If sectors of an econo­
my are nationalised, it is because capital needs weak sectors to be operat­
ed by the state. Such arguments are inherently one-sided. When the
needs of workers are excluded at the outset and only capital's needs are
recognised, it cannot be considered surprising that a one-sided Marxism
will fnd in the results of all real struggles a correspondence to capital's
Yet, this problem is not unique to those who have followed Marx. The
same functionalist argument can be found in Capital itself. Regardless of
his account of the struggle by workers to limit the working day and of the
resistance by capital, Marx nevertheless could comment that, due to the
deterioration of its human inputs, 'the limiting of factory labour was dic-
Silncs of Cpltll
tated by the same necessity as forced the manuring of English felds with
guano' . The limiting of the working day, in short, occurred (was 'dictat­
ed') because it corresponded to capital's requirements (just as farmers
had to replenish the fertility of the soil) . That clear functionalist statement
appears even though Marx later commented that capital concerns itself
with the degradation of the human race as little as with 'the probable fall
of the earth into the sun' and must be 'forced by society' to do SO.33
A similar problem is apparent in Marx's description of the value of
labour-power as determined by the 'value of the necessaries required to
produce, develop, maintain, and perpetuate the labouring power'. 34 The
premise is that, since the worker unfortunately depreciates and has a lim­
ited life, the maintenance of the use-value of this instrument with a voice
includes expenditures not only to redress its daily wear and tear but also
for those 'means necessary for the worker's replacements, i.e. his chil­
dren'.35 Since 'the man, like the machine', Marx proposes, 'will wear out,
and must be replaced by another man', there must be sufcient neces­
saries 'to bring up a certain quota of children that are to replace him on
the labour market and to perpetuate the race of labourers'. 3
Frankly, to propose that the value of labour-power contains provisions
for the maintenance of children because capital wants future recruits
twenty years hence - rather than because workers have struggled to
secure such requirements - is a teleological absurdity! However, it is a
logical result of the disappearance of wage-labour for itself from Capital.
Marx himself must bear responsibility for some of the functionalist absur­
dities of his disciples.
For those who have followed Marx in this regard and who have fur­
thermore treated Capital as a completed epistemological project, the
results have been disastrous. Failing to grasp what is missing from
Capital, failing to investigate the worker as subject, they are left with the
Abstract Proletarian, the mere negation of capital. That productive work­
er for capital within the sphere of production (i.e. the wealth producer)
and epitomised as the factory worker, that productive instrument with a
voice which can gain no victories which alow it to take satisfaction in
capitalist society (any apparent victories being in fact those of capital) ,
that not-capital who is united and disciplined as the result of capitalist
development - the Abstract Proletarian has no alternative but to over­
throw capital.
c. The dependence of the wage-labourer
The expropriators, however, have not been expropriated. And, it is not at
all difcult to grasp why when we focus upon the worker rather than upon
capital as the subject. For, it is the dependence of the wage-labourer upon
capital, that need and dependence which grows larger in proportion to
capitalist production, which becomes critical to understand. Consider the
position of the wage-labourer. In order to satisfy her needs, she must
secure use-values from outside her own process of production. Under the
prevailing circumstances, she must take the only potential commodity she
has, living labour capacity, and fnd the buyer for whom it is a use-value -
capital. To be for self, the wage-labourer must be a being for another. We
have here the worker as wage-labourer for self - as one who approaches
capital as a means, a means whose end is the worker for self.
Capitalism, in short, encompasses not only a relation in which the
I S31
Historical Materialism
worker is the mediator for capital in securing its goals (C-WL-C) but
also a relation in which capital is the mediator for the worker in securing
hers (WL-C-WL) . Once we articulate this second side, there is no mys­
tery behind the dependence of the worker upon capital. Within this rela­
tion, workers need capital; it must appear as the necessar mediator for
the worker. The maintenance and reproduction of capital remains a nec­
essary condition for the reproduction of the worker as wage-labourer. A
Marx noted in the Grundrisse, if capital cannot realise surplus-value by
employing a worker, then:
labour capacity itself appears outside the conditions of the reproduction of its
existence; it exists without the conditions of its existence, and is therefore a
mere encumbrance; needs without the means to satisfy them; .... 37
The worker, accordingly, is produced as one conscious of her depen­
dence upon capital. And, everything about capitalist production con­
tributes not merely to the relation of dependence but also to the 'feeling
of dependence' .'8 The very nature of capital is mystified - 'all the produc­
tive forces of social labour appear attributable to it, and not to labour as
such, as a power sp
inging forth from its own womb'. Having surren­
dered the right to his 'creative power, like Esau his birthright for a mess
of pottage', capital, thus, becomes 'a very mystical being' for the worker
because it appears as the source of all productivity.J9
Fixed capital, machinery, technology, science - all necessarily appear
only as capital, are known only in their capitalist form:
The accumulation of knowledge and of skill, of the general productive forces of
the social brain, is thus absorbed into capital, as opposed to labour, and hence
appears as an attribute of capital.. . .'0
Thus, as Marx commented, this transposition of 'the social productivi­
ty of labour into the material attributes of capital is so frmly entrenched
in people's minds that the advantages of machinery, the use of science,
invention, etc. are necessarily conceived in this alienated form, so that all
these things are deemed to be the attributes of capital' . 41 In short, wage­
labour assigns its own attributes to capital in its mind because the very
nature of the capital/wage-labour relation is one i which it has already
done so in reality.
Insofar as this sense of dependence upon capital is regularly repro­
duced, capital can safely leave its own condition of existence, the mainte­
nance and reproduction of the working class, to the worker's own drives.
The very process of capitalist production produces and reproduces work­
ers who consider the necessity for capital to be self-evident:
The advance of capitalist production develops a working class which by educa­
tion, tradition and habit looks upon the requirements of that mode as self-evi­
dent natural laws. The organisation of the capitalist process of production,
once it is fully developed, breaks down all resistance."
Breaks down all resistance! In the light of Marx's comment, how can
we possibly talk about the durability of capitalism and the passivity of the
Silences of Cpltl'
working class as anomalies? Indeed, given Marx's statement that 'the
great beauty of capitalist production' consists in its ability to constantly
replenish the reserve army of labour and thereby to secure 'the social
dependence of the worker on the capitalist, which is indispensable', how
can we talk about the revolt of the working
lass (however well it may be
'trained, united and organised') ?43 On the contrary, as Marx noted about
developed capitalism:
In the ordinary run of things, the worker can be left to the "natural laws of pro­
duction", i.e. it is possible to rely on his dependence on capital, which springs
from the conditions of production themselves, and is guaranteed in perpetuity
by them."
By education, tradition and habit, the worker in developed capitalism
necessarily looks upon the requirements of capital as common sense; and,
that feeling of dependence is indispensable in ensuring capitalism's nec­
essary premise and making it an organic system - in perpetuity.
III. Understanding the silence of Capital
So, if the book on Wage-Labour was so important to an understanding of
capitalism (rather than of just the logic of capital), why didn't Marx write
it? To answer this requires us frst to be absolutely clear as to why Marx
wrote Capital (and, indeed, volume l over and over again) . The answer is
precisely his understanding of the dependence of the worker upon capi­
tal. Given the inherent mystifcation of capital, demystification is a neces­
sary condition for workers to go beyond capital.
For this very reason, Marx considered it essential to reveal the nature
of capital, to reveal what cannot be apparent on the surface - that capital
itself is the result of exploitation. To counter the inherent mystification of
capital required the theory of Capital. Signifcantly, however, for this par­
ticular purpose only Capital - and not the originally projected six books
(or even the frst three) - is required; indeed, only Volume I of Capital is
Capital was Marx's attempt to make the proletariat 'conscious of the
condition of its emancipation', conscious of the need to abolish capital's
ownership of the products of labour - i.e. 'to inscribe on their banner the
revolutionar watchword, "Abolition of the wages system! "'45 That was a
limited object but, nevertheless, a crucial one given Marx's understanding
of capital's inherent tendency to develop a working class which looks
upon capital's requirements as 'self-evident natural laws' .
If we fail to recognise that limited object, however, we may misunder­
stand entirely Capital's place and importance. In the absence of the
demystification of capital, there is no going-beyond capital. Immiseration,
crises, stagnation, destruction of the natural environment do not lead
beyond capital because so long as capital appears as the source of all
wealth and as the only means to satisfy their own needs, workers are nec­
essarily dependent upon it. Thus, Capital is not merely a moment in the
understanding of capitalism as an organic system; it is also a moment in
the revolutionary struggle of workers to go beyond capital. Marx did not
write his projected volume on wage-labour because, ultimately, he was
less interested i the completion of his epistemological project than in his
Historical Materialism
revolutionary project. What E. P. Thompson forgets is that, as Engels
indicated in speech at Marx's graveside, 'Marx was before all else a revo­
lutionist' .46
IV. The most serious silence
And, yet, the failure to focus upon the worker as subject and upon the
process by which the worker produces herself has had a serious effect
both upon Marxism and upon the revolutionary project itself. If one pro­
ceeds simply from the contradictions inherent within capital, the central
issue may become (as it has for adherents of the Regulation School) an
explanation of how capital nevertheless is reproduced; investigation
accordingly focuses upon the particular modes of regulation which man­
age to sustain capitalist relations. If, on the other hand, we begin from a
consideration of the worker as subject, the central question becomes -
how, under the necessary circumstances, can capital not succeed? We are
led necessarily to the central place of the process of class struggle in
Marx's theory.
The silence of Marx in Capital has meant a de-emphasis upon the
process of struggle itself as a process of production. Just as every activity
of the worker alters her as the subject who enters into all activities, simi­
larly the process in which workers struggle for themselves is also a
process in which they produce themselves in an altered way. They devel­
op new needs in struggle, an altered hierarchy of needs. Even though the
needs which they attempt to satisfy do not in themselves go beyond capi­
tal, the very process of struggle is one of producing new people, of trans­
forming them into people with a new conception of themselves - as sub­
jects capable of altering their world.
Nothing is more essential to Marx than this conception. The failure to
understand the centrality of 'the coincidence of the changing of circum­
stances' and of self-change - that coincidence that can only be under­
stood as 'revolutionar practice' - is the failure to understand the dynam­
ic element without which there can be no end to the feeling of depen­
dence and thus no transcendence of capita!!47 Understanding struggle as
a process of production is the most serious gap as the result of Marx's
failure to go beyond Capital . Limited to Capital, we have only the
mechanical laws of capital, a structure without subjects, a one- sided
Marxism - or, to be more exact, the absence of Marxism!
