Structural Anti-Roma Racism in Italian Middle Schools

: This study examines persistent racism, despite its formal denial, through an interdisciplinary approach that combines critical race theories, whiteness study and social psychology. It questions whether the analytical and empirical distinction between different forms of prejudice—despite their coexistence—can yield positive outcomes in the fight against racism. Considering the case of the Roma community in Italy, specifically within the universalistic approach of the Italian school system, the level of prejudice was measured in a sample of 305 middle school teachers. The main findings confirm the persistence of blatant prejudice concealed beneath a veneer of subtle racism and reveal that an educational institution’s rhetorical commitment to democracy may not substantially impact the behavior of democratic teachers when interacting with Roma individuals. The systemic inequalities ingrained within the education system are reminiscent of colonial times, where practices that once suppressed the potential of Africans are now used against Italian Roma citizens. This article finally suggests how teacher training might be improved to reduce racism, based on the different profiles of prejudicial attitudes detected among teachers.


Introduction
The conclusion of World War II, although marked by the condemnation of fascist and Nazi atrocities, did not signify the end of racism.Instead, it led to the dissolution of a clear division between two racial approaches: naturalism and historicism.Western rationality has historically used these approaches to assert the superiority of its identity [1]."Naturalism" between the 18th and 20th centuries posited the innate inferiority of certain human groups, while "historicism", dominant post-1945, attributed inferiority to historical and cultural factors.If naturalism lies at the very base of the simplest forms of contemporary racism and presents mostly recognizable forms, historicism presents subtler forms.This latter approach permeates the logics of democracy and progressivism, and it is often difficult to see how its underlying assumptions affect different racial groups.
Indeed, blueprints, guidelines, and official documents on the matter of education, integration, and the like are frequently based on conceptualizations of otherness that share the same roots as naturalism.In other words, those who support this approach seem to view "history" as a mere sequence of facts that are not rooted in the experience and perspectives of any one subaltern group.What was at stake in the process of the subjugation of radical others-in terms of logic, worldviews, losses, etc.-is missing in most of these documents, and undoubtedly in those that address education.Historicists advocate for agency through education for racially inferior individuals, aiming to unlock their universal potential, while naturalists reject this notion [1].This analysis is grounded in the premise that these two traditions of racial thought have coexisted within states and individuals, as posited by critical race theory and social psychology studies.The newer ideology of "cultural pluralism", which seemingly promotes biological equality and ethnic diversity, has been implicated in justifying racism [2], and while post-Holocaust anti-racist norms formally deny racism, they contribute to its persistence [2][3][4].
In light of these premises, this article aims to reveal the persistence of a naturalist logic against the Roma within the universalistic historicist approach of the Italian school system, questioning whether the analytical and empirical distinction between forms of prejudice, despite their coexistence, can yield positive outcomes in the fight against racism.
Prejudice, particularly anti-Gypsyism 1 presents a significant challenge to the integration of Roma individuals into mainstream society.Despite national and European integration policies, which have been scrutinized for harboring hidden prejudice [5], anti-Gypsyism remains pervasive.However, there is a notable lack of research in social psychology that specifically addresses this form of prejudice.The gap is even more significant in studies with an interdisciplinary approach.Recent empirical studies have demonstrated that overt negative stereotyping of the Roma strongly impacts the perceptions of the dominant group regarding Roma integration [6].Thus, current empirical research on anti-Roma racism is crucial for bridging the gap between Roma and non-Roma students.
Even though they constitute only 0.2% of the population in Italy, the Roma 2 , who mostly hold Italian citizenship, are perceived as culturally distinct [7,8], and simultaneously, as inferior.Furthermore, despite data showing that Roma children perform comparatively poorly in school [9], there is broad international consensus that structural discrimination does not exist, supporting the popular understanding that racism cases are the result of individual errors.Ethnographic studies expose the continued existence of segregated classes for Roma [10,11], while inclusive classrooms do not ensure non-discriminatory treatment due to widespread prejudice among the Italian population.In fact, according to the longitudinal comparative analyses of data from the European Values Survey (EVS) from 1999 and 2008, Italy has the highest level of anti-Roma prejudice, as seen from the percentage of respondents who declared they would not want Roma neighbors [12].In the school context, teachers' prejudices are based on cultural attributes such as low motivation, limited family support, and poor hygiene [13].
In the Sicilian city under scrutiny here, a settlement for Yugoslav Roma was founded during the 1990s due to the Kosovo conflict.This settlement initiated segregated education schemes reminiscent of the so-called "Lacho Drom" classes in Northern Italy, which were exclusively designed for Roma students during the Seventies and shaped the imagination of teachers across generations.Despite the efforts of a Roma inclusion project that was also implemented in this city (2013-2023) [14], recent research has found that first-grade secondary school teachers in schools with Roma pupils maintain that the presence of "culturally diverse students" does not necessitate changes in teaching methods [15], effectively making them invisible.
This research is positioned within the body of writing that challenges the persistence of a naturalist logic that discriminates against Roma and its continuity in the post-Auschwitz era [16].It includes different aspects of the same logic, such as institutional racism [17], eugenic local policies [18], biological racism [19], and the long tradition of psycho-social racism and psycho-moral racial conceptions [20].This study adds an analysis of racism in education, acknowledging the historical backdrop of Italian colonialism.
Using the concept of the "boomerang effect" [21], this article argues that Italian colonial racism resurfaces in contemporary European education, impacting both Roma students and non-Roma educators.Cesaire's historical interpretation of the boomerang effect predicts the "process of barbarization" of Europeans at home [21] and views Hitler's ascent not as an aberration but as a symptom of a morally corrupt culture brought about by colonialism.The creation by Europeans of a hierarchy of human worth as a defense against brutal acts towards non-Europeans has degraded their culture and entrenched racism.This historical influence is examined through empirical research using Pettigrew and Meerten's prejudice scales [22], in order to discern racial ideologies and reveal insights into the implicit structure of racism in schools.
An interdisciplinary methodology was used to examine racism in educational institutions, integrating critical race theories-including whiteness studies and postcolonial studies-with social psychology.Ladson-Billings suggests that the effectiveness of critical race theory in elucidating inequalities in education requires a continuous critique that extends even to "legal victories" [23].
In the Italian context, the transition from special Roma-only classes to mixed classes during the 1980 calls for a critical examination through the lens of race, and it is imperative to reflect on why such a system leaves Roma students behind.Whiteness, which is often not perceived or recognized as a racial identity, comes with privileges.
As McIntosh observed, there are certain dominant societal components that perceive racism as a phenomenon that disadvantages others, but they are not taught to recognize the concomitant aspect of white privilege that benefits them [24].When examining how educators construct their white identity and become aware of their relationship to racism, it is evident they are immersed in a white worldview that is so ingrained [25,26] that they are seemingly innocent [27].By "whiteness", we also mean the belief that the dominant group sees reality as objective.This perception prevents educators from engaging with the discriminated subjects and their perceptions of reality, their life goals, and their feelings [2].The combination of whiteness studies and cognitive psychology helps transcend the individual and structural limitations inherent in addressing racism and highlights how the denial of racism helps to maintain oppressive structures [28] and negatively influences educational practices, leading to distrust and self-fulfilling prophecies [29,30].
According to Giuliani and Lombardi-Diop [31], the Italian national identity evolved through forms of opposition to "the other"-initially southerners and later Africans.Thus, discourse surrounding the Roma is entrenched in a colonial racist framework, as Clough Marinaro notes [19].The author considers them post-colonial subjects, even though their discriminatory treatment precedes colonial policies.For example, during the colonial era, hygiene was used to assert Italian racial superiority, creating distinct social structures for natives and white colonialists [31].Quite typically, today, in everyday discourse, the racialization of Roma in Italy perpetuates colonial ideologies, associating nomadism with dirtiness [19].In post-war Italy, thirty years after reconstruction, the creation of nomad camps clashed with modern values of domestic comfort and cleanliness, and consequently, the camps came to symbolize a pre-modern unhygienic lifestyle.Likewise, Picker notes that nomad camps and segregated classes for Roma students reflect Italy's colonial legacy [32].
In this article, "prejudice" and "racism" are used interchangeably, recognizing the intrinsic link between the individual and the structural.According to Essed's concept of "everyday racism" [2], racism is a phenomenon that simultaneously encompasses both macro-structural and micro-individual dimensions.It is deeply ingrained in everyday discourse and practices, rendering it an invisible yet pervasive reality.Irrespective of its mode of expression, it invariably carries serious implications due to the simultaneous action of its two components: cognitive (prejudice) and behavioral (discrimination).Even when racism is manifested by an individual, such as a teacher, their thoughts and actions continue to perpetuate the racial disparities that have historically pervaded the system.These disparities are reproduced at a microlevel, transcending individual consciousness.The term "Race" is preferred over "ethnicity" to dispel the myth that ethnicity is a scientifically valid concept; instead, race is acknowledged as a social construct used to mask economic and political inequalities through biological and cultural arguments.
This article consists of three sections.The first section delineates pivotal research on the history of the Roma in Italy, examined through the lens of postcolonial and critical race theory [33].The second examines prejudice from an intra-individual standpoint, drawing from social psychology literature.It examines prejudiced attitudes by means of a questionnaire distributed to teachers in a Sicilian city and explores the impact of interracial contact in the classroom on the reduction of prejudice.The third section makes recommendations for tailored anti-racist training based on categories of racism while acknowledging certain research limitations.In the light of these premises, the article highlights the importance of acknowledging white privilege.
In this latter regard, both authors acknowledge their white privilege in their interactions with local Roma individuals.This privilege is evident in instances of preferential treatment from political figures (including receiving attention over the phone, gaining access to offices, earning trust for initiatives, even when the Roma represent the primary focus, etc.).

