Ethnic Repertoires
among native Dutch
Amsterdam residents
Exploring vocal majority dispositions using
Q-methodology
Masterthesis Thijs Lindner Studentnumber: 6141994
Supervisor: Jan Willem Duyvendak Second Reader: Giselinde Kuipers Research Master Social Sciences University of Amsterdam Submitted on: 31-1-2018
1 [Abstract]
This paper analyzes how ethnic Dutch residents of two Amsterdam neighborhoods evaluate and reflect on ethnic discourse. This is done by q-methods, in which online surveys, q-sorting and interviews were held with residents. This resulted in four distinct patterns to evaluate, view, describe and reflect on issues of ethnicity, integration and ethnic belonging, so-called ethnic
repertoires. This paper argues that among upper-middle class residents in Amsterdam, a progressive
consensus over the subject matter does not preclude a hegemonic disposition. Rather, there are distinct differences in the how ethnicity, integration and ethnic belonging is understood and conceptualized, which implicates, and at times reinforces, symbolic boundaries over the imagined ‘us’ versus a generalized ethnic ‘other’.
2
Introduction
In ethnically diverse societies minority status no longer implies a secondary social status and presumes equality to ethnic natives. Formally, citizenship status secures this principle of civil egalitarianism. Stricter notions of nativist exclusionism, however, impair this basic constitution. The vocal majority in the Netherlands is increasingly subjected to more lenient notions of the ethnic ‘us’ versus the ethnic ‘other’. It is therefore crucial to analyze how the vocal majority that critically views, or opposes populism and nativism, views ethnic belonging and ethnic minorities (ethnic repertoires). This paper analyzes how repertoires link with ethnic dispositions among native Dutch. Dispositions are norms of specific manners, behavior and ways of thinking or viewing phenomena and objects, socialized by distinct cultural contexts, such as ‘ethnocentrist
dispositions’ or ‘aesthetic dispositions among cultural elites’. In this paper, dispositions are treated as system of belief (c.f. Lahire 2003). Ethnic dispositions, therefore, are cultural distinct representations of ethnic minority groups or ethnicity in general.
Natives, or, in more inclusive terms, ethnic Dutch1, are in most contexts still ‘the’ vocal majority (c.f. Kremer 2016). Mainstream vocal majority dispositions are often viewed through hegemonic progressive consensus (c.f. Kešić and Duyvendak 2016) Yet, when considering normative interpretations of integration and ethnic belonging, there is less of a cohesive ‘we’, but rather a contested arena, primarily articulated through politically oriented nativism versus
multiculturalism or inclusivity. Right-wing nativist sentiments are criticized thorough and are under great attention from both scholars and public media. Progressive notions by the majority on the subject matter, rather, have received less scholarly attention. Therefore, this study focuses on how ethnic Dutch in a liberal setting view ethnic minorities. Ethnic repertoires among
residents from two neighborhoods in Amsterdam are studied. Residents from this city are chosen because Amsterdam is known for its tolerant, progressive and cosmopolitan climate.
Residents from: the Indische buurt (‘Indian2 neighborhood’) and the Museumkwartier
(‘Museumquarter’) participated in the research. These neighborhoods differ in the level of ethnic mix and socioeconomic background. The Indische buurt as currently gentrifying, formerly lower class, ethnically highly diverse area, and the Museumkwartier as an affluent, mainly ethnically homogenous (ethnic Dutch) neighborhood. Different levels of ethnic mix might alter how residents from different urban areas view co-national ethnic minorities, and therefore this is taken into consideration.
1 See Slootman 2014.
3
The paper is structured around the central question how ethnic Dutch Amsterdam residents evaluate ethnic discourse, and how ethnic repertoires are constructed in the process of evaluation. In total, 29 respondents participated in the study. Q-methodology was used to discover and analyze ethnic repertoires of the resident groups. Respondents filled in a q-sort and were interviewed about the general topic of ethnicity and ethnic belonging in Amsterdam and in the Netherlands. The results show an overall appreciation of progressive dispositions, yet there are fundamental distinctions in the interpretation of minority position and boundaries around the ethnic ‘us’ versus a generalized ‘other’.
In the next section, an overview of the literature around the main concepts used to empirically study the question at hand will be provided. Then, methods, case selection and data collection are discussed. In the fourth section, the empirical results are presented and discussed. Lastly, this paper concludes with a discussion of the implications of the findings.
4
Theoretical Orientations
In this section the central theoretical concepts and orientations are expanded on that are used to analyze ethnic repertoires in a Dutch setting. First, the conceptual standing of this paper within the field of existing literature is discussed. Then, the central concepts ethnic repertoires and ethnic
dispositions will be considered. Thirdly, focus is on the ethnoracial hierarchy and its relation to the ethnic mainstream. In the fourth part, contrasts between the endorsers of ‘superdiversity’ literature and
its criticizers are discussed. In the last part the concept of culturalization of citizenship and its orientations are examined.
Conceptual & academic problems
In light of current debates on consequences of the educational gap – especially considering support for or opposition towards ethnic newcomers – there is a tendency towards thinking that ethnically exclusive representations of newcomers is part of a lower class disposition (the ‘class-principle’). In other words: ‘higher educated are not guilty of racist discourse, because they tend to use inclusive language, while lower educated are, since they face economic competition’ (c.f. Lamont et al., 2016). Other studies (such as van der Waal and Houtman 2011; Wekker 2016), however, show that notions of ethnic exclusionism among natives cannot be unilaterally and exclusively tied to political orientation or educational level; thus the ‘class-principle’ is
problematic. For example, van der Waal and Houtman (2011) show that levels of ethnocentric sentiments are not the same among lower educated groups in industrial and postindustrial cities, since postindustrial cities foster a more tolerant cultural climate. Moreover, Duyvendak (2011) and Schinkel (2008) assert that notions of culturalized otherness and cultural exclusion are widespread in the Netherlands, also among higher educated individuals. This paper further examines the problematization of the class principle with respect to the process of discursive ethnic boundary making.
A primary feature of the ethnic majority is a capacity to define and vocalize boundaries of inclusion and exclusion (Kremer 2016). Identification and ascription of ethnic minorities are a primary feature in the process. Ethnic belonging is constructed through shared norms and notions of recognition, which frames the ethnic majority as a precondition for mainstream affinity. Power hierarchies, especially those hierarchies (partly) based on ethnic classification, are discursively framed and structured by the ethnic majority, natives (‘autochthones’) in the
Netherlands. Autochthones are therefore depicted as the norm (Huijnk et al. 2015), which sets basic conditions of cultural, social and emotional value orientations (such as norms of gender relations and sexuality). Currently, there is great attention to study the influence of superdiversity
5
with respect to notions of integration (see Meissner and Vertovec 2015; Vertovec 2007, 2017). However, in this line of research there is a tendency to obscure the lasting effect of the native majority over cultural ethnic representations (c.f. Kremer 2016). This paper analyzes how ethnic Dutch, the ethnic mainstream, in the Netherlands make sense of, and structure ethnic repertoires.
