Preference and prejudice: Does intermarriage erode negative ethno-racial attitudes between groups in Spain?

Introduction

Intermarriage or mixed unions (i.e. unions between individuals from different racial or ethnocultural groups) have traditionally been considered a key indicator of integration and have been equated with the notion of a society without differences. Mixed unions have been thought to imply the permeability of social boundaries, the lessening of social distance, and the absence or weakening of fixed ethnic identities, negative attitudes, and racial or cultural stereotypes and prejudices (Gordon, 1964; Kalmijn, 1998; Pagnini and Morgan, 1990). Moreover, mixed unions are often conceptualised as constituting a bridge between different cultures – a space where the differences and identities related to ethnic origin, class, and gender intersect and are contested, which, in turn, usually bring about positive processes of both social integration and interethnic understanding (Barbara, 1985; Bystydzienski, 2011; Rodríguez-García, 2006; Varro, 2003; Vucinic-Nescovic, 2002). Furthermore, while endogamy still prevails, even in online dating (Feliciano et al., 2011; Potârcă and Mills, 2015), the ever-growing increase in intermarriage across national, cultural, racial, and religious boundaries worldwide (Heikkilä and Yeoh, 2010; Rodríguez-García, 2015) can be seen as a sign of the diminishing barriers to social interaction across groups – and as an important step in the lessening of ethno-racial distinctions in forthcoming generations (Alba and Nee, 2003).

But is intermarriage the ultimate boundary breaker? To date, several studies point out the actual complexity of the link between intermarriage and assimilation (Collet and Santelli, 2012; Marcson, 1950; Miguel-Luken et al., 2015; Rodríguez-García, 2006, 2012, 2015; Rodríguez-García et al., 2015, 2014a; Song, 2009). The idea of intermarriage as a panacea for social harmony and cohesion – like the idealised vision presented in a Benetton advertisement – might be too simplistic.

First, intermarriage does not imply the renunciation or reconciliation of the two partners' differences or beliefs, which can result in conflict. In fact, a number of studies show that marriage dissolution rates are higher among the intermarried than among endogamous couples, owing to socio-economic disparities (Goldstein and Harknett, 2006) as well as cultural dissimilarity (Kalmijn et al., 2005; Smith et al., 2012). Social discrimination and lack of support from third parties, such as friends and family, and, in particular, rejection of the intermarriage by parents are also factors explaining intermarriage conflicts and dissolutions (Bratter and Eschbach, 2006; Bratter and King, 2008; Cerroni-Long, 1984; Hohmann-Marriot and Amato, 2008; Milewski and Kulu, 2014; Moscato et al., 2014).

Second, intermarriage and racially mixed individuals do not necessarily lessen the significance or salience of ethno-racial boundaries, inequality, and discrimination, as Daniel (2003) has argued for the case of South Africa and Brazil, and Okamura (2010) has for the case of mixed-race Hawaiians. Other recent empirical studies conducted in Spain have reported similar findings (Rodríguez-García et al., 2014b, 2015; Rodríguez-García and Miguel-Luken, 2015). The classic work of Roger Bastide (1961) had already noted that intermarriage does not presuppose the erosion or absence of prejudice towards other groups, especially in highly ethnically segmented social contexts. In fact, it could be argued that the gender imbalances frequently observed in the composition of mixed unions are often the product of the exoticisation of certain groups (e.g. Latin American and Asian women, black men). Such stereotypes and constructs, often rooted in a colonialist or imperialist past, might foster higher rates of couplings between, for example, white men and East Asian women, who have traditionally been portrayed in western societies as feminine, exotic, docile, and seductive (Lee, 2004; Lee, 2015; Nemoto, 2006), or between white women and black men, who have traditionally been portrayed as exotic and hypersexual (Fanon, 1967; Kempadoo, 2004). In this respect, mixed marriages can also reflect socio-ethnically stratified social structures (Rodríguez-García, 2007; Rytina et al., 1988). Overall, the substantial rise in intermarriage globally has not abolished negative attitudes towards intermarriage itself or towards multiracialism (Aspinall and Song, 2013; Bratter and Eschbach, 2006; Daniel, 2003; Herman and Campbell, 2012; Okamura, 2010; Rodríguez-García, 2015; Wang, 2012).