Historical Analysis
Within critical race theory studies, Goldberg [1,34] has noted the coexistence of naturalism and historicism in the formation and practices of European states, dating back to the early stages of modern political theory.However, Goldberg acknowledges that the predominance of one racial approach over another depends on the racial group under consideration.
In the context of Italy's treatment of the Roma, there has been constant oscillation between two racial conceptions on both institutional and ideological levels.For instance, during the fascist period (unlike the Third Reich, which considered all attempts to educate "Gypsies" as meaningless [35]), Italy established a school inside a Roma concentration camp.This discovery highlights an attitude towards re-education and historicism, even during the peak of naturalist and pseudo-scientific racism [36].Ideologically, during this period, the Roma were considered to be of Indian and Aryan origins, yet they were described as a race of criminals [37], polluters [38], and were subject to elimination [39] due to their nomadic lifestyle.
Although the historicist conception prevailed over the naturalist stance in the 20th century, especially from 1945 onwards, the case of the Roma community illustrates the persistence of the latter.
In the 1960s, the first European-level document addressing Roma issues proposed the establishment of camps to sedentarize nomads as a response to the educational challenges posed by their nomadic lifestyle.In Italy, this mission was undertaken by the Opera Nomadi, a moral organization used by the state to establish Roma camps in various Italian cities.These camps were accompanied by segregating educational interventions introduced during the 1960s within the camps themselves and later in special classes exclusively for Roma children in educational institutions [16].
Within the framework of liberal democracy and legislation aimed at integrating students with disabilities, the special and differential classes established in the 1960s were abolished and replaced by support teachers who worked in regular classes, as mandated by Law No. 57/77.In this context, the creation of differential classes known as "Lacho Drom"  constituted a form of discriminatory treatment towards the Roma community, reaffirming naturalist approaches that echoed those applied to Africans in the colonies [32].
Over the next two decades, "Nomadic camps" became the focal point for non-Roma education initiatives for the Roma people [40].Indeed, after the closure of the "Lacho Drom" classes, Roma students continued to be monitored through inclusion programs, in which non-Roma educators facilitated coercive education, with school enrollment becoming a requirement for their stay in the camps.
Beginning in the 2000s, Roma oppression in Italy resulted in institutional discrimination, including forced evictions.Legislative Decree No. 122/2008 deemed Roma camps a state of emergency, comparing the Roma themselves to a natural disaster.Racial slurs in the media and racially driven acts of violence against the Roma community signaled a shift away from historical racist ideologies, which had portrayed them as free-spirited nomads, towards the adoption of a "naturalist racial conception" [32], which redefined the Roma as a contaminating difference, suggesting that biological racism portrays the Roma as physically inferior, polluting, and sometimes inhuman [19].
In 2009, despite Roma inclusion in mainstream classes, a Milanese school considered "at risk" due to its proximity to a Roma camp used Ministry of Education funding to build showers in order to promote the integration of Roma students.Rejecting accusations of racism, the principal argued that her hygiene program was more effective than a hundred conferences on racism.This school project, known as "acqua e sapone" (soap and water), reflected the re-emergence of Neo-colonial forms of racism while denying its existence [41].The concept of a civilizing mission [42] was rhetorically replaced by the myth of integration.

