CAIRN-INT.INFO : International Edition

1The method of collective interviews allows us to take a fresh look at how we approach the “classic” sociological question of relationships between individuals and the collectivities to which they belong from an empirical point of view. In the most contemporary uses of this method in the social sciences, these “two inseparable levels of the human universe” [1] are subject to varying degrees of scrutiny. For some, the discussion group per se constitutes the most pertinent unit of analysis. It is a matter of finding the “shared frames” between people who have certain characteristics in common (socioprofessional categories, sex, age, etc.). [2] Comparison between different groups, [3] which generally bring together people whose social relationships are well-established, constitutes the principal analytical operation. It is often used to create a research setting judged to be more favorable for understanding the behaviors of those with the least cultural capital. [4] Other researchers concentrate their attention more on individuals to understand how they behave in a given situation of interaction. [5] While comparison between groups is not ignored, it is less central. These analyses privilege interactional dynamics playing out within groups that may be socially heterogeneous or may bring together people who do not know each other. These interactions, in particular in the case of disagreements, supposedly allow us to understand more precisely what research participants have in mind when they give their opinion. In defending their points of view, they clarify their arguments, their values, and their belief systems. [6]

2In an attempt to go beyond this alternative, it seems promising to draw on the reflexive work begun by Jocelyn A. Hollander, who aims to understand the ways in which the context of collective interviews shapes the modalities and the nature of the verbal exchanges recorded. She suggests considering together on the one hand what the group produces, and on the other hand what individuals may produce within this group. From here, the methodological stakes consist “not in getting from the participants in the collective interview their ‘truth’, but rather to understand and analyze the multiple and complex interactional forces that lead participants to share certain truths, to hold back others, and to elaborate new versions of reality in a given context”. [7] These suggestions are based on an exploratory study about perceptions of violence against women, a theme which is intensely personal and that reveals itself to be particularly sensitive to the discussion setting. During this research, Hollander conducted individual interviews with people who had participated in her collective interviews. These participants explained that their comments were restricted by the group setting and then spoke about the theme under discussion in relatively different terms. Hollander thus posits that the space of what is welcome, important, or risky, even impossible, to say varies as a function of the quality of the social relationships (“sociability”) of the participants and of the degree of familiarity between them (members of the same family or work colleagues, for example), of the degree to which participants’ social statuses are homogeneous (for example, groups that are mixed or not), of the theme of the conversation (controversial or, inversely, likely to bring people together), but also of the place in which the conversation takes place. In other words, the analysis of collective interviews would benefit from considering as indivisible the “shared frames” at the level of the group and the types of interaction at the heart of these collective situations.

3To extend this perspective, which aims not to separate the individual from the collective, we must emphasize that while the individual interview proves itself to be more favorable for gathering biographical elements and for the detailed description of trajectories over the course of which the habitus takes shape, collective interviews allow us to understand how these social dispositions are used within the framework of a discussion. They provide the opportunity to observe people as they converse and express their conceptions of the world, and how they perceive, classify and represent things. Collective interviews thus reveal a specific part of the system of dispositions that make up the habitus, the part that corresponds to its cognitive and interpretive dimensions, and that we might term a schema. Acquired and elaborated through layers of social experiences, schemas reveal themselves as being inseparably both individual and collective. Defined by Alfredo Joignant as “cognitive predispositions”, [8] they are characterized by their durability and their transferability and distinguish themselves in this way from opinions. Opinions, intended to be spoken publicly, are thus perceived as being more sensitive to the context as well as to the theme of a conversation. These opinions are often the result of schemas that, as social dispositions, make up the habitus. Indeed Pierre Bourdieu uses the concepts of “schema” and “disposition” in a relatively interchangeable fashion. He uses the word “schema” essentially to evoke modes of classifying and of perceiving the world. Thus, just like dispositions, schemas must be studied in context since their transferability always includes a measure of adjustment. [9] In his cumulative and critical assessment of the debates surrounding “schema theory”, Joignant underscores that, for cognitivists, “the social functions of schemas are not all that different from those fulfilled by similar sociological notions (such as Bourdieu’s habitus)”. [10] It remains true that research about the cognitive structures of agents tends, for the most part, to be asociological. [11] Whether they are supporters or detractors of schema theory, researchers in social psychology rarely take into account the social anchorings and uses of schemas in the sense that their work is generally based on experimental or statistical research into “the life of the mind” [12] of a generic individual. I aim to complement these cognitive approaches by emphasizing, in my use of the notion of the schema, how it is at the same time socially constructed [13] and also generates perceptions of the world. [14]

4Thus, can collective interviews as a methodology contribute to sociological knowledge concerning schemas of understanding of a public problem? This article claims that collective interviews, to the extent that they allow for the articulation of comparative analyses and interactional approaches, enrich the understanding of mechanisms through which schemas are deployed, adjusted, and maintained. It calls for a balance between inter- and intra-group analyses, with the first being situated at the level of the collective and the second at the level of the individual, in order to shed light on the dialectic of schemas. Schemas generate interpretive repertoires that are of varying substance on the one hand, and regular in their social distribution on the other. Moving forward, collective interviews encourage us to interrogate the impact of sociabilities [15] on the mobilization of schemas.

5To test these propositions that are inseparably both methodological and theoretical, I will draw on a dissertation project in which twelve conversations on climate change were recorded, which brought together people who knew each other (the composition of the groups is presented in the annex). [16] The comparative analysis of these conversations brings out socially situated ways of understanding the climate problem in particular and public problems in general. But, for reasons linked to the study’s protocol, to the profiles of the study participants, and to the public construction of the problem under discussion, these conversations do not lend themselves to an analysis of sharply contrasting positions. In our corpus, the processes of collective negotiation of opinions and of judgments operate essentially in cooperative mode and are rarely conflictual. [17] It is thus important to pay special attention to those interactions that are falsely “harmless”, arising less from disagreement than from agreement, but during which a space of what can be said is delineated and schemas are mobilized.

6The first part of this article interrogates the subtle mechanisms that govern the mobilization of schemas within discussions. Once they are in agreement, notably on what is worth their attention and thus worth being discussed, the participants share visions of the world. These interactions, although rarely conflictual, are part of participants’ maintenance of their interpretive principles. When collective interviews reveal different situations of interaction they lead us to deeper contextualization of these ways of thinking. Taking a more exploratory perspective, the second part thus proposes ways forward for analyzing how far sociabilities impact the conditions of expression of schemas. Faced with the difficulties that there can be in objectivizing the influence of sociabilities, this section suggests in particular the importance of taking into serious consideration the degree of conversational routine among the participants. [18]

The silent mobilization of schemas in conversations

7In order to discuss collective interviews as a methodology as well as the data that they enabled me to collect, we must first return to the conditions of production of this study. One of the initial objectives was to evaluate the impact of routine sociabilities on how people understood the climate question. It thus seemed preferable to obtain exchanges which drew on common experiences. This strategy implied bringing together individuals who had known each other for several years, who had shared histories, who had faced challenges of the same type and among whom the distribution of roles was well established. In fact, “the more we have groups of everyday peers, the more we have a chance of recording ordinary representations”; [19] in other words, representations that are shared by participants and that are the most faithful to the reality of their daily interactions.

A model seeking plausible sociabilities

8In order to obtain groups of participants who knew each other, directions were given to intermediaries [20] to solicit from their entourage people “who are used to discussing things together” and invite them to come and talk about “social questions”. Bringing together people who know each other also represents a means of constituting groups that are more likely to be homogeneous from the perspective of social conditions of existence. In order to proceed to a comparative analysis to appreciate the effect of social variables, the ensemble of the groups must however form a contrasted sample. [21] This imperative is reinforced here by the fact that the analysis of second-hand statistical data attests to the discriminating role of economic and cultural resources in the social distribution of practical relationships to the climate problem. [22] I thus ensured that I contacted intermediaries with unequal amounts of economic and cultural capital. While other variables such as age, sex, or where people live probably have an impact on the ways in which people conceive of the climate problem, it proved to be a delicate matter to intersect all of these criteria when working on a qualitative sample of fifty participants. I thus primarily considered education levels, income, and professional and socioprofessional categories during recruitment as well as during the analysis.

