tinuous genesis of discontinuity. When we retrospectively project the
concept of artist before the 1880s, we commit absolutely fantastic
anachronisms: we overlook the genesis, not of the character of the artist
or the writer, but of the space in which this character can exist
as such.
And the same is true of politics. We take the risk of formidable historical
fallacies when we fail, as do some historians who, today, take a
fancy to "political philosophy," to pose the question of the social
genesis of the political field (Bourdieu 1981a) and of the very notions
that political philosophy eternalizes by treating them as transhistorical
essences. What I just said about the words "art" and "artist" would
apply to notions such as "democracy" and "public opinion" (see Bourdieu
1979e, Bourdieu and Champagne 1989, Champagne 1990). Paradoxically,
historians often condemn themselves to anachronism
because of their ahistorical, or dehistoricized, usage of the concepts
they employ to think the societies of the past. They forget that these
con cepts and the reality they capture are themselves the product of a
historical construction: the very history to which they apply these
concepts has in fact invented, created them, oftentimes at the cost of
an immense— and largely forgotten—historical work.41
3 The Logit of Fields
The notion of field is, together w ith those of habitus and capital, the central organizing
concept o f your w ork, which includes studies o f the fields o f artists and
intellectuals, class lifestyles, Gnndes etoles, science, religion, the field of pow er,
of law , of housing construction, and so o n /2 You use the notion o f field in a highly
41. This fruitful tension between history and sociology encouraged by Bourdieu is
particularly well illustrated by the historical research of his colleagues and collaborators
Christophe Charle (1987,1990,1991), Dario Gamboni (1989), Alain Viala (1985) and Victor
Karady, who has undertaken an ambitious long-term project in the historical sociology
of Hungary and other Eastern European countries (see Karady 1985, Don and
Karady 1989, Karady and Mitter 1990). On the question of historical discontinuity and
the temporal rootedness of conceptual categories or episternes, there are many parallels
between Bourdieu and Foucault, some of which can be traced directly back to their
common training in the history of science and medicine under Canguilhem (Bourdieu
1988e: 779). The major differences are rooted in Bourdieu's historicizing of reason via
the notion of field.
42. On the intellectual and artistic field, see Bourdieu 1971a, 1975b, 1975c, 1983a,
1983d, 1988a; on the space of classes and classlifestyles, Bourdieu 1978b, 1984a, 1987b; on
9 4 I P ie rre B ourdieu an d Loic J. D . W acq u an l
The Purpose o f R eflexive Sociology (The Chicago W orkshop) I 95
technical and precise sense th at is perhaps partly hidden behind its commonsense
m eaning. Could you explicate w here the notion comes from (for Americans, it is
likely to evoke th e "fie ld theory" o f Kurt Lewin) and w hat its m eaning and theoretical
purposes are?
I do not like professorial definitions much, so let me begin with a brief
aside on their usage. I could refer here to L e m etier de sociolog u e (Bourdieu,
Chamboredon, and Passeron 1973), which is a didactic, almost
scholastic, book,43but a book which nevertheless contains a number
of theoretical and methododological principles that would make
people understand that many of the gaps or shortcomings for which I
am sometimes reproached are in fact conscious refusals and deliberate
choices. For instance, the use of .open con c ep t s4* is a way of rejecting
cultural goods, Bourdieu 1980h, 1985d, and Bourdieu and Delsaut 1975; on the religious
field, Bourdieu 1971b, 1987h, Bourdieu and de Saint Martin 1982; on the scientific field
Bourdieu 1981d, 1987e, 1990e; on the juridical field and the field of power, Bourdieu
1981a, 1986c, 1987g, 1989a, and Bourdieu and de Saint Martin 1978,1982, 1987; the field
of private housing construction is explored in Bourdieu et al. 1987 and in the articles
that make up the March 1990 issue of A ctes de la recherche en sciences sociales.
Others studies of fields conducted at the Center for European Sociology include,
in ter alia, the fields of comic books (Boltanski 1975) and of children's book publishing
(Chamboredon and Fabiani 1977), the field of the French university and intellectuals at
the turn of the century (Charle 1983 and 1990, Karady 1983, Fabiani 1989), the field of
power under the Third Republic (Charle 1987), and the fields of religion (Grignon 1977),
the arts and sciences in the classical age (Heinich 1987), seventeenth-century literature
(Viala 1985), the management of the "elderly" (Lenoir 1978), peasant trade-unionism
(Maresca 1983), social work (Verdes-Leroux 1976,1978), political representation (Champagne
1988, 1990), and feminist studies in France (Lagrave 1990).
43. This book (whose translation was for years blocked for obscure copyright reasons
and has just been published by Walter de Gruyter) is essential to an understanding
of Bourdieu's sociological epistemology. It consists of a dense exposition of the
foundational principles of "applied rationalism" in the social sciences, and of a selection
of texts (by historians and philosophers of science, Marx, Durkheim, Weber,
Mauss, and other sociologists) that illustrate key arguments. Each comprises three
parts which theorize the three stages that Bourdieu, following French epistemologist
Gaston Bachelard, considers central to the production of sociological knowledge and
that he encapsulates in the following formula: "Facts are conquered [through rupture
with common sense], constructed, confirmed ( lesfaits son t con qu is, con stru its, con states)’’
(Bourdieu, Chamboredon, and Passeron 1973:24). A worthwhile critical introduction to
Bachelard's philosophy can be found in Tiles 1984; see MacAllester 1991 for a selection
of texts.
44. For examples of criticisms of Bourdieu for the lack of closure or rigor of his concepts,
see DiMaggio 1979:1467, Swartz 1981:346-48, Lamont and Larreau 1988:155-58.