Social Psychology Studies
Within social psychology studies, the accepted model of prejudice acknowledges the complex coexistence of direct and indirect forms of prejudice [43].However, unlike insights from critical race theory, there is a greater focus on the utility of distinguishing between the two types of prejudice.
This differentiation captures an ambivalent post-modern racial attitude whereby individuals may adhere to egalitarian values, presenting as non-racist, yet harbor negative beliefs towards non-white groups, often unconsciously.This nuanced understanding is crucial for identifying individuals who engage in cold and indirect discrimination within normative contexts that prohibit overt racism.This is addressed by constructs referred to as aversive and subtle racism, considered conceptually equivalent even though the application context differs-America for aversive racism [29,44] and Europe for subtle racism [45].
In the 1990s, Pettigrew and Meertens introduced pivotal measures to assess "new" subtle forms of racism alongside the more traditional colonial form, creating a scale that measures both types of racism-prejudice [22].Arcuri and Boca highlight a strong correlation between the two types of prejudice [43], measured according to this scale in various research contexts [43].However, they argued that to distinguish latent from manifest prejudice, it is necessary to identify a variable differentially linked to both forms of prejudice.For example, in their study and in Rattazzi and Volpato's research in Italy [46], political affiliation shows a differential association with the two forms of prejudice.Progressives score high solely on the subtle prejudice scale, while conservatives exhibit elevated scores not only on the subtle but also on the manifest scale.This suggests that in socially unconstrained contexts, progressives may manifest racist behaviors contrary to their political group's norms, as teachers might deviate from democratic values when they feel they are not being observed.
In another Italian study, it is the cultural level that is thought to differentially link these two types of prejudice.When comparing groups of parents, the author found no link between subtle prejudice and educational attainment.However, it was noted that cultural levels influence the expression of prejudice, channeling it toward more socially acceptable forms, except for highly prejudiced individuals who intensify such attitudes [47].
In the present research, classroom contact between teachers and Roma students is considered pivotal in distinguishing individual prejudices.
Contradictory results emerge from the literature on inter-group contact.According to the meta-analysis conducted by Pettigrew and Tropp [48], inter-group contact reduces prejudice, including hidden forms [49].Particularly positive emotions are elicited by contact and have a greater influence on prejudice, reducing fear, for example, than past experience and stereotypes.However, studies with Roma communities in Italy indicate that contact itself fails to mitigate "prejudiced attitudes" [50], and in some cases, the effects of contact depend on other variables, as studied by Allport [51], including the status of the out-group (whether equal or asymmetric), the amount, and the type of contact.
Pertinent to this article is the Spanish research carried out by Gómez-Berrocal and Moya [52] using Pettrigrew and Meertens' scales [22] to measure prejudice against Roma amongst high school students.They identified ambivalent prejudice, with subtle manifestations prevailing in line with anti-prejudice norms, while overt prejudice was characterized by feelings of threat and avoidance in intimate relationships.
This article establishes the link between Pettigrew and Meertens' theorized manifestation of individual prejudices [22] and state racial formations [1] (as discussed in the historical analysis).

Research Context
The Sicilian city where this study was conducted is geographically located between wealthier and more impoverished regions of the Mediterranean and is linked both to historical Italian emigration routes and the current migratory flows from Africa.With an estimated population of 217,000 inhabitants, about 5.1% of the city's residents are foreigners.
While some social psychologists [43] have noted a possible decrease in prejudice towards African immigrants among Sicilians compared to other Italian regions, no scientific studies have confirmed a similar trend towards the Roma.Public discourse at local, regional, and national levels perpetuates cultural stereotypes and portrays the Roma community as a threat [53], contributing to the normalization of the forms of prejudice analyzed here.
If we consider the Roma's migration patterns in Italy since the 15th century, the most notable flow pertinent to the urban context studied here involves Roma from former Yugoslavia [54].In the 1990s, a densely populated settlement for Yugoslav Roma was established in this city due to the conflict in Kosovo.Their status was mainly that of refugees.Though the settlement was dismantled in 2012, it influenced educators' perceptions of the Roma.The "nomad camp" became the starting point for education initiatives for the Roma put forward by local Catholic volunteers, who initially provided literacy classes, showers, and a bus service so as to integrate Roma students into state schools [55].However, after many families were relocated in 2012, the camp's after-school service was suspended for three years, resulting in academic setbacks for Roma students.After a hiatus, after-school activities resumed in 2015, with increased Roma involvement through their association.Since 2021, local institutions have formally withdrawn permission for the Roma to use local spaces for after-school activities, despite the fact that Roma participation has led to improvements in their academic performance.
Lower secondary education in Italy, which is commonly referred to as "middle school", is the second stage of the first cycle of compulsory education after primary school.This phase of education lasts three years and is intended for students aged 11 to 13.
In a political climate marked by rising racist extremism and its denial [56], there is no comprehensive national program specifically dedicated to the integration of Roma students, except for the one implemented in 13 Italian metropolitan cities [14].Moreover, they are frequently classified as foreign students, even though they were born in Italy, due to the requirements of Italian citizenship legislation, which only allows the children of foreign parents to begin the process of obtaining Italian citizenship at the age of eighteen.Both the 2007 document on the intercultural education of foreign pupils [57] and the two national strategies for Roma inclusion [9,58] recognize Anti-Gypsyism as a distinct form of racism, with intercultural education aiming to combat this through an awareness of Roma history.When comparing the new "National Strategy" with the previous one, it becomes evident that the focus has shifted from cultural to socioeconomic factors regarding the educational challenges faced by the Roma community.However, child labor and a distrust for school are regarded as Roma-specific concerns, and the strategies advocate for Roma engagement and family involvement without addressing the broader context of Italy's racial and colonial history [58].
Since 2013, this research city in question has participated in the aforementioned Roma Children Inclusion program, extended until 2023, focusing on education and local service networks.This program declared that its direct beneficiaries were Roma minors (3-14 years), and its indirect beneficiaries were all the non-Roma children in the participating schools, as well as teaching staff and social workers [14].
Although these interventions targeted too few teachers in schools with Roma students from the local immigrant Roma community, this study's findings, and indirectly those of another local researcher [13], indicate the ineffectiveness of these interventions, including intercultural methodology workshops and Roma history lessons.
Italy's national legal framework prohibits overt racism and penalizes speech or expressions that incite hatred or discrimination based on race, ethnicity, or religion, but enforcement of these laws is still deficient.According to Article 604bis of the Italian Penal Code, such offenses carry a maximum sentence of up to one year and six months, but this provision is not consistently implemented.There have been relatively few judicial rulings, despite the fact that political officials regularly incite hatred against the Roma community.

