ISSN 1355-6509 © St Jerome Publishing, Manchester
The Translator. Volume 11, Number 2 (2005), 219-236 ISBN 1-900650-86-X
Bourdieu the Ethnographer
The Ethnographic Grounding of Habitus and Voice1
Institute of Education, University of London & Ghent
Abstract. A remarkable feature of the way in which Bourdieu’s
work has been adopted in studies of language in society is the
emphasis on Bourdieu as a macro-sociologist providing insights
into the larger processes of structuration in tightly integrated First-
World societies. Yet, Bourdieu himself consistently emphasized the
ethnographic epistemological foundations of his work, and espe-
cially in the last years of his life abundantly acknowledged
influences from ethnography. This paper delves into Bourdieu’s
views on ethnography-as-epistemology, arguing that one of
Bourdieu’s central concepts, habitus, should be seen as inextri-
cably linked to situated ethnographic inquiry. Taking habitus as
an ethnographic concept, we may find better ways of investigat-
ing problems of voice – the conditions for speaking in society.
This point is illustrated with examples from the Belgian asylum
procedure, where habituated conversational practices by the in-
terviewer simultaneously appear to contain proleptic moves that
‘prepare’ the story of the applicant for the next step in the asylum
procedure. We see in this form of simultaneity the on-the-spot,
layered deployment of macro-social (institutional) conventions
through conversational, co-operative practices.
The concerned was interrogated on November 23, 1993 at the [High
Commissariat for Refugees and Stateless Persons] in the presence of
1 I thank Moira Inghilleri for gently twisting my arm to write this article, thus allowing
me to revise and expand the short essay on ‘Bourdieu the ethnographer’ which I had
written as the introductory paper for a panel on ‘Bourdieu and Ethnography’ (Interna-
tional Pragmatics Conference, Toronto, July 2003). Marco Jacquemet, Jane Hill, Chris
Stroud, Adrian Blackledge, Stef Slembrouck, Jim Collins and Moira herself all com-
mented perceptively on that essay. I wish to specially thank Katrijn Maryns, with whom
I have worked for several years on these issues, and whose wonderful dissertation (Maryns
2004) provided me with further inspiration for this article. Just as this paper went into
production, a special issue of Ethnography came out devoted to Bourdieu and field-
work. Although I could not completely incorporate that material in this paper, I
recommend Loic Wacquant’s introductory chapter (Wacquant 2005): a fascinating and
insightful reflection on Bourdieu’s basis in fieldwork, largely corroborating the analysis
Bourdieu the Ethnographer220
[name], his attorney.
He claimed to be a ‘political informant’ of the MPLA. On October
18, 1992 however, he passed on information to UNITA. At the UNITA
office, however, he met with Major [name], who works for the MPLA.
Two days later, Major [name] had the concerned arrested. Fearing
that the concerned would give the Major away at the trial, [name of
the Major] helped the concerned to escape. The concerned fled to
[locality] where a priest arranged for his departure from Angola. The
concerned came, together with his wife [name and register number]
and three children, through Zaïre and by plane, to Belgium. They ar-
rived on May 19, 1993.
It has to be noted that the concerned remains very vague at certain
points. Thus he is unable to provide details about the precise content
of his job as ‘political informant’. Furthermore the account of his
escape lacks credibility. Thus it is unlikely that the concerned could
steal military clothes and weapons without being noticed and that he
could consequently climb over the prison wall.
It is also unlikely that the concerned and his wife could pass the
passport control at Zaventem [i.e., Brussels Airport] bearing a pass-
port lacking their names and their pictures.
Furthermore, the itinerary of the concerned is impossible to verify
due to a lack of travel documents (the concerned sent back the
The statements of the concerned contain contradictions when com-
pared to his wife’s account. Thus he declares that the passports which
they received from the priest [name] were already completely in or-
der at the time they left Angola. His wife claims that they still had to
apply for visa in Zaïre.
This is the text of an official letter sent by the Belgian asylum authorities to
a refugee from Angola (the original text was in Dutch bureaucratese; the
version above is my own translation). Such letters are part of the asylum
application process, which in itself usually involves three stages: (1) an ini-
tial interview on the causes and reasons for seeking asylum and on the ‘facts’
of the case, conducted by the Immigration Office (Dienst Vreemdelin-
genzaken); (2) in case of a negative initial verdict, a second ‘deeper’ interview
conducted by the High Commissariat for Refugees and Stateless Persons
(Hoog Commissariaat voor Vluchtelingen en Statenlozen); and (3) in case
of a second negative verdict, a hearing before the Board of Appeal of the
Supreme Court (Raad van State) (Maryns 2004). This particular letter comes
from the second stage of the process. The High Commissariat informs the
man about the rejection of his asylum application and provides reasons for
As a text-artefact, the letter represents an advanced stage in a long cycle
Jan Blommaert 221
of entextualizations of ‘the story of the applicant’ – a text trajectory which
contains numerous interventions in the story by various actors: summaries,
translations, reformulations, interpretations, evaluations (see Silverstein and
Urban 1996; also Sarangi and Slembrouck 1996). The text trajectories are,
as a rule, multilingual – applicants from other countries tell their stories to
Belgian asylum officials who are normally native speakers of either Dutch
or French, sometimes in uneasy varieties of French or English, sometimes in
what is claimed (or believed) to be their native language with the assistance
of an interpreter – causing, as can be imagined, nightmarish communicative
events. Rejection of applications is the rule as well. Belgium has the highest
percentage of rejected asylum applications in the whole of the EU, and an
overwhelming number of these rejections are based on assessments of the
‘quality’ of the applicant’s story. They are, in other words, based on ele-
ments gathered during the text trajectory sketched above (Maryns 2004,
Blommaert 2001a, 2001b, 2005, Inghilleri 2003).
