Date post: | 27-Dec-2015 |
Category: |
Documents |
Upload: | marcoz-antonioz-gaverioz |
View: | 39 times |
Download: | 5 times |
Theor Soc (2006) 35:237–264
DOI 10.1007/s11186-006-9004-y
Northern theory: The political geography of generalsocial theory
Raewyn Connell
C© Springer Science + Business Media B.V. 2006
Abstract The relationship between geopolitical position and general social theory is exam-
ined by a detailed reading of three important texts, Coleman’s Foundations of Social Theory,
Bourdieu’s Logic of Practice, and Giddens’s Constitution of Society. Effects of metropoli-
tan position are traced in theoretical strategies, conceptions of time and history, models of
agency, ideas of modernity, and other central features of their theorizing. Four textual moves
are identified that together constitute the northernness of general social theory: claiming uni-
versality, reading from the center, gestures of exclusion, and grand erasure. Some alternative
paths for theory, embodying different relations with the global South, are briefly indicated.
But one should not lose sight of the real.Frantz Fanon, Black Skin, White Masks
In a short but disturbing paper examining the 14th World Congress of Sociology, Heinz
Sonntag, a former president of the Latin American Sociological Association, demonstrated
the institutional dominance of world sociology by academics from the rich countries of
the global North. In terms of organizational authority within the International Sociological
Association, in the convening of Congress sessions, and in the authorship of papers, the same
pattern repeats itself–massive predominance of the developed countries.1
As Sonntag would doubtless agree, the professional organization of sociology is not the
root of the problem. Vast international inequalities of resources, especially in the size and
wealth of higher education systems, shape all academic disciplines. But global inequali-
ties may also be embedded within a discipline, in the way intellectual workers define their
problems and carry out their work.
It is time we explored this issue for a key element of sociology’s disciplinary culture,
general theory. Social theory is overwhelmingly produced in the global North. This is perfectly
well known, but – except in a specialized literature of “post-colonial theory” – remains
R. ConnellFaculty of Education and Social Work, A35, University of Sydney, NSW 2006, Australiae-mail: [email protected] Heinz R. Sontag, “How the sociology of the North celebrates itself,” ISA Bulletin 80 (1999): 21–25.
Springer
238 Theor Soc (2006) 35:237–264
unspoken in our theoretical discourse. It is one of those uncomfortable facts “in front of
your nose,” as George Orwell put it in a memorable critique of political thinking, “which are
obvious and unalterable, and which will have to be faced sooner or later.”2
I propose to study this uncomfortable fact by a close reading of influential texts of general
theory. I focus on three of the most influential theorists of the last generation, James S.
Coleman, Anthony Giddens, and Pierre Bourdieu. From their oeuvre I focus on the books
that most explicitly state their general theoretical perspectives.
A small set of texts cannot represent the whole of social theory, but if we are to examine
the genre at all, these seem a good place to start. They come from three countries influential
in the history of sociology, and represent contrasting styles of theoretical work – one building
a tightly-knit propositional system, the second an elaborate scheme of categories, and the
third a practical tool-kit for analysis. The authors all have reputations as major theorists.
Their work is, for instance, prominent in Charles Camic and Neil Gross’s 1998 survey of
“contemporary developments in sociological theory.” The Web of Science on-line database
provides solid evidence that these particular texts are widely known and used. In the last ten
years, Giddens’s The Constitution of Society has 2279 citations recorded, Bourdieu’s TheLogic of Practice has 1236, and Coleman’s Foundations of Social Theory has 1860.3
Although it may not be popular bedtime reading, general theory is much admired; it has
a certain hegemony in the collective life of sociologists. Books of general theory will, we
expect, tell us what the most important features of the social world are and what the best way
to understand them is.
I value what such texts try to do. General theory is important in enabling social science to
be a cultural force. But the way theory is done may also be severely limiting. In this article, I
raise the question of what in the genre of theory (rather than what propositions in particular
theories) we need to re-think, to allow social science to play a larger role in the world.
Northern choosers: Coleman’s Foundations of Social Theory
James S. Coleman’s Foundations of Social Theory was published in 1990 as the summation of
a very distinguished intellectual career. The author had been for three decades a leading figure
in US sociology, working in fields as diverse as youth studies, quantitative methodology,
educational inequality, and rational choice theory. Famous far beyond sociology for the
“Coleman Report” on race and schooling, Coleman also had an agenda for the re-making of
the discipline, which this book spells out.
Foundations is perhaps the most single-minded solo flight in recent sociology. Across
a thousand pages it makes a heroic traverse of sociological problems ranging from social-
ization and the family to corporate management, the state, and revolution. Coleman shows
in every chapter how existing knowledge can be re-written in a single language of choices
and choosers. In the final section of the book, this re-writing evolves into a mathematical
formalization, presenting algebraic models of social processes, strongly influenced by game
theory. The book was greeted by some reviewers as the most important piece of social theory
2 Bart Moore-Gilbert, Postcolonial Theory: Contexts, Practices, Politics (London: Verso, 1997); George Or-well, In Front of Your Nose: Collected Essays, Journalism and Letters, Volume IV, 1945–1950 (London: Seckerand Warburg, 1968).3 Charles Camic and Neil Gross, “Contemporary developments in sociological theory: Current projectsand conditions of possibility,” Annual Review of Sociology 24 (1998): 453–476; Web of Science,http://www.isinet.com/products/citation/wos/.
Springer
Theor Soc (2006) 35:237–264 239
since Parsons’s Structure of Social Action, and Coleman as a “master of social thought” on
a par with Weber and Durkheim.4
Although the book offers hardly any new concepts and is noticeably isolated from other
theoretical trends in sociology, Foundations is important because of the point of view it crys-
tallizes. Coleman had been a leading advocate and practitioner of methodological positivism,
which has dominated empirical sociology in the United States since the 1930s but has had
limited impact on theory. This book is modern positivism’s great moment in sociological
theory. Its arguments are particularly timely because of the current dominance of economics
in the Western social sciences and public policymaking. Coleman’s model for theory of-
fers a solution to sociology’s current dilemma of marginalization, orienting the discipline
consciously towards the hegemonic science.5
Ambition
Coleman’s theoretical ambition is announced in his first sentence: “A central problem in social
science is that of accounting for the functioning of some kind of social system.” “Some kind”
becomes “any kind,” through an extremely abstract definition of what a social system is. A
social system is a set of individuals linked by transactions, in which they must engage to
satisfy their own interests because the other individuals have some control over the resources
they need. The interplay between individual and system, the micro-macro link, becomes a
formative problem in Coleman’s theorizing, and is generally a central problem in modern
positivism.6
Less noticed, because it is so common in sociological theorizing, is Coleman’s assumption
that this language of individual and system, interest, control, and resource, micro and macro,
is of universal relevance. The concepts can be applied in any time and place. This is in accord
with the epistemology of the positivist school. The attempt to make universal statements,
“highly generalized propositions” (in Marion Levy’s phrase) that could be tested empirically,
was always their key strategy of theory-building. Coleman’s ambition, consistently, is to
produce a universally applicable account of the functioning of social systems.7
The two starting points
Coleman is explicit, indeed insistent, about what his starting-point is: “the individual,” also
called “the person” or “the natural person.” These are the “elementary actors” of social
theory, up to the point where “corporate actors” are introduced – but the corporate actors
have already been deduced from the individuals. Resources and rights may be transferred to
corporate actors, but they begin with individuals. In one of the few passages where Coleman
4 Peter Abell, “Review article: James S. Coleman, Foundations of Social Theory,” European SociologicalReview 7/2 (1991): 163–172; Michael Hechter, “Review of James S. Coleman, Foundations of Social Theory,”Public Choice 73 (1992): 243–247; Thomas J. Fararo, “Review of James S. Coleman, Foundations of SocialTheory,” Social Science Quarterly 72/1 (1991): 189–190.5 For the earlier methodological position, see James S. Coleman, “The methods of sociology,” in R. Bierstedt,editor, A Design for Sociology: Scope, Objectives, and Methods (Philadelphia: American Academy of Politicaland Social Science, 1969), 86–114.6 James S. Coleman, Foundations of Social Theory (Cambridge: Harvard University Press, 1990), 1, 29.7 Marion J. Levy, Jr, “Scientific analysis as a subset of comparative analysis,” in J.C. McKinney and E.A.Tiryakian, editors, Theoretical Sociology (New York: Appleton-Century-Crofts, 1970), 100; cf. Hubert M.Blalock, Jr. and Ann B. Blalock, editors, Methodology in Social Research (New York: McGraw-Hill, 1968).
Springer
240 Theor Soc (2006) 35:237–264
approaches eloquence, he insists that even in the processes where sovereignty is transferred
to collectivities, “individual persons do have primacy.”8
Some critics, such as Neil Smelser, have seen this as the central weakness of Coleman’s
work, a paradoxical attempt to construct a social science from individualist assumptions.
Coleman is sensitive to the charge of exaggerated individualism – the ghost of Durkheim
can be heard off-stage, groaning – but he has an answer to the charge in his later institutional
analysis. He does get to collective processes eventually.9
The more important problem about this starting point is what kind of individual is being
brought into play. Coleman is sharply critical of the “intellectual disarray” in sociology
resulting from varying conceptions of the person:
The correct path for social theory is a more difficult one: to maintain a single conception
of what individuals are like and to generate the varying systemic functioning not from
different kinds of creatures, but from different structures of relations within which these
creatures find themselves.10
So what kind of creature does Coleman maintain? When we examine what the “natural
persons” do in his text, it becomes clear that they are creatures of a very specific kind. They
pursue their own interests, they make calculations about costs and benefits, they bargain
with others, they give up rights or receive rights, they engage in purposive actions towards
a goal. In short, they behave like entrepreneurs in a market – all the time. Olof Dahlback
put it succinctly: Coleman’s theory assumes “that individuals are rational and that they are
egoistic.”11
This is not surprising. It is, after all, the model of the individual in marginalist economics
from which Coleman was borrowing. This model provides the assumptions required to set
up the formalization of social exchanges in Part V of the book, “The Mathematics of Social
Action.”
But this shows that Coleman is not quite accurate in claiming the individual as the “starting
point” of his theory. Equally, his starting point is a concept of the market – the social structure
that gives rise to that particular kind of “individual.” Coleman is more sociological in his
underlying reasoning than he admits himself.
