This is an open access article distributed under the terms of the Creative Commons
Attribution License (Formed by different ethnic groups, the so-called Gypsies or Roma people have witnessed over the centuries violations and exclusion practices. Bearing in mind that, in the contemporary social context, media production has a great influence on the way people understand the world in which they live in, this study aimed to analyse the social representation of the Roma ethnic group provided by some news articles published in well-known newspapers in Brazil and Italy.
Different types of media play a key role in the process of construction of social thought (Höijer, 2011; Polli & Camargo, 2015), since they contribute to the production and dissemination of meanings regarding the most diverse social objects (e.g., Kroon et al., 2016; Polli & Camargo, 2015; Villano et al., 2010). These socially shared meanings are called social representations (Moscovici, 2012).
Social representations are forms of common-sense knowledge, i.e. a set of ideas, beliefs and values that help individuals to understand the world and to create a common social reality, allowing them to make the unfamiliar familiar. They facilitate social communication and serve as guides for individuals’ conduct in social interactions (Jodelet, 2001; Moscovici, 2012; Sammut et al., 2015; Vala & Castro, 2013). Social representations thus have an intimate relationship with communication (Moscovici, 2012; Sammut et al., 2015), insofar as “one conditions the other because we cannot communicate unless we share certain representations, and a representation is shared and enters our social heritage when it becomes an object of interest and of communication” (Moscovici & Marková, 2000, p. 274).
In this sense, Moscovici (2012) differentiated three systems of communication (in addition to conversation), which can be used by the press: diffusion, propagation, and propaganda. Diffusion involves different targeted audiences and entails coverage of a variety of topics, with an emphasis on the transmission of information, without the explicit intention to convince the public (Polli & Camargo, 2015; Sammut et al., 2015). Propagation, on the other hand, is aimed at a specific audience and intends to reconcile and organize information about a social object, in order to make it compatible with the values of the social groups involved. Finally, propaganda is used in conflict relations in which there is a threat to the group’s identity; and it generally involves persuasion, by means of approaching themes in a dichotomized and simplifying way, hence favouring the interests of certain groups over others (Cabecinhas, 2009; Moscovici, 2012). Thus, the media, such as newspapers, are crucial actors both in maintaining the already existing social representations within specific social contexts and in generating new social representations (Höijer, 2011).
Two interdependent processes are at play in the genesis of social representations: (i) objectification, which allows individuals to simplify information, via the materialization of ideas and/or concepts into something concrete, making reality more tangible; and (ii) anchoring, a process by which the unfamiliar is classified and accommodated within a system of more familiar categories (Moscovici, 2012; Sammut et al., 2015; Vala & Castro, 2013). Studies that focused on the analysis of sociogenesis have contributed to the debate on numerous psychosocial processes, such as the role of metaphors in the justification of social practices (Castro & Castro, 2018), the processes of discrimination and ontologization of ethnic minorities (Pérez et al., 2007), the personification or figuration of social objects such as suicide (Ordaz & Vala, 1997), the naturalization of history and collective memories (Hakoköngäs & Sakki, 2016), or even the relationship between iconic dimension and affective field in the identity context (Bonomo & Souza, 2010).
Ambivalent Representations of Gypsies in Italy
The national and international waves of migration that have crossed Europe in recent decades have produced a redefinition of relations between majority and minority groups within the borders of many countries. Among these is the group called ‘Gypsies’, whose presence in some countries has been decisively reshaped due to the recent increased mobility.
The existence of strongly negative attitudes, rooted in time, and discrimination by Europeans towards Gypsies is evidenced by several sources. Piasere (2004, 2012, 2015) shows that in Europe, anti-Gypsyism is neither a recent phenomenon, nor is it limited to Italy alone, and psychosocial research conducted in several countries (Bigazzi, 2009; Drydakis, 2012; Kende et al., 2017; Kende et al., 2022; Sam Nariman et al., 2020; Servidio et al., 2020; Tileagă, 2006; Villano et al., 2017) confirm its widespread character. The Roma have been, and still are, a target of prejudice, marginalization, and social exclusion across Europe, especially in Eastern-Central European countries. According to the Council of Europe estimates, in 2019 the presence in Italy of individuals belonging to different Gypsy ethnic groups (Roma, Sinti and Caminanti) is between 120,000 and 180,000 people (about 0.23% of the population). Half of the Roma living in the country is of Italian nationality, only 3% is nomadic, while most of the Roma population is sedentary. The Italian percentage is among the lowest recorded in Europe. Half of them, who arrived in Italy between the fifteenth century and 1950, have Italian citizenship, while the rest come for the most part from former Yugoslavia or are EU citizens of Romania. Although many Roma and Sinti in Italy are settled and have no experience of nomadism behind them, about a third are confined in the so-called ‘nomad camps,’ a condition that even documents of the European Commission define as ‘segregation’ and which is linked to the granting of nomadic status. The structural changes that have taken place in recent decades in European and Italian society have inhibited this character of being Roma, with the result of emptying of meaning and functionality of the life habits of the Gypsy people.
In Italy, the prejudice against Gypsies is very marked. Recent research shows that the stereotypes attributed to a specific group of Gypsies (the Romanian Roma) and the emotions felt towards them are overlapping, but the intensity of emotions, especially the positive ones, is very low (Villano et al., 2017). In the Italians’ imagination the Gypsy category emerges with an independent image compared to other groups: they are afforded the status of ‘different/other’ par excellence, and this is a very consolidated stereotype (Costarelli, 1999; Meneghini, 2017; Servidio et al., 2020; Zani & Kirchler, 1995).
Another recurrent stereotype in representing Gypsies is that of filth. The idea that the Gypsy does work that brings him/her into contact with filth (chimney sweep, dishwasher, etc.) is often present in research. Piasere states that Gypsies are seen as those who spread filth and disease and that filth is considered the "only human thing about gypsies" (Piasere, 1991, p. 181). Fontanella et al. (2016), in a research conducted on a sample of 530 participants, found that the attitude towards Gypsies is independent of gender and age of the participants. In summary, when there is a comparison between outgroups, Gypsies are always the worst rated by Italians, suggesting the idea that they can be considered as pariahs of our societies or as outsiders in the social map of human identity.
Trying to synthesize the stereotypical image of Gypsies that emerged in Italian studies over the last thirty years (Berti et al., 2013; Costarelli, 1999; Fontanella et al., 2016; Meneghini, 2017; Piasere, 2012; Villano et al., 2017; Volpato & Durante, 2008), there seem to be two images of Gypsies. One is the Gypsy of whom there is a perception of physical proximity, the recipient of a negative stereotype, widely shared and consolidated, towards whom hostility is not hidden because it is justified by the very behaviour of the outgroup: they are the ‘criminal’ Gypsies towards whom ‘one cannot but feel hatred’ and is ‘afraid to show hostility’. On the other hand, a non-threatening Gypsy figure emerges, linked to a positive image of freedom, a source of artistic inspiration, endowed with extraordinary musical skills. He is the ‘romantic’ Gypsy (the image of the Spanish Gypsy). The idea that there may exist a sort of ambivalence towards this sociocultural group, a phenomenon that for Moscovici (1981) is a constant in the behaviour of the majorities towards minorities.
