Cultural Sociology http://cus.sagepub.

com

The Work of Culture

Tony Bennett
Cultural Sociology 2007; 1; 31
DOI: 10.1177/1749975507073918

The online version of this article can be found at:

http://cus.sagepub.com/cgi/content/abstract/1/1/31

Published by:

http://www.sagepublications.com

On behalf of:

British Sociological Association

Additional services and information for Cultural Sociology can be found at:

Email Alerts: http://cus.sagepub.com/cgi/alerts

Subscriptions: http://cus.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations (this article cites 2 articles hosted on the

SAGE Journals Online and HighWire Press platforms):
http://cus.sagepub.com/cgi/content/abstract/1/1/31#BIBL

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 Cultural Sociology
Copyright © 2007
BSA Publications Ltd®
SAGE Publications
Los Angeles, London,
New Delhi and Singapore
Volume 1(1): 31–47
DOI: 10.1177/1749975507073918

The Work of Culture

■ Tony Bennett
Open University, UK

ABSTRACT
The formulations of cultural sociology have a tendency to merge culture and the
social so closely that they become indistinguishable from one another. Drawing on
Foucauldian governmentality theory and actor network theory, this article argues
that it is preferable to examine the processes through which culture is separated
off from the social via the production of distinctive cultural assemblages.The kinds
of issues that need to be taken into consideration to account for the work that
goes into making culture as a publicly differentiated realm are identified. Attention
then focuses on the kinds of work that culture does in being brought to act on
the ‘working surfaces on the social’ that are organized in the relations between
social and cultural knowledges. The argument is exemplified by considering how
the assemblages of Aboriginal culture produced by Baldwin Spencer enabled the
production of a new surface of social management through which the relations
between white and Aboriginal Australia were organized in the context of the
Aboriginal domain.

KEY WORDS
Aboriginal domain / absolute racism / actor network theory / assemblage /
governmentality / the social

Introduction

he launch of a new journal under the title Cultural Sociology is as good a

T moment as any to reflect on the terms in which sociology and culture have
been related to one another, and to ask whether the time is now ripe for
new terms of theoretical and methodological engagement between the two.
I want to suggest, though, that this should include some consideration as to

31
Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007
© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
32 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

whether the rubric of ‘cultural sociology’ is necessarily the best one for this
purpose. Its value as an open-ended title for a journal that aspires to serve as a
venue for broad debate on these questions is, of course, clear. It serves this pur-
pose better than ‘the sociology of culture’, and not just because this formula-
tion implies a field of inquiry that is limited to the institutions and practices
comprising the arts, cultural industries and media sectors. It is rather the theo-
retical implication that ‘society’ is to be invoked as an explanatory ground in
relation to those institutions and practices that is the problem.
‘Cultural sociology’ is much looser and more pliable in these respects. It
encompasses the arts, cultural and media sectors but as parts of a broader soci-
ological canvas which, indeed, is often stretched so as to be coterminous with
that of sociology as a whole. Resonating to the logic of ‘the cultural turn’, it
implies that as there is no determining ground of the social that exists outside
culture, with the implication, in many formulations, that the very idea of a non-
cultural sociology would be an oxymoron. It is also a formulation that estab-
lishes useful historical affiliations. After all, wasn’t ‘the sociological tradition’,
from Durkheim to Parsons, concerned precisely with the role of norms, beliefs,
customs, etc. – and so, in its extended sense, culture – in the organization of
social life?1 And, finally, it makes good international connections with both the
French and American schools of cultural sociology.
Yet I do have some doubts about the theoretical value of bringing culture and
sociology together on these terms chiefly because they often bring in tow a ten-
dency to merge culture and the social so closely together that they become indis-
tinguishable. This is not to dispute the prevailing contention that, in a general
sense, cultural practices are implicated in the make-up and organization of social
relationships – although this has become so familiar a claim that its value is now
more-or-less doxological, a ritual invocation that occludes more than it reveals.
For, if analysis does not push beyond such general formulations to consider more
closely the varied mechanisms through which culture and the social are con-
nected, it can only too easily result in a set of ghostly, disembodied agents – val-
ues, beliefs, meanings, narratives – being credited with the ability to perform
heroic tasks: securing social cohesion, or bringing about civic renewal, for exam-
ple. Moreover, it often seems that this is to be accomplished quite effortlessly
without any distinctive kind of work being involved in either producing culture
as a specific realm with identifiably specific and concrete actors, or organizing the
interfaces through which culture is able to connect with and act on the social.
As a corrective to this, then, I want to propose some principles for
analysing the relations between culture and the social that will, first, accord sig-
nificant attention to the work of making culture. This involves, I shall suggest,
a focus on the material processes through which culture has come to be, and
continues to be, differentiated from the social, the latter understood, like cul-
ture, as itself a historical rather than an anthropological reality. It also involves
attending to the further work through which culture, in its varied differentiated
forms, then acts on the social through the ‘working surfaces on the social’ that
it produces. These are organized by specific cultural knowledges and techniques

