Does the superdiversity label stick? Configurations of ethnic diversity in Dutch class rooms

Introduction

The impact of migration on societies has become a prime focal point in international sociology. Especially in Europe, the superdiversity concept has inspired a booming number of publications after Steven Vertovec’s introduction of the term in 2007. Are we witnessing the take-off of a new ‘grand idea’ (Alba and Foner, 2014) of superdiversity that will shed new light on our understanding of migrants in relation to non-migrants? Is it helpful to discover what we missed so far? It is too early to tell, since the provision of sufficient clarity on what superdiversity actually means, what it claims and how to theorize and operationalize it has not kept pace with the increasing popularity of the term (Crul, 2016).

The aim of this article is to contribute to the exploration of the heuristic value of the term superdiversity by assessing its applicability to the experiences and interactions of migrant and non-migrant students in the classes of a Dutch university of applied sciences. It will focus particularly on social and cultural aspects within institutional settings like education since these aspects may be profoundly shaped by such institutions. I argue that superdiversity’s heuristic value depends on two questions. First, does superdiversity’s connotations of positive excitement reflect the lived and subjective experiences of these students in their daily interactions and meaning-making? Second, is their reality in a more objective sense marked by something new, more, or more complex than we knew before, as the term superdiversity suggests?

Findings provide negative answers to both questions. First, superdiversity’s key experiential associations turn out to be irrelevant to the main concerns of these students, i.e. to enjoy education and to construct their identities as well as to avoid ethnic boundary constructions. They reject notions of diversity as something ‘super’ or as something exciting. Second, findings also show that key descriptive connotations of the superdiversity concept hardly apply to the two configurations of ethnic diversity found in this university’s classes, i.e. commonplace and essentialist diversity. There are no reasons to suppose that these configurations represent anything new, more complex or more multidimensional than we knew before. They also contrast with superdiversity’s expectation of leaving behind the ethno-focal lens.

The article will first discuss the main claims and connotations attributed to the term superdiversity. Next, it will relate superdiversity’s claims of more (complex) diversity in a demographic sense and successively in a social and cultural sense to the evidence put forward so far. Subsequently, the case under study and its applied methods will be outlined: 125 interviews with students and teachers of this Dutch university of applied sciences. The ensuing findings part assesses whether respondents’ statements reflect superdiversity as subjectively lived experiences followed by a classification of findings into two configurations of diversity to see whether superdiversity may serve as an objective description of the reality in the classes of this university. Finally, the consequences of this study’s findings for our understanding of superdiversity will be debated.

Superdiversity and its risks

Diversity studies concern the social scientific enquiry of modes, mechanisms and outcomes of social differentiation in complex social environments. They include research on ethnic diversity, i.e. real or assumed social and cultural differences between people that are somehow related to different origins and backgrounds. However, when studying ethnic diversity, we should avoid taking ethnic categories as fixed and for granted and we need to be sensitive to complexity and multidimensional articulations with other differences such as age, gender, class, migration status, etc. This demarcation of the field of (ethnic) diversity studies by Steven Vertovec (2015) in his edited volume on diversity studies is particularly relevant for migration’s impact on society.

Within this field, Vertovec (2007) also introduced his concept of superdiversity to refer to his London findings. Its inhabitants stem from a wide spectrum of countries and ethnic backgrounds, which creates complex configurations of population compositions, migration flows and channels, legal statuses and human capital that involve migrants and non-migrants (Meissner and Vertovec, 2015). Especially in Europe, his term has inspired a burgeoning of studies. In 2016, the term was already used in more than 325 publications, ¹ but this booming literature fails to provide a clear definition of the term superdiversity. Some fuzziness surrounds superdiversity, as Fran Meissner (2015) argues.