The analysis so far reported in relation to Gypsies in the Italian context suggests, on the one hand, that the more the Gypsy is perceived as a neighbour, the more he/she is seen as a source of threat, on the other hand, that contact is rarely there because it is avoided a priori. In fact, the processes of ontologization towards Gypsies identified by Marcus and Chryssochoou (2005) and Pérez et al. (2007), and found in the Italian context by Berti et al. (2013), have shown that the perception of threat from Gypsies concerns the fact that they may jeopardize the human and cultural superiority of the group (majority). Italians’ prejudice against Gypsies seems to be based on a lack of knowledge of their cultures, coupled with a lack of contact that favours rigid stereotypes and a prejudice that is constantly more negative, contemptuous, persistent and brazen than other target groups. As a demonstration of this, in fact, in Italy the Roma population is often relegated to liminal areas of urban cities and its members live in campers, caravans or shacks crumbling located in camps of the suburbs.
The Social Invisibility of Roma People in Brazil
The history of Roma people in Brazilian territory has been characterized by different forms of violence since their arrival, after being banned from Portugal in the 16th century, as a consequence of different decrees regulating anti-Roma policies in Europe (Costa, 2005; Lyra Junior, 2020; Moonen, 2011, 2013; Teixeira, 2008). However, in Brazil, Roma people still suffer from a process of social invisibility (Almeida et al., 2013; Kemper, 2013; Murta et al., 2016). Despite five centuries of presence in the country, official data on Roma population are scarce, estimating that there are more than 800,000 people belonging to different Gypsy ethnic groups (Calon, Rom and Sinti) in national territory, according to the Instituto Brasileiro de Geografia e Estatística (IBGE, 2014).
For Andrade Júnior (2013), throughout this trajectory of violations, Roma people have managed to survive the hegemonic cultures and forms of sociability that are against their way of life. Such perspective is also corroborated by Fazito (2006), who analyses that the very name ‘Gypsies’ is “a stereotype elaborated based on collective representations, experienced by individuals from different cultural traditions over centuries of contact” (p. 689). Reflecting on the genesis of these stereotypic meanings, historian Moonen (2011) warns that they emerged in the 15th century, when different constructions were sedimented in these anti-Roma images, namely: “1) they were nomads, who never stopped for long in the same place; 2) they were parasites, who lived begging or taking advantage of the credulity of the people; 3) they were averse to regular work; 4) they were dishonest and thieves; 5) they were pagans who did not believe in God and also had no religion of their own” (p. 45).
Only in the last two decades further studies have been carried out on Roma people in Brazil in different fields of knowledge. Generally speaking, these investigations show that the Roma have been mainly represented based on: (i) a romanticized idea, as party-goers, who enjoy dance, music, and colourful clothes, and who are symbols of freedom; (ii) mystical images, being seen as clairvoyants and as people who hex others; (iii) stereotypes linked to crime, such as thieves and treacherous; and (iv) generalized meanings, within the framework of a different or exotic culture (Andrade Júnior, 2013; Bonomo et al., 2011; Fazito, 2006; Lima et al., 2016; Melotti et al., 2018; Moonen, 2011, 2013; Moscovici, 2009). Regarding the negative effects of these representations, different studies conducted with non-Roma have identified the existence of discriminatory practices directed at Roma people (Lima et al., 2016), often guided by the fear of the Roma ethnic group (Bonomo et al., 2011).
Such historical evidence and their anchoring in the present temporality highlight the need to elaborate, strengthen and consolidate pro-Roma public policies in Brazil (Brasil, 2013). Despite some advances concerning legal instruments (e.g., Brasil, 2002), several challenges remain in terms of initiatives to guarantee Roma’s rights and their inclusion in Brazilian society.
The Current Research
Aim and Hypothesis
Considering that the negative and/or romanticized stereotypes about Roma – which integrate the non-Roma hegemonic social imaginary (Bonomo et al., 2017; Cavalcante, 2013; Moscovici, 2009; Rizzi, 2021; Villano et al., 2017) – are frequently disseminated by different media (Kroon et al., 2016; Richardson, 2014; Schneeweis & Foss, 2017; Tremlett et al., 2017), the aim of this study was to investigate the social representations of Roma people in news articles from leading newspapers in Italy and Brazil.
The reason to analyze the newspapers of these two countries is driven not only by taking into account the fact that the authors of this article have been collaborating for years now, carrying out cross-cultural research on the topic of social representations of Gypsies, but especially by the fact that in both countries, during the selected period of investigation (2017 and 2018), there was the rise to power of right-wing party coalitions that publicly expressed their negative attitudes towards certain ethnic minorities, including Gypsies. This has encouraged us to analyze how newspapers presented news about Roma people, bearing in mind the Social Representation Theory.
On the basis of the preceding theoretical considerations, we can expect that in Italian newspapers there are more articulated representations of Roma people, especially negative ones, which can be traced back to the political events and news items that occurred in Italy during the period under examination. On the other hand, for Brazilian newspapers, since the Roma people are almost invisible to the community, we expect more crystallized representations anchored to other social objects of the collective imagination.
Materials and Method
Information Sources and Procedures for Data Collection
For the data collection, the Italian and Brazilian newspapers of reference were used, with digital edition and the highest number of online accesses in the years 2017 and 2018, namely: La Repubblica and Libero, in Italy; and Folha de São Paulo and O Globo, in Brazil. With regard to Italian newspapers, AUDIWEB was used as a source of information for the choice of newspapers. Brazilian newspapers were selected from information provided by the Communication Verifier Institute, since these newspapers occupied, respectively, the first and second positions in the national ranking of greatest dissemination in the reference years for data collection (see Table 1).
Table 1
Information About Newspapers
| Newspapers | Founded | Average circulation daily – Digital edition
|
|
|---|---|---|---|
| 2017 | 2018 | ||
| La Repubblica | 1976 | 1,445,789 | 2,567,569 |
| Libero Quotidiano | 2000 | 173,979 | 392,428 |
| Folha de São Paulo | 1921 | 167,592 | 194,855 |
| O Globo | 1925 | 99,689 | 173,527 |
As a strategy for data collection, the online search tool of each newspaper was used, by establishing the keywords and the period of publication of the news articles (from January 1, 2017 to December 31, 2018, totalling 730 days for data collection). In order to gather all the news articles from that period that addressed the topic ‘Gypsies,’ the following terms were used for the online search: cigano/a, ciganos/as (Gypsy/Gipsies), and nômade/s (nomad/nomads) (Brazilian newspapers); zingaro/a, zingari/e (Gypsy/Gipsies), and nomade/i (nomad/nomads) (Italian newspapers); and Rom (Romani), sinti (Synti), and Calon (both Brazilian and Italian newspapers). Throughout the data collection procedure, 2,491 news articles were selected: 315 in Brazilian newspapers (187 in Folha de São Paulo and 128 in O Globo) and 2,176 news articles in Italian newspapers (1,752 in La Repúbblica and 424 in Libero).