9Just as I sought to constitute groups which it was hoped would reproduce likely interactions, so the whole research design was concerned with a quest for realism. The discussions generally took place at the home of one of the participants or in places where the participants habitually met, such as in the offices of an association. The presence of familiar landmarks and the establishment of a “relaxed” context aimed to show the participants that they could behave as normal, in other words without paying too much attention to adjusting their dispositions to the situation. In the same vein, the desire to avoid constituting groups of more than six people is explained by the improbable character of conversations where there are more speakers. This quest for realism should nevertheless not be confused with the demands of an ethnographic study, for example. I did not seek to make the research protocol disappear entirely – so a researcher was present, plus a means of recording (including a video camera) and remained visible, which could have affected how the participants positioned themselves. Researchers should also be conscious of the exceptional nature of a discussion this long (an hour and a half on average) on a given theme with a given group of people. [23]

10This desire to remain close to socially plausible conditions of conversation also explains the choice not to use stimuli to activate the discussions. Collective interviews that study the “receptions” of media messages are often embellished with audio, video, or paper prompts to which the participants must react. What is observed then corresponds to a short-term negotiation of a specific message. By not privileging one “text” over others, what was at stake here was more to capture “secondary receptions” [24] in which different symbolic resources may be mobilized by the participants. Hence the interview guide was intentionally open and imposed little, in order to allow the participants to appropriate the theme “in their own way”. The interviews were organized in three phases. The participants were first invited to react to the introduction of the climate theme. The discussions were introduced as follows: “So today, I would like us to talk about climate change.” The questions that followed essentially sought to see if they considered climate change as a problem, and, if so, why and how they qualified this problem: “But then, is climate change a problem, do you think? If so, what do you think the problem is?” After a few questions on how participants kept themselves informed on this mediatized issue, [25] questioning then focused on causal responsibilities: “And, in your view, what causes climate change? Do you have an idea of what we should do to combat climate change?” Finally, the third part of the interview gathered together the opinions of the participants on the different actors in the problem, whether they were scientists, associations, journalists, political leaders, or even individuals themselves (“What about you, do you try to reduce your energy consumption?”). As for the directions for moderating the group, [26] the strategy was not so much to encourage debate, by for example pointing up divisions between the participants, as allowing the conversations to follow their “natural” course. Collective interviews, moreover, constitute a different social situation from that of an individual interview, in particular in relation to the position of the researcher. In the latter case, the anticipation of (assumed) judgments by the researcher is undoubtedly more likely to affect the attitude and comments of the participant, while the collective dimension tends to dilute the scholarly and intrusive nature of the interaction. [27]

11Finally, a word on the coding of the data. In order to more easily identify when people spoke or be able to isolate the contributions of one participant in particular, it proved useful to transcribe the conversations (all of which were video-taped) in Excel. Without being a decisive factor, using this software turned out to be advantageous for the analysis of interactions. At low cost, it enabled us to sort or qualify certain passages, and thus to usefully regroup data and zoom in on particular elements to study the dynamics of conversations and the ways in which different participants took part in the discussions. Coding in Excel facilitated, for example, the identification of sequences where the conversation moved away from the researcher. [28] Positing that the group appropriates the problem during their interactions, the analysis was concentrated on those slices of conversation during which the researcher does not speak. One of our hypotheses was that the themes that provoke exchanges between participants correspond to what interests them and brings them together. This strategy thus allowed us – at the cost of a long and slow process in order to make comparisons – to identify who is speaking (about what, why, etc.), and to better understand how the climate problem is collectively understood in the different social milieus studied, insofar as as they are “represented” in the groups brought together.

Table 1

A corpus and contrasted conversations[29]

Table 1
Group identification Number of participants Conversational dynamic Characteristic modes of approaching the climate problem within this conversation Prevalence of conflictual interactions2 Working classes Working-class neighbors 6 Three participants out of six sustain the conversation: the couple who is hosting and the participant with the highest level of education Based on what is said in the media and with an interest in savings to be made on domestic energy Very weak Working-class family 6 A young adult couple tend to do all the talking. Unobtrusive comments and little interaction from the other participants Essentially based on personal experience (practical observations, work etc.) Occasional conflict between the young couple Upwardly mobile working-class family 3 Balanced, even if in the background arguments are exchanged, mostly between the father and the daughter Material considerations/the father occasionally “politicizes” the problem based on his passion for cars (referring to laws, for example) Occasionally between father and daughter 1. Intermediate social positions Intermediate 50-somethings 4 Only one person remains a little on the sidelines during the discussion The most educationally qualified person offers more political interpretations. Otherwise practical, everyday and local notions are foremost Very weak Intermediate, 25 years old 4 Balanced: two couples who are used to discussing things together On the basis of regular reports on the weather forecast and in relation to domestic tasks. One participant distinguishes himself by referencing the Al Gore film which he is alone in having seen Very weak Retired people 6 The two participants with the lowest educational qualifications take little part in the conversation Many personal anecdotes; numerous digressions Very weak Teachers 3 One participant is clearly more active than the other two. As a teacher of agronomy he presents himself as an expert on the matter On the basis of environmental issues, which also refer back to lifestyle questions Weak. The most knowledgeable person adopts engaged viewpoints Upper social classes Students at the Sorbonne 4 Balanced: careful to listen to others On a universalizing and “cultivated” basis Sensitive, as in a televised debate Young educated Parisians 4 Two participants contribute more than the others, feeling themselves more qualified to do so: one in relation to the subject matter, the other – a journalist – in relation to the requirement to converse By bringing up travel; a discussion of solutions to the problem; the participants see themselves as actors Very weak Young Parisian lawyers 5 Balanced; enthusiasm for the exercise Weak knowledge of the topic, extended to the environment in general. Political and moral interpretations of the issues Very weak Parisian intellectuals 3 One participant, an American women, says less Consideration of the scientific, political and moral issues within the problem Very weak Parisian medical professionals 2 Balanced Political and scientific approach Sensitive in that these two participants know each other very well but do not share the same political preferences

A corpus and contrasted conversations[29]

*NB: To facilitate the identification of the groups, they are labeled as a function of social properties that characterize them in relation to the other groups. These labels nevertheless remain arbitrary. The methodological annex provides a more detailed presentation of the groups. Moreover, the tripartite presentation is open to discussion and certain groups (such as “teachers”) are somewhat tangential. The classification of the groups into one or other of the three categories relies on taking into account an ensemble of elements, including professional and socioprofessional category, their trajectory, their level of income, or even where they live. This presentation primarily serves to help the reader find landmarks within the sample.

Conversations with few conflicts

12The twelve recorded conversations about the climate problem confirm that “the real politicization of discussions – in the sense that the interlocutors recognize, regarding a question of collective interest, the existence of divergent points of view – is indeed a rare phenomenon”. [30] The majority of verbal exchanges that we collected do not allow an explanation of positions as ordered around previously existing divisions. Our participants do not approach the discussion as an experiment during which they must defend controversial points of view, and rely on alliances or positions which are risky because they reveal situated ideological and social affiliations. Three elements contribute to explaining this first observation.

13A first reason is presumably linked to how the climate question is publicly evaluated. In the way it has been shaped in the general media in France since the end of the 1990s, this problem belongs more to those causes which are “without adversaries” [31] than to issues that are debated and that split opinions. Conflicting understandings of this theme are thus easily avoidable, and probably judged as illegitimate given the amount of emphasis the media gives to the unifying dimensions of the problem. [32] Conversations regarding other public problems that are the objects of more explicit political codification – such as insecurity; the reform of systems of retirement, health, unemployment benefits, or education; the regulation of immigration, or even the regulation of genetically modified organisms – are doubtless more likely to take a more conflictual turn. However, explaining the “smoothness” of the conversations collected in terms of this one effect of mediatization would amount to attributing significant weight to the media. A second reason is thus connected to the dispositions of the 50 participants who make up the groups. We did not create groups which had a specific link to the climate question (for example a group of environmental activists, tourism professionals or automobile professionals). This choice is explained by the hypotheses of this research. After having studied, in terms of the media offerings, the overall output of reporting focused on climate change on French evening television news shows on the channels TF1 and France 2 between 1997 and 2006 – in other words on the channels the most accessible to the largest number of people – we wanted to survey groups whose sociabilities were not based on a particular relationship to the climate question. [33] While some of our participants expressed a strong awareness of the environment, none believed themselves to be engaged in the environmental cause. They were thus not likely to radicalize the terms of the exchanges and they accepted the unifying character of the theme. At most they wondered about the ephemeral and superficial dimension of the “environmental trend”. It is probable that groups whose members know each other via a specific relationship to climate issues would have produced discussions which adopted more entrenched positions. [34] Finally, the moderation strategy sought more to let proceedings be guided by the participants than to orient them toward discussions where they would have to expose, defend, or refine their points of view. The guiding idea behind this mechanism was to let the different groups appropriate the theme of the discussions. This implied not imposing one aspect of the problem on them more than another; but instead observing which issues they discussed, how they discussed them, and what their points of reference were. It was thus not appropriate to seek to create conflict within the exchanges in order to see how the groups themselves framed the problem.

14These three elements of the study converge and explain in large part why the exchanges rarely included participants adopting positions but instead proceed in the style of “it goes without saying”. Arguments were rare and the exchanges brief (which is clear from the frequent interventions of the researcher). This leads us to place emphasis on comparison between the groups to pull out “common framings” whose social underpinning is thus confirmed (see below).

