Open Journal of Political Science
2014. Vol.4, No.1, 1-7
Published Online January 2014 in SciRes (http://www.scirp.org/journal/ojps) http://dx.doi.org/10.4236/ojps.2014.41001
1
Manuel Castells and Historical Materialism
Mark Cowling
School of Social Sciences and Law, Teesside University, Middlesbrough, UK
Received October 20th, 2013; revised November 27th, 2013; accepted December 10th, 2013
Copyright © 2014 Mark Cowling. This is an op en access article distributed under the Creative Co mmons Attri-
bution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited. In accordance of the Creative Commons Attribution License all Copyrights ©
2014 are reserved for SCIRP and the owner of the intellectual property Mark Cowling. All Copyright © 2014
are guarded by law and by SCIRP as a guardian.
Considering the very large volume of work produced by Castells, the relative scarcity of secondary litera-
ture discussing his theories is somewhat surprising. Castells does not really try to situate his theories
about the network society in relation to Marxism. The argument of this paper is that Castells provides a
plausible account of some major features of the means of production in the contemporary phase of capi-
talism, and also of the effects of this on society more generally. His theories could form the basis of an
academically respectable version of historical materialism. Some problems with his theories are dis-
cussed.
Keywords: Marxism; Castells; Informationalism; Historical Materialis m; Internet
Introduction
Castells started as a political exile from Franco’s Spain but
soon sprang to fame as a lecturer in Paris specializing in urban
issues. At the beginning of his career he undoubtedly adopted a
version of Althusserian Marxism, and was dismissed from his
post at Paris X University, Nanterre for his role in the events of
May 1968 (Castells, 1977, 1972). He was then employed in-
stead at the École des Hautes Études en Sciences Sociales, and
1979 became a professor at the University of Berkeley, Cali-
fornia. He has recently moved to the University of Catalonia.
The high point of his theoretical trajectory and 25 books is
undoubtedly the trilogy: The Information Age: Economy, So-
ciety and Culture (Castells, 1996, 2004, 2000).
Castells undoubtedly started from a Marxist framework, but
he now does not claim that his approach is Marxist, and does
not particularly try to draw Marxist conclusions from it. How-
ever, the central logic of his approach is to link a change in the
mode of production to widespread general change in society,
which is a distinctively Marxist way to look at things. His basic
idea is that in the last 30 years or so of the 20th century there
was a funda mental s hift in the way in whic h the capita list mode
of production operates. The new social structure he terms in-
formationalism (Castells, 1996) In the industrial mode of de-
velopment the main source of productivity lies in introducing
new energy sources or using them in different places. In the
informational mode of development the source of productivity
lies in knowledge generation, processing and communication.
Information processing is focused on improving the technology
of information processing. There the chief aim is the production
of knowledge (Castells, 1996, p. 17).
Informational capitalism has two fundamental distinctive
features: it is global, and it is structured to a large extent around
a network of financial flows. Capital works globally as a unit in
real time (Castells, 1996, p. 471). “Financial capital needs··· for
its operation···knowledge and information generated and en-
hanced by information technology. This is the concrete mean-
ing of the articulation between the capitalist mode of produc-
tion and the informational mode of development” (Castells,
1996, p. 472).
This new form of society is based on networks. “Networks
are the fundamental stuff of which new organizations are and
will be made” (Castells, 1996, p. 168). The networked enter-
prise is: “that specific form of enterprise whose system of
means is constituted by the intersection of segments of auto-
nomous systems of goals” (Castells, 1996 p. 171). “The net-
work enterprise makes material the culture of t he i nformational/
global economy: it transformed signals into commodities by
processing knowledge” (Castells, 1996, p. 172). “My hypothe-
sis is that, as the process of globalisation progresses, organiza-
tional forms evolve from multinational enterprises to interna-
tional networks” (Cast e ll s, 1996, p. 192). Information-proces-
sing is central to the new configuration of the mode of produc-
tion:
Computer software, video production, microelectronics de-
sign, biotechnology based agriculture, and so on, and many
other critical processes characteristic of advanced economies,
merge inextricably the informational content with the material
support of the product, which makes it impossible to distin-
guish the boundaries between “goods” and “services” (Castells,
1996, p. 205).
Who are the capitalists in this new set up? Not the legal own-
ers of the means of production. Some actors at the top of this
global capitalist system are indeed managers, as with Japanese
corporations. Others could be identified under the traditional
category of the bourgeoisie as in the overseas Chinese business
network s. In United States there is a mixture of traditional
bankers, nouveau riche speculators, self-made geniuses turned
entrepreneurs, global tycoons and multinational managers.

