As the largest ethnic minority in Europe, the Roma represent 10 to 12 million people (Commission européenne, 2020). Originating in northwest India, they left the region around the tenth century to reach the Middle East, then, via Mesopotamia, arrived in the Byzantine Empire. Around the fourteenth century, several groups of Roma settled in the Balkans. Although the majority remained in Eastern Europe, from the fifteenth century, groups dispersed in several migratory waves throughout Western Europe, particularly toward the Iberian Peninsula. Other northward movements took place around the sixteenth century (Liégeois, Reference Liégeois2007; Piasere, Reference Piasere2011; Delépine, Reference Delépine2012). The various migratory directions taken by the Roma generated distinct names and identities. The fall of the Berlin Wall and the European Union's accession of Hungary in 2004, as well as of Romania and Bulgaria in 2007, led to new migratory flows and accelerated the mobilities of these populations from the east to the west (Liégeois, Reference Liégeois2007: 23; Legros and Vitale, Reference Legros and Vitale2011: 4; Delépine, Reference Delépine2012: 6). However, whether in Western or Eastern Europe, their living conditions have been marked by prejudice, discrimination and social exclusion (Commission européenne, 2011: 2). As a result, 40 per cent of Europeans refuse to have Roma as neighbours (Barbulescu and Beaudonnet, Reference Barbulescu and Beaudonnet2013: 9), while 80 per cent of these do not have direct contact with them (Eurobarometer 393, 2012: 12).
Considering the problem that these statistics raise, the present study is interested in the factors that contribute to such attitudes toward the “Roma minority”Footnote 2: namely, the mental state of individuals, which predisposes them to act in a certain way—positive or negative—in contact situations. Attitudes are defined here as “a psychological tendency that is expressed by evaluating a particular entity with some degree of favour or disfavour” (Eagly and Chaiken, Reference Eagly and Chaiken1993: 1). Many social scientists argue that the psychological process that gives rise to attitudes is stimulated by an individual's external environment (Bergamaschi, Reference Bergamaschi2011: 95). In this way, the analysis of social interactions has often been privileged. Our study proposes to offer a comparative analysis of the impact of the different types of “intergroup contact,”Footnote 3 as well as the influence of the media on the attitudes of members of two host communities toward the Roma. The central question guiding our research is the following: How are attitudes toward the Roma constructed?
This investigation reveals a gap to fill in the literature on attitudes toward minorities. Several studies have been devoted to this subject in contexts of diversity (McLaren, Reference McLaren2003; Tucci, Reference Tucci2005; Schneider, Reference Schneider2008; Meuleman et al., Reference Meuleman, Davidov and Billiet2009; Rydgren and Ruth, Reference Rydgren and Ruth2013; Havekes et al., Reference Havekes, Coenders and Dekkers2014; Schemer, Reference Schemer2014) and in the case of certain minorities, notably Muslims (Savelkoul et al., Reference Savelkoul, Scheepers, Tolsma and Hagendoorn2011) and Blacks (Tropp, Reference Tropp2007), but no research seems to have been carried out on relationships that involve members of Roma communities. And yet the Roma are one of the most discriminated- against minorities in Europe (Mayer et al., Reference Mayer, Michelat, Tiberj and Vitale2014). Furthermore, it is important to analyze the social and political relationships as well as the social representations that concern this minority group, since anti-immigrant attitudes vary according to the origin and characteristics of each group (Card et al., Reference Card, Dustmann and Preston2005: 11).
Several scholars have addressed the question of the construction of attitudes toward ethnic minorities in order to better understand the mechanisms that generate discrimination. These studies use different disciplinary approaches, notably in the fields of social psychology, sociology and political science. However, few studies combine intergroup contact theories, originally developed by psychologists and sociologists, with those related to the influence of local integration policies or media content, mainly mobilized by researchers in political science and communication. Although these theories come from distinct disciplinary approaches, the advantage of combining them, as we hypothesize, provides perspective while revealing complementarities. Our study is thus part of a multidisciplinary approach combining social psychology—more specifically, the study of intergroup dynamics—and an analysis of public policies. We will test four theories to account for relations between Roma and non-Roma: the contact theory, the halo effect, the impact of municipal Roma integration policies and the influence of the media.
In the following pages, we lay out the situation of the Roma in France, presenting an overview of research that provides some theoretical expectations regarding the formation of attitudes toward the Roma. These take the form of hypotheses and are tested empirically through the analysis of semi-structured interviews carried out in two municipalities in the Paris region: La Courneuve and Ivry-sur-Seine. We then present the methodology used for data collection and empirically test the validity of the assumptions discussed in the theoretical section. The results show that since the Roma are subject to negative representations—and these are overpublicized—and are also victims of spatial segregation and social exclusion, there are few opportunities for quality intergroup contacts. This creates a negative perceptual bias and explains why attitudes toward these populations are more negative than those toward other minorities on the continent. However, the implementation of municipal policies in favour of the integration of the Roma makes it possible to improve the living conditions of the members of this group and, in so doing, deconstruct certain prejudices attributable in particular to their precarious situation. Finally, our analysis illustrates the way in which the media activate, nurture or consolidate ways of perceiving the Roma.
