Political Support in Times of Personalized Politics. A Qualitative Inquiry of French Citizens' Representations About Political Actors and the Political Regime

This article investigates how citizens speak about representative democracy and questions their perceptions of representation and of the democratic regime they live in, by mobilizing the distinction between diffuse and specific support, in a context of personalized politics. It shows that political actors and their performances are at the core of citizens' perceptions. I investigate citizens' representations of the political field through an original qualitative fieldwork, composed of couple interviews with French citizens, under the Sarkozy presidency. Studies about the political support of citizens often mobilize quantitative surveys to measure the degree of support and satisfaction. I rather choose the qualitative approach to grasp perceptions of political field through discussions about political and societal issues. Couple interviews offer an adequate framework to observe political opinions that are built in daily life. Representations of the political field are mainly dominated by the role of political actors. Political parties and institutions are rarely mentioned. Politicians are systematically held accountable, and are often criticized in citizens' discussions. The existing literature has often distinguished specific and diffuse support. My analysis tends to show that the weakness of the former through personalization can undermine the support for the regime. However, alternatives to representative democracy remain underexplored and even not considered. Overall, these representations depend on sociopolitical factors, such as political convictions or social backgrounds.


INTRODUCTION
Analyses of citizens' relationship to politics reveal a certain disaffection with politics through a number of indicators, such as the decline of electoral turnout in Western democracies (Franklin, 2004), partisan disaffiliation (Mair and Van Biezen, 2001) but also the increasing distrust in political leaders (Pharr and Putnam, 2000;Hay, 2007;Citrin and Stoker, 2018). In France, the distance between those who govern and those who are governed is growing and the feeling that "politicians do not care about people like us" now concerns nearly 85% of the population in 2019 (when it concerned 40% of the population at the end of the 1970s) 1 . Recent studies show that populist movements and parties are largely based on an anti-elite discourse (Mudde, 2007), irreconcilably pitting "the people" against the political elites.
However, these findings should be qualified and put in perspective. First of all, available data does not show a massive and irremediable decline in political confidence: they rather show variations according to the institutions, actors and countries (Thomassen, 2016). Disaffection with politics can also be interpreted in different ways: some authors see it as a democratic renewal and a revitalization of commitment toward democracy, notably through the increase in protest practices and their increased legitimacy (Norris, 2011). These authors argue that we witness a transformation of the modes of the relationship with politics: "elite-directed" political activities would be replaced "elite-challenging" practices (Inglehart, 1977;Dalton and Welzel, 2014). Finally, a lot of studies insist on the legitimacy that citizens always give to democracy and their positive judgment about it (Norris, 2011;Grossman and Sauger, 2017 for the French case): the democratic regime is therefore not called into question, but representation experiences a "crisis." Some authors therefore evoke "the danger of deconsolidation" and see the political disaffection as structuring problems that could have irremediable consequences on the regime (Foa and Monk, 2016).
This distinction between regime and actors is thus particularly central to analyze the assessment of individuals' relationships to politics, and it also refers to Easton's distinction between diffuse and specific support (Easton, 1965). Diffuse support is a stable source of support for institutions and the political system: it is based on citizens' loyalty to the system, as well as the legitimacy they grant it. It is to be understood as "a reservoir of favorable attitudes or goodwill that helps members to accept or tolerate outputs to which they are opposed or the effect of which they see as damaging to their wants" (273). Diffuse support is conceived as a stock of trust and applies to the political system and the political community. Specific support, on the other hand, concerns political authorities and depends on the evaluation of the actions and performance of incumbents. It is therefore potentially more fluctuating than diffuse support. Some authors have therefore tackled the challenges of operationalizing these dimensions. Following Dalton (1999), Norris proposes indicators that enable to capture both types of support: adherence to democratic values and principles is an element in the assessment of diffuse support whereas satisfaction with the incumbent political leaders testifies to specific support (Norris, 2011). In this paper, specific support will be captured by dissatisfaction toward the political authorities (and especially politicians), whereas diffuse support will refer to a process of disaffection that can characterize citizens. This paper deals with the central question of political support in contemporary democracies characterized by personalized politics. Do people only criticize the government's performances or does their disaffection go beyond specific actors and affect the support for the political regime? Do citizens make this distinction between actors, institutions and regime when they 1 See the Political Trust Barometer of CEVIPOF, https://www.sciencespo.fr/ cevipof/sites/sciencespo.fr.cevipof/files/CEVIPOF_confiance_vague10-1.pdf. talk about politics? Do they point to a crisis of representation, without questioning the democratic regime? Easton underlines the necessity to maintain the distinction between specific and diffuse support from a theoretical and empirical point of view, even if the difference can sometimes be tenuous (Easton, 1975). This paper assesses the distinction between diffuse and specific support in a context of personalized politics but shows how these two kinds of support can enhance or diminish each other. It also aims to anchor this reflection in the prism of people's sociopolitical characteristics. I then explore the ways in which citizens represent themselves in the political field, and the relationship they have with both institutions and political actors. What form(s) do citizens' criticisms and support for the political field take in their discourse? My analysis focuses on the ways in which people think about politics and democracy, according to their socio-political characteristics. It is based on a qualitative field, consisting of semi-directive interviews with couple, the details and issues of which I set out below.

PERSONALIZATION AND POLITICAL SUPPORT
I choose to specifically question the links between the process of personalization that has been observed for several decades in contemporary democracies and political support. Indeed, citizens' satisfaction with politicians is seen as one of the central dimensions of specific support (Norris, 2011).
A lot of political science scholars study personalization since the 1970s. Farrell (1971) observed that "in almost all political systems, executive dominance and the personification of this domination in a single leader is a central fact of political life": the personality of the leaders thus becomes central in the apprehension of the political space and its functioning. Personalization (Renwick and Pilet, 2016) and presidentialization (Poguntke and Webb, 2005) characterize contemporary democracies-whether parliamentary, presidential or semi-presidential. By example, in France, this process of personalization took hold during the Fifth Republic and particularly with the direct election of the President of the Republic since 1965 (Delporte, 2008): the presidential election became the center of the French political life (Grunberg and Haegel, 2007). Some authors propose a further conceptualization of personalization, focusing on "centralized and decentralized personalization" (Balmas et al., 2014). This distinction makes it possible to refine the mechanisms of personalization in order to better understand who precisely benefits from transfers of power. "Centralized personalization implies that power flows upwards from the group (e.g., political party, cabinet) to a single leader (e.g., party leader, prime minister, and president)" and "decentralized personalization means that power flows downwards from the group to individual politicians who are not party or executive leaders (e.g., candidates, members of parliament, and ministers)" (37). The authors point out that centralized personalization is the most widespread phenomenon. The literature is also interested in personalization indicators and how these phenomena can be measured and evaluated.