Procedures for Data Analysis
Having as starting point two data corpora (Italian and Portuguese), the textual analysis resource of the T-Lab software (Lancia, 2004, 2020) was used to process the data. Such program consists of a set of linguistic, statistical, and graphic tools for a quali-quantitative analysis of textual products. Several studies (De Rosa & Holman, 2011; Di Stefano, 2011; Mannarini et al., 2020; Rochira et al., 2020) have been developed with resources from the T-LAB software, evidencing its contributions in investigations of different methodological and thematic orientations, in view of the diversity of possible analyses generated by the program.
The T-LAB allows lemmatization1, categorization and the development of three main types or sets of analyses: co-occurrence analyses, thematic analyses, and comparative analyses. In this study, the corpus was first lemmatised, reducing them, for example, to the infinitive form in the case of verbs or to the masculine singular form in the case of nouns, and then semantically categorised in order to reduce the number of words in the dictionary (e.g. in the Italian corpus the words ‘theft’, ‘robbery’, ‘mugging’ and ‘pickpocketing’ were unified under ‘theft’, while in the Brazilian corpus the words ‘music’, ‘singer’, ‘piano’, ‘jazz’ and ‘song’ were categorised under ‘music’). This operation was done with the help of two independent judges and, in case of ambiguities or doubts, a third judge was appointed.
Then, using as data input the 100 most important categories (according to the criterion of maximum frequency and semantic relevance), for the databases of both nationalities a lexical correspondence analysis (CA) was carried out by using the tool ‘Co-words analysis’ (Lancia, 2007).
Lexical CA (Benzécri, 1973; Lebart et al., 1995) is a quali-quantitative technique of factor analysis applied to categorical data, based on the co-occurrence matrix.
The lexical CA allows to extract a certain number of factors (equal to the total number of categories minus 1) in order to synthesise the significant information contained in the co-occurrence matrix. Furthermore, for each factor extracted it is possible to assign a coordinate to each category, which allows to create graphs illustrating the distribution of the variables in the space, the factorial plan.
This technique is helpful for a qualitative reading of the factorial plan based on relationships of proximity/distance and/or similarity/dissimilarity between the categories: the closer they are, the more co-occurrences they have in the same lexical context (in our case, the newspaper articles); vice versa, the more distant they are, the less frequently they co-occur. Moreover, starting from some indicators produced by the analysis, called ‘Test Values’, it is possible to establish whether a category is actually represented in the factorial space, according to the criterion V-Test criterion ≥|1.96|, corresponding to the most commonly used statistical significance (p ≦ .05) (Lebart et al., 1995).
Results
The Social Representation of Roma People in Italian Newspapers
For the Italian daily newspaper database, the 100 categories used in the analysis are reported in Table 2, with an occurrence varying between 2,394 and 212 and an average frequency of 481.15. Through lexical correspondence analysis, 2 factors were extracted which explain 49.63% of the variance (36.39% F1 and 13.24% F2).
Table 2
List of 100 Categories From the Italian Database: Frequency and V-Test Values on F1 and F2
| Categories | F | V-Test F1 | V-Test F2 | Categories | F | V-Test F1 | V-Test F2 |
|---|---|---|---|---|---|---|---|
| Roma camps | 2,394 | -11.52 | -13.46 | Prison | 371 | -20.40 | 14.98 |
| Salvini | 1,643 | 46.17 | 16.75 | Police | 371 | -20.42 | 9.80 |
| Young | 1,556 | -23.25 | 14.34 | Group | 364 | -2.99 | 2.65 |
| Mayor | 1,160 | 17.81 | -16.22 | Suburbs | 362 | -6.95 | -7.54 |
| Home | 1,087 | -12.27 | -3.44 | Resident | 360 | -12.83 | -6.93 |
| Politics | 1,037 | 26.48 | 2.18 | Naples | 359 | 0.64 | -8.27 |
| Italy | 1,035 | 20.47 | 11.83 | Community | 336 | -5.04 | -2.76 |
| Rome | 997 | 4.79 | 4.31 | Investigation | 330 | -21.58 | 12.94 |
| Family | 949 | -17.16 | -4.09 | Theft | 324 | -21.02 | 17.03 |
| Council | 935 | 12.66 | -18.64 | Region | 317 | 9.93 | -11.30 |
| Child | 930 | -20.90 | -0.54 | Europe | 311 | 17.06 | 6.08 |
| Italian | 901 | 11.45 | 11.94 | Blaze | 311 | -19.10 | -9.72 |
| City | 841 | 3.46 | -11.52 | Emergency | 308 | 3.44 | -16.87 |
| Sons | 794 | -23.30 | 9.46 | Beppe Grillo | 299 | 32.36 | 4.95 |
| To live | 776 | -16.55 | -6.62 | To hit | 299 | -11.34 | 11.01 |
| Minister | 717 | 34.45 | 10.52 | Turin | 295 | -4.06 | -0.10 |
| Citizens | 664 | 2.28 | -8.30 | Virginia Raggi | 293 | 16.13 | -7.78 |
| Woman | 614 | -20.68 | 14.39 | Fire | 282 | -18.40 | -14.19 |
| Government | 609 | 35.35 | 7.78 | Immigr. policies | 281 | 4.38 | -7.08 |
| School | 587 | -9.19 | -5.66 | To die | 280 | -8.50 | 4.92 |
| To clear out camp | 582 | -2.24 | -18.88 | Violence | 279 | -5.81 | 10.84 |
| To arrest | 561 | -30.19 | 22.68 | Officers | 278 | -16.43 | 6.36 |
| Street | 561 | -10.22 | -6.71 | Indictment | 277 | -14.81 | 10.83 |
| Area | 539 | -6.50 | -18.76 | To steal | 277 | -13.06 | 14.96 |
| Problem | 530 | 7.52 | -6.79 | Scampia | 275 | -7.86 | -17.91 |
| Rights | 525 | 8.44 | -0.19 | Leader | 271 | 29.12 | 10.81 |
| Neighbourhood | 521 | -11.08 | -10.77 | Luigi Di Maio | 270 | 33.37 | 12.36 |
| Lega Nord | 508 | 34.82 | 9.00 | Foreigner | 269 | 1.48 | 7.47 |
| Milan | 501 | -4.19 | 5.94 | Territory | 266 | 3.36 | -9.55 |
| Racism | 497 | 10.43 | 7.36 | Gypsy | 264 | -3.53 | 8.26 |
| To migrate | 492 | 20.90 | 4.60 | Meeting | 262 | 6.43 | -4.19 |
| To report | 487 | -13.06 | 8.68 | Pow. of attorney | 259 | -19.27 | 6.41 |
| To occupy | 476 | -2.25 | -13.40 | Vote | 259 | 23.75 | 5.96 |
| Law | 474 | 7.41 | 3.82 | Matter | 255 | 5.33 | -6.79 |
| Social | 469 | 1.82 | -11.04 | Leghista | 250 | 26.94 | 7.58 |
