1Many works in political science have shown how valuable exploring the universe of private life can be in understanding the political attitudes and behaviors of individuals in the public sphere. From the 1950s, the works of the Columbia school then those of the Michigan school attributed great importance to the place and role of emotional interactions with loved ones. They established the influence of primary groups on the forms and intensity of politicization, and demonstrated the impact of familial transmission on the shaping of ideological and partisan choices. In the field of electoral analysis, these seminal works showed that voting is not merely an individual act: it also involves other people in a person’s immediate social circle, especially the family. The circle of family and friends, because it is a site for exchange and discussion, steers and influences individual voting choices far more than any media propaganda. Family is the first source of options for citizens’ decisions, especially their electoral decisions: 35% of French people say that they give greater weight to the opinion of a family member in this domain (19% take into account the opinions of experts in the media and 14% the opinions of their friends).  The importance of these identification mechanisms in the parental transmission of political choices has long been established in studies on political socialization. 
2Recent electoral sociology shows a relative autonomization of electoral behaviors and a relative weakening of dispositional sociological factors in explaining both voting and non-voting.  Nevertheless, the political socialization processes within the primary group always play a decisive role in the making of citizens.  Despite competition from other bodies, and even though it is less prescriptive concerning social, cultural and ideological allegiances than in the past, the role of the family context in political learning and in influencing electoral choices continues to be key.  For the vast majority of French people, family political tradition and the political choices of one’s parents remain influential. Two thirds of them (66%) favor continuity with the political choices of their parents (left-wing, right-wing or neither), whereas only a small minority (11%) break away and change political side.  This role has been confirmed for ideological orientations, but also for behaviors. It has been possible to measure a reinforcement of electoral participation within the household when members of the same family live together.  Living together, sharing everyday exchanges and discussion, thus encourages family members to vote.
3It has long been recognized that a conjugal relationship influences people to vote.  Those living as a couple are more likely to participate in elections than those who are not a couple. In certain situations, being in a conjugal relationship may even compensate for increasingly low turnout in working-class areas, particularly among women.  Moreover, the effect of conjugal or familial transitions on the conditions of political participation has been demonstrated, showing significant differences between women and men. We may thus observe decreased participation among women in cases of separation or divorce, whereas this type of event does not affect male participation.  Contextual approaches to individual politicization phenomena are thus crucial in understanding how opinions are formed and how certain behaviors spread, including electoral behavior.  They give an understanding of politics in the context of both social interactions and psycho-affective interactions, which explain the politicization of individuals. 
4Political discussions in families and close relationships are spaces of confrontation and deliberation which influence choices and behavior.  By reinforcing a mutual agreement or accommodating disputes, they become a space for experiencing politics, a sort of “rehearsal” for deliberations in the public sphere.  They are, moreover, quite widely conditioned by the socio-historical contexts and national cultures which inform the manner in which the issues and antagonisms that prevail in the public sphere are expressed and confront one another. Research has shown that the intensity of informal conversations about politics and how these take place in everyday and ordinary networks of exchange depend on the history and length of democratic experience in the countries where they take place. 
5Although the involvement of these exchanges in the process of political socialization is a recognized phenomenon, and although they clearly influence many judgments and opinions, the degree of this influence is difficult to define. The results of existing work in this field of research are often ambivalent, and sometimes conflicting. Several works have clearly demonstrated the influence of demographic and socioeconomic factors on the frequency of political discussions.  Others have shown the role of strictly political factors such as partisan identification, the degree of interest in politics, or the level of knowledge.  Moreover, several works have examined how political homogeneity (versus heterogeneity) in the interpersonal and affective networks of individuals affects the conditions of politicization, but without any definitive conclusions. It remains difficult to make any definitive statement on the impact of diverging opinions. Some believe that they reinforce the cognitive capacities and knowledge level of individuals, whereas others see them as a hindrance to political participation, and yet others assert that they have no proven influence.  Certain studies even state that opinions, formed in the context of exchanges and interactions, belong less to those who form them, and more to the networks within which they take shape. This invites us to consider the plurality of opinions held by individuals, as well as how these opinions are modified depending on the discussion groups within which they develop. 
6Although politics is far from the only subject discussed in families, the family is nevertheless the context in which politics is most discussed. How these discussions take place, their content, and the place of voting within them, are little or poorly understood in French political sociology. In order to fill this gap, this study draws primarily on original research specifically focused on understanding affective interactions concerning politics in the individual’s inner circle, but also on other national and European quantitative surveys.  This study allows us to evaluate the place and intensity of political opinions in private conversations, within couples, in families, or among friends. It identifies certain social, psycho-affective and political drivers for this form of politicization via intimate life, which takes place in the space of interpersonal and private relationships. It therefore falls within an approach which takes into account the private infrastructure of choices made in the public sphere to explain the place and content of politics as it is refracted in individual existences. Beyond this, it aims to contribute to an exploration of politics with unclear and often hidden boundaries between the private and public spaces.
Talking politics: who discusses it with whom?
7France is not the country in which politics is most discussed. It is more frequently talked about in Germany, Austria, and Denmark. However, it is discussed more in France than in most countries in southern Europe.  The frequency of political discussions depends on the circumstances. During electoral periods or times of extreme polarization concerning social or societal issues, and also during major political news events (affairs, political-financial scandals, attacks, etc.), discussions occur more frequently. During France’s last presidential election in 2012, political conversations were common in households. Almost half of French people (48%) said that they often discussed the electoral campaign with their close social circle, family, friends, or work colleagues; 39% sometimes discussed it and only 13% never discussed it. 
8In the long-term, the place of political conversations has been somewhat strengthened. In the space of about 40 years, Eurobarometer surveys have indicated a large drop in the number of people who never discuss politics. From 1975 to 2006, measured in non-electoral contexts and therefore in periods of lower politicization, the percentage fell from 36% to 27%.  Nevertheless, the number of people for whom political exchanges are frequent remains quite low: 23% habitually discuss politics with four or more people, 39% with two to three people, 16% with just one person, and 22% with nobody. The relational space of these exchanges is wider for men than for women: 26% of men habitually discuss political matters with four or more people (compared to 20% of women). 
9Although political discussions have spread and become more common, the frequency of these conversations with close friends and family nevertheless varies greatly according to family configurations and the nature of interpersonal connections. Thus, many French people (69%) say that they often or quite often discuss politics with their partner. In contrast, they discuss it much less with other members of their family: only 38% say that they discuss it with their father, and 30% with their mother. Only 39% discuss politics with their child. Similarly, politics is little discussed between siblings (only 26% discuss it with their brother and 22% with their sister).  Political exchanges are therefore most cultivated in the most private context, but also within a chosen relationship of affinity.
10Women stated that they discuss politics less often than men, and that when they do, they prefer to do so in the private sphere. Men more readily discuss politics with a wider circle, with friends and work colleagues. In the Médiapolis survey (2009), 31% of men said they talk politics with friends (24% of women) and 16% with work colleagues (8% of women). Women mainly discuss politics in their family circle: 34% of them said they mainly talk about politics with their partner (compared to 22% of men), and 28% with a member of their family (also compared to 22% of men); 42% of women (compared to 27% of men) said that in forming an opinion, they mainly take into account the views of a family member.