15But this low level of conflict in the exchanges does not constitute a sufficient reason to take shortcuts in the analysis of these discrete interactions. On the contrary it seems a promising line of enquiry to understand what is playing out behind the scenes in these conversations. This implies articulating a comparative and interactional approach. Comparison leads to the discovery of two types of interactions governing how schemas are enacted. The first concerns the collective “management” of agreements and disagreements. There is an (often unspoken) stake in agreeing on what matters. The second is found in the sharing of experiences and of points of view. Recounting anecdotes or “an idea that just came into my head” (“that makes me think of…”) rarely pursue a deliberative objective. But it effectively inscribes the problem in different registers and categories. Schemas, through these “anecdotal” exchanges, generate interpretive repertoires adjusted to the situation. This point encourages us to defend a voluntarily loose usage of the concept of a schema, which is linked to the notion of fluidity. [35]

Saying what matters: the symbolic meaning of agreements

16If the cohesion of a collective depends on its power of exclusion, [36] Elias and Scotson also show that “gossip reinforces pre-existing cohesion”. [37] In other words, the “we”, which is often founded on the identification of a “they”, also demands upkeep. In taking seriously the discussions surrounding shared points of view, to which there is no need to return, we observe the routine ways in which a “we” is created. One of the functions of “chatting” is without a doubt to create symbolic links among the members of the group. In certain cases, this takes the form of non-verbal expressions (a wink or a nod of the head) that signifies agreement with what seems to the protagonists as obvious. The excerpt that follows shows how three Parisian lawyers, when they mention possible solutions to the climate problem, tacitly converge in valuing highly the responsibility of each person in relation to environmental problems.


[“Young Parisian lawyers” group]
Lisa – Yes, I think that before we get to the large scale, it must first enter into people’s heads.
Jean-Pierre – Yeah, these are small reflections: “Can you turn off the light?”; these are not discussions, but reflections! “Don’t leave the faucets on”, little things like that!
Géraldine – I’m the one who usually says these things.

18The conversation continues about the way education allowed them to internalize these gestures that have become “automatic”. Hence it is not just a question of agreeing on what must be done to preserve nature, but also to show (to themselves) that they share the same norms. Without seeming to do so, this interaction indicates to the participants that they have in common relationships to education and to self-discipline. The comparative analysis then reveals that, far from being unique, these moral understandings of ecology are found within the majority of groups whose members possess a high degree of cultural capital. Being somewhat in thrall to the logic of learning, the participants with the highest level of education in the groups believe that “the environment is like the alphabet, you have to learn it” (Guillaume, high school agronomy teacher). To the extent that the public construction of the climate problem operates in an individualizing register that is favored by the accentuation of the norms of “environmental citizenship”, [38] these groups show themselves to have a higher likelihood of aligning their practices with dominant models. But as we move down the social hierarchy, these understandings of the issues in civic and pedagogical registers occur less often, doubtless because the social profits that they can bring diminish or are less clearly perceived. [39] Environmental preoccupations increasingly tend to be relegated behind material, schooling, professional, and health preoccupations, etc. Members of the “rural workingclass” group thus frequently mention domestic practices for saving energy (geothermal energy, ground-coupled heat exchangers, insulation, biofuels, etc.). These are not so much justified by a concern for “civic virtue” as by their adaptation to the constraints these participants face.


[“Rural working-class” neighbors group]
Researcher – What do you find interesting in this energy-saving equipment?
Sylvie – First off, the savings.
General consensus: Marie, Henri, and Thierry repeat – Yes, the savings…
Thierry – In the end, it’s more expensive to buy, but then you make it up in the long run and then you’re not polluting the planet!
Marie – Exactly, it leaves you with a clear conscience.

20Even though they are gestured towards, the symbolic benefits linked to energy saving remain secondary to the expected economic gains. The collective and simultaneous exclamation, “yes, the savings”, translates a shared vision of what is at stake. This agreement operates in relation to the hierarchy of justifications and concerns the secondary nature of environmental and moral motivations within this hierarchy. While these participants do indeed feel that they are adopting environmentally virtuous behaviors, promoting (themselves via) these attitudes is less important for them than for other participants who do not share their social conditions. For them, education does not represent the best way to modify individual behaviors, which should instead be constrained, particularly through mechanisms that “hit the wallet”.

21The comparisons among the groups thus confirm the strength of social dispositions [40] to the extent that they show that the mobilization of schemas generates socially differentiated understandings of the norms of “environmental citizenship”. While the higher-level social categories and the most educated participants of the mid-level groups almost exclusively emphasize moral and civic motivations, working-class and mid-level groups almost systematically privilege economic motives, without necessarily being unaware of environmental issues. For the former, environmental discipline can be transmitted and absorbed while, for the latter, it must be imposed from the outside instead. Thus, despite its qualitative nature, comparative analysis allows us to perceive some regular features. These are not measured in terms of statistical frequency, but they are revealed by the persistence and consistency, within relatively equivalent social conditions, of certain ways of understanding the problem.

Not-so-“innocuous” discussions

22When people reach agreement on what matters these occasions are generally accompanied by the retelling of anecdotes or opinions whose “innocuous” nature should not mask their contribution to socialization processes. [41] It is useful here to wonder what participants do when they relate their experiences, their points of view, the media’s take on issues, or other types of knowledge, without necessarily seeking to impose their views. At the very least, they collectively define a coherent and legitimate frame of interpretation in relation to their common social characteristics. [42] They reaffirm norms, they re-evaluate the salience of the different aspects of the problem, they situate themselves, with their peers, against this multidimensional question. If we adopt the distinction proposed by Michael Schudson, the majority of the discussions in our corpus arise from the “sociable model of conversation”, and, more rarely, from the “problem-solving model”. [43] The following exchange illustrates these conversational practices for which what’s at stake is not to defend a particular position or to solve a problem, but to kill time, to become engaged, to play along with the discussion, etc.


[“Retired, mid-level” group]
Michel – They push buying energy-saving light bulbs now, rather than ordinary ones!
Jean-Claude – They’re expensive!
Reine – They’re expensive but you win out. I use them! And I see the savings…
Jean-Claude – Yes, I use them too.
Michel – With energy-saving bulbs, you need 15 watts to replace 75…That’s what they say, but when you turn on these light bulbs, you don’t see well right away, that comes a bit later! It’s not all that bothersome if you don’t need bright light.
Reine – I have five and I only turn one on!
Jean-Claude – Yes, but Reine is paying attention to her wallet.
Michel – But that’s only a small saving!
Jean-Claude – But it’s still a saving. Between a 15W and a 75W, the relationship is one to five!
Michel – Ok, but everyone always told me that for light bulbs, energy consumption was negligible!

24Within this group of six retired people, it is often a matter of sharing information, of teaching and learning from others. The disagreement over the futility of savings linked to low-consumption light bulbs poorly disguises the more general agreement around the necessity of saving money and energy. In other words, the discussion does not seem to impact points of view. Discussing low-consumption light bulbs is not, however, neutral. It assumes seeing this topic as pertinent both as a subject of conversation, and in terms of the assumed expectations of the other people present. Not all examples, anecdotes, experiences, or points of view can be mobilized in all circumstances, to the extent that we do not think everything we say and we do not say everything we think. What makes an experience mobilizable is particularly dependent on the anticipated reactions from the other people present. Over the course of the interactions, interpretive repertoires are thus constituted, which, because they satisfy the participants, become legitimate. This legitimacy, tacitly elaborated, comes through compromises or the use of “naturally”s, which makes the collective interview a situation of interpersonal communication that “supposes and produces mutual knowledge and recognition”. [44] Exchange transforms the things which are exchanged into signs of recognition. Thus, for the groups the most concerned with developing distinctive interpretations, understanding the climate question in terms of “rain and good weather” and, like Aurélie, thinking of “melting ice and the seasons we no longer have” [45] is simply unthinkable. The weather is not an appropriate approach for everyone and it is important for certain groups to avoid these concerns, judged as “trivial” because particularly associated with those who are less “cultivated”. For example, four history students at the Sorbonne, members of a student journalism association, immediately situate their discussion in a register with intellectual and universalizing pretensions.


[“Students at the Sorbonne” group]
Adrien – I think it’s the challenge of the 21st century. It’s a new challenge, there are no predecessors in history, and it’s inscribed in a perspective of globalization because it concerns humanity as a whole. Jean-Louis – It’s a subject that interests me a lot, but I have a hard time seeing where things are at the scientific level, because even if there is the IPCC, they don’t know if the rising temperatures come from the center of the earth. After all, we know that there are ice ages and inter-glacial periods, so maybe we’re entering a new era.

26Dependent as much on social dispositions as on the interaction, the narratives discussed by the participants orient the use of schemas of understanding. It’s a truism to say that one does not understand the climate problem in the same way when it is discussed based on facts from daily life (be that in terms of the weather or of energy management in the home) as when those discussions are based on political or scientific perspectives. The example of relationships to the international dimensions of the climate problem is significant. The participants who do not travel much feel “quite small” up against the size of the problem, like “a drop of water in the ocean”. They make veiled allusions to China or to the United States to indicate that these countries have not ratified the Kyoto protocol, or to observe that they have a far greater impact on the problem than France does. In the groups in which the participants travel frequently and far, the conversation may extend to these aspects of climate change, which then become the grist for more ostentatious attitudes. The register is then that of an eye-witness account that allows the speaker to move from a position as distant spectator of the diplomatic parrying, to that of spokesperson for the realities of the problem.


[“Young educated Parisians” group]
Sophie – The Chinese don’t pay for electricity based on their usage, they pay a fixed rate. So, they do not pay attention to what’s on. On the other hand, in all the places where I’ve been, I saw that everyone had solar panels. And that, in America, it’s the same, we’ve seen it.
Thomas – I find that in China, pollution is more obvious: they all have masks, there’s lots of smog, and the trees and grass aren’t green.