Coleman is well aware that there are many social situations that are not competitive
markets, for instance authority relations. But he consistently analyzes non-market structures
by bringing into play “a set of independent individuals” that consists of only market actors,
calculating and bargaining. In this respect, his sociology is strikingly contemporary. It is a
grand generalization of the vision of people and social relations characteristic of modern
neo-liberalism.12
Theoretical strategy
Coleman follows the time-honored strategy of moving from (apparently) simple to (appar-
ently) complex phenomena. Indeed this provides the architecture of the book as a whole,
8 Coleman, Foundations, 3, 32, 367, 493, 531.9 Neil J. Smelser, “Can individualism yield a sociology?” Contemporary Sociology 19/6 (1990): 778–783.10 Coleman, Foundations, 197.11 Coleman, Foundations, passim, e.g. 34–37; Olof Dahlback, “Review of James S. Coleman, Foundations ofSocial Theory,” Aeta Sociologica 34 (1991): 139–140.12 Coleman, Foundations, 66 et seq.
Springer
Theor Soc (2006) 35:237–264 241
as well as the shape of many moves within it, e.g., “Social relations between two persons
are, of course, the building blocks of social organization.” This allows him to start with
radical abstraction and simplification, construct a less-simple derivation, and then compare
the product with some actual set of events.13
An illuminating case in point is his discussion of Palestinian resistance to Israeli oc-
cupation. He examines these events as an illustration of the relation between frustration
and the outbreak of revolutions. This “case” is not forced on him by any logic of argu-
ment. Coleman has not been analyzing, for instance, Islam’s relationship to the European
world, which would require him to look at these events and no others. The case is simply
an example of a certain kind of relationship. Another case from another period of history,
indeed any other case from any period of history, would serve equally well. The theoret-
ical strategy thus leads to a consistent disembedding of actual events from their historical
contexts.14
In place of historical time, Coleman’s argument works with an abstract time. Processes
occur with a before and after, but not with a date. Alternatively they are abstracted from time
altogether, e.g., the indifference curves of the formalized “linear system of action.” To put
it another way, in the positivist theoretical strategy, history is treated as homogeneous and
non-cumulative. Historical events do not change the logical structure of later events; there is
no dialectic here. (However a disjunction is assumed in Coleman’s treatment of modernity,
discussed below.)15
The site
Coleman’s actors move in an energetic dance, calculating, bargaining, and exchanging, on a
featureless dance floor. It is not entirely accidental that his visual models of action systems
resemble teaching diagrams for the fox-trot and the jazz waltz. The featurelessness of the
dance floor follows from the ahistorical method. In each derivation, the same limited set of
elements and possible relations is set in motion. The theoretical logic will not work, any
more than one can dance a fox-trot, if the dance-floor is lumpy with footprints from previous
dances or with the bodies of previous dancers.
To use another metaphor, Coleman’s own: at each important step in the argument Coleman
has to imagine a space in which the building (he repeatedly invokes “building blocks”) of
the social system can go ahead. His theory is an account of a building operation, an account
that presupposes the cleared space of the building site. His book has no name for this space,
in which the “set of independent individuals” that provide his “theoretical foundation” can
be conceived to exist. It is a significant silence. As I show later, we can find and name this
space, but only by stepping outside Coleman’s text.16
Raiding history
Coleman’s theoretical strategy, precise about derivations and formalization, is much vaguer
about the role of evidence. This was a key point for sociological positivism a generation
13 Coleman, Foundations, 43.14 Coleman, Foundations, 484–486.15 Coleman, Foundations, 30, 190, 213, etc.; indifference curves, chapter 25.16 Coleman, Foundations, 66. For an example of diagrams, see 889.
Springer
242 Theor Soc (2006) 35:237–264
ago, and marks a certain distance between Coleman’s positivism and strict sociological
empiricism. Coleman is not very much concerned with verification or falsification. But
he is consistently concerned to illustrate his argument. Brief worked examples pepper the
text.17
The principle that the theory is universally relevant allows Coleman to dip into any period
of history for these examples. As the book unfolds, examples are plucked from modern US
demography, a theatre fire, transnational corporations, US high schools, the South Sea bubble,
a student demonstration, medieval European land tenure, the constitution of the USSR, a
printing union, Eskimo polar bear hunts, and many more. In this respect, Foundations ofSocial Theory is strikingly traditional. This is the way evidence was deployed in Sumner’s
Folkways and other books of the pre-World War I era, though one must admit Sumner had a
richer store of ethnographic detail.18
Again a strong assumption of homogeneity is at work. Illustrations from any place, any
time, have the same relevance. Indeed imaginary examples have the same standing for Cole-
man as real ones. The text works as if the theory describes not just the real social world but
the only conceivable social world.
But a few of the examples feel different. Most of Coleman’s cases are drawn unproblem-
atically from the life of North America and Europe in the twentieth century. At one point in
the text, however, Coleman speaks of “primitive” societies, at another of “primitive tribes,”
at another of “natives” (citing, for the first and only time, Frantz Fanon). Late in the book
he gives, with an air of amusement, the example of a Bedouin husband riding while his wife
carries a burden on foot – and an American wife takes the family car. Early in the book, two
such cases pop up together: “nomadic tribes of the Sahara” dividing rights to a camel, and
Eskimos dividing the carcass of a bear.19
It seems that, despite the assumption of homogeneity, there is a heartland of Coleman’s
sociology, and also an exotic periphery. This is not just a matter of a few colorful and amusing
examples. There is something significant in the theorizing here – a dichotomy that goes back
to the earliest days of sociology.
Modernity
Well into the text, at chapter 20, Coleman opens a discussion of “modern society.” What is
distinctive about the modern, Coleman proposes, is a predominance of “purposively con-
structed” relationships over “natural” ones. This is part of:
a long-term historical development in which the primordial, natural environment is
replaced by a purposively constructed one. The change occurs in both the physical
environment and the social environment.20
This means the predominance of “the new, purposively constructed corporate actors” over
“primordial ties and the old corporate actors based on them (family, clan, ethnic group, and
community)”. A society of a new type has been produced over the last few centuries.
17 For the earlier emphasis on verification, see Hans L. Zetterberg, On Theory and Verification in Sociology,revised edition (Towota: Bedminster, 1963). I am grateful to C. Calhoun for calling attention to this difference.18 William Graham Summer, Folkways: A Study of the Sociological Importance of Usages, Manners, Customs,Mores, and Morals (Boston: Ginn and Company, 1906 [reprinted 1934]).19 Coleman, Foundations, 325, 607, 480, 783, 59.20 Coleman, Foundations, 552.
Springer
Theor Soc (2006) 35:237–264 243
Here Coleman is replaying the argument of an earlier book, The Asymmetric Society.
This line of thought is central to his political agenda for sociology, since he sees the loss of
“primordial ties” as constituting a deep social crisis. Coleman notes a similarity to Weber’s
story of rationalization. I would go further, and say his theory at this point depends on the
very traditional figure of sociological thought that constructs a global difference between
the modern and the primitive. This grand ethnography, characteristic of nineteenth-century
evolutionary sociology, is reproduced even in Coleman’s reply to critics of Foundations,
where he evokes “a fundamental structural difference between the societies now emerging
and all those that have gone before.”21
Coleman does not speak of “capitalism,” because he has no theory of accumulation.
He generally lumps the state and corporations together, on one side of a divide that has
“the family” on the other. This is actually more like Spencer than like Weber. Modernity
is both the creation of the new purposively constructed corporate actors (in more familiar
terminology, large-scale organizations) and the dissolution of the old. This yields a fluid
world of “freestanding” corporate actors “without a fixed relation either to natural persons or
to other corporate actors.” In fact this is our good friend, market society. Coleman’s theorizing
thus arrives at what it presupposed at the start.22
The map of the world
The exotic examples now fall into place. The “primitive tribes” whose members hunt bears,
cut up camels, and make the wives walk, are beyond the edge of the modern.
At a couple of points in the text, this edge is almost in view. One is the discussion of the
Palestinian revolt. The Palestinians are being drawn into “prosperity” by the Israeli economy,
yet turn against it, and start throwing rocks and committing arson. However, Coleman’s
interest is not in how this conflict of cultures and interests arose; it is in how well the course
of events matches theories of frustration and revolution.
Although his account of the “constitution” of a social system is overwhelmingly a consen-
sus theory (drawing on social contract models with a whiff of Parsons), Coleman acknowl-
edges that some systems are coercive. He calls the very coercive ones “disjoint constitutions”
where one set of actors creates arrangements that “impose constraints and demands on a dif-
ferent set of actors.” That might sound to you or me like the definition of an empire, or perhaps
the structural adjustment policies imposed on Latin America by US banks and the IMF. But
Coleman’s principal example is Stalinist paper constitutions that defined the workers as
beneficiaries and other classes as targets!23
Coleman’s account ignores the whole historical experience of empire and global domin-
ation. He never mentions colonies. He treats slavery briefly elsewhere, mainly in terms of
the intellectual problem that slavery creates for an exchange theory of society. (His mem-
orable solution is that it is rational for the slave to accept enslavement if the alternative is
death.)24
21 James S. Coleman, The Asymmetric Society (Syracuse: Syracuse University Press, 1982); James S. Cole-man, “The problematics of social theory,” Theory and Society 21/2 (1992): 263–283. For the idea of grandethnography, see R.W. Connell, “Why is classical theory classical?” American Journal of Sociology 102/6(1997): 1511–1557.22 Coleman, Foundations, 579.23 Coleman, Foundations, 327–328.24 Coleman, Foundations, 327, 86–88.
Springer
244 Theor Soc (2006) 35:237–264
Despite the universal ambitions of the theory, then, Foundations misses or misrepresents
vast tracts of human history, and ignores the majority of contemporary social experience.
This is a striking asymmetry, which, as we shall see, is not unique to Coleman.
In summary
Coleman’s general theory builds a picture of the person and social relations that is drawn from
recent European and especially North American social experience, reflecting the hypertrophy
of the market. His central model of social process presupposes a cleared space and suppresses
historical time. His theoretical strategy for the most part homogenizes history and social
experience, though it allows a linear narrative of modernization. There is every indication
that it is difficult for Coleman to “see” any experience different from that of his own society,
except through a residual idea of the primitive. Yet the form of the theory makes universal
claims about social systems and processes. Thus, market society and the bargaining individual
become the standards by which we understand all social process.