Materials and Methods
This study is part of a broader, ongoing research project on racism that uses both quantitative and qualitative methodological approaches [59].The qualitative methods include interviews with teachers and Roma activists, which will be the subject of a forthcoming article.
The empirical study 3 presented in this article includes a descriptive analysis aimed at investigating the presence and manifestation of prejudice among educators.In addition, an exploratory analysis has been carried out to examine the differences in prejudice between teachers who have dealt with Roma students and those who have not.There is also a comparative analysis to assess the significance of the differences in expressed prejudice between these two groups of teachers.
The objectives of the empirical study are as follows: 1. To identify the type of prejudice exhibited by Italian teachers in relation to Roma students 2. To determine the relationship between the prejudice of Italian educators and their contact with Roma students.
The hypotheses of this study are as follows: Hypothesis 1a.Italian teachers exhibit higher levels of subtle prejudice than blatant prejudice in relation to Roma group in general.

Hypothesis 1b.
The widespread presence of subtle prejudice is accompanied by a substantial persistence of overt prejudice among Italian teachers.

Hypothesis 2a.
It is further assumed that the contact experience does not reduce prejudiced attitudes towards the Roma group in general, due to the asymmetric relationship established in the education system [51].

Hypothesis 2b.
An increased rate of teachers who exhibit high levels of overt and subtle prejudice (classified as fanatical and subtle) is expected among the group of teachers who have had contact experiences with Roma students in class.
Therefore, the variables in this study are as follows: The presence of prejudice among teachers, the type and intensity of prejudice (subtle or overt) expressed by teachers, and the categorization of teachers into general profiles based on scores obtained on prejudice scales.
Additionally, the independent variable (contact) is analyzed to determine its effect on the dependent variables (subtle prejudice form and overt prejudice form).
Middle school teachers were chosen for this study because, more than any other level of education, they reflect the democratization process within schools.This began in the 1960s with the reform of the middle school (Law No. 1859 of 31 December 1962), which paved the way for mass enrollment.Previously, after primary school, education was divided into vocational schools and three-year secondary schools, with entrance exams that determined pupils' early possibilities and reflected their social status.
Purposive sampling was used, and different schools in the city were invited to participate in this study, with the sample including teachers who agreed to participate in this study.The total population of first grade secondary school teachers in the reference municipality in Sicily was 1999, so in order to obtain a representative sample, a calculation based on the tables provided by Arkin and Colton [60] was performed.
The sample is statistically significant, with a confidence level of 95% and an error of 5%.The original sample consisted of 305 teachers; however, the scoring procedure for the ad-ministered test revealed that 10 of the respondents, all white Italians, had a prejudice score equal to the theoretical median.According to the protocol, they were excluded from the subsequent analysis.The final sample consisted of 295 teachers, 81% of whom were female and predominantly between the ages of 40 and 60 (69.8%).Overall, 36.7% reported having at least one Roma student in their class, and less than 7% of respondents reported having participated in training aimed at integrating Roma children in the classroom (Table 1).The statistical tests used aimed to be rigorous, both in terms of the distribution of the sample and the testing of the results (non-normal distribution) and also in terms of taking previous studies into account in order to compare the results.
In view of the object of the investigation and the subjects involved, this research adopts a dominant interpretative stance within a scientific frame.