M. COWLING
2
Some public corporations are capitalist actors. In Russia we
have the survivors of the communist nomenclatura competing
with wild young capitalists. “And all over the world, money-
laundering from miscellaneous criminal businesses flows to-
wards this mother of all accumulations that is the global finan-
cial network” (Castells, 1996, p. 473). There is not a global
capita list c la ss, but the re i s an inte gra te d globa l ca pi tal ne tw ork.
“While capitalism still rules, capitalists are randomly incarnated,
and the capitalist classes are restricted to specific areas of the
world where they prosper as appendixes to a mighty whirlwind
which manifests its will by spread points and futures options
ratings in the global flashes of computer screens” (Castells,
1996, p. 474).
Castells sees this new economy as inimical to organized la-
bor, and says relatively little about a working class response to
the changed situation: “Under the conditions of the network
society, capital is globally coordinated, Labor is individualized.
The struggle between diverse capitalists and miscellaneous
working classes is subsumed into the more fundamental oppo-
sition between the bare logic of capital flows and the cultural
values of human experience” (Castells, 1996, p. 476).
The central role of information processing in the new devel-
opment of the capitalist mode of production makes it possible
for money to be shifted around the world with extreme rapidity
and also for technology and manufacturing to move between
states and to be coordinated a t a dista nce. What is developing is
no less than a global economy: “A global economy is··· an eco-
nomy with the capacity to work as a unit in real time on a pla-
netary scale (Castells’ emphasis)” (Castells, 1996, p. 92). An
illustration of this development is that the share of trans-border
financial flows for major market economies increased by a
factor of about 10 in 1980-1992 (Castells, 1996, p. 93).
Castells acknowledges that the economy is not yet fully
global:
Markets, even for strategic industries and major firms, are
still far away from being fully integrated; capital flows are re-
stricted by currency and banking regulations (although the off-
shoring of financial centers and the prevalence of computer
transactions tend to increasingly circumvent such regulations);
the mobility of labor is undermined by immigration controls
and people’s xenophobia (Castells, 1996, p. 97).
In order to seriously come to terms with Castells’ analysis, it
would be necessary to get to grips with his use of the term
“network”. It makes sense for rapid computerized exchanges of
information as part of financial dealings or of dispersed manu-
facturing and design across the globe or of lateral exchanges of
information and ideas between people at the same level in dif-
ferent enterprises. There is also nothing wrong with arguing
that the core of the global economy is located in the USA, Ja-
pan and Western Europe; although with the rise of China and
India this will doubtless change. But to describe this triadic
dominance as a “network” seems inappropriate. However, Cas-
tells is much more plausible when he claims that this new eco-
nomic pattern based on information is having enormous effects
on advanced societies, what would have been described as
Third World countries, played a major role in the fall of the
Soviet Union, and is linked to substantial changes in class
structure, the decline of the patriarchal family, the role of poli-
tics and the media, the form taken by social movements and
new opportunities for organized crime.
Many of the developments since 1970 that Castells is ana-
lyzing have featured in the work of other writers, and are fre-
quently described as a shift from Fordism to post Fordism, or as
the ineluctable rise of globalisation. Many other writers give an
account of globalisation which incorporates substantial out-
sourcing, rapid financial flows, increased inequality within the
advanced countries matched by increased global inequality,
power exerted over thousands of people in one continent by
decisions made in another, the triumph of neoliberalism (Held
& McGrew, 2001, p. 4; McGrew, 1997, p. 8; Giddens, Jameson,
& Wallerstein, quoted in Beynon & Dunkerley, 2000; Gill,
1995, p. 406). To some extent this type of description can be
regarded as an empirical account of changes. The interesting
feature of Castells’ work is that the central mechanism of the
change is a technological one, i.e. the development of informa-
tion technology. How does this link with historical material-
ism?
I have argued elsewhere that several interpretations of his-
torical materialism are possible. The version that makes the best
link with Castells, and which has good links to many appropri-
ate places in Marx, is technological determinism, most ably
defended in the work of Jerry Cohen (Cohen, 1978). Cohen’s
account is a subtle and well-argued elaboration of the famous
quotation from Marx’s The Poverty of Philosophy: “[t]he hand-
mill gives you society with the feudal lord, the steam-mill, so-
ciety with the industrial capitalist” (Marx & Engels 1975, Vo-
lume 6, p. 167 the 50 volume Marx and Engels Collected
Works, which will henceforth be abridged to MECW.). A
number of other interpretations are possible, however. These
are briefly described below. For the purposes of the present
discussion I want to take Cohen’s interpretation as the starting
point. I provide a very brief exposition and critique of the other
approaches, with a view to demonstrating that Cohen’s ap-
proach, whilst by no means perfect, provides us with both a
very plausible account of Marx, and a theory that can be ap-
plied to the real world of today. The other interpretations are
less good in both respects. One particular problem with the
interpretations which I shall be discussing below is that they
make no direct reference to state action, but various forms of
state action have been very important in advancing informatio-
nalism. It is now necessary to continue with a brief exposition
of more of Castells’ ideas.