The Stigmatization of the Roma in France
Our study highlights the case of the Roma in France, this country being of particular interest due to national circumstance. With its “republican model,” France rejects the concept of ethnic minorities and adopts assimilationist policies toward newcomers (Rothéa, Reference Rothéa2003). These policies effectively fail to recognize, and in so doing stigmatize, such communities. This stance generates discrimination toward the most fragile minority groups, in particular with regard to housing or work (Kamiejski et al., Reference Kamiejski, Guimond, De Oliveira, Er-Rafiy and Brauer2012: 53). In this context, the presence of Roma on French territory constitutes a real “problem to be solved.” It revives the question of public order, frequently invoked by the state, and is justified with reference to Roma behaviours, considered deviant. Both the political right and left employ demagogic rhetoric to essentially criminalize Roma groups (Fassin et al., Reference Fassin, Fouteau, Guichard and Windels2014). The fight against crime is used as a pretext by the government to destroy their camps (Kropp and Striethorst, Reference Kropp and Striethorst2009). The major controversy that arose following Nicolas Sarkozy's speech on engaging in a war against traffickers and delinquents (Grenoble speech, July 2010), where he declared that he wanted to “put an end to the wild Roma settlements,” (Élysée, 2010) as well as Prime Minister Manuel Valls' declarations regarding the intransigent nature of the Roma (Libération, 2013), constitute examples of speeches by heads of state. The argument of Roma delinquency threatening the republican order justifies the need to set up derogatory and discriminatory administrative practices toward them and transforms them into scapegoats that serve as a potential outlet for mob violence (Legros and Vitale, Reference Legros and Vitale2011: 19). Roma in France are among the populations most confronted with rejection and exclusion by majority groups (Mayer et al., Reference Mayer, Michelat, Tiberj and Vitale2014; CNCDH, 2016: 13).
In the context of this article, we are interested in recent immigrants, particularly those linked to the two consecutive enlargements of the European Union in 2004 (Hungary and the Czech Republic) and 2007 (Romania and Bulgaria). The majority are from Romania and, to a lesser extent, from Bulgaria and Hungary. Together they form a relatively small population compared to other communities, estimated at—according to civic organizations and government sources—between 15,000 and 20,000 people. And yet they attract a disproportionate amount of political and media attention.
Multiple Perspectives on Intergroup Relations
The first two theories highlight the impact of quality contact or lack thereof on individual attitudes toward ethnic minorities. The third focuses on municipal policies for the integration of immigrants and their effect on intergroup contacts, while the last presents the influence of the media in shaping anti-immigrant attitudes.
Contact theory
Contact theory is most frequently drawn upon by researchers working on desegregation and intergroup relationships (Schofield, Reference Schofield1991: 359). Developed by Gordon W. Allport (Reference Allport1954, Reference Allport1958), this theory sets out four conditions that positively affect intergroup relations: (1) The status of group members where contact is made must be equal. (2) Groups must share and work toward a common goal. Practising a team sport is one such example. (3) There should be cooperation rather than competition between groups. (4) Standards—support from authorities, laws or customs—to make contact socially acceptable must be put in place (Allport, Reference Allport1958). According to Thomas F. Pettigrew, Linda R. Tropp, Ulrich Wagner and Oliver Christ (Reference Pettigrew, Tropp, Wagner and Christ2011), these conditions deemed necessary by Allport facilitate the formation of positive attitudes toward Others. A meta-analysis of 515 studies concerning intergroup contact carried out by this group of researchers showed that individuals who have quality interactions see their prejudices decrease by 29 per cent, while those whose interactions do not meet Allport's conditions nonetheless see them decrease by 20 per cent.
First developed in the field of social psychology, contact theory has been successfully taken up by political scientists. The application of this theory to desegregation policies in American schools has opened up promising leads for improving intergroup relations (Schofield and Eurich-Fulcer, Reference Schofield, Eurich-Fulcer, Brown and Gaertner2001: 574). One of the intrinsic mechanisms of this theory is that quality intergroup contact—that is, that which satisfies the four favourable conditions—modify stereotypes and negate negative prejudices by allowing members of the majority group to learn more about members of the minority group (Pettigrew, Reference Pettigrew1998: 70). These contacts generate new behaviours, such as acceptance, affection and trust, which are precursors to changes in attitude toward the other group. As a result, as demonstrated by Alberto Voci and Miles Hewstone (Reference Voci and Hewstone2003) in their analysis of contacts between Italians and immigrants, quality contact reduces anxiety-related emotions generated by the presence of the Other (see also Dovidio et al., Reference Dovidio, Gaertner and Kawakami2003). Such interactions encourage feelings of empathy toward a stigmatized group and, as such, develop one of the essential conditions that Pettigrew adds to the theory: intergroup friendship relationships (Pettigrew, Reference Pettigrew1998; Dovidio et al., Reference Dovidio, Gaertner and Kawakami2003; McLaren, Reference McLaren2003).
The halo effect
Quality intergroup contact helps to reduce prejudice against a minority group, since members of distinct groups get to know each other. Conversely, ignorance promotes the development of negative attitudes (Pettigrew, Reference Pettigrew1998: 71). One of the precursors of the theory explaining this phenomenon, called “halo effect,” comes from the psychologist Solomon E. Asch. In his article “Forming Impressions of Personality,” he presents experiments in which psychology students must express their impressions of a fictitious person whose personality traits have been partially shared with them. He explains the formation of impressions based on the interpretation and selection of information: “We look at a person and immediately a certain impression of his character forms itself in us. . . . In what manner are these impressions established?” (Asch, Reference Asch1946: 258). Asch demonstrates that individuals tend to judge a stranger by a feeling that comes from a characteristic that occupies a central place in their configuration.
The halo effect is therefore a bias that affects perceptions and leads to the adoption of prejudices, positive or negative, toward individuals one does do not know directly. These prejudices are based on an impression that can be produced by various factors—in particular, by the media and the stereotypes that circulate and are perpetuated in society with regard to a group. Mainly used in social psychology, this theory has also helped explain certain phenomena of interest to political scientists, including the construction of negative attitudes toward minorities. In his study on support for the Front national in France, Pascal Perrineau (Reference Perrineau1997) shows that this support is more marked in peripheral regions than in those where there is a high concentration of immigrants. Fears or worries about the presence of immigrants are thus more imaginary than real. These feelings of hostility are the result of distant and superficial intergroup contact that does not allow for the deconstruction of prejudices and stereotypes (Rydgren and Ruth, Reference Rydgren and Ruth2013; Bilodeau and Turgeon, Reference Bilodeau and Turgeon2014).