Instrument
The questionnaire used in this study consists of the overt and subtle prejudice scale [22], adapted into Italian by Arcuri and Boca [43], and following the item order proposed by Manganelli Rattazzi and Volpato [44].The instrument was validated by its authors in a large-scale survey conducted in various European countries to assess the attitudes of native populations towards recently arrived immigrant groups.The authors used data from seven independent national samples collected at the end of 1988 as part of the Euro-barometer survey No. 30 conducted by the European Community [61].Despite the fact that the scale was conceptualized and developed in the late 1980s, it is still used today as a suitable tool for evaluating new expressions of prejudice towards different groups of people [62,63].
The scale can be adapted to the target group by making slight modifications [43] 4 and consists of twenty indicators divided into two sub-scales.
The first sub-scale represents overt prejudice and focuses on two dimensions of the naturalistic form of racism in the colonial era and Nazi and fascist regimes: perceiving the out-group as a threat to one's own group and the avoidance of any contact with out-group members.
The second sub-scale represents subtle prejudice and evaluates "socially acceptable" expressions of prejudice that can stem from the following dimensions: the defense of the traditional values intrinsic to one's own group, the suppression of positive emotions, and the exaggeration of cultural differences.
Dimensions of direct prejudice include indicators related to the perceived dishonesty of out-group members, the perception of descending from populations considered less capable than Italians, attitudes towards sexual intercourse and intermarriage with members of the Roma group, and finally, work interaction (especially when a Roma employee holds a superior position within an organization).
Dimensions of indirect prejudice include standard in-group (in our case, Italian society) values such as "striving harder", the refusal of any favoritism for minorities; in essence, these values emphasize individual responsibility for success or failure according to Neoliberal arguments.Moreover, in the dimension termed "suppression of positive emotions", prejudice is examined as the denial of positive feelings, represented by two indicators concerning the lack of admiration and solidarity towards the Roma.This solidarity refers to the refusal to acknowledge positive feelings such as empathy or support for the Roma community.Lastly, the exaggerated perception of differences in language, religious practices, sexuality, and values passed down to future generations indicates negative prejudice towards the out-group.
The questionnaire employed a six-point Likert response scale, without a neutral point, for agreement/disagreement responses and a five-point Likert scale for similarity judgments.
The instrument allows for the classification of participants into three categories: "Fanatics", characterized by high levels of overt and subtle prejudice; "Democrats", with low levels of prejudice in both sub-scales; and "Subtles", who show low levels of overt and high levels of subtle prejudice.
To assess the reliability of the instrument, Cronbach's Alpha values were calculated for both subscales.The results show high internal consistency: α = 0.805 for overt prejudice and α = 0.839 for subtle prejudice.These values indicate greater reliability in measuring subtle prejudice compared to previous studies [22,43,52].Therefore, the results indicate that the prejudice measurement instrument used in this study is reliable and valid for assessing levels of overt and subtle prejudice towards Roma groups in this research area.According to the authors, their prejudice scales should be moderately correlated, expecting the coexistence of different forms of prejudice from the beginning.In this study, the scores of the two scales show a correlation (Spearman's coefficient) of α = 0.71, indicating a significant positive correlation between the two forms of prejudice.This result is consistent with similar previous studies where the correlation coefficient ranged from a minimum of α = 0.48 to a maximum of α = 0.70 [43].

Procedure
The questionnaire was distributed to 16 school complexes (lower secondary schools) located in an entire Sicilian municipality using various distribution methods [64].To preserve the confidentiality of the schools involved in this study, their names have been censored.
Approximately 25% of the sample omitted answering questions 15 and 23, which were related to the sexual sphere.Therefore, for greater reliability of results, it was deemed appropriate to exclude these questions from the questionnaire and adequately recode the calculations of the test score and its sub-items, using the following new theoretical median values as references: Overt prejudice = 31; Subtle prejudice = 30.5.

Results
The obtained data were analyzed using two statistical programs, R (version 4.3.2) (https://www.r-project.org/) and SPSS (version 27) (https://www.ibm.com/spss).To verify the normality of the variables, the Shapiro-Wilk test was applied.Since the dimensions of prejudice did not follow a normal distribution (as indicated by the p-values obtained from the Shapiro-Wilk test), a non-parametric inferential approach was adopted for the statistical analyses (Table 2).The results obtained from the Wilcoxon signed-rank test reveal that the mean score for subtle prejudice is just slightly above the theoretical median, while the mean score for overt prejudice is below the theoretical median.Scores below the theoretical median indicate low prejudice, while scores above the median indicate high prejudice.This positioning indicates that the participants exhibit a medium-to-high level of subtle prejudice and a medium-to-low level of overt prejudice (compared to the theoretical reference means).Additionally, the statistical analysis showed that the mean score for subtle prejudice is significantly higher than the mean score for overt prejudice.
It can be observed that in the scale for subtle prejudice, the dimension of cultural differences is the highest, followed by the dimension of defense of Italian values.On the other hand, in the scale for overt prejudice, the dimension of "Fear and Threat" emerges as predominant compared to the other dimensions (Table 3).The data presented in Table 4 illustrates the percentage results for teachers who scored higher than the theoretical median in prejudice.It is evident that a notable proportion of teachers have a score for overt prejudice higher than the theoretical median (15.6%), while a more significant percentage of teachers exhibit a higher score for subtle prejudice than the theoretical median (51.86%).All these results highlight the existence of a significant number of teachers with high prejudice, which manifests in both subtle and overt forms.
The results regarding the percentage distributions for the different typologies of teachers based on the scores obtained in the scales (Table 5) reveal a majority of subjects classified as "Democrats-Equalitarians" (low level in both prejudices), a minority of subjects classified as "Fanatics-Bigots" (high level in both prejudices), and 37.63% of subjects classified as "Subtles-Hidden" (low overt prejudice and high subtle prejudice).Cases classified as "Errors" are individuals with high explicit-blatant prejudice and low implicit-subtle prejudice.These cases are theoretically uninterpretable; those who discriminate directly will also do so indirectly; therefore, their extremely limited presence within the classification of teachers in different typologies (Table 5) confirms the validity of the measurement instrument used [46].
The analysis also focused on the variable of contact between teachers and Roma pupils in the classroom.
To evaluate the validity of the hypothesis that contact experience does not reduce prejudiced attitudes but can increase the profiles of subtle and fanatical teachers, the Mann-Whitney U test was used to compare the scores related to the "contact" variable.
Analyzing the data in Table 6, a significantly lower mean score of subtle prejudice was found in the group of teachers who had Roma students in their class compared to those who did not have this experience.In particular, a significant decrease was observed in the dimension of suppression of positive emotions.Furthermore, a reduction was noted in the dimension of fear and threat within the context of overt prejudice.Although an overall reduction in prejudice score was observed in the presence of contact, there were no significant influences on the rest of the dimensions: "anti-intimacy" of direct prejudice and the two dimensions of subtle prejudice, namely "exaggeration of cultural differences" and "defense of in-group values".Finally, Table 7 shows the percentage distribution of teachers based on prejudice profiles differentiated by contact with Roma students.In the subset of teachers with reported contact, a statistically significant reduction is observed in the prevalence of individuals with a 'Subtle' profile compared to those without contact.Similarly, there is a significant decrease in the percentage of individuals with a "Fanatic" profile among educators with contact in comparison to those without contact.Additionally, it is noteworthy that the prevalence of a "Democratic" profile is higher in the contact group, though this difference is not statistically significant.

