Reading guide to: Bourdieu P. 'Understanding', in Bourdieu P. et al.  (1999) The Weight of the World: Social Suffering in Contemporary Society, Cambridge: Polity Press.

Dave Harris

This is a difficult piece to summarize because it is very densely argued and frequently illustrated with examples from the substance of the book itself. I have tried to organize my notes thematically on this occasion, but the chapter itself is heavy with insight and it will reward careful reading.

Problems with existing methods of research (my headings throughout)

Theoretical reflections on methods are not very helpful [including Habermas's?], and instead it is best to take actual examples of interactions between researchers and subjects. Writings on methodology don't take us very far either, since they often attempt to justify standardized procedures: these fail to grasp the  'almost infinitely subtle strategies that social agents deploy in the ordinary conduct of their lives' (607).

Scientistic procedures are inadequate, but so are 'the anti-scientific caveats of the advocates of mystic union' (607). A note on page 608 makes the point that both quantitative and qualitative methods constrain the interaction that can take place, and this is often unrealized by exponents -- even  'the ethnomethodologists, whose objectivist view of the social world leads them to ignore the effects exerted by objective structures not just on the interactions they record and analyze... but also on their own interaction with those who are subjected to their observation or questioning'.

The distortions of research
Research is always an intrusion into social interaction. Researchers and respondents are likely to have different objects, and this distance must be understood as having effects. If researchers occupy a higher place in the social hierarchy, possessing different kinds or amounts of linguistic capital, these effects may be amplified. Thus there is always the potential for symbolic violence.

Preferred approaches and procedures
Sociologists must be aware of the distortions introduced by the research relationship. This requires a deep kind of reflexivity, which allows sociologists to monitor interviews on the spot, and to improvise hypotheses. This cannot be produced by a procedure as much as a  'sociological  "feel" or sociological "eye"' (608). Researchers must gain knowledge of their own presuppositions and reflect on the effects of the research itself. All research involves constructions of knowledge, and it is essential to become aware of the work of construction and to master its effects.

What is required is  'active and methodical and listening, as far removed from the pure laissez-faire of the non-directed interview as from the interventionism of the questionnaire' (609). It requires interviewers to make a special effort to understand the  'singularity of a particular life history', but combined with  'methodical construction, founded on the knowledge of the objective conditions common to an entire social category' (609). An aside on 'intrusion' on page 610 notes how difficult it is to pursue such active listening even in ordinary conversations, while researchers can often produce 'complete amazement' followed by 'a polite response' to an inappropriate question.

This requires careful attention both to the conscious elements of interaction, such as those signs designed to encourage collaboration, and to the very structure of the relationship. Thus interviewers were encouraged to choose respondents from people they knew or from people to whom they would be introduced, since  'social proximity and familiarity provide two of the conditions of  "non-violent"  communication' (610). Such proximity encourages lower levels of symbolic threat -- 'that subjective reasoning [will not be] reduced to objective causes'-- and permits constant interchanges of verbal and non-verbal signs which show  'immediate and continuously confirmed agreement' (610). It is  'favourable to plain speaking' (612) . It might even be possible to encourage role play (that is, trying to pass as a member of a particular group). However, a promising approach, based on work by Labov, was to train as researchers people who already had familiar access to the sorts of respondents required. In such circumstances, interviewers and respondents are more prepared to tolerate symbolically insensitive questions, especially ' brutally objectifying questions': both participants are able to see the threat, and both share the risk of being objectified.

Constructing an interpretation
However it is necessary to move beyond just 'collecting "natural discourse"'to begin to analyze it. Here we must recognize that we are constructing discourses scientifically  'in such a way that [they] yield the elements necessary for [their] own explanation' (611). Non-professional researchers find this especially demanding, and many were unable to do it -- their interviews were just  'dropped from the book' (612).  [A bit of academic symbolic violence seems to have been practised here then. This subsequent stage, deciding how to use the data that has been collected reverts to an unequal power relationship between professional and non-professional interviewers -- at this stage there is no shared ownership. This double level has also been identified as crucial in modifying the apparently fully participatory nature of  British 'action research']. As a note on page 612 points out, research interviews always run the risk of turning into two extremes:  'total overlap between investigator and respondents, when nothing can be said because, since nothing can be questioned, everything goes without saying; and total divergence, where understanding and trust would become impossible'.