9 6 I R e tie B ourdieu and Lotc J. D . W acquant
positivism— but this is a ready-made phrase. It is, to be more precise, a
permanent reminder that concepts have no definition other than systemic
ones, and are designed to be p u t to w o rk em pirically in sy stem atic
fashio n . Such notions as habitus, field, and capital can be defined, but
only within the theoretical system they constitute, not in isolation.45
This also answers another question that is often put to me in the
United States: why do I not propose any "laws of the middle range"? I
think that this would first of all be a way of satisfying a positivistic
expectation, of the kind represented in earlier times by a book by Berelson
and Steiner (1964) which was a compilation of small, partial
laws established by the social sciences. This kind of positivistic gratification
is something that science must deny itself. Science admits
only systems of laws (Duhem showed this long ago for physics, and
Quine has since developed this fundamental idea).46And what is true
of concepts is true of relations, which acquire their meaning only
within a system of relations. Similarly, if I make extensive use of correspondence
analysis, in preference to multivariate regression for instance,
it is because correspondance analysis is a relational technique
of data analysis whose philosophy corresponds exactly to what, in my
view, the reality of the social world is. It is a technique which "thinks"
in terms of relation, as I try to do precisely with the notion of field.47
To think in terms of field isto thin k relation ally .48 The relational
45. The distinction between relational or "systemic concepts" (rooted in a theoretical
problematics of the object) and "operational concepts," defined in terms of the pragmatic
requirements and constraints of empirical measurement, is elaborated in
Bourdieu, Chamboredon, and Passeron 1973: 53-54.
46. The now famous "Duhem-Quine hypothesis" states that science is a complex
network that faces the test of empirical experience as a whole: evidence impinges not
orfany particular proposition or concept but on the entire net they form.
47. The technique of correspondence analysis is a variant of factor analysis developed
by the school of "French Data Analysis" (J. P. Benzecri, Rouanet, Tabard, Lebart,
Cibois), which has elaborated tools for a relational use of statistics that are increasingly
being employed by social scientists in France, the Netherlands, and Japan in particular.
Two useful and accessible presentations in English are Greenacre 1984 and Lebart et al.
1984; correspondence analysis has recently been included on standard computer packages
by SAS and BMDP.
48. Bourdieu (1982a: 41-42, my translation) explains: "To think in terms of field demands
a conversion of the whole ordinary vision of the social world which fastens only
on visible things: the individual, this en s realissim u m to which we are attached by a sort
of primordial ideological interest; the group, which is only in appearance defined solely
by the temporary or durable relations, formal or informal, between its members; and
The Puipose o f R eflexive Sociology (The Chicago W orkshop) I 9 7
(rather than more narrowly “structuralist") mode of thinking is, as
Cassirer (1923) demonstrated in Su bstan z beg riff u n d Fu n ktion sbeg riff,
tTie hallmark of modem science, and one could show that it lies behind
scientific enterprises apparently as different as those of the Russian
formalist Tynianov,49 of the social psychologist Kurt Lewin, of
Norbert Elias, and of the pioneers of structuralism in anthropology,
linguistics and history, from Sapir and Jakobson to Dumezil and LeviStrauss.
(If you check, you will find that both Lewin and Elias draw
explicitly on Cassirer, as I do, to move beyond the Aristotelian substantialism
that spontaneously impregnates social thinking.) I could
twist Hegel's famous formula and say that the real is the relation al: what
exist in the social world are relations—not interactions between
agents or intersubjective ties between individuals, but objective relations
which exist "independently of individual consciousness and
will," as Marx said.
In analytic terms, a field may be defined as a network, or a configuration,
of objective relations between positions. These positions are
objectively defined, in their existence and in the determinations they
impose upon their occupants, agents or institutions, by their present
and potential situation (situ s) in the structure of the distribution of
species of power (or capital) whose possession commands access to
the specific profits that are at stake in the field, as well as by their objective
relation to other positions (domination, subordination, homology,
etc.).
In highly differentiated societies, the social cosmos is made up of a
number of such relatively autonomous social microcosms, i.e., spaces
of objective relations that are the site of a logic and a necessity that are
sp ecific an d irredu cible to those that regulate other fields. For instance,
the artistic field, or the religious field, or the economic field all follow
specific logics: while the artistic field has constituted itself by rejecting
even relations understood as in teraction s, that is, as intersubjective, actually activated
connections. In fact, just as the Newtonian theory of gravitation could only be constructed
against Cartesian realism which wanted to recognize no mode of action other
than collision, direct contact, the notion of field presupposes a break with the realist
representation which leads us to reduce the effect of the en v iron m en t to the effect of
direct action as actualized during an interaction."
49. Jurii Tynianov (1894-1943) was, with Roman Jakobson and Vladimir Propp, a
leading member of the Russian Formalist school which advocated a structuralist approach
to the study of literature and language.
9 8 I P ierre B ourdieu an d LoTc J. D . W acquant
or reversing the law of material profit (Bourdieu 1983d), the economic
field has emerged, historically, through the creation of a universe
within which, as we commonly say, "business is business," where the
enchanted relations of friendship and love are in principle excluded.
You often use the analogy of a "g am e" to give a first intuitive grasp o f w hat you
understand by field .
We can indeed, with caution, compare a field to a game (jeu ) alt
though, unlike the latter, afield is not the product of a deliberate act
of creation, and it follows rules or, better, regularities,50that are not
explicit and codified. Thus we have stakes (en jeu x ) which are for the
most part the product of the competition between players. We have
an in v estm en t in the g am e, illu sio (from lu du s, the game): players are
taken in by the game, they oppose one another, sometimes with ferocity,
only to the extent that they concur in their belief (dox a) in the
game and its stakes; they grant these a recognition that escapes questioning.
Players agree, by the mere fact of playing, and not by way
of a "contract," that the game is worth playing, that it is "worth the
candle," and this collu sion is the very basis of their competition. We
also have tru m p c ards, that is, master cards whose force varies depending
on the game: just as the relative value of cards changes with
each game, the hierarchy of the different species of capital (economic,
social, cultural, symbolic) varies across the various fields, fin other
words, there are cards that are valid, efficacious in all fields—these
are the fundamental species of capital—but their relative value as
trump cards is determined by each field and even by the successive
states of the same field^
This is so because, at bottom, the value of a species of capital (e.g.,
knowledge of Greek or of integral calculus) hinges on the existence of
a game, of a field in which this competency can be employed: a species
of capital is what is efficacious in a given field, both as a weapon
and as a stake of struggle, that which allows its possessors to wield a
power, an influence, and thus to ex ist, in the field under consideration,
instead of being considered a negligible quantity. In empirical
work, it is one and the same thing to determine what the field is,
where its limits lie, etc., and to determine what species of capital are
50. On the difference between rules and regularities and the equivocations of structuralism
between those two terms, see Bourdieu 1986a, and 1990a: 30-41.
The Purpose o f R eflexive Sociology (The Chicago W orkshop) I 99
active in it, within what limits, and so on. (We see here how the notions
of capital and field are tightly interconnected.)