11The surveys therefore confirm the importance of the private sphere in the politicization of women. However, women are less interested in politics than men and have a lower level of political knowledge.  In 2014, there was a 19 point difference between the declared interest of men and women in politics.  This lower female interest in politics may explain the fact that these discussions have a lesser place in women’s everyday worlds, including with close family and friends. Political discussion is therefore less conditioned by a gender effect itself than by the structural and constant effect over time of lower female politicization.
12The intensity and frequency of political discussions are closely linked to individuals’ degree of politicization. All other things being equal, although the criteria of individual sociocultural context come into play, the most defining variables are political attitude and behavior. People talk politics with those close to them, primarily because they are interested in politics. It is the individual level of politicization which counts, and which ultimately explains the intensity and frequency of the discussions. In this respect, these results are in line with the conclusions of preceding works. 
13In 2012, in the weeks leading up to the presidential election, people claiming to be very or quite interested in politics were 5.3 times more likely to discuss the electoral campaign often with friends and family than those who were not interested.  The level of political confidence appears to play a decisive role. Those asserting their political confidence in the left or the right to govern are respectively 3.4 times and 2.3 times more likely to discuss politics than those with confidence in neither side, who are much more distant from the political system. Although these “neither-nors” talk politics much less, no significant correlation can be observed between ideological orientation and the frequency of political discussions in the everyday environment. Politics is not discussed more on the left than on the right. Political behavior has been confirmed as a sociological phenomenon. Couples in which both partners belong to the socio-professional categories of employees or workers are three times less likely to regularly discuss the electoral campaign in their family or social circle.
14It is not the oldest who discuss the electoral campaign the most with their family or social circle. The elderly discuss politics less than younger age groups. In fact, compared to those aged 65 and over, 25-34 year-olds are 5.5 times more likely, and 35-49 year-olds are almost three times more likely to regularly discuss the electoral campaign with their family, friends or work colleagues than they are never to discuss it. Although we can observe a characteristic age factor establishing a link between aging and electoral participation, no such effect can be observed concerning the intensity of verbalization of politics in daily exchanges.  The relatively youthful profile of political conversation has been proven in earlier works.  It concerns less the very young than the generations at the start or in the middle of working life, engaging with politics as a result of concrete political programme issues. This observation qualifies a presupposed de-politicization in the generational dynamic; or at least, if politicization is decreasing, this does not imply disinterest in public affairs in exchanges between friends and family. Even in a context of great distrust towards the authorities and governing figures, politics occupies people’s attention, at least in terms of spectacle, its media coverage, and its debate.
15Revealing political choices involves committing oneself and opens the door to an intimate space which some might want to protect. It concerns a person’s relationship to the world and to others, it reveals their moral judgements, that which is sayable and therefore visible as soon as we reveal our convictions. These convictions are the core of personal and intimate identity, and making them public means taking risks, particularly the risk of discord and conflict.  [MM1]A majority of French people (61%) actually prefer to discuss politics with people who share their opinions, and women even more than men prefer situations where opinion converges to those where it diverges. 
16Political conversation is facilitated by the convergence of choices and ideas.  This has already been shown in the pioneering investigation by Paul Lazarsfeld and his team, at the time of the 1944 American presidential election. They observed that political discussions mostly take place between people on the same political side.  There is therefore a preference for homogamous political choices when it comes to discussion.  In fact, homogamy actually leads to more frequent discussions. There is still more political discussion with the father (38%) than with the mother (30%), although the difference is very minor (-3 points) in situations of left-wing homogamy (left-wing father/mother and ego). In right-wing families, and in cases of right-wing homogamy (right-wing father/mother and ego), the difference between father and mother is much more pronounced (-10 points). While the individual’s ideological orientation (be it left- or right-wing) appears less decisive for the frequency of political discussions with friends and family than their level of politicization, in environments of political homogamy, there is still a pronounced gap between the left and the right, which defines different customs. A homogeneous left-wing political universe between ego and both parents delineates a specific space of political socialization and creates conditions for politicization in which political discussion still seems to be widespread, or in any case more widespread than in family contexts of right-wing homogeneity (Table 1). Although political discussion is more deeply rooted and more habitual in left-wing circles, this is also because the political culture of the left implies more intense politicization, if only through its objectives to contest and transform the existing social and political order.  This is even stronger in a homogeneous and convergent interpersonal and affective context. However, in the recent period in France, the politicization of right-wing circles has been expressed during certain large-scale demonstrations translating highly polarized socio-cultural divides, such as the Manif pour tous against homosexual marriage. The increased diffusion and legitimation of protest practices and of the use of demonstrations in right-wing circles may also explain the blurring of differences between the left and right concerning involvement in political conversation.
Frequency of political discussions with father and mother according to family contexts of political homogamy and heterogamy (%)
Frequency of political discussions with father and mother according to family contexts of political homogamy and heterogamy (%)
17Although homogeneous left-wing family contexts enable political conversation to unfold more than in other types of family situation and political relationships, the place of such conversations in the exchanges between parents and children fades in comparison to the content of exchanges between partners. In couples, the frequency of political discussions is much greater than between parents and children. Over two thirds (69%) of people living as a couple say that they often discuss politics with their partner. The same increased verbalization in contexts of political homogamy is seen between partners, particularly where opinions converge on the left. When both partners are left-wing, 37% discuss politics very frequently, compared to 32% when both partners are right-wing.  In politically heterogamous couples, conversations are much less frequent: only 15% say they discuss politics very often, and 39% rarely or never discuss it. Once again, it is when both partners share the same absence of choice and distance regarding the political system that political discussions are at their rarest: 49% of couples in this situation rarely or never discuss politics (Table 2).
Frequency of political discussions within couples according to conjugal contexts of political homogamy and heterogamy (%)
Frequency of political discussions within couples according to conjugal contexts of political homogamy and heterogamy (%)
18Political discussions involve agreement more than disagreement, in the relationships of affinity which characterize a couple as well as in the forced and non-chosen relationships which define the links between parents and children. A left-wing homogamy still favors exchanges, but it seems to play an even more marked role in reinforcing political discussion in the context of relationships between ego and parents. In family or conjugal configurations characterized by a refusal to be positioned either on the left or the right and by political heterogeneity, there is significantly less discussion. No matter who the protagonists in political conversation are, the presence and recognition of an ideological affiliation, especially a left-wing one but also a right-wing one, is a factor of politicization which fosters a space of discussion and debate.
Discussing politics: subjects, issues and conflicts
19The family table remains an emblematic place for political discussion. Family memory is a mine of anecdotes and accounts, of which the traces very often remain distinct.  Asked about the attitude they adopt if a political discussion is at risk of degenerating at the family table and they disagree with what is being said, only a minority of individuals continue the discussion at all costs, risking dispute and conflict. Only 8% continue the discussion to defend their ideas, even if this makes them angry, and only 3% leave the table. The vast majority avoid conflict: 51% change the subject of conversation, while 37% continue the discussion, but without going so far as to get angry (Table 4).