Who speaks here? Whose voice do we hear? Surely there is a stentorian
bureaucratic voice; but that voice has completely absorbed and appropriated
the voice of the applicant. In fact, most of what we read in the text is a
summarized version of the applicant’s story, framed in a metapragmatic evalu-
ative grid that casts doubt on or disqualifies episodes, fragments or the whole
of the story. That is, it casts doubt on or disqualifies utterances made by the
applicant at one point. What we read in the text is the end product of a se-
quence of entextualizations of an orally produced, performed narrative, which
we know was produced in a particular language, with a particular rhythm,
prosody and intonation, generically formatted into different episodes, plots,
side-plots and sub-plots, accompanied by gesture and facial expression, and
in response to prompts given by an interlocutor (whose utterances are also
orally produced and performed in the sense given here). The interaction it-
self is a highly specific, genred event in which the conditions and expectations
are not necessarily equally well understood by both parties (Maryns 2004).
The text does everything to elide these performance features (and does a
remarkable job, to be sure); but it is good to remember that intertextuality
here is not an abstract transfer of textual substance but a process that has to
deal with performed, situated narrative. This, as mentioned above, raises
issues of voice. The transformations of the story entail changes in the condi-
tions for articulating subjectivity. We see in the transfer from performed
narrative to bureaucratic text-artefact a transfer of one form of subjectivity
into another: from a situated, conditioned and contextualized subjectivity
articulated in performed narrative, to a decontextualized ‘pure’ subjectivity
which ascribes the profoundly manipulated final and evaluated version of
the story to the applicant himself, as in, “the concerned remains vague”, “the
statements of the concerned contain contradictions”. This is subjectification
in the sense of Foucault (2003); the subject as a product of power, as an
Bourdieu the Ethnographer222
effect of evaluative statements by others on one’s (auto-?)biography. The
statements of the applicant were made under strict conditions of power, dur-
ing a genred event known as the application interview. Their subsequent
entextualizations were also produced within a particular regime of power in
which ‘the story of the applicant’ was scanned for contradictions, unlikely
episodes, incoherent bits or passages lacking detail and precision. But when
this work is over, the story is re-attributed to the applicant: it has become
‘his own’ story again (see also Barsky 1994, Briggs 1997). So, who speaks?
According to the bureaucratic voice, it is the applicant who has spoken, but
not well enough.
In what follows, I intend to elaborate on the tension between subjectivity
and situatedness – a tension which is crucial to any understanding of voice.
Voice here refers to the capacity to make oneself understood as a situated
subject. It is a matter of ‘brought abouts’ and ‘brought alongs’, of perduring
conditions for articulative performance as well as of on-the-spot deployment
of resources and means.2 And I will argue that Bourdieu’s concept of habi-
tus can be useful in interpreting the tension, provided we see habitus as
ethnographically grounded, i.e., as allowing for the situated, performed
subjectivities we address here. Recall that Bourdieu used habitus in a var-
iety of (situated) formulations; the shorthand definition I shall use in the
remainder of this paper is this (Bourdieu 1990a:54):
the structures characterizing a determinate class of conditions of ex-
istence produce the structures of the habitus, which in their turn are
the basis of the perception and appreciation of all subsequent experi-
ences. The habitus, product of history, produces individual and
collective practices – more history – in accordance with the schemes
generated by history.
Note the emphasis on the historical embeddedness of habitus. Bourdieu does
emphasize the dimension of durability in anything he says about habitus –
habitus as a system of perduring conditions for thought and action, as a sedi-
ment of structure in our agency – but he does so within a historical, not a
timeless frame. That means he does so within a frame that allows for consid-
erable change, even within the same synchrony since different historically
grounded forms of habitus may be involved in the same event. Habitus is
durable, but not static. In order to understand this important nuance, we need
to turn to Bourdieu-the-ethnographer. Bourdieu used an ethnographic epis-
2 The perduring conditions on communication can be called ‘pretextual conditions’
(Maryns & Blommaert 2002): those aspects of competence that people carry along as
‘baggage’, involving control over and access to language and communicative resources
such as linguistic variety, literacy skills, genres, registers, accents, and so forth.
Jan Blommaert 223
temology in his work, and this is what should prevent us from seeing struc-
ture (sedimented in habitus) as static.
1. Bourdieu the ethnographer
In studies on language in society, Bourdieu is often seen as a macro-socio-
logical theorist, whose main contributions lie in the field of theorizing social
action and structuration, symbolic power relations and capital, and habitus.
A glance at the politics of citation of his work in mainstream pragmatics and
sociolinguistics testifies to this: Bourdieu is most often referred to when
concepts such as habitus and field are brought into analyses, as well as when
language is defined as a symbolic commodity with exchange value (the main
source for such references would be his translated collection Language and
Symbolic Power, Bourdieu 1990b). Furthermore, one would see references
to his work confined to analyses of First-World societies, some analysts of
other societies effectively contesting the validity of his theoretical apparatus
for non-western societies and social processes (Haeri 1997, Stroud 2002).