Agents of the gavotte: Giddens’s Constitution of Society
In 1984 Anthony Giddens published The Constitution of Society, with the subtitle “Outline of
the Theory of Structuration.” This text too was the culmination of a long project. His approach
can be seen developing through New Rules of Sociological Method, which gave an account of
practical action, Central Problems in Social Theory, which expanded on the relation between
action and structure, and A Contemporary Critique of Historical Materialism, which criticized
Marx’s view of world history and proposed an alternative.25
The Constitution of Society offers a summary (in fact, three summaries) of the matured
structuration framework, a detailed exposition of some of its themes, and some illustrations
of how the perspective could be applied. As a bonus, Giddens appends to most of the chapters
short essays on other theorists, in the style of his Profiles and Critiques in Social Theory,
explaining where their work corresponds to, and where it falls short of, structuration theory.
All the theorists discussed are men, and all are First World.26
Ambition
The task Giddens set himself in the series of books from New Rules to Constitution was
a reformulation of social theory as a whole, the reconciliation of conflicting intellectual
traditions, and the creation of a consistent conceptual framework for social research and
social critique. This magnificent project involved an enormous effort of synthesis, on a scale
hardly matched in modern social thought except by Habermas. It is wider in scope than
Bourdieu’s project and intellectually deeper than Coleman’s. In the Constitution, Giddens
criticizes and incorporates research ranging from psychoanalytic accounts of the development
of trust to Goffman’s anatomies of encounters, debates on the origins of the state, innovative
25 Anthony Giddens, New Rules of Sociological Method (London: Hutchinson, 1976); Anthony Giddens, Cen-tral Problems in Social Theory: Action, Structure and Contradiction in Social Analysis (London: Macmillan,1979); Anthony Giddens, A Contemporary Critique of Historical Materialism. Vol. I: Power, Property, andthe State (London: Macmillan, 1981).26 Anthony Giddens, Profiles and Critiques in Social Theory (London: Macmillan, 1982).
Springer
Theor Soc (2006) 35:237–264 245
work in geography, and the empirical sociology of education, taking in Parsons, Blau, and
Foucault along the way.
This tremendous range of reference makes sense because the object of knowledge is so
broad. Giddens says at the start:
The basic domain of study of the social sciences, according to the theory of structuration,
is neither the experience of the individual actor, nor the existence of any form of societal
totality, but social practices ordered across space and time. Human social activities, like
some self-reproducing items in nature, are recursive. . . To be a human being is to be a
purposive agent, who both has reasons for his or her activities and is able, if asked, to
elaborate discursively upon those reasons.27
The field of theory, then, is unbounded. It concerns social practices and human beings ingeneral. The theory of structuration embraces all social relations, all social structures, and
all societies. So Giddens can dip into the story of neo-Confucian China, then ancient China,
the financial moguls of the City of London, a car factory, a concentration camp. Because
the theory concerns all possible social relations, Giddens, like Coleman, has no hesitation in
analyzing imaginary examples as well.28
Like Coleman, however, he draws almost no examples from the colonized world. A striking
example is Giddens’s discussion of the development of autonomy in chapter 2. Giddens makes
effective use of Erik Erikson’s psychoanalytic model of human development. But he makes
no use of Erikson’s famous cross-cultural analysis in the very book, Childhood and Society,
being quoted. The result in Giddens’s hands is a universalized, completely abstracted, account
of human development – very much at odds with the emphasis on plurality and diversity in
the modern sociology of childhood.29
The business of theorizing
Giddens frames his task, in the “Introduction,” in terms of the history of social theory and
philosophy – for instance, coming to terms with the “linguistic turn.” He repeatedly re-writes
familiar sociological or psychological concepts in the language of structuration, much as
Coleman re-writes in the language of markets and choice.30
Giddens also undertakes to transcend dichotomies in existing theory. Although he does
this for various minor issues, by far the most important is the dichotomy (also transcended
by Bourdieu) between objectivism and subjectivism. Transcending this dichotomy leads
to Giddens’s basic principle of the “duality of structure,” whose child is structuration it-
self, “the structuring of social relations across time and space, in virtue of the duality of
structure.” The fundamental concept in Giddens’s theory thus arises, not from any con-
frontation with social problems, crises or transformations, but from a refined professional
27 Anthony Giddens, The Constitution of Society: Outline of the Theory of Structuration (Cambridge: PolityPress, 1984), 35.28 Giddens, Constitution, 165–168, 319–326, 128, 62; for imaginary examples 8–11, 81–82, etc.29 Erik H. Erikson, Childhood and Society (London: Imago, 1950); for a good example of modern childhoodresearch, see Marjorie Faulstich Orellana, Barrie Thorne, Anna Chee and Wan Shun Eva Lam, ‘Transnationalchildhoods: the participation of children in processes of family migration.’ Social Problems 48/4 (2001):572–591.30 Giddens, Constitution, xxii, 193ff.
Springer
246 Theor Soc (2006) 35:237–264
practice – reflection on the internal antinomies of a European/North American intellectual
tradition.31
The unbounded object of knowledge and the theorist’s willingness to re-write other peo-
ple’s work in a more abstract language give rise to a characteristic feature of Giddens’s
writing. The text of Constitution alternates between critical commentary on existing liter-
ature and frequent bursts of definition and concept-elaboration. Even a favorable reviewer
such as Jonathan Turner, when Constitution first came out, was moved to remark on the
“definitional texture” of the book. The glossary at the end is needed.32
This is the opposite of Coleman’s strategy of taking the smallest set of categories for the
longest possible walk. Giddens’s work reads as if the vastness of the field creates vacuums
that theory must expand to fill. The result is often both enthusiastic and banal – as we see
in a model of social change so generalized that it covers every episode in the history of the
world, yet says almost nothing about them.33
The knowledgeable agent
Where Giddens is in no degree banal, where he has a strong line and argues eloquently for
it, is in the theory of the agent. Giddens’s agent is not only active, as with Coleman and
Bourdieu, but also knowledgeable:
The knowledge of social conventions, of oneself and of other human beings, presumed
in being able to “go on” in the diversity of contexts of social life is detailed and dazzling.
All competent members of society are vastly skilled in the practical accomplishments
of social activities and are expert “sociologists.” The knowledge they possess is not
incidental to the persistent patterning of social life but is integral to it. . . Human agents
always know what they are doing on the level of discursive consciousness. . . .34
There are moments when Giddens on practical consciousness sounds very like Bour-
dieu on practical logic, as will be seen, but here the contrast is marked. Bourdieu emphaz-
ises misrecognition, Giddens emphasizes knowledgeability and competence. Accordingly,
in Giddens’s writing there is little of the irony one finds in Bourdieu’s.
Giddens’s idea of the knowledgeable agent is drawn, as the allusions in this quotation
suggest, from the later Wittgenstein and from ethnomethodology, not from Marxist theories
of praxis (where the idea of purposive and skilful action was both more collective and more
closely tied to social transformation). This genealogy has two consequences. Giddens’s
“agent” is an individual and is abstract.35
By “abstract” I do not mean that real individuals are entirely absent from Giddens’s text,
though they mostly are. More importantly, agency is understood in terms of the universal
requirements of the duality of structure. Consider, for instance, Giddens’s very effective
argument against the positivist search for “laws” in social science: “according to the view
31 Giddens, Constitution, 46, xx, 26, 162, 376.32 Giddens, Constitution, 31, 35, 176, 244; Jonathan H. Turner, “Review Essay: The Theory of Structuration,”American Journal of Sociology 91/4 (1986): 969–977.33 Giddens, Constitution, 244ff.34 Giddens, Constitution, 26.35 Karel Kosık, Dialectics of the Concrete: A Study on Problems of Man and the World (Dordrecht: Reidel,1976); John W. Murphy, “Yugoslavian (praxis) marxism,” Current Perspectives in Social Theory 3 (1982):189–205; for the agent as individual, see Giddens, Constitution, 163.
Springer
Theor Soc (2006) 35:237–264 247
suggested here, it produces a form of reified discourse not true to the real characteristics of
human agents.”
The “real characteristics” are the competencies that allow actors to constitute and re-
constitute social systems through their routine activities and interactions. To Giddens, these
capacities appear the same in all times and places, because what the agent is required for, in
the theory, is always the same. The recursive “stratification model of the agent” is therefore
described in terms as universalized as the model of the duality of structure.36
Yet where Coleman’s and Bourdieu’s agents are tacticians and bargainers, always with a
sharp eye out for a deal, Giddens’s agents are much more subdued and orderly. Giddens does
not, in fact, have a market model of the person. His accounts of agency emphasize routine,
trust, and coordination, the interlocking of activities between different agents. If Coleman’s
tacticians seem to be weaving across the floor in a fox-trot, Giddens’s diagrams seem to be
maps of a stately gavotte, executed by a ballroom full of well-trained dancers.37
Explaining the social
The agent may be an individual, but Giddens is emphatic that his theorizing does not start
with the individual, that to him, society is equally real. This is certainly true: Giddens’s
concept of agency does depend on a notion of the social order. But so does his concept of
the social depend on the notion of agency. In fact, the principle of the “duality of structure”
locks the two levels together logically. One is not emergent from the other, as in Coleman’s
theorizing, or (from the other end) in Althusser’s and Foucault’s conceptions of the subject.38
Giddens theorizes the social in two divergent ways. In the first mode, he is concerned with
how society is possible, how organized social existence can occur and persist. As Urry put it,
much of Constitution is “principally concerned with constituting an ontology of the social.”
Concepts such as “structuration” relate to these questions, and their extreme abstraction
results from Giddens trying to give answers that will be valid for any known, or any possible,
form of human social existence. Hence, such enormous categories as “reciprocity between
actors in contexts of co-presence” (English translation: “people doing things together face-
to-face”).39
Concern with the classic conservative problem of how society is possible leads Giddens
to re-define some social-scientific concepts drastically. “Structure” itself is one of these
concepts. Rejecting both the notion of discoverable empirical pattern (as in Lazarsfeld’s
latent structure analysis), and reversible system of transformations (as in Levi-Strauss’s
structural anthropology), Giddens arrives at this definition:
Structure thus refers, in social analysis, to the structuring properties allowing the “bind-
ing” of time-space in social systems, the properties which make it possible for dis-
cernibly similar social practices to exist across varying spans of time and space and
which lend them “systemic” form.40
36 Giddens, Constitution, 179, 5, 29.37 Giddens, Constitution, 29.38 Giddens, Constitution, 163.39 John Urry, “Book review: The Constitution of Society,” Sociological Review 34/2 (1986): 434–437; Giddens,Constitution, 28.40 Giddens, Constitution, 17.