Discussion
Regarding the first Hypothesis 1a on the greater prevalence of the subtle component of prejudice in the reference sample, the results confirmed the pre-eminent presence of subtlehidden racism [43,45,46] in institutional contexts such as schools, where open expression of racism is prohibited, and in the social group represented by teachers, where respect for equality norms is more deeply rooted.
This hypothesis is further confirmed by comparing the teachers in this study to a subgroup of graduates analyzed in another study by Arcuri and Boca [43], which revealed an inverse correlation between education level and prejudice.Although the sample of teachers is homogenous in terms of a higher level of education, it shows a higher level of subtle prejudice, while overt prejudice is lower.
The empirical data support Pettigrew and Meertens' theories regarding the diffusion of socially acceptable forms of racism in regulated normative contexts, where there is a greater social desirability represented by expressions of subtle prejudice compared to those of overt prejudice [46].
Although Italian teachers seem to express their racism more subtly within educational settings, their level of direct, manifest prejudice is also important.In fact, when compared with the prejudice levels of a sample of high school students in Spain [52], it appears to be higher.Indeed, an in-depth analysis of data related to Hypothesis 1b, which focused on the "persistence of overt prejudice", affirmed a substantial presence of teachers within the sample who exhibited pronounced overt prejudice.Notably, the dimension of "fear and threat" garnered a higher score compared to other dimensions, indicating a prevalent inclination in the sample to perceive Roma individuals as a threat, aligning with the tenets of naturalist racism.If overt prejudice is mainly based on fear and threat, subtle prejudice is based on cultural differences.The results indicate an oscillation between a historicist approach that perceives cultural differences and defends Italian values and a naturalist approach characterized by overt prejudice primarily rooted in fear of the Roma.These results align with the limited literature on racism against Roma in Italy, which highlights the exceptional position of Roma individuals as a minority seen as both dangerous and reformable [16][17][18][19][20][21][22].
Further results partially contradict Hypothesis 2a, which posits that contact between teachers and Roma students in the classroom does not positively influence attitudes of prejudice due to the asymmetric relationship in the educational context [51].On the contrary, it was found that contact has a positive impact on indirect prejudice, especially at the emotional level.Teachers who have had contact experience show less suppression of positive emotions such as solidarity (subtle prejudice) and exhibit less fear towards the Roma group (Fear and Threat dimension of overt prejudice) compared to colleagues without contact experience in the classroom.
These last two results challenge Allport's theory [51] and confirm research on contact, as reported by Pettigrew and Tropp [48], emphasizing the importance of emotions and the reduction of fear in transforming prejudiced attitudes positively.
However, this experience does not question the presumed superiority of the teachers' own group and culture, which is based on white-centric and Neo-liberal ideas (indirectly measured by the dimension "defense of in-group values" and the dimension of exaggeration of cultural differences).Both dimensions remain unchanged among teachers.The ideas against contact between races in the form of overt prejudice (anti-intimacy dimension), which are typical of colonial-naturalist ideologies and led to segregation policies towards Roma individuals due to their association with nomadism and deviance, also remain unchanged.
Contrary to Hypothesis 2b, which predicted an increase in the rate of subtle and fanatical teachers in the group of teachers who had contact experience with the Roma in the classroom, the results show a decrease in both typologies.
However, no statistically significant differences were found in the proportion of teachers with a "democratic" profile.It has been theorized that this typology is composed of low-bias individuals who challenge entrenched social stereotypes.They are more capable than their "subtle" counterparts of replacing cultural stereotypes with personal beliefs, such as equality [22].
In Pinelli's study, conducted in Lecce among parents whose children participated in an intercultural school project, 79.63% of parents with a university degree or higher were classified as "democrats", and showed low prejudice against immigrants.In contrast, only 46.78% of teachers in the current study fell into the "Democrat" category.This discrepancy could be due to the fact that Pinelli's sample consisted of parents who were highly sensitive to intercultural issues due to their participation in numerous intercultural initiatives [47].
However, Pinelli found that a high level of education does not necessarily go hand in hand with a more democratic attitude.Our findings, in fact, confirm this view, despite the fact that Pinelli focused on the concept of "education" and we deployed the notion of "contact" with Roma students.

Limitations
The investigation conducted in the school context confirms the extreme difficulty of investigating the topic of racism due to social desirability, which, in this case, is exacerbated by the characteristics of the participants combined with the use of an intrusive and directive instrument like a questionnaire [64].All these factors, in combination with the possibility of compromising the teachers' privacy due to the various procedures for questionnaire distribution and collection, led to potential distortions in order to avoid stigma or negative impressions of the school.
Approximately 25% of the sample omitted responses to questions 15 and 23, which pertained to the sexual domain, embracing direct and indirect forms of prejudice, such as sexual contact with a member of the out-group and exacerbation of perceived cultural and sexual practices.Such an omission would require an explanation that is difficult to provide, but it can be interpreted both as a symptom of the limitations of the tool (unclear items, unsuitable questions, etc.) and as a sign of the salience of the questions.In this latter case, the refusal of the interviewees to provide an answer might indicate that these participants felt that their views were inappropriate and might reflect their inner racism and prejudices.
In the sample, which consists of 81% females, there were no significant differences in prejudice between genders across all the scales considered.Although men tended to score higher, the difference was not statistically significant.Given the predominance of females working in the teaching profession in Italy and Sicily and their consequent influence on the education system, it is crucial to examine the impact of gender on racism.
The views of non-Roma women on gender issues in the context of Roma are influenced by "whiteness" because Roma are perceived as belonging to an antiquated, oppressive, male-dominated Roma culture.It is advised that non-Roma women's perspectives on gender-sensitive issues, which were not explored in this article, be further investigated in the local context where this research on racism is carried out.The authors' positions as non-Roma female researchers significantly influenced their research, prompting them to seek insights from international Roma feminist scholars.This aspect of this research will be further explored in a forthcoming publication.
Although the article provides a snapshot of the Italian educational landscape, the survey was conducted solely in one city in Sicily.It would therefore be crucial to extend this research to other Italian regions to compare and contrast attitudes towards the Roma.