Nevertheless, writers who play a central role in this literature (e.g. Blommaert, 2013; Meissner, 2015; Meissner and Vertovec, 2015; Vertovec, 2007) associate the term superdiversity with several specific meanings. They claim that we witness a diversification of diversity, above and beyond what we have previously known in terms of spread, speed and scale of diversity, fomenting complex sets of configurations, representations and encounters. Other robust claims suggest that superdiversity has become the central characteristic of our times (Maly, 2016; Nicholls and Uitermark, 2016; Phillimore, 2015). ² Jan Blommaert (2013, 2015) and others write about superdiversity in quite excited and superlative terms, arguing that it is hugely complex and unpredictable, even requiring new paradigmatic approaches.

How to operationalize such ‘an excess of revolutionary rhetoric’, to use Lars Hinrichs’s (2015) words? ³ First, the prefix ‘super’ suggests that there are new or more complex realities that are experienced by people in the field (not just by scholars like Jan Blommaert) as something that arouses their excitement, something that reflects their lived experiences and concerns in everyday life in a positive way. This is superdiversity’s experiential dimension. Second, the term also suggests that there is something more, or more complex, or qualitatively new to be found in diverse realities compared to what we have previously known. It would urge us to leave behind the ‘ethno-focal lens’ (Meissner and Vertovec, 2015), i.e. an essentialist focus on migrants as representing ethnic minorities or categories. These analytical or descriptive dimensions of superdiversity can be confirmed when studies demonstrate such more complex, multidimensional or non-essentialist new realities.

Before assessing the heuristic value of superdiversity as lived experience and/or as an analysis or description of reality, some risky trends in the literature need to be discussed. The current popularity of the term superdiversity as a new ‘grand idea’ (Alba and Foner, 2014) may seduce superdiversity authors to selectively pick evidence that fits the term, to ignore countervailing evidence and to abstain from looking for sufficient evidence at all, as pointed out by Aneta Pavlenko (in press) in her review of superdiversity sociolinguistic studies. Several authors tend to exaggerate the applicability of the term superdiversity. Titles like ‘in times of superdiversity’ (Maly, 2016) or ‘in the era of superdiversity’ (Nicholls and Uitermark, 2016; Phillimore, 2015) or ‘in a world of superdiversity’ (Schrooten et al., 2016) suggest that superdiversity has become the defining trait of our times and world, without sustaining such bold claims with adequate evidence. Many studies have put ‘super’ before ‘diversity’ without indicating why (e.g. Becci et al., 2017). Why would ‘diversity’ have ended and turned into ‘superdiversity’? Based on which criteria? Such questions remain open.

Superdiversity authors put forward questionable claims about a wider historical, social and cultural context. For example, in his study of the Belgian town of Ostende, Ico Maly (2016: 718) concludes that ‘rapid social changes’ have taken place, but has only demographic data and linguistic signs in public spaces to account for his conclusion. Despite his ethnographic claims, he presents no voice or view of any of the town’s inhabitants. In a similar linguistic landscape study in the Antwerp neighbourhood of Oud-Berchem, Jan Blommaert (2013, 2014) provides interesting insights behind the scenes of a variety of public linguistic signs. He argues that this variety is related to ‘massive demographic, social and cultural shifts over the past two decades’ (2014: 432), but fails to present any data to sustain such claims. He labels these shifts as superdiversity, but without indicating why.

More (complex) demographic diversity?

There is not enough space in this article for an exhaustive overview of the relationship between the literature on superdiversity and empirical evidence. Nevertheless, based on the work of key authors in the debate, work that is recurrently referred to by others, some questions need to be raised as to the heuristic value of the term superdiversity, both in terms of the lived experiences of the people they write about and as an analysis or description of their reality. As to the former, despite ethnographic pretensions on the part of several superdiversity authors, the voices of the people whom they write about are often absent in their work or fail to express the positive excitement the superdiversity literature suggests. That casts doubts on whether superdiversity’s connotation of positive excitement reflects their experiences. More research on the subjective and experiential dimension of diversity is needed to be able to tell whether this connotation applies.