Another issue emphasized in this study is that ethnic minority belonging in the Netherlands (as elsewhere) is highly politicized. Since the 1990s, this has been fueled by state policies which yield increasing restrictions on civic belonging (c.f. Uitermark, Duyvendak, & Rath, 2014), and as result, notions of citizenship have become culturalized (Duyvendak, Tonkens, and Geschiere 2016). Moreover, there is an overall heightened sense of the ‘radical Islam other’, which has concurrently resulted in public association with fundamentalism and terrorism if one has distinct Arabic looks. This study analyzes how the discursive politicization of ethnic minority belonging is present among the vocal majority, and in particular the majority that is known to assert a more tolerant and liberal conception towards immigrants and integration.
Also, this study elaborates on the notion that normative identification of ethnic others among native white Dutch yields notions of ethnic in- and exclusion. Relatedly, it is important to understand how notions of, and notions and towards discrimination work3. This study discerns the public understanding of discrimination and grasps its meaning viewed as a practice of
discursive and social sense-making. This enables to take an objective look at repertoires of ethnic evaluation. Two emblematic normative repertoire strands are considered here: the ‘nativist strand’ (yielding exclusionary boundaries of ethnic distinction) and the ‘ethnically inclusive strand’
(emphasizing openness towards ethnic others). Both normative strands analytically categorize belonging based on ethnicity. Although the ‘ethnically inclusive strand’ is more prevalent in mainstream media outlets, members of ethnic minority groups might experience particular discursive mainstream repertoires as discriminatory. In this regard, Wekker (2016) shows that the Dutch colonial past has framed certain repertoires within a racially denominated black-white dichotomy (often implicitly yielding the black as negative), such as the tradition of Black Pete; the blackfaced servant of Saint Nikolaas. In effect, this has unconsciously led to discursive practices of ethnic distinction-making in contrast to a public feeling of innocence toward ethnic
boundaries in Dutch identity and culture (Ibid.). This study analyzes how both nativist discourse as well as ethnically inclusive discourse seemingly relate to (and/or reinforce) patterns of ethnic evaluation. In effect, three normative strands of ethnic repertoires are distinguished: nativist, inclusive and neutral. This last normative strand acts as a less politicized middle ground between
3 Mind that the concept of discrimination does not tell anything about perceived or experienced forms of
discrimination (or racism), the concept here merely adheres to the notion that repertoires of (group) identification among native white Dutch are (partly) structured using ethnic categorization and boundary making.
6
the two ‘extreme end’ repertoires. This strand was taken into account, since participants in this research were to be able to identify towards at least one normative strand, without being felt antagonized. However, all three normative strands analytically categorize on the basis of ethnic belonging, and are, in essence, practices of discursive sense-making. How these normative strands associate with repertoires of ethnic evaluation, and how they are interpreted is of concern in this study.
This study mostly takes an inductive approach, since the goal is to explore vocal majority dispositions on issues of ethnicity and ethnic belonging. With respect to this, it is important to note that within the research design one crucial aspect was more deductively approached: residential location and neighborhood composition with respect to ethnic heterogeneity. This is done because level of residential ethnic diversity contributes to a sense of group belonging and solidarity. See for example literature on the ‘neighborhood effect’ (c.f. Wilson 1987) and residential repertoires (Blokland 2003; Mepschen 2015). These studies show that one feels (dis)associated with neighboring residents on the basis of shared characteristics, such as class, ethnic background, age, educational level, and culturally defined norms and values, which construes notions towards ethnic others. Level of ethnic mix therefore (partly) determines individual repertoires about ethnic minorities. In this regard, this paper explores if different settings with respect to ethnic diversity impact ethnic dispositions differently.
Ethnic repertoires & dispositions
As mentioned in the introduction, ethnic repertoires are patterns of evaluation of ethnic minorities. Ethnic minorities are groups considered not belonging to the ethnic mainstream (more on this in the next subsection). These patterns of evaluation are based on symbolic boundaries: “[…] conceptual distinctions that we make to categorize objects, people, practices, and even time and space” (Lamont 1992: 9). Lamont’s notion of repertoires is used to study how ethnic Dutch Amsterdammers view ethnic minorities and integration. Here, repertoires are “a set of tools that [are] available to individuals to make sense of the reality they experience” (Lamont et al. 2016: 21). Repertoires are thus mental representations constituted through shared
understanding and inter-subjectivity, which shows a discursive orientation from certain putative groups and cultures, and renders a mechanism of cultural identification and distinction visible (c.f. Lamont 1992; Lamont and Molnár 2002). In the case of ethnic repertoires, the symbolic boundaries investigated in this paper are about the conceptual distinctions of ethnic belonging; identifiers of ethnic and civic membership that are discursively employed to construct a sense of groupness and of a social hierarchy. Often, these distinctive markers of ethnicity are subject to
7
debate; what do they mean and what is implicitly meant when they are used? Micro-level subjective evaluation of ethnic repertoires are complex identification mechanisms. Individual repertoires are interpreted and expressed through a context of interaction and based on norms of the putative cultural group, and are therefore conceptualized as dispositions (Lahire 2003; Swartz 2002). Dispositions are operationalized in this study as repertoires of ethnic evaluation4. In doing so, Lahire’s notion of distinguishing dispositions to act and dispositions to believe is followed (Lahire 2003). The first notion refers to habits to act. The latter notion maintains that mental
conceptions, such as sentiments, representations, principles, passions, doctrines, logics, convictions, and values, are disposed by a culturally structured discursive rationale. When individuals therefore evaluate statements on ethnic others (non-ethnic Dutch) they employ rationalizations and logics that are in line with socially accepted and culturally determined norms of viewing and thinking about these groups. Ethnic dispositions are thus norms and discursive representations referring to ethnic and civic position; they show how people think about
boundaries around the ‘Us’ and the ‘Other’. Consequently, studying ethnic dispositions by ethnic Dutch mainstream groups shows how symbolic hierarchies of ethnicity and ethnic belonging are constructed and attested. However, analyzing ethnic dispositions limits itself in this paper to mental logics. Its relation to actual behavior is not analyzed, since this paper is limited to discursive analysis.
