Inspired by a seminar given at EHESS in 1982, published in different versions and for the first time in the form of a book in 1989 (Grignon & Passeron, 1983, 1985, 2015), Le Savant et le Populaire has not lost any of its relevance. The sociological questions that Jean-Claude Passeron and I wrote about have become “societal questions” that are constantly rehashed in the media; the term “populism” is increasingly popular. The relations between the dominant and the dominated classes, between the “elites” and the People, “real people” or “the man in the street,” figure more and more prominently at the centre of opinion debates. The same can be said of that which concerns the “acceptance” of immigrants from foreign cultures: should they assimilate into our culture to be integrated into society, or preserve their cultural identity (i.e., their otherness) at the risk of becoming isolated and excluded?
We are concerned by those topics, but they are not what we are most interested in: we deal with populism in sociology and literature, not in politics. These topics can add obstacles and risks for the sociologist who ventures to study popular and dominated classes and cultures. The political stakes that this study involves run the risk of waking the intellectual “who we know sleeps within every sociologist” (Grignon & Passeron 2015: 117), and of skewing our interpretations, or even our observations, by subjecting them to ideological biases and prejudices. For we sociologists, the relevance of Le Savant et le Populaire is due to the permanence of the sociological questions it addressed, which have never stopped being asked. Regarding eminently social questions, we strove to take the most sociological point of view possible by reviewing and formulating the questions that they pose to sociology. Le Savant et le Populaire is an epistemological exercise, an application of the Bachelardian vigilance to the empirical and mental operations and instruments involved in the sociology of working-class cultures. The theoretical reflections that we pursue, the concepts that we elaborate on, come from a critical inventory of the tools at our disposal; we endeavoured to find out their presuppositions and implicit orientations; what they allow and what they prevent from being seen and known, their limits, their traps, and their drifts.
“Society” and “culture:” these concepts, which are the basis for and at the centre of our analyses, are examples of the kinds of words which are both ordinary and scholarly, which mean “something to everyone” because no one knows exactly what they mean. Let us say that society is what the sociologist studies and culture is what is studied by ethnologists; in this case, it truly is the point of view that creates the object. In Le Savant et le Populaire, we decided to “limit the question to the forms of domination that are at play within a society” (18), in this case contemporary French society. This was a true sociological choice. Sociologies are most often national in scope; the sociologist studies, in the language spoken in the country, the society of which he or she is a member, to which he or she “belongs,” on the national, or on a regional or local scale. The sociologist therefore finds himself confronted with the obstacles and limitations the particularities of his object present to generalization and theorization; but this is also the case for ethnologists and historians. The sociologist is the only one, however, to enjoy advantages and suffer from the handicaps that come with proximity: on the one hand, intimate knowledge, familiarity with a culture that he belongs to, most often as a native; on the other hand, risks of ethnocentrism, prejudice tied to social position, and an inability to question the taken-for-granted aspects of the society he studies, and of its culture.
Strictly hierarchical and integrated, as much socially as culturally (compulsory education, the use of French as the common language), this society presented contradictory characteristics that made it and continue to make it particularly interesting. But the social composition of the working classes has changed and the rereading of Le Savant et le Populaire invites us to take these changes into account. The characteristics of the dominated cultures and their relationship with the dominant cultures depend on the properties of the groups that support them. Since the 1980s, the size of the traditional popular classes has diminished; this is especially the case with farmers and, to a lesser extent, workers (who still in 2015 account for more than one fifth of the labour force); this is also true of workers of peasant origin. At the same time, the population of more or less acculturated immigrants, bringing with them many foreign cultures, which are set apart by two essential elements, language and religion, has grown. To what extent can we consider these “faraway cultures,” once they have drawn closer and become part of this society, to be “other cultures” in the same way as pre-colonial cultures? Can we apply ethnographic cultural relativism to them without difficulty and without risk? Both autonomous and dominated, these cultures constitute a typical case that the concepts elaborated in Le Savant et le Populaire, such as autonomy and heteronomy, should help to disentangle the ambivalence.
Cultural relativism is as necessary a passage for the ethnologist of popular cultures, who must maintain his distance from his culture of origin to draw nearer to the one he studies, as it is for the sociologist when he attempts to consider these cultures as autonomous. But we run the risk of passing from cultural relativism to a cognitive relativism which, in its extreme form, is based on a radical scepticism for which it is impossible to distinguish what is true from what is false because truth does not exist1. Relativism is fashionable; it applies to intellectual fashion victims, including sociologists through sociologism, which continually tries to go further in the reduction of every element of every culture to arbitrary constructions, all equally deprived of necessity and practical and technical efficiency, that need only be deconstructed to be explained.