As to superdiversity in a descriptive and analytical sense, Vertovec’s 2007 London article has triggered a number of urban studies in Western Europe (e.g. Crul et al., 2013; Nicholls and Uitermark, 2016) that discovered a similar wide spectrum of national and ethnic backgrounds of inhabitants. The non-migrant population has become a minority here and migrants no longer fit into a limited set of relatively large categories. This demographic argument associated with superdiversity seems compelling. Moreover, Maurice Crul (2016) demonstrated the articulation of migration categories with stratifications in educational performance, underscoring superdiversity’s claims of complexity and multidimensionality.

If we take a longue durée perspective (Braudel, 1982), the newness of Vertovec’s (2007) and others’ findings on urban demographic fragmentation becomes far less evident, though. Already in the seventeenth century, Dutch cities like Amsterdam and Leiden were populated by a variety of people stemming from different places, who made up large parts of their populations (see Grigg, 1980). The distances these people had to cross when migrating to these cities were substantially smaller than now, but that does not mean that in cultural or social terms these people were less diverse. Demographic diversifications of origins, gender, age, class and migration channels as well as their functions as hubs in transnational migration flows were not much different in these seventeenth-century Dutch cities compared to contemporary urban centres.

Even going back just a few decades, claims of new diversity are not always convincing either. Jozefien De Bock (2015) showed that the population of and migration to Ghent from 1960 to 1980 were not very different from today. Pavlenko (in press) points to the fact that most Eastern European cities are marked by exactly the opposite from superdiversity, i.e. substantially less demographic diversity than one or two centuries ago. We need longitudinal and historical research to be able to tell. Until then, claims of new, more, or more complex demographic realities remain unsubstantiated.

More (complex) social and cultural diversity?

Even if contemporary urban spaces embrace a wider spectrum of countries of origin of their migrant population and more complex articulations with other demographic categories (like age and gender), with migration flows and channels, legal statuses and human capital, does that entail also more diversity in social relations and in what people actually do, think and create? Does superdiversity provide added value to analyse such social and cultural matters? In her excellent ethnography of the London borough of Hackney, Susanne Wessendorf (2014) demonstrates that demographic complexity intersects with differences in class, ‘race’, ‘ethnicity’, age, legal status, lifestyles, identities and so on. However, may we infer from those complexities and intersections a similar complexity in social and cultural diversity, i.e. in patterns of meaning-making, behaviour, identity formation and socialization?

Such an inference is questionable. It can only be sustained by perceiving migrants and non-migrants in a cultural essentialist way (Grillo, 2003), i.e. as bearers of different cultural repertoires and social relations that would stem from their group-belongingness and that they naturally would like to enact, both in private and in public. Such an inference would take the salience of ethnic or national identities in daily interactions for granted – i.e. that would adopt the ethno-focal lens superdiversity authors reject – as if these identities would naturally overrule or sideline more institutionally ‘nested’ identities of the same people, like professional or civic identities. Superdiversity authors hardly pay attention to the homogenizing and regulating impact of modern institutions.

Susanne Wessendorf (2014) found that, in Hackney, there is no such direct relationship between diversity in cultural and social terms, and demographic complexities and intersections. Hackney’s population represents a widely diverse demographic spectrum, but in public and parochial spaces (in-between public and private, like IT clubs or parents groups) most people behave in a rather homogeneous way following the same codes of civility and conviviality. Private life in Hackney is primarily structured along class lines. Only within each class, social milieus with particular cultural characteristics and social interactions emerge. Some social milieus have developed on an ethnic, national or racial basis, but not in an unmediated way. ‘Ethnicity’, ‘nationality’ and ‘race’ have become refigured through the ethnicization, nationalization or racialization of sports, music, dress, language (accents) and lifestyles, and moderated by legal status, racism and age, as Wessendorf (2014) shows. In short, it would be too simple to say that demographic fragmentation and intersections would entail ‘more’ or ‘more complex’ social and cultural diversity in Hackney.