If, then, we accept the importance of 'revolutionary practice', it is
clear what cannot be a basis for going beyond capital - the absence of
people in motion!S And, this is what we need to think about as we
approach the beginning of a new millennium. We need to think about
1 967, 1 00 years after Capital was first published. Capital meant some­
thing more then (as it did at its 50th anniversary) . Why? Not because its
account of capitalism was any truer then but because people were in
motion (and were changing themselves) . That is what made the theory of
Capital a use-value. A Marx well-knew, the struggle of workers is indis­
pensable for 'preventing them from becoming apathetic, thoughtless,
more or less well-fed instruments of production' . Athough Marx wrote
Capital to explain to workers why they were struggling, 'it is not enough
for thought to strive for realisation, reality must itself strive towards
Silnces of Cp't,
thought' .4"
When we teach about Capital, we need to teach what it left out, its
silences - i.e. not only what is in Capital but what is not. And, we have
to help to bring an end to that sience. This is at the core of a revitalised
Marxism, a Marxism that wil continue the revolutionar project that we
observe in Capital.
1 . Revised from the version presented at the Society for Socialist Studies panel:
'Capital After 1 25 Years' at the annual meetings of the Leared Societies of
Canada, Charlottetown, P.E.I., May 28-3 1 , 1992 and published subsequently
in Lebowitz 1 992b.
2. That these 'anomalies' are identified as separate is itself interesting. Burawoy
1 989, p. 51 .
3. ' I was delighted to find my theoretical conclusions fully confrmed by the
FACTS'. Marx 1 987, pp. 407-8.
4. Marx 1 977a, p. 381 .
5. Marx 1 981 , p. 754n and Marx 1 977a, p. 638.
6. Marx 1977b, p. 410; Marx 1 977a, p. 61 7.
7. Marx and Engels 1 976, pp. 487-8.
8. Marx 1981, p. 352.
9. Cohen 1 978, p. 1 75.
1 0. Marx 1977a, p.929.
1 1 . Thompson 1 978, p 65.
1 2. Thompson 1 978, p. 62.
1 3. Thompson 1 978, pp. 1 64-5.
1 4. Marx 1975, p. 304.
1 5. Marx 1 977a, pp. 71 8-19.
16. Marx 1975, pp. 241 -2 and Thompson 1 978, pp. 59, 60, 65, 1 63-4, 1 67.
1 7. Marx 1 977b, p. 278.
1 8. Marx 1 977a, p. 71 8.
1 9. Marx 1977b, pp. 520-1 and 264.
20. Lebowitz 1 992a, chapter 3; Lebowitz 1 982.
21 . Marx 1 977a, p. 772.
22. Marx 1 985, p. 1 46.
23. Marx 1977b, pp. 287, 409-10; Marx 1 977a, p. 1 062.
24. Marx 1977b, p. 287.
25. Marx 1977a, pp. 275, 655; Lebowitz 1 992a, Ch. 2.
26. Marx's reference to this section as a 'chapter' may be placed in context by not­
ing that it occurs in his 'chapter' on capita, which comprises pages 239 to 882
in this edition. Marx 1 977a, p. 81 7.
27. Marx and Engels 1 987, pp. 44-5.
28. Marx 1 977a, pp. 1 068-69.
29. Marx 1 977a, p. 1069. See Lebowitz 1 992a, chapter 5 for a consideration of
problems in the discussion of relative surplus-value when the assumption of a
fixed standard of necessity is relaxed.
30. Marx 1 977b, p. 597; Marx 1971 , p. 3 12.
3 1 . Engels 1 970. Engels's comment in his critique of the Erfurt Programme was:
'The organisation of the workers, their constantly increasing resistance, wl
Historical MateriaUsm
most probably act as a certain barrier against the increase of poverty.'
32. Among these is the tendency to combine and reduce the separation among
them. Cf Lebowitz 1 992a, chapter 4.
33. Marx 1977a, pp. 348, 381 .
34. Marx 1 985, p. 130
35. Marx 1 977a, p. 275.
36. Marx 1 985, p. 1 29.
37. Marx 1 977b, p. 609.
38. Marx 1977a, p. 936.
39. Marx 1 981 , p. 966; Marx 1 977b, p. 307.
40. Marx 1 981 , p. 694.
41 . Marx 1 977a, p. 1 058.
42. Marx 1 977a, p. 899.
43. Marx 1 977a, p. 935.
44.Marx 1 977a, p. 899.
45. Marx 1 985, p. 1 49.
46. Engels in Tucker 1 978, p. 682.
47. Marx 1 976, p. 4.
48. Note that this concept also points to the nature of the state necessary to go
beyond capital. See Lebowitz 1 995.
49. Marx and Engels 1 975, volume 1 2, p. 1 69; Marx 1 975, p. 183.
Burawoy, Michael 1 989, 'Marxism Without Micro-Foundations',
Socialist Riew, 1 9: 2.
Cohen, GA. 1 978, Krl Marx's Theory of Histor: A Defence, Princeton:
Princeton University Press.
Engels, Friedrich 1 881 , 'The Wages System', The Labour Standard, 21
May 1 881 .
Engels, Friedrich 1 970, 'Critique of the Draft Social-Democratic
Programme 1 89 1', Marism Today.
Engels, Friedrich 1 978, 'Speech at the Graveside of Karl Marx', The
Mar-Engels Reader, 2nd edition, edited by Robert C. Tucker, New
York and London: W.W. Norton.
Lebowitz, Michael A. 1 982, 'The One-Sidedness of Capital', Review of
Radical Political Economics, 1 4: 2.
Lebowitz, Michael A 1 992a, Beond Capital: Marx's Political Economy of
the Working Class, New York: St. Martin's Press.
Lebowitz, Michael A 1 992b, 'Capital After 1 25 Years', Socialist Studies
Bulletin, 42.
Lebowitz, Michael A 1 995, 'Situating the Capitalist State', in Marxism in
the Moder World: Confonting the New World Order, edited by
Antonio Calari et aI., London: Guilford Press.
Marx, Karl 1 971 , Theories of Surplus Value, volume III, Moscow:
Progress Publishers.
Marx, Karl 1 975a, 'Economic and Philosophic Manuscripts of 1 844', in
Mar and Engels Collected Works, volume 3, New York: International
Marx, Karl 1 975b, 'Contribution to the Critique of Hegel's Phiosophy of
Law. Introduction', in Mar and Engels Collected Works, volume 3,
Silences of CI pltl l
New York: International Publishers.
Marx, Karl and Engels, Friedrich 1 975c, Marx and Engels Collected
Works, volume 1 2, New York: International Publishers.
Marx, Karl 1 976, 'Theses on Feuerbach' in Marx and Engels Collected
Work, volume 5, New York: International Publishers.
Marx, Karl 1 976, 'Manifesto of the Communist Party' in Mar and
Engels Collected Works, volume 6, New York: International
Marx, Karl 1 977a, Capital, volume I, New York: Vintage Books.
Marx, Karl 1 977b, Grundrisse, London: Penguin.
Marx, Karl 1 981 , Capital, volume III, New York: Vintage Books.
Marx, Karl 1 985, 'Value, Price and Profit' in Marx and Engels Collected
Works, Volume 20, New York: International Publishers.
Marx, Karl and Friedrich Engels 1 987, Marx and Engels Collected Works,
Volume 42, New York: International Publishers.
Thompson, E.P. 1 978, The Povert of Theor, New York: Monthly
A note on alienation
J ohn Holoway
There are two different ways of understanding alienation: as a condition
and as a struggle. On this distinction turns the whole theory and practice
of Marxism.
The centrality of the concept of alienation for Marxism derives from the
fact that it is a discussion of power, and Marxism is a theory of power.
The basic idea of alienation is that under capitalism our power (our
creative power, the power to control our lives - a power that is exclusively
human, since there is no god nor other external force) is alienated,
appropriated by others. We are dis-empowered. The alienation of power
over one's own life is the loss of oneself, what Marx calls self-alienation.
The 'act of estranging practical human activity, labour' means that a per­
son's activity exists 'as an alien activity not belonging to him; it is activity
as suffering, strength as weakness, begetting as emasculating, the work­
er's own physical and mental energy, his personal life - for what is life
but activity? - as an activity which is turned against him, independent of
him and not belonging to him' . ' Alienation means that activity exists in
the form of suffering - that is surely the central point - and strength in
the form of weakness.
If humanity is defined by activity - a basic presupposition for Marx -
then alienation means that humanity exists in the form of inhumanity,
that human subjects exist as objects. Alienation is the objectifcation of
the subject. The subject alienates her or his subjectivity, and the subjectiv­
ity is appropriated by an other. The existence of 'activity as suffering,
strength as weakness' means the attribution of human activity, human
strength to an alien instance: no longer god, but the modern god of capi­
ta! in its multiple forms (value, money, state, etc.) . At the same time as
the subject is transformed into an object, the object which the subject
produces, capital, is transformed into the subject of society. The objectifi­
cation of the subject implies a subjectifcation of the object. This is the
core of the Marxist critique of capitalism.
The more common way of understanding alienation i s t o see i t as an
analysis of the human condition under capitalism. Our existence is a
alienated existence, capitalist society is an alienated society, we are
objects, we suffer, it is capital which is the subject, capital which acts.
This interpretation of alienation has deep roots. It is widely assumed
Note on aUenation
in Marxist discussion that the power of capital (or possibly the power of
the capitalist state or a particular fraction of capital) is the starting point
for refection. The idea that capital is the only subject is something that is
simply assumed in many analyses of capitalist development: the very con­
cept of 'Marxist economics' (instead of the Marxist critique of econom­
ics), for example, is based on the idea that capital is the subject of this
society. If capital is the subject of power, then we clearly are taken to be
the objects of that power, to be impotent, the victims of capital. Very easi­
ly, the theory of alienation becomes transformed into the ideology of the
victim, the lament so loved by the Left.
This interpretation of alienation implies a
lear separation between
alienation and non-alienation. The concept of alienation involves,
inevitably, the notion of non-alienation. We can speak of alienation only if
we have some idea of the opposite, of lack of alienation. If alienation is
understood as a description of the present, then non-alienation presum­
ably refers to the future (or possibly to the past) . This interpretation sug­
gests, therefore, a stark contrast between the present (the alienation
which characterises capitalism) and the future (communism defined as a
non -alienated society) .
However, in order to have a concept of a future without alienation and
to be able to criticise present alienation, it is necessary that there should
be somebody now who has been liberated from that aienation. The inter­
pretation of alienation as a condition implies, therefore, a distinction
between the emancipated (ourselves, of course) who have succeeded in
liberating ourselves from that condition (at least conceptually) and the
others, the masses, the common people. It implies a distinction between
true class consciousness (ours) and the ordinary or false consciousness of
the working class 'man in the street'. They suffer, we act. We understand,
they do not.