Conclusions
The empirical and historical research we carried out confirms that there is a boomerang effect in Italy.
The "barbarization process" described by Césaire in his analysis of colonialism is reflected in the current treatment of Roma in the Italian State.In fact, among Roma pupils, who are mostly Italian citizens, only one in a thousand has access to higher education.As mentioned in the introduction and further elaborated in the sections on data and analysis, Pettigrew and Meerten's prejudice scale was used to empirically discover prejudiced attitudes among teachers, which is indicative of colonial continuities.In this conclusion section, the type of racism measured in a specific context is explicitly discussed through the post-colonial framework, which draws upon colonial history to interpret the empirical data presented in the preceding sections.This interdisciplinary approach helps to reveal the implicit "colonial" structure of racism in Italian schools, which is reminiscent of historical colonial attitudes towards African children.In reference to the concept of "barbarization", the teachers exhibit a moral and cultural decline, notably in the dimension of "fear and threat", which is typical of colonial ideologies and is predominant among teachers.Additionally, racial contact, a crucial aspect of colonial racism, was highlighted.Empirically, it is significant that the dimension of racism based on ideas contrary to contact with the Roma group persists even among teachers who have Roma students in their classes.
Cultural justification for the low educational performance of Roma students relieves Italian educators and politicians of responsibility insofar as they normalize systemic racism against the Roma.
If critical historiography has questioned the notion of "good Italians" and their ostensibly benign colonial rule-especially with regard to school segregation for natives and the neglect of their education, equating these with severe colonial crimes [65,66]-the contemporary myth of the inclusive Italian school should also be scrutinized when examining Roma education.
As discussed above, the Italian school system began educating Roma using segregated systems (Lacho Drom classes), justified by the alleged need to adapt schools to Roma lifestyles.However, this resulted in a traumatic experience for the Roma children, who encountered racism for the first time, specifically within the school system [67].Thus, the supposed academic failure of many Roma can be attributed to European and Italian integration projects rather than to the Roma families themselves.
The "historical tragedy" for Roma families in Italy is that they have had to endure mediated interaction with wider society through segregation in camps.The living conditions in these camps often compel Roma children to identify themselves as "Gypsies", a term associated with racial stereotypes about Roma.The civilizing mission, which had progress as its goal, has proved to be hypocritical.For instance, many Roma families who migrated to Italy during the Kosovo war opted to remain primarily for their children's education [68].In this study, it was observed that these children face daily racial micro-aggressions in interactions and practices [69].
The anthropologist Cipriani instructed colonial administrators, including teachers, to educate African children according to their perceived natural abilities, which were deemed inferior to Italians [70].This attitude persisted, especially within the pedagogical framework established by Karpati, which formed the basis for the Italian Roma scholarization project and the "Lacho Drom" classes.This pedagogy perpetuated prejudices regarding the "primitive mind" of Roma students [20], attitudes that some teachers still endorse today.
It has been uncovered that approximately 22% of teachers who primarily interacted with second and third-generation Roma from Yugoslavia hold the belief, to varying degrees, that Roma students may underperform due to perceived inherent inferior abilities 5 .
One significant finding from this research is the manner in which the dichotomy between racial naturalism and racial historicism, ingrained in structural inequalities and everyday racism, operates within the education system.Despite legal provisions for equal access to education, teachers, as agents, may unintentionally contribute to the normalization of racism through their ambivalent attitudes and behaviors.
In these schools, historicist racial ideas appear to be more prevalent among teachers than naturalist ones.This is evident from the higher mean scores on subtle prejudice items, indicating that teachers perceive Roma individuals as different primarily due to cultural distinctions.These distinctions encompass values instilled in children, religious beliefs, and their manner of speaking Italian.
However, it is important to note that there is no significant impact of direct educational contact with Roma students on the dimension of anti-intimacy prejudice that has always characterized the naturalist racial conception.
The presence of Roma students in educational settings helps reduce perceived threats associated with the out-group and suppresses negative emotions, promoting a more inclusive classroom atmosphere.However, this positive effect may not extend beyond the school environment.When teachers do not feel governed by their institution's anti-discrimination norms and their professional roles, their naturalist racial biases are more likely to surface.This is particularly evident in their preference for avoiding close contact with Roma individuals in personal, marital, or professional situations, as suggested by manifest prejudice scale indicators.Teachers also continue to defend their in-group values, maintaining attitudes of subtle prejudice.
This result confirms the persistence of blatant prejudice concealed beneath a veneer of subtle racism, as reported in other research [52] and as described by Goldberg [1].
It is crucial to recognize the risks associated with not critically examining teachers' ambiguous racial attitudes in educational settings and the subsequent implications for pedagogical practices.These attitudes can manifest as Essed's everyday racism within schools, leading to marginalization, problematization, and containment, as well as mechanisms of pathologization, discouragement, and underestimation of pupils [2].
Moreover, empirical data reveal that contact appears to have no impact on the typology categorized as "democratic-egalitarian".
Comparing the percentage of Democrats in the two samples, among Democratic teachers, the results are indeed lower compared to the parents in Pinelli's research [47].
This finding suggests that the rhetorical commitment of education institutions to democracy may not substantially impact the occurrence of this typology when teachers interact with Roma individuals.It implies that one's democratic stance relies on universal principles that are detached from the specific situation of the Roma in the Italian school system.This result corroborates other research suggesting that European values such as universalism, humanism, egalitarianism, and fairness may be compromised in inter-group relations between European national majorities and the Roma community [6].
Considering that a high level of education is associated with a desire for impartiality, greater control over one's prejudices, and a preference for moderate ways of responding [71], the presence of both subtle and overt attitudes is not surprising.This observation can be interpreted in light of other research conducted with teachers in local schools with Roma students.In these studies, it is claimed that the inclusion of "culturally diverse students" does not require changes in pedagogical programming, planning, implementation, or assessment [15].This situation presents a dichotomy; on the one hand, there is evidence of existing prejudices based primarily on the exaggeration of cultural differences, as revealed by the survey.On the other hand, there is a refusal to adapt teaching practices, which teachers typically justify under the guise of student equity.This hypocritical universalism serves as a mechanism to deny racism [72] among teachers who, according to the theoretical implications of this study, persist in their belief in cultural superiority conferred by their white status.

Suggestions for Combating Prejudice among Teachers
Before presenting practical recommendations, we summarize the overarching conclusions from this study of teachers' prejudicial attitudes toward Roma in two points: 1.In the Italian context, the results reflect a racial climate that places race at the centeran important but often overlooked aspect of educational disparities.This has been highlighted in a different context by critical race theories applied to the American educational context [73].2. In contexts where racism is ostensibly not permitted, the historical coexistence of naturalist and historicist racial conceptions makes it difficult to recognize biases that perpetuate racism (as shown by the refusal to answer direct questions aimed at measuring racism, such as Items 15 and 23).Therefore, an analytical and reductive distinction between these racial ideologies within educational institutions is somehow "beneficial".It raises teachers' levels of awareness and shows the necessity of implementing anti-racism strategies [74].Finally, it advocates for liberatory education [75].
Structural interventions are necessary in order to challenge the long history of socialization into naturalist and historicist racial ideologies, both overt and subtle.These include halting evictions, putting a stop to the system of nomadic camps, and encouraging young Roma to become teachers [26].However, it is equally important to address the various expressions of racism among non-Roma teachers.The use of Pettigrew and Meertens' scales is recommended to identify biases among white individuals in positions of power in sectors such as employment, education, and health.
In light of the recognized ineffectiveness of integration projects for the Roma [9] and the very real prejudices that exist, policymakers are advised to refocus their efforts.If the Roma are the direct recipients of these interventions, but their views are not taken into consideration, they will continue to be perceived as "colonial subjects" who need to be civilized by the dominant group.Without this shift, interventions, including anti-bias training for educators, will remain ineffective.
It is essential to involve both Roma and non-Roma university researchers, professors, and activists so as to give a critical analysis of the "whiteness" in the education system in Italy.The crux of this debate lies not only in accessing initial and ongoing teacher training but also in analyzing the various approaches [76].These should include context-specific strategies rather than standardized methods across nations [6] and should be based on the standpoint of the Roma people.Moreover, an interdisciplinary framework incorporating critical race theory, whiteness studies, and post-colonial studies with social psychology is crucial for examining racism in educational settings.Anti-racism initiatives need to engage with a nuanced understanding of Italian anti-Gypsyism, which is intertwined with the dialectic of racial naturalism and historicism.The goal is to create truly inclusive and self-critical approaches where the voices of Roma students can be heard-especially now when the state school sector, despite its many criticalities, offers secure employment opportunities and attracts many Italians.
Methodologically, anti-racism courses for teachers should be developed in order to reflect on how white privilege is manifested in Italian schools.Once teachers become aware that their position as educators is expressed in the political biases of a culture and thoroughly comprehend the responsibilities they carry with regard to systemic racism against the Roma and other minorities in class, immense possibilities will open for all the social actors in the education system, and in particular, to expose how white superiority has been constructed up to the present.
Based on the interdisciplinary approach outlined in the introduction, the following strategies are suggested to help Italian teachers reflect on the themes of racism based on the types of prejudices detected: a.For "subtle" teachers who conform to existing norms, formally accepting all students while maintaining implicit prejudices, addressing their indirect racism requires raising awareness of their prejudices and stimulating their internal motivation to live according to egalitarian ideals [77].b.For teachers who express direct prejudice against Roma individuals ("Bigots"), it is important to organize experiential work that includes contact experiences.Such experiences have been shown to reduce fear towards the minority group.However, further investigation is needed to understand the persistence of anti-contact beliefs.In these cases, exploring the psychological causes, history, and quality of previous contacts with Roma individuals may provide further insights [78].c.In this research, the democratic typology proves to be problematic compared to other typologies when discussing interactions with the Roma.For teachers who are generally "egalitarian-democratic", the results regarding interracial contact in the classroom emphasize the need for a more engaged approach.This approach should ensure that the internalization of anti-racist norms translates into democratic behaviors.Despite the perceived constraints within the contemporary Neo-liberal education environment [79], a more proactive approach involves challenging what is considered "normal" in school practices.This includes addressing instances of "everyday racism", such as the provision of separate integration showers for Roma students in a school on the outskirts of Milan [41].Being democratic requires action, and teachers can, for example, write letters to address state injustices [80].
The inclusion of Roma history and anti-Gypsyism in teacher training, as proposed by the new National Roma Integration Strategy, cannot, as this study argues, promote democratic action.This is because this integration strategy does not adequately take into account the Italian colonial past and does not recognize the colonial continuities that affect Roma, leading to a state of "frozen anti-racism" [81].This contributes to the normalization and denial of racism among teachers.
Without a true deconstruction of the concept of Italian white innocence [27], it is impossible to empower teachers to empower their students [82].
Racism is a system of dominance, not just an individual act [25].This article suggests that, in order to overcome this social order, a different interdisciplinary framework needs to be developed and applied in the context of contemporary education in Italy.

Table 1 .
Descriptive analysis of the sample.

Table 2 .
Results of the normality test.

Table 3 .
Means and standard deviations for the scales and subscales (N = 295).

Table 4 .
Percentage of teachers scoring above the theoretical median (High Prejudice) (N = 295).

Table 5 .
Percentages of teacher typology in the sample (N = 295 teachers).

Table 6 .
Means and standard deviations of prejudice scale and sub-scale scores, overall and differentiated by contact with Rom students.

Table 7 .
Classification of teachers based on prejudice scales differentiated by contact with Rom students.