The Effects of the Rise of Informationalism in the
Advanced Countries
There are very strong differences between the occupational
structures of societies equally entitled to be considered as in-
formational. Japan and the United States represent the opposite
ends of the comparison although in all the advanced societies
theory is a “Common trend toward the increase of the relative
weight of the most clearly informational occupations (managers,
professionals, and technicians)” (Castells, 1996, pp. 217-218).
Crudely there are two informational models: the service econ-
omy model represented by the United States, UK and Canada
with a rapid phasing out of manufacturing employment after
1970 and with an emphasis on capital management services;
and the industrial production model represented by Japan and
largely by Germany which reduces the share of manufacturing
employment but continues to keep it at a relatively high level.
Producer services are much more important than financial ser-
vices in this model (Castells, 1996, p. 229).
Skilled workers in the North greatly benefited from global
trade because they took advantage of the higher economic

M. COWLING
3
growth and the international division of labor gave their firms a
comparative advantage. Unskilled workers in the North suf-
fered considerably because of competition with producers in
low-cost areas (Castells, 1996 p. 227). High unemployment is
not an inevitable consequence of informationalism; it is mainly
a European problem caused by state policies. In the Asian Pa-
cific overall employment has expanded substantially (Castells,
1996, p. 257).
Although the above is true at the global level, the conse-
quences for particular people in particular countries may be
dramatic. “The emergence of lean production methods goes
hand-in-hand with widespread business practices of subcon-
tracting, outsourcing, off shoring, consulting, downsizing, and
customizing.” The social costs of labour flexibility which this
precipitates can be high, but on the whole there are improved
family relationships and greater egalitarian patterns between
genders (Castells, 1996, p. 265). The ramifications of informa-
tionalism are accelerating the decline of the patriarchal family,
which is now a minority form (Castells, 1996, pp. 196-281).
The direct consequence of economic restructuring in United
States is that in the 1980s and 1990s family income has plum-
meted. Wages and living conditions continued to decline in the
1990s in spite of a strong economic recovery in 1993 (Castells,
1996, p. 264). By the early 1990s the top 1% of the population
owned 40% of all assets, double what it had been in the
mid-1970s, and at the level of the late 1920s, before progressive
taxation (Castells, 1996, p. 275). The ratio of total chief execu-
tive officer to total worker pay declined from 44.8 times for
1973 to 172.5 times more in 1995. Some 80% of American
households saw their share in national income decline from
1977 to 1999 by about 6% (Castells, 1996, p. 130). The Gini
coefficient rose from .4 in 1967 to 2.45 in 1995 (Castells, 1996,
p. 132). Poverty has increased. The percentage of persons with
income below the poverty line increased from 11.1% in 1973 to
13.3% in 1997. Some 14.6 million Americans in 1991 had an
income below 50% of the poverty level. Basically the informa-
tional economy and globalisation has caused this (Castells,
1996, pp. 134-135). In 1999 the new economy comprised 19
million workers while the old economy employed 91 million
workers. Education had become a critical resource. In 1979 the
average college graduate earned 38% more than the average
high school graduate, but in 1999 the difference was 71% (Co-
hen, 1996, p. 136). In 1999 only 13.9% of the American labour
force was unionized (Cohen, 1996, p. 137) - 12% in 2006 ac-
cording to Bureau of Labor Statistics—see Center for Econom-
ic and Policy Research:
http://www.cepr.net/index.php?option=com_content&task=vie
w&id=1019&Itemid=138
The change to an informational economy has put a down-
ward pressure on welfare states. Because firms can relocate
freely there follows “the downward spiral of social costs com-
petition”. In the past the limits to this have been the productivi-
ty and quality gap between protected workers from the ad-
vanced economies and less developed competitors and the role
of protectionism. Both of these are withering away, pushed
onwards by the World Trade Organization (Castells, 2000, p.
313) Mexican workers’ productivity in automobile factories
lags only about 18 months behind that of American workers.
Now there are similar trends in Asia. American labour produc-
tivity is still the highest in the world (Castells, 2000, p. 314).
Castells says that in a world where economies are increasingly
integrated “there seems to be little room for vastly different
welfare states, with relatively similar levels of labour produc-
tivity and production quality.” A global social contract would
be required to preserve the better welfare states, but in the libe-
ralized, network, global economy this is very unlikely and wel-
fare states are being downsized to the lowest common denomi-
nator. Finland has been an exception to this (Castells, 2000, p.