Other kinds of solidarities might be developed between researchers and respondents, not involving immediate proximity but equally acting as 'guarantees of sympathetic comprehension relations or childhood friendships ... affinities between women [which helped] them to overcome the obstacles linked differences of social situation -- in particular the fear of patronizing class attitudes which, when the sociologist is perceived as socially superior, is often added to the very general, if not universal, fear of being turned into an object' (612).

However, social distance is likely to remain between interviewers and respondents. Respondents are not always the underdogs, but can sometimes master the situation for themselves, of course. Considerable effort is often required to construct research interviews as 'natural', which conceals the effects of social distance. Some sociologists are able to signal (by 'tone' and by questions asked, the presentation and conduct of the interview) that they want to permit respondents to  'legitimately be themselves' despite the distance between them. Researchers are aware that they do not simply share the same point of view, but are still capable of  'mentally putting themselves in [the respondent's] place' (613). This is not merely a matter of 'projection' or empathy, but can only arise from a proper grasp of the social relations involved, such as  'the circumstances of life and the social mechanisms that affect the entire category to which any individual belongs' (613). This helps the respondent makes sense of the interview and the social situation out of which responses emerge. Interviewers are required to have considerable knowledge of the subject and of the social relations involved, far more than is required by more routine research.

This knowledge also helps 'constant improvisation of pertinent questions, genuine hypotheses' aimed at more complete revelations (613). However, researchers must still attend to others and display a  'self-abnegation and openness rarely encountered in every day life' (614). It is easy to be inattentive and to accept 'immediate half understanding'. The process is perhaps best understood as a form of spiritual conversion, a 'forgetfulness of self... a true conversion of the way we look at other people... if the capacity to take that person and understand and just as they are in their distinctive necessity... a sort of intellectual love' (614).

These preconditions permits an 'extra-ordinary discourse, which might never have been spoken, but which was already there, merely awaiting the conditions for its actualization' (614). A note suggests that we should 'aim to propose and not impose, to formulate suggestions sometimes explicitly presented as such... and intended to offer multiple, open-ended continuations to the interviewee's argument, to their hesitations or searchings for appropriate expression' (614 - 15). Respondents often see themselves as being offered a unique opportunity to explain themselves. In doing so, as  'an induced and accompanied self analysis', they can even experience a  'joy in expression' (615). Sometimes this leads to the expression of  'experiences and thoughts long kept unsaid or repressed' (615). However, an aside explores some further complexities:

(a) interviewees can take a chance to present themselves in the best light, sometimes censoring their opinions
(b) interviewees may construct a  'false, collusive objectification' of themselves, seemingly analyzing themselves but  'without questioning anything essential' (616)
(c) researchers can be swept along, and engage in  'a form of intellectual narcissism which may combine with or hide within a populist sense of wonder', losing their critical penetration in favour of a recognition of their own conceptions of disadvantaged groups  (616).
(d) respondents can take the interview over, asking and answering questions for themselves --  'the researcher is taken in by the "authenticity"  of the respondent's testimony... the respondent plays her expected part' (617), and both get seduced by what seems like the literary value of the speech.

Researchers should submit to the data. Paradoxically, this requires  'an act of construction' in order to properly hear what is being said, 'how to read in... words the structure of the objective relations, present and past', such as the educational establishments attended and their effects  (618). This does not reduce the individuality of the respondent but attempts to explain him or her as a  'singular complexity' (618). More is involved than simply collecting conversations and studying their dynamics -- interest lies in the  'invisible structures that organize' such interactions  (618).

Interviewers cannot eliminate themselves from the situation and just describe anyway. They need to adopt a  'realist construction', investigating underlying [sociological?] realities, and attempting to understand the [subjective?] realities as experienced by the respondents. This is much more fruitful than techniques like those employed in opinion polls, which appear neutral, but which impose a problematic nevertheless --'Their forced, artificial questions produce out of nowhere the artifacts they believe they are recording'  (619).  'Opinions' are volatile and can take many forms of expression, thus it is easy to impose a problematic on them. [Indeed, we know that this is considered to be one of the skills of researchers, or, indeed academics -- to impose some framework of meaning].