If alienation is the condition of present-day society, then there are two
possible ways of thinking about a transition to the non-alienated society
of the future. Either such a transition is impossible, there is no way out of
alienation, we who understand can only criticise, but without hope (the
position associated with the Frankfurt School, or with many left acade­
mics) ; or the only way of conceiving of the revolution is in terms of the
leadership of the emancipated - the vanguard party. Both concepts, the
pessimism of the Frankfurt School and the revolutionary optimism of the
Leninist tradition, derive from the same elitist interpretation of alienation
as a condition. Both avoid asking themselves the obvious question: how
did we, the emancipated, manage to escape fom the smog of alienated
There is, however, another possible interpretation of the concept of alien­
ation. That is to see it not as a condition but as a process, as a constant
In his discussion of alienation, Marx emphasises the importance of
understanding alienation in terms of activity. It is the alienation of work,
of practical human activity, which is the key to understanding the other
aspects of alienation, such as private property. In the last lines of the sec­
tion on estranged labour in the 1844 Manuscripts, Marx contrasts the
Historical Materialism
significance of alienation for the worker and for the non-worker (capital­
ist): 'it has to be noted that everything which appears in the worker as an
activity of alienation, of estrangement, appears in the non-worker as a
state of alienation, of estrangement'.2 That suggests a contrast between
the perspective of the non-worker or capitalist, according to which alien­
ation is a state or condition, and the perspective of the worker, for whom
alienation is an activity, a process.
Now, if alienation is an activity, then this implies that the worker is
constantly engaged in the production of his or her own alienation. It is
the activity (or doing) of the worker which produces his or her suffering
(passivity), it is the strength of the worker that produces his or her weak­
ness. It is the worker who produces private property and therefore capi­
tal. 'Just as he creates his own production as the loss of his reality, as his
punishment; his own product as a loss, as a product not belonging to
him; so he creates the domination of the person who does not produce
over production and over the product'. 3 Alienation is the production of
capital by the worker. From this follow two important points. The first is
that capital depends on labour. The daily existence of capital depends on
the daily repeated alienating activity of the worker, on the alienation of
work. And secondly, if this alienation is understood not as a state or con­
dition but as an activity, this means that it is not pre-determined. Activity
implies uncertainty, the possibility of failure, of openness. Alienation,
understood as activity, is always in dispute. The existence of capital,
therefore, depends on an alienation of work that is always in dispute. In
other words, alienation is the struggle of capital to survive, the struggle of
capital to subordinate labour, the everyday struggle by capital to over­
come its dependence on labour, to transform our doing into suffering,
our activity into passivity, our strength into weakness. Alienation is class
struggle, the struggle by capital to take away our power, the power of
labour; just as the alienated (or fetishised) forms analysed by Marx in
Capital (value, money, etc.) are forms of that struggle. Alienation is capi­
tal's unceasing struggle for power. Alienation is not an aspect of class
struggle: it is the struggle of capital to exist.
To understand alienation in this way implies a complete shift in perspec­
tive. If alienation is understood as a condition, the question of dis-alien­
ation is marginal, a matter for the future. And if disalienation is a matter
for the future, then present struggle can be understood as a means to
achieve the end of disalienation, a means which can be separated from
that end.
If, on the other hand, one understands alienation as present struggle,
then the question of disalienation (what the Zapatistas call, more elegant­
ly, 'dignity') is the question of how we live and struggle now. If the repro­
duction of capital depends on the struggle of alienation, then our strug­
gle, the struggle of labour against capital, is the struggle of disalienation.
Disalienation is not something in the future, it is not the post-revolution­
ary condition that we shall reach after passing through this vale of tears;
nor is it the privilege of the enlightened, of the emancipated few. On the
contrary, disalienation is here, now, in our existence as insubordinate
labour, in our existence not only within but against capital. Disalienation
Note on aUenation
is the unceasing rebellion of activity against passivity, of doing against
1 . Marx 1 975, p. 275
2. Marx 1975, p.282; emphasis in the original.
3. Marx, 1975, p. 279.
Marx, Karl 1 975, 'The Economic and Philisophical Manuscripts' in Mar
and Engels Collected Works, volume 3, London: Lawrence and
New from Workers' Libert:
Russia: From October to the Gulag
Lost Texts of Critical Marxism
To mark te 80t an iversary of te 1917 Revolution we reprint long
unavailable articles - and some classics - on both the Revolution and its
degeneration - including pieces by Max Shachtman, Hal Draper, Leon Trtsky,
Joseph Carter and James P. Cannon.
This book is essential for all those who want to debunk the prevailing
myths of Marxist theory and practice and recover the lost insights of some
of the century's most radical thinkers.
256 pages. Available now for only £5.99, including postage and packing,
from W Publications, PO Box 823, London SE15 4NA. Or call 0171-207-4673.
Cheques payable to W Publications'.
Workers' Liberty, PO Box 823, London SE15 4NA
Phone +4 171 207 4673/3997/0706/1688
Fa +4 171 277 8462
Globallsation: states, markets and
class relations
Peter Burnham
I ntrodudion
The concept of 'globalisation' increasingly dominates economic and polit­
ical debate in the 1 990s. However, despite a profusion of commentaries
and case studies on aspects of 'globalisation' such as 'Japanisation',
'Americanisation', 'McDonaldisation' and, of course, global information
technologies, there are few radical interrogations of the notion of 'globali­
sation/internationalisation' and little discussion of the theoretical implica­
tions of recent changes in the global political economy (GPE) . The cen­
tral argument of this paper is that in order to make sense of these devel­
opments a broad focus is required which begins by conceptualising the
changing nature of relations between national states in the global econo­
my and concludes by understanding these relations in class terms. 1 This is
not simply to restate the importance of an international relations or inter­
national political economy 'dimension', since these 'disciplines' fail
absolutely to relate 'interstate' restructuring to the re-composition of
class relations. Rather, the aim of the paper is to prompt a more general
theoretical reorientation towards understanding the process of interna­
tional restructuring as one undertaken by national states in an attempt to
re-impose tighter labour discipline and recompose the labour/capital rela­
tionship. My starting point, therefore, is that global capitalism is still
structured as an antagonistic state system, and that many of the changes
which characterise the global political economy are introduced by states
in an attempt to solve problems that have their roots in labour/capital
conflict. In summary form, the paper concludes that the concept of 'glob­
alisation' obscures more than it reveals and that Marx's understanding of
the relationship between labour, capital and the state remains a more pro­
ductive starting point for analysing contemporary global processes.
The first part of the paper discusses three popular approaches to
understanding the changing character of national states in the GPE.
Arguments premised on notions of the 'post-Fordist state' (Jessop), the
'new global order/withering away of the state' (Ohmae, Strange and
Held), and the 'internationalisation of the state' (Cox) are reviewed and
found wanting in three main respects. Firstly, they all counterpoise the
state and the market as two opposed forms of social organisation stress­
ing how globalisation gives the market 'power over' the state with' a
resulting loss of 'sovereignty' and 'national autonomy'. The weakness of
this view is that ' states and markets' are conceived in a reified 'thing-like'
manner brought into a purely external relationship with each another. I
1 1
will argue that notions of externality and structure should be replaced by
the dialectical categories of social process and contradictory internal rela­
tionship. In this way, we can understand the social constitution of states
and markets as differentiated aspects of the same fundamental social rela­
tions. Secondly, the approaches reviewed overestimate the extent to
which national states could 'control' capital before 1 973. In this sense, by
invoking historically inaccurate notions such as the Keynesian Welfare
State, they support the increasingly implausible idea of a watershed in the
global political economy occurring at some point in the 1 970s. Finally, if
labour/capital relations are deemed to deserve a mention, they tend to be
seen as external to the process of restructuring.
The second part of the paper offers an alternative view of the relation
between states and markets in the GPE which stresses the internal and
necessary nature of their interrelation and sees both as aspects of the
social relations of capitalist production. The global re-composition of
national states is not a process which diminishes state 'power' - to accept
this is to adopt a reified 'thing-like' view of the state) - but is in many
ways a state-led initiative whose primary aim is to restructure labour/cap­
ital relations. The fnal part of the paper makes this argument explicit by
drawing briefly on examples from Britain's membership of the Exchange
Rate Mechanism 1 990-92 and current debates which focus on the move­
ment towards central bank independence across the globe in the late
1 990s.
Part I: Three fallacies regrding the state in the global economy
1. The post-Fordist fllacy
Favoured by regulation theorists, state theorists, political sociologists and
many within the field of industrial relations, the notion of 'post-Fordism'
has led a charmed existence.2 Bob Jessop provides the clearest and most
sophisticated thesis claiming that the post-Fordist accumulation regime
associated with the interationalisation of production now requires the
establishment of a 'post-Fordist state'. ' The post-Fordist state is hyper­
liberal and is increasingly 'hollowed out' from below, from above and
horizontally. Some national state capacities are transferred to pan-region­
al or international bodies, others devolved to local levels within the
national state, and yet others are usurped by emerging local and regional
horizontal networks which by-pass central states and connect localities
and regions in several nations. In this form, Jessop's thesis appears deter­
minist and mono-causal and to suffer from the defects of structural func­
tionalism (admittedly he does term his analysis a 'thought experiment') .
State form is determined automatically by the 'needs' of the accumulation
regime· and identifying the way in which these needs are met apparently
constitutes an explanation of the social change and institutional arrange­
ments. Class relations and, by implication,
lass struggle are viewed as
external to the process of restructuring. Labour and the state itself are
depicted as powerless, passively responding to the demands of a unex­
plicated post-Fordist economy. Moreover, the 'post-Fordism thesis' relies
on a caricature of historical development predicated on identifying water­
sheds/sharp breaks in labour/capital/state relations and awkward ideal
types such as the Keynesian Welfare State.4
I S5s
Historical Materialism
2. The new global order Iwithering away of the state fllac
This popular approach builds on aspects of the post-Fordist state argu­
ment combined with liberal transformationist writing which sees the end
of the state as nigh. Recent arguments by Kenichi Ohmae and Susan
Strange point in this direction. This argument is developed albeit some­
what cautiously by David Held who notes that the existing strategies and
practices of the Left are outmoded. Current strategies may have been
appropriate to an earlier 'nation
l' stage of capitalism, but these are now
rendered obsolete by globalisation.5 The new global order has escaped the
democratic control of national governments, and so democracy must now
become a 'transnational' affair orchestrated through 'transnational civil
society' . Again, we are presented with a view of the state as powerless and
external to the market, with 'globalisation' tipping the balance towards
the market against the ailing democratic state. This extreme version of
the market power argument rests on a Fortune magazine/Economist view
of the internationalisation process. It seems unproblematically to main­
tain that we live in a borderless and stateless world dominated by stateless
firms with unlimited global options. Claims about the 'democratic' nature
of the nation-state prior to globalisation are grossly idealised and under­
estimate the importance of the state debates of the 1 970s and 1 980s, the
most penetrating of which underscored the enduring class character of
the capitalist state in whatever guise.6
3. The internationalisation of the state fllacy
Pioneered by Robert Cox and the neo-Gramscian school of international
political economy (IPE), also favoured by progressive liberals, the 'interna­
tionalisation of the state' is an idea which, at face value, has much to com­
mend it.7 Cox sensibly argues that we are witnessing the transformation
(the reorganisation of structure and role both internally and externally)
not the destruction of the state. His position rests on three points. Firstly,
states historically have acted as buffers/bulwarks protecting national
economies from disruptive external forces in order to sustain domestic
welfare/employment. Since 1 973 this priority, he notes, has shifted to one
of adapting domestic economies to the perceived exigencies of the world
economy. Secondly, this shift has affected the structures of national gov­
ernments. Agencies that act as conduits for the world economy have
become pre-eminent within governments (ministries of industry and
labour are now subordinate to ministries of finance) . Finally, we have a
transnational process of consensus formation in organisations such as the
International Monetary Fund (IMF), Organisation for Economic Co­
operation and Development (OECD) or the G7 that transmits guidelines
to dominant state agencies which in turn enact national policies. The
state's role is therefore one of helping to adjust the domestic economy to
the requirements of the world economy. The state is a transmission belt
from world to domestic economy - it has become 'internationalised' from
the 'outside-in'. It is a tributary to something greater than the state,
orchestrating 'governance without government' . Cox's thesis provides sug­
gestive themes for further research but his argument fails to convince on a
number of grounds. Historically, when since 1 900 did states pursue
domestic welfare as their centrl priorit? In terms of a theory of policy
making, Cox sanctions a simplistic fractionalist view of policy making at
the behest of fnance capital. He underplays the extent to which 'globalisa­
tion' is authored by states and is regarded by state agents (both liberal
market and social democrat) as one of the most efcient means of restruc­
turing labour/capital relations to manage crisis in capitalist society.s
Part I I : The recomposition ofthe
state and the restructuring of labour I capital
The most significant theoretical problem exhibited by the approaches men­
tioned briefly above is that they view the state and the market as opposed
forms of social organisation with globalisation tipping the balance in
favour of the market against the state. In this regard, they all follow
implicitly the liberal/realist position, which separates social reality and
looks for external linkages between disaggregated phenomena.9 However,
rather than viewig the relationship between states and markets as external
and contingent, an alterative methodology rooted in Marx's work is to
start from the premise that the relationship is internal and necessary
although, of course, the institutional form of this relationship varies given
the historical character of the class struggle. States are an aspect of the
social relations of production - a differentiated form of those relations.