314). To survive in a globalized economy the welfare state
needs to be connected to productivity growth in a virtuous cir-
cle by a feedback loop between social investment and economic
growth (Castells, 2000, p. 315, Castells & Himannen, 2004).
The new economy generates new forms of social exclusion,
notably of black Americans. Castells argues that there are what
he calls “black holes” of the informational economy: “The
fourth world has emerged, made up of multiple black holes of
social exclusion throughout the planet··· It is inseparable from
the rise of informational global capitalism (Castells, 2000, p.
160).” Black Americans have problems of unemployment be-
cause they are reluctant to undertake menial labour. In turn this
links to a higher rate of illegitimacy and a tendency for the
breakdown of family life (Castells, 2000, pp. 143, 146). Very
many black males end up in prison. In 1996 there were 600
prison inmates in United States per 100,000 residents, a rate
which had doubled in about 10 years (Castells, 2000, p. 148).
53% of these were black. Blacks were 40% of death row in-
mates. This is largely due to discrimination in sentencing rather
than because of the frequency of crimes (Castells, 2000, p.
150).
The Fall of the Soviet Union
Castells describes the Soviet system as statism, and argues
that the crisis of Soviet society from the mid-1970s onwards
was the expression of the structural inability of statism to en-
sure the transition towards the information society (Castells
2000, p. 8). Statism worked well in an industrial society. In the
1950s until the late 1960s and the Soviet Union generally grew
faster than most of the world. The annual growth of Soviet
national income was 7.2% from 1950 to 1960. It was 4.1%
1965 to 1970, 3.2% 1970-1975, then something close to stagna-
tion settled in. (Castells, 2000, p. 10). This is because the So-
viet Union missed the revolution in information technologies
that took shape in the world in the 1970s (Castells, p. 26).
There was a situation close to parity in computer design in the
early 1960s but by the 1990s there was a 20-year difference in
design and manufacturing capability (Castells, 2000, p. 30). In
the USSR typewriters were rare, carefully monitored devices;
two signatures were required for access to a photocopier, or
three for a non-Russian text. There were special procedures for
using long-distance telephone lines. The notion of a personal
computer was objectively subversive (Castells, 2000, p. 36).
Castells does not say this, but there is good reason to think
that the fall of the Soviet Union is responsible for several of the
well-known features of globalisation. The Soviet Union func-
tioned as a counterweight to the capitalist West and as a sort of
welfare state for Third World countries, offering them an alter-
native pattern of development and source of help. The demise
of the Soviet Union has left us with a single superpower. The
challenge to neoliberal globalisation now rests with an assort-
ment of relatively small and powerless social movements. The
rampant growth in inequality, both within advanced states and
between the advanced states and others, which has characte-
rized the period since about 1980, would have been much more

M. COWLING
4
subject to challenge if there had been an alternative ideology
and power base.
Margin al ization of Sub-Saharan Africa
Developing countries which have the political capacity to
develop an infrastructure which can take advantage of the in-
formational economy can advance very quickly. Those which
cannot are destined to languish (Castells, 1996, p. 105). Libera-
lization policies in Africa didn’t attract investment or improve
competitiveness, but destroyed large sectors of agricultural
production for local markets and in some cases subsistence
agriculture. The struggle for the control of the state became a
matter of survival. Tribal and ethnic networks were the safest
bet for support. The struggle to control the state was organized
around ethnic cleavages leading towards genocide and banditry.
This is rooted in “the political economy of Africa's disconnec-
tion from the new, global economy.” The new global economy
does not have much of a role for the African population. Pri-
mary commodities are useless or low-priced, markets are too
narrow, investment too risky, labour is not skilled enough,
communication and telecommunication infrastructure clearly
inadequate, politics too unpredictable, and government bureau-
cracies inefficiently corrupt (Castells, 1996, p. 105). The per-
centage of world trade to and from Africa roughly halved be-
tween 1980 and 1995; foreign direct investment, growing sub-
stantially elsewhere, is not attracted to Africa (Castells, 1996,
pp. 83-90).
Africa is by far the least computerized region of the world,
and does not have the minimum infrastructure to make use of
computers. In 1991 there was one telephone line for a hundred
people in Africa compared to 2.3 for all developing countries,
and 37.2 for industrial countries (Castells, 2000, p. 92). Castells
emphasizes the role of the developmental state in the rise of the
Asian tiger economies. Africa has the reverse. As Colin Leys
puts it: “few theorists of any of these persuasions [Marxists,
dependency theorists] expected the postcolonial state of all
ideological stripes to be corrupt, rapacious, insufficient, and
unstable, as they have almost all been” (Castells, 2000, p. 96).