Political discussion at the table. When a political discussion becomes heated at the dining table and you don’t agree with what is being said, what do you usually do?*,**
Political discussion at the table. When a political discussion becomes heated at the dining table and you don’t agree with what is being said, what do you usually do?*,*** CSP+ includes company directors, self-employed craft & tradespeople, retailers, executives, intellectual professions, and intermediate professions
** CSP- includes white-collar workers, manual labourers, and farmers
20In the distribution of answers, we can see certain significant differences compared to the social and political dynamics identified previously. Women are clearly more reluctant to risk any conflict. Meanwhile, 43% of men say they can continue the discussion to defend their ideas (compared to 31% of women). A majority of women (57%) prefer to change the subject (compared to 44% of men). The female role in the domestic space is oriented towards preserving family cohesion. The function of women is to maintain both relationships and opportunities for exchange and solidarity. Many publications have identified the specificity of their role and demonstrated the function of kinkeeping (being a “liason agent” in the family) attributed to them. 
21Young people are more willing than older people to embark upon a full discussion, even if this leads them to get angry: 16% of 18-24 year-olds are prepared to go this far, compared to 5% of people aged 65 and over, and 8% of the whole sample). This result leads us to qualify the observation of a supposed apathy among young people concerning political ideas. It confirms the effect of age on political discussions demonstrated above. Not only do older people discuss politics less, but they also try harder to avoid conflicts or adversity. Twice as many older people as young people prefer to change the subject when a family political discussion is at risk of becoming heated (61% of people aged 65 and over, compared to 31% of 18-24 year-olds).
22Individuals’ degree of politicization determines how they navigate the discussion. The more politicized the individual, the more committed they are to the discussion, and the more willing they are to risk conflict. Although 45% of those who indicate that politics plays a large role in what defines them continue the discussion to defend their ideas, 59% of those who accord it only a small role change the subject. Thus, differences of opinion have different consequences on conflict management between loved ones according to the political commitment of the individuals. Where they are heavily committed, they are more willing to face and debate disagreement, whereas those who are weakly committed prefer to avoid or deny disagreement. In fact, particularly in little-politicized environments with heterogeneous opinions, politics is often treated as a taboo in the family space, due to a complete ban on endangering family cohesion, both in its affective and symbolic economy and in concrete exchanges. 
23Ideological orientation has an impact on political dispute. The discussion is conducted differently according to whether individuals are left-wing or right-wing. Those on the left tend more towards argument and have a greater appetite for a battle of ideas, even though they still take into account the risk of conflict and some therefore choose to avoid it. Almost half (47%) continue the discussion to defend their ideas, but without getting angry, compared to just under a third (31%) of those on the right. The right prefer to avoid political discussion and confrontation: 58% said they changed the subject of conversation (compared to 39% of those on the left). We can thus observe a contrast in ideological approaches to dealing with political discussions and the possible consequences of disagreement for family harmony. Even in politically homogamous family contexts (ego/partner/father/mother all on the left, right, or neither), where we might expect more convergence of ideas, this difference persists.
24On the left, there is a pervasive culture of debate: 50% choose to continue the discussion to defend their ideas, but without getting angry (31% of those in a homogeneously right-wing context). On the right, avoidance or even ousting of politics from discussions at the family table prevails: 60% admit to changing the subject (compared to 41% in a homogeneously left-wing context). Despite the visible weakening – particularly in the younger generations – of the left-right divide in how people position and identify themselves in relation to the political sphere, the circulation of political talk among family and friends still takes place within clearly differentiated political cultures.
25The more meaning and belief people invest in politics, the more difficult it is to gloss over and silence it in the space of daily exchanges. Nevertheless, most individuals attempt to preserve relationships. Family comes before politics. Moreover, only a minority of French people (13%) say they often argue about the subject in their family. Family splits because of political disagreements do exist, but they are very marginal: only 4% say they have broken off a family relationship for this reason, generally with siblings and cousins. Parents and children very rarely fall out for political reasons. Outside of these extreme cases, political disputes nevertheless make themselves heard and impose themselves as an archetype of a scene which is at once trivial and emblematic of family life.
26France’s National Front is not only the leading subject of political dispute, but it is also the source of recurrent conflict, whatever the political leanings. National Front voters face more conflicts within the family about this subject than others. In the family space, diverging opinions are more difficult to accept and tolerate when certain protagonists make extreme choices. However, this is even more apparent when the far right is involved. Having a far right-wing partner and not sharing this political choice oneself is even more difficult to envisage than the possibility of a far left-wing partner. 
27Government policies cause differences of opinion which fuel many family discussions and disputes, particularly among far-left sympathizers. Of course, not all individuals are equally committed to conflictual discussions. Only a quarter (26%), who are among the most politicized, admit to trying to convince someone to change their ideas or opinions. In short, they argue but nobody changes sides. Even during electoral periods, only a minority of individuals (29%) admit to putting pressure on their interlocutor to change their voting decision.  Although politics and differences of opinion in the space of private life can encourage disagreements, these generally remain contained, and many tactics are used to tolerate political disagreement and maintain relationships. 
Disclosing or not disclosing your vote: means and uses of secrecy
28Political discussions broach a wide variety of subjects and themes. However, voting involves and exposes even more. Among the tactics used to avoid the risk of discord, voting secrecy ranks highly. It institutionalizes and legitimates taboo not only as a means for regulating conflict arising from diverging opinions, but also as a way of delineating a boundary to separate an intimate space which concerns only the individual from both the sphere of private exchanges and the public space. 
29Established in France since 1903, voting secrecy guarantees both voter freedom and genuine choices. Respect for secrecy is a negotiation which operates in affective exchanges, situated at the boundaries between public expression (the vote) and private conviction (personal opinion). According to Olivier Ihl, “the power of secrecy was to privatize both expression of opinion and the opinion itself”. He goes on: “The law confers the right to vote not as a matter of private law, but as a public function carried out in a personal manner”.  In electoral history, the customs and instrumentalizations of voting secrecy have varied between countries. Nevertheless, its establishment is part of a movement to individualize politics, whilst strengthening the sanctity of choice and the electoral process.  Because it allows a “safeguarding of personal opinion”, while promoting the democratic ideal of citizen involvement, it places the act of voting at the heart of individuals’ intimate lives and promotes secrecy to the level of a moral value. For Alain Garrigou, “the polling booth is a technology to provide intimacy. Designed to protect against intimidation, it places the voter in a situation where they can express the deepest part of their person, their intimate conscience”.  The recognition of the intimate inner choice as the final arbiter of the electoral decision dissociates the voter from their environment.  The polling booth symbolizes the separation between the citizen’s political role and their other social roles.  Thus, voting secrecy allows us to disclose our vote or keep it a secret. It guarantees freedom of voting and the absence of coercion, whilst serving the democratic ideal of citizen responsibility.  It allows collective involvement whilst preserving individual autonomy. It thus places the vote at the heart of the “democratic intimacy” model which according to Anthony Giddens characterizes late modern democracies.  A secret vote, although it implies a citizen choice which is necessarily part of the public sphere, ultimately remains a strictly personal and private choice. In this sense, more than any other form of political participation, the act of voting lies at the boundary between the private sphere and the public sphere.
30In the context of a psychic and affective economy of interpersonal connections within the family, in the couple, or between friends, voting secrecy avoids any risk of confrontation as well as any form of intrusion. The use of secrecy can function as a defense mechanism, similar to a “denial pact” in the words of René Kaës, aiming on the one hand to use denial and suppression to avoid divisions resulting from destructive urges, and on the other hand, to preserve connections through unconscious alliances implying self-effacement and the unspoken. 