| Nomadic | 466 | -12.99 | 0.74 | Facility | 250 | -5.37 | -8.84 |
| President | 461 | 14.01 | -4.79 | Cinque Stelle | 248 | 32.41 | 4.85 |
| Democratic Party | 460 | 30.47 | 2.58 | Solution | 247 | 1.17 | -13.16 |
| Different | 443 | -3.93 | 1.43 | To leave | 246 | -10.62 | -10.36 |
| Life | 441 | -6.64 | 3.13 | Ill. settlement | 244 | -6.77 | -10.26 |
| Association | 440 | -0.26 | -12.79 | Right wing | 239 | 18.53 | 2.29 |
| Man | 422 | -15.78 | 15.36 | Roma Families | 239 | -15.26 | -6.50 |
| Wastes | 419 | -9.37 | -17.05 | Illegal traffic | 239 | -1.79 | -1.33 |
| Car | 384 | -20.13 | 7.43 | Need | 233 | 4.69 | -1.38 |
| Victim | 382 | -19.38 | 16.04 | Shack | 231 | -12.85 | -11.25 |
| Project | 381 | 5.37 | -13.03 | Death | 220 | -11.62 | 8.45 |
| Fear | 380 | 0.32 | 3.54 | Culture | 218 | 3.94 | 0.44 |
| Safety | 379 | 7.27 | -6.68 | Camper | 217 | -19.99 | 1.77 |
| Carabiniere | 374 | -25.51 | 16.12 | Crime | 213 | -11.09 | 12.97 |
| Left wing | 374 | 24.44 | 4.95 | Census | 212 | 17.19 | 3.75 |
Figure 1 shows the categories with V-Test ≥|1.96| projected on the factorial plane obtained through the factorial analysis of correspondences. Different representations related to Roma people emerge in the four quadrants.
Figure 1
Correspondence Factor Analysis of Italian Data
In the lower left quadrant, we find a description characterized by the protagonists, also non-Roma, of the events reported in newspapers ('Roma families', 'family', 'resident', 'child' and 'community') and by the contexts in which Roma people live ('To live', 'street', 'Roma camps', 'suburbs', 'neighbourhood', 'home', 'school', 'Turin'), often characterised by conditions of degradation ('shack', 'To occupy', 'illegal settlement', 'area', 'wastes') which have come to the fore for particular news events ('Scampia', 'fire', 'blaze') which have been the pretext for animated debates for the 'solution of the problem' ('To clear out camp', 'association', 'To leave', 'facility').
In the lower right quadrant, the reference to the local political debate ('meeting', 'council', 'mayor', 'Virginia Raggi', 'President', 'region') on the 'social' 'problem' due to the presence of Roma in some specific cities ('Naples', ‘city’) and ‘regions’ of Italy have been the focus of discussions on the issues of ‘safety’ and the ‘need’ of the ‘citizens’, the ‘immigration policies’ and ‘rights’ of the Roma and the need to find ‘solution’ to the ‘emergency’ of the problem, through the implementation of ‘projects’ of intervention on the ‘territory’.
In the upper right quadrant, we find the themes connected to the cultural political debate that, at a parliamentary level, sees the Italian political parties clash on the politics to be adopted towards the Roma minority. We thus find the reference to the different political parties ('5 Stelle', 'Lega Nord', 'Democratic Party', 'leghista') and its representatives ('Salvini', 'Luigi Di Maio', 'Beppe Grillo', 'minister', 'leader', 'government') that conflict in the parliament in 'Rome' on opposing 'politics', on the one hand, the 'right wing' which, supported by the 'fear' of the 'foreigner' on the part of the Italians ('Italian'), proposes to carry out a 'census' in 'Italy' of all Roma, on the other hand the 'left wing' which accuses opponents of 'racism', claiming that a census would be unconstitutional ('law'). The debate on advanced policies at European level ('Europe') towards Roma and migrants in general ('To migrate') completes the picture just described, the topic that focuses on the term 'culture', which is declined both as a culture of tolerance towards minorities and as a recognition of the Roma population as bearers of traditions and their own culture. Finally, the category 'vote' refers to the national elections held on 4 March 2018 and all the topics falling within this third quadrant characterized the long electoral campaign advanced by the political parties.
In the upper left quadrant, we find terms that refer to the news events involving some individuals of Roma ethnicity as both victims and perpetrators. Therefore, there are several terms semantically referable to the ‘crime’ ('crime', 'To steal', 'theft', 'To hit', 'violence', 'death', 'life', 'To die', 'indictment', 'prison', 'To report', 'investigation', 'To arrest'). Then some categories appear that indicate the actors involved in the actions ('carabiniere', 'police', 'officers', 'victim', 'young', 'sons', 'man', 'woman', 'group', 'power of attorney'), sometimes described with negative and stereotyped terms ('nomadic', 'gypsy' and 'different'). Then the picture is completed by some references to specific details of what happened ('camper', 'car', 'Milan').
In order to give a more qualitative reading of the factorial plan, we can observe how it highlights three thematic cores. The first, located at the top left, defined as ‘The facts of criminal news’, is represented by the news that narrate criminal events where individuals of Roma ethnicity have been involved in the role of both victim and perpetrator. The second, in the lower part of the plan, labeled ‘The emergency of Roma camps’, refers to those political debates that have occurred at the local level (in some Italian cities) on the governance of territories (the camps) where Roma groups live, following news events that have brought to the fore situations of strong degradation. The third, on the top right, collects the nucleus of news appeared on newspapers centered on the national and international debate on policies of intervention towards the Roma population.
The Social Representation of Roma People in Brazilian Newspapers
Regarding the Brazilian newspaper database, Table 3 shows the 100 categories used in the analysis, with frequency of occurrence that varied between 857 (Music) and 16 (Quilombola), with an average frequency per category of 101.33. Also for Brazilian data, through lexical CA, 2 factors were extracted which explain 20.30% of the variance (12.23% F1 and 8.07% F2).