It should be added that more recently conditions have become
better in much of the African continent. Civil wars have ended,
dictators have fallen, and economic development has somewhat
advanced.
The Asian Tiger Economies and Japan
In dramatic contrast to the fate of sub-Saharan Africa, Cas-
tells provides a fascinating analysis of the rapid growth expe-
rienced in post-war Japan, the role of the state in four of the
leading tiger economies of the Pacific-Singapore, South Korea,
Taiwan and Hong Kong, and the prospects for continued rapid
growth in China. He argues that there has been an important
role for what he identifies as the developmental state. He also
argues that the tiger economies experienced a crisis in the late
1990s in part because they needed to make the move from so-
cieties under the aegis of the developmental state to fully net-
worked advanced economies (Castells, 2000, pp. 212-230).
Castells and Ma rxism
Enough of Castells’ account of informational capitalism has
been given to demonstrate that it could form the basis of a
modernized version of historical materialism. It starts from
significant changes in the means of production which in turn
have profoundly affected the working of the capitalist mode.
Considerable work would be needed to analyse whether all the
linkages along the way are satisfactory, and, as suggested above,
Castells arguably operates with an unduly flexible concept of
network. He does, however, manage to explain major and sig-
nificant features of the development of capitalism over the last
40 years on a global basis. Not surprisingly, several difficult
questions remain. The changes which Castells discusses have
been accelerated in Britain and the United States by neoliberal
politics, notably pursued by Reagan, Thatcher and Bush junior
and senior, but also pursued in a diluted form by Clinton and
Blair. Are these figures simply going with the flow, or are they
significantly accelerating it? President Obama has a clear com-
mitment to curb some of the wilder excesses of US capitalism
and to develop universal health care, both of which policies
involve restraining features of the informational economy.
These policies should be much more difficult to implement than
their neoliberal predecessors. What should be said about the
rise in spending on education and the NHS under New Labour?
What about the various measures being taken by governments
to rein in bankers in the wake of the credit crunch? To what
extent can states and other actors mitigate the anti-egalitarian
features of informational capitalism? To what extent can they
resist the tendency of their economies to export jobs to China,
India and other countries offering advanced facilities and cheap
labour? Growth has generally been fairly steady in Germany,
which suggests it is possible to buck the neoliberal trend. Is it
fair to say that Castells provides us with a satisfactory version
of historical materialism which explains some very general ten-
dencies in capitalism, but that one would expect any explana-
tion at this level of generality to have quite a number of exceptions?
In order to get to grips with this question, the obvious start-
ing point is to review previous interpretations of historical ma-
terialism. There are at least six of these, as can be seen from
what follows (Cowling & Manners, 1992, pp. 9-29). As I indi-
cated above, the best of these is technological determinism.
The idea of technological determinism as a way of making
sense of Marx’s theory of history is best developed by Cohen
(Cohen, 1978). It did not, of course, start with Cohen. There are
some obvious passages in Marx which point in this direction.
Two of the most famous are:
The hand-mill gives you society with the feudal lord, the
steam-mill, society with the industrial capitalist (MECW, Vo-
lume 6, p. 167)
No social order is ever destroyed before all the productive
forces for which it is sufficient have been developed, and new
superior relations of production never replace older ones before
the material conditions for their existence have matured within
the framework of the old society’ (Preface).
Cohen’s version of technological determinism builds on pas-
sages of this sort. He attempts to establish an account of pro-
ductive relations which does not involve legal terms, by rede-
scribing the apparently-legal aspects of production relations in
terms of powers, which may be supported by law, but which are
by no means identical with it. He also offers an account of the
way the productive forces determine the economic structure and
the economic structure determines the superstructure in terms
of functionalist explanation: stability requires a legal expression
of production relations. The rest of this volume is concerned
with the further pursuit of these ideas. It should be pointed out
here that although I have described Cohen as a technological

M. COWLING
5
determinist, he does not use this term as a self-description.
Nonetheless, the general approach he adopts has usually been
seen, I believe rightly, as technological determinism, and his
own programmatic statements do not undermine my view of
him. In the preface to Karl Marx’s Theory of History: A De-
fense he says of his own approach:
··· it is an old-fashioned historical materialism which I de-
fend, a traditional conception in which history is fundamentally,
the growth of human productive power, and forms of society
rise and fall according as they enable or impede that growth
(Cohen, 1978, p. x).