The effect of municipal policies
Recall that Allport (Reference Allport1958) argues that to be successful, intergroup contact must be supported by authorities, laws or customs that establish a process of, and standards for, acceptance. In this study, we argue that municipal involvement in the integration of immigrants is essential to setting the framework and conditions for social interaction. And yet, as Aude-Claire Fourot (Reference Fourot2013) has argued, studies on the integration of immigrants often ignore municipal policies and their implementation. While municipalities avoid giving themselves an explicit mandate on the subject, it is worth noting that they do not respond in the same way to the challenges posed by the integration of immigrants. These actors have significant and politically inflected latitude when it comes to accommodating ethnocultural diversity on their territory.
Furthermore, as Thierry Berthet and Christian Poirier (Reference Berthet and Poirier2000) argue, the integration of immigrants is largely based on social interactions. These relationships are built by local integration policies that act as a framework for behavioural norms and social configurations. Thus, local policies can potentially forge relationships of domination or of more inclusive integration. In addition, since municipalities are on the front lines of the challenges related to immigration, they have a better understanding of the situation and can generate actions adapted to this reality.
All of this is even more relevant with regard to Roma integration. Faced with a situation in which France and the European Union continually trade responsibility over this matter, the municipalities become the guarantors of their living conditions. In the absence of a precise and coherent strategy, local initiatives vary from one municipality to another. Local authorities react differently to the installation of Roma camps on their territories. Some municipalities are attempting to offer an alternative to these precarious habitats, a solution which is, however, often reserved for a group of Roma deemed to be the most “deserving” and who meet certain criteria (Legros, Reference Legros2011; Bessone et al., Reference Bessone, Doytcheva, Duez, Girard and de Latour2014; Vitale, Reference Vitale2015). An alternative, in the form of “integration villages,” has been widely criticized for the logic of control and coercion that governs these spaces, in particular through the installation of fences, as well as monitoring devices and screened visits (Doytcheva, Reference Doytcheva2014: 4). To date, around 750 Roma migrants are supported in temporary accommodations and by integration schemes initiated by civil society actors and set up with the assistance of the municipalities (Dalibert and Doytcheva, Reference Dalibert and Doytcheva2014: 75).
There are, however, some municipalities (and, in some cases, the same) that do not hesitate to evacuate the Roma settlements without offering alternative accommodation. According to the French inter-ministerial circular of August 26, 2012, a social diagnosis of each family present in the settlement must be carried out before it can be dismantled. However, carrying out diagnostics allowing individual support, particularly in terms of housing and employment, is often an administrative formality. Access to basic services such as water, education, garbage collection and the dismantling of Roma settlements, with or without rehousing options, is dependent upon the goodwill of the municipalities (Cousin, Reference Cousin2013; Bessone et al., Reference Bessone, Doytcheva, Duez, Girard and de Latour2014; CNDH Romeurope, 2015; Vitale, Reference Vitale2015).
Media influence
The literature on the influence of the media in shaping anti-immigrant attitudes is extensive. First, several studies have shown that racial prejudice in the news is likely to help forge negative opinions concerning minorities (Dixon and Linz, Reference Dixon and Linz2000; Kellstedt, Reference Kellstedt2003; Mastro, Reference Mastro, Bryand and Oliver2009; Schemer, Reference Schemer2014). Indeed, prejudices can be created, maintained or increased in an individual confronted with negative racial connotations in the media. This phenomenon is called “priming” and refers to the activation of stereotypes in response to recent or frequent exposure to the stereotypical portrayal of an ethnic group (Schemer, Reference Schemer2014: 531).
In addition, the accumulation and repetition of media messages concerning migrants make it difficult for viewers to see beyond these preconceived conceptions (van Dijk, Reference van Dijk1980). Along these lines is the notion that the construction of prejudices against minorities goes through a socialization process generated by the media (Crépeau, Reference Crépeau1997): “The classical prejudice model suggests that racial and ethnic prejudice is not a response to threat but, first and foremost, socially learned feelings and sentiments” (Gorodzeisky and Semyonov, Reference Gorodzeisky and Semyonov2016: 334).
The media play an essential role in shaping attitudes, deciding which subjects are important and publicizing what appears to be public opinion on certain societal issues (Imbert, Reference Imbert1982; Sheets et al., Reference Sheets, Bos and Boomgaarden2016). In their work on the link between the media and support for right-wing populist parties, Penelope Sheets, Linda Bos and Hajo G. Boomgaarden (Reference Sheets, Bos and Boomgaarden2016) demonstrate that negative discourse on immigration consolidates anti-immigrant attitudes. This plays a key role in the politicization of everyday experiences (Mutz, Reference Mutz1994; Kinder, Reference Kinder1998). Media coverage of immigration issues informs people and politicizes these issues in the minds of citizens, often leading to negative attitudes (Hopkins, Reference Hopkins2010).
Media representation, in the case of Roma in France—and in several European countries (Barbulescu et al., Reference Barbulescu, Beaudonnet and Harell2015)—is often homogenizing and stigmatizing. The media portray the Roma as being at the heart of social problems and unease (Fassin et al., Reference Fassin, Fouteau, Guichard and Windels2014), notably poverty, delinquency, lack of housing and insecurity, without addressing the issue as a sociologist would. Media processing “produces modes of perception and interpretation of social reality, suggesting an extreme and distorted representation of the otherness of which the Roma are bearers” (Mayer et al., Reference Mayer, Michelat, Tiberj and Vitale2014: 206). For Marion Dalibert and Milena Doytcheva (Reference Dalibert and Doytcheva2014), the media employ “social branding” practices, assigning Roma “problematic” social attributes. They are differentiated from members of the majority by specific characteristics that foster exclusion processes. Repeatedly associated with a territory, whether it is a camp or an “integration village,” as well as with gendered representations deemed deviant, such as “the submissive woman” or “the overly virile or violent man,” the Roma are the target of identity discourse that further reinforces the border between the majority and the minorities.