Ethno racial hierarchy, the mainstream & ethnic repertoires
In general, ethnic minorities are identified as distinct from ethnic-nationals, which establishes a sense of discursive dichotomy: the native mainstream versus a diverse body of ethnic migrant groups. Immigrant integration occurs, according to Alba & Foner (2015), always in relation to mainstream society, which, “can be thought of as encompassing those social and cultural spaces where the native majority feels ‘at home’ or, in other words, where its presence is taken for granted and seen as unproblematic” (Ibid: 5). These authors state that the mainstream is formed by institutions such as schools and government, “[…] more informal settings such as
neighborhoods inhabited in large numbers by the native majority, and accepted ways of behaving, which needless to say, typically differs to some extent among different native subgroups (as defined by, for example, by social class or region)” (Ibid.). Moreover, through mainstream institutions, dominant societal norms become translated and enacted in local institutions (such as local norms of social interaction). Not belonging to the ethnic mainstream often results in being
4 Put simply, cultural dispositions are patterns of evaluation of cultural capital (such as having expert knowledge of
classical music). Not knowing the correct markers of high brow culture in a sense betrays a lower social status (c.f. Lamont 1992)
8
(discursively) depicted as an ‘outsider’, which can be thought of as being part of lower rungs of the ethnic hierarchy – the continuum between the (native) mainstream and ethnic minorities. How the ethnic hierarchy is constructed and shaped, which groups define the boundaries, and under what conditions groups can move up, is different for each national context. However, those who have most influence on the process of ethnic hierarchization are everywhere
considered the mainstream. Moreover, the mainstream occupies leading positions in the economy, the polity, the media, and other institutional sectors, and has access to these positions, or has accepted others who have met their standards (Alba and Duyvendak 2017).
In the Netherlands, distance from the mainstream by ethnic minority groups has been identified on the basis of level of ethnic segregation – i.e. the level to which minority groups are segregated from the mainstream (Huijnk et al. 20155). These scholars studied self-identified distance among the four largest non-western immigrant groups6. Their survey shows that Turkish and Moroccan Dutch identify themselves to be more distant from natives than Surinamese and Antillean Dutch do. The Turkish Dutch find themselves to be most distant: 54% feel moderately segregated and 20% feel fully segregated from the mainstream. These percentages are 47% and 15% respectively for Moroccan Dutch respondents. For Surinamese Dutch and Antilean Dutch respondents, these numbers are much lower7. The reason that Turkish and Moroccan Dutch feel relatively segregated from the mainstream is because of their religious orientation – being
Muslim. Their more strict notion of religiosity affects their convictions and orientations (such as more traditional orientations towards homosexuality and women), low level of interethnic contact and processes of strong identification with the country of origin and low identification with the Netherlands (Huijnk et al. 2015). Moreover, interviews with adolescent Moroccan and Turkish Dutch shows that religion is viewed as an important part of their identity, and therefore ‘identifying as a Muslim’ (over ‘being Dutch’, ‘being a Moroccan/Turk’, or ‘being a
Rotterdammer’) is for many interviewees their prime way of identification (Ibid.). Within this paper, notions of ethnic othering, through the lens of religious boundaries (‘being Muslim’), is therefore examined in greater detail.
The ambivalent status of Muslims in the Netherlands is exemplified by Uitermark, Duyvendak & Rath (2014). They studied how the mainstream formalized an ethnic hierarchy, by looking at how municipal governance structures constructed a political discourse of ethnic
5 The authors developed seven social-cultural categories that identified distance from the mainstream: ‘(full)
segregation’, ‘ethnic isolation’, ‘moderately segregated’, ‘isolated and oriented to the Netherlands’, ‘dualistic attachments’, ‘emphasis on the Netherlands’, ‘assimilation’.
6 Turkish Dutch, Maroccan Dutch, Surinamese Dutch, and Antillean Dutch.
7 17% of the Surinamese Dutch in total felt moderately or fully segregated, and 19% for the Antillean Dutch in total
9
hierarchization, which was (initially) aimed at rendering Islamist radicalization visible in
Amsterdam. In the early 2000s, the municipality emphasized on increasing the division between Muslims and non-Muslims, and developed a system that categorized Muslims on the basis of their supposed fundamentalist orientations. The authors explain the dynamics by these state-led policies and overt strategies of the local government as a reaction to an increased perception of radicalization among Muslim youth:
“There is a rough division between Muslims and non-Muslims and both groups are internally differentiated according to their putative civil virtue. Policies are thus based on a certain civil hierarchy: some people are regarded as better citizens than others and it is the government’s task to ensure that people move up in the hierarchy.” (Uitermark et al. 2014: 180)
Although Muslim groups were considered as part of the lower rungs of society before the introduction of these measures, this study shows how local policy measures increasingly (yet probably unintentionally) can affect and imputed social status.
Minority incorporation or mainstream exclusion?
When looking at the literature on the relation between the mainstream and ethnic minority groups in Western Europe, there are currently, roughly speaking, two contrasting perspectives. On the one side there is a group of scholars who consider (the increase of) immigrant groups in Western societies as a trend of social or cultural diversity, in which being native is less considered as the norm for notions of civic belonging. In this perspective old majority groups have become part of a diverse body of ethnicities and identification patterns. This group of scholars consists of, among others, Steven Vertovec (2015; 2007), Maurice Crul (2016; 2012; 2013), and Susanne Wessendorf (2014). On the other side, there is the group of scholars who problematize this notion, and remain ambivalent about the structural implications of increasing diversification of Western societies. This group consists of (among others): Jan Willem Duyvendak (2011; et al. 2016; Alba and Duyvendak 2017), Richard Alba (2015; Alba and Duyvendak 2017;), and Michèle Lamont (2016; 2014). They see that societies are affected by the increasing numerical presence of ethnic minority groups, but remain skeptical about their structural embeddedness in positions that are the heart of the mainstream.