His main reason for avoiding the term seems to be a reluc-
tance to accept possible connotations of “determinism”, al-
though Cohen’s views about the causal mechanisms of histori-
cal materialism are clear. He says of his version of historical
materialism that it “may be called technological, but the issue
of determinism will not be discussed” (Cohen, 1978, p. 147f).
This is a particularly well elaborated account, and the de-
scription of the other possibilities will be much briefer.
One of these is economic determinism, as propounded by
Richard Miller and William Shaw (Shaw, 1978, p. 72; Miller,
1984, pp. 8-9). In this version the idea is that the economy de-
termines other areas of society, but there is no particular ac-
count of the relationship between the forces and the relations of
production. This interpretation may be regarded as having cer-
tain advantages over Cohen’s theory: it specifies the mechan-
isms by which the development of the productive forces trans-
late into effective class action (a “motive” of more efficient
production, and the means to achieve it through an increasing
economic power base); it allows for stagnation due to revolu-
tionary failure, because the condition of a stable mode of pro-
duction is the ruling class’s economic dominance and not its
ability to promote productivity; and it does not demand that a
new mode of production be optimal for the development of the
forces of production. However, the distinction between the
economic base and the superstructure is not at all clear in this
interpretation.
Another possibility is an interpretation where the relations of
production are seen as determining the economic base and also
the superstructure. It is associated with Althusser at one stage
and also Hindess and Hirst at one stage (Althusser & Balibar
1970; Cutler, Hindess, & Hirst, 1975). Two particularly useful
ideas are developed within this perspective. One is that a mode
of production requires for its reproduction various precondi-
tions, such as the reproduction of the laborers or an appropriate
legal framework (Balibar in Althusser Balibar, 1970, pp. 2585-
2589). The other is the identification of transition with the
transformation of the forces of production by the relations of
production.
A further interpretation, which probably actually departs
from Marxism is skepticism. Marx is seen as alternating unsa-
tisfactorily between advocating a theory of history which gives
an account valid for all societies and allowing the validity of
exceptions. This leads to the view that any idea of economic
determination should be abandoned (Cutler, Hindess, Hirst, &
Hussain, 1970, Volume 1).
A yet further interpretation could be called class unitarianism.
This approach is particularly associated with historians such as
Hobsbawm, Hilton, Hill, and Thompson. The basic idea is that
that social and historical determination resides precisely in a
concrete and all-encompassing unitary class experience. Most
importantly, there is the unfavorable comparison with the base/
superstructure model which posits the economy as the deter-
minative basis of class as well as of society, and thereby justi-
fies the centrality of class analysis; by contrast, in unitarian
theory class is the determinative basis of society, but there is no
justification for this determination, only the sporadic and sub-
jective self-consciousness of class members and an empiricist
intuition of the objectivity of class existence. Despite its appeal
for historians whose task it is to apply the general principles of
historical materialism to an analysis of concrete history, this
theoretical paradox severely undermines the class unitarian
approach.
Moving on, there is an interpretation based on the idea of
praxis (Jakubowski, 1990, p. 37; Gramsci, 1971, p. 372). This
approach emphasises that history is determined by conscious
and creative human activity. Despite its association with Anto-
nio Gramsci, this approach suffers from a considerable degree
of vagueness. It leads Gramsci in the direction of relativism and
idealism.
Finally we come to the idea of an organic totality (Rader,
1979). This approach which is developed by Melvin Rader and
Bertell Ollman emphasizes that Marx’s central concepts are
importantly internally related, citing the problem that science,
law and property belong to both the base and the superstructure
(Rader, 1979, pp. 119-120). I do not consider that Marx’s con-
cepts are as tightly internally related as this approach suggests.
It is also not entirely clear where it leads. Thinking about a
human organism, we can easily do without hair; people can
manage without feet, although life becomes much more diffi-
cult. Readers may like to continue with this grisly conceptual
experi ment and then see if it ca n be applied to society . It seems
to me that the overall result is relatively vague.
The most plausible way of linking Castells and historical
materialism can be developed from the Cohen approach out-
lined above. The advance from one mode of production to
another occurs partly because the discarded mode of production
has started to act as a fetter on the forces of production, and
partly because a revolutionary class is committed to introducing
a new mode of production. Castells considers that the informa-
tional mode of production divides and weakens the working
class to the extent that he does not really use this term. He
therefore has virtually nothing to say about class struggle.
However, this plainly remains an aspect of economic and polit-
ical life in both advanced and developing countries. It is not the
informational economy on its own which is opposing health
reforms in the United States; there is a right wing coalition
involving parts of the Republican Party and those with an eco-
nomic interest in preserving the status quo.