Four Research Hypotheses
Based on our review of the literature, we can formulate several theoretical assumptions. First, contact between members of different groups improves understanding and attitudes collectively. Our first hypothesis is that members of host communities who have good contact with the Roma are less inclined to have negative attitudes toward them.
Second, superficial intergroup contact generates an imagined encounter, which increases negative attitudes toward a minority (Bilodeau and Turgeon, Reference Bilodeau and Turgeon2014). In fact, this type of contact does not allow for adjustments to preconceived ideas about a minority group; these remain and can be amplified at the sight of certain behaviours, which give way to extrapolation. From this follows our second hypothesis: that members of host communities who live near a Roma settlement and who do not have direct contact with the Roma are more likely to have negative attitudes toward them.
Third, some local contexts can limit—just as they can encourage—more advantageous forms of contact (Pettigrew, Reference Pettigrew1998: 78). By encouraging diversity, municipal integration policies can reduce discrimination against a minority group and improve intergroup relations (Berthet and Poirier, Reference Berthet and Poirier2000; Tossutti, Reference Tossutti2012). Our third hypothesis is thus that the implementation of municipal policies promoting the integration of the Roma encourages the development of better-quality intergroup contacts.
Finally, the media play a crucial role in the transmission of prejudices and in the reinforcement of racism toward ethnic minorities, since they predominantly report negative situations or events (van Dijk, Reference van Dijk and Cottle2000). The fourth hypothesis that we are therefore proposing to test is that the French media fuel stereotypical representations of the Roma. Media influence is seen as a factor that can limit the desire to get in touch with the Roma and amplify the halo effect. It also shapes attitudes toward the Roma by conditioning individuals to seek and maintain a certain representation of reality. The aim of our study is not to carry out a formal verification of this hypothesis but to provide illustrations as to how the media are likely to build or maintain certain attitudes.
Analysis of Attitudes toward the Roma in La Courneuve and Ivry-sur-Seine
A large part of the Roma in France lives on illegal land settlements on the outskirts of Paris (Bessone et al., Reference Bessone, Doytcheva, Duez, Girard and de Latour2014: 183). In 2016, the Inter-ministerial Delegation for Accommodation and Access to Housing (DIHAL, 2016) listed 6,615 individuals living in shantytowns in Île-de-France. Our own investigation focused on two municipalities in the Paris region: La Courneuve and Ivry-sur-Seine. The two have a large settlement (ranging from 300 to 600 inhabitants depending on the period) on their territory, long establishedFootnote 4 (since 2007 in La Courneuve and 2011 in Ivry-sur-Seine) and under threat of evictionFootnote 5 (94 Citoyens, 2015; Vincendon, Reference Vincendon2015). These communes share a similar socio-economic standing, the average income per household being lower than the average French income, and are ranked among the lowest standards of living in France (Boudet, Reference Boudet2014). Built around the 1950s as working-class cities for Paris, in a period of major industrialization, made in part possible by the construction of railways, they form what is known as the Parisian “red belt” (Pinson, Reference Pinson1992: 71–90).
However, local integration contexts differ from one municipality to another. There are more than 20 Roma families in Ivry-sur-Seine who have social housing, training, a job, and whose children go to school. These integration projects are encouraged and subsidized, among others, by the municipal government (Association Logement Jeune 93, 2015). A support group for the Roma has also been in place since 2011, bringing together many Ivry citizens who are fighting for an improvement in the living conditions of those families living in the rue Truillot settlement. Conversely, no project has been implemented by the municipal government of La Courneuve, and there is no support group helping families who have been living in the settlement on rue Pascal for several years.
Our analysis was based on semi-structured interviews that we carried out in the municipalities of La Courneuve and Ivry-sur-Seine (in June and July 2015) with residents (28 interviews in La Courneuve and 29 in Ivry-sur-Seine), agents from the two municipalities (4 interviews) and members of associations and support groups for the Roma (8 interviews). In order to measure attitudes toward the Roma, we asked the inhabitants of the two municipalities about their views on this minority and how they might deal with it. During the interviews, we presented respondents with two newspaper articles revealing discriminatory events involving the Roma in France in order to gather information on their attitudes. A respondent who does not object to the negative treatment of the Roma must therefore hold a negative attitude toward these populations. We asked the following questions: (1) What do you think of the Roma settlement at La Courneuve / Ivry-sur-Seine? (2) What do you think of the fact that Haubourdin's municipal government allowed the three trailer loads of manure to be dumped in front of a Roma settlement? (Reading of a newspaper article outlining this event: Libération [2015].) (3) What do you make of the Force Ouvrière union asking for separate buses for the Roma? (Following a reading of the Bourgneuf [Reference Bourgneuf2015] article.) (4) What do you think of the destruction of Roma settlements? (5) Do you think it is more difficult to start a friendship with Roma than with non-Roma?
The questions asked to ascertain the nature and quality of the contacts between each respondent and the Roma are as follows: (1) Do you live near a Roma settlement? (2) Do you have contact with Roma in your daily life (at work, in the supermarket, on the metro, on the street or other)? (2.1) If yes, what kind of contacts and in what milieu? (2.2) If not, do you have people around you who are in contact with Roma? If so, describe this relationship? (3) Do you have friends from the Roma community (people you meet regularly)? We collected data on the role played by municipalities in the integration of Roma by means of semi-structured interviews with municipal officials from La Courneuve and Ivry-sur-Seine, associations and support groups. We used this as an opportunity to ask them questions about the services offered to the Roma (education, drinking water, garbage collection and social assistance). Finally, in this research, we observed the media's influence through unprompted comments on the part of respondents, affirming their beliefs in certain stereotypes since they “saw it in the media.”