Previous studies have focused on attitudes towards interracial and international dating and on preferences and choices of marriage partners, predominantly in the United States (e.g. Dunleavy, 2004; Everts et al., 2005; Goldstein and Harknett, 2006; Herman and Campbell, 2012; Johnson and Jacobson, 2005), and, to a lesser extent, in Europe (Kalmijn and van Tubergen, 2006; Moscato et al., 2014). This paper will present selected findings from a larger research project ¹ on the nexus between mixed unions and sociocultural integration in Catalonia, Spain – a fertile research context that will be explained in the subsequent section. First, we analyse prejudices and prescribed preferences that the couples experience from their respective environments (emanating from the families of both the Spanish partner and the immigrant partner, and from the wider society). Second, we analyse biases and mate preferences verbalised by the members of mixed unions themselves towards their own communities, the Spanish community, or other groups.

In our study, we refer to ‘intermarriage’ or ‘mixed unions’ as marriages and common-law relationships formed by a foreign-born person and a Spanish-born person. These unions are also often mixed in terms of ethnocultural, racial, or religious differences. We have limited our sample to heterosexual couples. Although intermarriage is a well-established concept in the literature, we prefer to think of these unions in terms of ‘mixedness’, a more encompassing concept that refers not only to the couples, families, and individuals involved but also to the sociocultural processes of ‘mixing’; ‘mixedness’ not only describes the combination of national, racial, cultural, or religious differences but also signifies an active space that disturbs and contests social norms (Rodríguez-García, 2015: 3–4). And regarding ‘integration’, we understand this to mean individual and groups of newcomers or minority groups becoming full and equal members of the society in which they live, a process that encompasses legal–political, socio-economic, and cultural dimensions (Penninx and Garcés-Mascareñas, 2015). In previous publications, we have contested the direct relationship between intermarriage and other aspects of sociocultural integration, such as social embeddedness, identification with the society of destination, or second language acquisition (Miguel-Luken et al., 2015; Rodríguez-García, 2015; Rodríguez-García et al., 2015, 2014a). On this occasion, we focus on the sociocultural domain of perceptions, categorisations, and practices related to difference and diversity – or what Gordon (1964: 71) called ‘attitude’ and ‘behavioural receptional assimilation’; this area of investigation looks at the permeability of social boundaries and the absence or weakening of discriminatory practices and attitudes (including perceived discrimination) towards immigrants and minority groups.

Methodology

The data for this study were based on in-depth interviews conducted in Catalonia in 2013. The sample consisted of 58 immigrants in exogamous unions (heterosexual marriages and common-law relationships formed by a foreign-born immigrant and a Spanish-born individual) and 36 immigrants in endogamous unions (relationships where both partners were immigrants born in the same country), the latter of which served as a control group. The groups were matched in terms of relevant background characteristics, such as social class. For the purposes of this article, we will focus only on the exogamous group.

Our quota sample was further stratified by country of origin and gender. We selected immigrants from seven very diverse countries of origin: Morocco, Romania, Ecuador, Dominican Republic, Senegal, China, and Pakistan – all of which are key sending countries. Previous research on mixed unions has tended to focus on unions formed by the majority population and a single country of origin and/or ethno-racial group, predominantly black/white pairings and, to a lesser extent, Asian/white populations. Including a larger variety of countries of origin in a single study helps us to understand commonalities among immigrants and country-specific variations of stereotypes and prejudices that are experienced by groups.

Respondents were selected using personal connections, advertisements, and contacts with different migrant associations in different parts of Catalonia. In order to avoid the bias produced by snowball sampling, we instead sought respondents who were not connected among themselves, as we were interested in finding a variety of experiences.

After our participants filled out a form with basic socio-demographic information, we held semi-structured interviews with our respondents. These interviews included questions related to a variety of themes (e.g. migratory trajectory and context in which they met their partner, daily life, customs, socialising tendencies, identity, mate preferences, prejudices, discrimination, intergenerational dynamics, and so forth).

The interviews were held in either Catalan or Spanish, often in the respondents' homes, and took on average 2.5 h. All the interviews were recorded in their entirety, fully transcribed, and analysed with Atlas.ti software (v.7) for qualitative data analysis. An information sheet and a consent form were given to participants prior to conducting the interview, and anonymity was guaranteed.

Attitudes, preferences, and choices: External constraints to intermarriage

The family against intermarriage: ‘What have I done for you to marry a Nazarani?’