At each moment, it is the state of the relations of force between
players that defines the structure of the field. We can picture each
player as having in front of her a pile of tokens of different colors,
each color corresponding to a given species of capital she holds, so
that her relativ e fo r c e in the g am e, her position in the space of play, and
also her strateg ic orien tatio n to w ard the g am e, what we call in French her
"game," the moves that she makes, more or less risky or cautious,
subversive or conservative, depend both on the total number of
tokens and on the composition of the piles of tokens she retains, that
is, on the volume and structure of her capital. Two individuals endowed
with an equivalent overall capital can differ, in their position
as well as in their stances ("position-takings"), in that one holds a lot
of economic capital and little cultural capital while the other has little
economic capital and large cultural assets.fT o be more precise, the
strategies of a "player" and everything that defines his "game" are a
function not only of the volume and structure of his capital at the m om
en t u n der c o n sideratio n and of the game chances/Huygens spoke of
lu sion es, again from lu du s, to designate objective probabilities) they
guarantee him, but also of the ev olu tion o v er tim e of the volume and
structure of this capital, that is, of his social trajectory and of the dispositions
(habitus) constituted in the prolonged relation to a definite
distribution of objective chances^
But this is not all: players can play to increase or to conserve their
capital, their number of tokens, in conformity with the tacit rules of
the game and the prerequisites of the reproduction of the game and
its stakes; but they can also get in it to transform, partially or completely,
the immanent rules of the game. They can, for instance, work
to change the relative value of tokens of different colors, the exchange
rate between various species of capital, through strategies aimed at
discrediting the form of capital upon which the force of their opponents
rests (e.g., economic capital) and to valorize the species of
capital they preferentially possess (e.g., juridical capital).51 A good
number of struggles within the field of power are of this type, notably
51. For an illustration of the growing conflict between juridical and economic capital
involved in the rise of new legal professions (notably "bankruptcy experts") at the intersection
of the two fields, see Dezalay 1989.
100 I P ie rre B ourdieu an d Lolc J. D . W acq u an l
those aimed at seizing power over the state, that is, over the economic
and political resources that enable the state to wield a power over all
games and over the rules that regulate them.
This analogy displays th e links betw een th e core concepts of your theory, but it does
not te ll us how one determ ines the existence of a field and its boundaries.
[The question of the limits of the field is a very difficult one, if only
because it is alw ay s at stake in the fie ld itself and therefore admits of no a
priori answerj Participants in a field, say, economic firms, high fashion
designers, or novelists, constantly work to differentiate themselves
from their closest rivals in order to reduce competition and to
establish a monopoly over a particular subsector of the field. (I should
immediately correct this sentence for its teleological bias, the very bias
attributed to me by those who construe my analysis of cultural practices
as based on a search for distinction. There is a production o f difference
which is in no way the product of a search fo r difference. There are
many agents—I think for instance of Gustave Flaubert— for whom to
exist in a given field consists eo ipso in differing, in being different, in
asserting one's difference, oftentimes because they are endowed with
properties such that they should not be there, they should have been
eliminated at the entrance to the field.) Their efforts to impose this
or that criterion of competency, of membership, may be more or less
. successful in various conjunctures/Thus the boundaries of the field
I can only be determined by an empirical investigationjO nly rarely do
they take the form of juridical frontiers (e.g., n u m eru s clau su s), even
though they are always marked by more or less institutionalized "barriers
to entry."
* We may think of a field as a space within which an effect of field is
exercised, so that what happens to any object that traverses this space
cannot be explained solely by the intrinsic properties of the object in
j question/The limits of the field are situated at the point where the
~i effects of the field ceasej Therefore, you must try by various means
to measure in each case the point at which these statistically detectable
effects decline. In the work of empirical research the construction
of a field is not effected by an act of imposition. For instance, I seriously
doubt that the ensemble of cultural associations (choirs, theater
groups, reading clubs, etc.) of a given American state or of a French
region form a field. By contrast, the work of Jerry Karabel (1984) suggests
that major American universities are linked by objective rela-
tions such that the structure of these (material and symbolic) relations
has effects within each of them. Similarly for newspapers: Michael
Schudson (1978) shows that you cannot understand the emergence of
the modem idea of "objectivity" in journalism if you do not see that it
arose in newspapers concerned with standards of respectability, as
that which distinguishes "news" from the mere "stories" of tabloids.
It is only by studying each of these universes that you can assess how
concretely they are constituted, where they stop, who gets in and
who does not, and whether at all they form a field.
W hat are th e m otor causes o f the functioning and transform ation of a field?
I The principle of the dynamics of a field lies in the form of its structure j
and, in particular, in the distance, the gaps, the asymmetries betweenthe
various specific forces that confront one another. [The forces that|
are active in the field— and thus selected by the analyst as pertinent1
because they produce the most relevant differences— are those which
define the specific capital. A capital does n ot ex ist an d fu n c tio n ex cept in j
relation to a field. It confers a power over the field, over the materi-j
alized or embodied instruments of production or reproduction whose
distribution constitutes the very structure of the field, and over the
regularities and the rules which define the ordinary functioning of
the field, and thereby over the profits engendered in it.
As a space of potential and active forces, the field is also a field o f
stru g g les aimed at preserving or transforming the configuration of
these forces. Furthermore, the field as a structure of objective relations
between positions of force undergirds and guides the strategies
whereby the occupants of these positions seek, individually or collectively,
to safeguard or improve their position and to impose the principle
of hierarchization most favorable to their own products/The
strategies of agents depend on their position in the field, that is, in
the distribution of the specific capital, and on the perception that they
have of the field depending on the point of view they take on the field
as a view taken from a point in the fieldy/
52. Bourdieu takes pains to emphasize the discontinuity between a social field and a
magnetic field, and therefore between sociology and a reductionistic "social physics":
"Sociology is not a chapter of mechanics and social fields are fields of forces but also
fieldsof struggles to transform or preserve these fields of forces. And the relation, practical
or reflective, that agents entertain with the game is part and parcel of the game and
may be at the basis of its transformation" (Bourdieu 1982a: 46, my translation).
The Purpose o f Reflexive Sociology (The Chicago W orkshop) I 101
102 I Pierre Bourdieu an d Lotc J. D . W acquant
W hat difference is there betw een a field and an apparatus or a system as theorized
by Luhmann for instance?