28The comparative analysis of this exchange, which is based on an uncommon practice (traveling long distances) brings to light its socially situated character. This type of observation is found more rarely in the poorest groups, which confirms a certain regularity in the social distribution of understandings which make use of travel experiences. As for the study of the interaction, it demonstrates its “friend-making” dimension as well as its socializing character. The goal is less to find a solution or to impose a point of view than to make new friends by sharing life experiences and observations. But, over the course of these conversations where people get to know each other, a relatively durable space is consolidated, made up of vocabularies, themes, arguments, and anecdotes that are more or less legitimate and appropriate to a given situation. [46] “Innocuous” conversations allow individuals to reach agreement on what matters, what is interesting, what is (un)acceptable, (dis)honorable, etc. They thus support the sharing of common visions of the world, indications that people belong to the same social milieu. [47]

29These remarks question the nature of the verbal exchanges we obtained. What do these streams of conversation tell us? How can we make sense of them without over-interpreting them? At the end of the first section of this article, it seems that a comparison of the themes “that make people talk”, often by reinforcing categories of common agreement, is fruitful for at least two reasons. On the one hand, it allows for the discovery of social regularities in ways of understanding the problem, and on the other hand, it seems to allow us to go beyond simple expressions of schemas by giving access to the – discreet – mechanisms through which they are formed and maintained. [48] As we have just seen, two of these mechanisms are observable from our data: 1) the (implicit) hierarchization of ways of understanding these issues, and 2) the clarifying of agreements about what matters and about which issues are worth the effort of discussing.

30In order to move forward with this method, I will now discuss the effects of sociabilities on the mobilization of schemas. The heterogeneity of our sampling in this regard allows us to consider a number of avenues for analysis. Collective interviews can represent an effective means for clarifying relationships where the individual and the collective are intertwined.

The mobilization of schemas via sociabilities

31Depending on the people we are speaking with, certain principles for understanding issues are developed, or, on the contrary, closed off. For example, when Michel tries seven different times to politicize the topic by deploying an ideological schema in a critical vein, his attempts are systematically “diverted” by the rest of the group. The sociabilities of this group, whose consituents get together every week as members of an association of retired people, prove themselves to be unfavorable to political discussions that are doubtless judged too “serious” and “involving”. The protagonists present a weaker social homogeneity and a lower familiarity with each other than in other groups. Their conversational routines are less regulated. This form of sociability explains to a large extent why Michel’s “politicizing” interventions are never taken up by the other participants. When Michel, a retired employee from the national rail system, the SNCF, attributes the responsibility for the problems to collective entities, the other participants pretend that he has not said anything and react to a comment that had been made just before, or else they move on to something else. In the illustrations that follow, the remark on the lighting of cities is, for example, extended by a digression about a regional expression from Lorraine regarding candles, which refers back to the discussion that preceded, the theme of which was home lighting.


[“Retired, mid-level” group: “politicizing” interventions by Michel]
“Yes, but we’re not the ones who pollute the most!”
“After all, we’re not the main people responsible. Instead, it’s factories, the system, but us…”
“We are not talking about lighting in cities at night, because they stay lit!”
“No, what is more interesting is everything that is used but doesn’t serve any purpose!”
“We need to develop transport via rail, rather than roads.”
“To reduce greenhouse gases, we need to dramatically reduce the number of trucks on the roads […] with diesel, they give off a lot of pollution, even though they’ve improved the quality of the diesels.”
“We haven’t talked about the high speed trains[TGV]! The TGV Est goes at 320 km/h.”

33While Michel seems to find it necessary to present things differently by contesting the individualization of blame, he does not seek to impose this vision of the problem. We can, however, wonder how things would have turned out if he had been in the presence of friends with whom he shared this politicized reading of the social world. He would doubtless have developed these positions, which would have been taken up and discussed.

34It is thus important to study the “frames of participation” [49] within which schemas are mobilized in order to appreciate the effects of sociabilities on the ways they are expressed. [50] Three dimensions of these sociabilities seem to orient the ways in which the schemas are mobilized, particularly in terms of how people take part in conversations. The first two, relatively classic, dimensions relate to objective social data, in particular the social resources possessed by an individual and their relationship to the theme of the discussion. On a more interactional level, the third is linked to the degree of routinization of the conversation. I speculate further about this dimension below, which attests to the importance of recruiting pre-existing groups. This methodological choice allows us to observe that the more sociabilities are established, the more the participants are used to debating amongst themselves, the more the modalities of participation are regulated, and the more the schemas are used in a contentious manner. This, at least, is what the analysis of the rare sequences in which participants express their disagreements indicates. Finally, in a more conclusive manner, we will return to the contribution of this methodology to the sociological comprehension of schemas, by demonstrating that it can reveal their social dialectic.

Two confirmed social factors affecting participation in conversations

35How much people speak up and in what circumstances remains a function of the distribution of social resources among the participants. Even in groups with strong social homogeneity, certain attributes continue in many cases to keep people on the fringes of exchanges, particularly when the number of participants increases. Looking at the characteristics of those who do not say much, or anything, reveals that they are the furthest away from the rest of the group in terms of their relationship to at least one objective variable (cf. Box 1). This objective distance is transformed into a subjective distance, which takes the form of low participation.

Box 1. What those who say nothing or nearly nothing tell us

Déborah and Jonathan, “Working-class family group”, n = 6, nursery school assistant and electrician, 16 comments between them. Their withdrawal is particularly linked to the fact that they are the youngest (20 and 21 years old). This is compounded by Déborah’s lack of familiarity with the other members of the family and Jonathan’s immediately casual attitude regarding the study.
Reine and Sabine, “Mid-level retirees group”, n = 6, cook’s assistant and cook, 33 and 35 comments (for an average of 82 per participant). While these two participants do indeed speak up, their interventions are always very brief (“Fear does not avoid danger!” “We take showers!” “Well, politicians…”) and they rarely seek to take the initiative of a response or of an exchange. These half-hearted participations are doubtless due in large part to their educational capital, which is lower by some way than that of the other participants.
Henri, “Working-class neighbors group”, n = 6, farmer. 8 comments. This low participation is explained by his wife’s high investment in the conversation. She is the most educated of the group and seems to speak for the couple. In addition, Henri seems to be embarrassed by his strong Béarnaise accent, which sometimes makes him hard to understand.

36Beyond these borderline cases, age, educational level, gender, or socioprofessional situation impact how people participate in the conversations. When speaking up is reinforced by favorable social positions, participants more often express opinions, judgments, and standpoints which are more substantiated, and often more engaged.

37A second factor in participation is linked to the specific relationships that certain participants enjoy with the topic under discussion; [51] or even to their representations of the research situation (cf. Box 2). These elements, which are more exogenous to the group, can influence the ways in which participants become involved in the exchanges and deploy schemas. Thus, certain participants make use of experiences that are unique to them to strengthen and to legitimize their comments. For example, in the group “Mid-level, 25 years old”, Johan weighed all the more heavily on the conversation because he was the only one to have seen the Al Gore film “An Inconvenient Truth”. He explained that when he saw the film he became aware of the importance of the problem, [52] which allowed him to distinguish himself from his partner: “I used to say to myself, yes, this is what happens, I thought the same way you did [Aurélie, his partner]. Afterwards, I said to myself that I had been completely off the mark.” This also enabled him to back up some of his comments via the authority of the film: “It’s also what he was saying in the film, there are going to be huge migratory zones.” This specific (and unexpected) argumentative resource visibly reinforced the position occupied by Johan in the conversation.

38In a relatively similar way, Géraldine (“Young Parisian lawyers” group) positioned herself as an expert on the question by drawing on the experience of her father’s, a senior executive at Total, where he works on environmental concerns. She brought up on a number of occasions – and at length – the effort of economic actors to improve their durable development policies. In relation to her colleagues who claimed to take a “catastrophic” view of the climate question and who spontaneously expanded the topic to include environmental issues, Géraldine’s views on climate change were clear and well-argued. It follows that a relationship of “complicity” to the theme of the discussion can influence how certain schemas are mobilized or reinforce the investment of participants in the exchanges. [53]

Playing at being debating professionals

The representations of the research situation on the one hand, and the relationships to conversational practice on the other, also influence how people participate in exchanges. The conversation among four undergraduate students in history at the Sorbonne shows in a fairly ideal-typical way how a particular conception of the situation can lead to highly codified contributions.
The attitude adopted by the participants, who were also members of a journalism association, consisted in imitating televised debates (such as “C dans l’air”). The language used to “pick up on” or “come back to” the comments of another participant attest to the concern to behave like public discussion professionals. Similarly, the number of cultured references (theology, Luc Ferry, Jacques Attali, the use of journalistic adages like “Green is Gold”, etc.) were used to validate what was said, just as much as the speaker. The participants positioned themselves above the fray, seeking to analyze the theme via the noble categories of journalism and politics. The climate question was put into perspective in relation to other current events. Speculation abounded concerning “people’s” opinions and solutions to move the situation forward. References to the IPCC, to Nicolas Hulot or to Al Gore were used to support points of view. The conversation here became a question of “a game that we must take seriously”. The researcher did not experience any particular difficulties in launching the exchanges, which became self-managing and continued without his intervention.