Table 3
List of 100 Categories From the Brazilian Database: Frequency and V-Test Values on F1 and F2
| Categories | F | V-Test F1 | V-Test F2 | Categories | F | V-Test F1 | V-Test F2 |
|---|---|---|---|---|---|---|---|
| Music | 857 | 15.29 | -3.37 | Concentration camp | 77 | -11.18 | -10.56 |
| Brazil | 360 | -1.47 | 5.68 | To change | 77 | 0.74 | 1.96 |
| History | 243 | -2.82 | -5.10 | Government | 75 | -10.07 | 4.79 |
| Person | 234 | 3.40 | -0.45 | Novel | 73 | 4.38 | -6.77 |
| Work | 210 | 4.54 | -1.22 | Difficulty | 72 | -0.52 | 0.47 |
| Death | 200 | -4.82 | -7.35 | Period | 72 | 2.78 | -2.79 |
| Theatre | 194 | 9.57 | -4.52 | Director | 71 | 5.89 | -3.54 |
| Life | 189 | 0.80 | -4.43 | Party | 71 | 6.23 | 1.61 |
| Woman | 175 | -3.59 | -3.03 | France | 71 | 1.19 | 0.30 |
| Family | 173 | -2.49 | -1.73 | Character | 70 | 4.20 | -3.42 |
| Politics | 165 | -8.03 | 3.48 | Poetry | 69 | 4.91 | -5.33 |
| Italy | 163 | -6.32 | 4.09 | Region | 69 | -0.24 | 3.94 |
| Art | 162 | 9.85 | -2.69 | Brewery | 69 | 4.50 | 12.75 |
| To live | 161 | 2.87 | -0.97 | To exist | 67 | -0.04 | 5.48 |
| City | 154 | 4.08 | 5.69 | Movement | 64 | -0.02 | 2.27 |
| Group | 153 | 0.86 | -0.03 | Opera | 64 | 3.57 | -2.64 |
| World | 152 | 2.51 | 1.56 | Important | 63 | 1.40 | 0.68 |
| Film | 147 | 6.26 | -0.81 | Different | 61 | 0.61 | 1.15 |
| Public | 141 | 4.77 | 2.83 | TV | 61 | 2.21 | -3.25 |
| School | 140 | 1.64 | 2.62 | People | 59 | -3.28 | 2.36 |
| Child | 134 | -5.00 | -4.74 | Cinema | 58 | 4.95 | -1.38 |
| Culture | 133 | 3.94 | 2.19 | To help | 58 | 1.51 | 1.57 |
| Book | 122 | -1.06 | -8.06 | Pub | 57 | 6.00 | 13.11 |
| Dance | 116 | 7.83 | -2.38 | Carnival | 56 | 4.00 | 1.98 |
| To marry | 115 | -1.34 | -1.08 | To survive | 56 | -7.35 | -5.85 |
| Concert | 115 | 10.58 | -1.78 | Sidney Magal | 55 | 8.33 | -3.37 |
| Strong | 112 | 0.29 | -0.55 | Power | 53 | -2.92 | 2.28 |
| To write | 107 | 0.75 | -4.82 | President | 52 | -3.99 | 4.04 |
| War | 106 | -9.46 | -5.66 | Way | 52 | 1.59 | 3.80 |
| Nazism | 105 | -14.22 | -10.38 | Soap opera | 52 | 6.78 | -4.31 |
| Right | 102 | -6.94 | 3.11 | To fight | 51 | -3.64 | 2.08 |
| Beer | 102 | 5.91 | 15.34 | Inclusion | 51 | 0.47 | 0.82 |
| Germany | 102 | -10.60 | -4.36 | To defend | 51 | -3.65 | 3.32 |
| Traditional | 101 | -2.73 | 4.76 | Writer | 51 | 0.97 | -5.03 |
| Portugal | 101 | -1.62 | 2.71 | Spain | 51 | 0.24 | -1.12 |
| Street | 94 | 2.30 | 1.54 | Piece | 51 | 2.02 | -5.07 |
| Territory | 94 | -6.83 | 2.53 | Immigration | 50 | -7.74 | 5.16 |
| Europe | 94 | -5.67 | 0.58 | Social | 49 | -3.16 | 1.35 |
| Actor | 92 | 7.93 | -4.69 | Issue | 48 | 0.03 | 0.22 |
| USA | 91 | -0.44 | 1.91 | Jew | 48 | -10.09 | -3.72 |
| Community | 91 | -5.98 | 4.26 | Majority | 48 | -2.37 | 2.70 |
| Freedom | 87 | -0.34 | -1.48 | Law | 45 | -1.91 | 3.66 |
| Leadership | 87 | -8.07 | 5.18 | Latin American | 43 | 1.89 | -0.12 |
| Festival | 86 | 8.28 | 1.05 | Democracy | 43 | -3.68 | 3.53 |
| Young | 83 | 3.24 | -0.52 | Violence | 43 | -3.50 | 1.47 |
| Man | 81 | -0.47 | -1.45 | Black people | 42 | -1.62 | 0.33 |
| Indigenous | 81 | -2.30 | 4.26 | Salvini | 39 | -8.29 | 5.36 |
| Bolsonaro | 81 | -6.71 | 6.31 | Racial | 24 | -1.20 | 1.78 |
| Ideas | 80 | 2.05 | 1.66 | Minority | 19 | -5.00 | 1.66 |
| Camp | 78 | -10.03 | -3.14 | Quilombola | 16 | -4.13 | 4.17 |
Figure 2 shows the 82 categories with V-Test value ≥|1.96| projected on the factorial plane. Different representations of Roma also emerge for the Brazilian data.
Figure 2
Correspondence Factor Analysis of Brazilian Data
In the lower left quadrant, the ‘history’ of the ‘Nazism’ period is covered. Elements such as ‘camp’, ‘war’, ‘Jew’ and ‘Germany’ refer to this context of suffering of ‘families’ (‘woman’, ‘man’ and ‘child’) in an attempt to ‘survive’ the ‘death’ sentences in the ‘concentration camp’. Historical and biographical narratives are reported in the form of ‘book’ and described in the news articles.
In the lower right quadrant, content about the universe of ‘art’ predominates, with its literary, cinematographic, theatrical, television and musical manifestations (‘concert’, ‘cinema’, ‘TV’ and ‘theatre’) and its production modalities, such as ‘film’, ‘soap opera’, ‘opera’, ‘music’, ‘dance’, ‘poetry’, ‘piece’ and ‘novel’. Elements such as ‘character’, ‘actor’/‘director’, ‘writer’/‘To write’ and the Brazilian singer ‘Sidney Magal’ (singer of a famous song about the Gypsy Sandra Rosa Madalena) complement this image of the artistic-cultural productions and manifestations, based on the process of creating and directing the work and its interpretation for dissemination to the public. Categories like ‘period’, ‘person’/‘young’, ‘work’ and ‘life’/‘To live’ make up the stories and descriptions presented in the artistic works mentioned in these news articles.
In the upper right quadrant, there is, in a more independent and grouped way, the metaphor of Gypsies as nomadic, an idea that can be observed in the terms ‘bar’, ‘beer’ and ‘brewery’2, which address the context of production and consumption of the so-called Gypsy beers, which would be those produced without a fixed territorial reference or a factory of its own. ‘Carnival’, ‘party’ and ‘festival’ referring to the cultural habits, in which many non-Gypsies use elements of Gypsy ‘culture’ during this period of celebrations in Brazil.