Like “culture” and “society,” “domination” is one of those common but poorly defined concepts that we are obliged to use because they usually denote the questions and facts that we are concerned with. We strove to pinpoint their meaning not by looking for words that might explain their nature, but by collecting the things and facts—the “mechanisms’—to which they correspond and that they denote2. “Sociology of culture has to first break the monolith represented by the concept of dominated culture” (Grignon & Passeron 2015: 140). This is a task that is unequally difficult to accomplish; the notion of alternation allows us to distinguish between circumstances, fields, interactions, etc., and invites us to find empirical methods with which to do so. Conversely, the notion of ambivalence leaves the door open for interpretation, by suggesting that the most autonomous working class cultures are still, to some degree, dominated—that there is still some of the Same in the Other (and vice versa). “The mischievousness of the interpretative demon” (147) knows no bounds: the most factual descriptions, the most realistic and objective observations are open to interpretation. We interpret them as soon as we analyse them, seek to explain them, comment on them, present them; sociology of culture thereby comes dangerously close to “theories of suspicion,” to ideologies and to pseudo-science.
The definition of concepts leads back to the question of the relationship between working-class languages and the language in which the sociologist transcribes them. How can he represent the way subjects speak? To make their speech intelligible, one must translate it; but one cannot translate it without altering it, without causing it to lose its original meaning. George Sand very clearly formulated this aporia (291, note 5). The sociologist thinks and expresses himself spontaneously, naturally, in this dominant form of natural language that is the language of the well-read, the language of his culture, his mother tongue, the language in which he has been trained, that of his professional milieu and of the cultured audience to which his writings are directed. While popular speeches, vernaculars, and “lingoes,” are most often oral, sociology is written. The sociologist cannot do without the resources of “literary language which is doubly written” (291), especially its evocative power which helps to elaborate his concepts from observation and which, by tying these concepts to the imagination, allows what they represent to be imagined, and thus allows us to begin to know what they mean; take for example those accounts that are both suggestive and meaningful, which make the reader think by giving him something to see, cf. the works of Richard Hoggart, Marcel Maget, or Marcel Granet (Grignon, 2001). But this evocative power favours interpretative drifts, particularly those which are tied to the free association of words, images and ideas. It is therefore necessary to control it by trying to combine the conflicting principles and demands of literary and scientific writings. While the evocative power of literary language is based on connotation, scientific language depends on denotation. The meaning of a scientific word depends on its deictic capacity; we know what a scientific word means when we know precisely, without ambiguity, what it refers to (Servien, 1935). We will therefore allow ourselves to play with the connotations only for words whose denotation is known. We will be particularly careful regarding the connotations evoked by scholarly terms at every step of the research and for all of those who participate in it, including the interviewer and the interviewees, the sociologist who writes, and the reader whose expectations and reactions he anticipates.
As soon as one makes use of the resources and processes of literary writing, one risks giving in to the influence and attraction of literature. It is difficult for the sociologist who writes not to feel like or want to become an author; just as is the case with the intellectual, an author sleeps within the sociologist. Unless he stops attempting to paint a picture, the sociologist, like the historian, risks foregoing explaining, mixing art with analysis, reality with fiction, sacrificing exactitude to the illusion of truth, as Taine says (1858). When the characteristics that define them have been determined by their evocative power, the types that sociology tries to create approximate literary types and become portraits. Sociology therefore risks taking on literary representations of the worker, the farmer, or the immigrant, as folkloric and populist as they are miserabilist. “The author’s narcissism offers an impregnable bastion to class ethnocentrism” (Grignon & Passeron 2015: 313).
Translation and writing difficulties encourage us to pursue the epistemological exercise of Le Savant et le Populaire3. The languages used by the sciences, in which they think and express, are a determinative criterion of their differentiation and their categorization, if not of their hierarchy. Each science, each specialization, can be distinguished by its own words. In this respect, scientific languages have much in common with dialects, professional jargons, regional languages, local patois, etc. As particular, scholarly, and incomprehensible for the layman as they may be, they are variants of natural language. Their translation into ordinary language, for reasons of vulgarisation or instruction, poses the same problems as the translation of a speech or text into a “foreign” language. But the differences between scientific languages cannot be reduced to lexical differences. The principal, fundamental, opposition divides the sciences that think in mathematics and those that think in natural language: on one hand, the sciences of matter, which explain by formulating universal laws that allow prediction; on the other hand, the human sciences, which explain by finding causes, and reconstituting causal sequences that take place in an irreversible time (between the two is biological science, which is split between physics and history—molecular biology versus evolutionary biology or the biology of organisms).
In this regard, sociology belongs to the same family, the same subcategory of science as ethnology and history, that of collective human science. But, at the risk of forgetting that popular cultures are also dominated cultures, the sociologist who wants to be an ethnologist must remain a sociologist. That is why he must combine quantitative and qualitative methods: collecting and processing statistical data in order to establish regularities for lack of laws (the sociologist therefore uses mathematics, but as a tool and not as a language); explaining these regularities by reconstructing (through interviews, observations, case studies, etc.) the processes at the end of which the statistical relationships between the cause (explanatory variables) and effect (explained variables) take place (or not). By reconstructing individual stories, the sociologist gives himself the means of knowing how social determinisms act, but also why they sometimes do not act, and how some exceptional individuals can escape their social destiny. This is the case with Richard Hoggart; the story of his life allows us to know why the mechanisms of reproduction did not work, and, at the same time, to better understand how they function (Hoggart 1988, translated in French: 1991, 2013). But, for the sociologist, exceptional individuals are not unique; by comparing their respective stories, he can group them according to specific traits they share, putting together case clusters, thereby returning to the collective from the individual.