My mother did not like it [my marrying a Spanish woman] (…) In fact, I have a sister in the Netherlands who married a Dutchman recently [in 2010] (…), and my mother and my father did not go [to the wedding]; they didn't want to be involved at all. My mother says, ‘Are you punishing me? What have I done to make you marry Nazaranis?’ In Arabic, it is called Nazarani, from Nazareth (…) It's like an insult, like when you say ‘Moor’ here. (M1M, Moroccan man, 42 years old)

M1M is a Moroccan man who had been married to a Spanish woman for 15 years at the time of the interview. His parents live in the Netherlands, where they migrated during his youth. He decided to move to Spain, where he had friends, partially because he did not have a good relationship with his parents. He met his wife at a party. His mother did not like that he started seeing a Spanish, non-Muslim woman. His sister, who still lives in the Netherlands, similarly married a non-Muslim (Dutch) man in 2010, and the interviewee's parents did not attend his sister's wedding either. His mother used the term ‘Nazarani’ to refer to her son's and her daughter's respective partners, a derogatory way of referring to Christians, in much the same way that the term ‘moros’ (Moors) is used in Spain to refer to Moroccans and, by extension, Muslims/Arabs. The respondent's wife has not converted to Islam, nor has he ever even contemplated it, since he defines himself as a person with ‘no religion’. This case effectively illustrates the limits and barriers that families may impose on a union between two people who not only have different origins but also (and this seems to be fundamental) different religious backgrounds. This interview excerpt is also significant because it shows that negative familial attitudes expressed towards intermarriage may exist in both directions, from the ethnic minority group (within the migration context) towards the majority group and vice versa.

This section of the paper responds to the following questions: How do families react to the formation of a mixed couple? What are the main factors, if any, associated with family rejection? And do attitudes change over time? We will first concentrate on the families of the Spanish partner and then turn our attention to the families of the respondents themselves. Lastly, we will describe how rejection from family can sometimes change over time.

Of the 58 exogamous cases interviewed in this study, almost a third (17 cases) of the respondents made references to the objections that the family of the Spanish partner initially had or continued to have, and nine cases reported rejection from their own family. Six couples struggled with rejection from both families. These figures cannot be regarded as representative, as our cases were not randomly selected nor do we have a sufficient number of cases to be able to make statistical inferences. Nevertheless, it is meaningful that a third of our respondents (20 out of 58) reported opposition from one or both families. Our Appendix 1 summarises some of the characteristics (origin, gender, religion) of these 20 cases.

Opposition from the family of the Spanish partner

About two-thirds of the immigrant respondents had not experienced any marked rejections from the Spanish family. However, as indicated before, 17 respondents expressed that their partner's family had objected to their marriage/partnership or had posed barriers in one way or another. This ranged from criticising the attitudes or habits of the immigrant partner, to instilling a sense of guilt in both partners of the mixed union, to displaying stronger forms of rejection, such as not coming to the wedding or cutting off ties completely. Our findings show that the familial refusal to accept the foreign partner stems from both cultural and physical factors, which are often entwined. As Cea d’Ancona and Valles (2010: 292–293) have written, ‘In Spain, ethnic racism is … overshadowed by cultural racism, at least in the surveys of attitudes to immigration. This is not so in qualitative studies, where references to skin colour are a constant in the expression of -philias and phobias…, prevail[ing] over country of origin’.

Table 1 shows that the two groups that most often reported rejection from Spanish families are Dominican women and Muslim men from predominantly Islamic countries.

OPEN IN VIEWER

Table 1.

Respondents who experienced rejection from the family of the Spanish partner.

Origin	Female	Male	Total
Chinese	0 out of 5	0 out of 3	0 out of 8
Dominican	4 out of 5	1 out of 5	5 out of 10
Ecuadorian	0 out of 3	0 out of 3	0 out of 6
Moroccan	1 out of 5	4 out of 5	5 out of 10
Pakistani	–	2 out of 3	2 out of 3
Romanian	0 out of 8	0 out of 2	0 out of 10
Senegalese	1 out of 4	4 out of 7	5 out of 11
Total	6 out of 30	11 out of 28	17 out of 58

Source: Authors' compilation from our own data.