An essential difference: struggles, and thus historicity! I am very
much against the notion of apparatus, which forTnels"the Trojan
horse of "pessimistic functionalism": an apparatus is an infernal
machine, programmed to accomplish certain purposes no matter
what, when, or where.53 (This fantasy of the conspiracy, the idea
that an evil will is responsible for everything that happens in the social
world, haunts critical social thought.) The school system, the
state, the church, political parties, or unions are not apparatuses
but fields.|ln a field, agents and institutions constantly struggle,
according to the regularities and the rules constitutive of this space
of play (and, in given conjunctures, over those rules themselves),
with various degrees of strength and therefore diverse probabilities
of success, to appropriate the specific products at stake in the game.’'
/Those who dominate in a given field are in a position to make It'*
function to their advantage but they must always contend with the
resistance, the claims, the contention, "political" or otherwise, of
the dominated^
Now, under certain historical conditions, which must be examined
empirically, a field may start to function as an apparatus.54When the
dominant manage to crush and annul the resistance and the reactions
of the dominated, when all movements go exclusively from the top
down, the effects of domination are such that the struggle and the
dialectic that are constitutive of the field cease. There is history only
as long as people revolt, resist, act. Total institutions— asylums, prisons,
concentration camps— or dictatorial states are attempts to institute
an end to history. Thus apparatuses represent a limiting case,
what we may consider to be a pathological state of fields. But it is a
limit that is never actually reached, even under the most repressive
"totalitarian" regimes.55
53. "As a game structured in a loose and weakly formalized fashion, a field is not an
apparatus obeying the quasi-mechanical logic of a discipline capable of converting all action
into mere ex ecu tion " (Bourdieu 1990b: 88). See Bourdieu 1987g: 210-12 for a brief
critique of the Althusserian concept of "legal apparatus."
54. For historical examples of the opposite evolution, from apparatus to field, see
Fabiani (1989: chap. 3) on French philosophy at the end of the nineteenth century, and
Bourdieu (1987i) on the birth of impressionist painting.
55. The notion of apparatus also makes it possible to elude the question of the production
of social agents who can operate in them and make them operate, a question
The Purpose o l R eflexive Sociology (The Chicago W orkshop) I 103
r As for systems theory, it is true that it has a number of surface similarities
with field theory. One could easily retranslate the concepts of
"self-referentiality" or "self-organization" by what I put under the ,
notion of autonomy,Jn both cases, indeed, the process of differentiation
and autonomization plays a pivotal role. But the differences between
the two theories are nonetheless radical. For one thing, the ■
notion of field excludes functionalism and oirganicism: the products I
of a given field may be systematic without being products of a system,
and especially of a system characterized by common functions, inter-,
nal cohesion, and self-regulation— so many postulates of systems
theory that must be rejected. If it is true that, in the literary or artistic
field, for instance, one may treat the stances constitutive of a space of ,
possibles as a system, they form a system of differences, of distinctive '
and antagonistic properties which do not develop out of their own
internal motion (as the principle of self-referentiality implies) but via
conflicts internal to the field of production. The field is the locus of 3
relations of forCfe— and not only of meaning— and of struggles aimed
at transforming it, and therefore of endless change. The coherence -t
that may be observed in a given state of the field, its apparent orienta- j
tion toward a common function (in the case of the French G ran des eco - j
les, to reproduce the structure of the field of power; see Bourdieu
that cannot be dodged by a field analysis insofar as "a field can function only if it finds
individuals socially predisposed to behave as responsible agents, to risk their money,
their time, sometimes their honor or their life, to pursue the games and to obtain the
profits it proposes" (Bourdieu 1982a: 46; see also Bourdieu's [1987i] analysis of the historical
genesis of the artistic field as the "institutionalization of anomie" in aesthetic
matters).
The fictitious character of the notion of apparatus is further emphasized by Bourdieu
(1988i) in his critique of the notion of "totalitarianism" as developed by French
political theorists such as Lefort and Castoriadis, following Hannah Arendt. For Bourdieu,
the very concept of "totalitarianism" is what Kenneth Burke would call a "terministic
screen" which has masked the reality, however repressed, of ongoing social
contention in Soviet-type societies, just as, in the case of the court society under the
absolute monarchy of Louix XIV, "the appearance of an apparatus, in fact, conceals a
field of struggles in which the holder of 'absolute power' himself must participate"
(Bourdieu 1981c: 307). At the same time, Bourdieu (1981a) has highlighted opposite tendencies
in the functioning of the political field, where a range of factors related to the
lack of cultural capital among the dominated classes tend to foster the concentration of
political capital and therefore a drift of leftist parties toward an apparatus-like functioning.
For an analysis of the French Communist Party that critically assesses tendencies
and countertendencies toward "totalization" and of the social fabrication of members
fit to carry them out, see Verdes-Leroux 1981 and Pudal 1988,1989.
104 I P ierre B ouidieu an d Loic J. D . W acquant
1989a) are born of conflict and competition, not of some kind of immanent
self-development of the structure.56
second major difference.is that a field does not have parts, components.
Every subfield has its own logic, rules and regularities, and
each stage in the division of a field (say the field of literary production)
entails a genuine qualitative leap (as, for instance, when you
move down from the level of the literary field to that of the subfield of
novel or theater) .5lA iv e r y field constitutes a potentially open space
of play whose boundaries are dy n am ic borders which are the stake of
struggles within the field itself. A field is a game devoid of inventor
and much more fluid and complex than any game that one might ever
design. But to see fully everything that separates the concepts of field
and system one must put them to work and compare them via the
empirical objects they produce.58
B riefly, how does one carry out the study of a field and w hat are th e necessary steps
in this type of analysis?
An analysis in terms of field involves three necessary and internally
connected moments (Bourdieu 1971d). First, one must analyze the
position of the field vis-a-vis the field of power. In the case of artists
and writers (Bourdieu 1983d), we find that the literary field is contained
within the field of power where it occupies a dominated position. (In
common and much less adequate parlance: artists and writers, or intellectuals
more generally, are a "dominated fraction of the dominant
56. The necessity expressed in the structure and functioning of a field is "the product
of a historical process of progressive collective creation which obeys neither a plan
nor an obscure immanent Reason without being for that abandoned to chance" (Bourdieu
1989a: 326). Luhmann's conception of law as a system is briefly discussed in Bourdieu
1987g: 212; for a methodical comparison of Bourdieu and Luhmann, see Cornelia
Bohn's (1991) H abitus u n d K ontex t.