His oeuvre is often presented as closed, bounded, finished.3
Bourdieu’s work contains more that is of crucial importance for the de-
velopment of a critical science of language in society, in particular when we
focus on Bourdieu-the-ethnographer. There are, I believe, at least three mo-
tives for this: (1) Bourdieu’s own interest in ethnography; (2) Bourdieu’s
own use of ethnography; and (3) what I would like to call the ethnographic
invitation in Bourdieu’s work. Taken together, they offer us another Pierre
Bourdieu and another oeuvre, probably much closer and much more directly
useful to the study of language in society.
Bourdieu’s lively interest in ethnographic work is a matter of record. In
fact, he was instrumental in introducing important work from the American
ethnographic tradition to a Francophone and wider interdisciplinary audi-
ence. Erving Goffman was greatly admired by Bourdieu (Bourdieu wrote
an obituary on Goffman), and as one of his last publishing projects he co-
edited a collection of papers by Aaron Cicourel translated into French
(Cicourel 2002). He was also a founding board member of the journal Eth-
nography (founded in 2000) and contributed important papers to its first
few issues. He had a keen interest in and a profound understanding of con-
versation analysis, ethnomethodology, Batesonian anthropologies of
communication, and Batesonian ethnography as exemplified in Naven
3 Two recent collections of papers engage in detailed discussions on the value of
Bourdieu’s work for comparative sociolinguistic studies: Parlenko and Blackledge (2002)
and Fenigsen (2003). Particularly successful applications of Bourdieu’s insights in so-
ciolinguistic analysis are Heller (1999, 2002) and Jaffe (1999).
Bourdieu the Ethnographer224
(Bateson 1958). Some of his close collaborators (for instance Loïc Wacquant
and Yves Winkin) can clearly be situated within that school of ethnography.
The second and third reasons deserve some more comment, for they are
more fundamental. I will first discuss the role of ethnography in Bourdieu’s
own work; after that I will turn to the ethnographic invitation in his work.
The backbone of Bourdieu’s own work is ethnography, and remarkably,
he is consistently explicit about this. His use of ethnographic vignettes in his
works (e.g. The Logic of Practice or Homo Academicus) is more than a
stylistic feature of his writing, it is a crucial ingredient of his theorizing.
Throughout his work, whether discussing Algerian workers, immigrants in
the Parisian banlieue, French academics or the French middle class, there is
a sense of lived and experienced reality. Bourdieu starts from a fully ethno-
graphic attendance to microscopic detail of human activity and works his
way up to generalizable patterns of behavior and social organization. A con-
cept such as ‘sens pratique’ is richly ethnographically grounded, and so is
habitus. And even the English version of Distinction (one of his most socio-
logically-oriented works) starts with the announcement that “it can be read
as a sort of ethnography of France” (Bourdieu 1986: xi, emphasis added).
Let me elaborate on this, using two particularly illuminating texts: the
paper on ‘Making the economic habitus’ (Bourdieu 2000), published in
the first issue of Ethnography, and the preface to the 1990 Polity Press
edition of The Logic of Practice (Bourdieu 1990a). The first one can be
read as a declaration of faith in ethnography, where Bourdieu expands on
the way in which he depended on ethnography for constructing theoretical
concepts such as habitus. The second is a very informative intellectual
In ‘Making the economic habitus’, Bourdieu revisits his 1960s fieldwork
in Algeria. The sense of experienced reality is there in the first sentence: “I
witnessed, in Algeria in the 1960s, what with hindsight appears to me to be
a veritable social experiment” (Bourdieu 2000:18). A motif has been flagged:
the layers of reinterpretations that tend to cover fieldwork impressions, in
themselves valuable because of the deeply contextualized nature of percep-
tion and understanding in the field. Bourdieu now intends to go back to the
unmitigated, raw, contextualized fieldwork observations and ‘peel off’ these
layers (ibid., emphasis in original):
Without repeating the details of already published analyses, and giv-
ing priority to unpublished information preserved in my fieldwork
notebooks, I would like to outline briefly what appeared to me with
total clarity in that quasi-laboratory situation, namely, the mismatch
between economic dispositions fashioned in a precapitalist economy
and the economic cosmos imported and imposed, oftentimes in the
most brutal way, by colonization.
Jan Blommaert 225
The purpose of the paper is theoretical, to expose the fallacies of rational
actor models in economic theory and to pit it against the notion of economic
habitus. Such a habitus, Bourdieu insists, is the product of particular histori-
cal conditions; the economic dispositions have a social and historical genesis.
[i]t was no doubt because I found myself in a situation where I could
directly observe the disarray or the distress of economic agents de-
void of the dispositions tacitly demanded by an economic order that
for us is entirely familiar – in which, being an embodied and there-
fore naturalized social structure, they appear as self-evident, necessary
and universal – that I was able to conceive of the idea of statistically
analysing the conditions of possibility of these historically constituted
In this fragment we see a harmonious move from ethnography-as-direct-
observation to statistical analysis. The move might puzzle hardcore
ethnographers, so some clarification may be in order. I turn to the second
text for that, the preface to The Logic of Practice (Bourdieu 1990a). Let us
recall that the book not only develops the concepts of habitus and practice,
but also makes crucial epistemological points about objectification and
objectivity-in-subjectivity. The preface retrospectively couches this argument
in an intellectual travelogue.