Notably, the four Dominican women who reported rejection from the families of their partners are all phenotypically distinguishable from the majority of Spanish society. One of the Ecuadorian women who was not phenotypically distinguishable as a ‘foreigner’ reported that she herself had not experienced rejection but she also commented on the importance that racial appearance has in the Spanish context:

He [my partner] has told me that some people in his family had a bit of suspicion before they met me (…) I think it was his aunt who asked, ‘Where is she from?’ And she didn't like it. And then she saw a photo and said [now in a positive tone], ‘Oh! But she is light-skinned!’ (E1M, Ecuadorian woman, 36)

Furthermore, in the Spanish imagination, but also globally (Kempadoo, 2004), Dominican women, and Caribbean women in general, are associated with exoticism, hyperfemininity, and hypersexuality (Roca and Urmeneta, 2013), and with infidelity. For example, the following interviewee claims that she represented to her mother-in-law the stereotypical Dominican woman, overtly sexual and with a low level of culture, with whom the son should never have married. At the time of the interview, the interviewee was in the process of divorce, and she blamed the breakup to a large degree on the continual pressure and verbal attacks the couple had received from her mother-in-law, to which she considered that her ex-husband had not responded adequately:

His mother interfered too much; she made me feel bad (…), as if I was illiterate: ‘And in your country, do you have this? Are there apples in your country?’ I thought to myself, Does she think that I come from the jungle? (…) If I gave my opinions, it was like, ‘Well, what do they know? If it was not for us, they would still be walking around in a loincloth.’ (D2M, Dominican woman, 32)

The other group of people in our study that experienced rejection from the families of their Spanish partners were Muslim men: Moroccans, Pakistanis, and Senegalese, both practising and non-practising. These ethnicities are especially minoritised groups within Spanish society as they have a different religious background; they come from relatively poor countries; and they are often phenotypically distinguishable within Spanish society. The Muslim population from the Maghreb (chiefly from Morocco) has particularly been historically stigmatised in Spain as perpetual foreigners (‘the Moors’) since the time of the Crusades (Cebolla and Requena, 2009), then further discredited globally by the stigma of fundamentalism since the 9/11 attacks in the United States (Gottschalk and Greenberg, 2008). Notably, a number of studies in Spain (e.g. Arjona and Checa, 2014; Rodríguez-García et al., 2015; Rodríguez-García and Miguel-Luken, 2015) find that immigrants from the Maghreb are the ones that people express the strongest rejection towards intermarrying. We would argue that this rejection has a component based on physical traits: the ‘non-white’ phenotype of many North Africans, certain ways of dressing (e.g. donning the hijab), and ‘foreign’ accents become signifiers of being a Muslim and elicit social distancing and discriminatory reactions from both the mainstream society and other non-Muslim immigrant groups. ‘Looking Muslim’ therefore becomes a low-status racial attribute in the Spanish context, as this appearance is associated with negative stereotypes and prejudices (see Rodríguez-García, 2015).

Based on our study results, we would add gender as an important intersectional factor as well, as the rejection of Muslims as desirable marriage partners (for one's children, in this case) is often more directed towards men. The following interview excerpt from a Senegalese/Spanish couple illustrates the intersectionality of factors that contributes to familial rejection of the immigrant partner (at the time of the interview, this couple had recently become parents, but the family had refused to come to see the child):

R2 (Spanish partner): They [my family] were radically opposed. The beginning was very hard, because at home they did not even want to meet him, and in fact they didn't. I told them, ‘Don't just look at the surface; get to know him and then judge!’ But no… since we started living together, I have had no relationship with my family. They made me choose, and I chose him.

I (interviewer): And now your parents have a granddaughter…

R1 (respondent): Yes, they [my in-laws] know, but they have not had any reaction. And sometimes we all bump into each other in the street…

I: And they don't say hi or anything?

R2: Nothing, nothing. Because I left home with nothing: no money, no suitcase, only with what I was wearing. We were a normal family, but they could not cope with it; all this went beyond them, because he was black, a Muslim, came [to Spain] in a small boat, has no education… (S12M, Senegalese man, 34, and his Spanish partner)

Religion, more specifically the possibility that a son or daughter starts a relationship with a person of a different religious affiliation, is indeed a key factor in the way the family and wider society react overall to the formation of the mixed couple (Arweck and Nesbitt, 2010). This reality was highlighted in various interviews. For example, the following excerpt comes from a Spanish Catholic woman married to a Senegalese Muslim man. With regard to the family reaction to their relationship, she emphasises that what was most difficult for her family was the religious difference: ‘Well, it was difficult for my parents (…) Mostly the fear of him being a Muslim. I think it scared them a little, yes. Actually, what we hear about the Muslim world here is usually not very good…’ (S1M, Spanish partner of a Senegalese man, 35).

In some cases, the problem of acceptance by the Spanish family does not only concern the union of the son or daughter with a Muslim partner, but also – and primarily – their child's conversion to Islam. For example, a Senegalese woman who defines herself as a conservative and very religious person who grew up in an equally conservative family, both in a cultural and religious sense, indicated that for her, a prerequisite for marriage was that her partner converts to Islam. Her partner's family had great difficulties accepting his conversion.