Their 'power' derives fom the ability to reorganise labour/capital relations
within and often beyond their boundaries in order to enhance the accumu­
lation of capital both domestically and globally. 10 This interpretation of
Marx's approach to the state has most consistently been developed under
the auspices of the Conference for Socialist Economists and in the joural
Capital and Class. Rooted in Marx's account of the 'fetishism of com­
modities', this approach dissolves the state as a category and sees it as a
'rigidifed' or 'fetishised' form of social relations. I I National states exist as
political 'nodes' or 'moments' in the global flow of capital and their devel­
opment is therefore part of the antagonistic and crisis-ridden development
of capitalist society. Recent changes in the GPE are thus predominantly
about reorganising rather than bypassing states. This recomposition is
undertaken actively by states as part of a broader attempt to respond to a
crisis of labour/capital relations - manifest in national terms as fscal cri­
sis, declining productivity, lack of competitiveness etc.- and to restructure
those relations. In the last 25 years, the strategies adopted by 'advanced'
capitalist states to 'manage' labour and money look remarkably different
from the received models of post-war regulation. In many cases, overtly
'politicised' forms of labour management and inflation control involving
voluntary restraint, incomes policies and the machinery of tripartism, have
now given way to 'depoliticised' forms of management. These 'marketise'
aspects of state activity and shift responsibility for management onto inter­
national regmes and independent organiations and in this way 'interna­
tionalise' economic policy making to the advantage of national govern­
ments. This, of course, is a long-standing strategy of national states seen
for instance in the return to the Gold Standard in the 1 920s wt its asso­
ciated claims of 'automatic regulation'. The distinctive aspect of current
developments, however, is the way in which states are able to capitalise on
changes which have occurred in fnanci
l markets in the last few decades,
to shore up demands for low infation strategies which are focused primar­
ily on capping wage settlements.
In the last 20 years, the re-regulation of financial markets, in the
Historical Materialism
shape of increased capital mobility and a genuinely integrated global mar­
ket, has undoubtedly affected the fiscal and monetary choices open to
goverments.12 Since foreign exchange dealers prefer to hold currencies
backed by anti-infationary policies, the search for counter-infationary
credibility is of paramount importance both rhetorically and materially.
Whilst in the post-war era, currency markets were driven largely by cur­
rent account imbalances, now interest rate differentials are the prime
determinant of exchange rate movements. The 1 980s and 1 990s have
seen a general convergence in interest rates and any state which adopts a
significantly looser monetary policy than the prevailing level risks a
depreciation in its currency. Since the sheer scale of fows in the foreign
exchange markets rules out reserve intervention as anything more than a
short-term policy, monetary policy is, for the moment, inextricably tied to
exchange rate management. The net result of this fiscal and monetary
environment is a clear incentive for governments to enhance their
counter-infationary credibility. 13
Whilst it is common to view these developments as giving 'power' to
markets over states, an alternative would be to see the re-regulation of
fnancial markets as providing the strongest possible public justification
governments can muster for maintaining downward pressure on wages to
combat infation and thereby achieving price stability. States, of course,
were responsible for the creation of the Eurocurrency markets that devel­
oped largely out of interstate rivalry focused on finance. 14 In many key
ways, therefore, the re-regulation of financial markets enhances the
'power' of the state vis-a-vis the working class since it can be argued
forcefully that price stability really is the crucial determinant in the GPE
and lack of 'competitiveness' translates directly into a loss of jobs and
In this environment where the maintenance of price stability is publicly
acknowledged as the first objective of monetary policy, and high wage
settlements are seen as a major cause of infation, governments across the
world in advanced capitalist states have attempted to control inflation by
adopting 'rules -based' rather than ' discretion-based' economic
strategies. I' 'Rules based' approaches attempt to build counter-inflation­
ary mechanisms into the economy by off-loading part of the governmen-
1's responsibility for economic policy onto a non-governmental body. This
can be achieved in two ways. Firstly, an international regime, usually an
international monetary mechanism (Gold Standard, Exchange Rate
Mechanism), that sets definite rules, can be held responsible. This
attempts to build 'automaticity' into the system formally limiting govern­
ment room for manoeuvre. The second strategy is to off-load responsibil­
ity onto a national body, which is given a definite role in statute and
thereby greater independence from the government (for example moves
towards central bank independence) . Whereas 'rules-based' strategies are
attempts to 'depoliticise' the government's economic policy making,
thereby shielding the government from the political consequences of pur­
suing deflationary policies, 'discretion-based' approaches are highly
politicised since national governments play the central role in controlling
inflation, usually through formal incomes policies. Whilst 'discretion­
based' strategies offer maximum room for manoeuvre and enable govern­
ments to gain immediate credit for successful outcomes, they also carry a
higher risk that economic crisis will become a political crisis of the state
itself. In Britain following the governing cri ses of the
Heath/Wilson/Callaghan years, 'discretion-based' strategies have gradu­
ally, and falteringly, been replaced by the search for policies grounded in
more publicly credible 'rules' . In the context of the last twenty years this
shift is clear and was emphasised dramatically by Gordon Brown's deci­
sion to grant 'operational independence' to the Bank of England within
five days of becoming Chancellor. Nonetheless, this process is not depict­
ed accurately in terms of unilinear historical development. State manage­
ment of labour and money proceeds via a series of oscillations to secure
the maximum effectiveness of specifc policies: from fxed to foating to
fixed exchange rates; fom rules to discretion to rules; tighter and looser
monetary policies etc. Although much of this may seem somewhat
removed from a discussion of the restructuring of class relations, the real
signifcance of the shif from 'discretion' to 'rules', or from politicised to
de-politicised management, lies in how governments use the language of
'external commitments' to legitimate the recomposition of labour/capital
relations in the guise of global competitiveness.'
These remarks are best
illustrated with a brief example drawn from the experience of the British
state in the 1 990s.17
Part I I I : From ERM to central bank independence
Since 1 990 the British state has sought to externalise the imposition of
monetary discipline firstly through membership of the European
Exchange Rate Mechanism (ER) and then by moving, at least rhetori­
cally, to restructure the institutional relations between the Treasury and
the Bank of England. John Major and Nigel Lawson agreed with the
Bank of England in early 1 990 that Britain would reap substantial bene­
fits from joining the ER. In particular, by ruling out periodic exchange
rate realignments, both sides of industry, in Major's view, would be
forced to face the long-standing problem of inflationary wage settlements.
Margaret Thatcher's remedy for the 1 980s, that policy should focus on
the control of monetary aggregates (the Medium Term Financial
Stategy), weakened as the Government found it increasingly diffcult to
control the money supply and as different measures of money contradict­
ed each other . 18 The prerequisite for sustained ER membership was a
reduction in unit labour costs through lower wages and the intensifcation
of work. The Bank of England made this quite clear: 'The Governor has
emphasised that henceforth companies can have no grounds for expect­
ing a lower exchange rate to validate any faiure to control costs. The
greater stability which ER membership offers Sterling against other
European currencies should, in itself, be welcome to business as it will
enable frms to plan and invest with greater certainty. If companies recog­
nise that they are now operating under a changed regime, the benefits of
lower infation will accrue sooner, and at a lower cost in terms of lost
output, than could otherwise be expected. But if they fail to recognise the
constraints under which they now operate the outcome wlprove painful
to them." 9 With inflation rising, the ER offered the government the
opportunity to have monetary discipline 'implemented from without' .20
Clearly, the Major administration hoped it could be insulated from the
unpalatable consequences of 'economic adjustment' by shifting responsi-
Historical MatriaUsm
bility onto an international regime. The government hoped that ERM
membership would force employers to compensate for the high interest
rate pressure on profts by confronting their labour force to secure lower
wage rates and increase output per worker. A falling exchange rate would
no longer compensate sluggish productivity or enable wage negotiators to
agree 'unacceptably' high claims. In essence, the ERM replayed the
episode of Britain's return to the Gold Standard in 1 925. The 'politics of
austerity' could now be legitimated in the language of globalisation with
'exteral commitments' uppermost. In many respects, and from the vew­
point of the governors, ERM membership 'worked'. Infation fell from
10.9% in October 1 990 to 3. 7% by September 1 992. Norman Lamont
heralded the 'sea-change' in attitudes to inflation that occurred in Britain,
particularly amongst trade union leaders, since joining the wide band in
October 1 990. But above all else, this act of de-politicisation enabled
Major to preside over the second worst recession since the Second World
War and survive for a further term of offce. In addition to undermining
resistance to government policies (signifcant anti-poll tax riots occurred
in April 1 990) , it secured wage restraint forcing British workers into
deteriorating conditions for longer hours.