Springer
248 Theor Soc (2006) 35:237–264
A hard-edged concept is thus dissolved into – to coin a term – “structurishness.” As
Giddens develops the argument, it becomes clear that “structure” is whatever the theorist
needs to postulate, in order to account for the persistence of any kind of social order.
The concern with “how to go on,” not only at the level of the individual playing a language
game but also at the level of society, leads Giddens to an equally abstracted treatment of power.
“Power” is disconnected from inequality and oppression, re-defined in Parsonsian style as
“the means of getting things done,” and made a property of all action:
. . . action logically involves power in the sense of transformative capacity. . .. Power is
not intrinsically connected to the achievement of sectional interests. . .. In this concep-
tion the use of power characterizes not specific types of conduct but all action. . .. We
should not conceive of the structures of domination built into social institutions as in
some way grinding out “docile bodies”. . . .41
A few pages down the track, this blandness has become an explicit conservatism: “power is
not an inherently noxious phenomenon,” and we can never have a society without domination,
whatever the socialists say. There is a significant contrast, at least in emphasis, with Giddens’s
writing in other books of the period, such as the Contemporary Critique of 1981, and TheNation-State and Violence, which appeared in 1985. In the first of these, Giddens proposes
a strong concept of exploitation. In the second, where Giddens is concretely studying the
history of the European state system, he has a strong emphasis on coercion, military force,
and war. One could not say that Giddens’s books contradict each other, but it does seem that
the task of creating universal theory in Constitution leads to a marked de-politicization of
concepts related to power.42
The gaze on history
Yet Giddens is also aware of the glorious diversity of human social experience. He has
read widely and is interested in history. So he has a second mode of theorizing, in which he
elaborates categories of social situations and processes: types of time, types of regionalization,
types of context, types of constraint, types of society, types of resources, and so forth.43
These categories too are abstract, but in a different way from the “structuration” categories.
They are meant to catch the ways in which situations differ, rather than what is common and
necessary to all social processes. Nigel Thrift has suggested that the nub of Giddens’s whole
argument is that “social theory must become more contextual.” These are the categories that
allow him to map the diversity of social action’s contexts.44
With Giddens’s enthusiasm for definition and his fertility in elaborating concepts, they
add up to a tremendous grid, through which one can gaze on human history from a great
height, seeing where each episode fits in an intelligible scheme. This view-from-above on
the whole story of human civilization gives a grandeur to Constitution more reminiscent of
Spencer and Comte than of Giddens’s contemporaries in the social theory trade.
41 Giddens, Constitution, 15–16, 283.42 Giddens, Constitution, 32, 256ff., 283; Anthony Giddens, The Nation-State and Violence: Volume Two of AContemporary Critique of Historical Materialism (Cambridge: Polity Press, 1985).43 Giddens, Constitution, 35, 121, 132, 176, 181–182, 258.44 Nigel Thrift, “Bear and mouse or bear and tree? Anthony Giddens’s reconstitution of social theory,” Sociology19/4 (1985): 609–623.
Springer
Theor Soc (2006) 35:237–264 249
The most important part of this grid defines types of society. Giddens is no functionalist,
and so he does not fall into the trap of assuming societies are neatly bounded systems. He is
also no evolutionist, Marxist, or Hegelian, so does not assume any unfolding of a grand logic
in history. But he comes up with a hierarchy that has a little flavor of both – and is, in its way,
as traditional as Bourdieu’s and Coleman’s schemes of modernity and pre-modernity.45
Giddens’s grand ethnography is a three-fold scheme distinguishing:
(1) Tribal society
(2) Class-divided society (roughly, with cities but without factories)
(3) Class society, or capitalism
This is obviously intended as a historical order, the later listed arising after the former,
though Giddens insists it is not an “evolutionary scheme.” The different types of society are
distinguished by different “structural principles” and are marked by different “contradictions”
(another concept Giddens re-writes and de-politicizes). The traditional character of Giddens’s
thinking is especially clear in relation to his first category. Tribal societies are closer to nature;
they are “cold,” i.e., not adapted to change; they are dominated by kinship and tradition; they
are segmented; etc.46
How are these types of society related to each other? To Giddens, the most important
point is that they are logically distinct. If a society is one, it is not the other. However,
once the later forms come into being, different types of society can co-exist, in contact with
each other within an inter-societal system. Giddens invents the term “time-space edge” to
define where one structuring principle gives way to another, or as we might say in ordinary
language, where one type of society encounters another. The relationship across a time-
space edge may be one of domination or of symbiosis, the concept itself is neutral. Thus,
Giddens arrives at a way of referring to something like imperialism without uttering the
word.47
Missing the empire
The relationship that Constitution does not theorize is colonization, the structuring principle
it does not explicitly name is imperialism, and the type of society that never enters its
classifications is the colony. (“Colonization” appears once in the index – as a reference to
Goffman’s research on asylums.)
For a world-spanning book of general social theory, written in the heartland of the greatest
imperial power the world ever saw, this is interesting. There seems to be something in
Giddens’s project and frame of reference that makes it difficult to address this aspect of
global history. The struggle for de-colonization was certainly one of the most dramatic and
important changes, on a world scale, in Giddens’s lifetime. All that the theory of structuration
can find to say about it is: “What is a ‘liberation movement’ from one perspective might be
a ‘terrorist organization’ from another.”48
That is it. That quotation is all there is to say about anti-colonial movements, de-
colonization, neo-colonialism, and post-independence struggles, as far as Constitution is
concerned.
45 Giddens, Constitution, 163ff, 236, etc.46 Giddens, Constitution, 182, 193ff.47 Giddens, Constitution, 184, 244, 164.48 Giddens, Constitution, 337.
Springer
250 Theor Soc (2006) 35:237–264
In Giddens’s other books there is a little more attention to the issue. The Nation-State andViolence offers a critique of world-systems theory, suggests why European sailors outgunned
others, and includes “colonized” and “post-colonial” in a classification of types of state. But
the limits of Giddens’s thinking are indicated by the fact that colonizing states are not named
in the same classification. The metropole vanishes from view (subsumed under the category
of “classical” states). Nowhere in Giddens’s writing of the 1970s and 1980s did the social
relations of empire come into focus as a major issue. And when, in the late 1990s, he came
to write a book about globalization, Runaway World, it was to persuade us that we are all
becoming inter-dependent, democratic, and de-traditionalized, and that old-style imperial
domination is no more.49
The problem is not just the absence of factual detail about the majority world in a book of
general theory. The model of types of society, i.e., the part of the theory where Giddens is the
most traditional sociologist, leads to a doctrine that systematically downplays the significance
of imperialism and the experience of conquered and colonized societies. In a crucial passage
of Constitution, Giddens explains that modern capitalism, the third type of society, is not like
the others, and did not evolve out of them. Rather, it resulted from “massive discontinuities”
that were:
introduced by the intertwining of political and industrial revolutions from the eighteenth
century onwards. The distinctive structural principle of the class societies of modern
capitalism is to be found in the disembedding, yet interconnecting, of state and eco-
nomic institutions. The tremendous economic power generated by the harnessing of
allocative resources to a generic tendency towards technical improvement is matched
by an enormous expansion in the administrative “reach” of the state. . ..50
That is to say, Giddens sees modernity as an endogenous change within Europe (or “the
West”), producing a pattern that is afterwards exported to the rest of the world. This is, of
course, the standard sociological view of the origins of modernity, encapsulated in ideas of
the “industrial revolution” and the “democratic revolution.” The crucial shifts occur around
the late eighteenth century. This picture is partly derived from Comte, partly from Marx,
and has been re-worked by Foucault and others in a darker, but structurally similar, view of
modernity and Enlightenment.
The actual dating is important, as we can see in the best-known alternative account. In
Immanuel Wallerstein’s model of the capitalist world-economy, the crucial shift is in the
sixteenth century. To Wallerstein, capitalism involved from the start a colonial economy:
within Europe, in relation to Poland and Scandinavia; overseas, with the conquests of the
Spanish and the Portuguese, followed by the Dutch, the British, and the French; and overland,
with conquests by the Russians and the North American European settlers. All of these
conquests were underway before the late eighteenth century, some were long over, and the
imperial powers had already fought wars over the spoils. In Wallerstein’s account, conquest
and colony/metropole relations are not a by-product of what Giddens blandly calls “the
increasing ascendancy of Western capitalist societies.” Rather they are constitutive of modern
capitalism as a system.51
49 Anthony Giddens, The Nation-State and Violence: Volume Two of a Contemporary Critique of HistoricalMaterialism (Cambridge: Polity Press, 1985), 269; Anthony Giddens, Runaway World (London: Profile Books,second edition 2002).50 Giddens, Constitution, 183.51 Immanuel Wallerstein, The Capitalist World-Economy (Cambridge: Cambridge University Press, 1979);Giddens, Constitution, 185.
Springer
Theor Soc (2006) 35:237–264 251
This is a very loaded difference. Giddens’s view implies that “the West” is dominant, not
because it conquered the rest of the world, but because of its “temporal precedence.” The
West industrialized and modernized first:
Rather than seeing the modern world as a further accentuation of conditions that existed
in class-divided societies [i.e., type 2, urban/agricultural but pre-industrial societies],
it is much more illuminating to see it as placing a caesura upon the traditional world,
which it seems irretrievably to corrode and destroy. The modern world is born out of
discontinuity with what went before rather than continuity with it. It is the nature of
this discontinuity – the specificity of the world ushered in by the advent of industrial
capitalism, originally located and founded in the West – which it is the business of
sociology to explain as best it can.52
Other social orders are passing away, not because Europeans with guns came and shattered
them, but because modernity is irresistible. On this point, Giddens remained entirely consis-
tent, because this was to be the core of his model of globalization, too.
In summary
Giddens undertakes a sweeping reformulation of social theory, operating entirely within a
European/North American intellectual tradition and trying to resolve its antinomies. The
object of knowledge is an unbounded concept of the social, the central theoretical categories
are stated in universal terms. A non-market but highly abstracted model of the knowledge-
able agent is developed. Concepts such as power and change are formulated in an abstracted
and de-politicized way. The conceptual system constructs a universal grid for viewing hu-
man history, as a system of differences among social forms. A grand ethnography is con-
structed, seeing modernity as discontinuous from other forms of society, the product of
endogenous change within “the West”. The whole issue of colonialism and empire is thus
occluded.