The correct way is not to leave things alone, because this just leaves accepted beliefs unchallenged --  'the terrain is then free for pre-constructions or for the automatic effects of social mechanisms at work in even the most elementary scientific operations (conception and formulation of questions, definition of categories for coding, etc)' (620). Instead, active criticism of common-sense is required to take on common representations, including those in the media, which interpret adversely the experience of the disadvantaged. Ordinary people do not have access to social science, nor do they always say what they mean. By contrast sociology is in a position to challenge reconstructions and presuppositions, and the apparent spontaneity of opinion. Thus in researching hostility to foreigners, especially among those who do not know any, sociologists can understand it as 'displacement', accounting for contradictory experiences among the petty bourgeoisie, for example [farmers and small shopkeepers are the specific cases given here]  (621). These social contradictions are  'The real basis of the discontent and dissatisfaction expressed... in this hostility... people are... both unaware of  [them] and, in another sense, know them better than anyone' (621). The role of the sociologist here is 'like a midwife' [the process is compared to psychoanalysis earlier], but this is again a craft rather than an abstract way of knowing, following a  'real  "disposition to pursue truth"' (621), which often leads to improvisation.

Transcription and analysis also offer problems. There is no neutral way to transcribe and avoid interpretation. The style chosen in this book reflects  'the pragmatics of writing' (622), illustrating 'sociologically pertinent features' by the deliberate use of subheadings, for example. There is a need to try to be faithful to the contents of the interview, while retaining an interest in readability, which forbids, for example, describing intonation, rhythm, voice, gesture and so on -- and a note on page 622 reminds us that  'irony, which is often the product of a deliberate discrepancy between body and verbal symbolism or between different levels of the verbal message, is almost inevitably lost in transcription. And the same goes for the ambiguities, double meanings, uncertainty and vagueness so characteristic of oral language'.  Transcription is literally rewriting. The aim is to offer true self-expression rather than literal speech -- for example to manage hesitations, interruptions, digressions, ambiguities, references to concrete situations and so on. These often have to be omitted, since they can make transcriptions unreadable.

These transcripts can  'have the effect of a revelation, especially for those who share some general characteristics with the speaker' (623). [This certainly describes the effect on me of some of the passages in the book]. Such emotional effects  'can produce the shifts in thinking in seeing that are often the precondition for comprehension... but... also generate ambiguity, even confusion, in symbolic effects' (623). For example, it is difficult to report racist remarks without seeming to legitimate racism, or offer personal descriptions (for example of a hairstyle) without referring to personal aesthetics.

The reader
The dilemmas affect all those who intrude by publishing -- to balance the need to influence the reader while permitting the 'risks [in] allowing people free play in the game of reading, that is, in the spontaneous ...constructions each reader necessarily puts on things read. This game is particularly dangerous when applied to texts which were not written and which, for this reason, are not protected in advance against feared or rejected readings, above all, when applied to the words of speakers to do not speak like books [and thus] have every likelihood of not finding favour in the eyes of most readers, even those with the best intentions' (623 - 4). The team noticed that some non-specialist readers read the interviews merely as confidences or gossip, and took the opportunity to socially differentiate themselves.

For this reason it was necessary to intersperse the transcripts with headings, subheadings and introductory sections to enable readers to reconstruct the writers' stance. It is essential to get people to read the transcripts with  'sustained, receptive attention', as if they were philosophical or literary texts (624).  [A note on page 624 reminds us how difficult this is, since we commonly mix together readings of texts and judgements about the social standing of the writer --'Nothing escapes the logic of the academic unconscious which guides this a priori distribution of respect or indifference', and less specialist readers have even less chance to escape prejudices]. It is necessary to try to persuade readers to develop a suitable attitude towards the words they are to read  [the same one as is required for interviewers, in effect]. This is particularly difficult in doing justice both to subjects and to readers, and in particular trying to be objective and yet avoiding the effect of putting the subject  'in the dock' (625). Again, attention to detail is required to maintain the point of view of the writing, e.g. refusing to  'slide unconsciously' from the personal voice to the voice of science  (625).

Sociologists have to manage their own peculiar point of view, which is to take the point of view of others and resituate it within social space -- this is possible only by remaining objective about all possible points of view. This in turn requires sociologists to objectify themselves, while remaining aware of their own place in the social world and trying to reconstruct the point of view of others in other places, 'to understand that if they were in their shoes they would doubtless be and think just like them' (626).

More Bourdieu here