Established sociabilities: the social meaning of disagreements

39Our analysis also calls for a more detailed exploration of another variable that corresponds to the degree of routinization of the conversations. [54] By studying pre-existing groups, we build in the means to interrogate the impact of conversational routines on the ways participants take part in the discussion, as well as on how schemas are mobilized. From this point of view, the few sequences where participants enter into conflict deserve special attention. Because they involve a significant dose of subjectivity and because they bring into play the power relationships between the protagonists, they constitute moments that we can reasonably assume present a high degree of sensitivity to conversational routines. [55] The two examples that follow indicate that the participants who enter into disagreements generally occupy central positions in the discussion. The centrality of these positions, measured among other ways by the frequency of their comments, is explained on the one hand by the nature of their social characteristics but also by the history of their sociability within the group.

40In the “Working-class family” group, Laetitia, age 24, and Ludovic, her 26-year-old partner, did not agree on the risks linked to climate changes. According to him, as long as people can adapt there is no reason to change and, for change to be effective, it must be collective; so, for Ludovic, individualism and egoism take precedence over altruism. To this position, which was simultaneously “fatalist” and “pragmatist”, his girlfriend opposed a more “optimistic” and “voluntarist” vision. Laetitia criticized Ludovic’s attitude as being at the origin of the problems and pleaded for environmental discipline. This example can be analyzed in at least two ways. On the one hand, it illustrates the relationships of working-class people to climate issues (see above) via Ludovic’s economic realism, while nevertheless reminding us that these relationships are not univocal, since Laetitia outlines a vision that is closer to that characteristic of higher social categories – her upwardly mobile trajectory as well as her professional socialization [56] explaining in large part this difference with her partner’s point of view. On the other hand, we may wonder what this exchange tells us about the impact of sociabilities on the formation of opinions. The compromise sketched out by Ludovic at the end of the excerpt (“It’s true that everyone should practice some sort of self-discipline today regarding energy, but who is disciplined today? We have to be realistic too”) seems to indicate that his point of view is malleable, or at least situated within a tension between agreement about the seriousness of the problem (and thus about the necessity to act) and a desire to remain as close as possible to daily realities (and thus the pointlessness of acting). The discussion and presentation of a divergent opinion thus pushed Ludovic to complicate his point of view. But for how long? What would happen if he had to speak alone or without being contradicted? Inversely, if Laetitia found herself in a group of confirmed environmentalists, wouldn’t she demonstrate more pragmatism?

41Collective interviews thus raise the question of the variability and of the ambivalence of positions adopted. [57] Here, Ludovic showed himself to be receptive to his girlfriend’s counterarguments and recognized the legitimacy of her points of view, to which he had probably already been exposed. It remains true that the conversation observed, no matter how realistic it might be, corresponds to only one situation of sociability among other possibilities. The interaction between Ludovic and Laetitia thus only imperfectly informs the effects of the conversation on what the two participants think of climate issues, but it does provide information on the modalities of their participation in this discussion. Laetitia and Ludovic tended to monopolize this exchange (they total 185 comments out of a total of 245), which brought together Laetitia’s parents, her brother, and his girlfriend. But these other four protagonists did not engage at any time in a controversial argument. They did not all have “access” to disagreement and this works to reveal how participants invested themselves in the discussion. By expressing their disagreement, Laetitia and Ludovic conformed to a role that seemed routine to them within the framework of this family sociability.

42The conversation which brought together Aurore and her parents presented a comparable situation in terms of the division of positions within the exchange. The conversation essentially took place around the comments of Aurore and her father. The mother, who intervened often, but more briefly and expressing her opinion less often, played a more mediating role.


[“Upwardly mobile working-class family” group]
Philippe – In any case, our hands and feet are tied, what do you expect us to do!
Aurore – But if you think like that…
Philippe – What are you suggesting we do?
Aurore – I don’t really know, set up focus groups.
Philippe – Today, I heard that they were going to introduce a tax for those who release more than 200 grams of CO2, but I think it’s yet again more a question of money than anything else!

44The modalities of participation in this conversation must first be related to the social properties of these three participants, properties that make sense in relation to one another. Thus, only the mother, who has a CAP diploma [certificat d’aptitude professionnelle, a vocational diploma], is not in work or studying. The father is an industrial draftsman and also has a CAP. Aurore, who has completed the first year of a master’s degree in the management of organizations in an international context, thus has a significantly higher level of educational qualification than her parents. This contributes to explaining her central role in the discussion. As for the polarization of the discussion between the father and the daughter, it seems to refer to well-established conversational routines, as indicated by the tone of the sequence in which Aurore was annoyed by the fatalism of her father, who challenged her to find solutions: “Aurore: yes, but if you think like that… Philippe: What are you suggesting we do?”

45Between Ludovic and Laetitia or between Aurore and her father, a higher degree of conversational routine seemed to lessen the risk incurred when disagreements came up. The reactions were less uncertain, argumentative markers more tested and familiar. Our data in fact indicates that this high degree of routinization of conversations is found more often within groups composed of members of the same family. [58]

46Sociabilities influence spoken contributions and how schemas are mobilized, be that through the social properties of the protagonists, the specific relationships of some with the topic of discussion, or the conversational routines of the group. They orient the collective “management” of agreements and disagreements, as well as the background elements brought together to interpret the climate question. Nonetheless, the effect of sociabilities should be approached carefully. Our observations suggest the importance of dissociating two levels of discourse that also constitute two levels of analysis.

Recognizing stability in diversity: regularity and adjustments of schemas

47At a first level, that of the substance of the conversations and of modes of mobilization of schemas (conflictual or not; themes addressed; anecdotes recounted; lexical registers; associations of ideas, etc.), sociabilities are not neutral, particularly regarding the conflictual content of discussions. Disagreements are generally expressed between participants whose sociabilities are already known. [59] On the other hand, the less the interaction is regulated, the more we observe the avoidance of disagreements, and exchanges being confined to agreements. [60] Beyond this, the anecdotes, the metaphors, the reasoning or the judgments expressed vary as a function of sociabilities. The ways in which they are used vary to the extent that they are adjusted to the (more or less) anticipated reactions of the other participants in the discussion. From one interaction to another, the same participant modulates the substance of his or her comments. The properties of the interaction seem to impact the instability and the “heterogeneity of enunciative positions”. [61]

48Nevertheless, this instability of enunciative positions is not without its limits, as dialogic approaches in social psychology might lead us to suppose. It is thus appropriate to note that at a second level, that of the contours of the space of the sayable and of the types of schemas mobilized, sensibility to sociabilities is weak. For example, Johan’s references to Al Gore’s film did not affect the way in which this group understood the climate question. Despite a few increases in general comments engaged in solely by this participant (on the distant consequences or the industrial causes of the problem), the interpretation of the group essentially draws on the daily realities on the basis of which these participants make sense of the world. Logical operations which enable the association of a topic with an experience, or determine what is worthy of interest, thus turn out to be less reliant on sociabilities. Although they may vary in substance, ways of understanding – in other words of selecting, of hierarchizing and of categorizing – the issues contained in a problem remain above all dependent on socializations.

49In sum, as illustrated by the diagram below, sociabilities influence how schemas are mobilized but only affect the margin of the type of understandings that they generate. In other words, social conditions being equal, the types of schemas mobilized and the contours of the space of the sayable vary little when the interactional context changes. Inversely, interpretive repertoires that “furnish” this space of the sayable can take many forms as a function of interactions. From this point of view, collective interviews also enable us to emphasize the capacity of schemas and social dispositions to adapt, a sometimes-neglected dimension of the “theory” of habitus. The system of schemas that constitutes the habitus produces coherence but is not by any means regulated like a metronome. From this, it seems dangerous to try to reify a concept which is highly dialectic, to the extent that it simultaneously produces both regularity and plurality. Instead, this study leads us to call for an analysis that seeks to recognize stability in the diversity of elements (anecdotes, reasoning, judgments, etc.) that fuel conversations. [62]

Figure 1

The social dialectic of schemas

Figure 1

The social dialectic of schemas


50Analyzing conversations among peers – here on climate changes – constitutes an effective means for capturing the social anchoring of schemas as much as their uses. The interactions (where the uses are played out) to which collective interviews provide access gain in effect from being studied in their full social context (in which schemas are anchored). From this, collective interviews contribute to a better sociological knowledge of schemas of understanding a public problem, particularly because they allow us to be specific about patterns and their degree of adjustment to circumstances.