Attitudes toward Roma: Construction and Deconstruction of Negative Perceptions
Rare intergroup contact
Our analysis shows that our first hypothesis concerning contact theory cannot be confirmed. Only two respondents stated that they had quality contacts with Roma. Such is a testament to the exclusion these individuals face. The first respondent (10C, age 24, school crossing guard, lives far from the Roma settlement)Footnote 6 previously worked with a Roma colleague and got on well with her. This respondent adopted a discourse denouncing all forms of discrimination against minorities. Thus, it is difficult to confirm whether the respondent already possessed a certain degree of empathy for the Roma or whether contact had positively affected her attitude toward them. Another respondent (22C, age 57, retired, lives far from the Roma settlement) said that he had friendly relations with Roma. He himself had initiated first contact with the aim of helping certain members of this minority, so he already held positive attitudes toward the Roma before developing friendships.
Although contact between Roma and members of host communities is quite limited in the two municipalities, a few respondents described their daily contact with Roma, particularly in the context of their work. This latter group's representation of this population is more positive and nuanced than of those respondents who have no contact with them. This shows that contact, even if it does not adhere to the conditions developed by Allport (Reference Allport1958), still manages to transform characterizations and deconstruct certain prejudices. Pettigrew's (Reference Pettigrew1998) theory that daily contact facilitates the formation of positive attitudes is therefore valid. Indeed, although the respondents hold certain prejudices toward the Roma, other prejudices seem to be dismantled in view of behaviours that do not line up with preconceived notions. Respondents adopt more moderate discourse accordingly. These intergroup contacts mainly take place in informal contexts: when Roma speak to the baker who sells them bread, to the cashier at the grocery store, and so on. Several respondents who have this kind of contact with Roma have a more positive attitude toward them—though their speech remains tinged with prejudice. For example, in the context of her work in customer service at a bakery, respondent 13C (age 34, baker, lives and works near the Roma settlement) comes into regular contact with those who come in to buy bread. Her speech in relation to these encounters is largely positive:
— What do you think of the Roma camp within La Courneuve?
— I don't mind. They're nice people. They come in, they say hello, they don't do anything objectionable. So far, they have always been correct: they have always paid, they have not asked for anything. . . . They are very nice. Since they have been here [in La Courneuve], aside from begging, I've never seen them steal anything. People automatically hold their bags a little closer in their presence, but no, I don't see them as being in any way villainous or aggressive. Of those I meet here, I haven't seen one that was aggressive. On the contrary, when they are given something, they even exaggerate on the thank you!
The same is true for respondent 17C (age 31, a business owner who lives far from the Roma settlement), who regularly interacts with Roma on the job. “They come here to buy things. Some are nice, some are not. Like anyone, anywhere.” Presumably, being in regular contact with Roma allows respondents to deny certain stereotypes about them. They consider the Roma to be “normal” people, who perform the same everyday actions as any other individual.
A bias affecting perceptions of minorities
Our second hypothesis concerning the lack of direct contact with the Roma and its effects is valid in light of the responses obtained. Living in proximity to a Roma settlement, and regularly seeing Roma people without actually speaking with them, does not allow for the deconstruction of stereotypes. In the absence of direct contact, an individual tends to confirm existing stereotypes, narrowly interpreting certain gestures and behaviours that the Roma project. Influenced by negative characterizations, first impressions of this group are formed and remain in the realm of the imaginary.
In addition, in the two municipalities, both among respondents who hold positive attitudes and among those who hold negative attitudes toward the Roma, the majority expressed at least one prejudice. This is all the more obvious among respondents who do not have contact with the Roma, thus demonstrating the extent to which prejudices are anchored in representations. The recurrence and prominence of these representations make it difficult to evade preconceived images, even for someone well disposed toward the Roma. The most frequent spontaneous prejudices expressed by the respondents are that the Roma steal (according to 15 respondents), are dirty (15 respondents), do not want to integrate or prefer to live in their community (17 respondents) and do not seek to educate their children (9 respondents). The majority of respondents, however, had never witnessed actions confirming such prejudicial remarks.
The halo effect is particularly present in the commentary of inhabitants of La Courneuve, where the settlement is geographically isolated from the rest of the population. It is also noticeable in the remarks of residents of Ivry-sur-Seine who live far from the settlement and who regularly see Roma but who do not maintain contact with them. Respondent 8C's comments (age 34, homemaker, lives far from the Roma settlement) effectively illustrate this phenomenon. She has never spoken to a Rom and knows no one who has contact with them. However, she feels “overrun” by the Roma, and the associations she has of them are negative:
— What do the Roma do when you see them?
— You know, they're Roma. . . .They steal!
— Did they ever steal from you, or do you know someone who has been a victim of this kind of theft?
— No, never.
Another example of negative representation emerges from respondent 20C's comments (age 31, butcher, works near the Roma settlement), who very rarely meets Roma when they come to buy products in the butcher's shop where he works. For him, their presence in the street is enough to elicit prejudicial commentary: “The Roma stay in their settlement. They do not venture out much. If you see them outside, it's because they're taking a walk or because they're scavenging, or they're begging. Generally, the Roma stick to themselves, they stay in their clan.” The same holds true for respondent 18C (age 29, plumber, lives near the Roma settlement), who associates the Roma with the negative impression he has of Gypsies. This is a testament to the associations and classifications people make with respect to marginalized individuals—as a category of people excluded from the rest of society.
— The Roma are dirty. No need to go to their camp to see what's inside. They do not know how to live, or they deliberately reject what it means to live properly!
— Why would they do it on purpose?