Terms such as ‘public’, ‘way’, ‘city’ and ‘street’ point to a practical dimension of appropriation of collective spaces, promoting ‘ideas’ that would highlight important ways ‘To change’ the ‘world’. Regarding the element ‘school’, it refers to three different uses: carnival samba school; music, dance, and theatre school; and formal school.
Finally, in the upper left quadrant, as a contemporary context, the ‘social’ question manifests itself in the ‘power’ relations between the ‘majority’ and a ‘minority’ that seeks ‘To fight’ and ‘To defend’ their right ‘To exist’ in their own terms. Identified within this social category, therefore, there are the local challenges of the Brazilian ‘region’, of a ‘people’ who suffer from ‘violence’. This dynamic is associated with the ‘traditional’ ‘community’, like the ‘Quilombola’ and the ‘Indigenous’, in addition to ‘immigration’ – groups that have in the ‘territory’ a common agenda for their struggle. Against this resistance ‘movement’, the dimension of ‘politics’ and the debate over ‘law’ and ‘democracy’ come into play, by questioning the ‘right’ of these peoples and social groups. Conservative discourse in right-wing ‘governments’, as in national contexts such as ‘Italy’ and ‘Portugal’ (in ‘Europe’), as well as in ‘Brazil’, is mentioned in the news articles as a contemporary challenge for minority groups that live on the margins of the neoliberal interests adopted by certain political ‘leaderships.’ The then Italian prime minister ‘Salvini’, as well as ‘Bolsonaro’, who at the that time was a ‘presidential’ candidate in Brazil, stand out in this set of meanings.
The integrated analysis of the results allowed the identification of four central themes: (1) ‘Forgotten Holocaust’, located in the lower-left pole, including elements that refer to the Jewish Holocaust, with only residual mentions of Roma people and without any specific reference to the Roma Holocaust (also known as Porrajmos); (2) ‘The Roma as a character’, in the lower right, which refers to the idea of Roma people as characters in different types of artistic productions (literature, cinema, music, soap operas, among others); (3) ‘The Roma as a metaphor’, in the upper right, which presents meanings associated with the image of parties and wandering life; and (4) ‘Traditional peoples and communities’, in the upper left, in which Roma groups are mentioned along with discussions about the rights of different ethnic segments (such as Quilombolas and Indigenous peoples, for example).
Discussion
Based on the results mentioned in the preceding section, we argue that the discourse expressed towards the Roma is culturally determined. Generally, the data that emerged from the analysis of the news in Italian and Brazilian newspapers reflect old and sedimented stereotypes of the Roma - linked to crime, music, folklore (sorts of ‘natural’ traits) - and this finding once again confirms that anti-Roma prejudice is ancestral and may function as a common marker for cultural identity. The Roma people represent not only an outgroup, but an outsider in the social map of human identity: Roma do not seem to exist as people, but only as a stereotype. Many studies on Romanophobia discourses (for example, Cervi & Tejedor, 2020 in Italy; Goodman & Rowe, 2014, Turner, 2002, Schneeweis, 2009 in the UK; Molnar, 2021, Tileagă, 2005, 2006 in Romania) highlight this dimension. As Tileagă (2006) shows in a brilliant paper on Roma people and discursive psychology, they are delegitimated and morally excluded from public discourse: talk about Gypsies can be extremely prejudicial, because they themselves are blamed for prejudice towards them on the grounds that they are not just different from ‘us,’ but are also a morally inferior group. Roma people are not seen as being part of the same moral community as the other groups, but only as a stereotype. As Anderson (1983) suggests, communities must be distinguished "by the style in which they are are imagined" (p. 16). And our article demonstrates that Romani people are consistently imagined, both in Italy and in Brazil, as separate and excluded, strongly tied to prejudices that time does not seem to have erased. This finding, therefore, supports the suggestion that a level of prejudice that would not normally be accepted towards minority groups is commonplace when discussing Gypsies. It appears that when it comes to Gypsies, the norm against prejudice is suspended: prejudice towards Gypsies is not denied but rather is presented as an inevitable outcome given the actions of Gypsies themselves (Goodman & Rowe, 2014). Actually, the crime cliché in Italy is a recurring theme in the studies on prejudice against the Roma, and our data on newspapers confirm this tendency. The ‘Gypsy issue’ seems to have become a stable trait of the Roma public discourse: our data express the full force of the age-old question that still sees the Roma as a ‘permanent’ emergency and problem: Roma are treated as “a special subgroup, occupying a spatial and temporal enclave, thus denying the immediacy of their existence” (Sigona, 2005, p. 750). This stereotype underlies many of the decisions made within the justice system and is the core concept upon which the Italian government has constructed the so-called ‘Nomad Emergency,’ giving it extraordinary legal powers over this population, despite the fact that very few Roma and Sinti are actually nomadic (Clough Marinaro & Sigona, 2011). This exclusion serves to (re)produce and perpetuate a social distance between the majority and the minority and to deny similarities between ‘us and them’ (Molnar, 2021; Villano & Zani, 2007). The construction of the discourses about Roma people as deviating from the norm leads people to see them as abnormal and, thus, to accuse them more readily. People categorize Roma outside the boundaries of normal human behaviour. This attribution serves as an explanation as to why the Roma people continue to be seen at the fringes of society, where it is impossible to build any space for dialogue, and this attitude is concretely visible in the Italian social policy of building ‘nomad camps’ for Roma people. Political authorities in Italy have constructed the Roma as a dangerous community that is not constitutive of the Italian nation, where the Roma are treated as the ‘other.’ Enforcing the separation of Roma from mainstream society through the use of ‘camps’ is a key factor in the perpetuation of their status as ‘inner enemies:’ Roma people live segregated in the camps, and thus the vicious circle persists.
In Brazil, even after five centuries, the image of Roma people in national territory has been socially constructed by means of a policy of forgetfulness, associated with the characterization of Roma based on meanings with negative or romanticized connotations, conceiving them as eternal foreigners (Almeida et al., 2013; Bonomo et al., 2017; Costa, 2005; Murta et al., 2016; Teixeira, 2008), dimension that has also been evidenced in the Italian context (Costarelli, 1999; Meneghini, 2017; Servidio et al., 2020; Villano et al., 2017; Zani & Kirchler, 1995). In fact, as Moonen (2013) warns, “Brazilians ignore, for example, that they already had a Roma president, or descendant of Roma, President Juscelino Kubitschek, JK” (p. 12), in addition to important intellectuals, authors and representatives of national art and literature. Although Roma characters from different audio-visual productions are widely known, as verified in the results of this study, there is an undeniable and expressive erasure of the contributions of Roma people to the construction of knowledge and relevant cultural products in the history of societies (Moscovici, 2009).