51But this article also opens up the possibility of reformulating the initial question to ask ourselves if schemas constitute a pertinent level of analysis to account sociologically for modes of understanding public problems. These problems are generally comprised of many different issues. The climate question thus poses the problem of industrial development, the control of energy production, urban planning, lifestyles, consequences on biodiversity, the unequal capacities of adaptation of societies, etc. The categories elaborated for the analysis of political behaviors, such as “competencies” or “attitudes” thus do not seem the most appropriate for understanding routine relationships to the problem. When we focus on forms of knowledge, on how qualified people feel themselves to judge, or on the affective and cognitive dynamics of the evaluation of certain issues (most often political), such categories seem less suitable for analyzing ways of understanding the problem in its entirety. They only capture the relationships to one or other of its dimensions (how it is managed politically or scientifically; its importance etc.); whereas the concept of a schema relates back to transferable categories of thinking – in other words, which could be elaborated regarding other questions and which, above all, are not linked to one dimension of the problem in particular. [63] We should therefore empirically update the transferable character of schemas – which is well developed by Pierre Bourdieu in Distinction[64] – particularly by interrogating the same participants on a number of occasions regarding different types of public problems. [65]

Methodological annex

The social morphology of collective interviews

Total: 12 groups, 50 participants

52Sample size is often invoked to criticize the pertinence of collective interviews. While the results put forward should be considered with caution, the criticism concerning thresholds beyond which the corpus becomes reliable, and the results valid is debatable. Such criticism tends to examine the practices of qualitative research through the lens of the issues and principles of quantitative research. According to Peter Lunt and Sonia Livingstone, collective interviews are subject to other validation criteria for results derived from qualitative methods and their objectives. [66] Collective interviews find through comparison, for example, the sociological meaning that quantitative methods obtain via representativity. “What we gain by extension (of the number of cases studied) is also lost in intensity, and vice versa”; [67] quantitative and qualitative techniques are complementary since they do not claim to respond in the same way to the same questions, nor to follow exactly the same methodological rules.

53The information presented below essentially focuses on the social properties of the participants. Our concern to collect this information is linked to the fact that we consider the role of social origins and conditions as central in how people appropriate a public issue. A section of the interviews was thus devoted to the administration of a questionnaire distributed individually so as to gather the sociographic properties of the participants.

Three groups from working-class milieus, 15 participants

54“Working-class neighbors” group, six participants: a couple and their seventeen-year-old son, who live in a village in the Béarn region, invited over a neighboring couple who additionally brought their thirteen-year-old son. These six participants see each other daily or nearly daily and thus share strong ties. None has passed the high school baccalaureate exam, other than the female neighbor, who is a lab assistant and has a BTS [technical diploma]. The adults are between the ages of 42 and 48, they are white-collar employees or rural workers, occupying social positions similar to that of their parents. None gave his or her income.

55“Working-class family” group, six participants: a couple aged 46 and 47, who live in a village in the Bouches-du-Rhône region, engaged in discussion with their two children, aged 24 and 20, who were accompanied by their respective partners, ages 26 and 21 years. With the exception of the girlfriend of the younger child, the participants have strong ties. None has passed the high school baccalaureate exam, and those who work are employees or workers, occupying social positions similar to that of their parents. Their declared income ranged from 700 to 1,600 Euros per month (net). [68]

56“Upwardly-mobile working-class family” group, three participants. Made up of a couple and their daughter, this group shared strong ties. The parents, aged 47 and 53, have CAP diplomas. The father is an industrial draftsman and the mother is a “housewife”. The daughter, aged 22, is a master’s student. All have upwardly mobile trajectories (the parents of the couple were rural workers). They live in a medium-sized town in Picardie and the father declared income of 1,870 Euros per month (net).

Four groups in intermediate social positions, 17 participants

57“Intermediate, 50-somethings” group, four participants: this group brought together two couples who live in a mid-size town in the Picardy region. Two participants are siblings (brother and sister). The ties between these members are strong. Aged between 46 and 57 years, they work in intermediate professions, according to the professional and socioprofessional categories defined by Insee. Two have high school baccalaureate degrees and the other two technical high school degrees. They have upwardly mobile trajectories and declared earnings of between 1,700 and 3,400 Euros per month (net).

58“Intermediate, 25 years old” group, four participants: this group brought together two couples who live in a small village in the Oise region; the women are cousins. The ties between them are strong. Aged between 25 and 28 years, three of them have undergraduate college degrees and work in intermediate professions, the fourth is a farmer and holds a specialized technical high school degree. They occupy social positions equivalent to their parents and declared earnings of between 1,300 and 2,000 Euros per month (net).

59“Retired” group, six participants: made of up three men and three women, this group brought together members of an association for retired people who meet up occasionally and whose ties together are weak. Only two have the high school baccalaureate degree, three are former low-level white-collar employees, two are former high school teachers, and one is a former manager. Aged between 65 and 79 years, they have positions that are either equivalent to or higher than their parents. They all live in a medium-sized city in the Meurthe-et-Moselle region and declared earnings of between 550 and 2,600 Euros per month (net).

60“Teachers” group, three participants: a couple invited a friend over, they share strong ties with him. The three participants have undergraduate degrees and teach in the same high school. Aged between 28 and 38 years, their status corresponds to that of their parents. They live in Lille and its suburbs and declared earnings of between 1,500 and 1,800 Euros per month (net).

Five groups from high social categories, [69] 18 participants

61“Students at the Sorbonne” group, four participants: four undergraduate students in history at the Sorbonne met at the headquarters for their journalism association. These participants, aged between 19 and 22 years, share weak links. They live in Paris and their parents are all high-level executives.

62“Young Parisian graduates” group, 4 participants: this group brought together three women and one man, aged between 24 and 26 years and who share strong ties even though two do not know each other (a pair of roommates invited their good friend as well as a work colleague who has become a friend but who did not know the other guest). Three are managers, and one a master’s student. They all live in Paris and have social positions higher than their parents. They declared earnings of between 850 and 1,800 Euros per month (net).

63“Young Parisian lawyers” group, five participants: this group brought together four women and one man, aged between 29 and 32 years, who work together but whose ties are rather weak (they do not spend time together outside work). Lawyers, living in Paris and its suburbs, they have equivalent statuses to their parents. They claimed earnings of between 1,100 and 2,700 Euros per month (net).

64“Parisian intellectuals” group, three participants: two women living together invited over a friend with whom they share strong ties. Aged between 43 and 50 years, they hold advanced degrees, master’s level and above. Like their parents, they work in high-level intellectual professions and declared earnings of between 1,500 and 3,700 Euros per month (net).

65“Parisians, medical professionals” group, two participants: the discussion between this surgeon and anesthetist took place at the surgeon’s house. Aged 39 and 47 years, these two long-standing friends share strong ties. They have higher social positions than their parents, live in Paris, and declared earnings of 4,000 Euros per month (net).