Southern tacticians: Bourdieu’s Logic of Practice
Pierre Bourdieu’s The Logic of Practice was also the child of a long gestation. Its first form
as an Outline of a Theory of Practice (Esquisse, which can also mean a sketch or draft)
was published in French in 1972; a revision was made for the English translation in 1977; a
further revision, meant to be definitive, hit print in 1980 as Le sens pratique, and was in turn
translated into English in 1990. But all this was only the later stage of an enterprise that began
in Algeria in the 1950s, as a study of Berber-speaking farming communities in Kabylia. A
large part of the Logic (and the Outline) describes the daily lives of these communities, in
dense ethnographic text interspersed with methodological comments.53
At the empirical level there could hardly be a greater contrast with Coleman’s and Gid-
dens’s texts. Here the focus is overwhelmingly on the global South. Nor was this an arbitrary
choice of subject-matter. Bourdieu’s Algerian experience was, on his own account, forma-
52 Giddens, Constitution, 239, 131.53 Pierre Bourdieu, Outline of a Theory of Practice (Cambridge: Cambridge University Press, 1977); PierreBourdieu, The Logic of Practice (Stanford: Stanford University Press, 1990); Pierre Bourdieu, “Retoursur l’experience algerienne,” in Franck Poupeau and Thierry Discepolo, Pierre Bourdieu, Interventions,1961–2001: Science sociale et action politique (Marseille: Agone, 2002): 37–42.
Springer
252 Theor Soc (2006) 35:237–264
tive in his “conversion” from philosopher to social scientist and in shaping his distinctive
approach to social science, especially his concern with reflexivity. As Tassadit Yacine shows
in some detail, the young Bourdieu became a field researcher in close collaboration with
Algerian students and colleagues such as Abdelmalak Sayad.54
This did not make Bourdieu an anthropologist in the conventional sense. As early as
1958 Bourdieu – as Sayad remarks, a “veritable entrepreneur scientifique” – had published
Sociologie de l’Algerie. Fourteen years later, when the Esquisse came out, Bourdieu had also
published influential work in the sociology of education and the sociology of culture. By
the time Le sens pratique came out he had also published Distinction, on class hierarchies.
By the time the Logic appeared in English, Bourdieu held the most prestigious academic
chair of sociology in France. Bourdieu was certainly grounded in anthropological theory – a
respectful discussion of Levi-Strauss is a point of departure for the Logic – but constantly
subverted the distinction by which anthropology studied the primitive and sociology the
advanced.55
Ambition
The Logic of Practice is an attempt to develop a credible basis for social-scientific knowledge,
in the form of an analytic strategy and conceptual language, and to show this approach at
work. What is at stake is more than academic. Bourdieu thinks his project has cultural,
political, and philosophical importance. As he says in characteristic rhetoric at the end of the
Preface:
By forcing one to discover externality at the heart of internality, banality in the illusion of
rarity, the common in the pursuit of the unique, sociology does more than denounce all
the impostures of egoistic narcissism; it offers perhaps the only means of contributing,
if only through awareness of determinations, to the construction, otherwise abandoned
to the forces of the world, of something like a subject.56
To get to the place where “something like a subject” will come into view, Bourdieu has
to deal with existing accounts of the social and subjectivity. His opening chapters therefore
critique both “objectivism,” as represented by structural linguistics, Levi-Strauss, and struc-
turalist Marxism; and (more summarily and angrily) “subjectivism,” as represented by Sartre
and rational choice theory.
Both critiques raise the question of the theorists’ own place in the theory. This point about
the structuralists is sustained through the book, and is a clue to Bourdieu’s intention. He is
trying to define limits of social science as well as state its foundations, and also to suggest
a view of intellectuals. On the one hand, he rejects structuralism because it takes a god-like
view of social reality. The theorists are not present in the world being theorized, therefore
cannot learn from analyzing their own social practice. In consequence, they impose a formal
logic on a world to which formal logic does not really apply. On the other hand, Bourdieu
54 Tassadit Yacine, “L’Algerie, matrice d’une oeuvre,” in Pierre Encreve and Rose-Marie Lagrave, editors,Travailler avec Bourdieu (Paris: Flammarion, 2003): 333–345; Tassadit Yacine, “Pierre Bourdieu, amusnawKabyle ou intellectuel organique de l’humanite,” in Gerard Mauger, editor, Rencontres avec Pierre Bourdieu(Bellecombe-en-Bauges: Editions de Croquant, 2005): 565–574; Abdelmalek Sayad, “Abdelmalek Sayad inInterview” (1996), reprinted in Derek Robbins, editor, Pierre Bourdieu (London: Sage, 2000): 59–77.55 Bourdieu, Logic, 21.56 Bourdieu, Logic, 16, 75, 41.
Springer
Theor Soc (2006) 35:237–264 253
rejects subjectivism because it refuses to recognize the constraints on social action, thinking
that practice can be understood purely from decisions of the will.
The domain
Although Bourdieu sometimes drops in a remark that “all societies” do this or that, he is not
after universal generalizations in Coleman’s style. Indeed, he is sharply critical of theorists
who think they have discovered transhistorical laws – this is not his idea of general theory.
Nor is he constructing a grand classificatory system as Giddens does. His project is more
epistemological and methodological than theirs.
Nevertheless Bourdieu does sweep across topic, time and place in fine style. While his
main examples come from Kabylia, he also offers an extended case study of a village in south-
western France, and from time to time offers examples from French politics, or French class
dynamics, or even further afield. For instance, in arguing that practices can be coordinated
without being governed by design or law, he remarks:
The coherence without apparent intention and the unity without an immediately visible
unifying logic (is this not what makes the ‘eternal charm of Greek art’ that Marx
refers to?) are the product of the age-old application of the same schemes of action and
perception which, never having been constituted as explicit principles, can only produce
an unwilled necessity. . .57
Universal social laws might be fetishes, but Bourdieu certainly works on an assumption
of the methodological homogeneity of the whole of human history. His theoretical tool-kit
is intended to work anywhere and everywhere. What Bourdieu universalizes is not a set of
propositions, but a scheme of analysis, expressed in a core set of concepts and examples of
how to use them. Rogers Brubaker nicely captures this by suggesting that what Bourdieu
offers is not a fixed propositional scheme but a theoretical habitus, a well-defined manner of
doing theorizing.58
The practical logician and his world
Bourdieu lays out the tool-kit twice in Logic: briefly in the preface, and more extensively
in Chapters 3, 7, and 8. These are the now familiar concepts of practice and structure,
strategy, social reproduction, habitus, field, symbolic capital, and domination. The concepts
of symbolic violence and the cultural arbitrary, central to Bourdieu’s sociology of education,
are not much in evidence in the Logic, but the rest of his contribution to modern sociological
theory is in view.59
My purpose here is not to criticize these concepts; I did this some time ago, and many
others have done so since.60 Rather, as with Coleman’s and Giddens’s concepts, I want to
57 Bourdieu, Logic, 13; for the village example, 147–161.58 Rogers Brubaker, “Social theory as habitus,” in Craig Calhoun, Edward LiPuma, and Moishe Postone,editors, Bourdieu: Critical Perspectives (Chicago: University of Chicago Press, 1993): 212–234.59 Bourdieu, Logic, 16; Pierre Bourdieu and Jean-Claude Passeron, Reproduction: In Education, Society andCulture (London: Sage, 1977).60 Craig Calhoun, Edward LiPuma, and Moishe Postone, editors, Bourdieu: Critical Perspectives (Chicago:University of Chicago Press, 1993); R.W. Connell, “The black box of habit on the wings of theory: Reflectionson the theory of social reproduction,” in Which Way is Up? Essays on Sex, Class and Culture (Sydney: Allenand Unwin, 1983); Derek Robbins, editor, Pierre Bourdieu (London: Sage, 2000); David L. Swartz and Vera L.
Springer
254 Theor Soc (2006) 35:237–264
ask what view of the world and its inhabitants is at work in them. At one level, there is a
striking similarity with Coleman’s theorizing and a contrast with Giddens’s. The “agent” in
Bourdieu’s world, the person who engages in practice, uses practical logic, and is the bearer
of the habitus, is very much a tactician, maneuvering for advantage in a world where he
confronts other tacticians, who are also maneuvering.
“Even when they give every appearance of disinterestedness,” Bourdieu remarks towards
the end of his exposition, “practices never cease to comply with an economic logic.” Strategies
are always seeking “profit” of one kind or another. Bourdieu’s peasant is a more sophisti-
cated bargainer than Coleman’s rational chooser, maneuvering simultaneously in several
dimensions of social reality, and letting some strategies unfold over long periods before a
return is reaped. But the vision of the agent as bargainer, and the social world as a ter-
rain of deals, is equally strong. Bourdieu energetically extends the market vision into ap-
parently non-market fields of social life, and goes even further than Coleman by dealing
extensively with cases in which the market logic is systematically denied by the people
themselves.61
This gives Bourdieu’s sociology a strongly ironic flavor. He is constantly debunking preten-
sions, and revealing the advantages sought by maneuvers that cannot be acknowledged as ma-
neuvers – which are therefore systematically mis-recognized. This remained a central theme
in Bourdieu’s writing about culture and social hierarchies after the Logic as well as before.
However, Bourdieu’s agent goes bargaining in a lumpier world than Coleman’s. Here the
debt to Levi-Strauss and Marxism is clear. There is no cleared space; the social world is
already shaped by structures, especially those of class and kinship. It is these structures that
give rise to the habitus, the internalized principles of action. (Here Bourdieu relies on a black-
box treatment of socialization.) These structures are re-generated through the deal-making of
the agents, who maneuver always within limits set by the habitus. Thus, Bourdieu’s theory
of practice becomes, systematically, a theory of social reproduction. Another layer of irony
is piled on.
Dance of the happy shades
A society or social formation, then, is at one level a self-regenerating set of structures, at
another level a set of agents engaged in an endless dance of strategizing, bargaining, and
exchange. Through this dance, whose rules are set by the structures, the structures reproduce
themselves. (“The strategies produced by the habitus . . . always tending to reproduce the
objective structures that produced them. . ..”) On this theme, Bourdieu’s theorizing most
resembles Giddens’s.62
The dance unfolds in time. Bourdieu insists strongly on this point. He has a whole chapter
on it, pointing out, for instance, that time is of the essence in gift exchange. It is his most
effective criticism of structuralism, and makes possible the realism of his vivid discussions
of practical strategies and tactics.63
Bourdieu’s “time” is, nevertheless, quite as abstract, quite as date-free, as the time invoked
in Coleman’s derivations. It is striking that a theorist as sophisticated as Bourdieu, and as
Zolberg, editors, After Bourdieu: Influence, Critique, Elaboration (Dordrecht: Kluwer Academic Publishers,2004).61 Bourdieu, Logic, 122, 109.62 Bourdieu, Logic, 61.63 Bourdieu, Logic, 105–106.