31Voting secrecy remains strongly rooted in family lives, and for many people, family discussions avoid the subject of voting. A majority of French people do not tell their father how they are voting (52%). They are less secretive with their mother (46%) or with their children (43%). Only a minority talk about their vote with siblings: just 43% tell their brother and 42% tell their sister.  With friends, secrecy is even more widespread (57%) (Table 5). Only conjugal relationships break the code of silence. The vote is shared property in the economy of the couple. Politics is most frequently discussed with partners, and it is also with partners that transparency of electoral decisions is most common. In fact, 81% of French people tell their partner who they are voting for, and the electoral act appears hard to dissociate from an affinity norm which assumes that this information will be shared, or at the very least that differences will be respected. However, this appears even easier given that three quarters of couples actually share the same political choices, frequently leading to the same vote.  Almost three quarters of people living as a couple (73%) vote similarly. At the moment of voting, 29% of respondents admitted that they could be influenced by discussions with their partner; this is less the case for discussions with family (17%) and with friends (16%). In all cases, the couple, even if not particularly politicized, offers a favorable context for building converging electoral choices: 65% of people who say that politics plays only a small role in what defines them still vote the same as their partner.
32Voting transparency has become a conjugal norm. Nevertheless, one in five couples keep their vote a secret, preferring to leave it unsaid despite their supposed intimacy, without any noticeable connection to age or significant social factors. Over almost forty years, there has been a movement towards voting convergence between partners. In 1978, 46% of couples said they voted differently. In 2011, only 22% said this.  The transformation fits with the rise of the “confluent love” model identified by Anthony Giddens.  This growing similarity comes in spite of greater recognition of individual autonomy of choice becoming an imperative in the conjugal order, and more widely in the family order.  It is also accompanied by greater transparency. Whereas in 1978, only 66% of French people said they knew who their partner voted for, in 2011, 88% said this: an increase of over 20 points. Intra-conjugal interactions arising from political convictions and ideas are judged more on the basis of a relationship of equality between the couple. In this domain, the couple wishes and claims to be equal. In fact, only 11% of those asked admitted that they explicitly sought to influence their partner politically. The vast majority of cases display equality and reciprocity. Either the influence is recognized as reciprocal (31% of cases) or, most frequently, it is asserted that each individual maintains their position without seeking to influence the other (52%). However, although women very clearly proclaim their free will and independence of views and opinions (with 60% of them considering that each partner maintains their positions without attempting to influence the other, compared to only 44% of men), men more commonly claim that they have an influence over their partner (17% compared to just 5% of women). These results match those found elsewhere, which despite showing increased autonomization of each partner in the conjugal relationship, nevertheless observe the continuation of the predominant – proclaimed or real – political influence of men.  Although partners underplay the processes of influence, the coming together of political choices within a couple demonstrates the impact they have (even an implicit impact) in everyday political interactions.  Moreover, there is a clear discrepancy between the normative discourse emphasizing a predominantly egalitarian model of conjugal politicization, and the reality of exchanges and practices in which the model of conjugal politicization remains largely gender-differentiated.
33Disclosure of voting choices to friends is not only much less widespread, but this possibility is also more dependent on age and qualifications.  Young people, although disposed to talk politics, are nevertheless more reticent than their elders to tell their friends how they are voting. The importance of the peer group, as well as the weight of a horizontal socialization which has come to dominate and prescribe norms and values in the construction of social, cultural and political identities, might explain this reluctance. Under the yoke of a “tyranny of the majority” in the words of Dominique Pasquier, young people, seeking a means of expression and personal realization between generational conformity and authenticity, can adopt strategies which involve dodging and avoiding certain subjects.  The vote is seemingly the product of a strictly personal space and an unsharable intimate experience. It is connected to the expression of an authentic and profound self which requires preservation. Although it has no impact on voting transparency in couples, the level of educational qualification creates significant differences in the circle of friends. Voting secrecy is even greater where qualification levels are lower. Between partners, no such effect is observed (Graphs 1 and 2).
Disclosure of voting choices to partners or friends according to age
Disclosure of voting choices to partners or friends according to age
34Just like the frequency of political discussions, voting transparency depends on how intensely individuals are politicized. People who accord little importance to politics are far more likely to keep their vote secret, whoever their interlocutor within the family group, including their partner. The disclosure of this choice depends on their level of political commitment: the lower it is, the more frequently they avoid the subject. However, in this case, this use of secrecy is less a result of a strategy to preserve confidentiality of choices or to avoid conflict than a situation of indifference or non-communication regarding the subject (Graph 3). It is also among people who are neither left- nor right-wing that secrecy is most pronounced: only 42% tell their children how they are voting. The silence which prevails is the fruit of a wider indifference and an absence of communication regarding this subject.
Disclosure of voting choices according to the place occupied by politics
Disclosure of voting choices according to the place occupied by politics
35In couples, the norm of voting transparency is generalized and depends little on political leanings (Graph 4). However, outside of the couple, voting secrecy is more characteristic in right-wing than left-wing circles. In the universe of the right, it is the product of a political habitus that prioritizes individual free choice as well as a relative privatization of convictions, and consequently it can be claimed as such. In the political culture associated with the left, the “us” takes priority over the “I,” and the disclosure of voting choices is therefore quite natural, whoever the interlocutors in the family and friendship circle. Where 55% of individuals considering themselves to be right-wing disclose their vote to their children, this figure is 73% for those considering themselves to be left-wing. Voting secrecy mainly concerns the traditional right and the sympathizers of Nicolas Sarkozy. The latter display the greatest voting silence in the friendship circle. The French National Front’s electorate is also heavily affected, but less than Sarkozy’s electorate. It is left-wing voters who appear the most transparent of all, whatever the interpersonal relationships examined. The differences are small, except that in this latter electorate, voting transparency is more frequent between ego and mother than ego and father. The dominant role of mothers in political socialization and in transmission is confirmed even for the most intimate exchanges which allow disclosure of voting choices.  Moreover, except among the electorate of the National Front, voting secrecy is less common in “descending filiation” than in “ascending filiation”; in other words, people are more likely to tell their children who they are voting for than they are to tell their parents. The perspective that even if voting intentions cannot be transmitted they may still act as examples to be followed, may motivate the disclosure of voting choices, particularly on the left.
Disclosure of voting choices during the first round of the 2012 French presidential election
Disclosure of voting choices during the first round of the 2012 French presidential election
36The more the individual moves in a homogeneous environment from the point of view of their political choices, the more likely they are to engage in voting transparency with their interlocutors (Tables 4 and 5). Situations of political heterogamy, and therefore of potential disagreement, are more likely to foster voting secrecy. This facilitating effect of political homogamy is even more apparent in left-wing circles. Voting transparency is highly dominant in these circles, whatever the type of interpersonal relationships: conjugal, of course (92%), but also parental (in the case in point, 69% tell their father and 72% tell their mother who they are voting for, compared to 48% and 54% respectively in the whole sample). This is also visible between siblings and in the friendship circle. In homogamous left-wing contexts, people are more likely to tell their brother or sister their voting decision (62% and 56% respectively, compared to 43% and 42% in the whole sample), but they are also more likely to tell their friends (62% when the friendship circle is homogeneously left-wing, compared to 30% when it is heterogeneous and politically divergent).