[Referring to her father-in-law's negative reaction when he heard the news of his son's conversion to Islam] It's normal, it's as if I were to tell my father that I was going to leave my religion. That's hard. (S13M, Senegalese woman, 33)

With respect to the aforementioned groups (both Dominican women and Moroccan, Pakistani, and Senegalese men), there is also often the suspicion of so-called marriages of convenience, contracted for the sole purpose of obtaining quicker access to Spanish nationality and improving living conditions in Spain. Apprehension from in-laws because of this concern is exemplified in the following interview excerpt:

Her mother and father opposed; they said, ‘This foreigner… bad people, he wants to marry for papers [meaning nationality] and then go away; why marry him?’ She said, ‘He is a very good person; I want to marry him.’ I said, ‘Look, I do not care about the papers…’ (P4M, Pakistani man, 56)

In the cases described so far, regarding both Dominican women and men from predominantly Islamic countries, family rejection is linked with an intersection of origin, phenotype, and sometimes religion, but also with social class factors. This prejudice is very clear in the case we described earlier of the Senegalese man (S12M) who, as described by his wife, is ‘black, a Muslim, came in a small boat, and has no education’. In Spain, the ‘invasion’ of illegal African migrants arriving on the shores of the Mediterranean, fleeing poverty and looking for a better life, receives much media attention. These immigrants are depicted as poor and illiterate. Similarly, in the aforementioned excerpt from the Dominican woman (D2M), social class was intertwined with the historical position of coloniser/colonised (the Dominican Republic was one of the first territories colonised by Spain). The mother-in-law's reaction to her Dominican daughter-in-law insinuated the former's belief of the sociocultural ‘primitiveness’ of the previously colonised native Dominican population. However, even an objectively higher social position does not always eliminate the prejudice associated with origin and phenotype, as is illustrated by the case of another Dominican woman. She is a university professor and her husband is a dentist, and she has a salary equivalent to that of her husband. The couple has a high social status. Nevertheless, she is still not accepted by her in-laws:

His mother has never accepted me for being Dominican. We never see each other. As for the relationship with [our daughters], their grandparents have never acted as their family, or she as a grandmother to them; they don't love them. (D8M, Dominican woman, 50)

Objections from the family of the immigrant partner

Until now, we have looked at opposition to intermarriage from the native-born partner's family towards the immigrant partner. But as we indicated before, respondents sometimes told us that their own families also objected in one way or another to the mixed union. This outcome was less often the case than rejection from the Spanish families, and many respondents told us that their parents accepted the partner and just wanted them to be happy. Again, when rejection by family occurred, it could take many forms, from a cold reception and distancing themselves, to trying to set up their son or daughter with an alternative partner from their own community, to stronger acts of rejection that had a deeper impact on the couple's lives. Table 2 shows that the participants who reported rejection from their own families were mostly Moroccans, Pakistanis, and Senegalese (eight of the nine cases reporting rejection).

OPEN IN VIEWER

Table 2.

Respondents who experienced rejection from their own family.

Origin	Female	Male	Total
Chinese	1 out of 5	0 out of 3	1 out of 8
Dominican	0 out of 5	0 out of 5	0 out of 10
Ecuadorian	0 out of 3	0 out of 3	0 out of 6
Moroccan	2 out of 5	1 out of 5	3 out of 10
Pakistani	–	2 out of 3	2 out of 3
Romanian	0 out of 8	0 out of 2	0 out of 10
Senegalese	1 out of 4	2 out of 7	3 out of 11
Total	4 out of 30	5 out of 28	9 out of 58

Source: Authors' compilation from our own data.

Again, differences in religion appear to be the dominant factor in the discourses of rejection expressed by the respondents' families. In five cases, the respondents had lapsed faith (they considered themselves to be non-religious or non-practising); in two cases, the respondents self-identified as Muslims, but the partner had not converted to Islam; and in one other case, the partner had converted to Islam, but the family still expressed rejection. A Senegalese man stated the following:

There are many mothers there [referring to Senegal] who do not want to see their child with a woman who doesn't have the same religion. Because they say that their child can change… that they will do something that the mother does not like, or will forget their family. (S1M, Senegalese man, 35)

Although more of the Muslim respondents reporting rejection were males, the strongest rejections were experienced by two Moroccan women and a Moroccan man, who consider themselves non-religious or non-practising. The two women, M2M and M3M (see the latter's interview extract below), reported that they had experienced strong family pressure (especially from the men of the family) not to ‘abandon’ the community and to follow the moral and religious obligations to be met by Muslim women. Notably, the two women belong to the Amazigh (non-Arab) ethnic group; we did not observe a similar rejection among the Arab Muslim women from Moroccan origin. As well, both women had migrated with their families to Catalonia during their youth, so the family, because of its proximity, could have had greater control over their lives.