Springer
Theor Soc (2006) 35:237–264 255
aware of colonialism, nevertheless constantly uses the “ethnographic present” – without the
slightest hesitation or discussion - in this, his fundamental conceptual treatise. This anomaly
tells us something important about his theoretical project.
Time in Bourdieu’s theorizing is not only abstract, it is circular. The ironic effects of
the habitus constantly bends events back into their former patterns. Bourdieu is well aware
that in real life things do change, and the good old habitus becomes partly irrelevant to
new circumstances. But structural change is not what his theorizing explains. It is a familiar
criticism of Bourdieu’s sociology that the conceptual tool-kit does not contain devices for
analyzing social transformation. To put it another way, agents and practices are able to be
theorized by Bourdieu’s model insofar as their activities correspond to the model of social
reproduction.
The dance of practice, then, is a danse macabre, in which the ghostly emissaries of the
structures perform their semi-scripted revels, and at the end of each cycle of practice sink
back into their graves, i.e. their places in the structures. Time is of the essence, in the steps
of the dance. But at the level of the whole, history is frozen.
Where the women are
To make the rough social psychology of the habitus work as a mechanism of reproduc-
tion, Bourdieu has to make a strong assumption of cultural homogeneity. Coleman blandly
presupposes consensus by freely-choosing individuals in the “constitution” of social sys-
tems. Bourdieu is tougher-minded than that, he can see domination clearly enough. But his
theorizing also persistently presupposes a unified, interlocking social order.
This may sound strange in the sociologist who made differences in cultural capital so
central to the sociology of education. Yet in the Logic, Bourdieu constantly presents the social
order as culturally homogeneous, and Margaret Archer has shown that a similar assumption of
homogeneity underpins Bourdieu’s educational sociology. The most striking example in the
Logic is the ideal-type model of “The Kabyle House,” presented as simple reality. Bourdieu
laughs this off as “perhaps the last work I wrote as a blissful structuralist,” but he thought well
enough of it to reprint it in another book, and the same rhetorical device occurs throughout
the Logic. In the Logic there seem to be no debates among the Kabyle, no religious tensions,
no radical movements, and no prophecy.64
This feature of Bourdieu’s theorizing is very marked in relation to gender, an issue to
which he gives a lot of attention in the Logic. He draws an absolute dichotomy between the
man’s world and the woman’s world, making clear inter alia that his energetic bargaining
“agent” is a man. The gender system is mapped as a simple dichotomy and a simple hierarchy.
In a vivid passage, where Bourdieu is explaining how the habitus is built into the body, he
describes the stances of the manly man (upright, alert, etc.) and the well-brought-up woman
(stooped, eyes downcast, etc.).65
This schematic and archaic model of patriarchy, so much at odds with contemporary gender
research, was worked out for Kabylia, but Bourdieu clearly thought it was not confined there.
In Masculine Domination, one of his last books, he presented the same idea as a model of
64 Margaret Archer, “Process without system,” Archives Europeennes de Sociologie 24/4 (1983): 196–221;Bourdieu, Logic, 9; Pierre Bourdieu, Algeria 1960: Essays (Cambridge: Cambridge University Press, 1979).65 Bourdieu, Logic, 217, 70–72.
Springer
256 Theor Soc (2006) 35:237–264
universal patriarchy. Because of Bourdieu’s fame as a theorist this badly outdated formulation
is now having a considerable influence in some areas of gender studies.66
Grand ethnography
In most of the text, Bourdieu treats the world of the Kabyle, the world of metropolitan France,
and other milieux, as methodologically continuous. His arguments flow without interruption
from one case to the other.
This does not mean that he thinks all societies are of the same type. His experience in
Algeria argued against that, in agreement with the established view in social science. In several
passages of the Logic, Bourdieu discusses the “pre-capitalist economy.” In the chapter on
“modes of domination” especially, Bourdieu dichotomizes in a very traditional sociological
manner.67
This consists of presenting opposing pictures of the modern and the pre-modern as types
of society. In Bourdieu’s pre-modern, the material economy and the symbolic economy are
inextricably mixed. In the modern, after “the disenchanting of the natural world reduced to
its economic dimension alone,” they are separated into distinct fields. In the pre-modern,
social advantage must be continuously re-created by personal attention and effort. In the
modern, this is accomplished by institutionalization. For instance, modern society allows the
free circulation of cultural capital through a system of credentialling.68
What is crucial here is the form of argument. Bourdieu does not concern himself in TheLogic of Practice with the practical relationship between modern and pre-modern social
formations (though his own research in Algeria and Bearn had given him data on this is-
sue). Rather, like Coleman and Giddens, Bourdieu as a general theorist constructs a grand
ethnography, a concept of modernity defined via a sweeping contrast with the pre-modern.
In Bourdieu’s case, the pre-modern is described in much greater detail, because he actually
lived and studied there. Yet the underlying frame of thought is similar.
Light in the house
In his preface to the Logic, Bourdieu tells a memorable story. He was admiring some pho-
tographs of storage jars he had taken during his old fieldwork. The reason the photographs
were so good was that the roof of the house where he found them had been missing. The roof
was missing because it had been destroyed when the French army expelled the occupants.
The passage that includes this story is, I think, the only mention in the Logic that a war, indeed
an exceptionally bitter war, of colonial repression and liberation was raging in Algeria during
the time Bourdieu was doing his research.69
This is really remarkable. How could such an event as the Franco-Algerian war not seem
relevant to the analysis of practice, when fine details of parallel-cousin marriage do? It is
not because Bourdieu did not know the story. He had been sent to Algeria to do his military
service, stayed to research and teach in a hostile environment, eventually left Algeria under
66 Bourdieu, Logic, 77–79; Pierre Bourdieu, Masculine Domination (Stanford: Stanford University Press,2001).67 Bourdieu, Logic, 113, 123, 126; R.W. Connell, “Why is classical theory classical?” American Journal ofSociology 102/6 (1997): 1511–1557.68 Bourdieu, Logic, 117, 129–131, 132.69 Bourdieu, Logic, 3.
Springer
Theor Soc (2006) 35:237–264 257
the threat of violence from colonialist die-hards. He worked with Algerian colleagues, did
some research under the eyes of the military in “relocation camps” designed to frustrate
guerilla warfare, and did other fieldwork among peasants carrying weapons. In texts of the
late 1950s and early 1960s he had written about the disintegrating effects of colonialism and
the colonial war, and in the Logic he proposed an ethic of human solidarity. But still Bourdieu
did not see the anti-colonial struggle as essential material for his own statement of general
theory.70
There may be some biographical background to this. The most famous theorist of this
struggle was Frantz Fanon, who not only overlapped Bourdieu’s time in Algeria but like
Bourdieu made research trips to Kabylia, before leaving the country to work openly for
the Algerian FLN (National Liberation Front). Fanon’s L’an V de la revolution algerienneappeared in 1959, The Wretched of the Earth (with a famous preface by Sartre) in 1961,
while Bourdieu was still deeply engaged with Algerian issues. These books deal directly
with the practice that was transforming the society Bourdieu wrote about, yet they are never
mentioned in The Logic of Practice. No other participants in the Algerian struggle have
their ideas considered in the Logic, either. Bourdieu had long been contemptuous of the
schematic theories of revolution that circulated on the French left. He regarded Fanon and
Sartre specifically as purveyors of myth; he supported the colonized but wished to distance
himself from the doctrine of the FLN. He seems to have considered his own early sociology
as a cold dose of facts needed to educate people on both sides of the Algerian struggle.71
Nevertheless, at the deepest level, it is not Bourdieu’s political history but his conception of
theory that makes the anti-colonial struggle irrelevant. To arrive at “something like a subject,”
the European conceptual framing is self-sufficient. In the center of this debate, as Bourdieu
knew it from his early studies in philosophy onwards, there were no voices from Africa or
Asia. Bourdieu’s own project of creating a universally applicable tool-kit gave him no reason
to search out colonial voices, because it made irrelevant the specific history of the societies
through which the tools are illustrated – much as Coleman’s and Giddens’s examples could
come from anywhere.
Nor does his tool-kit require him to address a liberation struggle as a social process. Since
Bourdieu’s solutions to structure/agency problems constantly tend towards a theory of social
reproduction, the question of the dynamics of change is marginalized. As we see in many
other writings, Bourdieu is scathing about privilege and social domination; yet projects of
social transformation frequently meet his sociological irony.72
The result, in the Logic, is a text with a structure Bourdieu doubtless did not intend, but
that is all too familiar in European writings about the majority world. Knowledge about a
colonized society is acquired by an author from the metropole and deployed in a metropolitan
debate. Debates among the colonized are ignored, the intellectuals of colonized societies are
unreferenced, and social process is analyzed in an ethnographic time-warp. The possibilities
for a different structure of knowledge that undoubtedly existed in Bourdieu’s early research
are never realized in the later theorizing.
70 Pierre Bourdieu, Algeria 1960 (Cambridge: Cambridge University Press, 1979); Bourdieu, Logic, 15,112–113, 27–28.71 David Macey, Frantz Fanon, a Life (London: Granta, 2000); Pierre Bourdieu, “Retour sur l’experiencealgerienne,” in Franck Poupeau and Thierry Discepolo, Pierre Bourdieu, Interventions, 1961–2001: Sciencesociale et action politique (Marseille: Agone, 2002): 37–4272 For his scathing view of educational hierarchies over three decades see Pierre Bourdieu and Jean-ClaudePasseron, Les Heritiers: Les etudiants et la culture (Paris: Editions de Minuit, 1964); Pierre Bourdieu, TheState Nobility: Elite Schools in the Field of Power (Cambridge: Polity Press, 1996).
Springer
258 Theor Soc (2006) 35:237–264
In summary
Bourdieu builds a picture of the agent and social practice that is extensively illustrated from a
colonized society but responds intellectually only to European conceptual debates. He treats
different societies and periods as methodologically homogeneous. He develops an account of
the agent as a bargainer and strategist, in all dimensions of social life. But he sees the practices
involved as constrained in ways that reproduce social structures – themselves understood in
terms of schemata that construct dichotomies but not dynamics. Practice unfolds in time,
but time is circular. The exception to this is a very broad distinction between the modern
and the pre-modern. Bourdieu draws this as a conceptual opposition between types of social
formation, not as a concrete relationship, in disregard of his own knowledge of French
colonialism in Algeria.