These indicators make it possible to distinguish between institutional personalization, media personalization and behavorial personalization, showing the spheres of the political space in which the personalization process takes place.
One of the major debates on personalization concerns the link that the phenomenon has with party change or decline (Rahat and Kenig, 2018). The dominant position is that personalization is linked to party decline, as attested by this definition of personalization as a "process in which the political weight of the individual actor in the political process increases over time, while the centrality of the political group (i.e., political party) declines" (Rahat and Sheafer, 2007, p. 65). The rise of personalization is associated with the decline in the centrality of political groups, such as political parties. Deciding on causality is not easy, but some authors argue for a strong connection between the two phenomena. For Balmas et al., "personalization implies a decline in the role of parties-a decline that is likely to be pronounced in some or all of the functions performed by political parties." Other authors, particularly party scholars, consider that the decline of parties does not necessarily imply forms of political personalization, as other collective actors may take over from the parties. While this point remains debated, empirical analyses nevertheless show many cases where personalized politics and party decline are linked. Electoral sociology literature also emphasizes the personality of candidates as one of the factors explaining electoral choices (Wattenberg, 1991), in a broader context of the decline in partisan identification and partisan dealignment (Dalton and Wattenberg, 2000), even if the "candidate effect" should not be isolated from partisan affiliation and ideological factors (Brouard and Kerrouche, 2017). Media contribute to reinforce the personalization in politics, by providing news media coverage on candidates and leaders rather than parties and organizations (Van Aelst et al., 2012).
Personalization implies a decline in the role of parties, or political parties have been central actors of the political space and its functioning. In this regard, some authors warn of the political consequences of personalization: "Personalization undermines political parties because it engenders support for an appealing leader, not for the ideas and programs of the party as an institution . . . As the basis of political support, personalization is transitory and fragile. Massive shifts in support occur when leaders change or lose their novelty or reveal previously unpublicized qualities" (Mancini and Swanson, 1996, p. 272). In Easton's conceptualization, the political system receives inputs, i.e., citizens' demands, and produces outputs, i.e., laws and public policies (Easton, 1965). In these mechanisms, political parties constitute essential channels for expressing and taking into account citizens' demands. But the phenomenon of party decline calls this role into question, as political parties lose their monopoly of "affective and cognitive centrality" (Wattenberg, 1991). Therefore, through which channels does this transmission, necessary for the proper functioning of the political regime, take place? When political parties are no longer the "focal point of politics" (Rahat and Kenig, 2018, p. 211) but politicians are, this change questions major dimensions of political support. Are political leaders able to collect citizens' demands and deliver outputs? Do they hold the trust of citizens, compared to political parties that seem more stable and consistent? The viability and sustainability of the democratic system is here at stake. In a context of political personalization, the question of political support and functioning of the democratic regime deserves to be asked. My paper therefore questions the political consequences of political personalization and seeks to understand whether and to what extent personalized politics are able to provide the basis for political support and legitimacy, from an analysis of citizens' representations and opinions.

MATERIALS AND METHODS
Most studies focusing on political support rely on quantitative analyses. The issues of operationalization and measures are central to these explorations. The "SWD" (Satisfaction With Democracy) indicator 2 is the recurrent measure used to evaluate general support (Quaranta, 2018). This measure has been criticized since it can have several meanings (Canache et al., 2001), particularly with regard to the object it concerns (regime, institutions, actors, outputs, etc.) (Ferrin, 2016). Some works choose to go through the evaluation of democratic principles, such as equal treatment, citizen participation, freedom of the media (Ferrin, 2016). The relationship to institutions and political actors is essentially captured through the level of confidence/trust that people have in them. By mobilizing national and international databases, these studies enable to assess the level, evolution and explanatory variables of these political attitudes and behaviors (Magalhaes, 2018). They report on individual variables, such as interest in politics, economic and social status, level of education, but also national characteristics, such as the length of time democracy has been established (Dalton, 2004) to explain support for the regime and institutions. The quantitative approach can be very useful to understand the level of satisfaction with democracy and the sociopolitical variables that explain the phenomenon. But it also imposes a way of framing the questions, preventing individuals from spontaneously talking about political actors and regime. My qualitative approach aims to go beyond this limit. I defend here the idea that a discussion framework encourages the collection of citizens' political representations and perceptions.
My article gives an account of the representations of individuals about the political field and the ways in which they talk about it, questioning the distinction between regime, institutions and actors. To do so, it mobilizes a qualitative analysis of a field survey composed of semi-directive interviews. I therefore rely here on interviews conducted as part of my doctoral thesis about the uses of information in the construction of the ordinary relationship to politics (Dolez, 2013). This fieldwork is composed of 27 semi-directive couple interviews conducted between 2010 and 2011 in France. Many way wonder to what extent empirical interviews dating from one decade ago are still relevant to address these questions in our contemporary context. In this period, Nicolas Sarkozy has been President of the French Republic since 2007 and his way to exert this role was described as "hyper-presidency" (Maigret, 2008). In a French system characterized by a long trend of presidentialization, President Sarkozy is the embodiment of personalization (Neveu, 2012) and his presidency constitutes a favorable context to observe and understand the links between personalization and assessment of the political regime. Some authors propose to consider the Sarkozy presidency as the "symptom of a heavy tendency, " that of the individualization of the political field (Le Bart, 2013). Subsequent presidencies, that of Holland and the current Macron presidency, have shown the same trend. The Political Confidence Barometer conducted by the CEVIPOF shows the extent to which, between 2009 and 2019, mistrust of politicians dominates. Politicians are mostly accused of being disconnected from everyday life and citizens and of being corrupt. During this period, the Presidents of the Republic and successive Prime Ministers have very low confidence levels, around 30% of the population. The years 2010 thus seem to be largely indicative of current trends (Cautrès, 2019). Admittedly, cross national analyses do not attribute a very high score to France in terms of political personalization (Rahat and Kenig, 2018, p. 199), yet individualized perceptions of the political field by citizens are very present.