Again, there is presumably some flexibility in the way an in-
formational economy operates. Consider the above discussion
of the way in which black American males are socially ex-
cluded and imprisoned at a rate much higher than appears war-
ranted by their numbers in the population or their behaviour. It
does not seem to me impossible that the harm reduction ap-
proach to drugs which is currently implemented in various
forms in Switzerland, Holland, Portugal and increasingly in
some Latin American countries, might spread to the United
States. If this was matched with a determination to make better
use of African American males, there is surely no overwhelm-
ing interest in maintaining the current very high rates of impri-
sonment. Other countries which have survived the move to an
informational economy tolerably well manage with much lower
rates of imprisonment.

M. COWLING
6
Although there is doubtless some flexibility in a world cha-
racterized by informational capitalism, it needs to be recog-
nized that the forces of production are currently developing
extremely rapidly, so it is not realistic to e xpect massive prole-
tarian discontent at capitalist stagnation. This can be seen in
relation to the current situation in regard to the so-called de-
clining rate of profit. Marx argues that as capitalism develops
each worker is matched by an increasing mass of machinery.
This means that the value of the machinery outstrips that of the
workers’ wages, but according to the labour theory of value the
sole source of value is labour. The rate of profit therefore tends
to decline over time. Although this looks plausible when one
thinks about each worker typically being linked to an increasing
quantity of machinery, the relevant issue is the relative value of
the machinery. The role of computers over the last 50 years or
so illustrates this point dramatically. Back in the 1960s com-
puters were so expensive that it was felt necessary to operate
them 24 hours a day. This is not now necessary. The most
commonly quoted way of expressing this advance is Moore’s
Law, which states that the number of transistors on an inte-
grated circuit will double every 18 months, and which also
translates into a rapid advance in computing power per unit cost
(Wikipedia, Moore’s law). The cost of hard drive space per
megabyte fell from $10,000 US in 1956 to about one cent by
the year 2000, and has continued to fall rapidly since that time
(http://www.littletechshoppe.com/ns1625/winchest.html). This
rapid fall in the cost of computing power has also reduced the
cost and enhanced the efficiency of a wide variety of machinery
used in production. Thus the common sense idea of the in-
creased role of machinery leading to a falling rate of profit has
not merely a logical flaw, but has also dramatically failed to
work in the real world for the last 50 years or so. Obviously,
along with this has gone a spectacular development of comput-
er applications. One tiny example amongst many is the devel-
opment of voice dictation to computers, which now works quite
well but which would have been entirely impossible using the
most powerful desktop computers of 25 years ago. I started
doing word processing in the 1980s using the university’s
mainframe computer, which had 2.8 MB of RAM. My desktop
computer now has 16 GB of RAM. The application of comput-
ers to production, together with the developing capacities of
China and India, have led to a dramatic cheapening of all kinds
of products. This in turn must be at least part of the explanation
for the relative quiescence of working people.
Marx’s account of historical materialism is, of course, linked
to the idea of socialist revolution. What sort of socialist politics
can be pursued in a globalizing world of informational capital-
ism? Traditional Communist politics, meaning political strate-
gies intended to introduce an economy and system of govern-
ment similar to that in the former Soviet Union, looks particu-
larly unattractive. To go through all the pain and bloodshed
involved in revolution, likely intervention, and isolation in or-
der to produce a system which cannot keep up with informa-
tional capitalism and which is likely to collapse does not make
any sense. Similarly, strategies based on social democracy but
which put the accent on changes within a nation state and
which aim to secure a substantial degree of isolation from the
global economy such as the Alternative Economic Strategy
advocated in Britain by the Labour left in the 1980s do not look
at all promising. There are plainly benefits from, for example,
trade with China or the export of technologically advanced
products such as pharmaceuticals which would be jeopardized
by such an approach. Moreover, as Castells points out, people
who make their living through working in the advanced coun-
tries are becoming increasingly divided between skilled and
unskilled labour. These divisions amongst working people are
fairly slight compared to the gap between low pay or welfare
benefits in the advanced countries and rates of low skilled pay
in countries such as China and India. Thus it is certainly a
scandal that the chief executive of Wal-Mart is paid 871 times
as much per hour as the $9.68 average hourly pay of US Wal-
Mart employees, and even more of a scandal that he gets about
50,000 times as much per hour as garment workers in China
and Bangladesh working for Wal-Mart subcontractors on $0.17
per hour. However, this still leaves a gap between the noto-
riously underpaid US Wal-Mart employees and the Chinese
garment workers such that the Wal-Mart workers are paid about
57 times as much as the Chinese workers (Anderson, 2005 at:
http://www.ips-dc.org/projects/global_econ/Wal-mart_pay_gap
.pdf 2005). This is a bigger gap than that between average chief
executive pay and average pay of workers in the United States
in 1973, when chief executives earned 44 times as much as
workers. The gap between capitalists and workers is not, of
course, normally expressed as a gap in pay, but the gap in pay
would certainly be part of the reason that workers and chief
executives would be assigned to different classes. Even if one
focuses on minimum wage Wal-Mart employees and assumes
that these are particularly unlucky Chinese workers who could
be paid double what they are getting if they moved to a better
employer there is still a massi ve gap.
