Immigration theory between assimilation and discrimination

ABSTRACT

While most immigration studies traditionally build on assimilation theory, references to discrimination are increasingly present in the field, usually in an oppositional way. This article attempts to rethink the coexistence of these two concepts in the immigration scholarship and analyse their relationship. I first review the analytical perspectives of assimilation theory and argue that, unlike assimilation, discrimination is not a migration-specific concept. I distinguish between three forms of discrimination facing the immigrant population: civic discrimination, discrimination against migratory attributes and ethnoracial discrimination. The empirical section uses self-reported measures of discrimination from recent large-scale French data to assess each of these three forms of discrimination and analyse their determinants. The results show evidence relative to each of these three forms and point to discrepancies with assimilation. In particular, assimilation-facilitating characteristics such as early childhood knowledge of French and education in France are associated with higher levels of perceived discrimination. Perceptions of skin colour and religious discrimination also appear to remain stable if not increase over immigrant generations from non-European origins. The conclusion discusses some avenues to bridge the gap between intergenerational perspectives of immigrant socioeconomic attainment and the more general scholarship on the reproduction of inequality.

KEYWORDS:

• Immigration
• assimilation
• discrimination
• France

Acknowledgements

The author would like to thank Haley McAvay and Ettore Recchi for their insightful comments on this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 ‘Assimilation is a process of interpretation and fusion in which persons and groups acquire the memories, sentiments, and attitudes of other persons and groups, and, by sharing their experience and history, are incorporated with them in a common cultural life’ (Park and Burgess 1921, 735).

2 The first most cited article in IMR is Levitt Schiller, Conceptualizing simultaneity: a transnational social field perspective on society’ (1272 citations according to Web of Science in July 2022, while Alba’s has 1074 citations).

3 https://fra.europa.eu/en/eu-charter/article/21-non-discrimination#:~:text=1.,sexual%20orientation%20shall%20be%20prohibited.

4 https://www.legifrance.gouv.fr/codes/section_lc/LEGITEXT000006070719/LEGISCTA000006165298/

5 Debate about the relevance of the distinction between ethnicity, race and nationality are recurrent (Brubaker 2009; Cornell and Hartmann 1998; Omi and Winant 1994; Wimmer 2013). I choose to use the expression ‘ethnoracial’ in a broad sense in reference to the varieties of markers that are at stake in symbolic categorisation of humans that might entail unequal treatment and social inequality. In the empirical section, I focus more specifically on skin colour, national origin and religious denomination on which I have some data.

6 In the sense that we tend to use the immigrant status as an identity category while it is not, namely because immigrants do not have the same national origin and because citizenship, race, ethnicity, religion and many other identity categories are powerful sorting factors within the immigration population. To use the words of a famous French activist, there is no such thing as Migristan (Herrou 2020).

7 There is a wide literature on attitudes toward immigration focusing on the degree to which people accept to receive inflows of migrants in destination countries and how they react to immigration policy. This literature, which is prolific in social psychology, sometimes refers to anti-immigrant prejudice and even discrimination (Esses 2021; Hainmueller and Hopkins 2014). In this article, I use discrimination as an interactional concept that entails unequal treatments on the grounds of individual characteristics pertaining to immigrant status.

8 Regardless of religiosity (which entails commitment to a belief system, faith, religious practices, etc.), religious denomination may be regarded as an ethnoracial marker in the sense it relates to socially distinctive identities. There is consistent evidence in France and more generally in Europe that (real or supposed) belonging to the Muslim religion is involved in meaningful categorisation (and consequently unequal treatment) regardless of religiosity (Adida, Laitin, and Valfort 2016; Stasio et al. 2021; Valfort 2020).

9 The data is a (cross-sectional) update of TeO1 (2008) that has been extensively used in migration studies over the last decade (Beauchemin, Hamel, and Simon 2010b). For more details on the survey and questionnaire (see Beauchemin et al. 2023) and for the first descriptive findings (see Beauchemin, Ichou, and Simon 2022; Lê et al. 2022).

10 La Guadeloupe, La Martinique, and La Réunion are the regions of origin of by far the largest numbers of overseas migrants in metropolitan France. More details about these categories of population in France can be found in (Beauchemin and Safi 2020).

11 For all categorical variables that have considerable numbers of missing values, I decided to include a missing variable category instead of eliminating the missing values from the sample. While this decision does not affect the findings, it maximises the number of observations used in the estimation sample as the models control for a wide variety of variables. Moreover, this strategy circumvents bias related to self-selection into missing values.

12 These categories’ coefficients in the regression on the number of discrimination items are qualitatively similar but less consistently significant.

13 The full results in Table A4 show that in France, all religious denominations tend to report significantly higher levels of religious discrimination in comparison with the ‘secularized’ group. Nevertheless, the coefficients for the Muslim denomination remain significantly higher than those for the Christian or other denominations.

14 G3 is grouped with the majority in a category label G3+. This is in fact the case in the overwhelming majority of immigration studies since they rarely collect data about grand-parental origin. If we include them in the estimation sample, the results remain the same (analyses not shown for the sake of concision).

15 Interestingly, while women tend to report higher levels of discrimination overall, the detailed findings in table A4 show that this is only the case for religious discrimination. Women actually declared less discrimination based on skin colour or origin. This finding points to the gendered dimension of these religious boundary-making dynamics, which draw on the stigmatisation of Muslim women in relation to the veil and Western norms of gender equality.