Fifty four people have been interviewed and are individuals of various social backgrounds, places of residence, levels of political interest and ideological preferences. The interviews lasted between 1 h and 15 min and 4 h. Interviewees come from a working (18), middle (20), or upper-class background (16). They live in different areas: Paris (8), the Paris region (16), and the Nord-Pas-de-Calais region (30). There are two reasons for this geographic diversification. First, the geographical variable makes it possible to evaluate the relationship of individuals to politics according to their territory, differentiated by the distance to the center of power. It enables me to take into account the distinction "center-periphery" which has been constituted as a central political cleavage (Lipset and Rokkan, 1967). Second, the geographical diversification is also a way to grasp individuals with various social backgrounds. My social networks are more diversified in terms of social backgrounds in the Nord-Pas-de-Calais region, where I grew up and where a part of my family lives, contrary to Paris, where the proportion of higher education graduates is more important than in the rest of France. The recruitment of interviewees was 2-fold. First, 40 interviewees were recruited through persons of my social networks that oriented me toward persons I didn't know. Second, 14 people were recruited by ads I put on the streets and in shops. The ad specified that I was looking for two people living together for an interview based on a discussion about society topics (see below). These participants were compensated up to 20 euros per person in gift vouchers. This compensation was an undeniable way to get access to working class people (10 out of 18), who are usually underrepresented in qualitative surveys. To interview on political topics lay citizens about who are not necessarily interested in politics, the compensation can facilitate the participation of socially diverse people (Duchesne and Haegel, 2004). The corpus has therefore been built on a logic of diversification, which makes it possible to exhaust the profiles and to obtain a varied panorama of relations with politics. The qualitative approach here allows us to grasp the representations and mechanisms of politicization of individuals.
The choice of a qualitative approach is compatible with the exploration of the relationship to politics and the political representations of citizens, as Brigitte Le Grignou points out: "claiming to grasp politics in its 'ordinary, ' everyday dimension implies recourse to observation, in-depth interviews, focus groups, or to qualitative techniques and methods alone likely to refine the data produced by questionnaires or surveys" (Le Grignou, 2003, p. 197). I chose to interview couples and to build the interview around discussion on political and social topics to observe, in interaction, people's perceptions and judgements.
Why using couple interviews? This familiar discussion framework refers to banal, everyday situations in discussions on political or social topics, and in this sense makes it easier for individuals to voice their opinions (Braconnier, 2012). Couple interviews also turn out to be an appropriate framework of discussion. This interview situation is less artificial than the face-to-face interview and enabled me to observe the interactions between citizens. Group interviews are particularly relevant in revealing arguments and belief systems (Kitzinger, 1994) or ways of approaching public problems (Comby, 2011). Group interviews therefore facilitate access to individuals' representations, as these are more easily expressed in conversations than in an interview. More specifically here, the couple framework is one of the major frameworks for constructing individuals' political visions. The couple framework seems to favor the expression of the opinions and representations of the interviewees. In fact, the face-to-face situation with the interviewer can be embarrassing for interviewees when they are confronted with a question they cannot answer. When there are two interviewees, interviewees can be supportive. However, the couple's setting can also be a constraint and may reflect certain biases. The presence of the couple has consequences on the content of the interview: in this sense, this interaction constructs what is said and what can be said for each of the participants. In some cases, couple interviews can reveal mechanisms of male domination, where, in the couple, men are usually more interested in politics than women (Sineau, 2013) and the former undertake the political work whereas the latter can censor themselves when talking about political issues.
Why building interviews from discussion topics (Gamson, 1992)? The interview guide was designed to offer both interviewees discussion topics, introduced as "society topics 3 ." It therefore does not address directly the question of the judgement and evaluation of institutions and political actors but the interview guide enables me to capture them thanks to a situation of discussion. In this sense, this use of the interview helps to avoid the pitfalls and criticisms that can arise when it comes to apprehending the relationship with politics. Nicolas Mariot wonders how political opinions and behaviors can be empirically grasped and indicates that "the interview forces reflexivity about practices" whereas in their everyday life, people do not need to justify their political practices (Mariot, 2010, p. 187). The interview would therefore tell interviewees to justify their behaviors and opinions and would thus struggle to grasp indifference to politics. In my case, the interviewees are not asked to produce a discourse specifically on the judgment of institutions, support for the regime or the relationship with political actors, but through discussions on political issues, they report representations, perceptions and opinions that provide information about these dimensions. In this way, I analyze the words they use and the elements they identify to talk about the political field. This detour through political issues is also a way to avoid domination effects by using a specialized vocabulary. I was indeed confronted with this issue in this study. At the end of the discussion, I asked a question on the evaluation of democracy: "We often wonder whether democracy works well in our societies, what do you think about it in the case of France?" This question did not turn out to be relevant, in particular because it might not be understood by some interviewees and might make them feel incompetent. A good example is the reaction of Patricia and Gérard to this question. Patricia, 50, is married to Gerard, 52. Holder of a vocational training certificate in carpentry, Gérard has been a roofer in a private company for 32 years. Patricia has a vocational training certificate in typing and is currently unemployed. In her last job, she worked as a domestic helper. They express little interest in politics, but the interview reports a lively discussion on the topics. To my question about democracy, they answer: This exchange shows the extent to which the wording of the question on the functioning of the regime is not adequate and echoes instead to an academic questioning 4 . The challenge is to interviews without preventing participants from bringing up the subjects that (pre)occupied them in particular. 4 Patricia's reply, in which she admits not knowing about the topic, could be inhibiting for the rest of the interview because she is put in default during the interview. This question is asked almost at the end of the interview, so it is not too damaging for the interviewees' willingness to talk. grasp people's representations and perceptions and to understand what structures them, beyond an assessment of their level of competence or political knowledge; to do this, issue-based discussions prove to be much more appropriate.