Evidence about meaning-making and patterns of behaviour in society does not point unequivocally to more diversity. For example, Dutch surveys on views on marriage, sexuality and family life show a strong consensus and homogeneity among both non-migrants and most migrants (Duyvendak, 2011: 87–92). Sociolinguists like Jan Blommaert, Ben Rampton, Karel Arnout, Max Spotti and others (e.g. Arnaut et al., 2016; Blommaert, 2013) highlight a vast variety in communicative, linguistic and semiotic forms in transnational frameworks. Their linguistic landscape and ethnographic studies (online and offline) have convincingly highlighted sociolinguistic variation and differentiation, but without actually showing that this variation or differentiation would reflect more or new diversity in socialization, meaning-making and behaviour in any way. We need more evidence to tell whether superdiversity’s claims of new, more, or more complex diversity actually apply to demographic, social and cultural reality of migrants and non-migrants. The question regarding the heuristic value of the term superdiversity, both as a reflection of lived experience and as an analysis or description of reality, remains unanswered.

Case and methods

The current study assesses this heuristic value in the case of the social and cultural reality of the class rooms of a Dutch university of applied sciences. Our study ⁴ was carried out in 2011, when this university had about 21,000 Bachelor students, enrolled in 46 programmes in 13 faculties. It is located in a metropolitan area, the largest municipality of which has a majority of inhabitants with a migration background (52.7%, cbs.statline.nl).

Due to legal privacy restrictions, the student administration had no reliable data regarding the ethnic composition of the student population. However, we included relevant questions in the anonymous questionnaire distributed among all Bachelor students in another project on ethno-migrant inequality in study performance. The sample represented a response rate of 12.6% (2593 students), reflecting similar indicators of cohorts, study success, gender composition and prior study backgrounds as the total student population (student administration’s data). Of the students’ sample, 55.4% are female, but gender ratios differ substantially between the programmes. On average, students are 21.5 years old with a standard deviation of 4.4 years. Many programmes have a part-time variant for elderly students. As to class background, 47.9% of the students’ parents had enjoyed academic or higher vocational education.

A majority, 58.9%, of respondents had a non-migration background. The remaining 41.1% had a first or second generation migration background stemming from a wide variety of countries within the officially used categories of ‘Western’, ‘non-Western’ and Indonesia. First generation migrant students included those who came with their parents to live in the Netherlands, but also those (8.8%) who came to the Netherlands to study in one of the international programmes. The ethno-migrant composition of the programmes varies strongly: some have no migrant students, whereas in other programmes they make up 75%.

Thus, on the one hand, the student population reflects a considerable diversity in intersecting social and demographic terms: background, migration status, gender, age, class and legal status. On the other hand, there is a majority of non-migrants. Consequently, this composition contrasts with the demographic composition of places like Hackney (Wessendorf, 2014) that have no such majority, but also with a situation with a few clearly defined minorities.

Interviews provided the main data for our analysis. In total, 71 students from five different faculties were interviewed: 41 with a first or second generation migrant background and 30 with a non-migrant background. In addition, 54 teachers from the same faculties were interviewed: 31 with a non-migrant and 23 with a first or second generation migration background. Respondents were spread as much as possible over various programmes, over age, over gender, and, regarding students, over various cohorts. This sample served to capture various experiences without taking the salience of these categories for granted (Wimmer, 2013). We did not start the interviews with an ethno-focal approach addressing respondents as belonging to any of these categories.

Interviews took an open exploratory approach. Our basic objective was to find out what really matters to these respondents with regard to diversity. We first focused on dynamics in class between students and between students and teachers. Subsequently, the role of diversity in these dynamics was discussed, asking for events in which diversity or ethnic identity became salient, the factors that triggered this salience, the behavioural consequences of this salience as well as the impact of such events on learning processes. With students, we focused on their own class and in some classes we selected several students to have multiple views on the same class. With teachers, we discussed their experiences in all their classes. We presented ourselves as researchers from another university and guaranteed anonymity and confidentiality of statements. There were no credits to be gained.