Bourdieu starts from an acute awareness of ‘framing’ in research. We all
enter our research sites under particular sociohistorical conditions, and these
have an effect on what we see, perceive and understand. Bourdieu was aware
of this during his 1960s fieldwork in Algeria. The country had just passed
through a traumatic war of liberation against French colonial rule, and the
impact on his fieldwork was considerable, Bourdieu himself of course being
French among the Algerians. There is a wonderful vignette he uses to illus-
trate this. He had recently looked at a photo taken in a house in Kabylia
during fieldwork, and he was surprised to see how well lit the indoors image
was, despite the fact that Bourdieu did not have a flash on his camera. The
reason was that the roof of the house had been blown off by a French
grenade during the war. For Bourdieu, this example demonstrated the way
in which the conditions for seeing, observing and grasping the observed
and experienced social reality were deeply influenced by historical proc-
esses determining (i.e., collapsing in) the fieldwork event. There was no
way in which he, at that time and place, could not be influenced by the
effects of the past war in Algeria. It affected his ‘field’, surely, but also his
way of entering and cruising through that field – his identity as an ethnog-
rapher. This is where ethnography, in Bourdieu’s experience, became an
Bourdieu the Ethnographer226
In order to escape the biasing effect of these conditions, Bourdieu ex-
plored two sets of measures. First, he emphasized the importance of revisiting
the same object over and over again – a point already illustrated in ‘Making
the economic habitus’ – as well as of comparison (his work in Algeria was
followed by ‘native ethnography’ in the Béarn and in the French academic
and educational system) and expansion (including more materials than just
those collected during fieldwork). Second, he turned to the kind of structur-
alism then advocated by Lévi-Strauss, in order to find a vantage point which
enabled scientific objectivity. In doing this, he attempted like Lévi-Strauss
to move from ethnography to ethnology – a search for transcontextual (or a-
contextual) ‘driving principles’ in the social system observed, by focusing
on correlations, contrasts and forms of systemic coherence.
Whereas the first set of measures was maintained throughout his oeuvre,
the second set – the turn to structuralism – was abandoned. The main reason
was again ethnographic experience. Instead of the strict, transparent and
mechanical schemes of structuralism, Bourdieu had encountered paradoxes,
contradictions and flexible potential in the field. Furthermore, he had experi-
enced experience, so to speak: the fact that the distance advocated in
ethnology is, in actual fieldwork conditions, overgrown with sharedness of
meaning, joint understandings of ‘the logic of the game’, and so on. In other
words, Bourdieu had ethnographically experienced the fact that the ethno-
logical claim to objectivity-through-distance generated another, and a
potentially more dangerous form of ethnocentrism than the one intrinsic to
his own observer’s – but participating and co-constructing – role in ethnog-
raphy. Bourdieu worries about the specific role of the observer, and this role
is not substantially different whether one investigates Algeria, the Béarn or
the Sorbonne. From that point onwards, ‘dispositions’ occur, and Bourdieu
theorizes how he himself became part of the object – the objectification of
subjectivity.4 This is also the point where he makes the shift from anthro-
pology (or ‘ethnology’, see below) to sociology: a science in which precisely
the objectification of subjectivity is central, and a science that can aspire to
eventually develop a subject.
Important for our purposes here is that Bourdieu, whenever he reacts
against anthropology, reacts against Lévi-Straussian ethnology and its “meth-
odologically provoked anamnesis” (2000:24) which suggests closure and
total strangeness – absence of shared understanding – between observer and
observed. And he replaces it with a ‘sociology’ which aims at recovering,
on the basis of an experienced logic of practice, the internal dynamics of
4 This merging of subject and object is of course epitomized in Bourdieu’s work on
academia, e.g., Bourdieu (1988); Bourdieu et al. (1994).
Jan Blommaert 227
social systems. The formulations are remarkably similar to some current theo-
rizing in and on ethnography, notably Hymes (1996) and Fabian (1991, 1995,
2001). To make this point clear: Bourdieu started from an inherited (Lévi-
Straussian) methodology which separated ethnography from ethnology, the
former being the descriptive, methodical apparatus preparing the ground for
the latter, which was seen as rigorously scientific, theoretical and interpre-
tive. Gradually, however, the distinction between ethnography and ethnology
was shown to be flawed, since the ethnographic stage already invoked is-
sues of interpretation and theory – in fact, became the ‘scientific’ stage tout
court. It is the rejection of the Lévi-Straussian sequential-hierarchical dis-
tinction between ethnography and ethnology that creates Bourdieu’s
sociology: a science that has abandoned ‘ethnological’ distancing as an epis-
temology and replaced it with an emphasis on the situated, experienced and
practised (i.e., ‘ethnographic’) aspects of reality.
Let us now return to ‘Making the economic habitus’ and observe how
Bourdieu, this so-called macro-sociologist, consistently emphasizes local,
situated, ethnographic understanding in the development of the habitus con-
cept. First, and related to his critique of Lévi-Strauss, he emphasizes the
incoherence and lack of (structuralist) structure in the observed facts (2000:23;
emphasis in original):
I would need to evoke here the long series of often infinitesimal ex-
periences which made me feel (éprouver) in sensible and concrete
fashion the contingent and arbitrary character of these ordinary be-
haviours that we perform every day in the ordinary course of our
economic activities and that we experience as the most natural things
in the world….
In fact, this lived experience tends to obscure the “implicit philosophy of
labour” (ibid: 24) – the habitus (ibid.; emphasis in original):
because nothing had prepared me to understand the economy, espe-
cially my own, as a system of embodied beliefs,
I had to learn, step by step, through ethnographic observation later
corroborated by statistical analysis, the practical logic of the
precapitalist economy, at the same time as I was trying as best as I
could to figure out its grammar.