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 33

of intervention which operate either directly, or in combination with the

operation of specific social knowledges, to format the social for specific kinds
of action and intervention.2
It is, then, these two aspects of the work of culture – the work of making
it, and the work it does – that I think need to be placed at the centre of any pro-
gramme for the analysis of the relations between culture and the social.3 I shall,
in developing this perspective, put three theoretical traditions into play as a
means both of identifying a somewhat fractured genealogy for the arguments I
shall propose and of elaborating their implications. They comprise, first, the
rapidly emerging field of post-Bourdieusian sociology which, shaped by its
interactions with Bourdieu’s key concepts, is currently poised somewhere between
proposing a series of internal corrections to and qualifications of Bourdieu’s
work, and going beyond it to map out a theoretical territory that is constructed
on different premises. I include here Bernard Lahire’s work on the sociology of
individuals and its critical implications for Bourdieu’s understanding of the
concept of habitus (Lahire, 2001, 2003, 2004), and the work of Antoine
Hennion and others in its concern to find a place for things in the networks of
relations that Bourdieu theorizes as fields (Hennion, 1997; Gomart and
Hennion, 1999; Hennion et al., 2005).4 Second, I shall draw on the work of
Bruno Latour, and the traditions of actor-network theory, science studies, and
practice studies as well as those tendencies in anthropology – the approaches to
art developed by Alfred Gell, for example (Gell, 1998) – which focus attention
on tracing the socio-material networks of relations through which heteroge-
neous elements (including human and non-human actors) are brought together
in specific forms of action and interaction. While there are close connections
between these traditions and post-Bourdieusian sociology, they clearly depart
from the explanatory logic of Bourdieu’s work in which a hidden structure,
brought to light by the sociologist, is invoked to account for the phenomenal
level of observable differences in dispositions and practices – differences in cul-
tural tastes, for example.5 In lieu of this, actor-network theory proposes a sin-
gle-levelled reality which, since there are no hidden depths or structures to be
fathomed, merges the process of explanation with that of description: to
describe how socio-material networks of relations are assembled, disassembled,
and reassembled in new configurations is – if the range of the networks that are
thus traced is extensive enough – also to explain how those networks are made
up and operate. And third, I shall draw on two aspects of Foucault’s work. In
the first instance, I draw on his concept of dispositif or apparatus in view both
of its influence on actor-network theory and of the respects in which Foucault’s
understanding of such apparatuses as always somehow ready made is usefully
corrected by actor-network-theory’s focus on the processes through which het-
erogeneous elements are assembled into specific apparatuses.6 In the second
instance, I explore the implications of Foucault’s work on governmentality
(Foucault, 1991) which, in historicizing the relations between culture, the econ-
omy and the social, offers an account of the social that has a sharper political
focus than the status that is accorded it in Latour’s work.

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
34 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

Culture and the Social: Principles of Analysis

Let me now, in the light of these general remarks, advance five principles
through which the terms for analysing the relations between culture and the
social to which they point might be put into effect:

(i) First, we need to rule out the possibility that culture might be distinguished
from the social in terms of a special kind of cultural stuff that is distinct
from a special kind of social stuff of which the social is made up. This
makes no more sense than it does to suggest, as is the case with those ver-
sions of the cultural turn that are still in thrall to the linguistic turn, in
either its structuralist or post-structuralist guises, that culture and the
social should be merged because they are said to be made up of the same
kind of stuff (organized structures of meaning).7 To follow Latour’s con-
tention that the social cannot be designated as ‘a special domain, a special
realm, or a particular sort of thing, but only as a very peculiar move-
ment of re-association and reassembling’ (Latour, 2005: 7) entails that we
cease to look for ways of differentiating culture and the social as different
realms made up of different kinds of things (representations, say, versus
material social relations). To the degree that both (and, indeed, the econ-
omy too) are made of similar heterogeneous elements, the differences
between them have to be sought in the manner and locations in which they
are assembled. Their status as different is thus not ontological but public:
they are distinguished from one another as different public organizations
of things, texts and humans that are able to operate on and in relation to
each other through the differences that have thus been historically pro-
duced between them.
(ii) The making of culture and its differentiation from the social is, above all else,
the work of institutions. It is here that the work of making culture is enacted
and through which the work it accomplishes is performed. To analyse the
work of making culture entails attending to all those processes of accumu-
lation, classification and ordering to which varied practices are subject and
through which their ‘culturalness’ is conferred on them. It is clear, for exam-
ple, to put the point historically, that the bringing together of varied kinds
of writing to form literature or of painting to constitute art is the result of
the work of classification and codification performed by literary and artis-
tic institutions, and the forms of cultural knowledge which they marshal,
work which always involves the differentiation of the ensembles it assem-
bles from other assemblages of knowledges and practices. It is equally clear
that the ‘culturalness’ that is produced in this way is not the same as culture
in some general semiotic sense (culture as meaning-making practices) but
rather involves the production of specific relations between the objects and
practices that are thus brought together and, as a consequence, the organi-
zation of distinctive meta-semiotic properties arising from their inscription
in specific, institutionally produced zones of cultural action.

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 35

However, I do not mean to suggest that such ‘culturalness’ can be only aes-
thetic in form or that the work of making culture is limited to the institutions
of aesthetic culture. We can, indeed, witness the same process in relation to
culture understood as ‘ways of life’. For apart from themselves being assem-
bled from heterogeneous bits and pieces, ways of life can be and frequently
are also subjected to distinctive institutional processes of accumulation and
assemblage that give rise to distinctive forms of ‘culturalness’ which serve as
means for acting on and regulating the relationships between ways of life.
This is clear from the history of the relations between anthropology and
collecting institutions in which the assembly of ways of life, in varied forms,
from Franz Boas’s life groups to Henri Rivières’s ensemble ecologique and
the eco-museum, has been closely related to the development of ways of
managing populations. Boas’s life groups and the distinctive regionally
defined and differentiated cultures these made visible, for example, played a
key role in the move away from general stadial models for the administra-
tion of the native peoples of North America to culturally specific pro-
grammes of administration that took greater account of tribal differences.8
Ways of life are equally susceptible to varied forms of ethnographic or sta-
tistical collection in ways that have had significant implications for the man-
agement of populations. The history of Mass Observation provides one
example of this in setting in motion a process of gathering ‘the masses’, their
views and their habits, that opened up new ways of acting on popular opin-
ion in the context of the wartime and immediately post-war administra-
tions.9 Pierre Bourdieu’s work provides another such example, Distinction
(1984) serving as a model for ways of netting different habitus statistically
which have been connected to ways of acting on those habitus via education
and cultural policies that are calculated to equalize life opportunities.10 The
second principle, then, concerns the need to attend to the institutional mech-
anisms, and the forms of expertise that they marshal, through which varied
things, texts and humans are brought together and, through this work of col-
lection and assembly, acquire new properties derived from the operations of
the different forms of cultural expertise which bring them together.