— Because these people have no life! In my opinion, even if they were given a house, they would not sleep in it. It's the same deal with Gypsies: you give them a house, but they camp in the garden.
A final example of the halo effect is that of respondent 4C (age 29, salesman in a furniture store, lives far from the Roma settlement), who has never spoken to a Rom and who only occasionally crosses paths with Roma on the street. However, when a Rom enters the store where he works, he is immediately suspicious, assuming that they are delinquents likely to steal from him. This description is quite common. Worth noting is that Roma are often depicted as thieves not because they are poor but because they are Roma.
They are not integrated at all. They keep to themselves. They are not otherwise part of the community and do not mix. They are not allowed in anywhere and as soon as they enter a place, people are immediately on them. As for myself, every time they come in here, I watch them. We don't trust them because we know they're not going to buy, so they might steal or do something. Everyone is wary of the Roma.
Municipal policies that promote positive attitudes
Although respondents’ attitudes toward the Roma are largely negative, we have observed a mechanism for attenuating and dismantling prejudice in Ivry-sur-Seine. This takes the form of municipal initiatives, which set out to improve the living conditions of the Roma while creating new opportunities for interactions that go beyond those described above. The hypothesis that municipal policies in favour of integration encourage intergroup contact is substantiated by discourse collected in Ivry-sur-Seine. Indeed, by establishing policies for Roma integration, the municipality has helped to activate socialization mechanisms promoting a spirit of acceptance of the Other. These local policies have an impact on the place that the Roma minority occupies in society and, by the same token, on intergroup attitudes.
The municipal government of La Courneuve, which considers the integration of the Roma a national issue, has not initiated such policies. The concern that helping the Roma could make them want to stay in the settlement is also provided as a reason for this municipal disengagement. The result is a clear lack of understanding between groups. On the other hand, the municipal government of Ivry-sur-Seine has implemented integration projects and encourages diversity. Although the Roma are not fully integrated in this municipality—they live in residential segregation and mainly hold informal jobs—the programs put in place generate less superficial intergroup relations, and certain prejudices toward the Roma appear to be lessening.
Essentially, the precarious conditions under which the Roma live generate responses and behaviours that reinforce already held prejudices. In particular, their extreme poverty and lack of resources mean that they tend to cluster in settlements often installed on private or municipal land. In addition, for cultural reasons, such as the fact that they do not all speak French, and because of their vulnerable situation, it is difficult for them to integrate into the job market, just as it is difficult for their children to go to school when they don't have enough to eat or dress accordingly. This situation creates a context in which each day is used for the provision of immediate needs, such as food and shelter. Improving the living conditions of the Roma, however, helps them to change some of their behaviours—or their responses to such conditions—and consequently, to dismantle the prejudices they are subject to. This is, moreover, what we observed in the discourse of respondents from Ivry-sur-Seine, where the municipality is actively involved in the integration of the Roma.
Access to Drinking Water and Toilets
Roma living in the settlement on rue Pascal in La Courneuve have no access to drinking water or toilets. They have to fetch water from fountains, while washroom facilities consist of the ground in a corner of the settlement. Consequently, some Courneuviens criticize the Roma for “stealing water,” and this without even paying taxes (Respondent 19C, age 20, plumber, lives near the Roma settlement). Another criticism is that the Roma “are dirty” and “stink.” These kinds of comments are found at the heart of negative, sometimes even hateful speech, particularly marked among respondents from La Courneuve. A logical response to this criticism would be that residents who do not have a source of water at their disposal see less regularly to their personal hygiene. The thought process of several respondents, however, is limited to the results and not the reasons behind them. This is the case with the following comments made by respondents who display particularly negative attitudes. To the article that we presented them with concerning the Force Ouvrière union's request for separate buses, replies resembled the like: “Ah yes! I have them on the bus and it's true that they smell, wow!” (Respondent 29C, age 58, works at Plaine Commune Habitat, lives far from the Roma settlement). And: “It is true that they have unpleasant odours. It is true” (Respondent 1C, age 36, real estate agent, lives far from the Roma settlement).
The same goes for prejudices concerning their unhygienic habits, since they urinate and defecate on the ground: “Did you see the Roma settlement, how they leave it when they leave? Dirty, not clean! They poop everywhere, they pee everywhere, they don't care! They throw food on the ground, everything, everything, everything! They are like that!” (Respondent 9C, age 40, unemployed, lives near the Roma settlement). This last response shows that the attribution of generalized characteristics to the Roma group comes from a visual experience that encourages the propagation of stereotypes. That this respondent does not have direct contact with Roma but lives near the settlement means that she has access to a strong visual repertoire, while her prejudices persist and are consolidated.
In Ivry-sur-Seine, the municipal government has implemented several initiatives to improve the basic living conditions of the Roma living in the settlement located on Truillot street, including access to drinking water and dry toilets. As a result, unlike in La Courneuve, no criticism was made of water theft or the fact that the Roma relieve themselves in the open.
Educating Children
A popular prejudice concerning the Roma is that parents do not want to send their children to school (CNDH Romeurope, 2013). This prejudice is more prevalent in the language of Courneuvian respondents than those of Ivryans. During the interviews, the deputy mayors of La Courneuve said that few Roma children attended school due to the shortage of teachers within their school system. In addition, according to a school mediator from the Association d'aide à la scolarisation des enfants roms et tsiganes du 93 (ASET 93),Footnote 7 they tried to provide Roma children in the municipality with access to education, a process that was not a success due to the lack of resources and volunteers as well as the difficult conditions under which schools in La Courneuve are currently operating.