37The predominance of “descending filiation” in disclosure of the vote – disclosing your voting intentions to your children – is particularly apparent in politically homogamous left-wing families: 84% of people who share the same left-wing ideological orientation as their partner and child reveal their voting choices to their child. In the same situation of homogamy but for right-wing families, the vast majority of parents are also transparent about their vote with their children, but the percentage is lower (70%). Whatever the interpersonal configurations, political heterogeneity and situations of diverging opinions are much less favorable to voting transparency. Only 43% of individuals not sharing the orientations of both or one of their parents tell their father how they are voting, and 50% tell their mother. Only 48% of people in a politically heterogamous couple or who do not share the same political choices as their child tell this child how they are voting. In environments where there is no left-wing or right-wing political affiliation, which are more distant from the political system and in which abstention is greater, silence concerning the vote is dominant. Only couples escape this silence caused by indifference or apoliticism. Even when partners share the same refusal to position themselves on the left or the right, two thirds of them (67%) nevertheless mutually disclose their votes.
38Although right-wing homogamy also facilitates voting transparency, its influence is lower than that seen in homogeneously left-wing environments. The political culture of the right, in which the individualization of the relationship to politics is traditionally greater, maintains a privatization of voting, with the culture of secrecy still remaining an important point of reference. This greater silence on the right has already been observed by Guy Michelat and Michel Simon in 1977, in their works on the connections between class, religion, and politics. They attribute it to the Catholic culture, which is traditionally more suspicious of politics and of the risk of division it brings. 
39The importance of voting secrecy in the space of family exchanges creates a situation where people know relatively little of how their friends and family vote. Only 36% of French people know for certain who their mother voted for, 42% say they just have some idea, and 22% have no idea. Concerning the father’s vote, the proportions are equivalent: 35%, 40% and 25% respectively. People do not know precisely how their children vote: 30% are certain, 43% have some idea, and 27% have no idea. Finally, those who do not know how their sibling votes are dominant: only one in five French people (20%) are certain of how their sibling voted, just over a third have some idea (35%), and 45% do not know at all. Even within couples where voting transparency and intense discussions are common, we can observe a significant margin of approximation. Although 63% say they know for certain how their partner voted, 25% only have some idea, and 12% have no idea.
Disclosure of voting choices according to situations of political homogamy or heterogamy: ascending filiation and descending filiation (%)
Disclosure of voting choices according to situations of political homogamy or heterogamy: ascending filiation and descending filiation (%)
Disclosure of voting choices according to situations of political homogamy or heterogamy: partner and friends (%)
Disclosure of voting choices according to situations of political homogamy or heterogamy: partner and friends (%)
40* * *
41This study of political discussion in the private space of family life, conjugal life, and life with friends cannot claim to cover the full complexity of affective exchanges and individual political dispositions, which are difficult to integrate and grasp via quantitative investigations and closed or multiple-choice questions. Nevertheless, it sheds light on certain mechanisms for the politicization of individuals in and through intimate life, as well as on certain phenomena which reveal the affective dimension of all political existence. It thus contributes to the identification of certain social, political, and psycho-affective drivers in the political socialization of individuals via their primary groups and close circles. What lessons can be taken from this?
42Firstly, the centrality of the conjugal relationship: in the circle of friends and family, the couple is an important space of socialization and politicization. Politics occupies a significant place in the affective economy, but also in everyday exchanges. People talk the most with their partner, and it is also with their partner that they state and discuss political choices, particularly voting choices, most transparently. The conjugal relationship assumes much greater mutual political knowledge than other relational circles. The intimacy of the couple accommodates the intimacy of the vote. The couple is not a place for confiscation of political expression. Quite the opposite, it appears as a basic cell of democratic life, a space of coexistence and experimentation with pluralism, a space of deliberation and undoubtedly also of politicization. Although not absent from certain conjugal experiences, the unsaid and secrecy are less common than in other relational circles.
43In the family sphere, voting disclosure is much more restrained than in the conjugal sphere. Although the family plays a role in the transmission of politics, this is not necessarily intentional or verbalized. The parental relationship appears to be less concerned by all forms of exchange, discussion, and confrontation. People discuss politics less with their parents or children than with their partner. They discuss it even less with their brothers and sisters. Nevertheless, mothers occupy a specific place. Although people generally discuss politics in the strictest sense less with their mother than with their father, and although the political register is less associated with the mother, she is nevertheless favored when it comes to disclosing one’s vote, particularly where ego and parents disagree politically. She plays a privileged role in the mechanisms of intimate politicization and political socialization, including everyday discussion, and informal and infra-political conversations with one’s children. It is undoubtedly because this discussion is more frequent and familiar that voting transparency, as a form of expression of a person’s most intimate views, is more commonly expressed and experienced within this relationship.
44Gender has significant effects on the intimate politicization of individuals in the circle of family and friends. Women discuss politics less outside of the family and private space than men. Consequently, this space is likely to play a much more significant role in their relationship to politics. For women more than for men, the intimate space is an important means of socialization and politicization. Moreover, women are more reluctant to engage in argument and confrontation. They are more concerned about eliminating the sources of conflict and discord for which politics can serve as a pretext. The unspoken, avoiding discussion, and even secrecy are all possible ways of regulating interpersonal and family exchanges. Women use them more than men. Undeniably, for women family comes before politics, and all risk of conflict is to be avoided. Although recognition of parity in the political world is accepted and obligatory, and women are on the way to occupying their full place there, this greater rejection of the conflictual dimension of politics demonstrates an anthropological resistance which could lead to a renewal of the way we do and discuss politics, and perhaps of the way we govern.
45Political conversation in the circle of family and friends, its frequency, and its affective implications are highly dependent on the individual level of politicization. It is important to underline this influence of the degree of political commitment on the conditions of exchange and circulation of political discussion. It is more important than most social and cultural parameters in explaining the different forms of possible affective involvement in politics, in both discussions and close relationships. People talk about politics and about their vote primarily because they are interested in politics. When they remain distant from politics, they also remain distant from exchanges about it. However, this does not mean that silence or the unspoken are necessarily associated with a situation of depoliticization. Politicization can be expressed in ways other than discussion. Although there is less transparency about voting intentions in right-wing environments, they show greater electoral and voting participation.
46Except when it concerns a position that is neither left- nor right-wing, and therefore at a certain distance from the political system, political homogamy between family and friends always favors exchanges and discussion more than heterogamy does. Like the laws governing individuals in situations of communication and information, political discussions assume selective display as well as an effort to reduce dissonance and divergence of opinions. They require agreement to be greater than disagreement, and a level of shared opinions which facilitates mutual expression and listening. Nevertheless, the significant difference between the left and the right reveals very different practices of political expression and contexts of politicization. Left-wing political homogeneity between interlocutors who are close favors more frequent discussion and higher voting transparency. Situations of right-wing homogamy are more frequently associated with silence, refraining from discussion, and a fortiori voting secrecy. Beyond the continuing specificity of the identity markers and habitus of a left-wing political culture, the clear reinforcement of the frequency of discussions and voting transparency by political homogamy shows the importance of affective factors in driving individual politicization.