Considering the paradoxical dynamics of social visibility-invisibility and discrimination-romanticization involving Roma people, Lyra Junior (2020) argues that it is “complex and, at the same time, contradictory, to state that they are invisible and victims of prejudice, simultaneously. After all, how can one be prejudiced against what s/he ignores? How can I reject what I don’t perceive to exist?” (p. 207). As it was possible to verify in the Brazilian news articles that we have analyzed, the narratives that are disseminated portray this reality and contribute to its maintenance. The Roma generally appear as characters (in films, soap operas, theatres, operas, books, etc.) or associated with other groups, being mentioned incidentally (in the midst of the debate over traditional peoples and communities in Brazil or within discussions on the Jewish Holocaust). In this dimension, it is worth noting that the so-called forgotten Holocaust or forgotten genocide of Roma people during World War II, in which thousands of Roma were murdered in Nazi concentration camps (Cavalcante, 2013), constitutes another warning about this process of marginalization of Roma people in the course of history via policies of forgetfulness.
Apart from the ‘character-Roma’ within the artistic-cultural productions, that is, considering only the ‘real-Roma,’ it is possible to identify the dynamics of social in/visibility that both sustains and is sustained by a complex set of relationships, which lead to several challenges, as denounced by the Roma themselves (Brasil, 2013). At the level of non-Roma hegemony, social exclusion has been a historical instrument for making Roma people socially invisible, adding to their suffering due to prejudice and discrimination, especially when they affirm their social identities and fight for the social recognition of their rights. Or as Kemper (2013) explains, the “invisibility of these excluded people only comes into matter when they disturb the social order” (p. 110), whether by questioning the instituted normative system, denouncing the violations they suffer, participating in politics and fighting for their rights, or even by simply existing in public spaces and within social interactions. The effects of these processes can be identified in the most diverse situations of violence experienced by Roma people, which include banishment, criminalization, dehumanization, among many others (Berti et al., 2013; Bonomo et al., 2011; Fazito, 2006; Kroon et al., 2016; Lima et al., 2016; Moonen, 2013; Moscovici, 2009; Pérez et al., 2007; Richardson, 2014; Teixeira, 2008; Tremlett et al., 2017).
Therefore, the objectification of the social representations of Roma, as our data have shown, seems to be anchored in meanings guided by different semantic patterns (Bonomo & Souza, 2010; Castro & Castro, 2018; Fontanella et al., 2016; Ordaz & Vala, 1997; Pérez et al., 2007): the personification of the Roma is based on characters from the artistic universe and elements that make up this scenario. This study shows that, once again, the discourse on Gypsies portrayed in the press is largely stereotyped: what persists is the image of the ‘Gypsy’ as the ‘child of the wind’ who enjoys freedom in love and spirits, has innate artistic talents and cherishes the small things in life: the ‘pervasive’ myth about Roma culture. Gypsyism as a metaphor for errant practices, such as the use of the term ‘Gypsy’ to designate breweries; or the merger of the specific Roma group into superordinate categories, such as ‘social minorities’ or ‘Jewish Holocaust.’ While the Roma may be considered exotic and interesting at a distance, they became deviant when enmeshed in the social mainstream.
As a reflex of this process of social construction, markedly guided by the dynamics of dehumanization of Roma people (Lima et al., 2016), the absence of narratives about Roma groups in Brazil is notorious in the analysed news articles, while in Italy the themes found depicted the Gypsy issue as a social problem and as a focus of political-cultural discussion. Despite advances in the proposition and implementation of pro-Roma public policies, the effort to marginalize the Roma individual in the sphere of public spaces and domains is a concrete reality, since the “figurative knot of the social representation of Roma penetrates and turns into a knot of a social representation of non-Roma, referred to as ‘Gadje’, which, to some extent, operates on their history, way of life and way of conducting themselves” (Moscovici, 2009, p. 667).
The historical temporality signals the process of anchoring of these meanings (Hakoköngäs & Sakki, 2016; Jodelet, 2001) based on markers and antinomies such as nomadism/sedentarism or, even, strange/familiar (Moscovici, 2009). In other words, the stereotypic meanings attributed to the so-called gypsies were mainly based on narratives launched by numerous societies that projected the idea of the outsider to be feared and blamed for the scourges that affected the ingroup (Fazito, 2006; Moonen, 2011).
The results of this study, as well as the reference literature, indicate numerous tasks in the field of psychosocial work with Roma groups. The process of invisibility of Roma people in Brazil, and the discrimination they suffer in Italy, reveal different kinds of discrimination that support anti-Gypsyism, which despite being systematically denounced by Roma associations and segments, still lurks in the ambiguities that remain present in a discourse that praises and romanticises the ‘character-Roma’ and Roma myth. In fact, as a research conducted by the Carlo Cattaneo Institute (Istituto Carlo Cattaneo, 2018) shows, 85% of Italians express negative feelings towards Roma, confirming that, although Italy is one of the countries with the lowest percentage of Roma and Sinti population, anti-Gypsyism is higher than the European average (Pasta, 2019). In our opinion, and as the results of this research show, the problematization of the Roma ‘issue’, becomes a discursive opportunity for the political party (e.g. Salvini's Lega Party in Italy) best able to symbolically articulate (and emotionally charge) the ‘issue’, presenting itself as the best option to solve the ‘problem’. Such discourses, however, discriminate and exclude the Roma individual, either through the dissemination of negative stereotypes as intrinsic elements of what it is to be Roma or through the efforts made to maintain the Roma as an undesirable and invisible people in the sphere of public approval (Bigazzi, 2009; Drydakis, 2012; Piasere, 2015; Rizzi, 2021; Sam Nariman et al., 2020; Sigona, 2005; Tileagă, 2006; Villano et al., 2017).
This study tried to understand the scapegoating of Roma in the media discourse and, in particular, with the role exerted by ‘media élites’ (van Dijk, 1998) in the (re)production of anti-Roma discourse. It became evident, then, how the mass media play a prominent role in the coverage of ethnic affairs, and they often do so in a way that promotes ethnic prejudices and, indirectly, discriminatory social practices based on negative beliefs about others. However, considering the key role of the media in not only reinforcing already existing social representations but also in producing and disseminating new ones (Höijer, 2011), several measures could be put in place by media outlets to challenge negative stereotypes regarding different minorities, such as the Roma.