(iii) The work that culture does as a consequence of being assembled in these
ways depends on how, once such assemblages have been stabilized into
institutionally durable forms, it is connected to the social from which it has
thus been differentiated. The key issues here concern (a) the respects in
which the forms of cultural knowledge that are involved in such processes
of separation and assembly simultaneously fashion specific properties for
the forms of culture they assemble (the designation of specific bundles of
objects, buildings, and customs as world heritage sites, for example), (b)
the forms for knowing, probing, investigating and formatting the social
that are made available via what Law (2004) calls the ‘method assem-
blages’ of the social sciences, in which distinctive epistemological positions
and practical techniques of knowing make the social knowable in ways that

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
36 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

are always-already connected to ways of acting on it, and (c) the respects
in which the relations between these cultural and social knowledges produce
distinctive ‘working surfaces on the social’, comprising both ways of for-
matting the social conceptually and distinctive technologies of intervention,
through which culture’s action on it can be modulated.
The historical processes through which Culture (in the sense of Western
high culture) was made tangible, visible and performable in new settings
(museums, art museums, concert halls, libraries, literary and scientific asso-
ciations) which assembled texts, artefacts, techniques and persons in new
relations to each other under the superintendence of new cultural knowl-
edges, has thus to be approached alongside an analysis of the development
of the social sciences and their role in laying out the social in forms which
made it amenable to corrective action via cultural means. Embryonic social
survey data thus played a key role in identifying male drunkenness as the key
point of action through which family life might be acted on and ameliorated
in late 19th-century Britain, just as it was the previous sequestration of art
and the new values produced for it in the history of post-Kantian aesthetics
that made it intelligible to propose that the social could be successfully man-
aged by building new networks for art into the social via outreach and exten-
sion programmes.11 The production of ethnographic collections of castes in
the context of the English colonial governance of India similarly played an
equally important role in the development, after 1857, of the caste system as
the ‘working surface on the social’ through which the subordinate sections
of the Indian population were to be managed (Dirks, 2001).
The current practices of world heritage similarly offer ample evidence of
the ways in which earlier cultural assemblages are disassembled as they are
broken down into their constituent heterogeneous elements by being dis-
connected from the more local assemblages that had characterized their sta-
tus as merely national or regional heritage. For it is only by being
disassembled in this way – and often quite violently, in the face of local
protest and opposition – that such ‘heritage elements’ can then be inscribed
in new networks as global actors that are thus hauled into the emerging
spheres of economic and civic action that are produced in the relations
between world heritage sites (Kirshenblatt-Gimblett, 2006).
(iv) It follows from these first three principles that particular attention needs to
be paid to the different forms of ‘culturalness’ that are produced by differ-
ent institutional assemblages if the kinds of action and interaction – of the
self on the self, of varied agents on each other – that are integral to the
‘working surfaces’ through which their work on the social is organized, are
to be adequately understood. This means that questions concerning the
agency of non-human actors – of objects and texts or, more generally, what
I shall call presentations – should be accorded an important place in analysing
the processes and mechanisms through which particular zones of action and
effectivity are organized for particular cultural assemblages. Work in the
traditions of science studies and practice studies is more helpful here than
the general formulations of actor-network theory, owing to the attention

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 37

that these traditions have paid both to the processes of fabrication through
which new entities are produced in custom-built scientific settings and to the
specific qualities that attach to these as a consequence.
The same is true of the anthropology of art owing to its focus on the ways
in which art acts as an operative mediator of social relations, and to its view
that objects acquire their artfulness by being mobilized in the context of spe-
cific social relations rather than possessing this as an immanent quality. The
potential for treating custom-built cultural settings as places where distinctive
cultural entities are produced through processes of working that are particu-
lar to them is evident in a good deal of recent work focused on museums, par-
ticularly anthropological ones (see, for example, Bouquet, 2001), and, in the
case of contemporary media, on film and television studios (Hemming, 2006).
What is less evident, however, is a widespread appreciation of the respects in
which such analyses can be taken further to examine the new forms of alliance
between human and non-human actors that such new entities make possible
and the ways in which these are harnessed to new ways of acting on the social.
The nexus of relations between community art, the institutions in which this
is produced, stored and accumulated, the forms of expertise that are involved
in its production and dissemination, the ways in which both these and com-
munity art practices are enrolled in governmental programmes aimed at, var-
iously, including excluded communities, building up community identity as
an interface between the political system and the individual, and the ways in
which such initiatives call on sociological accounts of community and the role
of social capital in its formation and maintenance: all of these are pointers to
the form that such analysis might take.
(v) Finally, there is no question in all of this of looking to develop an account of
culture as an anthropological constant that operates in the same way in all
kinds of societies. The remit of the programme outlined above is limited to
the forms of cultural assemblage that are associated with the development of
secular forms of cultural knowledge, the institutions in which these are set to
work, and the ways in which their operations – viewed in the light of the par-
allel development of the social and economic sciences – are related to the his-
torical separation of culture, the social and the economy as different public
organizations of people and things. This is not to suggest a distinction of a
fundamental kind between modern and pre-modern knowledge formations
and the manner of their functioning. Similar principles of analysis can be
used to study the make-up and operation of other knowledge systems and
their implications for regulating social conduct. David Turnbull has thus
interpreted medieval cathedrals as knowledge spaces which, like laborato-
ries, brought together specific resources, skills and labour in operating as
‘powerful loci of social transformation’ (Turnbull, 2000: 67) and he makes
the same case for the knowledge assemblies of indigenous peoples. The dis-
tinctiveness of the ways of interrogating the relations between culture and
the social that I am proposing thus consists in the focus on the operations of,
and the interactions between, historically new forms of cultural and social
knowledge in the context of the public differentiation of culture, the social