47Finally, this identification of certain politicization mechanisms at work in the private sphere reveals paradoxical tension between a movement of individualization and privatization which confirms the persistence and extent of voting secrecy on the one hand, and a movement of de-dramatization of political conflicts and of converging choices on the other. We can observe significant tensions within certain mutations of contemporary citizenship, where the intertwining of decisions produced both in the public and private sphere conditions another decision faced by the post-democratic individual: that between the desire for resemblance and respect for difference. If this does not explain the relative disconnection between the frequency of political discussions and whether or not people disclose their vote, it at least gives it a reason for existing. We can discuss politics, and even tolerate differences of opinion, whilst ultimately preserving the secrecy of the voting slip that we place in the ballot box.
Enquête Médiapolis, 2009.
The works of Philip E. Converse are a useful reference here, in particular the chapter “The nature of belief systems in mass publics” in David E. Apater (ed.), Ideology and Discontent (New York: The Free Press, 1964), 206-61; and “Nouvelles dimensions de la signification des réponses dans les sondages” in Jean-Gustave Padioleau (ed.), L’opinion publique. Examen critique, nouvelles directions (La Haye/Paris: Mouton/Éditions de l’EHESS, 1981), 189-201. See also M. Kent Jennings and Richard Niemi, Generations and Politics. A Panel Study of Young Adults and their Parents (Princeton: Princeton University Press, 1981); and the article by Annick Percheron, “Le domestique et le politique: types de famille, modèles d’éducation et transmission des systèmes de normes et d’attitudes entre parents et enfants” in Revue française de science politique, 35(5), 1985, 840-91.
See in particular Bruno Cautrès and Anne Muxel (eds), Comment les électeurs font-ils leurs choix? Le Panel électoral français 2007 (Paris: Presses de Sciences Po, 2009).
The works of Annick Percheron were pioneering in this domain. See La socialisation politique (Paris: Armand Colin, 1993). Also useful here is the article by Daniel Gaxie, “Appréhensions du politique et mobilisations des expériences sociales”, Revue française de science politique, 52(2-3), 2002, 145-78.
Over the past ten years or so, there has been increased interest in research on political socialization. Particularly noteworthy are: Katharine Throssell, “Tous les enfants de ma classe votent Ségolène”, Agora. Débats/ Jeunesses, 51, 2009, 65-78; Vincent Tournier, “Filiation et politique: la construction de l’identité et ses conséquences”, in Pierre Bréchon, Annie Laurent, Pascal Perrineau (eds), Les cultures politiques des Français (Paris: Presses de Sciences Po, 2000), 189-208; Sidney Verba, Kay Lehman Schlozman and Nancy Burns, “Family ties: understanding the intergenerational transmission of political participation” in Alan S. Zuckerman (ed.), The Social Logic of Politics. Personal Networks as Contexts for Political Behavior (Philadelphia: Temple University Press, 2005), 95-114; Alan S. Zuckerman, Josip Dasovic and Jennifer Fitzgerald, Partisan Families. The Social Logic of Bounded Partisanship in Germany and Britain (Cambridge: Cambridge University Press, 2007).
These results come from an analysis of the Cevipof online post-electoral survey carried out in May-June 2012 (n = 2,509). Regarding my own works on family political heritage (filiation politique) in France, see (among others) A. Muxel, Avoir 20 ans en politique. Les enfants du désenchantement (Paris: Seuil, 2010), ch. 3, “La famille joue-t-elle toujours un rôle dans la transmission politique?”.
Cf. François Buton, Claire Lemercier and Nicolas Mariot, “The household effect on electoral participation: a contextual analysis of voter signatures from a French polling station (1982-2007)”, Electoral Studies, 31, 2012, 434-47; and “A contextual analysis of electoral participation sequences” in Philippe Blanchard, Felix Bülhman and Jacques-Antoine Gauthier (eds), Advances in Sequence Analysis. Theory, Method, Applications (New York: Springer, 2014), 191-212.
On this, see Alain Lancelot, L’abstentionnisme électoral en France (Paris: Presses de Sciences Po, 1968); or Janine Mossuz-Lavau, Mariette Sineau, “Sociologie de l’abstention dans huit bureaux de vote parisiens”, Revue française de science politique, 28(1), 1978, 73-101.
See the very enlightening study by Céline Braconnier, “Voter ensemble: dispositifs informels de mobilisation et compensation des inégalités de politisation”, in Laurent Le Gall, Michel Offerlé, François Ploux (eds), La politique sans en avoir l’air. Aspects de la politique informelle, 19e-21e siècle (Rennes: Presses Universitaires de Rennes, 2012), 355-84.
Marieke Voorpostel and Hilde Coffé, “Transitions in partnership and parental status, gender, and political and civic participation”, European Sociological Review, 28(1), 2012, 28-42.
This is shown by C. Braconnier in Une autre sociologie du vote. Les électeurs dans leur contexte: bilan critique et perspectives (Cergy-Pontoise: LEJEP Éditions, 2010). See also S. Verba, Small Groups and Political Behavior (Princeton: Princeton University Press, 1961); and R. Robert Huckfeldt, Politics in Context. Assimilation and Conflict in Urban Neighborhoods (New York: Agathon Press, 1986).
By recently opening up a field of research on the theme of “Politics, the affective and the intimate”, I specifically addressed the contextualization issues arising from interpersonal and affective relationships. A useful reference here is A. Muxel, Toi, moi et la politique. Amour et convictions (Paris: Seuil, 2008), recently published in English as Politics in Private. Love and Convictions in the French Consciousness (New York: Palgrave Macmillan, 2014); and A. Muxel (ed.), La vie privée des convictions. Politique, affectivité, intimité (Paris: Presses de Sciences Po, 2014).
On this point, see Jeffrey Levine, “Choosing alone? The social network basis of modern political choice” in Zuckerman (ed.), The Social Logic of Politics, 132-51. Also relevant here is the work of Diana Mutz, whose research particularly focuses on the phenomenon of converging and diverging opinions. See in particular the article “The consequences of cross-cutting networks for political participation”, American Journal of Political Science, 46(4), 2002, 838-55. Finally, see also the works of Alan S. Gerber, in particular: “Disagreement and the avoidance of political discussion: aggregate relationships and differences across personality traits”, American Journal of Political Science, 56(4), 2012, 849-74.
This expression is used in the article by Pamela Conover, Donald Searing and Ivor Crewe, “The deliberative potential of political discussion”, British Journal of Political Science, 32, 2002, 21-62. Robert Huckfeldt too has shown that political conversations at home play a major role in the formation of opinions, particularly during electoral campaigns. See his work, Politics in Context. Assimilation and Conflict in Urban Neighborhoods (New York: Agathon Press, 1986). In France, some particularly interesting recent doctoral theses have focused on interactions about politics between family and friends: see Julien Audemard’s thesis, defended in 2013 at Montpellier University, “Influences interpersonnelles: comment les contextes structurent les opinions et les votes”, and Charlotte Dolez’s thesis, defended at Sciences Po Paris in 2013, “L’écume des news: sociologie des usages des informations à partir d’entretiens de couples”.