Following Britain's exit from the ERM the government began a search
for a 'new framework', a 'new anchor', in short, for another strategy to
depoliticise economic policy. On 8 October 1 992, in a memo to the
House of Commons Treasury and Civil Service Committee, Lamont set
out the policy framework, which would replace the ERM. Counter infa­
tionary credibility would now be sought by restructuring the institutional
relationship between the Treasury and the Bank of England in order 'to
make the formation of policy more transparent and our decisions more
accountable'.21 In November 1 992 Lamont set an inflation target of 1-4%
and asked the Bank to assess infation prospects in quarterly independent
reports. With Kenneth Clarke as Chancellor in May 1 993, the Bank was
given the right of deciding the timing of interest rates. Finally, in
February 1 994, Clarke outlined a new framework for monetary policy
decision making. Decisions concerning interest rates would be taken at
meetings between the Chancellor and the Governor and minutes would
be published with some time delay. Well-publicised disputes between the
Chancellor and the Governor have questioned the ability of these halting
moves towards central bank independence to constitute an effective
counter inflationary anchor. It is for this reason that the government in
1 995 reviewed recommendations to enshrine price stability in statute as
the primary objective of monetary policy and follow the Bank of France
by creating an independent Monetary Policy Committee within the Bank
to oversee the making of policy. In the wake of Blair's victory in May
1 99 7, 'New Labour' completed the reforms begun by Lamont and
Clarke. The move to introduce 'operational independence', broadly along
the lines of the New Zealand model, has now left the Bank 'free' to pur­
sue an infation target laid down by a Monetary Policy Committee.
Globalisation, we are told ad nauseum, is a multifaceted phenomenon,
which has transformed both ' structure and agency' . At the heart of the
globalisation thesis is the caim that the relationship between state, capital
and class has changed fundamentally. States are either 'withe
ing away'
or becoming 'internationalised'. Capital is now global and 'markets' hold
and dispense 'power' in both economic and political arenas. The working
class must adjust to a new fexible insecurity and can no longer rely on an
increasingly powerless 'democratic government' to fght its corer.
Although this characterisation of contemporary society appears per­
suasive it is at the same time unsatisfactory. It fails to convince on
grounds of theoretical adequacy or historical accuracy and has little to
offer Marxian political economy. In theoretical terms the guiding princi­
ple of Marx' s analysis of capitalism is the historical specifcity of bour­
geois relations and the dynamic of the class struggle which produces con­
stant change giving rise to new social forms. In the Communist Manifesto
for instance, Marx emphasises the extent to which 'constant revolutionis­
ing of production, uninterrupted disturbance of all social conditions,
everlasting uncertainty and agitation distinguish the bourgeois epoch
from earlier ones'. In place of the old local and national seclusion we have
'intercourse in every direction, universal interdependence of nations' and
the creation of new industries 'whose products are consumed not only at
home, but in every quarter of the globe'. The strong globalisation thesis
depends on a stagist theory of history (from 'national' to 'global') which
it seems was redundant as early as 1 848! In essence, the capital relation
has always been a global relation. Similarly, the thesis rests on a static
theory of the state, which implicitly identifes statehood with nationalism
and state power with control of territorial resources. The error of this
view was identified in the early 1 970s when Poulantzas corrected an early
globalisation thesis, noting that 'institutions or apparatuses do not 'pos­
sess' their own 'power', but rather express and crystallse relationships of
class power' .22 The globalisation thesis conceptually divides state and
market without investigating the underlying social relations which consti­
tute these phenomena. By viewing states and markets as differentiated
forms of the same fundamental social relations we reinstate the class
character of both and emphasise the weakness of counterpoising the
'democratic state' and the unbridled avarice of the market.
Marx's approach identifies the internal and necessary unity of state
and market in capitalism. Space is thereby opened up to view new devel­
opments in the GPE in terms of the reorganisation rather than the
bypassing of states. Moreover this re-composition is undertaken activ
by states as part of a broader attempt to restructure and respond to crisis
in capitalist society. From the viewpoint of state managers, 'politicised'
forms of labour management and infation control, involving voluntary
restraint, incomes policies and the machinery of tri-partism have, for the
moment, given way to 'de-politicised' forms of management. These mar­
ketise aspects of state activity and publicly shift responsibility for manage­
ment onto 'external' regimes and independent organisations. In this way
de-politicisation strategies aim to capitalise on recent changes in the GPE
to the advantage of national governments.
Through depoliticisation and externalisation, states have found a novel
(but not historically unique) way in which the mythical 'national interest'
can be brought to bear on wage settlements. It is important that radical
economists are awakened to viewing current changes as tactical moves,
albeit made
lumsily and hesitantly by governments who are faced with
Historical Materalism
permanent crisis management. The technicalities of new currency bands,
central bank independence, the re-regulation of financial markets and
single currencies should not obscure our perception that these are means
of depoliticisation designed to achieve the subordination of labour to cap­
italist command. 'Globalisation' is therefore as much a state strategy as it
is a market one.23 Although individual governments may be bucked by
markets, market-based solutions offer government opportunities to recast
labour/capital relations, ofen in the guise of controlling inflation, without
resorting to the overtly 'politicised management' of the 1 960s and 1 970s.
In short, it is much too simplistic to see 'globalisation' as transferring
'power' from states to markets. The task for Marxian political economy is
to retain a focus on the centrality of labour/capital/state relations and
conceptuaiise global processes in terms of the re-composition of relations
of domination and struggle.
1 . See also Harman 1 996.
2. See contributors to Bonefeld and Holloway (eds.) 1 991 .
3. Jessop 1 994.
4. For a general critique see Matthews 1 968; Tomlinson 1 981 ; and Glyn 1995.
5. Ohmae 1990 and 1995; Strange 1 996; Held 1 995.
6. For a good overview see Clarke 1 991 .
7. Cox 1 996.
8. See Panitch 1 994; Wolf 1 997. For an extended discussion of the limits of 'frac­
tionalist' approaches to policy-making see Clarke 1 978. A more general cri­
tique of Cox and 'neo-Gramscian' international relations theory can be found
in Burham 1 991 .
9. I n brief, as Gilpin 1987 argues, most orthodox international relations theorists
are 'liberal' in their approach to the 'economy' and 'realist' in their analysis of
the state. For an excellent critique of this orthodoxy in international relations
see Rosenberg 1 993.
10. For much greater elaboration see Holloway and Picciotto 1 977; Barker 1978;
Clarke 1 988; Burnham 1 990 and 1 995; Bonefeld, Brown and Burnham 1 995.
1 1 . See Holloway 1994.
12. The term 're-regulation' is preferred to 'de-regulation' since in the last twenty
years we have seen a complex process of the drafting of new regulations (often
new market-oriented rules) rather than a simple lifting of regulations. See
Cerny 1 993.
13. For further elaboration see Thompson 1994; also the debate between Hirst and
Thompson 1996, and Radice 1 997.
14. See Va Dormael
1 5. Kydland and Prescott 1 977; Keech 1992.
1 6.Also see Gil 1 995.
1 7. For a much fuller account see Bonefeld and Burnham 1 996.
18. See Bonefeld 1993; Grant 1 994.
19. Quoted in Smith 1 993.
20. Sandholtz 1 993.
21 . Jay 1 994.
22. Poulantzas 1973.
23. Pantch 1 994.
Barker, Colin 1978, 'A Note on the Theory of Capitalist States', Capital
and Class, 4: 1 1 8-26.
Bonefeld, Werner 1 993, The Recomposition of the British State During
the 1980s, Aldershot: Dartmouth.
Bonefeld, Werner and John Holloway (eds) 1 991 , Post-Fordism and
Social Form: A Marxist Debate on the post-Fordist State, London:
Bonefeld, Werner, A Brow and P Burnham 1 995, A Major Crisis? The
Politics of Economic Policy in Britain in the 1 990s, Aldershot:
Bonefeld, Werner and Peter Burnham 1 996, 'Britain and the Politics of
the European Exchange Rate Mechanism', Capital and Class, 60:
Burnham, Peter 1 990, The Political Economy of Postwar Reconstruction,
London: Macmillan.
Burnham, Peter 1 991 , 'Neo-Gramscian Hegemony and the International
Order', Capital and Class, 45: 73-94.
Burnham, Peter 1 995, 'Capital, Crisis and the Interational State System'
in Global Capital, National State and the Politics of Mone, edited by
Werner Bonefeld and John Holloway, London: Macmillan.
Cery, Philip (ed.) 1 993, Finance and World Politics: Markets, regimes
and states in the post-hegemonic era, Aldershot: Edward Elgar.
Clarke, Simon 1 978, 'Capital, Fractions of Capital and the State', Capital
and Class, 5: 32-77.
Clarke, Simon 1 988, Kenesianism, Monetarism and the Crisis of the
State, Aldershot: Edward Elgar.
Clarke, Simon (ed.) 1 991 , The State Debate, London: Macmillan.
Cox, Robert and Timothy Sinclair 1 996, Approaches to World Order,
Cambridge: Cambridge University Press.
Glyn, Andrew 1 995, ' Social Democracy and Full Employment', New Left
Review, 2 1 1 : 33-55.
Gill, Stephen 1 995, 'Globalisation, Market Civilisation and Disciplinary
Neoliberalism', Millnnium, 24: 3: 399-423.
Gilpin, Robert 1 987, The Political Economy of Interational Relations,
Princeton: Princeton University Press.
Grant, Wyn 1 993, The Politis of Economic Polic, Hemel Hempstead:
Harman, Chris 1 996, 'Globalisation: a critique of a new orthodoxy',
Interational Socialism, 73: 3-34.
Held, David 1 995, Democrac and the Global Order, Cambridge: Polity.
Hirst, Paul and Grahame Thompson 1 996, Globalization in Question: the
interational economy and the possibilities of goverance, Cambridge:
Holloway, John 1 994, 'Global Capital and the National State', Capital
and Class, 52: 23-50.
Holloway, John and Sol Picciotto 1977, ' Capital, Crisis and the State',
Capital and Class, 2: 76-1 01 .
Jay, P. 1 994, 'The Economy 1 990-1 994', in The Major Efect, edited by
Historical Materialism
Dennis Kavanagh and Anthony Seldon, London: Macmillan.
Jessop, Bob 1 994, ' The Transition to Post - Fordism and the
Schumpeterian Workfare State' in Toward the Post-Fordist Welfare
State, edited by Roger Burrows and Brian Loader, London:
Keech, W 1 992, ' Rules, Di scretion, and Accountability in
Macroeconomic Policymaking', Goverance, 5: 3: 259-78.
Kydland, Finn and E Prescott 1 977, 'Rules Rather Than Discretion',
Joural of Political Economy, 85: 3.
Marx, Karl and Friedrich Engels 1848, The Communist Manifesto, vari­
ous editions.
Matthews, R 1 968, 'Why Has Britain Had Full Employment Since The
War?', Economic Joural, 78: 555-69.
Ohmae, Kenichi 1 990, The Borderless World: power and strategy in the
interlinked economy, New York: Fontana.
Ohmae, Kenichi 1 995, The End of the Nation State: the rise of regional
economies, New York: Free Press.
Panitch, Leo 1 994, 'Globalisation and the State', Socialist Register, 1 994:
Poulantzas, Nicos 1 973, 'L'Internationalisation des Rapports Capitalistes
et l'Etat-Nation', Les Temps Moderes, February.
Radice, Hugo 1 997, 'The question of globalization', Competition and
Change, 2: 2.