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
38 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

and the economy that is both their outcome and – so long as these differen-
tiations remain durable – their condition.

The Limits of ‘Cultural Sociology’

I have suggested elsewhere (Bennett, 2007) that ‘culture studies’ might serve as
a useful banner for this programme of work. This is partly because its focus
echoes that of science studies in seeking to effect a series of transformations in
relation to the sociology of culture similar to those which science studies has
carried through in relation to the sociology of science. If I am hesitant about
‘cultural sociology’ as a broader rubric in which to situate these concerns, this
is partly because many of the formulations of cultural sociology are implicated
in the organization of the ‘working surfaces of the social’ that analyses of the
kind I am proposing would need to take as their object. The role played by the
concepts of social and cultural capital as well as by new, non-essentialist con-
ceptions of ethnicity in reformatting the social in ways that have had significant
consequences for the organization of the working surfaces on the social that
contemporary cultural policies and the practices of cultural institutions now
work with and through, is a case in point. For, in all these cases, what we con-
front are ways of reformulating the social theoretically by injecting relational
values and meanings into its very sinews with the consequence that it is these –
an always-already ‘culturalized social’ – that provide the surfaces through
which the social can be connected with and acted on.
However, my hesitancy about the currency of ‘cultural sociology’ is also
partly prompted by the consideration that when cultural sociologists have
engaged with the traditions of science studies and actor-network theory, the
results have often illuminated very little except a determination to hold on to
established theoretical positions, or to carve out seemingly new ones, at the
price of a lamentable failure to accord those traditions proper theoretical atten-
tion. In staking out his claim for a strong programme in cultural sociology, for
example, Jeffrey Alexander annexes science studies to a general statement about
the need to secure culture’s autonomy in relation to the social even though
the form in which he conceives that autonomy is clearly at odds with both the
theoretical formulations and methodological procedures of science studies.12
Bourdieu’s discussion of Bruno Latour’s work and the tradition of laboratory
studies more generally is equally dispiriting in this regard, although for the oppo-
site reason. Clearly unaware of the distinction between science studies and the
sociology of science, he equates Latour with the latter, and entirely misreads his
account of the role of inscriptions (that is, things translated into signs) as particu-
lar kinds of ‘immutable mobiles’ (capable of travelling yet remaining the same) by
interpreting this as a semiological vision of the world in which reality appears as
the adventitious fabrication (in the sense of invention) of signs (Bourdieu, 2004:
8–9, 21–9). He then proceeds to insist on the need for an objective sociology of
science in the form of field analysis to account for the processes through which the
autonomy of the scientific field is organized without paying any heed whatsoever

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 39

to the different ways in which science studies accounts for the production of
autonomous scientific entities and their differentiation from the social.
These misunderstandings are all the more surprising to the degree that Latour
has, on a number of occasions, clearly indicated that his conception of the forms
of displacement to which science studies subjects the sociology of science is mod-
elled on certain traditions of cultural analysis which he considers to have been
ahead of the game in this respect (see, for example, Latour 1998). This is especially
true of one of his more extended discussions of the concept of immutable mobiles
which he illustrates with reference to Svetlana Alpers’s account of the production
of the ‘distance point’ method of Dutch landscape painting, Elizabeth Eisenstein’s
account of the new mobility of inscriptions associated with the development of the
printing press, and Johannes Fabian’s account of anthropology’s visual mapping
of the relations between moderns and primitives (Latour, 1990).
Latour is attracted just as much to what these accounts don’t do as he is to
what they do. What they don’t do, he argues, is to try to account for the phe-
nomena they are concerned with by referring them to some already socially
determined set of interests or structure and read them, in however indirect or
mediated a way, as an expression of these. These are not, then, sociologies of
culture. But they are not cultural sociology either, concerned in some vague way
with the dissemination of meanings and values or compressing culture and the
social into an undifferentiated whole. Rather, what Latour values about them is
their concern with craftsmanship: that is, with particular ways of doing and
making, conducted in particular settings (the artist’s studio, the anthropological
field site), through which new immutable mobiles are produced, mobilized, and
subjected to forms of inscription which – combined with or conscripting other
mobiles and inscriptions – open up new ways of acting on the social. His focus,
that is to say, is on the work of culture in the two senses that I have defined it:
on the work through which new cultural entities are produced, and on the work
that these then do in being mobilized to act on the social through the working
surfaces that are produced by the intersections between particular social and cul-
tural knowledges.