For a comparative interpretation of political discussions according to socio-political and national contexts, see Ken’ichi Ikeda, Laura Morales, and Michael R. Wolf (eds), Political Discussion in Modern Democracies: A Comparative Perspective (London: Routledge, 2010). On taking into account the length of democratic experience, see in this book the analysis of the Hungarian case presented in the chapter by Oana Lup “The role of political discussion in developing democracies: evidence from Hungary”, 183-200. See also Rüdiger Schmitt-Beck and Oana Lup, “Seeking the soul of democracy: a review of recent research into citizens’ political talk culture”, Swiss Political Science Review, 19(4), 2013, 513-38.
See (among others) Nina Eliasoph, Avoiding Politics. How Americans Produce Apathy in Everyday Life (Cambridge: Cambridge University Press, 1998); or Charles Pattie, Ron J. Johnston, “Context, conversation and conviction: social networks and voting in the 1992 British General Election”, Political Studies, 47, 1999, 877-89.
See for example Pamela Conover, Donald Searing and Ivor Crewe, “The deliberative potential of political discussion”, 21-62.
For a useful chapter presenting the controversies around this point, see R. Huckfeldt, “Information, persuasion and political communication networks”, in Russel J. Dalton, Hans-Dieter Klingemann (eds), The Oxford Handbook of Political Behavior (Oxford: Oxford University Press, 2007), 100-22. See also D. Mutz, “The consequences of cross-cutting networks for political participation”; or Robert Huckfeldt, Jeanette Morehouse Mendez and Tracy Osborn, “Disagreement, ambivalence, and engagement: the political consequences of heteregenous networks”, Political Psychology, 26, 2004, 65-96. On the emotional consequences of political disagreement in interpersonal networks see Bryan M. Parsons, “Social network and the affective impact of disagreement”, Political Behavior, 32(2), 2010, 181-204.
On this point, see the very interesting theories of Alexis Ferrand in Appartenances multiples. Opinion plurielle (Villeneuve-d’Ascq: Presses Universitaires du Septentrion, 2011). He writes: “In a situation of uncertainty, an actor reflects and is able to position him- or herself thanks to the discussions which result from a relationship with another actor. The minimal (socio)logical consequence of this is that the origin of the opinion is not the actor’s mind, but the dialogical process constituting the relationship” (p.103, translated from the original French). Consequently, individual opinion is necessarily plural, particularly in the context of uncertainty. “An actor discussing with partners in relationships belonging to heterogeneous contexts can hold several opinions on the same subject according to the relationships in which these are expressed” (104).
“Famille, amour et politique” (“Family, love, and politics”) is an original survey which I have conducted as part of a research contract with the Cevipof (the Sciences Po Political Research Center). It was managed by Opinion Way in June and July 2011, using the CATI database. It targeted a representative sample of the French population aged 18 and over (n = 1,908).
The regular Eurobarometer surveys under the European Commission allow the comparison of a certain amount of socio-political data collected in all countries of the European Union.
Cevipof post-electoral survey, May and June 2012.
Cited by Nonna Mayer, based on data analyzed by Vincent Tiberj using the Mannheim Eurobarometer Trend File (1975-2006), in the chapter “Parler politique”, of her work Sociologie des comportements politiques (Paris: Armand Colin, 2010), 258.
Results taken from the Médiapolis survey, 2009.
Results taken from the survey “Amour, famille et politique”, Cevipof, July 2011. If the respondent had several children or several brothers and sisters, they were asked to answer regarding the eldest.
Useful references here are: Jean Chiche, Florence Haegel, “Les connaissances politiques”, in Gérard Grunberg, Nonna Mayer, Paul Sniderman (eds), La démocratie à l’épreuve. Une nouvelle approche de l’opinion des Français (Paris: Presses de Sciences Po, 2002), 273-87; J. Mossuz-Lavau, “Genre et politique. La marche vers l’indifférenciation”, in A. Muxel (ed.), La vie privée des convictions. Politique, affectivité, intimité (Paris: Presses de Sciences Po, 2014), 169-85; M. Sineau, “Les paradoxes du gender gap à la française”, in B. Cautrès, N. Mayer (eds), Le nouveau désordre électoral. Les leçons du 21 avril 2002 (Paris: Presses de Sciences Po, 2004), 207-29; Réjane Sénac, L’ordre sexué. La perception des inégalités femmes-hommes (Paris: PUF, 2007); Jeffery Mondak and Mary Anderson, “The knowledge gap: a reexamination of gender-based differences in political knowledge”, The Journal of Politics, 66, 2004, 492-512.
Baromètre Confiance politique, 5th wave, Cevipof. Survey carried out from 25 November to 12 December 2013 on a sample of 1,803 people, representative of the French population aged 18 and over and registered on the electoral lists. Results available on the Cevipof website.
These results come from the statistical analysis of data from the post-electoral survey carried out by Ifop for Cevipof in May and June 2012. N. Mayer, in her chapter “Parler politique”, having analyzed the data for the 2007 French electoral panel, makes the same observation. Although a certain number of socio-demographic variables have statistically significant effects on the tendency to talk politics, all other things being equal, it is familiarity with politics which is most important. “Subjective political competence influences the tendency to discuss politics more than objective competence” (translated from N. Mayer, “Parler politique”, 262).
Results taken from the Cevipof post-electoral survey, Ifop, May and June 2012.
See Pierre Bréchon “L’abstention: de puissants effets de génération?” in A. Muxel (ed.), La politique au fil de l’âge (Paris: Presses de Sciences Po, 2011), 91-111.
See P. Conover, D. Searing and I. Crewe, “The deliberative potential of political discussion”.
The article by Pamela Conover et al., cited above, includes this quotation from a study which provides a good summary of the personal commitment involved in making our political opinions public: “You hit someone’s nerve, and you know, politics is close to religion in that it’s my little politics, these are my beliefs, and when you step on them or threaten them there’s a flight or fight reaction.” (Article available online at https://www.researchgate.net/publication/231853364_The_Deliberative_Potential_ of_Political_Discussion, accessed 4 February 2016). Verbalizing one’s political choices has intimate implications which commit individuals in their affective relationships. They can’t be voiced to just anyone. Discussing politics requires a certain degree of proximity and opens the way to a privileged relationship. In my qualitative study on the place of politics and the impact of divergence or convergence of choices on affective relationships, one respondent stated that: “You become close to someone when you can talk about love and politics.” See my work, Toi, moi et la politique.
“Famille, amour et politique” survey, Cevipof, July 2011.
See the works of Lindsey Clark Levitan and Penny S. Visser, who show that individuals, whatever their social background, prefer exchanges with people who share their universe of choices, values, and convictions: “The impact of the social context on resistance to persuasion: effortful versus effortless responses to counterattitudinal information”, Journal of Experimental Social Psychology, 44(3), 2008, 640-9.
Paul Lazarsfeld, Bernard Berelson and Hazel Gaudet, The People’s Choice. How The Voter Makes Up His Mind in a Presidential Campaign (New York: Columbia University Press, 1944).
The indicator of political homogamy in the case of the political relationship with one’s parents (“ascending filiation”) was constructed based on the question: “Would you say that you are more left-wing, more right-wing, or neither? What about your father? And your mother?” (“Famille, amour et politique” survey, Cevipof, July 2011). Ego/father/mother triads sharing a left-wing, right-wing, or neither left- nor right-wing position are considered to be “homogamous”. When the positions of ego and parents, or the positions of the two parents are different, they are considered to be heterogamous.