All interviews have been transcribed on small forms on which I wrote a verbatim of the interview and some elements of analysis, collecting forms by theme. I analyzed this qualitative material by putting together forms on the same theme, and compared by this way individuals. I focused here on the verbatim dealing with the identification of political actors, the evaluation of elected representatives, the perception of the democratic regime and the role of citizens, that are the main elements of analysis of this article. My analysis is conceived as an exploration of citizens' representations of the political field, around the articulation between regime, institutions and politicians. First, I show that representations and perceptions of the political field are generally based on the identification of individuals, rather than on collectives and institutions. My second point then deals with the implications of this personalized relationship, and questions, in this context, the relationship to representative democracy. Finally, the discussion part interrogates the relationship between specific and diffuse support, in a context of personalized politics.

Dissatisfaction in the Context of Personalized Politics
When people talk about the political system, politicians are the central actors. My qualitative analysis shows that politicians are the actors who are the most mobilized in the speeches and structure individuals' representations of the political field.

Limited References to Political Parties and Movements
Before showing how the reference to politicians structures the representation of the political field, it is worth emphasizing the very low level of mobilization of political parties in citizens' speeches. It is certainly difficult to base a demonstration on the absence of something, but political parties are very rarely mentioned in the discourse of the interviewees. This result thus argues for the thesis associating party decline with political personalization.
There are, however, special cases in which political parties are mobilized as such. One example is Stéphane, 32 years old. He lives in a house in a suburban area of a small town in the North of France with his partner Mathilde, 20. Stéphane studied history until the bachelor's degree, he is a worker in an automobile factory and is a trade union representative. He considers that "you have to stop talking about right and left (. . . ) it means nothing to me. It's better to say UMP 5 or socialist. Even on TV, people talk about right and left all the time. It's better to speak in terms of etiquettes." He draws a parallel with the trade union world where "labels" are important. In a context where right and left no longer seem to be points of reference, the names of the parties, as well as the names of the trade unions, retain their relevance, particularly because Stéphane testifies to his knowledge of the proposals and ideological shifts of the organizations. Another case concerns Nordine, 38 years old. Together with Christelle, 36 years old, they live in a commune of Seine-Saint-Denis and have always lived in this department. Nordine holds a technical school certificate in electrical engineering and is a technical agent in the public sector. Throughout the interview, he evokes a strong criticism of politicians (see below) and, evoking childhood memories, he refers to the "communists": "Before the communists, they made these things [the MJC 6 ] work, and I remember, I was a lot with the pioneers of France, so it was the communists and we used to go on holidays a lot. That doesn't exist anymore!" The reference to the communist party, through the social and popular education activities developed in the Parisian suburban cities, is in fact no longer relevant and does not remain a structuring element of his political discourse.
More generally, even though the majority of the interviewees do not refer to political parties, the difference between the right and the left is widely used to talk about the political field and the variations in political positions (Tiberj, 2004). This is all the more true among the interviewees who report entrenched political convictions, who position themselves either on the left or on the right and who define the characteristics of each of these camps. This is the case of Pedro, 45 years old. He lives with his wife and their two sons in a commune in the south-eastern suburbs of Paris. He studied mechanics and automation until the advanced technician's certificate and is a production manager in a private company. Pedro assumes a left-wing stance and gives a simple explanation of the differences between the left: "we play the solidarity card (. . . ), sharing" and the right: "we play the card for ourselves, (. . . ) making the most of it." This is also the case for Jean-Pierre, 64 years old, retired from a job as a quality inspector in the automobile industry. He lives with his wife Micheline, 60, in a town in the North of France. Jean-Pierre declares himself "neither left nor right" and not very interested in politics, but he also shows a reading of the political field through the left and the right lens. Commenting on the political alternation in France, he says: "I have noticed, we change the right one time and the left one time. On the one hand, we fill the coffers, on the other hand we empty the coffers (. . . ) From my personal point of view, in 2012 the left will win, he [Nicolas Sarkozy, then president] has filled the coffers well, so we will empty them!" The representation between the left and the right opposing each other on how to manage public spending structures Jean-Pierre's vision 7 .
However, even if the terms right/left are still used by the interviewees, they seem, for a certain number of individuals, to have lost some of their relevance. The blurring of ideological boundaries between left and right then leaves room for the personalization of politicians. This is the observation made 6 The MJC (Maison des Jeunes et de la Culture) are youth cultural centers in cities. 7 This representation of the political space between left and right should be reassessed since the partisan system has been largely modified since 2017 (Gougou and Persico, 2017). At the time of the interviews and during the Sarkozy presidency, the political spectrum is deeply organized around the left/right cleavage. by Marie-Thérèse and Thierry. Marie-Thérèse and Thierry live in a commune near Lens and both declare themselves "rather interested" in politics. Marie-Thérèse, 46 years old, is a reader trainer and is positioned on the left. Thierry, 48, has a technical school certificate in electricity and works as a logistics receptionist in a supermarket and says he is "neither left nor right." Thierry explains the irrelevance of the left and the right by saying that they are "just words, " without ideological unity. Marie-Thérèse, following this remark, declares: "Today we are not really sure anymore, I find that people are looking for a person, in all political parties now it is a person who represents, it is not really an idea anymore, I have the impression that people are voting for a person (. . . ). So it's also a danger, because there are people who sometimes speak very well but then in actions it's different. I don't know if today people still recognize themselves in parties a lot." In a mechanism quite close to the "third person effect 8 , " Marie-Thérèse evokes the risk of personalization, caused by a loss of relevance of political parties. In this sense, she also confirms the observed processes of partisan disaffiliation and the rise of personalities in the structuring of the political field.