The northernness of general theory
Having considered differences in the ways Coleman, Giddens, and Bourdieu theorize the
social, we are now in a position to formulate the common ground. As I remarked at the start,
no small group of texts can perfectly represent the whole genre. Nevertheless the problems
that emerge from these three can be found very widely in modern sociological thought.
Metropolitan geo-political location finds expression in theory through four main textual
moves, which I call here (a) the claim of universality, (b) reading from the center, (c) gestures
of exclusion, and (d) grand erasure.
The claim of universality
In each of the three texts, there is a strong and repeated claim to universal relevance. Coleman’s
ambition to analyze any kind of social system, Giddens’s unbounded object of knowledge, and
Bourdieu’s generalized models of the agent and of reproduction, embody a common claim.
To these authors and many others, the very idea of theory involves talking in universals
that refer to the social, structure and agency as such, etc. It is assumed that all societies are
knowable, and that they are knowable in the same way and from the same point of view.
That this point of view originates in the metropole is not explicitly acknowledged. Indeed,
this fact has to remain tacit – for if it were made explicit, universal relevance would im-
mediately be called in question. Social scientists in the periphery cannot universalize a loc-
ally generated perspective because its specificity is immediately obvious. It attracts a proper
name, such as “Latin American dependency theory,” and the first question that gets asked is
– how far is this relevant to other cases? It is only from the metropole that a credible tacit
claim of universality can be made.
The claim of universality is not confined to the practice of making universal statements.
The claim can also be made through method. An example is the practice of re-writing other
social scientists’ work in one’s own conceptual language, a common practice of both Coleman
and Giddens. This re-writing is never just a translation. It is a subsumption, in which the
universal relevance of the preferred theory is implicitly claimed. Each re-writing is offered
as an example, with the implication that any other case could be subsumed in the same way.
Springer
Theor Soc (2006) 35:237–264 259
Reading from the center
Contributions to general theory are often presented as resolutions of some antinomy, problem,
or weakness in previous theory. All three of our texts present themselves this way. It is a
professional requirement – one must relate one’s work to the literature. But whose literature?
All three texts address problems that arise in a metropolitan theoretical literature, and no
other.
For instance, Giddens and Bourdieu both focus on the antinomy of objectivism vs. sub-
jectivism. This is a classic problem for European cultural and social sciences. But it is nota central problem for colonial intelligentsias, either in conquered cultures or colonies of
settlement. The reason is apparent when we look at what objectivism and subjectivism share.
They are alternative ways of picturing oneself at the center of a world, alternative models
of actions or systems with no specific external determinations. To take Bourdieu’s cases: In
Claude Levi-Strauss’s “objectivism,” the system of transformations that constitute a structure
of kinship or myth is logically a closed system; the theorist stands at the center from where
alone the transformations can be seen as a system. In Jean-Paul Sartre’s “subjectivism,” praxis
is analyzed in order to build an account of the collective agent of a unique world-historical
transformation, which alone will transcend the universal condition of scarcity. A general
social theory shaped around the objectivism/subjectivism problem necessarily constructs a
social world read through the metropole – not read through the metropole’s action on the rest
of the world.73
It is common for metropolitan theorists, using universal language but basing themselves
on personal knowledge or local research, simply to generalize the specific experience of
metropolitan countries. Coleman’s model of agency, based on the entrepreneur in the North
American market-place, is a classic example.
Bourdieu’s theorizing is more complex in this respect, since he is using a case study
from the South. But he achieves a similar effect by bracketing most of what makes that case
specific. For instance, he almost entirely deletes Islam from his picture of Algerians, so the
profit-seeking bargainer emerges as a pure type. Even his subtle and interesting argument
about the incoherence of practical logic depends both on postulating the indeterminacy that
is really a characteristic of the metropole, and on bracketing the colonial situation of the
Kabyle. Thus, Bourdieu places his case study in the magically undetermined space of the
ethnographic present.
A very important case of reading from the center concerns time. All three theorists treat
time as an important issue. The time their theories suppose is generally abstract, i.e., date-
free, and continuous (with Bourdieu’s circular time, and Giddens’s reproduction cycles, a
special case). In the grand ethnographies, they offer the world-time of an intelligible historic
succession (pre-modern to modern, pre-capitalist to capitalist, etc.).
Continuous time, and the time of intelligible succession, is time as experienced in the
metropole. Colonial time is different. In colonized and settler societies, time involves fun-
damental discontinuity. Time involves a succession that is, from within indigenous culture,
unintelligible. One cannot predict colonial conquest from within the social experience of
the about-to-be-colonized society. Let me give one example, from a multitude. In the early
nineteenth century, the British imperial state took over from the Dutch in South Africa, and
73 Claude Levi-Strauss, Structural Anthropology (New York: Basic Books, 1963); Jean Piaget, Le structural-isme (Paris: P.U.F., 1968); Jean-Paul Sartre, Critique of Dialectical Reason: I. Theory of Practical Ensembles(London: New Left Books, 1976).
Springer
260 Theor Soc (2006) 35:237–264
soon stepped up the level of colonial violence, burning and killing in Xhosa settlements. The
historian J. B. Pieres describes the Xhosa experience:
Total war was a new and shattering experience for the Xhosa. . .The havoc wrought by
the colonial forces was not only cruel but incomprehensible. . . Now that this foreign
entity had crystallized as a threat there was no telling where it would all end.74
For colonized cultures, conquest is not evolution, rationalization, or transformation, but
catastrophe. Colonization introduces fundamental disjunctions into social experience that
simply cannot be represented in metropolitan theory’s models of change through time. This
is not a matter of just one historical moment, the instant of conquest. It is carried forward
in the structure of colonial society, and carried forward again into the post-colonial world.
Ignoring this disjunction is a basic flaw in many contemporary accounts of globalization.
Gestures of exclusion
Theorizing addressed to problems arising in the culture of the metropole generally proceeds
by quoting and debating other texts from the metropole. The theorist’s reading list is always
an interesting document. Who is not on the reading list is as interesting as who is.
I have noted the significant absence of Fanon, indeed the whole Algerian liberation move-
ment, from Bourdieu’s exposition of the theory of practice. Theorists from the colonized
world are very rarely cited in metropolitan texts of general theory. There is a notable absence
of reference to Islamic thought, given the historic interplay of Islamic and Christian cul-
tures, and the wealth of Islamic discussions of modernity. It would be interesting to see how
Coleman’s “sovereign individual” would survive within a cultural presumption of “Tawhid,”
the unity of the divine and of the world, which has been seen by some intellectuals as the
foundation for an Islamic approach to science.75
At times, texts of general theory include exotic items from the non-metropolitan world.
Examples are Coleman’s references to Eskimos and Bedouin, and Giddens’s passage on
Confucian China. These add color to the texts, but do not affect their intellectual structure.
They do not introduce ideas from the periphery that have to be considered as part of the
dialogue of theory.
More integral to the theory (and perhaps explaining how the “color” items work) are
mechanisms that define “us” and “them.” Particularly important are the formulae I have called
“grand ethnography,” emphasizing the modern/pre-modern distinction. Grand ethnography
now often includes “post-modern society” as a category (though not in the three theorists
discussed here). Alternatively, “modern society” can be seen as expanding and swallowing
all the rest. This is Giddens’s view in Constitution, later elaborated in The Consequences ofModernity and Runaway World. It is a widespread sociological view of globalization.76
74 J.B. Peires, “Nxele, Ntsikana and the origins of the Xhosa religious reaction,” Journal of African History20/1 (1979): 53–54.75 Behrooz Ghamari-Tabrizi, “Is Islamic science possible?” Social Epistemology 10/3–4 (1996): 317–330. Forthe Iranian debate on modernity see Farzin Vahdat, God and Juggernaut: Iran’s Intellectual Encounter withModernity (Syracuse: Syracuse University Press, 2002) – a book that strikingly illustrates the problem of“reading from the center” since the treatment of Iranian thought is framed by Kant and Hegel!76 Anthony Giddens, The Consequences of Modernity (Stanford: Stanford University Press, 1990); AnthonyGiddens, Runaway World: How Globalisation is Reshaping our Lives (London: Profile Books, 2002).
Springer
Theor Soc (2006) 35:237–264 261
In either case, the social thought of colonized cultures is rendered irrelevant to the main
theoretical conversation. It is construed as belonging to a world that has been in one way or
another surpassed.
Grand erasure
Michel Foucault’s Discipline and Punish begins with a famous contrast: the spectacular
and brutal exemplary punishment of a would-be royal assassin in the eighteenth century,
compared with the regulated, time-tabled, concealed management of juvenile offenders in
the nineteenth century. Two different penal styles, remarks Foucault, defining “a new age for
penal justice.” By the mid-nineteenth century, punishment was becoming hidden, “punitive
practices had become more reticent. One no longer touched the body, or at least as little as
possible.”77
Quite in passing, Foucault notes that this change took place in “Europe and the United
States.” He gives no reason for confining the argument to those regions, but it is just as well
he did. If he had included the colonies, the argument would be false. One hundred years
after the execution of Damiens the regicide, when Foucault’s “reticence” was supposedly in
full flow, the British executed a large number of men they captured while suppressing the
“Indian Mutiny” in 1857–58. They did it in public, with exemplary brutality, including mass
hangings and floggings, caste degradation of leaders, and blowing rebels from the cannon’s
mouth. Public, spectacular, collective punishments remained a favored technique of British
and French colonialism far into the twentieth century. Notable examples are the punitive
massacres at Setif and Kerrata in 1945, to intimidate the populations of northern Africa, just
after France itself had been liberated from the Nazis.78
Following the suicidal atrocity of 9/11, the technique was again used by Western forces
on Muslim populations, in Afghanistan and Iraq. “Operation Infinite Justice” devastated a
society and destroyed a government in retaliation for the World Trade Center attack. The US
assault on Falluja in 2005 produced hundreds of deaths immediately, and more in the long
run, as punishment for the killing of four US mercenaries. “Shock and awe,” the slogan of
the US military during the attack on Iraq, is not a bad definition of what Foucault thought we
had left behind in the eighteenth century.