In both cases, the wider community was the source of family pressure. Respondents mentioned that when there were fewer Moroccans residing in Spain, they had more freedom to go out and dress as they preferred (this was especially stressed in the case of women), but when Moroccans became a bigger group locally, the respondents started to experience greater social control. The following interview extract is from one of the above Moroccan women, who maintained a five-year relationship with her current Spanish husband before they got married. It was a relationship carried out with much discretion, and her family was never aware of it. At first, she thought that her partner would just have to convert to Islam to be accepted by her family. But once her mother found out about the relationship, her mother told her that she would never accept the Spanish spouse, no matter what. From the interviewee's words, it is clear that the mother's outlook was not so much a matter of expressing family rejection, but rather it stemmed from a fear of the collective pressure should the daughter marry a Spaniard. The respondent explains:

We talk very little with my mother. She told me something that I still remember: ‘You're the only girl I have, and the only one who breaks with the customs and rules is you. I want you to be happy (…) I want you to marry a man who treats you like a princess, a queen.’ She would have been happy that I married [my partner], who cares about and loves me. But she just cannot accept it [the relationship with a Spaniard] publically, within the Muslim community, even if [my partner] converted to Islam. (M3M, Moroccan woman, 38)

Notably, this respondent's father spent two years without speaking to her, until her daughters were born. At the time of the interview, she and her husband still had quite a tense relationship with her family.

Evolution of family opposition over time

In order to better assess whether or not mixed unions erode ethno-racial boundaries and negative attitudes towards other groups, we must take the factor of time into account: that is, whether initial negative attitudes and familial rejection change with the passing of time.

In cases in which there was an initially strong rejection, this lack of acceptance typically remained strong over time. As the Moroccan man (M1M) cited earlier (on pages 5-6) indicated, more than 10 years after his marriage to a Spanish woman, his sister also married a non-Muslim partner and the attitude of his parents was similar to the attitude they had in his case. In some cases, the negative attitudes gradually softened to some degree regarding prejudices and stereotypes, but a certain distance continued to exist towards the ‘foreign’ partner. In these cases, it seemed as though the family showed a certain resignation to the situation, even though they still did not completely agree with it.

On the other hand, cases in which the rejection was initially milder, or less visceral, tended to demonstrate what we could call a ‘transformation of perception’: that is, the prejudices decreased with time, especially when the families got to know the partner better and associate with him or her on a frequent basis. In some cases, the initial barriers that had arisen from fear and distrust completely disappeared. As put by the following male interviewee:

Her mother was a little frightened at the beginning, but then, that was that. As I’ve told you, what is most important is day-to-day ordinary life – that is where you forge the relationship. (M6M, Moroccan man, 39)

Moreover, as shown in previous studies (e.g. Van Zantvliet et al., 2014), the rejection of, and negative pressure placed on, interethnic partnering by parents increases when their children, particularly their daughters, move from informal dating to formalised marriage, with a higher possibility of having children. Some of our respondents also referred to this dynamic. But once families realised that the couple was serious about their relationship, they sometimes altered their attitudes.

Of all the mixed couples in our study, a certain type of transformation was verbalised in 12 cases (see Appendix 1), indicating that family opposition to mixing sometimes decreases over time. It should also be noted that, in quite a few cases of family rejection, tensions or conflicts with parents already existed before the partner entered the respondent's life, and respondents were already more used to resisting familial pressure.

Prejudices from within: ‘I am not a racist, but I would never be with a black, black, black partner’

When asked whether cultural differences had been an obstacle in the relationship with the partner, many respondents indicated that cultural differences had, indeed, caused misunderstandings, differences of opinion, or arguments, but many also said that the important thing was to have mutual respect and patience and to communicate about such issues. Of course, we focused mostly on individuals who were currently in a mixed union, and the picture may have been different had we focused on individuals who had divorced or separated from their Spanish partners. In general, however, we observed that people in mixed unions became culturally more flexible, developing a better awareness of the more subtle cultural differences between Spaniards and their own community.