The reactions to what has been called "openness 9 " in the constitution of the Fillon government in 2007, under the presidency of Nicolas Sarkozy, are indicative of the place taken by political figures. Interviewees with strong political convictions and a high degree of politicization judge negatively this undertaking, as the political camps defend an irreconcilable world view. The conflicting view of politics is especially characteristic of individuals who are politicized and have political convictions structured by ideological affiliation. For the other interviewees, openness is not questioned in principle. It is then judged positively, as Sylvestre and Catherine do. Sylvestre, 55, and Catherine, 53, live in a large house near the city center of a medium-sized town in Nord-Pas-de-Calais. Sylvestre is a liberal veterinary surgeon and his wife, who has a 4-years degree in Germanic philology, has worked for a few years as an English and Dutch teacher and has been a housewife for many years. They are both located in the center right and evoke this experience of openness: Catherine: "at least they were remarkable people. And I liked that idea, they called it the openness to the left. Sylvester: Mmh, mmh (of confirmation). It was even surprising because it wasn't planned (. . . ) He [Nicolas Sarkozy] was very 8 The "third person effect" has been highlighted in the work on the effects of the media (Davison, 1983): it suggests that the individual who expresses himself or herself feels protected from the persuasive or manipulative effects of the media, but that he or she considers these effects are very strong for other segments of society. This mechanism is, for example, particularly clear in Marie-Thérèse's case. She shows a committed activist stance-she does indeed have a strong associative commitment-advocating awareness of voting and information and the rejection of extremes. This stance is coupled with a certain elitism, even a kind of condescension toward the local population, "the locals, " with whom she is in contact on a daily basis in her work as a teacher and teacher-trainer and in her numerous trips to various suburbs in the region. She thus reproaches them for their distant relationship to politics and their naivety in believing everything politicians say and promise. 9 This refers to the decision of Nicolas Sarkozy, newly elected President of the Republic, to entrust key ministerial posts to political figures, members of the Socialist Party or associated with the left (whereas he is a right-wing President). criticized at the time (. . . ) Catherine: I was grateful that he dared to do something like that. Then it wasn't really anything else, I mean Martin Hirsch 10 was still something. Sylvestre: There was Bernard Kouchner 11 at the Foreign Affairs Department. Catherine: That was a nice idea, but then it has been politically exploited, so you end up getting everything dirty. Well, maybe that was it, I'm naive, but maybe it was also a real openness." Catherine thus shows that she believed in this openness, and defends a less conflictual, more consensual vision of politics, relying on "remarkable" personalities. This aspiration to a nonconflictual vision of politics is shared by British citizens (Clarke et al., 2018).
Political parties are therefore very little present in individuals' representations of the political field. In the French context, the distinction between left and right remains mobilized, even if it loses its relevance and seems less structuring than it could be. The loosening of these points of reference is part of a perception of the political field dominated by political leaders. Specific support focuses on political actors, without precising if these actors are individual or collective. My analysis clearly shows that for citizens, individual actors matter (and not collective ones), because they mobilize spontaneously them when talking about politics.

A Deep Dissatisfaction Toward Politicians' Performances
Collective organizations are not mobilized in the representations of the political field. By contrast, politicians are often mobilized. In this regard, they are unanimously held accountable and are also, for some interviewees, systematically criticized. Easton places the issues of outputs and accountability at the heart of its conceptualization of specific support. Specific support then depends on positive or negative evaluations of the authorities. This centrality of political actors is found among all the interviewees. Depending on their level of politicization or the existence of political convictions, this reference is deployed in different ways. From an empirical point of view, it is necessary to listen to people to catch their perception about political action because evaluations "may also be stimulated not by explicit actions on the part of the authorities but by their perceived general performance" (Easton, 1975, p. 438).
Jean-Jacques and Claudine are two interviewees characterized by strong left-wing political beliefs. They show an important level of politicization and attention to public affairs. Jean-Jacques, 58, and Claudine, 54, live in an apartment in the 11th arrondissement of Paris. Both holders of a Master's degree in public law, Jean-Jacques is a senior manager in a property management company and Claudine is a claims inspector in the insurance industry. Throughout the interview, they adopt a stance of analysts and 10 Martin Hirsch was appointed High Commissioner for active solidarity by President Sarkozy in 2007. He used to be the president of a charity for the underprivileged. 11 Bernard Kouchner is a doctor and a politician who used to be a member of the Socialist Party. critics. They are constantly pointing out the problems and issues that society faces, linking them to their knowledge and using economic and political theories. For Jean-Jacques and Claudine, difficulties that France is experiencing are due to structural constraints and the organization of power. But politicians are still largely part of their discourse, in that they are accused of incompetence and lack of creativity. They are therefore not up to the economic, social, and ecological challenges facing the contemporary world. Jean-Jacques and Claudine then question the uniformity of the elite education, which prevents them from proposing new and innovative solutions. The couple does not have a vision of the political field structured solely by politicians, as they account for the role of institutions, particularly the European institutions, but criticism of the elites is a key point in their discourse.
Some interviewees, for their part, systematically perceive the political field through the criticism of its actors. It is a way of making sense to situations that seem illogical or incomprehensible. More broadly, these interviewees mobilize a reading perspective that puts their feeling of exclusion from political action at the heart of their perceptions. They all feel excluded from social measures and public spending, and for some of them, they declare that they are subject to a significant tax burden, but don't benefit (or not enough) from State services. These interviewees come from the working classes and also from the middle classes. They often encounter rather difficult living conditions, which make the feeling of not being considered in society stronger. Citizens judge political action according to what the State and the government provide them personally. They call in question politicians in general, in that sense that they are responsible for political action, but also in a more personal way, as Gwendoline's speeches show. Gwendoline is 29 years old, she is married to Franck, 31 years old. They have four children and live in an apartment in a set of building bars in a small commune next to Valenciennes. Franck stopped school at the age of 14 and his wife has the general certificate of secondary education. They both currently receive the social inclusion income, Franck works from time to time (on average 1 day a week) as a temporary garbage collector. Gwendoline has not worked recently. Their interview is based on the denunciation of "foreigners" that Gwendoline considers responsible for her own personal difficulties, in a logic of competition with social aids. In order to make sense to the permanent discrepancies between promises, announcements and reality, she considers that all aid is given to foreigners. In this denunciation, Nicolas Sarkozy [President of the Republic at the time] is directly targeted. Gwendoline then mentions the steps she takes to obtain a new home.