In 2007, Vertovec introduced the term ‘superdiversity’, which supposes a ‘diversification of diversity’ (Meissner and Vertovec 2015), emphasizing new (hybrid and/or intersecting) axes of social identification in contexts with large immigrant populations. A superdiverse context
10
native majorities and foreign ethnic minorities at a micro-level locale. Due to the “new
complexity, caused by the dynamic interplay of variables, current (West-European) societies have [thus] become ‘superdiverse’” (Vertovec 2007). This is made manifest by: net migration inflows, countries of origin, languages, religions, migration channels and statuses, gender, space/place, and transnationalism, which have (supposedly) all increased, diversified, or become more complex (Ibid.). These variables lead to a diversification of numbers; descriptive markers of increased complexity. Later, Vertovec (together with Fran Meissner, 2015) state that the superdiversity perspective provides tools to analyze diversification instead of diversity:
“The perspective opened with these changes in thinking is potentially that of a greater recognition of diversity as normalcy – and one difficult to negate through homogenizing narratives (which were arguably able to gain ground with reference to earlier migration patterns).” (Meissner and Vertovec 2015:550)
Here, we see that ‘diversity’ in superdiversity becomes discursively attested, by including the narrative of ‘normalcy’ and ‘recognition’. What is meant here is an orientation towards a more inclusive perspective, that acknowledges non-native ethnic background as normal instead of anomalous8. In a similar vein do Crul and Mollenkopf note that cities all over Western Europe have become increasingly diversified, where “the old majority group will simply have to adapt to its new minority position within a more diverse terrain of home and work” (Crul and Mollenkopf 2012). Thus, major cities in Western Europe are (or will become) majority-minority cities (Crul and Mollenkopf 2012; Crul et al. 2013). As a result, traditional native majorities lose their numeric majority position. One of these cities is London, where Wessendorf (2014) studied the
neighborhood Hackney. This neighborhood has become increasingly diversified in the last decades, and, in consequence, the area is an example of ‘commonplace diversity’ (Ibid.). This means that, because no singly ethnic minority group dominates the area, and there is large differentiation within ethnic groups, “cultural diversity is not seen as something particularly special, it forms part of their everyday lived reality and is not perceived as unusual” (Ibid. 2). Moreover, commonplace diversity refers to “the societal setting which has diversified over time, to the practices of the people who live there and deal with difference on a daily basis, and to the
8 The term ‘superdiversity’ is used in many different form and ways, and in numerous disciplines, both within and
outside of academia. Vertovec recently studied the manifold uses of the concept, by analyzing well over 300
academic articles, published in a variety of disciplines and subdisciplines (Vertovec 2017). What he shows here is that the concept was used and applied in quite diverging ways. At times its use was dubiously simplified. Yet, in other studies it yielded novel approaches to interpret and understand social forces surrounding the intersectional diversification of ethnicities, (im)migration patterns and integration. Still, it should be noted that among these manifold interpretations and applications of the concept, the role of natives or ethnic-nationals is rather unaccounted for. This further emphasized in a critical reflection on the ethnic mainstream in diverse societies by Alba and Duyvendak (2017).
11
individual processes of habituation in the context of repeated everyday interaction. Diversity thus becomes normal, taken for granted and unsurprising” (Ibid. 45). The concept of commonplace diversity is used in this paper to take account of a differentiated relation between ethnic-Dutch and ethnic minorities in increasingly diversified settings.
The critical scholars of the superdiversity literature mostly focus their critique on the notion that numerical diversity automatically leads to shifts in power relations; as if the
demographic shift to a superdiverse society yields no structural power conflicts and struggles to representation, emancipation and integration. They note that it seems that ethnic succession gets too much considered as an automatic process of adjustment, that has no structural constraints and implications for newcomers, ethnic minorities and the ‘old’ majority group. Mainly, this is translated in the idea that studies on superdiversity tend to neglect the role of the mainstream.
For example, Paul Mepschen studied ethnic repertoires of native residents in the Amsterdam borough Nieuw West (‘New West’), a borough with a relative high share of (former) immigrants (so-called ‘non-western allochthones’), mainly originating from Morocco. He finds that native residents here discursively mobilize a figure of the ‘ordinary’ person, “whose
resistance to immigration is framed as completely natural, as common sense” (Mepschen 2015:4). In effect, these residents unreflexively speak about allochthone Others during small talk with neighbors. According to Mepschen, this portrays the way in which the culturalist framework has established itself in making sense of the daily social context in which they live; “[it] is an idiom of classification that signifies the unreflective, everyday ways in which people live and construe boundaries between self and other” (Ibid.: 9). This results in a focus on ‘autochtony’: a process of boundary-making between those who are construed as guests or strangers and those who belong to the ‘soil’ or nation. In effect, a shared understanding of the “alterity of Others” shapes the local community of natives. In a rather pessimistic tone, he emphasizes that “[rather] than commonplace diversity, the dominance of culturalism in the Netherlands leads to a politicization of difference” (Ibid.: 22). This results in a culturalist worldview that attributes to a lived, bounded structure of co-habitation in Amsterdam New West. This papers presents similar discursive mechanisms among ethnic-Dutch. However, here the focus will be more on dispositions among (upper) middle-class groups instead of lower-class.
Hagemans, Hendriks, Rath & Zukin (Hagemans et al. 2016) studied two gentrified shopping streets in Amsterdam. In the case of the Javastreet (an increasingly popular shopping street in the Indische buurt), the authors found that ethnic shopowners face state-led
12
“The exact criteria for which businesses are “approved”, and which are not, are largely taken for granted and rarely commented on. Yet observations of new businesses on the street, and interviews with new business owners, suggest that ethnicity and cultural capital play a huge role.” (Hagemans et al. 2016:109)
The municipality used a narrow understanding of ‘diversity’ in policy measures, regulations and criteria of establishment for the transformation the Javastraat. The strategy of ‘diversification’ here was not aimed at eliminating the presence of ethnic shops, but to “cater to the tastes and
exotic fantasies of a native Dutch, middle class public” (Ibid.: 110, italics added). Urban renewal
here thus underscores the need for processes of cultural distinction and socioeconomic improvement, based on a culturalized notion of the ethnic Other. The boosting of land value through the displacement of ethnic shopowners by ‘diversifying’ the street with white middle class entrepreneurs, was a strategy of gentrification and a political discourse that ‘justifies’ ethnic displacement. The narrative of local state actors relied heavily on ‘increasing diversity’, yet not diversity of ethnic background or national origins, but diversity of a commercialized sense of the culturally exotic other (which was safeguarded by white native ownership). Ethnic diversity, or ‘superdiversity’, is an attractive term within city planning and policy-making, since it denotes social inclusion and ethnic integration. However, in the case of the Amsterdam New West, and the Javastraat, policies on ‘superdiversity’ were aimed to attract creative classes in order to gentrify poor and problematic neighborhoods, which resulted in (feelings of) residential
displacement. Consequently, in this paper, notions of (ethnic) diversity among residents of mixed urban areas are analyzed critically, by considering its in relation to ethnic dispositions.
Alba and Duyvendak (2017) note that “the growing diversity [which] appears in mainstream settings [gets] connected frequently with a brightening of [the ethnic] boundary, at least from the perspective of the majority, whose members consider some groups of newcomers to embody the antithesis of Dutchness”. This papers argues that the concept of superdiversity helps to analyze local arrangements of new intersectional (ethnic) migrant diversity. However, on a more structural level, it fails to capture the influence of ethnic-national mainstream thinking about ethnic Others as a whole.