If we can talk about a global proletariat it is plainly a very
divided one. Some sort of worldwide Trotskyist revolution
looks just as unlikely as its Stalinist rival. Actions to produce
rough equality between workers in the Third World and those
in advanced countries are likely to be resisted by the latter. On
the other hand there is at least a degree of common interest on a
range of issues: reasonable working conditions including both
hours and health and safety, at least a minimum level of welfare
state provisions, corporate responsibility and transparency, be-
ing able to join an independent trade union, some degree of
protection for the environment and the avoidance of global
warming. The interests of working people worldwide are not
served by imperialist adventures such as the recent American
and British occupation of Iraq. If Chinese and Indian growth
rates continue to be close to double digits while those in the
advanced countries are more modest the gap between wages
should lessen over time. In the meanwhile there is scope for
trade union and political lobbying to try to bring about mini-
mum standards. A contemporary example would be the January
2007 lobbying of the World Economic Forum in Davos by
trade union leaders seeking corporate responsibility. (See:
http://www.ituc-csi.org/spip.php?article604). There is also a
role for non-governmental organizations such as those involved
in the campaign to Make Poverty History (Castells, 2004, p.
331). Some of what are generally seen as anti-globalisation pro-
testers have objectives which are fully compatible with a so-
cialist agenda. None of this is as inspiring as the revolution
predicted and strived for by Marx, but it does suggest that so-
cialism is not entirely dead. There is some scope for socialist
initiatives which are going with the grain of history.
Conclusion
Although his theories need considerable further development,

M. COWLING
7
Castells offers an account of contemporary reality which is
highly plausible. Cohen’s interpretation of historical material-
ism provides a plausible account of Marx which is also capable
of being applied to developments in the real world. The above
article demonstrates that it can be linked to the ideas of Castells.
What emerges is a plausible account of the real world which
can be linked to historical materialism. It also links well with
much contemporary writing on globalisation. Marx is by no
means entirely vindicated; in particular, his account of socialist
revolution, which is linked to historical materialism, does not
currently look plausible in its original form. However, the ove-
rall picture which emerges is a surprisingly respectable version
of historical materialism.
REFERENCES
Anderson, S. (2005). Wal-Mart pay gap. Washington DC: Institute for
Policy Studies.
http://www.ips-dc.org/projects/global_econ/Wal-mart_pay_gap.pdf
Bureau of Labor Statistics—See Center for Economic and Policy Re-
search.
http://www.cepr.net/index.php?option=com_content&task=view&id
=1019&Itemid=138
Castells, M. (1977) The urban question: A marxist approach. Trans-
lated by Alan Sheridan, London: Edward Arnold, French original
published, 1972.
Castells, M. (1972). The urban question. A marxist approach (Alan
Sheridan, translator). London: Edward Arnold.
Castells, M. (1996). The rise of the network society. Oxford: Blackwell.
Castells, M. (2000). End of Millennium—Volume three (2nd ed.), Ox-
ford: Blackwell.
Castells, M. (2004). Power of identity. Oxford: Blackwell.
Cohen, G. A. (1978). Karl Marx’s theory of history: A defence, Oxford:
Oxford University Press.
Cutler, A. Hindess, B., Hirst, P., & Hussain, A., (1977). Marx’s capital
and capitalism today (Volume 1). London: Routledge and Kegan
Paul.
Gill, S. (1995). Globalisation, market globalisation and disciplinary
neo-liberalism. Millennium, Journal of International Studies, 24,
399-423.
Held, D., & McGrew, A. (2001). The global transformation reader.
Camb ridg e: Polity, Cambridge.
Little Tech Shoppe (Year).
http://www.littletechshoppe.com/ns1625/winchest.html
McGrew, A. (1997). The Transformation of Democracy. Cambridge:
Polity. p. 8; Giddens, Jameson and Wallerstein (2000). In J. Beynon,
& D. Dunkerley (eds), Globalisation: The reader (p. 4). New York:
Routledge.
Rader, M. (1979). Marx’s interpretation of history. Oxford: Oxford
University Press.
Trade Union Lobbying (2007).
http://www.ituc-csi.org/spip.php?article604
Wikipedia. Manuel Castells.
http://en.wikipedia.org/wiki/Manuel_Castells