For data analysis, Matthew Miles and Michael Huberman’s (1994) steps were used of data reduction (selective, open and axial coding), data display and drawing conclusions. The unit of analysis was what happened in class and how the individual respondents experienced class dynamics. First, the issues that respondents considered to be most important regarding diversity were singled out reading through the transcripts. That turned out (a) to be able to socialize with all in class, to enjoy learning and to develop their own individual identity as opposed to (b) being confronted with ethnic subgroup formation that harms socialization and learning, and curbs individual identity formation. Second, all codes that refer to (a) or (b) were identified and text fragments were coded accordingly. Third, statements on the relations between those codes per configuration were singled out. Thus, basically two different configurations emerged from a systematic review of the interview scripts: commonplace diversity and essentialist diversity.

Examples of both will be presented below in their typical forms, but in practice we found many shifting in-between situations. That means that nuances in the presentation of findings will focus on the findings within each type; space does not allow for a nuanced discussion of all in-between situations. Exact estimations of the distribution of the various configurations over the various classes cannot be provided, but essentialist diversity applies to a considerable minority of classes, commonplace diversity fits a majority of classes.

Superdiversity as lived experience at school?

Ever since the 1980s, official Dutch policy discourses look at migrants with a so-called ‘non-Western’ background through the ‘ethno-focal lens’. They are seen as exemplars of ethnic categories or minorities with deviating cultural norms and values. At first, their ethnic community membership was perceived as beneficial for their societal participation, but since the turn of the century migrants’ assumed cultural traits are portrayed as incompatible with dominant Dutch norms and values and held responsible for their problems in societal participation (Siebers and Dennissen, 2015; Duyvendak, 2011). In short, superdiversity authors may call upon us to leave behind the ‘ethno-focal lens’ (Meissner and Vertovec, 2015), but official Dutch discourses do exactly the opposite.

The same holds true for the university’s management. In 2008, the Equal Treatment Commission, the official Dutch tribunal to file a discrimination complaint, ⁵ had condemned one of its faculties for discrimination against two migrant staff members. Subsequently, the management issued an anti-discrimination plan that includes not only the instalment of confidentiality persons and discrimination complaints procedures, but also a large number of diversity initiatives to boost a more inclusive climate. Diversity courses and trainings became standard courses of many programmes, a diversity minor was developed and diversity chairs were installed.

However, these diversity initiatives contrast clearly with the experiences and views of the students themselves and do not arouse any enthusiasm among them at all. Diversity is not something they like to discuss. Students with a migration background have good reasons to avoid the subject and feel anxious about it. In the Dutch context, the risk of ethnic boundary making and exclusion (Baerveldt et al., 2007; Wimmer, 2013) against those students is real, fuelled by the nationalist discourses in media and politics discussed above (Siebers, 2017; Siebers and Dennissen, 2015). Thematizing diversity through such discourses entails a risk of triggering conflicts.

However, even in contexts in which such discourses are not relevant or salient at all, students with or without a migration background are unwilling to thematize diversity. To them, diversity is something normal, banal and part of everyday conviviality. It is commonplace (Wessendorf, 2014), not something to be singled out or highlighted. Respondents used the word gezeur (‘moaning’ or ‘fuss’ in English) to refer to diversity talk as something irrelevant to them. Consequently, in the beginning it was difficult to arrange interviews with them since they simply did not want to talk about diversity. Only after students told each other that we were interested in what does matter to them without a pro-diversity predisposition, they were willing to talk to us.

Both these politicized and banal elements came together in December 2009. The communication department had decided that the tradition of putting a large Christmas tree in the central hall had become a bit boring, so they decided to introduce a light show instead. Then, the website Geen Stijl (‘No Decency’) announced that the school had given in to so-called ‘Islamization’ threats, denying the ‘Christian’ background of its students. The issue was taken up by the conventional media and questions were even raised in Parliament in a nationalist tone.

Thereupon, a group of students, many with a migration and Muslim background, went to town, bought a number of Christmas trees and put them ostentatiously in the central hall of the main building. Their statement was that not only this politicization of diversity but this whole diversity gezeur should stop. They do not see diversity as something ‘super’, something that deserves celebration or that is relevant to them or triggers excitement. Their statement was clear: ‘Stop all this fuss about diversity’. In short, the lived experiences of these students differ clearly from the language of excitement and superlatives about diversity of some superdiversity authors.