Rosenberg, Justin 1 993, The Empire of Civil Society: a critique of the
realist theor of interational relations, London: Verso.
Sandholtz, Wayne 1 993, ' Choosing Union: monetary policy and
Maastricht', International Organisation, 47: 1: 1-39.
Smith, Da
vid 1 993, From Boom to Bust: trial and error in British eco­
nomic polic, London: Penguin.
Strange, Susan 1 996, The Retreat of the State: the difsion of power in
the world economy, Cambridge: Cambridge University Press.
Thompson, Helen 1 994, Jining the ERM: executive decision-making in
the UK, 1 979-1 990, London: unpublished PhD thesis, London
School of Economics and Political Science.
Tomlinson, Jim 1 981 , 'Why was there never a Keynesian Revolution in
Economic Policy?', Economy and Society, 10: 1 : 72-87.
Van Dormael, Armand 1 997, The Power of Mone, London: Macmillan.
Wolf, M 1997, 'Far from powerless', Financial Times, 1 3th May.
The rate of profit and
economic stalnation in
the United States economy
Fred Moseley
In the frst thirty years afer World War II, the US economy performed
very well.' The rate of growth averaged 4-5%, the rate of unemployment
was seldom above 5%, infation was almost non-existent ( 1 -2%), and the
living standards of workers improved steadily. These were the 'good old
days'. However, this long period of expansion and prosperity ended in the
1 970s. Since then, both the rate of unemployment and the rate of infla­
tion have been much higher than before, and the average real wages of
workers (Le. the purchasing power of wages) have declined some 20%.2
Productivity growth has also slowed down and the debt burden of both
capitalist enterprises and the Federal government has increased dramati­
cally. It is in this sense that we may refer to the 'economic crisis' of the
US economy over the last two decades. This crisis has certainly not been
as severe as the Great Depression of the 1 930s, but the economic perfor­
mance has been signifcantly worse than in the early post-war period.
1. The decline of the rate of proft
In my view, the main underlying cause of this poor performance of the
US economy over the last two decades was a very significant decline in
the rate of proft in the economy as a whole. The rate of proft is the main
determinant of the overall healh of capitalist economies. When the rate
of profit is high, capitalist economies are generally more prosperous:
business investment is high, unemployment is relatively low, and the liv­
ing standards of the population generally rises (such as occurred in the
frst 30 years after World War II) . On the other hand, when the rate of
proft is low, prosperity turns into stagnation and depression: business
investment is low or non-existent, unemployment is high, and living stan­
dards decline (such as has occurred in the last 20 years and during the
Great Depression and 1 9th century depressions) .
In the post-war period the rate of proft in the US economy declined
very significantly - approximately 50% - from 1 950 to the mid- 1 970s.3
This trend is shown in Appendix 1 . This signifcant decline in the rate of
proft seems to have been a general world-wide trend during this period,
including all the major capitalist countries.
In my view, this very significant decline i the rate of proft was the
main cause of both of the so-called 'twin evils': higher unemployment and
higher infation, and hence also of the declining living standards of recent
decades. A in periods of depression in the past, the decline in the rate of
profit has reduced both the funds and the incentive for capital investment,
and has therefore resulted in its decline, which in tur has resulted in
Historical MateriaUsm
slower growth and higher unemployment. 4 One important new factor in
the post-war period is that many governments in the 1 970s responded to
the higher unemployment by adopting Keynesian expansionary policies
(more government spending, lower interest rates, etc.) in attempts to
reduce unemployment. However, these government attempts to reduce
unemployment generally resulted in higher rates of infation as capitalist
enterprises responded to the goverment stimulation of demand by rais­
ing their prices at a faster rate in order to reverse the decline in their rate
of profit, rather than by expanding output and employment. We often
hear orthodox economists talk about 'wage-push' infation (Le. that high­
er wages has 'pushed up' inflation) . I think that what has actually
occurred is 'profit-push' inflation, as businesses have attempted to
reverse the decline of the rate of profit and these attempts have 'pushed
up' infation.
In the 1 980s, fnancial capitalists revolted against these higher rates of
infation and have generally forced governments to adopt restrictive poli­
cies (less government spending, higher interest rates, etc.) . The result has
been less inflation, but also higher unemployment and reduced living
standards. Therefore, government policies have affected the particular
combination of unemployment and infation that has occurred, but the
fundamental cause of both of these 'twin evils' has been the decline in the
rate of proft. The restrictive fscal and monetary policies of recent years
have certainly contributed to higher rates of unemployment and falling
living standards. But these restrictive policies are not the fundamental
cause of the crisis. Rather, these policies are a response to the infationary
effects of the prior Keynesian expansionary policies, which themselves
were a response to the decline of the rate of profit.
I also argue that the decline of the rate of profit was the main undery­
ing cause of other important aspects of the current economic crisis.
Another effect of the decline of investment, which itself was caused by the
decline of the rate of proft, was slower productivity growth. Because cap­
italist enterprises invested less, they invested in new technology less. New
technology is the source of productivity growth. Therefore, the decline of
investment spending resulted in a slowdown of productivity growth. This
in turn contributed to the increase in the rate of infation. Orthodox
economists have not yet been able to explain the productivity slowdown
adequately: they still refer to the slowdown as a 'mystery'. But from the
point of view of Marxian theory, the productivity slowdown is not a 'mys­
tery'. The fundamental cause of the productivity slowdown was the
decline of the rate of profit which reduced capital investment in new tech­
One way that capitalist enterprises have attempted to maintain invest­
ment, in spite of the decline of the rate of profit, has been to rely more
and more on debt to finance capital investment. This strategy has main­
tained investment spending somewhat, but has also increased the debt
burdens of many capitalist enterprises. An increased debt burden means
an increased danger of bankruptcies in the event of fture recessions, and
thus also the dangers of a deeper depression with much higher rates of
unemployment. Increased debt makes it possible to prolong an expansion
in the short-run, but it also increases the danger of an even deeper crisis
i the long-run.
The Unite States eonomy
The disruptions of OPEC (Organisation of Petroleum Exporting
Countries)in the 1970s can also be understood as the result of the world­
wide decline of the rate of proft which also affected oil companies. Oil
companies are among the largest companies in the world. Five of the ten
largest companies in the US economy are oil companies. These compa­
nies responded to the decline of the rate of proft as all companies gener­
ally did: they devised strategies to try to restore their rate of profit. The
main strategy of the oil companies was to create an artifcial 'shortage of
supply' in order to increase the price of oil. This strategy was of course
very successful: the price of oil quadrupled, which in turn contributed
further to the increase in the general rate of infation. The oi-producing
countries of the Middle East gladly went along with this strategy, and
even took advantage of the situation to increase their share of oil profts,
at the expense of the oil companies. Orthodox economists treat the
OPEC disruptions as an 'exogenous shock' - an accident, due to political
causes. But from the point of view of Marxian theory, these OPEC dis­
ruptions were due to endogenous causes, and to the decline of the rate of
profit in particular. In the 1 970s, I wrote a paper entitled 'Capitalism is
running out of gas', in the sense that it was the real gas (or fuel) of capi­
talism - proft -which was declining. This was the fundamental problem
behind the 'oil shortage'.
When I first started studying economics, I was struck by how ofen I
heard the phrase ' exogenous shock', especially in macroeconomics.
Amost everything was explained in macro-economics by 'exogenous
shocks' : OPEC, bad government policies, bad weather, increases in
women ad young people in the labour force, etc. I began to get nervous
- when was the next 'exogenous shock' going to drop suddenly from the
sky? We shall see that this is in striking contrast to Marxian economics,
which is able to explain economic crises, not as accidents or due to exter­
nal causes or 'exogenous shocks', but rather as due to internal, systematic
causes, and in particular to the decline of the rate of profit.
It is striking how many of the most important phenomena of the cur­
rent economics crisis can be explained as a result of the decline of the
rate of proft. Therefore, the most important task in a further analysis of
the causes of the current crisis is to explain this signifcant decline in the
rate of profit A Marxian explanation of this decline of the rate is present­
ed in Section 2 below.
Capitalist enterprises have responded to this decline in the rate of
profit, not only by raising their prices at a faster rate, but also by various
other strategies to increase their rate of proft. The most important of
these has been the attempt to reduce the wages of workers by means of
direct wage cuts and/or by moving their operations to areas of the world
with low wages. This has been the main driving force behind 'globalisa­
tion': to increase the rate of proft of capitalist enterprises. These wage
cuts have of course contributed to the declining living standards of recent
Aother strategy has been to make workers work harder and faster.
Economists call this an increase in the 'intensity of labour' ; workers call it
'speed-up'. An increase in the intensity of labour increases the value pro­
duced by workers without increasing their wages, and therefore increases
the rate of proft. There are of course no direct estimates of the intensity
Historical MateriaUsm
of labour, but there is much indirect evidence that the intensity of labour
has increased significantly in many capitalist enterprises in recent years.
The higher rates of unemployment of recent years has itself contributed
to this increase in the intensity of labour, as workers have been forced to
compete with each other for the fewer available jobs by working harder.
One common business strategy has been to 'downsize' - i.e. lay off
10-20% of a firm's employees and then force the remaining employees to
do the work of the laid-off employees. This method also generally
increases the intensity of labour even before the workers are laid off, as
all workers work harder so that they will not be among those who are laid
NAFTA (The North American Free Trade Agreement) has been
another strategy by US capitalists to increase their rate of proft. Above
all else it is an attempt by US companies to restore their profitability. US
companies hope that NAFTA will increase their rate of profit by giving
them free access to Mexican markets and to cheaper Mexican labour.
However, this increased competition from US companies will also have
inverse negative effects on Mexican companies and will force many of
them into bankruptcy (especially small and medium-sized business) .
Therefore, any increase in the rate of profit of US companies as a result
of NAFTA will come at the expense of Mexican companies. In this way,
the worst effects of the crisis of profitability are shifted from the US to
Therefore, we can see that the attempts by business enterprises to
increase their rate of profit in recent decades has in general been very
painful for workers - it has resulted in higher unemployment, higher
inflation, lower living standards, and increased stress and exhaustion on
their jobs. Marx's general law of capital accumulation - that the accumu­
lation of wealth by capitalists is accompanied by the accumulation of mis­
ery by workers - has been all too true in recent decades.
However, the startling fact is that, despite the decline in real wages
and the increase in the intensity of labour, the rate of profit in the US
economy still has not increased signifcantly. The rate of profit i n the US
today is still 30-40% below the higher levels of the early post-war period
(see Appendix 1 ) .5 The restrictive government policies of recent years
have still not achieved their main goal: to increase the rate of proft. To
use my gasoline metaphor again: capitalism is still 'low on gas', or low on
its essential fuel of proft. This lack of a significant increase in the rate of
proft is the main reason why the performance of the US economy has
continued to be poor in recent years. Therefore, from my point of view,
the most important questions to be answered in a further analysis of the
causes of the economic crisis of recent decades are the following: what
were the causes of the significant decline in the rate of proft in the early
post-war period? Why hasn't the decline of real wages and increase in the
intensity of labour of recent years resulted in a more significant increase
in the rate of proft? And, fnally, what is the likely trend in the rate of
profit in the future? What are the chances of a signifcant increase in the
rate of proft, which would make a full and lasting recovery from the cur­
rent crisis and a return to the more prosperous conditions of the early
post-war period possible?