Accumulating Culture,Working on the Social

I want, however briefly, to offer a worked example of the principles for analysing
the relations between culture and the social that I have advanced to this point.
I shall do so by taking my bearings from Pandora’s Hope in which Latour notes
the significance, in the case of many sciences, of the relations between expedi-
tions and museums to the processes through which ‘nonhumans are progres-
sively loaded into discourse’, citing the galleries of the Museum d’Histoire
Naturelle and the collections of the Musée de l’Homme as places in which ‘all
the objects of the world are thus mobilised and assembled’ (Latour, 1999:
99–101). The relations between Baldwin Spencer’s and Frank Gillen’s various
collaborative expeditions to the desert regions of Central Australia and the col-
lections of Aboriginal materials that Spencer was thus able to amass at the

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
40 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

National Museum of Victoria (which he directed) offer a good means of exploring

the implications of this perspective.13
Three aspects of these expeditions are worthy of note. First, they formed
part of a key moment in the development of the fieldwork phase in anthro-
pology which significantly reorganized the relations between a range of agents
compared to those that had characterized the earlier phase of anthropology. In
the latter, collections had typically been accumulated via networks of commer-
cial or amateur collectors either plying their wares independently or acting on
instructions, with museums in colonial contexts acting as intermediaries for the
transit of objects from local sites of collection to metropolitan centres for their
interpretation and synthesis by ‘armchair anthropologists’. Fieldwork anthro-
pology substituted a new set of relations which, as Howard Morphy puts it,
combined ‘theorist and ethnographer in the same person’ (Morphy, 1996: 138).
Second, this meant that the distance travelled by the immutable mobiles Spencer
collected was considerably shortened: in the main, from Central Australia to
Melbourne rather than from Central Australia to London via Melbourne.
While continuing to trade duplicate specimens with European and American
museums, Spencer was a leading figure in staunching the free-flow of
Aboriginal objects to Europe and America by supporting the introduction of
export licensing requirements, and he contributed considerably to their local
accumulation by donating his own collections to the National Museum of
Victoria whose ethnographic holdings increased from 1200 to over 36,000
items during his period as director (Mulvaney and Calaby, 1985: 252).
Spencer was thus able to install the immutable mobiles he accumulated at
the National Museum of Victoria in two sets of relations at once. So far as the
international scientific community was concerned, they provided the material
basis and warrant for the inscriptions through which, in journal articles and
books, the various kinds of data that he and Gillen had accumulated were
recombined in accounts of Aboriginal culture that had considerable influence
on the development of European anthropology and, via Durkheim, sociology.14
However, the in situ presence of the collections in his museum also gave him a
large number of, in Latour’s terms, ‘well aligned and faithful allies’ that he
could ‘muster on the spot’ (Latour, 1990: 23) in the context of specifically
national political and administrative fields.
Third, if all of these conditions made it possible for Spencer to become a
political actor on the international and national stages, it is because they made
it possible for him to make and to mobilize a specific production of Aboriginal
culture. By bringing together artefacts, photographs, films, and sound record-
ings from diverse locations, combining these in new ways, simplifying and con-
densing them by subjecting them to further processes of inscription, Spencer
produced something that had not existed before: Aboriginal culture, not as a set
of autochthonous realities that preceded his inquiries, but as a new surface of
connection between white and black Australia, a surface that organized new
sets of governmental and administrative interfaces through which the former
might act on the latter and which, in turn, made Aboriginality performable in
new ways.

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 41

His activities were, of course, by no means unique in this regard. However,

Spencer’s role was a particularly significant one in contributing to the develop-
ment of what Tim Rowse has called ‘the Aboriginal domain’: that is, the
domain of institutions, practices, customs, languages, etc., which, by organiz-
ing a zone of interaction between them, provided the means through which dif-
ferent Aboriginal communities were absorbed into the colonial state (Rowse,
1992). Rowse identifies the development of rationing to compensate for indige-
nous food supplies displaced by cattle grazing as a key moment in the develop-
ment of this domain (Rowse, 1998). Introduced in 1894, rationing was the
basis for the development of a new mode of government that regularized
(unequal) exchanges across the colonial frontier and, in place of racial violence,
substituted an administrative infrastructure for managing these exchanges.
Spencer was to become directly involved in the administration of ‘the
Aboriginal domain’ through his role as Special Commissioner for Aboriginals,
Northern Territory. What concerns me here, however, are the more general aspects
of the influence he acquired through the ways in which the Aboriginal culture that
he produced contributed to the organization of the exchanges that took place
between the white and black in the Aboriginal domain of the early 20th century.15
There are three main issues here. The first concerns his racialization of Aboriginality
in the role he accords race, rooted rigorously in bloodline and skin colour, to
account for the shared substratum of Aboriginality that gives Aboriginal culture its
underlying unity in spite of the differences in customs and practices that character-
ize its variants among different tribes. The second has to do with the dual tempo-
rality that is attributed to Aboriginal culture as a result of the contradictory way in
which he inserts it into evolutionary time. For, on the one hand, it is a culture that
lacks an inner temporality having been – in Spencer’s view – more-or-less static for
millennia. This is the consequence of its grounding in a race which, lacking any com-
petition owing to Australia’s isolation from other races and from aggressive fauna,
had allegedly failed to evolve. On the other hand, once confronted with such com-
petition as a result of European occupation, the Aborigine, Spencer argues, faces
annihilation. Aborigines thus enter into evolutionary time for only a split second, at
the very moment they are destined to leave it by being driven to extinction. Third,
however, Spencer provides for a different temporality at the level of the individual.
While the laws of racial competition destine ‘full-blood’ Aborigines to extinction,
managed miscegenation provides a route through which half-castes and their off-
spring might survive, but only on the condition that they cease to be Aboriginal in
the sense that is, for Spencer, defining and constitutive: that is, racially.
Spencer was not, of course, unique in holding such views. The importance
of his work consisted rather in the manner in which he assembled artefacts,
photographs, and recordings from diverse locations into a production of
Aboriginal culture as racially grounded that had scientific authority.16 The key
role that he accorded the tribe as a major unit of analysis and the manner in
which he construed its relations to race are especially important in this regard
and, as Kevin Blackburn (2002) notes, a significant departure from earlier tra-
ditions in Australian anthropology. These were typically concerned to identify
the existence of different nations at the supra-tribal level: not, though, nations