57. The concept of field can be used at different levels of aggregation: the university
(Bourdieu 1988a), the totality of disciplines or the faculty of the human sciences; in the
housing economy (Bourdieu 1990c), the market made up of all home-builders or the
individual construction firm "considered as a relatively autonomous unit."
58. Contrast, for instance, the way in which Bourdieu (1990b, 1990c, 1990d; Bourdieu
and Christin 1990) conceptualizes the internal dynamics of the industrial sector of
single-family home production in France as an economic field and its interface with
other fields (notably the bureaucratic field, i.e., the state) with Luhmann's (1982) and
Parsons and Smelser's (1956) abstract theorization of the boundaries between the economy
and other formal subsystems.
class.") Se£ond,. one must map out the objective structure of the relations
between the positions occupied by the agents or institutions
who compete for the legitimate form of specific authority of which
this field in the site. And, third, one must analyze the habitus of
agents, the different systems of dispositions they have acquired by internalizing
a determinate type of social and economic condition, and
which find in a definite trajectory within the field under consideration
a more or less favorable opportunity to become actualized.
The field of positions is methodologically inseparable from the
field of stances or position-takings (prises de position ), i.e., the structured
system of practices and expressions of agents. Both spaces, that
of objective positions and that of stances, must be analyzed together,
treated as "two translations of the same sentence" as Spinoza put it. It
remains, nevertheless, that, in a situation of equilibrium, the space o f
position s ten ds to com m an d the space o f position -takin g s. Artistic revolutions,
for instance, are the result of transformations of the relations of
power constitutive of the space of artistic positions that are themselves
made possible by the meeting of the subversive intentions of a
fraction of producers with the expectations of a fraction of the audience,
thus by a transformation of the relations between the intellectual
field and the field of power (Bourdieu 1987i). And what is true of
the artistic field applies to other fields: one can observe the same "fit"
between positions within the academic field on the eve of May 1968
and the political stances taken by the various protagonists of these
events, as I show in H o m o A c adem ic u s, or between the objective position
of banks in the economic field and the advertising and personnel
management strategies they deploy, etc.
In other words, the field is a critical m ediation betw een the practices of those who
partake of it and the surrounding social and economic conditions.
First, the external determinations that bear on agents situated in a
given field (intellectuals, artists, politicians, or construction companies)
never apply to them directly, but affect them only through the
specific mediation of the specific forms and forces of the field, after
having undergone a re-stru ctu rin g that is all the more important the
more autonomous the field, that is, the more it is capable of imposing
its specific logic, the cumulative product of its particular history. Seco
nd, we can observe a whole range of structural and functional hom olog
ies between the field of philosophy, the political field, the literary
The Puipose o f R eflexive Sociology (The Chicago W oiksho p) I 105
106 I Pierre B ourdieu an d Laifc J. D . W acquant
field, etc., and the structure of social space (or class structure): each
has its dominant and its dominated, its struggles for usurpation and
exclusion, its mechanisms of reproduction, and so on. But every one
of these characteristics takes a specific, irreducible form in each field
(a homology may be defined as a resemblance within a difference).
Thus, being contained within the field of power, the struggles that go
on in the philosophical field, for instance, are always overdetermined,
and tend to function in a double logic. They have political effects and
fulfill political functions by virtue of the homology of position that obtains
between such and such a philosophical contender and such and
such a political or social group in the totality of the social field.59
A third general property of fields is that they are sy stem s o f relation s
that are in dep en den t o f the popu lation s w hich these relation s defin e. When I
talk of the intellectual field, I know very well that in this field I will
find "particles" (let me pretend for a moment that we are dealing with
a physical field) that are under the sway of forces of attraction, of repulsion,
and so on, as in a magnetic field. Having said this, as soon as
I speak of a field, my attention fastens on the primacy of this system
59. "The specifically ideological function of the field of cultural production is performed
quasi-automatically on the basis of the homology of structure between the field
of cultural production, organized around the opposition between orthodoxy and heterodoxy,
and the field of struggles between the classes, for the maintenance or subversion
of the symbolic order. . . . The homology between the two fields causes the
struggles for the specific objectives at stake in the autonomous field to produce eu phem
ized forms of the ideological struggles between the classes" (Bourdieu 1979b: 82, translation
modified).
At the core of Bourdieu's theory of symbolic domination is the notion that ideological
legitimation (or "naturalization") of class inequality operates via a correspondence
which is effected only between systems. It does not require that cultural producers
intentionally endeavor to mask or to serve the interests of the dominant—indeed, the
function of "sociodicy" of culture is more effectively fulfilled when the opposite is true.
It is only by genuinely pursuing their specific interest as specialists in symbolic production
that intellectuals also legitimate a class position: "Ideologies owe their structure
and their most specific functions to the social conditions of their production and circulation,
i.e., to the functions they fulfill first fo r the specialists competing for the monopoly
of the competence in question (religious, artistic, etc.), and secon darily and
incidentally for the non-specialists" (Bourdieu 1979b: 81-82, my emphasis).
for analyses of how the homology with the structure of class relations obtains and
with what effects, see Bourdieu and Delsaut 1975 on high fashion, Bourdieu 1980a on
tastes in theater and art, Bourdieu 1988b on philosophy and Bourdieu 1989a on elite
professional schools.
The Purpose o l R eflexive Sociology (The Chicago W orkshop) I 107
of objective relations over the particles themselves. And we could say,
following the formula of a famous German physicist, that the individual,
like the electron, is an A u sg ebu rt des Felds: he or she is in a sense
an emanation of the field. This or that particular intellectual, this or
that artist, exists as su ch only because there is an intellectual or an artistic
field. (This is very important to help solve the perennial question
that historians of art have raised time and again, namely, at what
point do we move from the craftsman to the artist? This is a question
which, posed in this fashion, is almost meaningless, since this transition
is made progressively, along with the constitution of an artistic
field within which something like an artist can come to exist.)60
The notion of field reminds us that the true object of social science *
is not the individual, even though one cannot construct a field if not
through individuals, since the information necessary for statistical j
analysis is generally attached to individuals or institutions! It is the j
field which is primary and must be the focus of the research operations.