Since the education of Roma children in La Courneuve is a real challenge for civil society actors, the assumption that Roma children do not go to school persists, which generates frustration: “They should be forced to bring their children to school” (Respondent 26C, age 30, works in airport security, lives far from the Roma settlement). Those Roma children who do not go to school are often seen in the streets of La Courneuve during the day: “The children are there in the morning, when the others are at school . . . imagine bringing one's children to the store, when the others are at school!” (Respondent 11C, age 55, grocery store security guard, lives far from the Roma settlement). These sightings also contribute to the criticism that the Roma use their children for begging instead of sending them to school:
I feel sorry for them on account of their housing conditions . . . but on the other hand, they send the children to beg! I feel uncomfortable passing by. It's a problem. When I come home from work, I have to say no, no, no, well, it's a bit of a bother. (Respondent 1C, age 36, real estate agent, lives far from the Roma settlement)
Inciting pity is how they go about it. I have seen them do it: they have crutches, they limp, but to take public transport, they run after the bus, the crutch under their arms! This is inappropriate! They use every tactic! They use children for begging and dogs so that people will donate. And we, we make generalizations about what we do not know . . . (Respondent 28C, age 57, works at Plaine Commune Habitat, lives far from the Roma settlement)
Respondent 28C's comments illustrate particularly well the intrinsic mechanism of the halo effect—that is to say, assigning to a group of people the behaviour observed in some people, as if this behaviour were common to all.
Conversely, the municipal government of Ivry-sur-Seine has made schooling compulsory for children of the settlement and has ensured that this ruling goes smoothly. City council members and volunteers were going to the settlement every morning to dress the children, wash them and take them to school. Mehdi Mokrani, deputy mayor for youth and social and solidarity policy in Ivry-sur-Seine, informed us during an interview that all children are now in school and that there is no longer any need for outside intervention. As the comments gathered during our interviews demonstrate, schooling the children has made it possible to dismantle the assumption that Roma children cannot be integrated. It also seems to have undone the stereotype of unruly Roma children whose parents use them to beg. No one sees them hanging around the streets of Ivry-sur-Seine on school days. As Roma parents are seen as wanting to educate their children, several respondents recognize this as a concerted effort on their part to integrate:
— Do you think that the Roma are integrated within Ivry-sur-Seine?
— In Ivry, there are little ones who have backpacks and who go to school. (Respondent 21I, age 29, delivery driver, lives far from the Roma settlement)
Intergroup Proximity
The impacts of the halo effect are reflected in the comments of respondents in the two municipalities. However, the fact that the municipality and certain members of the population of Ivry-sur-Seine offer support to the Roma for an integration process results in a change in perceptions toward them. First, our interviews revealed that since the Roma have access to and are encouraged to use municipal services, the other inhabitants of Ivry-sur-Seine meet them in places not associated with stereotypes, such as in the schoolyard, unlike the subways around where they are known to beg and steal from passers-by. An image of the Roma as “ordinary” people who do their grocery shopping “like everyone else” or who wait in line to buy their bread from the bakery can then be established and, as a result, eliminate certain prejudices. Further, the education of all the children of the settlement has encouraged intergroup contacts and promotes the idea that the Roma can and want to integrate. Proximity between groups within the framework of everyday situations allows for the development of new perspectives and thus more positive mutual awareness. The implementation of local integration policies can lead to the creation of mixed social situations and makes it possible to deconstruct prejudices held on either side:
We meet them every day: in the street, in the supermarket, in the bakery, in line . . . it's not a big deal. We're used to it. I have not had the opportunity to make Roma friends. But to the young people who hang around and some elders whom we recognize, whom we see regularly, we say hello. It's just a matter of respect. (Respondent 21I, age 29, delivery driver, lives far from the Roma settlement)
I saw them at the school party. They were completely . . . normal! (Respondent 17I, age 43, engineer, lives far from the Roma settlement)
I get the sense that they are integrating more and more. Already, we see them in supermarkets; before we didn't see them at all in the supermarkets. Before they were only on the street and now they go out in small groups with their little family. Like normal people. (Respondent 8I, age 25, caregiver for the elderly, lives far from the Roma settlement)
The Media and the Construction of Negative Attitudes toward the Roma
To reiterate, in the semi-structured interviews, no question made direct reference to the media, the answer to this question being overly complex to deal with within the framework of a qualitative analysis, and an awareness of how the media influences attitudes being rare. The analysis does allow for spontaneous comments from respondents who expressed (negative) knowledge of the Roma and whose source of information is the media. It is, however, necessary to mention that these comments can proceed from a post hoc rationalization—that is to say, where individuals with negative attitudes toward the Roma invoke what they observe in the media, finding comfort in them or as a means of justifying their comments.
First, some respondents mentioned being aware of behaviours or actions attributed to the Roma by the media, including newspapers and television reports. In this way, part of the negative perception of the Roma resides in the public imagination, having been socially constructed by the media. For example, although respondent 12C (age 27, lawyer, goes to the church next to the Roma settlement every Sunday) condemns racist and discriminatory actions against the Roma, she claims to have several prejudices against them, notably from the media:
— How do you perceive the neighbourhood relations between the Roma and the non-Roma?
— Filled with prejudice! Because it's true that in the face of the unknown, we're afraid. And we don't have the same way of life. We have it against them and they have it against us, I believe. . . . Prejudice can be unfounded. I am not in constant contact with the Roma, so I cannot say. . . . But from what we see on television, they are described as being non-sedentary people, and therefore really do not have the same way of life as us.
The comments made by this respondent show that the combination of media representation and the halo effect diminishes the desire to interact with Roma. It also reinforces the idea that the lifestyle of the Roma is not compatible with that of the members of the majority group. This idea was widely propagated by the media and by political elites, including Manuel Valls, who, as minister of the interior, proclaimed that “these populations have lifestyles extremely different from ours, and are obviously in conflict [read: with the majority]” (Dalibert and Doytcheva, Reference Dalibert and Doytcheva2014: 78; Fassin et al., Reference Fassin, Fouteau, Guichard and Windels2014: 30).