The ‘grammar’ here refers to the structural (but sociologically structural,
in the sense outlined above) properties of the object. These formal properties
remain important in his work. In Cicourel’s view, this qualifies Bourdieu as a
‘structuralist’, for he still reaches out towards “levels of theory and data which
are removed from the moment-to-moment interactional implementation of
Bourdieu the Ethnographer228
locally instantiated social organization” (Cicourel 1993:89). But more cen-
tral to Bourdieu’s concerns, I believe, is the structuring in this “system of
embodied beliefs”, a system which can only be recovered by accepting one’s
own habitus as something that needs to be put up for questioning. And this,
he consistently stresses, is a matter of ethnography, or to use his terms, of
For Bourdieu, ethnography is an epistemological issue. In its traditional,
codified form it evokes frightening questions of power and unilateral inter-
pretation; but when brought down to the level of practice (in Bourdieu’s
sense) driven by habitus, it becomes a site for constructing subjective knowl-
edge and questions about knowledge. These questions have validity for theory,
for they involve questions of contextualized understanding versus universal
or transcontextual patterns and models. It is by accepting ethnography as the
epistemological point of departure for theoretical questions that Bourdieu
can come up with theory. It is a mature position; he accepts ethnography in
its fullest sense, including the inevitable quagmires of subjectivity, bias and
‘doing-as-if’ in the field, and in that sense prefigures what later came to be
known as Critical Ethnography.
This brings me to my final point. Perhaps most importantly, Bourdieu’s
work contains what could be called an ‘ethnographic invitation’, a call to
empirically explore in micro-ethnography the structures suggested in his work,
an appeal to continue thinking theoretically while we work ethnographically
(see Cicourel 2003 for a discussion of this issue). There is no closure of the
conceptual framework in his oeuvre; on the contrary, there is a clear sugges-
tion that single cases, even if they don’t speak to the totality of the population
or the system, can speak to theory. As he states, “I believe it is possible to
enter into the singularity of an object without renouncing the ambitions of
drawing out universal propositions” (Bourdieu 1986:xi).
Herein lies the true fertility of Bourdieu’s oeuvre. He not only demon-
strates that ethnography need not be Kleinarbeit, but he also whips us into
ethnographic inspection of what he offers in the form of theoretical models
and claims – ethnography to him is the epistemological tool to arrive at theory.
He does so by emphasizing over and over again the biographical, experien-
tial ethnographic basis of his own theorizing.
2. Habitus and voice
Let us now return to our example of the asylum procedure. We see a sys-
temic (i.e., not an incidental) mismatch between the stories produced by the
applicants and the criteria for assessment used by the Belgian authorities,
expressed, for instance, in the huge rejection quota. In terms of voice, it is
clear that asylum seekers systemically fail to get their message across, to
make themselves understood. I emphasize the systemic aspects of this phe-
Jan Blommaert 229
nomenon for several reasons.
One reason is that we see rather obvious recurrent patterns in the text
trajectories of applicants’ stories, patterns in which the authorities appear to
use a particular rational, linear, detailed and ‘factually’ coherent narrative as
a model for assessing the applicants’ performance. Such models are implic-
itly sketched in guidelines for interviewers (Maryns 2004:54-56). So the text
trajectories are regulated, or to use a term more suggestive of power, they
are regimented. The authorities use a particular regime of discourse in the
transformation of performed narrative into ‘official’ text-artefacts. This has
an effect on what we define as our object of inquiry and what kinds of phenom-
ena we accept as evidence for our claims; this is therefore an epistemological
argument. We are looking for patterns that may reveal features of the discur-
Another, entailed reason is that in our analysis, we should not assume
that we can focus on the details of single cases in order to solve this larger
question of patterns in the text trajectories. In other words, a ‘pure’ dis-
course analysis in which we try to reduce the problem to ‘misunderstandings’
between applicant and interviewer will not do. Surely, situated misunder-
standings (in the sense of, e.g., conversation analysis) occur; the reasons for
such misunderstandings, however, are rarely purely situational and most of-
ten involve factors that surpass the single case. We are facing ‘frames within
frames’, to use Goffman’s (1974) terminology, and the larger frames also
matter. Such frames involve institutionality, sociolinguistics and other de-
termining aspects of the events – aspects that define the limits of what can
happen in such situated events, and that provide ‘formats’ for the single cases.
This is an analytical argument, and taken together with the epistemologi-
cal one given above, we now arrive at the following formulation of the problem.
What we need to investigate are the larger patterns of pre-structuring in
asylum procedure cases, but we need to do this ethnographically, with an
eye for the way in which larger patterns are deployed and played out during
concrete steps in the procedure.
Habituated, ‘normal’ patterns of behaviour are the switchboard through
which such larger patterns can be converted in practices that are being per-
ceived as ‘good’, ‘regular’, ‘normative’, and so on. Institutional rules are
converted into institutional routines, and such routines (habitually – i.e., re-
lated to the habitus) organize behaviour, experience and practices. As
Inghilleri (2003) demonstrates, they also organize the discourse people use
and thus provide the motives and the procedures for converting one dis-
course into another through interpreting, translation, summarizing, taking
notes, or even conversational involvement. Let us examine this process in
Such discursive conversions, as said before, are the key to the procedure,
and they happen instantaneously. The interviewers listen to the applicant’s
Bourdieu the Ethnographer230
story and question him or her on aspects of it; the answers of the appli-
cant are immediately converted into summaries, in Dutch or French,
word-processed on a computer-generated standard form, often already in-
corporating legal-categorizing terminology. Also, interviewers engage in
conversational interaction with the applicants by means of backchanneling
cues, reformulations and clarifications, summaries, and so on.