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
42 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

as sovereign political entities but nations in the biblical sense of cultural groups
sharing a common descent. Spencer’s erasure of this level of analysis deprived
such supra-tribal relations of any political or administrative significance. In
their place, he substitued a racial conception of Aboriginality as the common
factor which cohered the customs and practices of different tribes into a single
(albeit tribally differentiated) whole and, thereby, produced the race as such,
even if through the mediation of different tribes, as the surface that government
was to act upon.
Patrick Wolfe (1991) has argued that it was this racialized production of
the Aborigine as an absolute and undifferentiated other that underpinned the
development of new strategies of assimilation in the context of the early-20th-
century development of the Aboriginal domain. The polar opposition it con-
structed meant, Wolfe argues, that everything that was only ‘part-aborigine’
had to be counted as ‘non-aborigine’, thus opening up a space in which ‘half-
castes’ could be plucked off from the ‘full-blood’ and absorbed into white soci-
ety through a programme of epidermal-cum-cultural transformation that would
eventually prepare them for citizenship. These distinctions provided the crucial
terms in which the relations between the Aboriginal domain and the distribu-
tion of citizenship rights and duties were managed in the context of the inde-
pendent polity that was developed in the first two decades following the 1901
Federation of Australia (Clarke and Galligan, 1995). However, it was not until
the inter-war period of the 20th-century that the biological incorporation of
‘half-castes’ into the wider population was, from being regarded as a more-or-
less natural process, adopted as a strategy requiring coordinated state interven-
tion, most notably in Western Australia and the Northern Territory.
Building on the developmental gap that had been opened up between the
‘half-caste’ and the ‘full-blood’ through casual miscegenation, by promoting a
combination of ‘interbreeding between White and part-Aboriginal Australians,
and the curtailment of unions between full- and part-Aborigines’ (McGregor,
2002: 288), such policies instituted programmes of managed miscegenation that
aimed to take the Aborigine through a series of way-stations that were defined
in terms of both skin colour and the modes of segregated abode that Spencer
came to favour in the Northern Territory. The intelligibility of these pro-
grammes was the result of the new presentations of Australia’s indigenous peo-
ples as a racial unity that rolled out from the ways in which they and their
artefacts and rituals had been assembled to form a ‘working surface on the
social’ through which such peoples could be acted on by means of civilizing
programmes which formatted the social in terms of essentialist racial divisions
inscribed in contradictory relations to the vectors of evolutionary time.

Conclusion

One of the more memorable of Frederic Jameson’s many injunctions, and the guid-
ing principle of his The Political Unconscious (Jameson, 1981), is that we should,

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 43

as social and cultural analysts, always aim to historicize. Although, for reasons
that I won’t go into here,17 Jameson was unable to put this methodological maxim
into effect except abstractly, the maxim itself has always seemed to me an impec-
cable one. The key questions on which its implementation turns concern the level
of analysis at which it is to be put into effect and, consequently, what it is that
most needs to be historicized. My contentions in this article propose two answers
to these questions. The first, in line with my reading of the implications of gov-
ernmentality theory, is that our understandings of both culture and the social
need to be radically historicized if we are to produce an adequate basis for
understanding the specific contemporary forms of their interrelations.
The programme that I have proposed would, in this respect, find its place
as part of a historical sociology (in the end, the only kind worth having or aim-
ing for) that would seek to offer a more detailed and nuanced understanding of
historically varied orderings of the relations between culture, the social and the
economy – and of the mechanisms producing and sustaining their separation as
distinctive public organizations of the relations between people and things –
than are available from the generalized accounts of modernity and postmoder-
nity. The second contention is that, to do this, it is also necessary, now, to his-
toricize the doxological formulations of those tendencies, strongly associated
with the rubrics of cultural sociology, to merge culture and the social into one
another. It is, I have suggested, only by looking to those heterodox tendencies
emerging both from within sociology and outside it that it will prove possible
to unclasp culture and the social in ways that will make clear both the work
that goes into the making of the former and the work that it performs in rela-
tion to a social that is not simply its doppelgänger.

Notes

1 This is Göran Therborn’s (1976) contention in his influential account of classi-

cal sociology’s differentiation from Marxism in terms of its focus on the role of
cultural and ideational factors – what he called ‘the imagined community’ – in
contradistinction to Marxist materialism.
2 I borrow the notion of ‘formatting’ here from Timothy Mitchell’s (2002)
account of the relations between theories and techniques of economic regula-
tion and the economy as preferable to more conventional accounts of the econ-
omy and the social as discursive constructs.
3 The sketch for such a programme that is offered here is deliberately somewhat
abstract and schematic, even manifesto-like in some places. However, the main
planks of the argument are ones I have discussed at greater length elsewhere
(see Bennett, 2005, 2007).
4 See Albertsen and Diken (2004) for an excellent discussion of these tendencies.
5 The issues here are complex, particularly when Bourdieu’s interpretation of
field theory is viewed in conjunction with his concept of habitus and the role
that the relations between these play in his concern to overcome the bifurcation
of structure and agency by positing, in the notion of field, a relational structure