In trying to answer these important questions about the trends in the
Te Unit State eonomy
rate of profit, it is very surprising and disappointing that orthodox eco­
nomic theories are of absolutely no help. Orthodox economic theories -
especially macroeconomics - almost completely ignore the rate of profit.
Microeconomics does have a theory of proft or interest: indeed, it has
several mutually exclusive theories, the most important of which is the
'marginal productivity of capital' theory. But this theory is completely sta­
tic and provides no theory of trends over time, and is now in general dis­
repute because it has been shown to be logically contradictory (this con­
clusion of the 'capital controversy' is accepted by the top neo-classical
economists such as Samuelson) . This theory is still taught to students,
without mentioning the problems, but more and more it is simply being
ignored. And this theory is not used at al in orthodox macroeconomics -
either neo-classical or Keynesian. The marginal productivity theory may
serve a useful ideological purpose, but it is of no use for macroeconomic
analysis of the real economy. And orthodox macroeconomics has no the­
ory of profit at all. It attempts to provide a theory of capitalism without
including profit or the rate of profit as a variable! I was shocked to dis­
cover this when I frst started to study orthodox economics in college. I
had already studied Marxian economics outside the university, and
Marx's emphasis on the rate of proft as the key variable in understanding
capitalist economies seemed entirely reasonable to me. So I could not
believe it when I realised that orthodox macroeconomics attempts to pro­
vide a theory of capitalism which does not include the rate of proft. I
consider this to be a fatal weakness of orthodox macroeconomics, espe­
cially as a way of understanding economic crises. Without a theory of
profit, a theory of capitalism is not worth much.
Consequently, orthodox analyses of the causes of the current econom­
ic crisis also completely ignore the rate of profit and the very significant
decline in the rate of proft in the early post-war period. Instead, ortho­
dox analyses of the current crisis emphasise factors such as government
policy mistakes, the disruptions of OPEC, a mysterious decline of pro­
ductivity, etc. I have argued above that all of these other factors are them­
selves the result of the decline of the rate of profit, which is the funda­
mental cause of the current crisis. By ignoring the rate of proft, orthodox
analyses miss this fundamental cause and remain on the level of superfi­
cial effects.
The only economic theory that provides a substantial theory of the rate
of proft and its trend over time is Marxian economic theory. In fact, the
rate of profit and its trend over time is the central question addressed by
Marxian economic theory. The rate of profit is the main variable in
Marxian theory, in striking contrast to orthodox theory in which the rate
of proft is completely missing.
Therefore, if we want to understand the causes of the decline in the
rate of profit and its likely trend in the future, the only economic theory
available to us is Marxian economic theory. It is often said these days that
Marxian theory is 'dead' or 'obsolete' . But this assertion is simply false.
Marxian economic theory is still very much alive. There are many excel­
lent Marxian economists around the world using Marxian theory to
understand contemporary capitalism, including the current economic cri­
sis. Indeed, Marxian theory is absolutely essential if we want to try to
understand the rate of profit and its trends. There is simply no alternative
Historicl Materialism
theory of the rate of proft available.
Let us turn now to a
loser examination of Marxian theory and its
explanation of the decline of the rate of pr-oft in the post-war US econo­
2. Marian theory of the decline of the rate of profir
The main point of the Marxian theory of proft is that proft is produced
by the labour of workers, because the value added by the labour of work­
ers to the price of the products is greater than the wages workers are
paid. This con
lusion follows from the labour theory of value, which is
usually interpreted by orthodox economists as a theory of individual
prices, because that is the main question of orthodox economics. But this
is a misunderstanding. The Marxian labour theory of value is mainly a
theory of proft.
Marxian theory concludes that the process of technological change -
an inherent, ever-present feature of capitalist economies - tends to
reduce the number of workers employed in relation to the total capital
invested in machinery, etc. (or tends to increase the 'organic composition
of capital') . Technological change is usually 'labour-saving' - i.e. the
result is usually that it requires less labour to produce the same quantity
of output, and at the same time requires bigger and more complex
machines, which in turn require more and more capital investment (in
more technical terms, technological change increases the composition of
capital) . But, since according to Marxian theory, profit is produced by
workers, if fewer workers are employed in relation to the total capital
invested, then the amount of proft produced will also decline in relation
to the total capital invested. In other words, the rate of profit will decline
(since the rate of profit is the ratio of the amount of proft to the amount
of capital invested) .
Marxian theory argues further that this negative effect of 'labour-sav­
ing' technological change on the rate of proft can be partially offset by
the increased productivity, which also results from the new technology. If
wages remain constant, then an increase of productivity results in an
increase in the amount of proft produced by each worker (in technical
terms, the rate of surplus-value increases) . Therefore, the negative effect
of fewer workers employed is partially offset by an increase in the amount
of proft produced by each worker. This positive effect of new technology
and higher productivity is reinforced in other ways to increase the
amount of profit produced by each worker - such as wage cuts and
increases in the intensity of labour, as discussed above.
However, Marxian theory argues there are inherent limits to the
increase in the proft produced by each worker. The amount of profit
produced by each worker cannot be increased indefinitely. The main limit
is that there are only so many hours in the working day, and it becomes
harder and harder to increase the profit produced by each worker in a
given working day. Another limit is the resistance of workers, who usually
fght against wage cuts and fght for higher wages and a share of the ben­
efits of the increased productivity. A a result of these limits, Marxian
theory concludes that technological change will eventually cause the rate
of profit to decline.
According to Marxian theory, the rate of proft tends to decline as a
Te Unit Stt eonomy
result of technological change. Since technological change is an inherent
characteristic of capitalist economies, Marxian theory argues that the
decline of the rate of proft is not an accident, or due to causes external to
the capitalist economy, but is instead the result of capitalism's own inter­
nal dynamics characterised by continual technological change.
The above trends, predicted by Marxian theory, correspond to what
happened in the post-war US economy. Technological change had the
offsetting effect of reducing the number of workers employed in relation
to the total capital invested. Technological change also had the effect of
increasing the amount of proft produced by each worker. But, as pre­
dicted by Marxian theory, the capital invested per worker increased faster
than the proft produced per worker (Le. the composition of capital
increased faster than the rate of surplus-value) , so that the rate of proft
declined, as we have seen.
Another important determinant of the rate of proft according to
Marxian theory, which Marx himself did not emphasise, but which seems
to have been important in the post-war US economy, is the distinction
between production labour and non-production labour (in technical
terms, the distinction between productive labour and unproductive
labour) ,1 My explanation of the decline of the rate of profit, based on
Marx's distinction beteen productive labour and unproductive labour,
has been criticised by David Laibman8 and Stephen Cullenberg.9 My
responses to these criticisms are presented in Moseley 1 994a and 1 994b.
According to Marxian theory, profit is not produced by all employees of
capitalist enterprises, but only by workers engaged directly or indirectly in
production activities (actually making something or designing something)
and not by non-production employees. There are two main groups of
non-production employees. The first group we may call 'supervisory
employees', which includes managers, supervisors, 'bosses' in general.
The second group of non-production employees we may call 'sales'
employees, which includes sales and purchasing employees, advertising,
fnance, etc.
According to Marxian theory, since non-production employees do not
produce profit, if the number of non-prouction employees increases
more rapidly than the number of production workers (who produce prof­
it), this wll cause the rate of proft to decline, since costs are increasing,
but proft is not. This is what happened in the post-war US economy: the
number of non-production workers in relation to production workers
increased very signifcantly - almost doubled - and this contibuted t
the decline in the rate of profit.!O This trend of a relative increase of non­
production employees also seems to have been due at least in part to
technological change which increased the productivity of production
workers more rapidly than that of non-production workers, and which
therefore required more and more sales workers to sell this ever-increas­
ing output of production workers. 1 !
Therefore, according to Marxan theory, there were two main causes
of the decline of the rate of profit in the post-war US economy: an
increase in the capital invested in machinery, etc. per worker (Le. an
increase in the composition of capital), and an increase in the relative
number of non-production workers within capitalist enterprises (i.e. an
increase in the ratio of unproductive labour to productive labour) .
Historicl Materialism
According to my estimates, these two adverse trends contributed roughly
equally to the total decline of the rate of proft. Both of these trends were
themselves the result of technological change, an inherent systematic fea­
ture of capitalist economies. Therefore, the decline of the rate of proft in
the post-war US economy was not due to accidental, external causes, but
rather was the result of the inherent dynamic of technological change.
3. How to increase the rte of profit
We turn now to the next question: what does this Marxian theory imply
about the necessary adjustments that must be made in order to end the
current crisis in the US economy and return to the more prosperous con­
ditions of the early post-war period? First of all, Marxian theory implies
that the key adjustment that must be made is that the rate of profit must
be increased. Prosperity is possible in capitalism only when the rate of
profit is high. What then is necessary to increase the rate of profit,
according to Marxian theory?
Wage cuts are certainly one important way to increase the rate of
profit. However, Marxian theory suggests that wage cuts by themselves
are not likely to raise the rate of proft enough, because the prior decline
of the rate of proft was not caused by wage increases. What is required
in addition, according to Marxian theory, is a reversal of the two trends
that caused the prior decline of the rate of profit ¯ ( 1 ) the capital invested
in machinery, etc. per worker must be reduced and (2) the relative num­
ber of non-production workers in capitalist enterprises must also be
The main way that the capital invested in machinery, etc. per worker
has been reduced in depressions of the past has been through the wide­
spread bankruptcies of capitalist frms, which themselves were caused by
the combination of falling profits and rising debts, as discussed above.
What normally happens in bankruptcies is that surviving frms purchase
the productive assets of the bankrupt firms, and due to the desperate
nature of these sales, the surviving frm is able to purchase these assets at
a very low price, thereby reducing the amount of capital invested per
worker and raising their rate of proft. This process of bankruptcies con­
tinues until the amount of capital per worker has been reduced enough
and the rate of profit increased enough in the economy as a whole for
new capital investment to resume and for a period of recovery and expan­
sion to begin. However, this adjustment requires widespread bankrupt­
cies of capitalist frms, which obviously worsens the crisis in the short­
The main way to reduce the number of non-production employees
would be to lay off large numbers of these non-production employees
(managers, sales, workers, etc.) . The 'downsizing' of recent years has
been largely concentrated among 'supervisory' employees (especially
middle managers) . However, this 'downsizing' has not as yet been signif­
cant enough to have a large effect on the overall proportion of non-pro­
duction workers in the economy as a whole. Of course, if the displace­
ment of non-production workers were to occur on a more signifcant
scale, then the rate of unemployment would increase sharply, especally
among these groups of occupations.