The following example illustrates such synoptic reformulating statements
(Maryns 2004:219). The interview was recorded in late 2002 in the inter-
view rooms of the Immigration Office (Dienst Vreemdelingenzaken) in
Brussels, and it involves a Flemish official interviewer (I) and a male asy-
lum applicant from Cameroon (AS). The interview was conducted in English,
and it may be useful to point out that neither of the interactants was a native
speaker of that language. The applicant argues that during an incident with
law enforcers, both his and his family’s identity documents were taken away.
However, he fails to identify what particular documents he is talking about,
and the interviewer provides a helpful gloss. This is a small negotiation – or
conflict – over terminological appropriateness, and I will mark the relevant
terms in this exchange in bold:
AS: … that my documents and inside the file that is so because
my head was down but my eye ...
I notify my (inaudible) and my different different document
y-my-my certificate of my children…
I: Identity documents and (inaudible)
AS: yeah, yeah, yeah, my-my identity documents
The interviewer, thus, provides a helpful gloss, ending the applicant’s strug-
gle to find the right term. But simultaneously, he provides a summary and a
register-specific gloss for the applicant’s words (‘identity documents’), a
regimented term that can be inserted in the official report of this interview.
The next example is a fragment of a Board of Appeal Hearing, involving
a female judge (I) who conducts the interview with another refugee from
Cameroon (AS). The court proceedings are in Dutch, and therefore an inter-
preter (T) is provided to translate the Dutch statements into English and vice
versa. We see how an interviewer (I), through an interpreter (T) challenges
the plausibility of the applicant’s (AS) claim that there are no telephones in
the town where he lived (English glosses are given in italics):
AS: yeah there is no telephone in C.
T: Er is geen [there is no]
I: =geen telefoon in C . en het is onmogelijk om om het S SDF te
contacteren … Ik bedoel we kunnen zo makkelijk SDF mensen
contacteren email of fax … en en de SDF leden zelf kunnen dat
Jan Blommaert 231
niet en en wij kunnen het in een vloek en een sakker .. ik bedoel
da’s een beetje xxx [no telephone in C. and it is impossible to to
contact the S SDF .. I mean we can so easily contact email or fax
the people of the SDF … and and the SDF members themselves
cannot do that and and we can do it in a skip and a jump I mean
that is rather (inaudible)]
Here, the judge connects the statement of the applicant with general criteria
of plausibility, derived from previously given information on the applicant’s
involvement in an opposition movement. The applicant’s claim is ‘framed’
in terms of the plausibility criteria used in the assessment of applications in
general, and this framing assumes the shape of conversational involvement
– a challenge to the previous remark and a call for more elaboration and
additional information from the applicant.
Such discursive conversions provide the routinized, habituated layer of
‘professional’ bureaucratic practices; they involve a reduction of a poten-
tially infinite range of differences to a closed set of established categories,
some of which may favour the applicant’s case while others may jeopardize
it (Inghilleri 2003). Thus, such routinized practices are intrinsically evalu-
ative and judgmental: they induce a particular kind of ‘understandability’ in
relation to the statements of the applicant, a proleptic understandability in
terms of criteria used in the next step of the procedure. They only occur
during the moment of performance (the interview) because of the existence
of ‘next steps’ in the procedure, the bureaucratic text-trajectory.
Two frames of historicity and views of factuality are combined in this
move. As for historicity, the momentary frame of performance is maintained.
As seen in the example, the interviewer often uses such converted regimented
statements as conversationally relevant turns in the interview, signalling con-
sent and understanding and often triggering affirmations from the applicant.
But another historical frame is being introduced as well, that of the next
steps in the procedure. This frame detaches what is being performed from its
moment of occurrence as well as from the particular conditions-of-use of
summarizing statements and requests for clarification during the interaction,
and propels it into a different environment with different participants, codes,
rules and norms. This is the moment where situated subjectivity becomes
decontextualized, ‘pure’ (or purified) subjectivity – where the story no longer
develops on the basis of situated understandings alone, but becomes a ‘case’
ascribed to a particular individual. And it becomes a ‘case’ through a form
of simultaneity in which situated criteria for understanding are being used
alongside established patterns of procedural understanding, and the latter
criteria are, as a rule, only known to the interviewer.
This is where the second meeting of frames comes in. The established
procedural criteria are presented and (habitually) perceived as ‘getting the
Bourdieu the Ethnographer232
facts right’ in the ‘case’. In other words, they induce a register for talking
about facts in the specific application case, and as with all registers, the act
of enregistering marks a clear distinction between what belongs to the regis-
ter and what does not (Silverstein 2003). This is a very concrete matter. The
interviewer makes on-the-spot decisions about which parts of the performed
narrative ‘belong’ to the range of factuality and which parts are redundant.
Interpreting and translation are, of course, clear cases in point. But such
decisions are often also interactionally flagged during the interviews, e.g.,
by means of the synoptic reformulations or the evaluative framings seen in
the examples above. They thus take the shape of interactionally produced,
supportive and collaborative expressions of conversational understanding.
It is again a case of simultaneity where two acts of a fundamentally different
order (here to be taken literally) are blended in one performed sequence, and
where we see the process of extraction of the story to the next steps develop
in the form of an interactional pragmatics.