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
44 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

that is operated through the cultural unconscious of the habitus. While the
illusio of those who commit to the stakes of the game which define any partic-
ular field is an essential aspect of a field’s constitution, so that the field cannot
be defined as a pre-given structure, only the objectifying and distancing proce-
dures of the sociologist can bring to light the relations between different capi-
tals that are in play in different fields and which, in the social field as a whole,
organize the relations between different fields. I draw here on John Levi Martin’s
(2003) discussion of field theory.
6 In one of his clearest definitions of the dispositif Foucault states that the dis-
positif is both a structure of heterogeneous elements and a ‘certain kind of gen-
esis’, but he is vague as to what the latter is and does not provide any specific
tools for its analysis. See ‘The confession of the flesh’ in Foucault (1980).
7 I have discussed this in more detail elsewhere (see Bennett, 2002).
8 I draw here principally on the account in Hinsley (1981).
9 See New Formations 44, a special issue on Mass Observation, and, on this
point in particular, Hubble (2001).
10 There are, however, significant differences in the kind of policy adjustments to
the relations between habitus that Bourdieu envisaged at different points in his
intellectual career. See Ahearne (2004) for a useful discussion of these.
11 See, for more detailed elaborations of this argument, Bennett (1995, 2000).
12 I refer here to the discussion in the first chapter of Alexander (2003). I have
developed my criticisms of Alexander’s position here at greater length in
Bennett (2007).
13 Spencer, a product of Mancunian liberalism, was originally trained as a natural
historian and was an early supporter of Darwinism. His substantive position at
Melbourne was as professor of biology but his interests progressively shifted to
anthropology. Spencer met Gillen, a postmaster at Alice Springs, during the 1894
Horn expedition to Central Australia. They subsequently developed a lifelong
collaboration in which the distinction between the roles of field collector and
university-based interpreter, while remaining intact, was also significantly blurred.
14 The role of Spencer’s work on the Arunta in shaping Durkheim’s account of the
elementary forms of religious life has been subject to extensive commentary and
is discussed by both Wolfe (1991) and Morphy (1996).
15 I have discussed aspects of these elsewhere: see Bennett (2004: 154–8).
16 There are many texts here but for an economical and representative example of
the ways in which he assembled Aborigines from different tribes, their tools,
weapons and rituals into a racially defined common Aboriginality, see Spencer
(1914).
17 I have discussed this aspect of Jameson’s work elsewhere (see Bennett, 1990:
205–17).

References

Ahearne, Jeremy (2004) Between Cultural Theory and Policy: The Cultural Policy
Thinking of Pierre Bourdieu, Michel de Certeau and Régis Debray. Coventry:
Centre for Cultural Policy Studies, University of Warwick, Research Paper, no. 7.
Albertsen, Niels and Diken, Bülent (2004) ‘“Artworks” Networks: Field, System or
Mediators?’, Theory, Culture & Society 21(3): 35–58.

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 45

Alexander, Jeffrey C. (2003) The Meanings of Social Life: A Cultural Sociology.

Oxford: Oxford University Press.
Bennett, Tony (1990) Outside Literature. London and New York: Routledge.
Bennett, Tony (1995) ‘The Multiplication of Culture’s Utility’, Critical Inquiry
21(4): 859–89.
Bennett, Tony (2000) ‘Acting on the Social: Art, Culture, and Government’,
American Behavioural Science 43(9): 1412–28.
Bennett, Tony (2002) ‘Culture and Governmentality’, in J. Bratich, J. Packer and
C. McCarthy (eds) Foucault, Cultural Studies and Governmentality, pp. 47–66.
New York: SUNY Press.
Bennett, Tony (2004) Pasts Beyond Memory: Evolution, Museums, Colonialism.
London and New York: Routledge.
Bennett, Tony (2005) ‘Civic Laboratories: Museums, Cultural Objecthood, and the
Governance of the Social’, Cultural Studies 19(5): 521–47.
Bennett, Tony (2007) ‘Making Culture, Changing Society: The Perspective of
Culture Studies’, Cultural Studies 21(4/5): forthcoming.
Blackburn, Kevin (2002) ‘Mapping Aboriginal Nations: The “Nation” Concept of
Late Nineteenth Century Anthropologists in Australia’, Aboriginal History 26:
131–58.
Bouquet, Mary (ed.) (2001) Academic Anthropology and the Museum: Back to the
Future. New York and Oxford: Berghahn Books.
Bourdieu, Pierre (1984) Distinction: A Social Critique of the Judgement of Taste.
London: Routledge and Kegan Paul.
Bourdieu, Pierre (2004) Science of Science and Reflexivity. Chicago, IL: University
of Chicago Press.
Clarke, Tom and Galligan, Brian (1995) ‘Aboriginal Native and the Institutional
Construction of the Australian 1901–48’, Australian Historical Studies 26(105):
523–43.
Dirks, Nicholas B. (2001) Castes of Mind: Colonialism and the Making of Modern
India. Princeton, NJ: Princeton University Press.
Foucault, Michel (1980) Power/Knowledge: Selected Interviews and other Writings,
1972–1977 (Colin Gordon ed.). New York: Pantheon Books.
Foucault, Michel (1991) ‘Governmentaity’, in Graham Burchell, Colin Gordon and
Peter Miller (eds) The Foucault Effect: Studies in Governmentality. London:
Harvester Wheatsheaf.
Gell, Alfred (1998) Art and Agency: An Anthropological Theory. Oxford: Clarendon
Press.
Gomart, Emilie and Hennion, Antoine (1999) ‘A Sociology of Attachment: Music,
Amateurs, Drug Users’, in John Law and John Hazard (eds) Actor Network
Theory and After. Oxford: Blackwells/The Sociological Review.
Hemming, Emma (2006) ‘Into the Newsroom: Applying Actor Network Theory to an
Investigation into the Way in Which Constellations of Humans and Technologies
Construct News Facts’, conference disk, Media Change and Social Theory,
CRESC (ESRC Centre for Research on Socio-cultural Change).
Hennion, Antione (1997) ‘Baroque and Rock: Music, Mediators and Musical
Taste’, Poetics 24: 415–35.
Hennion, Antione, Teil, Geneviève and Vergnaud, Frédéric (2005) ‘Questions of Taste’,
in Bruno Latour and Peter Weibel (eds) Making Things Public: Atmospheres of
Democracy, pp. 670–9. Germany and Cambridge, MA: ZKM and MIT Press.