Gwendoline: "We wrote to him [to Nicolas Sarkozy] on the Internet for an accommodation saying we were living in a twobedroom apartment with four children and he did not help us anyway. (. . . ) And I have a friend, she's Arab, she wrote to Sarkozy and Sarkozy found her a house, I don't think that's normal! (. . . ) Then they say that the French become racist but it's because of him!" The "Arabs"-who are confused in Gwendoline's remarks with "foreigners" -are a privileged target-especially in a context of crisis (Noiriel, 2006)-insofar as they too would be favored by the President's actions.
Gwendoline: "Yeah, he [Nicolas Sarkozy] gives more aid to foreigners than to French people. The French need it much more than foreigners. Also, he says there are not enough houses for people: he gives them to others, to foreigners, rather than helping the French! (. . . ) He [Nicolas Sarkozy] is not there to help the French, he is not there to help us. In fact, he does what he wants, he wants it to start all over again like in the old days, the rich on one side, the poor on the other. We have been sidelined, we are low class for him, that's all." The President of the Republic is therefore directly and personally held responsible accused of favoring other categories of the population ("foreigners, " "rich") at the expense of Gwendoline and her family. He is the person in charge of political action, and therefore the main target of criticism.
The interviews thus reveal a general perception of a decline in the ideological structuring of political debates and the weak presence of political parties as landmarks. Most of the relationship with the institutional world is, in fact, envisaged through politicians, sometimes without any connection to more structuring reference points. The latter are then both key players and heavily criticized. Interviews show a weak specific support which goes together with personalized politics.

The Implications of Personalization
What are the implications of such a critical perception of politicians? The challenge here is to understand the consequences of these politician-centered representations, both on citizens' opinions and on support for representative democracy.

Personalization and Politicization
First of all, my interviews show that personalized politics and a high level of criticisms toward political leaders can have consequences for citizens' political orientations of citizens.
Indeed, the systematic mistrust and accusation of politicians can lead to a willingness to settle certain issues individually rather than collectively. This mechanism is particularly present in Nordine (38, technical officer). Throughout the interview, Nordine denounces the politicians in charge of exercising power. He points out the difference between the circulation of impressive amounts of money on the political scene and the fact that he and his household receive nothing from the State. He finds it difficult to understand how such sums of money can be mobilized in a context of economic crisis when political leaders often claim the necessity to cut expenses. On the issue of retirement pensions, the challenge is precisely for him to understand the discrepancy he perceives between the amounts of money wasted by politicians (political scandals, travel expenses of ministers, etc.) and the need to extend the contribution period for retirement. He gives meaning to this paradox by being suspicious of politicians who only want to fill their own pockets. For him, more money is needed since politicians have deliberately misappropriated the money spent on pensions. Nordine feels he has been deceived by politicians. This creates a political demand which is not compatible with the left-wing position that he declares at the beginning of the interview. Indeed, the lack of trust for politicians leads him to formulate the idea that some issues hitherto managed by the State should be taken care of by the citizens themselves, individually, so as to prevent any dishonest action by politicians. Here, the widespread mistrust of politicians supports a mechanism in favor of the withdrawal of the state or even depoliticization.
Another implication concerns the criteria used to judge politicians. All the interviews mention, to varying degrees, political scandals and accusations of corruption or crime among politicians. Politicians are repeatedly criticized in the interviews for their lack of integrity (Lascoumes, 2010). For the majority of the interviewees, honesty is one of the central criteria for judging elites and one of the qualities necessary for claiming power, even if it goes beyond ideological orientation. Régis' comments bear witness to this. Régis, 52, has been living with Fatima, 50, for 12 years. They have an 11-years-old son and they live together in an apartment in a recent building not far from the center of a medium-sized town in the North of France. Régis dropped out of school in 4th grade and is an unskilled worker in a large car factory in the region. Régis and Fatima have a very ideological reading of political problems and issues, positioning themselves on the left and against the brutality of the liberal system. They show disappointment with left-wing governments and are very critical of current politicians, who lack of strong political beliefs and are disconnected with reality. They analyze the political issues, by linking them to a well-founded criticism of the liberal system. Régis defends a conflicting vision of politics, between the left and the right. However, when discussing the government and its constitution, he indicated that it is necessary to have "especially honest people, whoever they are." Integrity thus takes precedence over ideological orientation, and the exemplary nature of the politician replaces the conflicting vision of politics and governance.

Citizens and Representative Democracy
How, in this context, is the place of the citizen thought out? What are the consequences of negative evaluations of politicians for representative democracy? In fact, people consider the regime and their role in it differently according to their sociopolitical characteristics. Three perceptions have been identified and show different democratic linkages between those who govern and the governed.
A first group of people consider the democratic regime exclusively through the figure of political leaders, in this context of personalized politics. This appeal to strong political leaders is linked to a rather catastrophic view of the country's situation. These people are characterized by personal experiences of significant social and financial difficulties and by a feeling of downgrading or abandonment. Their interpretation of the world and of political and social events is based on an anxious vision and shows very little optimism about the future. This vision mainly concerns interviewees aged 50 and over. Among these interviewees, political decisions are judged to be ineffective and not strictly applied. In this context, an authoritarian reassertion of the use of an effective leader is seen as the solution. What matters is that someone knows how to take the right decisions and, above all, how to enforce them. The authoritarian imposition of decisions thus proves to be a relevant solution to improve political action. For example, Marie-Paule (59 years old, former administrative agent, early retired in a situation of invalidity, positioning at the far right) thus indicates that she would like Nicolas Sarkozy to say that "during 5 years it is me and I impose things, you accept them, well so much the better, you do not accept them, well so much the worse!" For these people, politicians are accused of being too soft and not enough strict about rules compliance. But criticisms are not limited to politicians, they also concern the way democracy functions in our country, and the need for a strong Presidency. That said, this point of view is not totally free of ambivalence. Moreover, Marie-Paule considered earlier in the interview, with regard to the social protest about the pension reform (in 2010), that the demonstration was an important tool for the citizen and that it was essential to ask people's opinions before putting in place reforms. For this first type of perception of the democratic regime, the authority of a leader imposing one's decisions without referring to citizens is clearly considered, especially regarding the issues of security and justice.