What gets thrown out? Often, what disappears from the reformulations
(and thus also from the interviewer’s report of the interview) are things that
are hard to understand interactionally: situated misunderstandings, either the
effect of differential linguistic competence, accent or differences in lexis
due to particular varieties of the language used, or the effect of an overload
of detail often confusing to the interviewer (see Blommaert 2001b for ex-
amples). That is, misunderstandings that belong to the moment of performance
become absences in the next steps of the procedure; they are erased as po-
tential ‘facts’ to be considered in this case. Names of people, objects and
places, but also particular anecdotes and longish, detailed subnarratives on
the situation in the home country are common victims of such erasures. For
the applicant, these elements may contain crucial ‘facts’. Their interactional
deployment, however, either allows or compels the interviewer to strike them
from the record – that is, to strike them from ‘the story of the applicant’ that
will be judged in the next steps. It is again a proleptic move, simultaneously
deployed while satisfying interactional conventions or responding to the
interactional dynamics in the interview.
The guidelines for interviewers emphasize that the interview should not
be concluded unless the interviewer is satisfied that he or she has obtained
all the facts necessary for processing the case. It should be clear, though,
that the delineation and identification of facts relevant to the case is already
a far-reaching intervention into the story of the applicant, and that it involves
insertion of the story in a discursive regime over which the applicant has no
control. The applicant has effectively lost his or her voice in the process. For
our purpose here, the point is that the interviewer deploys institutional strat-
egies in routinized performance. That is, the institutional habitus is manifest
in ordinary, habituated, common-sense conversational practices of consid-
erable versatility, which involve the applicant as well. Both collaborate
Jan Blommaert 233
towards the performance of the narrative; but the interviewer simultaneously
extracts the situated narrative to the level of ‘a case’, proleptically shaping
the situated interaction in view of requirements of the next steps in the pro-
cedure. Production and evaluation proceed together, every productive
conversational move being also, at once, an evaluative move which detaches
the story from the situated subjectivity during the moment of performance
and moves it to the level of ‘pure’ subjectivity, i.e. the subjectivity which is
the product of power. There is no abstract institutional machinery involved
here, no faceless technical apparatus testing and clinically observing. What
we see is conversational involvement enacted in such a way that it enables
institutional judgement to proceed afterwards on the basis of a particular,
genred and regimented version of the story. The link between both worlds of
practice is the professional habitus, the embodied ‘normalcy’ of institutional
interactions, with its particular rhythm, pace, turn-taking patterns, tone and
key (Inghilleri 2003, Slembrouck 2004). It is the particular form of habitus
deployed here which accomplishes and regulates the on-the-spot meeting of
two different sets of frames, the particular kinds of simultaneity discussed
here. It accounts for this strange phenomenon in which someone may lose
voice even while he or she is using it.
The examples discussed here are by no means exceptional, to be sure. The
kind of heteroglossia described is probably just that, a macroscopic instan-
tiation of Bakhtinian heteroglossia. The advantage offered by institutional
encounters such as the asylum procedure is precisely that it offers us a mac-
roscopic and almost grotesque vantage point. Problems of voice are problems
of inequality, and as such they will occur in every environment where in-
equality is a feature of structure (Hymes 1996, Blommaert 2005). The point
here, however, is that such structured forms of inequality in language re-
quire an analytical toolkit that addresses them as features of structure, not
necessarily of intentional deployment in interaction. We need to be able to
address the fact that heteroglossia often assumes rather predictable, familiar
shapes – the fact that, as Bourdieu so brilliantly demonstrated, our thoughts
and actions seem to drift into specific directions and not in others. And this
is where habitus may be a useful concept. It offers us a perspective on the
not necessarily innocent nature of routinized behaviour, on the fact that rou-
tines may be the points where patterns of inequality enter into our everyday
behaviour, and that these patterns of inequality lead to patterning in our rou-
tines as well.5
5 It is interesting to see that Aaron Cicourel’s essays on medical reasoning, edited by
Bourdieu and Winkin (Cicourel 2002), make exactly this point. Cicourel, like Bourdieu,
Bourdieu the Ethnographer234
This is why I believe there is some virtue in reading Bourdieu from an
ethnographic viewpoint. The case he argued was a case for situated de-
ployment of structure in activities we perceive as self-evident and normal.
It explains why we shift into particular forms of performance when dis-
cussing particular topics with particular people, as well as why we wear
neckties to do so on certain occasions. Like every classic ethnographer,
from Marcel Griaule to Evans-Pritchard, Victor Turner and Clifford Geertz,
there is a danger that he should be read as a describer of abstract structure.
It is more profitable, I believe, to read his works as examples of ethnography-
as-epistemology, a way of conceiving knowledge as always requiring a
foundation in lived and experienced reality and practice. Habitus is ex-
actly the concept that should allow such a reading.
Deptartment of African Languages and Cultures, Ghent University, Rozier
44, B-9000 Ghent, Belgium. firstname.lastname@example.org
Barsky, Robert (1994) Constructing a Productive Other: Discourse Theory and
the Convention Refugee Hearing, Amsterdam: John Benjamins.
Bateson, Gregory (1958) Naven: Survey of the Problems Suggested by a Com-
posite Picture of the Culture of a New Guinea Tribe, Stanford: Stanford
Blommaert, Jan (2001a) ‘Context is/as Critique’, Critique of Anthropology 21(1):
------ (2001b) ‘Investigating Narrative Inequality: Analyzing Asylum Seekers’
Stories in Belgium’, Discourse & Society 12(4): 413-49.