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
46 Cultural Sociology Volume 1 ■ Number 1 ■ March 2007

Hinsley, Curtis M. Jr. (1981) Savages and Scientists: The Smithsonian Institution
and the Development of American Anthropology, 1846–1910. Washington,
DC: Smithsonian Institution Press.
Hubble, N. (2001) ‘Charles Madge and Mass-Observation are at Home: From
Anthropology to War, and After’, New Formations 44: 76–89.
Jameson, Frederic (1981) The Political Unconscious. London: Methuen.
Kirshenblatt-Gimblett, Barbara (2006) ‘World Heritage and Cultural Economics’,
in Ivan Karp, Corinne A. Kratz, Lynn Szwaja and Tomas Ybarra-Frausto (eds)
Museum Frictions: Public Cultures, Global Transformations. Durham, NC:
Duke University Press.
Lahire, Bernard (2001) ‘De la théorie de l’habitus à une sociologie psychologique’,
in Bernard Lahire (ed.) Le travail sociologique de Pierre Bourdieu: Dettes et
critique, pp. 121–52. Paris: Éditions La Découverte.
Lahire, Bernard (2003) ‘From the Habitus to an Individual Heritage of Dispositions.
Towards a Sociology at the Level of the Individual’, Poetics 31: 329–55.
Lahire, Bernard (2004) La culture des individus: dissonances culturelles et distinc-
tions de soi. Paris: Éditions la découverte.
Latour, Bruno (1990) ‘Drawing Things Together’, in Michael Lynch and Steve
Woolgar (eds) Representation in Scientific Practice. Cambridge, MA: MIT Press.
Latour, Bruno (1998) ‘How to be Iconophilic in Art, Science, and Religion?’, in
Caroline A. Jones and Peter Galison (eds) Picturing Science, Producing Art.
New York and London: Routledge.
Latour, Bruno (1999) Pandora’s Hope: Essays on the Reality of Science Studies.
Cambridge, MA: Harvard University Press.
Latour, Bruno (2005) Reassembling the Social: An Introduction to Actor-Network-
Theory. Oxford: Oxford University Press.
Law, John (2004) After Method: Mess in Social Science Research. London and
New York: Routledge.
McGregor, Russell (2002) ‘“Breed Out the Colour”, or the Importance of Being
White’, Australian Historical Studies 120: 286–302.
Martin, John Levi (2003) ‘What is Field Theory?’, American Journal of Sociology
109(1): 1–49.
Mitchell, Timothy (2002) Rule of Experts: Egypt, Techno-Politics, Modernity.
Berkeley, CA: University of California Press.
Morphy, Howard (1996) ‘More Than Mere Facts: Repositioning Spencer and Gillen
in the History of Anthropology’, in S.R. Morton and D.J. Mulvaney (eds)
Exploring Central Australia: Society, Environment and the 1894 Expedition.
Chipping Nortin, NSW: Surrey Beatty.
Mulvaney, D.J. and Calaby, J.H. (1985) ‘So Much That is New’: Baldwin Spencer,
1860–1929: A Biography. Melbourne: Melbourne University Press.
Rowse, Tim (1992) Remote Possibilities: The Aboriginal Domain and the
Administrative Imagination. Casuarina (Darwin): North Australian Research
Unit, Australian National University.
Rowse, Tim (1998) White Flour, White Power: From Rations to Citizenship in
Central Australia. Cambridge: Cambridge University Press.
Spencer, Baldwin (1914) ‘The Aboriginals of Australia’, in G.H. Knibbs (ed.) Federal
Handbook prepared in connection with the eighty-fourth meeting of The
British Association for the Advancement of Science held in Australia, August
1914. Melbourne: Commonwealth of Australia.

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.
 The Work of Culture Bennett 47

Therborn, G. (1976) Science, Class and Society: On the Formation of Sociology and
Historical Materialism. London: New Left Books.
Turnbull, David (2000) Masons, Tricksters and Cartographers: Comparative
Studies in the Sociology of Scientific and Indigenous Knowledge. London:
Harwood Academic Publishers.
Wolfe, Patrick (1991) ‘On Being Woken Up: The Dreamtime in Anthropology and
in Australian Settler Culture’, Comparative Studies in Society and History
33(2): 197–224.

Tony Bennett

Tony Bennett is Professor of Sociology at the Open University, where he is also a Director
of the ESRC Centre for Research on Socio-cultural Change, and Professorial Fellow in the
Department of English with Cultural Studies at the University of Melbourne.
Address: Sociology Discipline, Faculty of Social Sciences, The Open University,
Walton Hall, Milton Keynes MK7 6AA, UK.
E-mail:T.Bennett@open.ac.uk

Downloaded from http://cus.sagepub.com by Marta Cabrera on July 16, 2007

© 2007 SAGE Publications and the British Sociological Association. All rights reserved. Not for commercial use or
unauthorized distribution.