  • [1]
    Norbert Elias, Qu’est-ce que la sociologie? (La Tour d’Aigues: L’Aube, 1991), 156.
  • [2]
    William Gamson’s study is clearly written from this perspective when he studies how groups with low educational capital interpret four contemporary themes. The use of collective interviews is justified by “a group dynamic that leads participants to elaborate on common framings, even if they do not agree on the solutions”: William Gamson, Talking Politics (New York: Cambridge University Press, 1992), 128.
  • [3]
    In this article, the word “group” refers to the “discussion groups” that we analyzed and not to social groups making up society (designated here as “social milieus” or “social categories”).
  • [4]
    For Éric Darras, collective interviews represent a rich methodological approach for reaching the least well off “in the sense that being together would facilitate working-class people speaking up”: Éric Darras, “Le pouvoir de la télévision? Sornettes, vieilles lunes et nouvelles approches”, in Antonin Cohen, Bernard Lacroix, and Philippe Riutort (eds), Les formes de l’activité politique: Éléments d’analyse sociologique. 18e-20e siècles (Paris: PUF, 2005), 471.
  • [5]
    In France, Florence Haegel and Sophie Duchesne seek for example to uncover “how an individual politicizes his point of view” through conflictualization or specialization: Sophie Duchesne and Florence Haegel, “La politisation des discussions, au croisement des logiques de spécialisation et de conflictualisation”, Revue française de science politique, 54(6), 2004, 878.
  • [6]
    Jenny Kitzinger, “The methodology of focus groups: the importance of interaction between research participants”, Sociology of Health and Illness 16(1), 1994, 113. Indeed, according to Alfredo Joignant, “focus groups are better suited to revealing very general political competencies and productive forms of knowledge that would be otherwise invisible to researchers, to the extent that these competencies and forms of knowledge are more discernible in situations of conversation than in the framework of an interview”: Alfredo Joignant, “Compétence politique et bricolage: Les formes ‘profanes’ du rapport au politique”, Revue française de science politique, 57(6), 2007, 803.Online
  • [7]
    Jocelyn A. Hollander, “The social contexts of focus groups”, Journal of Contemporary Ethnography, 33(5), 2004, 632. I thank Guillaume Garcia for suggested this text to me.Online
  • [8]
    Alfredo Joignant, “Pour une sociologie cognitive de la compétence politique”, Politix, 17(65), 2004, 162.
  • [9]
    Pierre Bourdieu, Méditations pascaliennes (Paris: Seuil, 1997), 166-7.
  • [10]
    A. Joignant, “Pour une sociologie cognitive…”, 169.
  • [11]
    From this point of view, the work of Richard R. Lau in social cognition is an exception. This author attempts to link types of schemas to social properties, such as level of education or income. Cf. Richard R. Lau, “Political schemata, candidate evaluations, and voting behavior”, in Richard R. Lau, and David O. Sears (eds), Political Cognition (Hillsdale: Erlbaum, 1986), 95-126.
  • [12]
    James H. Kuklinski, Robert C. Luskin, and John Bolland, “Where is the schema? Going beyond the ‘S’ word in political psychology”, American Political Science Review, 85(4), 1991, 1341-80 (1346). These authors, who question the contributions of schema theory, call for a more systematic recourse to experimental techniques to capture “the finer grain of stored cognition, much less cognitive processes” (1348).Online
  • [13]
    Émile Durkheim and Marcel Mauss, “De quelques formes de classification: Contribution à l’étude des représentations collectives”, L’Année sociologique, 6, 1903, 1-72. The authors call for a sociology of knowledge and, beyond that, of the logical operations of human understanding, their origin and their functioning in society.
  • [14]
    This last point allows us to underline that through schemas a process of selection, hierarchization, and categorization of different issues is operating. The concept of schema thus comes close to the notion of “framing”, but it is distinguished to the extent that it accentuates the pre-existing, durable, and adjustable character of the interpretations mobilized by socialized individuals. On the relevance of the notion of “framing”, which implies simultaneously analyzing media mechanisms and social dispositions, see Jacques Gerstlé (ed.), Les effets d’information en politique (Paris: L’Harmattan, 2001).
  • [15]
    In the generally accepted meaning in French-language literature, by sociability I mean the forms taken by the social relationships of an individual. Interactions are thus inscribed in more or less formalized sociabilities. On the origin and the uses of the concept of sociability in sociology, see Carole-Anne Rivière, “La spécificité française de la construction sociologique du concept de sociabilité”, Réseaux, 123 (2004), 207-31.
  • [16]
    The part of the study that is discussed in this article comes from my thesis: Jean-Baptiste Comby, “Créer un climat favorable. Les enjeux liés aux changements climatiques: valorisation publique, médiatisation et appropriations au quotidien” Paris, Université Paris II-Panthéon Assas, 2008. This thesis was also the basis for a research report submitted to the Agence de l’environnement et de la maîtrise de l’énergie (ADEME, Agency for the environment and energy management) in November 2008 and entitled “Les mécanismes sociaux en jeu dans la formation des opinions sur les enjeux liés aux changements climatiques (2006-2008)”. The financing of this report allowed for the recruitment of a research officer, Grégory Mercier, whom I would like to thank for his close and rigorous collaboration in this work.
  • [17]
    On the distinction between conflictual and cooperative logics in political discussions, see Sophie Duchesne and Florence Haegel, “What political discussion means and how the French and the (French speaking) Belgians deal with it”, in Ken’ichi Ikeda, Laura Morales, and Michael Wolf (eds), Political Discussion in Modern Democracies: A Comparative Perspective (London: Routledge, 2010), 44-61.
  • [18]
    I would like to thank Guillaume Garcia, Florence Haegel, Matthieu Grossetête and Vincent Goulet for their indispensable comments on previous versions of this text.
  • [19]
    W. Gamson, Talking Politics, 192.
  • [20]
    These intermediaries were (more or less close) acquaintances of the researchers.
  • [21]
    As Sophie Duchesne explains, when it is a question of qualitative methods, “speaking of representativity thus only means that we strive to bring together people with all of the characteristics which might result in differences relating to the representations under study” (Duchesne, “Pratique de l’entretien dit ‘non directif’”, in Myriam Bachir (ed.), Les méthodes au concret: Démarches, formes de l’expérience et terrains d’investigation en science politique (Paris: PUF/CURAPP, 2000), 11.
  • [22]
    Cédric Planchat, “Protéger l’environnement: un objectif pour une grande majorité de Français”, Insee première, 1121, January 2007: study undertaken in January 2005 in partnership with ADEME, the Institut français de l’environnement, and the Ministry of the Environment.
  • [23]
    Moreover, we might wonder whether a researcher who strives to make exchanges last would not thus be adapting the research design to the dispositions of the most culturally equipped, in other words to those who are more likely to have more things to say and to be more at ease in a discussion that lasts “abnormally” long.
  • [24]
    Dominique Boullier, La télévision telle qu’on la parle: Trois études ethnométhodologiques (Paris: L’Harmattan, 2004).
  • [25]
    The interviews were carried out between June and December 2007. The period of recording is often seen as a strong factor of bias. The news in the days or weeks preceding the interview could influence the content of the discussions recorded. However we should not overestimate this criticism which bears the imprint of mediacratism. It is often the case that what is being studied is more than just how participants reprise media discourse. At the same time, clarifying the perspectives and the ways in which media messages are discussed can indicate how the problem is appropriated.
  • [26]
    The interviews were moderated in part by the author and in part by G. Mercier, a temporary researcher recruited for this study. To standardize our procedures, we moderated the first interview together, then, using a video-recording, we clarified the directions based on the different scenarios anticipated (for example: never give our opinions asked; re-cue and look directly at those who speak less, etc.).
  • [27]
    Gérard Mauger, “Enquêter en milieu populaire”, Genèses, 6 (1991), 134.Online
  • [28]
    These passages are not all that numerous, and a large part of the transcription indicates fewer exchanges between the participants, and more, mostly collective, responses to the questions of the researcher.
  • [29]
    In the sense in which the term is used by Haegel and Duchesne, who characterize conflictual political discussions by the fact that they implicate the social identity of the speakers and that they are underpinned by (often pre-existing) political divisions: S. Duchesne and F. Haegel, “What political discussion means and how the French and the (French speaking) Belgians deal with it.”
  • [30]
    S. Duchesne and F. Haegel, “La politisation des discussions…”, 883.
  • [31]
    Philippe Juhem, “La logique du succès des énoncés humanitaires: un discours sans adversaires”, Mots, 65, 2001, 9-27. On the depoliticization of the climate problem in France, see Jean-Baptiste Comby, “Quand l’environnement devient ‘médiatique’: Conditions et effets de l’institutionnalisation d’une spécialité journalistique”, Réseaux, 157, 2010, 157-87
  • [32]
    This argument echoes the thesis put forward by Élisabeth Noëlle-Neumann, who supports the idea that dominant media framings generate a “spiral of silence”. To avoid disapproving judgments and isolation, people prefer to conform to the opinion that they perceive as being or about to become dominant. For these people, the media thus represent an effective means for understanding a certain state of opinion. Cf. Élisabeth Noëlle-Neumann, “The spiral of silence”, Journal of Communication, 24, 1974, 43-54.
  • [33]
    The information produced in televised news shows are not addressed to specifically implicated audiences but rather to “anyone who comes along”. We thus decided to recruit based on variables shared by the ensemble of television audiences of news shows, to the detriment of the “specific interest in the issues” variable. We also wanted to capture the general dynamics for understanding climate issues and not the effect of activist or professional dispositions.
  • [34]
    Greg Philo shows that groups understand news topics differently depending on the objective relationships that they maintain with these questions. To demonstrate this, he developed a research protocol in which he asked groups of police and union activists to construct a news bulletin based on images taken from a report by the BBC on a miners’ strike. Greg Philo, Seeing and Believing: The Influence of Television (London: Routledge, 1990).Online
  • [35]
    J. H. Kuklinski, R. C. Luskin, and J. Bolland, “Where is the schema?…”, 85. The authors regret the still vague character of “schemas” and the fact that there has not been a cartography of political schemas. Inversely, my approach posits that “the habitus is what makes agents behave in a certain way in certain circumstances […] The habitus is in part linked with the undefined and the vague”: Pierre Bourdieu, “Habitus, code et codification”, Actes de la recherche en sciences sociales, 64(1), 1986, 41 (emphasis mine).Online
  • [36]
    Richard Hoggart, La culture du pauvre (Paris: Minuit, 1970).
  • [37]
    To use Norbert Elias and Francine Muel-Dreyfus’s expression, “Remarques sur les commérages”, Actes de la recherche en sciences sociales, 60 (1985), 27. This article refers to Chapter 7 of the work by Norbert Elias and John L. Scotson, Les logiques de l’exclusion: Enquête sociologique au cœur des problèmes d’une communauté (Paris: Fayard, 1997).
  • [38]
    J.-B. Comby, “Quand l’environnement devient ‘médiatique’…”. On this point, see also Sophie Dubuisson-Quellier, La consommation engagée (Paris: Presses de Sciences Po, 2009). Online
  • [39]
    As Luc Boltanski had already observed in the 1960s: “If it is true that ‘virtue requires a certain affluence’, the material conditions of possibility of ascetic morals are rarely met in the working classes” (Luc Boltanski, Prime éducation et morale de classe (Paris: Éditions de l’EHESS, 1984), 119).
  • [40]
    Daniel Gaxie, “Appréhensions du politique et mobilisations des expériences sociales”, Revue française de science politique, 57(6), 2007, 145-78.
  • [41]
    Taking equally seriously these shared interpretive territories and their function (particularly in relation to identity) in conversations, Michael Billig speaks of a “kaleidoscope of common sense”: Michael Billig, Talking of the Royal Family (London: Routledge, 1992), 48.Online
  • [42]
    With Erving Goffman, it’s the status of speaking in interaction that we must then consider. According to him, “nothing is more effectively the foundation of a common engagement than speaking. Words are the biggest and best way to attract the speaker and the listener inside the same space of attention, into the same schema of interpretation applying to the aforementioned space.” Erving Goffman, Façons de parler (Paris: Minuit, 1987), 80.
  • [43]
    Michael Schudson, “Why conversation is not the soul of democracy”, Critical Studies in Mass Communication, 14(4), 1997, 299. Jenny Kitzinger also notes the importance of the discussions during which the participants “speak for pleasure” and which belong to the “performance of and [a] way of ‘being sociable’”. Such discussions crystallize representations that have become common, in the double meaning of being banal and being shared. See Jenny Kitzinger, “Le sable dans l’huître: analyser des discussions de focus group”, Bulletin de psychologie, 57(3), 2004, 302.Online
  • [44]
    Pierre Bourdieu, “Le capital social: Notes provisoires”, Actes de la recherche en sciences sociales, 31(1), 1980, 2.
  • [45]
    “Intermediate, 25 years old” group.
  • [46]
    In these conditions, “people talk primarily with others who share their values and they expect that conversation will reinforce them in the views they already share”. M. Schudson, “Why conversation is not the soul of democracy”, 302.
  • [47]
    Here we see the relevance of constituting pre-existing groups, in the sense that familiarity among the participants encourages these logics that would undoubtedly have been less perceptible in so-called “artificial” groups (and within which the “sociable model of conversation”, often facilitated by knowing each other, would function with more difficulty).
  • [48]
    Inversely, theorists of the schema in social psychology think that we can only have access to the empirical consequences of schemas and not to the schemas themselves. Cf. Milton Lodge, Kathleen M. McGraw, Pamela Johnston Conover, Stanley Feldman, and Arthur H. Miller, “Where is the schema? Critiques”, American Political Science Review, 85(4), 1991, 1365.
  • [49]
    Erving Goffman, La mise en scène de la vie quotidienne: La présentation de soi (Paris: Minuit, 1973).
  • [50]
    In the observations that she undertook within local associations of immigrant youth (two leisure associations and one with a social vocation), Camille Hamidi shows that “the logics of politicization do not depend exclusively on individual attributes, but also on the effects of context”. Camille Hamidi, “Éléments pour une approche interactionniste de la politisation: Engagement associatif et rapport au politique dans des associations locales issues de l’immigration”, Revue française de science politique, 56(1), 2006, 14.
  • [51]
    G. Philo, Seeing and Believing…, 86-110.
  • [52]
    Which does not, in fact, prevent him from continuing to make economic preoccupations prevail over environmental issues: “As for us, for heating, I try to consume less but it’s only for financial reasons. My first idea is not to say to myself ‘hey, I’m going to use less heat because of global warming’.”
  • [53]
    The example of Henri nuances this observation. Possessing a unique relationship to the issue that is being discussed does not always take the form of higher participation in the interaction. Thus, in his group, Henri is the most directly implicated by the problem, in that he is a farmer. But, he does not draw on this practical relationship to the issues to take a greater part in the conversation.
  • [54]
    The expressions “conversational routines” or “degree of routinization” refer to habits and to conversational reflexes between the participants. It is difficult to reduce them to a particular marker: instead they appeared when we analyzed the conversations and even more when we watched the recordings and considered non-verbal behaviors. Conversational routines take into account, for example, the propensity to initiate or put an end to an exchange. This line of enquiry is merely explored here, and could be deepened based on other studies that are more explicitly conceptualized to work on it.
  • [55]
    Here I draw on the previously cited works by Haegel and Duchesne, as well as those by Billig, according to whom: “the radical arguments that can be put forward when an individual expresses/imposes his or her position are socially produced, and the complex rhetoric of these positions as well as the counter-reactions that they provoke must be analyzed in relation to this argumentative context”. Michael Billig, Ideology and Opinions. Studies in Rhetorical Psychology (London: Sage, 1991), 81.
  • [56]
    As a salesperson in interior decoration, Laetitia is developing in a professional milieu where the attention paid to one’s surroundings and to the environment is intense. In particular, the interior decoration magazines that she reads regularly include many references to environmentalism in daily life.
  • [57]
    On the difficulty for participants to maintain coherent positions over the course of a conversation, see Darras, “Le pouvoir de la télévision?…”.
  • [58]
    The families we brought together have histories defined by close proximity, with many years spent under the same roof.
  • [59]
    Here we coincide with the conclusion drawn by Schudson, who observes how, in conversation with people that they know, “people may test their opinions, to be sure, and venture ideas that may not be warmly received, but they do so knowing that they agree on the fundamentals and that the suggestions that they are making will not jeopardise the relationship” (M. Schudson, “Why conversation is not the soul of democracy”, 302).
  • [60]
    These suggestions must be relativized as a function of social variables, such as the volume and the structure of educational capital, that lead to unequal deliberative positions. International comparisons show that national traditions can also orient the inclination to engage in debate: cf. Sophie Duchesne, Elizabeth Frazer, André-Paul Frognier, Guillaume Garcia, Florence Haegel, and Virginie Van Ingelgom, “Europe between integration and globalisation: social differences and national frames in the analysis of focus groups conducted in France, Francophone Belgium and the UK”, Politique européenne, 30(1), 2010, 67-105.
  • [61]
    Anne Salazar Orvig and Michèle Grossen, “Représentations sociales et analyse de discours produit dans des focus groups: un point de vue dialogique”, Bulletin de psychologie, 57(3), 2004, 263-72.
  • [62]
    In the same vein, sociologists of language underscore the relevance of “linking social situations and discursive regularities”. François Leimdorfer, “Registres discursifs, pratiques langagières et sociologie”, Langage et société, 124, 2008, 10.
  • [63]
    According to Conover et al., attitudes are focused on one unique object while schemas are more general: “Attitude structures characterized by the sum of all beliefs about an object are not the same as schematic structures characterized by categories and hierarchical structures.” M. Lodge, K. M. McGraw, P. Johnston Conover, S. Feldman, and A. H. Miller, “Where is the schema? Critiques”, 1366.
  • [64]
    Pierre Bourdieu, La distinction: Critique sociale du jugement (Paris: Minuit, 1979).
  • [65]
    The intersection between this research and the research lead by Mathieu Grossetête – in the same vein – regarding highway fatalities tends to show that with groups form equivalent social conditions, these two public problems are understood based on similar, even comparable, schemas. In particular, the “civic” relationship to the issues is found more frequently for members of the most educated social categories. These observations thus suggest a certain transferability of schemas. See Jean-Baptiste Comby and Mathieu Grossetête, “Les classes sociales face à la ‘sécurité routière’ et au ‘changement climatique’”, presentation at the “30 ans après La distinction” conference, Paris, 6 November 2010.
  • [66]
    Peter Lunt and Sonia Livingstone, “Rethinking the focus group in media and communication research”, Journal of Communication, 46, 1996, 93.Online
  • [67]
    D. Gaxie, “Appréhensions du politique et mobilisations des expériences sociales”, 172.
  • [68]
    NB: the indicated income corresponds to information given by the participants; these are thus orders of magnitude to be interpreted with prudence, in the sense that they do not all mean the same thing and are often given “from memory”.
  • [69]
    These five groups live in Paris, and the sampling thus does not include groups from higher social categories living in the provinces.