Colonial war is erased from Bourdieu’s Logic. Colonial relationships of all kinds are
erased by using the ethnographic present. The inherently divided culture of colonialism
cannot be modelled in Bourdieu’s reproductionism nor in Coleman’s derivations. Nor, for
that matter, can it be modelled in sociological functionalism or anthropological structuralism,
nor in ethnomethodology’s notion of the competent member of a culture, which underpins
Giddens’s notion of the agent.
The defining politics of colonial and post-colonial society cannot be modelled by the
de-politicized notions of power in Giddens and Coleman, nor by exchange theory generally.
The impossibility is sufficiently indicated by Coleman’s ludicrous attempt to theorize slavery
within rational choice theory, where the slave is supposed to have bought his right to stay alive.
The erasure of colonial experience and social process is so common in metropolitan social
theory that it usually goes unnoticed. It may even be built in to the orthodox disciplinary
view of a key issue, as in the erasure of imperialism from mainstream sociological accounts
of the emergence of modernity.
77 Michael Foucault, Discipline and Punish: The Birth of the Prison (New York: Pantheon, 1977), 7, 11.78 Saul David, The Indian Mutiny, 1857 (London: Viking, 2002); 145–146.
Springer
262 Theor Soc (2006) 35:237–264
Something even more drastic can happen. In discussing Coleman’s text, I noted how his
account of social system-building presupposes a featureless, cleared space, and I suggested
we could discover where that space is. It is certainly not in Europe. But it does exist in the
neighborhoods of Chicago and Sydney.
Chicago was an exemplary new town in a colony of settlement, where space was cleared
by the “westward expansion” of the United States – a process of eliminating the society, and
much of the population, that had been there before. Sydney was also a new town, the main
port of entry for the British conquest of Australia. In British law, Australia at the time of
settlement was understood as terra nullius, “land belonging to nobody.” An entire continent
could therefore be claimed for the crown and distributed at the colonial government’s pleasure.
The deep connection that a whole existing population had to the land was simply obliterated.79
Terra nullius, the colonizer’s dream, is a sinister presupposition for social science. It is
invoked every time we try to theorize the formation of social institutions and systems from
scratch, in a blank space. Whenever I see the words “building block” in a treatise of social
theory, I wonder who used to occupy the land.
Looking south
In the body of this article, I have shown deeply problematic features of general sociological
theory as practiced by some of its most eminent figures. I now turn briefly to the question
of alternatives. Can we have social theory that does not claim universality for a metropolitan
point of view, does not read from only one direction, does not exclude the experience and
social thought of most of humanity, and is not constructed on terra nullius?
I believe we can. In fact, we have a good deal of it already – though not much is on the
reading lists of courses in sociological theory. There are even moments in texts of general
theory that suggest new possibilities, moments when the edge of the metropole flickers into
view, when light comes through the roof.
The alternative to “northern theory” is not a unified doctrine from the global South. No
such body of thought exists nor could it exist. Indeed, one of the problems about northern
theory is its characteristic idea that theory must be monological, declaring the one truth in
one voice. It seems to me that a genuinely global sociology must, at the level of theory as
well as empirical research and practical application, be more like a conversation among many
voices.
On this view, elements of a far more inclusive sociology exist in a number of well-
established bodies of thought. One is the Islamic debate about modernity already mentioned.
Another is the African discussion of “indigenous knowledges” and the possibility of an
African renaissance. A third is the theorization of autonomy, dependence, and globalization
conducted in Latin America. A fourth is the international feminist critique of metropolitan
hegemony, and the development of global dialogue among different feminisms. A fifth is the
Indian debate on what Ashis Nandy calls “culture, voice and development.”80
79 Murray Goot and Tim Rowse, editors, Make a Better Offer: The Politics of Mabo (Sydney: Pluto PressAustralia, 1994).80 Farzin Vahdat, God and Juggernaut: Iran’s Intellectual Encounter with Modernity (Syracuse: SyracuseUniversity Press, 2002); Catherine A. Odora Hoppers, editor, Indigenous Knowledge and the Integration ofKnowledge Systems: Toward a Philosophy of Articulation (Claremont: New Africa Books, 2002); MiltonSantos, Por uma outra globalizacao (Rio de Janeiro: Editora Record, 2003); Chilla Bulbeck, Re-OrientingWestern Feminisms: Women’s Diversity in a Postcolonial World (Cambridge: Cambridge University Press,
Springer
Theor Soc (2006) 35:237–264 263
This list is far from complete, but may serve to indicate the wealth of resources for
theorizing the social that can be seen when we look beyond the metropole.
The extreme abstraction in the three theories examined in this article suggests another
line of thought. Creating a separate domain of general theory is not the only way to do
conceptualization in social science. It may be that this form of specialization, on the face
of it reflecting the metropole’s claim to a view of the whole, uniquely embeds metropolitan
hegemony. If this is so, we may very fruitfully draw on other social-scientific traditions that
have linked theorizing with social struggle, and that have tried to democratize the production
of knowledge.81
Other paths for theory, therefore, exist. As we follow them, a number of problems promptly
arise. Do non-metropolitan intellectuals also write northern theory? Certainly. I can speak
with some authority here, as I have done it myself – Gender and Power, for instance, has most
of the marks of “northern theory” identified in this paper. Yet when social scientists in the
periphery theorize as if from the North, they create profound difficulties for the understanding
of social relations in their own regions. The dilemmas of an intellectual formation such as
“Australian sociology,” as traced in the recent history by John Germov and Tara McGee,
largely arise from this process.82
On the other side, can metropolitan intellectuals escape the effects explored in this article?
Indeed they can. But there are costs in doing so, including the very heavy commitment of
time involved in cultural re-tooling, and risks to professional credibility (consider what an
acceptable citation list is for a paper in a “mainstream” North Atlantic journal). And there
are many difficulties for metropolitan social theorists in entering dialogues with the majority
world – among them difficulties of language, of limited personal contact, and ethical problems
about the appropriation of knowledge.
In this article, to get the analysis going, I have operated with the simplest possible
metropole/periphery model. Yet both terms in this dichotomy are complex. The production of
theory is a very different enterprise in affluent peripheral countries such as Australia and poor
peripheral countries such as Indonesia. This difference partly corresponds to the distinction
between settler colonialism and commercially-driven conquest, and thus involves different
historical trajectories. The social theory of a writer like Nandy in India, drawing directly on a
very rich indigenous intellectual history, will read differently from theory produced in settler
communities such as the Afrikaners in South Africa or the British in Australia.83
From the periphery, the metropole often appears as a solid bloc, edged with privilege. But
the metropole too has its hierarchies, exclusions, and struggles for legitimacy. This is perfectly
shown by Pierre Bourdieu’s dramatic rise from a lower-middle-class family in a remote rural
corner of France to a pinnacle of influence in the most prestigious institutions of Paris. It
is not surprising that these hierarchies are a central theme of his sociology of intellectuals.
1998); Vinay Lal, Empire of Knowledge: Culture and Plurality in the Global Economy (London: Pluto, 2002);Ashis Nandy, Bonfire of Creeds: The Essential Ashis Nandy (New Delhi: Oxford University Press, 2004).81 Not a new idea – cf. Sven Lindqvist, Grav dar du star: Hur man utforskar ett job (Stockholm: Bonniers,1978).82 R.W. Connell, Gender and Power: Society, the Person and Sexual Politics (Cambridge, Polity Press, 1987);John Germov and Tara Renae McGee, editors, Histories of Australian Sociology (Melbourne: MelbourneUniversity Press, 2005).83 For the Afrikaner case, see Hermann Gilomee, “‘Survival in Justice’: An Afrikaner Debate over Apartheid,”Comparative Studies in Society and History 36/3 (1994): 527–548; for the Australian, see John Germovand Tara Renae McGee, editors, Histories of Australian Sociology (Melbourne: Melbourne University Press,2005).
Springer
264 Theor Soc (2006) 35:237–264
Across the Atlantic, similar points are made in Camic and Gross’s “new sociology of ideas.”
Internal hierarchies, such as the internal colonialism affecting the Black population of the
United States and the guest workers of Europe, must affect the way the metropole operates
in the global production of social knowledge.84
For all these reasons we should not underestimate the difficulties of a more inclusive
theoretical project. But do we have any choice? It seems to me that the project of metropolitan
social theory, in which the work of Giddens, Bourdieu and Coleman represent genuine
pinnacles of achievement, is now exhausted. The problems mapped in this article cannot be
overcome within this tradition of thought.
At the same time, social science is very much needed as a cultural force, with the worldwide
triumph of the neo-liberal market agenda, and the retreat from dialogue on the part of dominant
powers, both political and economic. And theory is a crucial part of what makes social science
a cultural force. But under the conditions we now face, monological northern theory cannot
do the job. We really have no choice but to face the difficulties of “doing theory” in a globally
inclusive way.85
Acknowledgments This article was first presented at the Theory Section mini-conference on TheoreticalCultures, at the 99th annual meeting of the American Sociological Association, San Francisco, 17 August2004. I am grateful to the discussant, Craig Calhoun, to Michele Lamont and Neil Gross, and to all otherparticipants. Excellent advice for revision was given by Theory and Society reviewers. In the longer run, thisarticle derives from my “Classical Theory” course at the University of California, Santa Cruz; my thanks toparticipants and colleagues, especially Behrooz Ghamari-Tabrizi, John Sanbonmatsu, Paul Lubeck, and TerryBurke. I am grateful for advice and materials from Frederic Vandenberghe and Dean Ashenden. The projectwas encouraged by University of Sydney colleagues in a recent exploration of social theory, Toni Schofield,Robert van Krieken, and Julian Wood. This article benefited greatly from the dedicated research assistance ofJohn Fisher and Molly Nicholson.
About our contributor
Raewyn Connell is University Professor at the University of Sydney, and author, co-author,
or editor of nineteen books, including Ruling Class Ruling Culture (1977), Making the Dif-ference (1982), Gender and Power (1987), Schools and Social Justice (1992), Masculinities(1995), The Men and the Boys (2000), and Gender (2002). Connell is an Editor of Theoryand Society. A contributor to research journals in sociology, education, political science, gen-
der studies, and related fields, her current research concerns social theory, neo-liberalism,
corporate masculinities, gender practices, and intellectual labor.
84 Charles Camic and Neil Gross, “The new sociology of ideas,” in Judith R. Blau, editor, Blackwell Companionto Sociology (Cambridge: Blackwell, 2001).85 The case for the specific historical importance of sociology (broadly understood) in a period of neo-liberaldominance is made in R.W. Connell, “Sociology and world market society,” Contemporary Society 29/1(2000): 291–296.
Springer