Superdiversity may not reflect students’ lived and subjective experiences, but it still may serve as a description of their reality in a more objective sense. Our data analysis showed that their diversity-related reality can best be typified into two configurations of diversity, one called commonplace and the other essentialist, with many in-between situations.

Commonplace diversity in class

I think, it is more allochthone than autochthone, but the allochthones are very mixed as well. I feel it’s a very mixed class. … Everyone is simply who he is, we support each other.

This quote is from a female student in Human Resource Studies. Halimah ⁶ was born of Moroccan working class parents who started their own business. She was in her early twenties when we interviewed her. Her account illustrates a kind of diversity found in several classes.

Halimah’s class has a superdiverse demographic composition, i.e. there is no clear majority – the autochthones (non-migrants) are a minority – whereas those with a migration background (called allochthones) are very mixed as well. She said that everyone’s identity is recognized, students support each other and the atmosphere is gezellig (Dutch for ‘cosy’, ‘pleasant’ or ‘friendly’). She was proud of her class and claimed it was the best of her cohort’s four classes. She said that only at the start of the year, when students were new to each other, did they tend to look for fellow students with a similar ethnic or linguistic background to socialize with, but subsequently students socialize more with fellow students with a similar performance level. That is, the institutionally ‘nested’ identity as students takes over from ethnicity in guiding their socialization.

In a similar positive and self-assured way, she talked about herself. Although her educational career had proceeded far from smoothly, she emphasized being in charge of her life. She is quite selective: she considers what suits her or fits her identity, what feels good and what does not, and makes her own choices about what to incorporate into her identity. That varies according to the stage of her development, she said.

She talked about her identity as a ‘combination’ of different aspects, roles and fields she operates in and had learned to switch between. Regarding school, she said that she simply wants to be a good student and considered its diversity initiatives to be patronizing. She said that it is perhaps nice to know about the cultural backgrounds of other students, but it is not relevant at school. She did not connect her positive views about her class and herself with notions of diversity. She sees no need to thematize diversity; ethnic or national identities in terms of groupness (Brubaker, 2002) are simply not ‘nested’ at school.

Nobody is ‘them’. … It [her Moroccan identity] is fine as long as it is part of me and does not turn me into being part of the Moroccans.

Where does that leave her ethnic or religious identity? To Halimah, ‘we’ or group-belongingness refers either to humanity or to all students in class, not to something like a Moroccan or Islamic community. She does not accept the idea of ethnic or religious collectivities (Boli and Elliott, 2008) in society or class with boundaries between them. Therefore she said: ‘Nobody is “them” as opposed to “us”. … I do not see how a society can be successful if you start thinking like “them” and “us”.’

She rejects her Moroccan identity as ‘having’ to socialize with fellow Moroccans and ‘having’ to do things in a Moroccan way. She does not want to be addressed as a member or representative of a category like Muslims or Moroccan community since that produces stereotypes and limitations, she said. She stated that at a particular moment, she had decided to wear a head scarf, but stopped doing so when others took it as symbolizing such community membership. If the unequal labour market participation of Moroccans is discussed in class, she does not feel it is addressed to her.

However, she did not deny her Moroccan and Muslim identities either. They are important components of her identity, turn her into a ‘special me’ and allow her to profile her own sense of individuality. She talked about ‘her’ particular Moroccan and Muslim norms and values as resources for building her own subscribed identity. They enrich her own person, she argued. She is quite selective regarding these resources, as her head scarf decision illustrates. She does observe Ramadan, but does not pray five times a day. She celebrates her birthday at home in a Moroccan way, but also the Dutch tradition of Saint Nicholas and Black Pete on 6 December to please her little brother.

Most of her statements about ethnic and religious components of her identity refer to private life, but not all. At school, she likes to talk with fellow students about her choices regarding elements from her Moroccan and Muslim backgrounds that profile her individual identity. She was optimistic and reassuring in her statements despite having experienced serious instances of discrimination, including by some teachers.