------ (2005) Discourse: A Critical Introduction, Cambridge: Cambridge Uni-
Bourdieu, Pierre (1986) Distinction: A Social Critique of the Judgment of Taste,
------ (1988) Homo Academicus, Stanford: Stanford University Press.
------ (1990a) The Logic of Practice, Cambridge: Polity Press.
------ (1990b) Language and Symbolic Power, Cambridge: Polity Press.
emphasizes the ethnographic, experiential and even autobiographical grounding of his
cognitive sociology, and he also sets out to identify the structures articulated in every-
day professional routines in medical settings. Bourdieu and Winkin in their introduction
to Cicourel (ibid.) also comment on style and approach in sociology. Cicourel, accord-
ing to Bourdieu and Winkin, on the one hand represents the most rigorous, disciplined
and aridly factual kind of sociology; yet in selecting the facts he uses, he strictly con-
fines himself to data collected through ‘deep’ ethnographic experiences of his own. The
description of Cicourel’s approach is obviously reflexive of Bourdieu himself.
Jan Blommaert 235
------ (2000)‘Making the Economic Habitus: Algerian Workers Revisited’, Eth-
nography 1(1): 17-41.
------, Jean-Claude Passeron and Monique de Saint-Martin (1994) Academic Dis-
course: Linguistic Misunderstanding and Professorial Power, Stanford:
Stanford University Press.
Briggs, Charles (1997) ‘Notes on a ‘Confession’: On the Construction of Gen-
der, Sexuality, and Violence in an Infanticide Case’, Pragmatics 7(4): 519-46.
Cicourel, Aaron (1993) ‘Aspects of Structural and Processual Theories of Knowl-
edge’, in Craig Calhoun, Edward LiPuma and Moishe Postone (eds) Bourdieu:
Critical Perspectives, Cambridge: Polity Press, 89-115.
------ (2002) Le raisonnement medical: Une approche sociocognitive, edited and
with an Introduction by Pierre Bourdieu and Yves Winkin, Paris: Seuil.
------ (2003) ‘Understanding Bourdieu’s Concepts of Habitus and Symbolic Power
as Cognitive and Social Processes: Some Empirical Suggestions’, Paper pre-
sented in a panel on ‘Bourdieu and Ethnography’, International Pragmatics
Conference, Toronto July 2003.
Fabian, Johannes (1991) Time and the Work of Anthropology, Chur: Harwood
------ (1995) ‘Ethnographic Misunderstanding and the Perils of Context’, Ameri-
can Anthropologist 97(1): 41-50.
------ (2001) Anthropology with an Attitude, Stanford: Stanford University Press.
Fenigsen, Janina (ed) (2003) Misrecognition, Linguistic Awareness, and Lin-
guistic Ideologies: Ethnographies and Approaches, Special issue, Pragmatics
Foucault, Michel (2003) Abnormal. Lectures at the Collège de France 1974-
1975, New York: Picador.
Goffman, Erving (1974) Frame Analysis: An Essay on the Organization of Ex-
perience, New York: Harper & Row.
Haeri, Niloofar (1997) ‘The Reproduction of Symbolic Capital: Language, State
and Class in Egypt’, Current Anthropology 38(5): 795-816.
Heller, Monica (1999) Linguistic Minorities and Modernity, London: Longman.
------ ( 2002) Eléments d’une sociolinguistique critique, Paris: Didier.
Hymes, Dell (1996) Ethnography, Linguistics, Narrative Inequality: Towards
an Understanding of Voice, London: Taylor & Francis.
Inghilleri, Moira (2003) ‘Habitus, Field and Discourse: Interpreting as a Socially
Situated Activity’, Target 15(2): 243-68.
Jaffe, Alexandra (1999) Ideologies in Action: Language Politics on Corsica,
Berlin: Mouton de Gruyter.
Maryns, Katrijn (2004) The Asylum Speaker: An Ethnography of Language and
Communication in the Belgian Asylum Procedure, Unpublished PhD Dis-
sertation, Ghent: Ghent University.
------ and Jan Blommaert (2002) ‘Pretextuality and Pretextual Gaps: On Re/de-
fining Linguistic Inequality’, Pragmatics 12(1): 11-30.
Parlenko, Aneta and Adrian Blackledge (eds) (2003) Language Ideologies in
Multilingual Contexts. Special issue, Multilingua 21(2-3).
Sarangi, Srikant and Stef Slembrouck (1996) Language, Bureaucracy, and
Bourdieu the Ethnographer236
Social Control, London: Longman.
Silverstein, Michael (2003) ‘Indexical Order and the Dialectics of Sociolinguis-
tic Life’, Language and Communication 23: 193-229.
------ and Greg Urban (eds) (1996) Natural Histories of Discourse, Chicago:
University of Chicago Press.
Slembrouck, Stef (2004) ‘Reflexivity and the Research Interview: Habitus and
Social Class in Parents’ Accounts of Children in Public Care’, Critical Dis-
course Studies 1: 91-112.
Stroud, Christopher (2002) ‘Framing Bourdieu Socioculturally: Alternative Forms
of Linguistic Legitimacy in Postcolonial Mozambique’, Multilingua 2(2-3):
Wacquant, Loïc (2005) ‘Following Pierre Bourdieu in the Field’, Ethnography