This does not imply that individuals are mere "illusions," that
they do not exist: they exist as ag en ts— and not as biological individuals,
actors, or subjects— who are socially constituted as active and
acting in the field under consideration by the fact that they possess
the necessary properties to be effective, to produce effects, in_ this
field./And it is knowledge of the field itself in which they evolve that
allows us best to grasp the roots of their singularity, their p oin t o f v iew
or position (in a field) from which their particular vision of the world
(and of the field itself) is constructed.
This is because, a t every m om ent, there is som ething lik e an "adm ission fe e " that
each field imposes and which defines, elig ib ility for participation, thereby selecting
certain agents over others. .
People are at once founded and legitimized to enter the field by their
possessing a definite configuration of properties. One of the goals of
research is to identify these active properties, these efficient charac-
60. Bourdieu's analysis of the historical formation of the artistic field in late nineteenth-century
France and of the correlative "invention" of the modern artist is the centerpiece
of a forthcoming book entitled The Econ om ics o f C u ltural G oods. For preliminary
sketches, see Bourdieu 1971a, 1971c, 1971d, 1983d, 1988d. A concise statement ofhis sociology
of aesthetics and art is Bourdieu 1987d; several of these articles are contained in
Bourdieu forthcoming c.
108 I Pierre Bourdieu an d Loic J. D . W acquant
teristics, that is, these forms of sp ecific capital. There is thus a sort of
hermeneutic circle: in order to construct the field, one must identify
the forms of specific capital that operate within it, and to construct the
forms of specific capital one must know the specific logic of the field.
There is an endless to and fro movement in the research process that
is quite lengthy and arduous.61
To say that the structure of the field—note that I am progressively
building a w o rkin g definition of the concept—is defined by the structure
of the distribution of the specific forms of capital that are active in
it means that when my knowledge of forms of capital is sound I can
differentiate everything that there is to differentiate. For example, and
''T*this is one of the principles that guided my work on intellectuals, one
cannot be satisfied with an explanatory model incapable of differentiating
people— or, better, positions—who ordinary intuition in
the specific universe tells us are quite different. In such a case, one
should search for what variables have been omitted which permit us
to differentiate. (Parenthesis: ordinary intuition is quite respectable;
only, one must be sure to introduce intuitions into the analysis in a
conscious and reasoned manner and to control their validity empirically,62
whereas many sociologists use them unconsciously, as
when they build the kind of dualistic typologies that I criticize at the
beginning of H om o A c adem ic u s, such as "universal" vs. "parochial" intellectuals.)
Here intuition raises questions: "Where does the difference
come from?"
[ One last and critical point: soc ial ag en ts are n ot " partic les" that are
| mechanically pushed and pulled about by external forces. They are,
rather, bearers of capitals and, depending on their trajectory and on
the position they occupy in the field by virtue of their endowment
(volume and structure) in capital, they have a propensity to orient
61. For a detailed illustration of this "hermeneutic circle," through which the population
of relevant individuals or institutions and the efficient assets or forms of capital
are mutually specified, see Bourdieu's study of the reform of governmental housing
policy in France in the mid-1970s (Bourdieu and Christin 1990, esp. 70-81).
62. "Far from being, as certain 'initiatory' representatives of the 'epistemological
break' would have us believe, a sort of simultaneously inaugural and terminal act, the
renunciation of first-hand intuition is the end product of a long dialectical process in
which intuition, formulated in an empirical operation, analyses and verifies or falsifies
itself, engendering new hypotheses, gradually more firmly based, which will be transcended
in their turn, thanks to the problems, failures and expectations which they
bring to light" (Bourdieu 1988a: 7).
The Puipose o i R eflexive Sociology (The Chicago W orkshop) I 109
themselves actively either toward the preservation of the distribution
of capital or toward the subversion of this distribution. Things are of
course much more complicated, but I think that this is a general proposition
that applies to social space as a whole, although it does not
imply that all small capital holders are necessarily revolutionaries and
all big capital holders are automatically conservatives.
Let us grant th at th e social universe, a t least in advanced societies, is m ade up of a
num ber of d ifferentiated fields th at have both invariant properties (this justifies the
project of a general theory of fields) and varying properties rooted in th eir specific
logic and history (which requires a genetic and com parative analysis of each of
them ). How do these diverse fields relate to one another? W hat is the nature of
th eir articulation and th eir differen tial w eight?
The question of the interrelation of different fields is an extremely
complex one. It is a question that I would normally not answer because
it is too difficult, and I risk saying things that are relatively
simple and might thereby reawaken modes of analysis phrased in
terms of "instance" and "articulation," that allowed some Marxists to
give rhetorical solutions to problems that only empirical analysis can
tackle'. I believe indeed that there are n o tran shistoric law s o f the relation,
betw een fields, that we must investigate each historical case separately
Obviously, in advanced capitalist societies, it would be difficult to
maintain that the economic field does not exercise especially powerful
determinations. But should we for that reason admit the postulate of 1
its (universal) "determination in the last instance"? An example from I
my research on the artistic field will, I believe, suggest how complicated
this question is.
When we study this question historically, we observe that a process
began with the Quattrocento which led the artistic field to acquire
its true autonomy in the nineteenth century. From then on,
artists are no longer subjected to the demands and commands of
sponsors and patrons, they are freed from the state and from academies,
etc. Most of them begin to produce for their own restricted market
in which a sort of deferred economy operates (Bourdieu 1983d,
1987i). Everything would lead us to believe that we are dealing with
an irreversible and irresistible movement toward autonomy, and that
art and artists have once and for all achieved their freedom from external
forces. Now, what do we observe today? A return of patronage,
of direct dependency, of the state, of the most brutal forms of cen-
110 I P ieire Bourdieu and Loic J. D . W acquant
sorship, and suddenly the idea of a linear and indefinite process of
autonomization is reopened. Look at what happened to a painter
such as Hans Haacke who uses artistic tools to question interferences
with the autonomy of artistic creation.63 He exhibited at the Guggenheim
Museum a painting displaying the origins of the financial resources
of the Guggenheim family. Now, the Director of the Museum
had no alternative other than to resign or be dismissed by his funders,
or to ridicule himself in the eyes of artists by refusing to exhibit
the painting. This artist gave a function back to art and immediately
he ran into trouble. Thus we discover that the autonomy acquired by
artists, originally dependent for both the content and the form of their
work, implied a submission to necessity: artists had made a virtue out
of necessity by arrogating to themselves the absolute mastery of the
form, but at the cost of a no less absolute renunciation of function. As
soon as they want to fulfill a function other than that assigned to
them by the artistic field, i.e., the function which consists in exercising
no social function ("art for art's sake"), they rediscover the limits
of their autonomy.