Culturalization of citizenship
Culture has become an important factor in the way how ethnic minority groups in the
Netherlands are viewed by the mainstream. Moreover, cultural otherness in this respect is crucial. Muslim immigrants are thereby increasingly viewed as culturally incompatible towards Dutch values of liberal progressivism, especially in terms of gender roles and sexuality (Slootman and
13
Duyvendak 2015). The way how Dutch citizens think of, speak of, and evaluate ethnic others are strongly influenced by this discourse surrounding aspects of culture, such as religion, language and history. See, for example, the previously mentioned study of identification towards the ethnic mainstream among ethnic minorities in the Netherlands (Huijnk et al. 2015). Primarily, culture becomes an identifier of ethnic in- and exclusion. As a consequence, everyday social interaction between residents in ethnically diverse cities is heavily impacted by a sense of culturized
otherness.
Why is it that culture takes such an important role in ethnic boundaries and ethnic repertoires in the Netherlands? An explanation is offered in the culturalization of citizenship literature. In short, this is about “a process in which what it is to identify as a citizen is less defined in term of civic, political or social rights, and more in terms of adherence to norms, values and cultural practices” (Duyvendak et al. 2016). In the Netherlands, its roots can be found in the development of cultural homogeneity, that traces back to the 1960s. In this period
progressive attitudes on political emancipation, social mobility, secularization, individualization, and cultural pluriformity became widely supported norms (Kennedy 1995), which resulted in a progressive liberal hegemony in the decades hereafter. This led to a ‘progressive consensus’: a national identity marked by uniform views on values and norms (Kešić and Duyvendak 2016). Opinions on issues of gender equality, sexuality, and family are nowhere in Europe valued as uniformly progressive as in the Netherlands. As a result, diverging from these norms – cultural differentiation – is regarded as a growing problem (Slootman and Duyvendak 2015).
Culture can be mobilized in two ways. On the one side, there is a functional way, which includes speaking the country’s dominant language in public, gaining knowledge of its history and traditions as well as its conventions in politics, education and on the labor market (Duyvendak et al. 2016). The other way is an affective approach. The idea here is that ‘feeling rules’ (Hochschild 2003) prescribe how citizens are expected to feel, such as in relation to the nation, or towards their co-citizens. A consequence is that the required feelings of loyalty to the nation-state become a demand from citizens as proof of loyalty. Newcomers – immigrants – who do not manage to feel at home in the Netherlands should either move to another country, or, even ‘better’, move ‘back home, to the country from which they came in the first place’. Being born and raised here does not intrinsically suggest a heightened sense of national affective belonging, in this view. Duyvendak (2011) explains this by an increased focus on ‘local home feelings’, present among natives in both Western Europe and the United States. Nativism, increased awareness of local belonging, and sentiments of nostalgia are ‘onmipresent’ in this regard (Ibid.) and contribute to a politicization of ethnic othering and immigrant integration discourses; a ‘crisis of identities’ if you
14
will. In this study, it is investigated how culture gets employed in the evaluation of ethnic minorities and ethnic belonging among groups that commonly not associate with nativist sentiments.
Within the Dutch discourse on immigrant integration, ethnic minorities are considered objects of problematization. According to Willem Schinkel (2013), this is currently due to the culturalist discourse in governmental agencies. Here, cultural differences are viewed as
incompatible with each other and therefore should not be mixed. As a result, immigrant cultures (and thereby ethnic minorities) are discursively separated from ethnic mainstream culture. An example of this is the aforementioned publication on self-identified distance of ethnic minority groups from the ethnic mainstream. It is important to note that scholars studying ethnic
integration remain critical and skeptical towards the use of terms that emphasize a certain level of
groupness (c.f. Brubaker 2002) and thereby construct social boundaries. Special attention is
therefore paid in this paper to how respondents employ and operationalize a sense of racialized and ethnic groupness.
15
Methods
Using Lamont’s notion of repertoires renders visible dispositions about ethnic minorities as constructed through everyday social interaction and collective representations. However,
dispositions and repertoires are articulated through personal and subjective feelings. Therefore, in this study subjectivity plays an important role methodologically, which is difficult to measure. Q-methodology is therefore used. This method allows to structurally map groups of respondents, and their interconnectedness, on the basis of shared sorting of statements9.
In this study, respondents were asked to sort a set of statements about ethnicity, minorities and integration. The statements represent different ways to discursively evaluate the subject, and are based on the three predetermined normative strands: exclusionary nativist, ethnically neutral, and inclusionary pluralist. Respondents were asked to rank-order the set, from “disagreement” to “agreement”, in a prefixed scheme. All statements were placed on this axis, the q-sample framework, as to allow for systematic ordering (q-sorting), with many statements placed in between, and only a few on the extreme ends (c.f. Brown 1993). At both extreme ends, only one statement was to be placed. At the following end, two statements were placed. For the third level, three statements were placed. And in the middle column, four statements are to be placed (the neutral middle ground). This creates a simple bell-curve shape (see figure 3 in the appendix for a q-sort scheme). As a result, respondents were compelled to order the statements in respect to one another, and were thus forced to show subjective evaluation. This way, primarily, they were bound to evaluate statements on a controversial and sensitive topic – ethnic minorities and integration. If respondents evaluated statements individually (common in standardized surveys) they would be more inclined to opt for a ‘gray middle’, which would result in political correct (and thereby skewed) data. This would not capture shared subjective dispositions on a sensitive topic as well. Moreover, in-depth interviews about the subject, without q-sorting, would also more likely result in less valid and skewed data, since respondents are then directly confronted to express their subjective evaluation about ethnic minority groups to the researcher directly. Also, subjective dispositions are then not recorded in a structural fashion.
Respondents were questioned about the general topic of ethnicity and ethnic belonging before sorting the statements. The sources of their viewpoints on the topic were taken into account also, by questions about personal aspects, such as current or former employment, education level, subjective feelings of local attachment, and the place were respondents grew up. During and after q-sorting, respondents were asked to speak of the difficulties in the sorting process and were asked to elaborate on their opinions about the statements, on ethnicity and its
16
role in everyday life, their overall viewpoints on ethnic minorities, and the sources of these viewpoints.
Data was collected in two ways. First, an online q-sort framework was constructed with Easy HtmlQ software10. Next to the q-sort framework – in which respondents were instructed how q-sorting works and were guided step-by-step through the process – a survey was
implemented within the online component. Here, the topics mentioned above were surveyed as well. The second way of data collection was via face-to-face interviews. Here, respondents completed the q-sort as described above, and during the process were primed to explicate their responses. This way, in-depth qualitative material was recorded, which helped to analyze evaluative considerations by respondents.