A second group of people consider the democratic regime under the prism of power delegation. Even if they can be critical toward politicians, they do not question the legitimacy of representation. These people are usually interested in politics and position themselves at the center on the left-right scale. For these interviewees, as indeed for all the interviewees, politicians are not exempt from criticism, such as disconnection with reality, lack of honesty and the systematic search for positions of power at the expense of ethical political practice. Yet they declare that politicians are strongly legitimate. Decline in political ideologies, difficulties in mastering political issues and the lack of understanding and reference points in the political world may ultimately result in a legitimate delegation to political leaders. Politicians are thus seen as the organizers of political debates but also as those who are responsible for generating ideas and solutions to society's problems. Gérald's speech is typical of this position. Gérald is 56 years old, he is the IT director at the town hall of a medium-sized town in the North of France. He is at the center of the political spectrum. At the time of the discussion on the situation of undocumented migrants, Gérald has difficulty forming an opinion on this issue and, above all, in seeing how this situation could be resolved. Gérald's hesitation in finishing his sentence "there is no solution" and the change to the personal pronoun "I" are, in my opinion, indicative of the mechanism of delegation and trust. Gerald cannot express the fact that there are no solutions, but simply that he does not have any: there are therefore certainly solutions that must be found by politicians, those who are responsible for doing so. This sentence then reflects the importance of political action and the role of politicians in the resolution of situations perceived as problematic. These individuals have trust in the ability of politicians to resolve difficulties and do their utmost to solve problems, even if they can raise doubts. Citizens seem to ultimately hope that politicians will take the right decisions, but they are still not sure about it.
Gérald: "It's necessarily up to the government to make the decision, because at a given moment, you have to trust the people, you have to think that they are sufficiently competent, after all the opinions they've taken to check their hypotheses (. . . ) The government was elected to make decisions, well, it makes them and then you hope that it's not wrong." In Gérald's view, it is clear that the government is the legitimate body to make decisions, but he nevertheless keeps some doubts about this capacity. In any case, in the management of public affairs, the citizen is not called upon to take a more important place than he currently has through voting.
Competence is therefore at the core of their political representations. For some interviewees, the principle of delegation remains the key mechanism of how democracy functions. To counter the negative evaluations of political outputs, they consider the role of "experts." Christine and Michel are quite typical of this position. Christine, 55, is married to Michel, 58. She is a liberal dental surgeon. Her husband, an insurer for 25 years, is currently the manager of a professional reintegration company. Coming back from a conference on the limits of democracy, the couple then mobilize the reflections they heard there. First of all, they lay the stress on the ignorance of politicians about some topics and the need to resort to experts who provide a reliable and truthful vision of the issues at stake. Michel then proposes that democracy should become "regulatory rather than participatory, " which means that "depending on the opinions of experts, there are sometimes decisions that have to be taken that are not going to please and that cannot be taken according to a democratic mode of operation." He applies this to the question of pensions, where the whole population will say that they want to work less when it is necessary to lengthen the contribution period: "if we want it [the pension system] to hold, we have to make the decision, and that cannot be done with participatory democracy, where we ask everyone for their opinion (. . . )." The couple concludes by saying that "that's democracy, we have to keep it, but on certain points there are limits." The place of the citizen is seen here rather as an obstacle to effective and relevant decision-making, and the response to current challenges (the environment, for example) consists in giving a central place to experts, which echoes the principle at the core of the "stealth democracy" (Hibbing and Theiss-Morse, 2002).
The last way to consider the role of citizens in the democratic regime that I identified concerns a small part of the interviewees. For them, the principle of representation can be questioned and procedures where the citizens can participate more have to be considered. These people have strong political beliefs, often positioning themselves at the left wing and had experiences of associative or political involvement. This is the case of Christiane, 68, retired after having worked as a nurse for a few years. She is a "very left-wing, anarchist" activist and has commitments to associations (housing rights, for example). When she talks about representative democracy, she keeps a watchdog position, because the elected representatives "may represent [her], but she does not expect 100% of them, [she] does not trust them 100% even if [she] voted for them." Christiane also proposes to set up a system of "popular vote, " in which citizens are consulted regularly, since the election of representatives is not conceived as an end to citizen participation.
Widespread criticism of politicians can thus have consequences on the representations of the democratic system, seen mainly through its actors. The appeal to authoritarian leaders able to maintain law and order and also to listen to people's wills concern rather old citizens. They have a pessimistic perception of the situation, come from a working-class or middle-class background and have the feeling of downward mobility. In the second perception, delegation is seen as legitimate, and trust toward politicians or experts does not seem to be too much attacked, even if politicians remain the major targets of negative political evaluations. This perception concerns citizens who position at the center of the political scale, come from a middle-class background and have an average interest for politics. Finally, a few interviewees want to reassess the role of citizens in representative democracy. They usually are very interested in politics and have civic or political involvements, which explain why they want to give citizens' a more important role. The solutions induced by dissatisfaction with political performance vary, and the citizen's place in this reflection remains limited. For that matter, does this widespread dissatisfaction affect the level of diffuse support?

DISCUSSION: WHEN A WEAK SPECIFIC SUPPORT UNDERMINES DIFFUSE SUPPORT
I would like to question the implications of such a personalized and critical representation of the political field on the perception of the regime by reiterating the distinction between specific and diffuse support. More specifically, I argue that the weakness of specific support through distrust of politicians and dissatisfaction with government performance can ultimately undermine diffuse support through support for the regime and its functioning. Easton has already mentioned that "if discontent with perceived continues over a long enough time, it may gradually erode even the strongest underlying bonds of attachment" (Easton, 1975, p. 445). He also points out that a low level of specific support can undermine the level of diffuse support and thus contribute to a much wider criticism of the regime and institutions. As stated in the section on methodological issues, the direct question on the functioning of the democratic regime is not necessarily relevant for collecting the representations and opinions of individuals on this subject. Responses to this question mainly show an attachment to the democratic regimewhen interviewees understand what it refers to. They value its principles, and often make a comparison with other countries or with more authoritarian regimes, denouncing them and considering themselves lucky. My analysis proposes to go beyond their answers to try to understand how the overwhelmingly critical judgement of politicians and their performance can have consequences for attachment to the regime.