Other collected commentary shows the influence of the media in the construction of prejudices but also of collective fears and suspicions. A combination of negative representation and superficial intergroup contact reinforces stereotypes and affects attitudes toward the Roma. On the other hand, although some respondents find it difficult to distinguish between true and false in media reports about the Roma, negative news takes precedence. This is the case of respondent 25I (age 30, locksmith, lives far from the Roma settlement), who is suspicious of the Roma, in particular, because of everything that is said on television about them. This influences his interpretation of their behaviour:
Sometimes I see a poor man sitting there. And I see other people coming and talking to him. This same man gives over any money he has made, and the others then leave. I don't know if it's a network, but I think it is because you hear about it a lot on television. Frankly, after all, I don't know what I think about the Roma, I don't really have an opinion. But with everything we see on television . . . I think they tell both truths and falsehoods about the Roma . . . I don't know.
In addition, if migrant groups are essentially defined by the media, by political discourse, by administrative measures or by police attitudes, the stereotypes that result from these entities only consolidate fears and distrust of the Roma. This is reflected in the particularly negative comments of respondent 5I (age 58, baker, works near the Roma settlement), who feels “annoyed” by the presence of the Roma and thinks that they should be sent back to Romania:
— Have you ever been robbed by a Rom?
— No, it hasn't happened to me, but we know about it. And then, even in the metros in Paris, they know who is stealing and they make an announcement telling us to pay attention to our belongings. They know that they are there, and they know who they are, but there is nothing they can do about it because they are kids: they will arrest them and then release them. There is so much coverage on television about this.
— Do you have friends who have already had their personal effects stolen?
— No, not that I know of. But we often hear about it in the media.
Other respondents, without explicitly mentioning the media, echoed information about the Roma that they had not seen for themselves. The media and political speeches are a plausible source of explanation for their provenance, as well as other rumours everyone seems to be talking about without anyone seeming to have confirmed their veracity. Indeed, the media paint the subject with one brush and a limited colour palette, modifying the manner in which the Roma may be perceived and giving them visibility on the public scene. The same goes for the way social and political elites manage what they designate as the “Roma problem,” and this does not fail to publicly disseminate a certain image of the Roma, of which here are some explicit illustrations:
They accumulate funds here, after they leave for Romania, and they come back. . . . And when they go to Romania, they build castles. (Respondent 23C, age 52, unemployed, lives far from the Roma settlement)
When we hear that the government has been giving them money to leave, but that they take the money, go home and then come back for more. . . . They make a business out of everything. They have a disreputable side. (Respondent 29C, age 58, works at Plaine Commune Habitat, lives far from the Roma settlement)
Sometimes I put things down and I'm worried they'll steal them. They steal strollers and babies from these buildings. (Respondent 11I, age 27, homemaker, lives near the settlement)
There are many who could leave because they have very nice houses: they are better off there than here. It's true. Everyone knows it. (Respondent 6I, age 50, restaurateur, works near the Roma settlement)
Conclusion: Municipal Actions That Favour the Integration of the Roma
Our study aims to improve the understanding of the mechanisms for the construction of attitudes toward the Roma, with particular reference to French society. Interviews conducted in two municipalities in the Paris region allowed us to illustrate the impact of social interactions and intergroup contacts, local integration policies, as well as the media influence on perceptions and representations of migrants in precarious situations in their host societies. In particular, we have shown that social interactions and quality contacts between Roma and non-Roma are rare. They can even create a feeling of anxiety in certain individuals, who fear being robbed or attacked. Intergroup contacts are thus influenced by the beliefs of members of host communities about these newcomers.
However, informal and daily interactions are taking place in the two municipalities, particularly among merchants. Our study reveals that these contacts develop more when commingling is encouraged by the municipality, as is the case in Ivry-sur-Seine, where the involvement of the municipality in the integration of the Roma has made it possible to create a new social configuration, thereby mitigating certain factors that contribute to the dissemination and maintenance of stereotypes. Improvement of the Roma's daily living conditions, in particular with the installation of drinking water in the settlement and the accompaniment of children to school, allows them to transform some stigmatized practices, linked to the precarious conditions under which they live. These transformations aid in dismantling certain prejudices. For example, the sight of Roma bringing their children to school helps to undo the notion that they do not want to educate them. Further, interactions with Roma in the contexts of daily and socially accepted activities, in the schoolyard or at a municipal event, gradually introduces them into the same category as members of the majority.
Municipalities play a crucial role in the integration of Roma. They have an impact on the place that this minority occupies within the social space of the host society and, consequently, on intergroup attitudes. Finally, the media consolidate or influence the construction of these attitudes. The media-forged link between crime and Roma migrants generates fears among members of the majority group and, with these, a desire to exclude them. Taken together, media representations build an image of the Roma that is not easily overcome. As we have shown, this creates situations in which it is not easy for individuals to distinguish between what is true and what is not.
However, our study does not allow us to analyze the effects of quality intergroup contacts, as articulated by Gordon Allport (Reference Allport1954, Reference Allport1958). It would therefore be interesting to examine a municipality where Roma integration is more advanced—that is to say, where these populations live in and among the majority, in terms of housing, and have stable employment. Furthermore, considering the recent wave of attacks, as well as the rise of the extreme right in France, the increase in the level of tolerance toward the Roma, noted in the study carried out by the National Consultative Commission for Human Rights (CNCDH, 2016: 12) between 2014 and 2016, must still be assessed.Footnote 8 As we suggested in this study, local mobilizations and initiatives can be a plausible explanation for the change in attitudes toward the Roma. There may be other explanations, which deserve further consideration. The fact remains that such explanations are consistent with the central thesis of this study—namely, that local inclusion and awareness work can reduce prejudices, both individually and collectively held, toward migrants in precarious situations.