The set of statements was constructed from concourse (Brown 1993), various notions which capture how a topic is being discussed and viewed in public. Through a combination of the three normative strands and ten different subtopics, 25 statements were devised. Figure 1 in the appendix shows all statements (in Dutch) and the various subtopics11. Most importantly, the concourse captures common sense repertoires; ways of communicability to speak of ethnic others that are used frequently in daily interactions. This does not imply that the q-sample is eliminated from controversial or offensive statements. Some expressions in the statements are blunt and explicit. This way, respondents were prompted to respond and thereby offering discursive evaluation. The q-sample thus expresses a variety of ethnic repertoires. However, the aim was not to construct a q-sample which secures full representation. In doing so, I follow Steven Brown, political psychologist and q-methodologist, who states that “[…] meanings are not be found solely in the categorical cogitations of the observer, but as well (and even more
importantly) in the reflections of the individual, as he or she sorts the statements in the context of a singular situation” (Brown 1993: 101). Thus, in Q-methods, the aim is not to account for complete representation of a social phenomenon, but to account for different types of evaluation and reflection, as long as the topic under study is broadly covered (c.f. Watts and Stenner 2005: 75).
After fieldwork, principal component analysis (PCA) was used to identify underlying evaluative dimensions among all completed q-sorts. This allowed to uncover which statements were most often selected at the extreme ends; the factors. The factors discovered depict typical
1010 Easy HtmlQ is open source based software, that allows researchers to design and compose their own q-study.
This is done with by html coding, which, on the one hand enables the survey to be filled in online by respondents, and, on the other hand, makes it easy for researchers to implement new parts, questions and design strategies.
11 Mind that not for all topics statements have been devised for the three normative frames. This is because it was
chosen to use a q-sort scheme with 25 statements (see figure 3 in the appendix for an example). This amount was chosen because it allowed to cover a wide variety of evaluative topics without becoming overly complex and manifold for the respondents.
17
ways to identify ethnic others, or, in other words: ethnic repertoires. Moreover, the factors discovered consist of clusters of respondents who think alike on the subject of ethnic minorities. In effect, the main outcome of the q-sorts, the factors, are based on clustered respondents. This method allows to inductively discover the underlying dimensions of clustered subjectivity; it shows correlations between residents who think and evaluate alike about the topic of this study. Subsequent interpretation of these factors was done with the help of interview material.
18
Case selection & data collection
Since different perspectives and conceptions on ethnic belonging are of prime concern in this study, two different urban areas were used to find respondents. These areas differed in the level of ethnic mix (one mainly ethnic Dutch and the other more heterogenous) and socioeconomic background. This latter aspect is taken into consideration, because dispositions and repertoires towards ethnicity and integration are strongly construed by shared class norms. Moreover, by searching for patterns of repertoires among natives, using neighborhoods as starting point to observe ethnic dispositions is relevant, since this allows to “[…] observe patterns of everyday group formation without presuming ex-ante that these necessarily cluster along ethnic lines” (Wimmer 2012: 12).
The city of Amsterdam is relevant, because large parts of the inner city are gentrifying rapidly (roughly demarcated by the beltway A10), due to a combination of large-scale selling of social housing and the renewal of post-war housing blocks by governmental agencies (see Hochstenbach 2017). Still, large parts of the inner city are diverse in both socio-economical dimensions (as a consequence of large scale mixing of public housing with private ownership housing) and ethnic dimensions – a consequence of the status of ‘gateway city’ to incoming immigrants. Since gentrification often results in an increase of residential homogenization (at least in respect to residential income level), the inner city of Amsterdam can be regarded as urban
gentrification frontier (c.f. Smith 1996), in which long-term and less affluent residential groups are
running the risk of being displaced, such as ethnic migrants. As a result, lower class (ethnic) minority groups are being surreptitiously foisted to an influx of (upper) middle class natives and cosmopolitan (international) elites, yielding new boundaries (see for example Alba and Foner 2015; Crul and Mollenkopf 2012; Huijnk et al. 2015). This new diversification could potentially affect the way how these new native gentrifiers think about ethnic others.
The Indische buurt is a neighborhood in the eastern part of the city, and the Musuemkwartier is in the southern part. The Museumkwartier has a share of only 8% non-Western allochthone residents12, and is part of the affluent southern borough. The Indische Buurt, on the other hand, used to be an impoverished neighborhood, with a relative high number of criminality and low residential educational levels. The Indische buurt has since become an
12 Source: Bureau Onderzoek en Statistiek (O&S) Gemeente Amsterdam
(http://www.ois.amsterdam.nl/feiten-en-cijfers/# 22-3-2017). The share of autochthone Dutch in the Museumkwartier 63% and a share of 29% Western allochthones. The difference between ‘Western’ and ‘non-Western’ allochthones is problematic, since it indicates a socioeconomic spatial hierarchy; a Japanese would be considered Western allochthone, whereas a Surinamer is considered a non-Western allochthone. Put simply, non-Western allochthones are ethnic groups commonly associated with ethnic Othering and non-mainstream belonging, and is therefore subject to a debate of integration, while the Western Allochthone is not.
19
ethnically diverse neighborhood, with 50%13 of its residents being of non-Western allochthone ethnic background. In superdiverse terms, this neighborhood can be considered as part of the city with no numerical ethnic majority; it is an example of the trend of minority-majority
urbanization where the old native majority is losing its (numeric) position over a hybrid of ethnic minorities (Crul et al. 2013). Next to being an ethnically diverse part of the city, it has become a frontier for gentrification, that is also occurring at the rest of the inner city. Housing prices and land values are increasing manifold here14. The Museumkwartier on the other hand has since its development in the mid 1800’s been constructed for the affluent and wealthy. Here, the urban landscape is green and lively, large manors and well maintained parks determine the streetscape, and the well-known museums and tourist attractions (such as the Rijksmuseum, the Van Gogh Museum, and the Concerthall) form the heart of its bourgeois and upper class image. This expresses itself in the residential composition, where the ‘allochthone’ group almost exclusively consists of Western allochthones – wealthy expats, who, in colloquial terms, are part of the ethnic mainstream.
In total, 29 residents participated in the research. 15 have been interviewed face-to-face, while 14 participants filled in the online q-sort and survey. Initially, respondents were found via snowballing and personal networks. In terms of accessibility, it turned out that finding
respondents in the Museumkwartier was more difficult than in the Indische buurt using these sampling strategies. Therefore, local community centers were approached. Through these, more respondents were reached and found in the Museumkwartier15. By Q-methodology standards (c.f. Brown, 1993) a respondent group ranging from 20 to 50 people are ideal sizes to draw valid conclusions from. Therefore, the aim was to reach 30 respondents in total. A lower number meant lower chance to find various underlying evaluative dimensions. After testing with the data from 29 respondents, it turned out that six underlying evaluative dimensions could be discovered, and therefore it was decided that more data was not required, since this would take a lot of time to process, while it would arguably not result in more valid results or conclusions. Respondents of all age cohorts participated in the study. However, since both neighborhoods differed in demographic composition, sampling focused to find particular respondents in line with local composition that were of interest for this study. This means that for the Indische buurt mostly young urban professionals participated, and for the Museumkwartier middle-aged and elderly
13 In O&S statistics, the Indische buurt is divided into an Eastern and Western part. The Eastern part has a share of
53% non-Western allochthones, while the Western part has a 47% share.