Nevertheless, she specified her agency, explaining that she was very active in volunteers’ organizations and in a student association that helped especially students with a Moroccan background with language courses in Dutch and Arabic, study trips abroad, evening dinner at Ramadan and bringing them in contact with senior Moroccans to learn from their labour market experiences. In other words, the configuration in her class enables agentic forms of sociability, identity formation and education processes.

Essentialist diversity in class

… I said, I am fed up with being asked where I come from, whether I am Turkish or Moroccan, which is not what I am. I am fed up with the fact that these things are considered. I just want to … feel like a human being, not like Dutch or like non-Dutch, that it does not matter anymore who I am. I want to be seen as a human being.

Karolina, a 22-year-old student with a first generation Macedonian background, made this emphatic statement in a discussion in her class. She explained to us that her class is split up between an ‘autochthonous’ and a few ‘allochthonous’ groups according to background classifications, mainly Turkish and Moroccan. Karolina’s class reflects a more classical demographic composition of a non-migrant majority category and a few migrant categories.

The crucial issue, though, is that in Karolina’s class these categories have become salient and turned into groupness. Her 24-year-old class mate with a first generation Turkish background, Pelin, confirmed that the class room is physically split between the ‘allochthones’ and ‘autochthones’, each sitting on their side of the room. If you enter class, she said, all are watching whether you will sit with ‘your’ ethnic group. Teams recruit their members for team assignments along ethnic lines. Groups avoid each other and sometimes there are tensions and conflicts between them. Pelin said that in class she does not dare to express that she has non-migrant class mates as friends. She has to represent the views that ‘allochthones’ are supposed to represent in class discussions, so there is a strong pressure to close ranks as an ethnic group. For this reason, Miriam, their 23-year-old non-migrant class mate, said she felt frustrated in discussions in class. She talked about a ‘wall’ typifying the boundaries between ethnic groups.

This groupist behaviour and these strong ethnic boundaries developed over time. Unlike Halimah’s class, this ethnic subgroup formation and ethnic groupist identity formation did not disappear but deepened in Karolina’s class over time, sustained by several incidents. One such incident was sparked by a film shown in class about homosexuality. A male ‘Moroccan’ student started to make vomiting gestures and expressed his disgust of homosexuality. He called it ‘dirty’ and ‘an illness’. A non-migrant bisexual student reacted furiously, saying that she felt attacked and hurt by his comments. Other non-migrant students supported her and migrant students backed up the ‘Moroccan’ student, triggering accusations and insults between ethnic groups. This conflict had already lingered on for over a year when we interviewed these students.

Miriam explained that these subgroups stigmatize each other, addressing members of other groups not as individuals but as exemplars of their ethnic subgroup. They feel pressured to adhere to their own assumed subgroup identity in a ‘black and white’ fashion, she said. These ethnicized processes sustain themselves in a self-propelling way, i.e. independently from the intentions and wishes of the students involved. Pelin, Miriam and Karolina regret these processes. Karolina’s appealing speech in class, quoted above, to stop ethnicizing dynamics was met with a loud applause by everyone, but after a while the same ethnicizing patterns re-emerged. Both students and teachers indicated that such dynamics disrupt learning processes.

Karolina expressed her frustrations in the strongest words. As the only student with a Macedonian background, she felt left out by all groups. The salience of ethnic group identification is causing her:

… sleepless nights when I think I do not fit with the Dutch, I do not fit with the foreigners. In fact, I am nothing, I am nobody. That is the feelings you get at such moments.

Conclusions

Cultural contents related to their demographic or ethnic backgrounds do not enter directly into these students’ meaning-making, behaviour and identity formation in class. In Halimah’s case, her individual identity profiling (see Boli and Elliott, 2008) encourages her to draw selectively on symbolic contents and repertoires stemming from her background. In Karolina’s class, the dynamics of ethnic polarization incite students to bring forth those elements from the repertoires associated with their backgrounds that they can use as arms (see Barth, 1969). Teachers confirmed that in one-on-one conversations, students express more nuanced views than in class. Consequently, a diversity in origins and backgrounds does not directly produce a diversity in meaning-making, behaviour and identity formation. Thus, superdiversity authors’ demographic diversity and complexity argument does not necessarily translate into diversity or complexity in a cultural or social sense.