The impact of ideological orientation on the relationship to politics has been examined in many works in the fields of both political socialization and political sociology, in an attempt to explain the links between social and cultural background and individual political predispositions. It is not possible to cite all of these works. However, it is useful to refer to the recent contribution of Jennifer Fitzgerald, who confirms the specificity of the left-wing universe in terms of the political attitudes and behaviors of individuals: Jennifer Fitzgerald, “What does ‘political’ mean to you?”, Political Behavior, 35(3), 2013, 453-79.
The indicator of political homogamy in couples was constructed based on the question: “Would you say that you are more left-wing, or right-wing, or neither? What about your partner?” (“Famille, amour et politique” survey, Cevipof, July 2011). Couples in which both partners share the same left-wing, right-wing, or neither leftnor right-wing orientation are considered to be politically “homogamous”. When ego and partner have different political positions, they are considered to be “heterogamous”.
On the place of family meals in individual and collective memory, see my book, Individu et mémoire familiale (Paris: Hachette-Pluriel, 2007 [1st edition, 1996]).
See Carolyn Rosenthal, “Kinkeeping in the familial division of labor”, Journal of Marriage and Family, 47(4), 1985, 965-74; or Jean-Hugues Déchaux, “Les femmes dans les parentèles contemporaines: atouts et contraintes d’une position centrale”, Politiques sociales et familiales, 95, 2009, 7-17.
In my work Toi, moi et la politique, I present the different “faces” of political agreement and disagreement, particularly that of the Taboo (205-16).
See my chapter “La politique et les proches: faut-il être d’accord pour s’aimer?” in A. Muxel (ed.), La vie privée des convictions. Politique, affectivité, intimité (Paris: Presses de Sciences Po, 2014), 75-95. For a majority of French people, it is hard to imagine their partner having extremely right-wing ideas (impossible for 29% and difficult for 25%). They find the far left less off-putting (impossible for 12% and difficult for 24%).
Results from the analysis of the Cevipof post-electoral survey, May and June 2012. A logistic regression shows that people who say they are interested in politics are six times more likely to try and convince someone to change their vote than people who are not interested. This commitment to discussion appears even stronger when individuals are left-wing or right-wing. People who place their confidence in the left to govern are 2.4 times more likely, and people who place their confidence in the right are 2.1 times more likely to adopt this approach than individuals who classify themselves as neither left-wing nor right-wing. However, we can observe a slightly more widespread culture of debate and persuasion among left-wing families (1.5 times more likely to try and convince someone to change their opinion than in apolitical families which are neither left- nor rightwing; no significant effect can be observed for the right).
My work, Toi, moi et la politique, presents the different types of political disagreement and agreement through which convictions are negotiated in the space of private life.
Michaël Foessel makes an entirely heuristic distinction between intimate, private, and public. Intimate is not the same as private. It assumes a relationship to Others, it is the space in which otherness is negotiated. See La privation de l’intime (Paris: Seuil, 2008).
See Olivier Ihl, “Vote public et vote privé” in Pascal Perrineau, Dominique Reynié (eds), Dictionnaire du vote (Paris: PUF, 2001), 960-5 (960). Quotation translated from the original French.
For a historical perspective on voting secrecy, see Romain Bertrand, Jean-Louis Briquet, Peter Pels (eds), Cultures and Voting. The Hidden History of the Secret Ballot (London: Hurstand, 2007); or Yves Déloye, Sociologie historique du politique (Paris: La Découverte, 2003), ch. 4.
Alain Garrigou, “Le secret de l’isoloir”, Actes de la recherche en sciences sociales, 71, 1988, 22-45 (44); quotation translated from the original French.
A. Garrigou goes on: “Imposing the polling booth not only protects the voter from pressure, it also imposes a new definition of the voter as an abstract, rational individual who is dissociated from their social environment, in the accomplishment of a specifically political role.” (“Le secret de l’isoloir”, 44).
Stein Rokkan, “Mass suffrage, secret voting, and political participation”, Archives européennes de sociologie, 2, 1961, 132-52.
Annabelle Lever, On Privacy. Thinking in Action (New York: Routledge, 2011).
Anthony Giddens, The Transformation of Intimacy. Sexuality, Love and Eroticism in Modern Societies (Cambridge: Polity Press, 1992).
R. Kaës, Les alliances inconscientes (Paris: Dunod, 2009). “The specificity of this pact is that it is made to fulfill the defensive needs of subjects when they form a connection, and to maintain this connection. Such a pact therefore fulfills a meta-defensive function for each of the subjects involved in it. It should therefore be seen as a means of resolving intra-psychic conflicts and the conflicts involved in a configuration of connections” (114), quotation translated from the French.
Results taken from the survey “Famille, amour et politique”, Cevipof, July 2011. Where there are several brothers or sisters, the person was asked to answer in relation to their oldest brother or oldest sister.
Here, there is a risk of bias connected to conformity in responses. The results are established on the basis of ego’s declaration of similarity or non-similarity of their loved one’s vote to their own vote, and we cannot dismiss the possibility of this being overestimated. Nevertheless, the differences observed on the level of structural effects remain pertinent.
Ifop-La Vie Survey, 2-8 March 1978.
In the conclusion of The Transformation of Intimacy. Sexuality, Love and Eroticism in Modern Societies (Cambridge: Polity Press, 1992), 188-9, Anthony Giddens writes that: “The possibility of intimacy means the promise of democracy […] The structural source of this promise is the emergence of the pure relationship, not only in the area of sexuality, but also in those of parent-child relations, and other forms of kinship and friendship. We can envisage the development of an ethical framework for a democratic personal order, which in sexual relationships and other personal domains, conforms to a model of confluent love.”
On the changes to the family institution in contemporary democracies, see in particular Jean-Claude Kaufmann, La trame conjugale. Sociologie du couple par son linge (Paris: Nathan, 1992); François de Singly, Sociologie de la famille contemporaine (Paris: Armand Colin, 3rd edn, 2007); and Libres ensemble. L’individualisme dans la vie commune (Paris: Nathan, 2000).
Boris Wernli shows this for Switzerland. See B. Wernli, “Homo et hétérogamie dans les attitudes et le comportement politiques en Suisse”, Revue suisse de science politique, 12(1), 2006, 33-72.
See the study by Jennings and Stoker on how couples interact about politics: Laura Stoker and M. Kent Jennings, “Political similarity and influence between husbands and wives”, in Zuckerman (ed.), The Social Logic of Politics, 51-74.
Concerning the influence of the friendship circle on individual political participation, see Laurence Kotler-Berkowitz, “Friends and politics: linking diverse friendship networks to political participation”, in Zuckerman (ed.), The Social Logic of Politics, 153-70.
Cf. Dominique Pasquier, Cultures lycéennes. La tyrannie de la majorité (Paris: Autrement, 2005).
See the chapter entitled “Bounded partisanship in intimate social units: parents and children”, in Alan S. Zuckerman, Josip Dasovic and Jennifer Fitzgerald, Partisan Families, 91-122. See also A. Muxel, “Socialisation et lien politique”, in Thierry Blöss (ed.), La dialectique des rapports hommes-femmes (Paris: PUF, 2001), 27-44.
Guy Michelat and Michel Simon, Classe, religion et politique (Paris: Presses de Sciences Po, 1977).