14 https://www.nrc.nl/nieuws/2016/04/19/straks-wonen-alleen-nog-de-rijken-in-de-stad-1610078-a1032373
(22-3-2017)
15 In particular, I owe gratitude to Annelies Mansfeld, volunteer at local community center ‘De Nieuwe Coenen’,
20
participated. This latter group was focused on, since this residential group makes up the majority of the demographic composition of the area. Figure 2 in the appendix shows an overview of the respondents, except two respondents whose sort did not load significant on any of the factors that were used for analysis (more on this in the next paragraph). Their interview material however was used for the overall interpretation of discursive patterns of ethnic dispositionality.
21
Factors & evaluative patterns
Q-sort data from the face-to-face interviews was put in the online survey as well, so that all q-sort data could be processed and analyzed in the same manner. Analysis was done using KenQ
Analysis16, an open source based web-app to analyze q-sort data, specifically designed for data collected via html code. Here, all 29 individual Q-sorts were coded into a covariance matrix. After this, eight principal components remained which explained the underlying structure of the matrix. Six of these principal components have an eigenvalue greater than 1, which means their individual addition positively attributes to the overall explained variance of the matrix. After Varimax-rotation to maximize variance explained between the extracted factors (c.f. Watts and Stenner 2005), it appeared that these six factors together explain 81% of the total covariance of the matrix (see table 1 below). Each factor consists of particular grouped q-sorts, and therefore illustrate idealized evaluative patterns among clustered respondents (henceforth referred to as composite factors). These idealized evaluative patterns depict distinct ethnic repertoires. Consequently, having found six underlying factors shows that there is not a single (hegemonic) evaluative pattern when sorting statements about ethnic belonging and integration. This indicates that there are multiple ways to talk and think of the subject among ethnic Dutch Amsterdam residents. In line with standard Q-methodological approach were individual respondent’s factor loadings calculated using the Standard Error (SE), with the expression 1/√N, where N, equals the number of statements (25). Thus, the SE is 0.2. Respondents with factor loadings in excess of 2.58(SE) were considered significant at the 0.01 level, and are then illustrative of the relationship between individual q-sort’s and the composite factor type (c.f. Kroesen and Bröer 2009; Rajé 2007). This means that respondents with factor loadings in excess of ±0.51 were considered as representative for the thought pattern present in the factor on which they load. As a result, there are four remaining factors which have two or more respondents that load significantly on it (factors 1, 3, 4 and 6). The composition of these factors are further analyzed below, and account for a total of 68% explained variance of the covariance matrix.
Overall, the composite factors show that respondents positively value statements aimed at inclusivity and ethnic diversity17. Similarly, all factors show that respondents devalue statements with a nativist, restrictive, racist, or discriminative tone towards ethnicity and integration.
16 See https://shawnbanasick.github.io/ken-q-analysis
17 One respondent (Janneke, 85, housewife) however sorted statements somewhat differently; she selected a couple
of statements as positive that were geared towards a more discriminative and nativist tone. For example, she selected statement 5 as almost fully agreeing with (‘I do not associate someone with a dark skin color with being Dutch’) and statement 10 as somewhat agreeing with (‘An allochthone does less or not at all belong to the Dutch culture in my opinion’). She was the only one who loaded significant on factors 2 and 3, which both are not included in the analysis. Therefore, her individual sort does not affect the results further analyzed.
22
However, this aggregated result does not tell us about distinct patterns of an inclusive and progressive orientation on the subject matter. The more interesting question rather is how the four factors differ and on what basis. The interview data shows that there is quite some variance in how ethnic belonging and integration is normatively and conceptually interpreted. This will be further discussed in the following subsections. Interview recordings were transcribed in detail and coded using Atlas.TI. The coding scheme was based on the topiclist (and was hence
predetermined), in order to enable structural analysis of the interview material in line with the discovered evaluative dimensions.
Composite Factor
Evaluative term
(repertoire) Religionᵃ Cultureᵃ
Race & aestheticismᵃ Family & sexualityᵃ Class & criminalityᵃ Explained variance (%) 1 'Relative & conditional' + vs. + + vs. - - vs. -- ++ vs. -- + vs. ++ 24 3 '(de)emphasizing class culture' -- vs. + ++ vs. + + vs. + + vs. -- -- vs. ++ 11 4 'Embracing multiculturalism' - vs. + ++ vs. ++ - vs. - + vs. - - vs. + 14 6 'Considering boundaries & egalitarianism' ++ vs. ++ + vs. ++ + vs. -- + vs. -- + vs. + 19 Table 1: Evaluative patterns and criteria of judgment. The two ratings are about condemnation (negative pole) vs. approval (positive pole) on the five main themes of the statements.
ᵃ --: plays no thematic role, -: plays a small thematic role, +: plays a respective role, ++: plays an important role. Thematic role rating is based on the combined z-scores of the general themes.
Table 1 above also shows the importance of general themes of the three factors. The individual statements are constructed along five general themes (religion, culture, race & aesthetics, family & sexuality, and class & criminality). These general themes are drawn from the specific topics by which the statements were devised. These general themes are further used below to analyze which evaluative aspects are important in a particular repertoire.
23
Repertoire A: Ethnicity as relative and conditional
Factor 1 (see below) positive’s side – ‘agreement’ – consists of statements which frame ethnicity as contingent in relation to other dimensions of difference (such as gender, class, or education level). Among the statements positively valued we see that ethnicity gets deemphasized as a structural marker of identity. Rather, statements on this side view ethnicity as a relatively unimportant dimension of social difference. This notion of ethnic conditionality is for example found in statement 14 (z-score 1,52), which states that ‘allochthones are not more or less civilized than
autochthones’, and statement 1 (z-score 1,42), which that ‘Fundamentalism is found everywhere and is not related to ethnicity’. Although a discursive dichotomy between ethnic Dutch (‘autochthones’) and
ethnic Others (‘allochthones’) is used in these statements, its structural effect is explicitly yielded as irrelevant. Moreover, respondents in this repertoire often decidedly remarked the irrelevance and inappropriateness of this discursive dichotomy. Rhetorically, respondents often compared negative stereotypes of ethnic minority groups to negative aspects of the ethnic majority. As a result, ethnic stereotypes and prejudices are put into perspective and get trivialized. Bob18 (29, producer advertising agency) for example, says the following about a statement which frames allochthones as uncivilized and unmannered (statement 1519):
“The word allochthone there is already odd. It is a generalization, because Asians, Arabs, Africans, South-Americans, they are all put into the same pot. […]. And what the hell is civilized? […] We Dutch eat at 6 o’clock and if somebody is at our door we send him away because we are eating. And we think of ourselves as civilized then. In many other cultures, if you’d do something like that, you’d be considered a rude asshole. That’s two different values next to each other. If people associate being civilized with being hospitable, then autochthones are the least civilized of all!”