Since the mid- 1970s in the US economy, as we have seen, wages have
Te Unite States eonomy
been signifcantly reduced and this has indeed contributed to an increase
in the rate of profit (while at the same time contributing to an increase in
the poverty of workers) . However, the other two adjustments necessary
to increase the rate of profit - a reduction in the capital invested per
workers and a reduction in the relative proportion of non-production
workers - have not yet happened. The capital invested per worker has
levelled off, but the relative proportion of non-production workers has
continued to increase and has continued to have a negative effect on the
rate of proft, which has largely offset the positive effect of wage cuts on
the rate of profit. This is the main reason why the rate of profit has
increased so little since the 1 970s, in spite of the reduction of wages (and
increase in the intensity of labour) - because non-production workers
have continued to increase more rapidly than production workers.
4. The fture?
What does this Marxian theory imply about the future course of events in
the US economy? This theory implies, first of all, that the future of the
US economy, like its past, will depend mainly on the trend in the rate of
proft. If the rate of proft increases significantly, then perhaps the US
economy will return to the more prosperous conditions of the early post­
war period. However, if the rate of profit remains depressed, then a
return to prosperity is not very likely. Instead, the US economy will con­
tinue to be characterised, at best, by continued stagnation and declining
living standards, and, at worst, by a deepening depression, similar to the
1 930s, and perhaps even worse.
Furthermore, this theory suggests that the future trend of the rate of
proft itself depends on the three factors discussed above: the wages of
workers, the capital invested per worker, and the relative proportion of
non-production workers. Wages will probably continue to decine, and
this will continue to have a positive effect on the rate of proft - although
of course at the same time declining wages will also continue to have neg­
ative effects on the living standards of workers. The capital invested per
worker will probably remain more or less constant, as in recent years.
However, the relative proportion of non-production workers
ill proba­
bly continue to increase, which will continue to have a negative effect on
the rate of proft. The net effect of these opposing trends is diffcult to
predict with precision, but extrapolating from the recent past, it does not
appear very likely that there will be a significant increase in the rate of
profit in the foreseeable future. In the absence of such an increase in the
rate of proft, the US economy will probably remain stuck in the stagna­
tion of recent decades. Wages will continue to be cut and the liing stan­
dards of workers will continue to decline. But this will not raise the rate
of profit enough to generate another investment boom and an eventual
return to more prosperous conditions. This dismal course of events
seems to be the best we can expect. Eventually, the conditions of slow
growth and stagnation could deteriorate further into deep depression,
and then of course things would be much worse. In either case, contrary
to the optimistic predictions of government oficials, the economic crisis
of recent decades is not yet over - nor even 'almost over' . Marxian theory
suggests that the worst is yet to come.
Furthermore, according to Marx's theory, there is not much the gov-
Historical Materialism
ernment policies can do to avoid this gloomy prospect. In the early post­
war period, economists and politicians thought that Keynesian policies
eould guarantee more or less prosperous conditions forever. This confi­
dence has long since disappeared. The expansionary policies tried in the
1 960s and 1 970s only resulted in higher rates of inflation, without
increasing the rate of proft significantly. Even though expansionary gov­
ernment policies may give the economy a temporary boost in production
and employment in the short-run, these policies do not solve the funda­
mental problem of insufcient proftability, and in some cases even make
this fundamental problem worse. 13
More recently, government policies are often seen as the cause of our
economic problems, rather than as the solution to these problems.
However, according to Marxian theory, government policies are neither
the cause of, nor the solution to, our economic problems. The cause is
the decline in the rate of profit, and this problem cannot be solved by
government economic policies. Restrictive government policies - the
opposite of the expansionary policies of the 1 960s and 1 970s - may
make it easier for businesses to cut wages, but these policies have litte or
no effect on the other determinants of the rate of profit: the capital
invested per worker and the relation proportion of non-production work­
On the other hand, it also appears unlikely that the US economy is
about to fall into another deep depression, similar to the Great
Depression. The rate of profit has increased somewhat and appears to be
not so low as to cause widespread bankruptcies of capitalist firms. If such
a deeper depression were to threaten, emergency government policies
would probably be enacted, which although they would not be able to
solve the underlying problem of insufcient profitability, they would
probably be able to prevent, or at least delay, a slide into deeper depres­
Of course, from a broader, global perspective, much of the world is
already in a depression, especially in Latin America and Mrica. These
developing countries have suffered the most from the current crisis of
capitalism. For example, in Mexico, sub-employment has increased dra­
maticaly (unemployment is often disguised as 'self-employment', e.g.
selling candy on the streets), real wages have been cut by half or more,
and many small and medium-sized enterprises have gone or soon will go
bankrupt. There is not much hope of a recovery in these countries unless
the advanced countries grow more rapidly, which we have already seen is
not very likely. There seems to be a greater risk that these countries will
deteriorate even further, which would have important repercussions on
the US economy. In addition, the continuing depression and increasing
misery in these countries may lead in the years ahead to more organised
social movements in opposition to capitalism such as the Zapatistas in
Mexico and the trabajadores sin tierra (workers without land) i Brazil.
Such social movements would also have repercussions back on the US
and other major capitalist countries.
Therefore, as we move into the 21 st century, it seems very likely that
capitalism will remain i a condition of stagnation and crisis for the fore­
seeable future. The rate of profit is still too low to make possible a faster
rate of expansion and a return to more prosperous conditions. Capitalism
Te Unite State eonomy
may be able to avoid another Great Depression for the foreseeable future,
but it is extremely unlikely that capitalism will be able to provide a retur
to prosperity and improving living standards for the vast majority of the
world's population, and will probably instead continue to produce lower
living standards and increasing misery.
Whether or not these conditions of stagnation and deteriorating living
standards wl generate signifcant oppositional movements in the US and
elsewhere is a very important question that is beyond the scope of this
paper. However, further consideration of this question of the future of
social movements should take into account the very strong probability
that the objective social conditions will continue to deteriorate in the
years ahead for the vast majority of the world's population, including
those in the United States.
Appendix 1
United States of America: Rate of profit 1947-1994,
48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94
The Unite States eonomy
1 . This paper is a revised version of a public lecture I presented on May 5 1 997,
at the Universidad Autonoma Metropolitana (lztapalapa) in Mexico City where
I was a Visiting Professor for the academic year 1 996/97.
2. This percentage applies to private wages paid to workers by employers. The
average 'social wage' (the government services provided to each worker: edu­
cation, health care, income assistance, etc.) has declined even more.
3. Further discussion of these estimates, and a description of the sources and
methods used to derive them, are presented in Moseley 1 991 , especially chap­
ter 4 and Appendix B.
4. A referee of this journal commented that my analysis made 'too easy an associ­
ation between the rate of profit, levels of investment, and productivity increas­
es'. I agree that more work needs to be done, both theoretically and empirical­
ly, to establish these relations. A important recent work on this subject is
Crotty 1 993. To me, it seems very plausible that the very significant decline in
the rate of profit shown in Appendix 1 contributed to the signifcant decline of
investment spending over this period, which in turn contributed to the slow­
down in productivity growth.
5. Moseley 1 997, presents a more complete discussion of these more recent esti­
6. A more formal presentation of this Marxian theory of the decline of the rate of
proft is presented in Moseley 1 991 , chapters 2 and 4.
7. Marx's distinction between productive labour and unproductive labour is con­
troversial and not accepted by all Marxian economists. Moseley 1 991 chapter 2
discusses this distinction further, along with recent criticisms of this distinction
and my responses to these criticisms. However, the validity of Marx's distinc­
tion between productive and unproductive labour cannot be decided on the
basis of theoretical arguments alone, but also depends on the explanatory
power of this distinction. I argue that this distinction provides a superior expla­
nation of the decline of the rate of profit in the post-war US economy than
competing explanations (especially the profit squeeze theory) and that this
explanatory power is important evdence of the validity of Marx's distinction.
8. Laibman 1 993.
9. Cullenberg 1 994.
lO. Moseley 1 991 , chapter 4.
1 1 . Moseley 1991, chapter 5.
1 2. Over the last year or so, the rate of unemployment in the US has dropped to a
20-year low of around 5%, leading many to conclude that the recent period of
stagnation is now over and that the US is entering a new phase of expansion
and prosperity, similar to the early post-war period. However, an increasing
percentage of the new jobs generated in the US economy in recent years has
been in low-paying jobs, especially in services and retail trade, and many of
these have been part-time jobs. The average real wage did indeed increase last
year, for the first time in several years, but only a scant 1 %, which is a long
way from recovering the 20% decline of real wages of the last two decades.
The recent 'boom' has been fuelled primarily by increased consumer spending,
not by investment spending, which remains low due to low proftability. The
increase in consumer spending has been financed to an increasing extent by
debt, as consumers seem to be making purchases (of homes, autos, household
appliances etc.) that they have put off for years now that their economic situa­
tion is at least improving a little bit. The increase in consumer spending has
Historical Materialism
also been fuelled by the stock market boom (which is itself based more on
speculation than on improved proftability) , as stock owers feel richer and
hence are willing to spend more. The inevitable end of this consumer spending
spree will probably lead to another recession, and perhaps even worse.
13. The pioneering analysis of the 'limits' of the mixed economy was presented by
Paul Mattick, most completely in Mattick 1 969. Mattick argued in the 1 950s
and 1 960s, almost alone among economists, including Marxist economists,
that sooner or later the effectiveness of Kynesian policies would come to an
end and that the post-war boom would be followed by yet another period of
capitalist crisis. The events of the last 20 years have dramatically confirmed the
validity of Mattick's analysis, based on Marx's theory of the faIling rate of prof­
Crotty, James 1 993, 'Rethinking Marxian Investment Theory: Keynes­
Minsky Investment Theory, Competitive Regime Shifts and Coerced
Investment', Review of Radical Political Economics, 25: 1 : 1 -26.
Cullenberg, Stephen 1 994, 'Unproductive Labour and the Contradictory
Movement of the Rate of Proft: A Comment on Moseley', Review of
Radical Political Economics, 26: 1 1 1-2 1 .
Laibman, David 1 993, 'The Falling Rate of Profit: A New Empirical
Study', Science and Society, 57: 223-33.
Mattick, Paul 1 969, Marx and Keynes: The Limits of the Mied Economy,
Boston: Porter Sargent.
Moseley, Fred 1 991 , The Falling Rate of Profit in the Postwar United
States Economy, London: Macmillan Press.
Moseley, Fred 1 994a, 'Unproductive Labour and the Rate of Profit: A
Reply to Laibman', Science and Societ, 58: 84-92.
Moseley, Fred 1 994b, ' Unproductive Labour and the Causes of the
Decline in Rate of Profit: A Reply to Cullenberg's Comment', Review
of Radical Political Economics, 26: 1 21-8.
Moseley, Fred 1 997, 'The Rate of Proft and the Future of Capitalism',
Review of Radical Political Economics, 29: 4, forthcoming.
Weisskopf, Thomas E. 1 979, 'Marxian Crisis Theory and the Rate of
Proft in the Postwar US Economy', Cambridge Joural of Economics,
3: 341-78.