The concept of superdiversity was introduced to depict the demographic diversity and complexity of migrants and non-migrants in Western European cities where non-migrants have become a minority (Vertovec, 2007). This demographic composition holds true for Halimah’s class, whereas a more classical demographic composition of a non-migrant majority with a few migrant minority groups is visible in Karolina’s class. Although they do coincide in these two classes, the overall data show that there is no coincidence between demographic composition and the kind of diversity configurations in class. A superdiverse demographic composition is not necessarily linked to commonplace diversity nor is a classical demographic composition related to essentialist configurations. So, demographic diversity does not determine social and cultural configurations of diversity. More important are the creolizing or essentializing ways of using and processing the cultural contents and repertoires stemming from origins and backgrounds.

Our findings regarding the social and cultural aspects of diversity in the classes of this university differ from what might have been expected from a superdiversity point of view. On the one hand, superdiversity does not reflect students’ subjective and lived experiences. Students do not express the positive excitement several superdiversity authors associate with diversity nor does the term cover the issues that do matter to them, i.e. that class dynamics are in sync with institutional demands to facilitate education, socialization and creolizing identity formation. Most students prefer to avoid the subject of diversity, either because it represents a risk to them or because it seems irrelevant. To them, diversity talk is gezeur (fuss), definitely not something ‘super’.

On the other hand, findings also question the applicability of the term superdiversity as an objective description of cultural and social reality for several reasons. First, in contrast with superdiversity authors’ call to discard the ethno-focal lens, students in essentialist diversity classes adopt this very same lens themselves in processes of ethnic polarization in line with official Dutch discourses on migrants and migration. In commonplace diversity classes, ethnic identity remains part of individual identity profiling. In either way, as disruptive groupness or as constructive individuality, ethnic identity does play a role in the meaning-making and socialization of these students.

Second, several concepts do apply to our data: such as conviviality, creolization and commonplace diversity in Halimah’s class, and ethnic boundary making and groupism in Karolina’s class. However, the authors who have developed these concepts do not argue that these concepts represent anything substantially new or more (complex) currently than before. The two configurations of diversity found in this study hardly represent anything ‘new’ or ‘more’ than what we already knew about conviviality, creolization and ethnic boundaries elsewhere and previously.

Third, this study’s findings do not suggest that the configurations of diversity in these classes are particularly complex or multidimensional. In Halimah’s class, the articulation of ethnic backgrounds with age, gender, socioeconomic class and religion is quite unproblematic. In Karolina’s class, gender, sexuality and religion do play a role, for example in the incident on homosexuality discussed above, but subordinated to the dynamics of basically ethnic polarization (see Siebers, 2017). These dynamics drive students towards subgroup homogenization and complexity reduction, simplifying mutual ascriptions to one dominant ethnic identification, i.e. towards the opposite of complexity and multidimensionality.

Thus, the main connotations of the term superdiversity do not reflect the main concerns and subjective experiences of these students nor do they apply to these configurations of diversity. The term superdiversity has no heuristic value for understanding these students. This study has its limitations, though. Longitudinal research is recommended to get an even better insight into whether diversity is becoming more complex or multilayered. Observations would be helpful to get a more detailed picture of dynamics in class. Quantitative data would be welcomed to assess in more detail the relative weight of commonplace versus essentialist diversity, preferably not limited to this school. More knowledge is needed to explain why commonplace diversity manages to develop in some classes, while essentialist diversity succeeds in shaping the dynamics in other classes. The key to such explanations may be found in the interplay between the regulatory impact of institutions and the influence of ethno-nationalist discourses (Siebers, 2010, 2017) promoting invented collectivist identities (Boli and Elliott, 2007).