Interestingly, we also asked whether the respondents had certain preferences and/or self-imposed barriers in terms of partnering. At first, many respondents did not comment on specific preferences for a certain type of partner according to origin, phenotype, culture, or religion. They often stated that they simply had fallen in love with someone, and that they had never really thought about whom they might prefer or with whom they would not couple. In fact, in the interviews, we felt that the question took respondents both in mixed and endogamous couples by surprise. The idea that 'love knows no bounds' was recurring:

I didn't say, ‘Look, I'll fall in love with a Spaniard,’ or ‘I'll fall in love with a French woman,’ because the heart wants what it wants, and I instantly fell for her. (D3M, Dominican man, 37)

In love, there is no age, no skin, no religion, no country, no anything. No continent. Love is love. Whether [the person] is Indian or Chinese, whatever. (S13M, Senegalese woman, 33)

Yet, when we probed a little further, asking about specific populations (e.g. ‘So you also could have married, say, a Chinese or Moroccan man?’), a fair number of the intermarried interviewees did reveal ethno-racial prejudices (particularly when referring to marriage preferences for themselves and for their children) and discriminatory attitudes towards other immigrant minority groups and also towards their own culture. It is striking how respondents in mixed couples recurrently used constructions of the following type: ‘South Americans are like this…’, ‘I do not like latinas because they always…’, or ‘all Chinese are…’ The interview extract below is a clear example of the prejudices that we sometimes encountered among respondents:

I'm not a racist, but I've never been with a black, black, black person. I have black friends. They are good people – I have nothing against them – but I would never be with a black partner (…) I do not like their colour, their hair… The Chinese are in the same group for me as blacks (…) To see my daughter partnering with a Moroccan, Pakistani, Arab, Russian, or one of those strange things, I would not approve. European, yes. (D9M, Dominican woman, 45) ²

Figures 1 and 2, using ideas/categories verbalised in the interviews, present the main reasons for respondents’ rejection of a hypothetical mixed union with certain ethno-racial groups. The figures, though revealing prevalent stereotypes, should by no means be regarded as representative of all the barriers and prejudices that one community has with respect to members of its own community or other communities. We have differentiated between prejudices verbalised by our exogamous female respondents regarding men of their own group and from other communities (Figure 1) and prejudices that our exogamous male respondents expressed regarding women of their own background and from other communities (Figure 2), with women clearly referring to a greater number of ‘barriers’ for interethnic partnering. Between both sexes, the three most recurrent themes were gender inequality in the relationship (machismo), religious differences (Islam), and (hyper)sexual/promiscuous behaviour. OPEN IN VIEWER

OPEN IN VIEWER

Figure 2.

Stereotypes, prejudices, beliefs, and perceptions that male respondents of different origins have about women from their own country and from other countries of origin. Note: Ovals in grey refer to groups who did not speak about this issue in their interviews. Source: Authors' compilation from our own data.

As Figure 1 shows, the perceptions that female respondents have regarding men from their own country and from other communities are often closely linked to the issue of machismo and the little respect that men can have for women. This belief can also refer to the respondent's own community (e.g. exogamous Moroccan women may reject Moroccan men because they feel they tend to be sexist). The following interview excerpt from a Dominican woman exemplifies this pattern:

I have lived with a Dominican guy (…) And I have lived with a Spanish one [her current partner], and they're nothing alike! The Dominican is lazier, more macho, but much more! (…) Womanisers, lazy, sexist, and everything you can think of. There may be Spaniards like that too! (…) But really, as I have lived with both, and now that I am not in a relationship with a Dominican, I would not go there again, at all! (D2M, Dominican woman, 32)

As we can see in this excerpt and in the one that follows, Spanish (western/European) men often appear in the narratives as partners with whom it is possible to establish a relationship with greater gender equality. As one female Romanian respondent expressed:

Here, women are the boss; men are very submissive. How a couple is understood is very different in Romania than in Catalonia. My ex-husband [Romanian] did nothing. Romanian men are very macho. They are very poorly brought up, with too much permissiveness, and they become tyrants. You do not receive respect from them; they are not kind to you; you have nothing (…) So, imagine after having had a relationship like this before and now being in a relationship of this type [with her Spanish partner], in which the man can't do enough to please me… (R3M, Romanian woman, 49)

In sum, although the members of mixed unions seemed to have developed greater cultural flexibility within their own relationship, they still expressed strong prejudices towards people from other countries of origin, and often also towards people of the opposite sex from their own country of origin. In this sense, ethno-racial prejudices had not exactly been eroded.

Conclusion

In this article, we have challenged the idea – rooted in classic assimilation theory (Gordon, 1964) – that mixed unions clearly weaken social boundaries and negative ethnocultural and racial attitudes.