[-This is only one example, but it has the merit of reminding us^that
relations between fields—the artistic and the economic field in this
case— are not defined once and for all, even in the most general tendencies
of their evolutionJThe notion of fiqld does not provide readymade
answers to all possible queries/in the manner of the grand
concepts of "theoreticist theory" which claims to explain everything
and in the right order. Rather, its major virtue, at lea&tinmy eyes, is
that it promotes a mode of construction that has to berethought anew
every time. It forces us to raise questions: about the limits of the universe
under investigation, how it is "articulated," to what and to what
degree, etc. It offers a coherent system of recurrent questions that
saves us from the theoretical vacuum of positivist empiricism and
from the empirical void of theoreticist discourse.
In a recent issue of Actes de la recherche en sciences sociales (March 1990) devoted to
the "Economy of Housing," that is, the set o f social spaces th at have to be taken
into account to understand th e production and circulation of this peculiar economic
good th at the single-fam ily home is, you have been led to analyze the genesis of
63. The sociological significance of Haacke's work is underlined by Howard Becker
and John Walton (1986).
The Purpose o f R eflexive Sociology (The Chicago W orkshop) I 111
stale policies which, in this case, enter directly in the determ ination of the functioning
of an economic m arket. In so doing, you have begun to outline a theory of the
state as a sort of m e ta -fie ld .M
Indeed, it seems to me that, when you take a close look at what goes
on inside what we call the "state," you immediately annul most of the
scholastic problems that scholars, armchair Marxists and other speculative
sociologists, keep raising about the state, that quasi-metaphysical
notion that must be exploded in order to "go to the things
themselves," as Edmund Husserl said in a different context. I think
for instance of the consecrated theoretical alternative between "correspondence"
(or dependance) and "autonomy." This alternative presupposes
that the state is a well-defined, clearly bounded and unitary
reality which stands in a relation of externality with outside forces
that are themselves clearly identified and defined (for instance, in
the case of Germany, on which so much ink has been spilled because
of the famous Son derw eg , the traditional landed aristocracy of the
Junkers, or the wealthy industrial bourgeoisie, or, in the case of
England, the urban entrepreneurial bourgeoisie and the country gentry).
In fact, what we encounter, concretely, is an ensemble of administrative
or bureaucratic fields (they often take the empirical form of
commissions, bureaus and boards) within which agents and categories
of agents, governmental and nongovernmental, struggle over this
peculiar form of authority consisting of the power to ru le via legislation,
regulations, administrative measures (subsidies, authorizations,
restrictions, etc.), in short, everything that we normally put under the
rubric of state policy as a particular sphere of practices related, in this
case, to the production and consumption of housing.
The state, then, if you insist on keeping this designation, would be
64. The analysis of the structuring role of the state in the economics of housing is
found in Bourdieu 1990b, and Bourdieu and Christin 1990. Bourdieu was first led to
address the question of the state frontally in ha n oblesse d' Etat, when he came to the
conclusion that the "contemporary technocracy" are the "structural (and sometimes
genealogical) inheritors" of the n oblesse de robe which "created itself [as a corporate
body] by creating the state," and formulated the hypothesis that "the state nobility . . .
and educational credentials are bom of complementary and correlative inventions"
(Bourdieu 1989a: 544, 540). Bourdieu's course at the College de France in 1988-91 has i
been devoted to this topic, in the form of an investigation of the genesis and effects of l
the modem state understood as the organizational expression of the concentration o f.
symbolic power, .or "public trove of material and symbolic resources guaranteeing pri- j
vate appropriations" (Bourdieu 1989a: 540).
112 I Pierre Bourdieu and Loic J. D . W acquant
the ensemble of fields that are the site of struggles in which what is at
stake is— to build on Max Weber's famed formulation— the m on opoly
o f leg itim ate sy m bolic v io le n c e ,65 i.e., the power to constitute and to impose
as u n iv ersal a a d u n iv ersally app licable within agiven "nation," that
is, within the boundaries of a given territory, a common set of coercive
norms. As I showed in the case of state housing policy in France
between 1970 and 1980, these fields are the locus of a constant confrontation
between forces belonging both to the private sector (banks
and bankers, construction and architectural firms, etc.) and to the
public sector (ministries, administrative divisions within these ministries,
and the g ran ds co rps d' E tat who staff them),66 that is, sub-universes
themselves organized as fields that are both united by and
divided over internal cleavages and external oppositions. The notion
of "state" makes sense only as a convenient stenographic label—but,
for that matter, a very dangerous one— for these spaces of objectiv e relation
s between position s of power (assuming different forms) that can
take the form of more or less stable networks (of alliance, cooperation,
clientelism, mutual service, etc.) and which manifest themselves in
phenomenally diverse interactions ranging from open conflict to
more or less hidden collusion.
As soon as you examine in detail how "private" agents or organizations
(say, banks interested in the passing of certain regulations likely
to boost the diffusion of given kinds of housing loans), which are
themselves in competition with one another, work to orient "state"
policy in each of their domains of economic or cultural activity (the
same processes can be observed in the case of an educational reform),
how they form coalitions and ties with other bureaucratic agents
65. For developments, see Bourdieu 1989a: part 5, and Bourdieu and Wacquant 1991:
100: "The state is in the final analysis the great fount of symbolic power which accomplishes
acts of consecration, such as the granting of a degree, an identity card or a certificate—so
many acts through which the authorized holders of an authority assert that
a person is what she is, publicly establish what she is and what she has to be. It is the
state,-as the reserve bank of consecration, that vouchsafes these official acts and the
agents who effect them and, in a sense, carries them out via the agency of its legitimate
representatives. This is why I distorted and generalized Max Weber's famous words to
say that the state is the holder o f a m on opololy , n ot on ly ov er leg itim ate phy sical v iolence, but
‘ ov er legitimate sy m bolic v iolence as well."
66. The gran ds corps are corporate bodies made up of graduates of the country's top
G randes ecoles which traditionally reserve for themselves certain upper-level administrative
positions within the French state. (On G randes ecoles, see p. 231, n. 22.)