These measures were widely discussed in previous publications concerning media coverage of migration/migrants (e.g., International Organization for Migration [IOM], 2019; Office of the United Nations High Commissioner for Human Rights [OHCHR], 2019; Triandafyllidou, 2017) and of minorities in general (e.g., Respect Words, 2017), as well as in studies regarding media coverage of Roma people, more specifically (e.g., Istituto Internazionale di Scienze Mediche [IISMAS], 2014). Following the recommendations presented in these publications, some of the ideas that could foster a more balanced reporting on the Roma, are: a) apply appropriate terminology according to international law, respecting human rights principles; b) avoid pejorative terms and stereotypes, for instance, not mentioning the Roma origin of perpetrators when reporting crimes, when this information is not useful to understand the news, or not referring to Roma people as clandestine, foreigners, among others expressions; c) include Roma as sources in news articles and share more positive stories involving them; d) recruit Roma individuals to work in newsrooms and provide diversity training to the teams.
Affirmative actions are therefore a possible and strategic tool for the long and much needed effort of social change, especially by the effective implementation of public policies in the different regions of the country, guaranteeing quality access to the fundamental rights of Roma people and respecting their ethnic-cultural references (Moscovici, 2009). And yet, in order to reduce prejudice, we can glimpse a way forward. Not certain solutions, absent in this field, but we can think about some reflections and actions aimed at a possible change. The first is presented to us by Martha Nussbaum in her book Not for Profit. Why Democracy needs the Humanities (Nussbaum, 2017, p. 26), a praise of humanistic culture, often forgotten by democracies and the education of young people, says the author. Humanistic culture that lays the foundations and provides the tools to "think critically, transcend localisms, address the world problems as 'citizens of the world', to sympathetically portray and recognize the category of the other". To counteract prejudices, which sometimes lead to the ‘disgust’ of other people, Nussbaum argues that it would be necessary to teach authentic things about different groups (about ethnic minorities, religious, gender, and so on), while promoting critical thinking and encouraging responsibility in the younger generations and, to this end, we believe that a fundamental role can be played by schools of all levels, because it is the first social agent that should be in charge of education for social inclusion.
In our study, very reductive representations of the Roma people emerge from the press: they are criminals, a problem to be solved, or simply characters from literature or a stereotypical symbol of freedom, all representations that generate mistrust and fear (Bigazzi, 2009; Lima et al., 2016; Moscovici, 2009; Villano et al., 2017). We are sure that the educational and inclusive action of schools can help transform and, above all, extend not only these representations of the Roma, but also those produced about other minorities.
However, for this to occur, the political will of different countries is required to constantly monitor the social representations that are formed in a given social context towards minorities subject to discrimination in order to develop and apply, on the instructions of the Ministries of Education, public education policies, which promote, for example: (1) the development of studies and research on the history, the traditions and the ways of life of Roma people; (2) the periodic updating of curricula, monitoring the existence and dissemination of discriminatory content regarding minority groups, such as the Roma; (3) the adjustment of educational materials (such as textbooks, booklets, dictionaries, among others) that contain negative stereotypes about Roma; (4) the elaboration of educational projects aimed at recognizing and valuing ethnic diversity in the daily life of the school community; (5) the inclusion, in the academic calendars, of festivities and commemorative dates that celebrate the culture, the identities, and the knowledge of minority groups; and (6) the implementation of training activities involving the education professionals in order to constantly improve their qualifications on issues such as ethnic diversity, social minorities, prejudice and social discrimination (Brasil, 2002, 2006, 2012, 2020). Not only that, this ministerial operation should be carried out with particular attention to the risk of introducing, for example through the narratives present in the textbooks, contents proper to the so-called hidden curricula (Apple, 2004; Giroux, 1978; Rahman, 2013), knowledge systems proper to the dominant culture that lead "students are socialized ... to accept ideologies related to power, knowledge, and social stratification that maintain inequities and injustice (Anyon, 1981)" (Thomas & Dyches 2019, p. 602), with the result that social representations that one would like to deconstruct are, on the contrary, reinforced.
Finally, in this study we have seen how in discursive production, ideologies usually operate indirectly, first through the construction of attitudes and knowledge about social groups These social representations of events interact with models of context that people dynamically construct of a communicative situation, which inevitably influence discourse structures. This happens most clearly at the level of discourse content or meaning, that is, in what people say: the themes (topics) they choose or avoid, the standard topoi of their arguments, the implicit or explicit information, the meanings placed at the beginning or in the background, the details specified or not, and so on.
We also know that information that is favorable to one's group or unfavorable to the outgroup will tend to be important and explicit. Information that portrays us negatively (or others in a very positive light), on the other hand, will tend to remain implicit, hidden, vague, and undetailed. For example, intonation and accent in words and sentences can make meanings more or less salient, as can visual structures in newspapers such as page layout, font size and type, color, or photographs.
In conclusion, it should be noted that the links between discourse and ideology go both ways. Not only do ideologies influence what we say and how we say it, but the reverse is also true: we acquire and change ideologies through reading and listening to a lot of different information and positions, both written and oral. Speech is the most important of these social practices, and it is the only one that can directly express and thus transmit ideologies. Therefore, it is necessary for the academic community to continue working on the analysis of discourses towards minorities, in order to try to counteract mental schemes that are all too often stereotypical and generalized.
Limitations
This study tried to identify the field of meanings associated with Roma people, based on the information disseminated by the press. Among the limitations of the study, it is the non-use of social media as another source of data, since such communication channels have been considered the main means of access to information by the general population. Furthermore, in the case of the Italian data, it is noteworthy that an analysis focused on the gender category was not carried out. Such analysis could provide important information with which to understand the social discrimination suffered by Roma people in that territory, since Roma women have been constructed in the social imaginary as representatives of Gypsyism in general and also suffer from the intersection of this phenomenon based on the social markers of gender and ethnicity.
A further limitation of this research is due to the qualitative-quantitative method adopted in that it allows us to capture only the main thematic cores that emerge from the articles published in the newspapers. A future line of our research could be oriented towards the use of more qualitative methodologies, such as the traditional content analysis, but also the discourse analysis, which would allow us to grasp the complex articulation of the contents handled in the news and how newspapers contribute to the reification and legitimization of different belief systems that concern the ethnic Gypsy population.
Despite these limitations, as social psychologists. we believe it is important to consider both individual actions and the consequences on the individual, as well as the importance of institutional norms, educational practices, and policies that can generate exclusion and marginalization, as in the case of the Romani people. Cultural, social and political factors contribute to create (possible) experiences of exclusion at various levels - individual, community, social - affecting different groups - migrants, LGBTQI+, women, minorities. Thus, there is a need to work at the policy level to foster a climate of inclusion while respecting everyone's rights. Supporting the importance of policies in generating a climate of dialogue and communication is crucial, because laws are part of the system and form the images and words with which people see and think about the society of which they are a part. In fact, studies have shown that countries in which diversity policies are promoted generate more cultural norms about integration and inclusion (Guimond et al., 2013). All this obviously relates to the discourse of the importance of ideologies also conveyed by mass media and social media in shaping the attitudes of individuals and groups in a society. They reflect and reinforce the psychological and social needs, motivations, and characteristics of individuals and groups, forming mental schemata? that help people explain, justify, or criticize aspects of reality (Jost, 2006).