The systematic perception of a gap between politicians and citizens, and between announcements of measures and their implementation, feeds the idea that citizens' demands are not considered.
The majority of interviewees underlines the disconnection between themselves and the elites: politicians lose all connection with reality when they are in power, and do not realize what citizens experience in their daily lives. The "them vs. us" opposition is particularly structuring in the discourse of the interviewees. Jean-Jacques and Claudine, who declare strong political beliefs and a high level of politicization, indicate that "the elite forgets the everyday life of the people." Isabelle and Pedro both declare themselves to be on the left, even if this political position is no longer as obvious as it was a few decades ago. They both criticize the standardization of politicians and Pedro says: "they are people from the same place so they all think the same. They come from the same schools, they have the same friends (. . . ) so that's also what disconnected them [from the people]." For other interviewees, this disconnection is expressed in even more blunt terms. Gérard says he has little interest in politics and is "neither on the left nor on the right": the interview is dominated by his indignation toward the political system, multiplying the statements of misunderstanding and injustice. Speaking of Nicolas Sarkozy, Gérard retorts that "he doesn't care about us, " highlighting, in a strong personalization, the indifference of the President of the Republic for the citizens. This disconnection is also expressed in the feeling that politicians are apart, with multiple advantages, which differentiates them from citizens. Some practices can thus be criticized by citizens, because they favor the current political establishment of politicians, without aiming at the general interest. In this sense, the following elements are denounced: the fact of changing ministries during reshuffles without this being associated with a specific competence on the perimeter of action, or the possibility for ministers to reclaim their municipal mandates once they have left the government. For Jean Pierre (64 years old, retired from a job as a quality inspector in the car industry), "that is not normal. If I was dismissed from Mr. [his company], I would go directly to the ANPE [National Employment Agency]. Why did they. . . ? He's fired from the government (. . . ) he's going to come back to take his place, that's not normal!" The discrepancy can also be seen in the judgement of performance. Citizens highlight the distinction between promises and announcements made by politicians and actual implementation on the ground. People mention disappointment and even recurrent indignation about public action, which is either judged insufficient or does not live up to the announcements made. Grossman and Sauger also specify the characteristics of the French system that explain this discrepancy: for them, "the combination of the majority logic of the regime, its presidentialization and the use of a two-round voting system has largely contributed to inscribe in the French political landscape a fundamental contradiction between the need to raise high expectations among voters in order to hope to survive the first round and the observation of a powerlessness to satisfy them in the register of daily government action" (Grossman and Sauger, 2017, p. 157). The two authors also show that the "honeymoon" available to the newly elected President is becoming shorter and shorter: the popularity curves are falling very quickly and there are many more people who do not trust the President than others (131). Gwendoline's comments are particularly illustrative of this discrepancy: she mentions a series of measures that were supposed to concern the unemployed, of which she and her husband are a part of.
Gwendoline: "He [Nicolas Sarkozy] says that there will be help for the Internet for the unemployed, for the telephone, we didn't get it! An EDF [Electricity supplier] help! We haven't seen it, the help, we haven't seen anything! (. . . ) I had seen that on TV, in the newspaper, so he promised, he promised again and all in all, nothing has been done." Here Gwendoline expresses her systematic disappointment with measures that feed her hope of being able to alleviate the financial difficulties she encounters on a daily basis. This gap between announcements and the application of measures is also reinforced by the development of selective and targeted social policies, at the expense of universal policies (Paugam and Duvoux, 2013). Access to aid is thus conditioned by thresholds and increasingly by the assessment of situations (Dubois, 2012;Lima, 2016). This can reinforce the impression of a gap between the announcements and reality, as Annie (47 years old, nurse) shows when she says, with regard to the aids put in place, that "there is always something that makes you not entitled to them." The disconnection between politicians and citizens and the recurrent dissatisfaction with public action call into question more broadly the capacity of institutions to take into account the needs of the population and thus call into question the functioning of the regime. Rachel (21 years old, in training as a health care assistant) underlines this distance between the people and the political actors, by noting that the regime is certainly democratic but not very capable of establishing bridges between the two: "and if we do something, it won't go up to the president, we would perhaps like to have more control over the people higher up. (. . . ) It's democratic but we are still too low." Beyond a deep dissatisfaction, it seems that means of communication between citizens and those who govern are not very effective. Trust and legitimacy (which are the two elements of Easton's diffuse support) toward the regime can then be damaged through a constant and massive dissatisfaction.

CONCLUSION
In a context of intense personalization, my article shows that citizens' representations of the political field are dominated by individual actors and the figure of politicians.
Politicians are at the same time the key players, the targets of criticism and those responsible for public action. Collective actors and institutions are still barely mobilized in the discourse and the latter are mainly perceived through the prism of politicians. This is why criticisms about politicians have consequences on the perception of the regime and democratic institutions.
This criticism is widespread among citizens, with varying degrees of intensity depending on how firmly political and ideological beliefs are rooted. Among those for whom the partisan structuring of the political space is weak, the negative judgment of politicians can be systematically mobilized as a reading grid of the world. While the literature has often separated negative evaluations of politicians' performances and the attachment to the democratic regime, my analysis suggests that the weakness of specific support may in fact undermine diffuse support. Indeed, mistrust of politicians, a feeling of disconnection with elites and widespread dissatisfaction with government performances maintain the idea of a problematic representation but also of an inability of institutions, captured through politicians, to take into account the needs and demands of citizens. In this system, citizens feel that they are not given much consideration. Individuals then differ in their vision of the political system: some value the use of an authoritarian leader who imposes decisions, while others insist on changes to the democratic system to give more room to experts and limit the expression of citizens. The principle of delegation to politicians is accepted and legitimate, but representation raises doubts and may require greater citizen control and stronger and more regular participation. Representative democracy is then faced with two options: to work for better representation or to give more space to citizens, in order to improve the ways of communication between citizens and their representatives.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

AUTHOR CONTRIBUTIONS
CD did the conception of the work, collected the qualitative data, analyzed the data, interpreted them, and wrote the entire paper.

ACKNOWLEDGMENTS
The author wants to thank Camille Bedock and Pierre-Etienne Vandamme, for their careful reading of previous versions of this article and their stimulating remarks.