H ie Purpose o f Reflexive Sociology (The Chicago W orkshop) I 113
whose preference for a given type of measure they share, how they
confront yet other organizational entities with their own interests and
resources (e.g., the properly bureaucratic capital of management of
regulations), you cannot but jettison all speculations about correspondence
and autonomy. To be truthful, I feel closer, on this count, to the
analyses of Edward Laumann (Laumann and Knoke 1988), though I
differ from him in other respects, than to those of Nicos Poulantzas
(1973) or Theda Skocpol (1979), to cite two names emblematic of traditional
positions on correspondence and autonomy. By this, I mean to
point out also that, in such matters as elsewhere, the "armchair Marxists,"
those materialists without materials, whom I ceaselessly opposed
at the time of their apogee in the 1960s, have done much to help
the perpetuation of scholastic issues.
More generally, this illustrates what makes for much of the difficulty
of my position in the sociological field. On the one hand, I can
appear very close to the "Grand Theoreticians" (especially the structuralists)
insofar as I insist on structural configurations that cannot be
reduced to the interactions and practices through which they express
themselves. At the same time, I feel a kinship and a solidarity with
researchers who "put their noses to the ground" (particularly symbolic
interactionists, and all those who, through participant observation
or statistical analysis, work to uncover and to debunk the
empirical realities that Grand Theoreticians ignore because they look
down upon social reality from such heights), even though I cannot
agree with the philosophy of the social world which often undergirds
their interest in the minutiae of daily practices and which, in this
case, is in fact imposed upon them by this "close-up view" and by the
theoretical myopia or the blindness to objective structures, to relations
of force that are not immediately perceivable, that this view
encourages.
W hat, then, would separate your analysis of the state as a set of partially overlapping
bureaucratic fields from Laumann and Knoke's (1988) notion of the "organizational
state" or from netw ork theory m ore broadly?
could recall here the distinction I established, against Max Weber in
particular, between structure and interaction or between a structuralrelation
which operates in a permanent and invisible fashion, and an'
effective relation, a relation actualized in and by a particular exchange)
(see Bourdieu 1971b, 1971e, 19 n fact, the structure of a field,
114 I P ierre Bourdieu an d Loic J. D . W acquant
understood as a space of objective relations between positions deIfined
by their rank in the distribution of competing powers or species
|of capital, is different from the more or less lasting networks through
which it manifests itself. It is this structure that determines the possibility
or the impossibility (or, to be more precise, the greater or lesser
probability) of observing the establishment of linkages that express
and sustain the existence of networks.j^The task of science is to uncover
the structure of the distribution of species of capital which
tends to determine the structure of individual or collective stances
' taken, through the interests and dispositions it conditions.] In network
analysis, the study of these underlying structures has Been sacrificed
to the analysis of the particular linkages (between agents or
institutions) and flows (of information, resources, services, etc.)
through which they become visible—no doubt because uncovering
the structure requires that one put to work a relational mode of thinking
that is more difficult to translate into quantitative and formalized
data, save by way of correspondence analysis.
I could pursue this argument by drawing on the research I have
been conducting over the past few years on the historical genesis of
the state. I could argue, to simplify greatly, that there has occurred,
since the construction of the dynastic state and, later, of the bureaucratic
state, a long-term process of concentration of different species
of power, or capital, leading, in a first stage, to private monopolization—by
the king— of a public authority at once external and superior
_to all private authorities (lords, bourgeoisie, etc.). The concentration
of these different species of capital— economic (thanks to taxation),
military, cultural, juridical and, more generally, symbolic— goes hand
in hand with the rise and consolidation of the various corresponding
fields.|The result of this process is the emergence of a specific capital,
properly statist cap ital, born of their cumulation, which allows the state
to wield a power over the different fields and over the various forms
of capital that circulate in them/This kind of m eta- capital capable of
exercising a power over other species of power, and particularly over
their rate of exchange (and thereby over the balance of power between
!■their respective holders), defines the specific power of the statej It follows
that the construction of the state goes hand in hand with the
constitution of the field of power understood as the space of play in
which holders of various forms of capital struggle in partic u lar for
. power over the state, that is, over the statist capital that grants power
The Purpose o f R eflexive Sociology (The Chicago W orkshop) I 115
over the different species of capital and over their reproduction (via
the school system in particular).
4 Interest, Habitus, R ationality
Your use of the notion of interest has often called forth the charge of "econom
ism ."17 W hat theoretical role does interest play in your m ethod of analysis?
The notion of interest imposed itself upon me as an instrument of
rupture with a philosophical anthropology, a naive conception of human
conduct that was dominant when I started working in the social
sciences. I have often quoted a remark of Weber about law which says
that social agents obey a rule only insofar as their interest in following
it outweighs their interest in overlooking it. This sound materialist
principle reminds us that, before claiming to describe the rules according
to which people act, we should ask what makes those rules
operative in the first place.
fThus, building upon Weber, who utilized an economic model to
uncover the specific interests of the great protagonists of the religious
game, priests, prophets, and sorcerers (Bourdieu 1971b, 1987h), I introduced
the notion of interest into my analysis of cultural producers
in reaction to the dominant vision of the intellectual universe, to
question the ideology of the freisc hw eben de In tellig en z \ l much prefer to
use the term U lusio, since I always speak of specific interest, of interests
that are both presupposed and produced by the functioning
of historIca21y delimited fields. .Paradoxically, the term interestiias
brought forth the knee-jerk accusation of economism.68Jln fact, the
67. E.g., Paradeise 1981, Caille 1981 and 1987a, Richer 1983, Adair 1984, Kot and
Lautier 1984, Ranciere 1984: 24, Joppke 1986, Sahlins 1989: 25. Thus Fiske (1991: 238)
lumps Gary Becker and Bourdieu together as defenders of "the selfish rationality assumption"
that constitutes one of his four models of social relations. The opposite interpretation
is vigorously defended by Harker, Mahar, and Wilkes (1990: 4-6),
Thompson (1991) and Ostrow (1990:117), among others, who commend Bourdieu for
his rejection of economism.
68. Bourdieu's opposition to economism is clear from his first ethnographic pieces
on the sense of honor among the Kabyles (Bourdieu 1965 and 1979d). It is argued at
great length in Esqu isse d' un e theorie de la pratiqu e and in The Logic o f P ractice: "Economism
is a form of ethnocentrism. Treating precapitalist economies, in Marx's phrase,
'as the Fathers of the Church treated the religions that preceded Christianity,' it applies
to them categories, methods (economic accounting, for instance), or concepts (such as
the notions of interest, investment, or capital, etc.) that are the historical product of