Focus groups encourage us to combine comparative analysis with approaches which pay more attention to interaction. They reveal how schemas, which also are social cognitive and interpretive predispositions, express themselves according to different forms of sociability. Data show that although the kind of schema used may be merely due to an individual’s social trajectory and thus vary little when the context changes, the ways in which these schema are used, on the other hand, demonstrate a higher sensitivity to “participation frames” (Goffman, 1973).

Jean-Baptiste Comby
Lecturer at the Institut français de presse, Université Paris-II Panthéon Assas, Jean-Baptiste Comby teaches sociology of the media. He is also a researcher at the Centre d’analyse et de recherche interdisciplinaires sur les medias (CARISM, Center of interdisciplinary analysis and research on media). In particular he has published: “Quand l’environnement devient médiatique: conditions et effets de l’institutionnalisation d’une spécialité journalistique”, Réseaux, 157-8 (December 2009), 159-90; “La contribution de l’État à la définition dominante du problème climatique”, Les Enjeux de l’information et de la communication (December 2009),; “La mise en scène d’une demande des habitants pour enrôler les collectivités territoriales dans la ‘lutte contre’ le changement climatique’, in Virginie Anquetin and Audrey Freyermuth (eds), La figure de l’“habitant”. Sociologie politique de la “demande sociale” (Rennes: Presses Universitaires de Rennes, 2009), 75-92; “Statistiques et imputations de responsabilité: ‘Les Français sont responsables de 50% des émissions de Gaz à effet de serre’”, Médiations et information, 28 (December 2008), 149-57. His research focuses on the one hand on the social construction of public problems, and on the other hand on the social impact of relationships to politics and to the media (CARISM, Université Paris II-Panthéon Assas, 4 rue Blaise Desgoffe, 75006 Paris
Translated from French by 
Jasper Cooper
Latest publication on cairn or another partner portal
Uploaded on on 03/03/2014
Distribution électronique pour Presses de Sciences Po © Presses de Sciences Po. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait