Catholic heritage, ethno-racial self-identification, and prejudice against Haitians in the Dominican Republic

ABSTRACT

Latin American cultures are significantly influenced by Catholicism, which partially but meaningfully represents the European legacy that situates Latin American societies into the Western world. This influence may be particularly relevant in the Dominican Republic, where local discourses have underlined the role of Catholicism as an essential characteristic of the Dominican identity. In this study, I use nationally- and regionally-representative survey data to examine whether Catholic self-identification is directly associated with self-identifications that downplay individuals’ own Afro characteristics, and whether individuals who self-identify as Catholic are significantly prejudiced against Haitians. I find regional-level evidence of direct associations between Catholic self-identification and self-identifications that downplay Afro characteristics. I also find national- and regional-level evidence of direct associations between Catholic self-identification and different indicators of prejudice against Haitians. These findings reveal the prevailing influence of Catholicism on ethno-racial self-identification and anti-Haitian sentiment in the Dominican Republic.

KEYWORDS:

• Catholicism
• ethno-racial identities
• prejudice against Haitians
• Dominicanidad and mestizaje
• Dominican Republic
• Latin America

Acknowledgements

I am deeply grateful to Alex Weinreb for sharing the San Benito data used in this study and for his helpful feedback. I also thank Néstor Rodríguez, Daniel Powers, the editors of Ethnic and Racial Studies, and the anonymous reviewers for their comments. Furthermore, I thank the Latin American Public Opinion Project (LAPOP), its major supporters (the United States Agency for International Development, the United Nations Development Program, the Inter-American Development Bank, and Vanderbilt University), Edward Telles, and the Project on Ethnicity and Race in Latin America (PERLA) for making the data available. Earlier versions of this paper were presented at the 2015 Congress of the Latin American Studies Association and the 2015 Population Association of America Annual Meeting.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 Mestizaje (Spanish) means ethno-racial mixture. Mestizaje ideologies in Latin America have disseminated discourses that underscore the positive value derived from racial mixing for incorporating indigenous and Afro-descendant populations in processes of nation-making (see Paredes 2017, 2018).

2 Henceforth, I use the term Afro to refer to the physical and cultural traits of Afro-descendants, as it is used in many Latin American countries.

3 The Concordat with the Vatican established during Trujillo’s regime still benefits the Dominican Catholic Church (e.g., via tax exemptions).

4 See, for instance, Jesuit projects that seek to help immigrants founded on intercultural objectives: Centro Bonó (http://bono.org.do), Centro de Formación y Acción Social y Agraria (http://cefasa.com/), or their collaborative efforts with other not-for-profit Catholic organizations (e.g., Obra de Cooperación Apostólica Seglar Hispanoamericana: http://www.ocasha-ccs.org/).

5 The unification of Hispaniola refers to the forceful annexation of Republic of Spanish Haiti into Haiti.

6 The relationship between Trujillo and the Catholic Church began to deteriorate by the end of the 1950s. After the regime, the political mediation of the Catholic hierarchy contributed to the legitimation of subsequent authoritarian governments. Nevertheless, a smaller section of the Dominican Catholic Church established a new alternative orientation founded on principles of social justice (Betances 2009).

7 This use contrasts with the use of indio in other indo-Latin American countries, which refers pejoratively to indigenous populations (Paredes 2018).

8 Racial projects highlight the relevance of ideologies as instruments of domination. Accordingly, Dominicans are influenced by mainstream Dominicanidad shaped as a racial project. This influence should not be interpreted as Dominicans’ denial of their “true” Afro identities (Wheeler 2015). Dominicans also have developed alternative discourses that counter mainstream Dominicanidad in different ways (Ricourt 2016; Simmons 2005).

9 I also fitted multinomial regression models with a nominal dependent variable with four categories: white, indio/mestizo, mulato, and black/Afro-Dominican. Based on these analyses, I decided to use a binary dependent variable as the optimal choice for this study. I also translated negro as black in order to connect the Afro-Dominican condition with the analytic notion of blackness. Nevertheless, it is necessary to mention that negro, as in other mestizaje contexts, can be interpreted in different ways (García-Peña 2015).

10 In Spanish: “¿Hasta qué punto está de acuerdo con que los hijos de inmigrantes haitianos nacidos en la República Dominicana sean ciudadanos dominicanos?,” and “¿Hasta qué punto está de acuerdo o desacuerdo con que el gobierno dominicano otorgue permisos de trabajo a los haitianos indocumentados que viven en República Dominicana?”

11 I fitted ordered logistic regression models to predict prejudice against the citizenship of Dominican Republic-born children of Haitian immigrants, and prejudice against work permits for undocumented Haitians who live in the Dominican Republic. These models, however, did not satisfy the proportional odds assumption even after I collapsed the number of categories in different ways. Therefore, I dichotomized these variables.

12 I translated the questions from Spanish to English. “En general, su opinión sobre los haitianos es muy negativa, negativa, ni negativa ni positiva, positiva o muy positiva” “Si es posible, ¿los evita? (a los haitianos)”.

13 See http://perla.princeton.edu/surveys/perla-color-palette/. I recoded this variable using zero for the lightest and grouped categories 8 and over for the darkest skin tone.

14 I did not include household income in the analyses of self-identification because its coefficient was statistically insignificant. The results presented in these analyses were not different after income was included as an independent variable.

Additional information

Funding

The San Benito survey was funded by a grant awarded by the National Institute of Child Health and Human Development [1R21HD054731-01A1] to Mariano Sana (P.I.) and co-P.I.s Alexander Weinreb and Guy Stecklov.