In line with previous studies that point to the complexity of the link between intermarriage and assimilation (e.g. Collet and Santelli, 2012; Rodríguez-García, 2006, 2007, 2012, 2015; Rodríguez-García et al., 2015; Song, 2009), our current study of 58 exogamous immigrants of varied origins residing in Catalonia, Spain, also found that members of mixed couples suffer social discrimination, particularly from their respective families, who may feel that their group identity and cohesion are threatened (Hohmann-Marriot and Amato, 2008; Kalmijn, 1998). This familial rejection is based on negative stereotypes and prejudices linked to the partner's origin, phenotype, religion (in particular, Islam), gender, or social class (and often an intersecting combination of these factors) and consequently, this disapproval is much stronger for mixed unions involving members of certain groups than it is for others. It is also important to note that opposition to intermarriage emanates both from the Spanish partners' families towards the immigrant partner or spouse and from the immigrant respondents' families towards the (majority-group) Spanish partner or spouse. Moreover, family rejection does not only depend on the family's own prejudices or preferences but also on the attitudes that the family observes in the wider society.

In order better to assess whether or not mixed unions erode ethnocultural boundaries and negative attitudes, we have further taken the factor of time into account. In cases of strong initial rejection (as was particularly the case for certain groups, such as Moroccan/Spanish intermarriages), we observed that the rejection tended to remain strong over time. If the initial rejection was milder or less visceral, however, attitudes were sometimes transformed over time by the everyday relationship of the persons involved. This informs us that, in some cases, but not in all, mixed unions can have a bridging effect on cultural differences. It must be acknowledged, though, that larger scale intermarriage between international immigrants and natives in Spain is still quite recent; therefore, continued research in this field in the future, once the country has had longer experience with immigration and the related phenomenon of intermarriage, will be needed to assess the true extent to which intermarriage and mixed families influence ethno-racial boundaries and attitudes.

Apart from focusing on the reactions that these mixed unions received from the families of the couples and from the wider society, we have also focused on the prejudices and preferences expressed by the members of the mixed unions themselves. This examination revealed that ethnic and racial prejudices (which were particularly salient when respondents were referring to marriage preferences for themselves and their children) and discriminatory attitudes towards other immigrant minority groups also exist among exogamous couples themselves. We found rejections existing in all directions: some Romanian and Chinese respondents expressed that they would reject Moroccans as partners; Moroccans and Senegalese stated they would not partner with Latin Americans; Latin Americans would not partner with ‘blacks’ or ‘Moors’; Romanians would not partner with Latinos or Chinese; Chinese would reject Latinos; Latin Americans would reject Pakistanis…and so on. These findings, again, show that mixed unions only erode ethnocultural boundaries and prejudices to a certain (and perhaps, limited) extent; the absence of prejudice is less likely when communities are more stigmatised to start with, as can be seen in the case of (especially male) Muslim immigrants from the Maghreb in Spain.

In conclusion, while acknowledging that mixed unions, or more broadly ‘mixedness’, can, to some extent, become an engine for the increased ethnocultural integration of immigrants, the weakening of social boundaries, and the growth of inter-group solidarity (Barbara, 1985; Bystydzienski, 2011; Gordon, 1964; Rodríguez-García, 2006, 2015; Varro, 2003; Vucinic-Nescovic, 2002), our findings contest the role of mixed unions as a clear diluter and harmoniser of differences.

We believe that intermarriage can be considered a micro-laboratory of intercultural relations that can help our understanding of intercultural dynamics at a macro level. Further, the study of mixedness can illuminate the socially transformative value of mixed unions/families while, at the same time, shedding light on the disheartening persistence of ethnic and racial divides. In this respect, we think that more research is needed in general in the field of perceptions, attitudes, and social inclusion/exclusion processes, including research on the children of mixed couples, whose lived experience might be at odds with theorised ideals of integration (e.g. encountering racial labelling and discrimination on the basis of their phenotype, and consequently experiencing ‘identity mismatch’ and ‘reactive ethnicities’ (Rumbaut, 2008)), as some recent studies have shown (Aspinall and Song, 2013; Rodríguez-García and Miguel-Luken, 2015; Simon and Tiberj, 2012). Finally, the factor of religion – arguably the most prominent divisive barrier in Europe – including a closer look at the racialisation of Islam (Meer, 2014; Rodríguez-García, 2015), is an area that, in our view, deserves more attention in future research on mate preferences and attitudes towards intermarriage.