A shared aspect among respondents here is that they respond to ethnic essentialism. This is done by comparing normative frames on ethnic minorities to normative frames of the ethnic majority, in order to contest ethnic differences. In doing so, the boundary between the mainstream and minority are seen as contingent and conditional to markers of social identification, other than ethnicity. Consequently, respondents draw attention to a false justification of symbolic groupist boundaries (c.f. Brubaker 2002). This is illustrated by the least preferred statements. Statement 20 (z-score -1,51) notes, for example, that ‘Someone with a foreign background is often poor and profits from
18 All names in this paper are fictional, since a statement of anonymity was provided for all participating residents.
Real names are known to the author.
24
our social provisions’, and statement 7 (z-score -1,58) states that ‘Dutch natives and allochthones cannot have a lasting relationship (such as a marriage)’.
Agreement (z-score) Disagreement (z-score) 14. Allochtonen zij niet meer of minder beschaafd dan
autochtonen (1,52) 20. Iemand met een buitenlandse achtergrond is vaker arm en profiteert van onze voorzieningen (-1,54) 1. Fundamentalisme komt overal voor en staat los van
etnische achtergrond (1,42) 15. Allochtonen zijn vaak ongemanierd en onbeschaafd (-1,47) 16. Niet-christelijke religies passen prima in de
Nederlandse cultuur (1,36)
7. Nederlanders en allochtonen kunnen geen langdurige relatie (zoals bij een huwelijk) hebben.
(-1,34) 24. Allochtonen zijn niet minder of sneller geneigd tot
criminele handelingen dan autochtonen (1,32) 3. Iemand die in de Islam gelooft vormt een hoog risico op een terroristische aanslag (-1, 28) 8. Allochtonen vormen een verrijking voor de
Nederlandse cultuur (1,12) 10. Een allochtoon behoort naar mijn idee minder of niet tot de Nederlandse cultuur (-1,14) 13. Fatsoen heeft niets met integratie te maken.
Asociale migranten verdienen dezelfde rechten als fatsoenlijke autochtonen (1.00)
5. Iemand met een donkere huidskleur associeer ik niet met een Nederlander (-0,85)
Composite factor 1: ‘Relative and conditional’. Highest scoring statements (positively and negatively) in Dutch, including statement number.
An interesting aspect of this composite factor, when looking at the positively valued statements, is the support for a neutral, subtle and pluralistic view on ethnic diversity and ethnic differences. See for example the tendency to present in an empirical and nuanced manner in statements 1, 24, and 13 above. As a result, dimensions of social difference (such as socioeconomic position) are viewed as relative and are strongly disputed as stemming from ethnic background. When respondents here speak about the conditionality of ethnicity or race, they often refer to nativist and populist sentiments – such as the increasing popularity for populist right wing parties20. Most prominently, they discursively oppose these sentiments. Viewing ethnicity as a conditional social dimension is thereby framed as an urgent response to nativist populism, among the respondents. This aspect also figured prominently during interviews with respondents from other repertoires. This will be further discussed at other sections below.
Figure 2 in the appendix presents the significant loading respondents for each composite factor. 11 out of 30 respondents load significant on this factor (which accounts for 24% of the explained variance). Respondents from the Indische buurt are the majority here. Also, most of the respondents are somewhat younger (below 40), apart from two respondents from the
Museumkwartier. Age is thus a shared characteristic of the respondents who correspond with this repertoire. Two out of four respondents from the Museumkwartier are also younger. These two
20 Most prominently, Geert Wilders of the Freedom Party (PVV) and The People Party for Freedom and Democracy
25
respondents stated during the interviews that they did not prefer to be identified as a “typical
resident of Oud-Zuid” (the affluent city borough in which the Museumkwartier lies), since they “are not as snobbish as the rest of their neighbors” (Bob, 29, producer advertising agency), nor interact with
any of them. This is highlighted by the fact that respondent Ella (28, service design consultant) had actually just moved from the Museumkwartier during the time the interview was held (two weeks prior, to be precise), after having lived there most of her life. These two respondents described a lifestyle that they found much more in tune with other parts of the city, that were supposedly “less dull and more vibrant”.
“Actually, I don’t go out much in my own neighborhood. I prefer to go to the Pijp or somewhere else. But I also see the Pijp more as my own neighborhood, since I also work there. Because I know the people at the bars there and in the shops. And this is something I do not have at my own neighborhood.” (Bob)
A relative young age and a preference for an urban hip lifestyle seems to be corresponding traits among respondents here. This is noteworthy, since these shared aspects come up less prominent among respondents in other repertoires. An explanation can be found in the fact that young, upper middle-class people in Amsterdam are more likely to view ethnic differences as contingent on account of having grown up in diverse (semi-)urban areas.As a result, they see an ethnic mix as ‘commonplace’ (Wessendorf 2014), and are drawn to urban areas which exhibit diversity (and are affordable). Diversity, namely, is viewed as an attribute of tolerance and an indicator of being part of the urban frontier. Rudi (27, psychologist) for example comments on the
overrepresentation of ethnic Dutch at community gatherings and local bars in the Indische Buurt: “[It] only attracts… you know… those couples between 25 and 35… […] and it surprised me that they are the
ones, the whites, who visit these things. That is not representative at all for who lives in this street. I was surprised by that.” Thus, her surprise about ethnic division in the local community implies that she views
ethnic diversity as commonplace and ordinary here. Commonplace diversity featured prominently among the interviews with most respondents from the Indische buurt – not only those
respondents who evaluated according to this repertoire.
The themes culture, class and criminality are found among the higher ranked negative statements. Many respondents emphasize a reluctance to symbolically link ethnicity, class and criminality – which is in line with decoupling class and criminality from an ethnic lens, that is: ethnic essentialism. Respondent Hannah (26, PhD candidate) duly notes this when she clarifies why she choose statement 2321 as most preferred:
26 “A non-Dutch background often gets linked to criminality, and as a result some people think that all people with a
non-Dutch background are criminals. Besides, criminality statistics often get explained incorrectly; the numbers show that certain ethnic groups are overrepresented, but that effect disappears when it is controlled for soc[io]-economic status and gender.”
In line with her rhetorical logic, this is a legitimate remark by the respondent. Moreover, she shows how ethnic essentialism can distort questions of causality, when studying ethnic inequalities and structural ethnic differences.
Some respondents in this repertoire however have an inclination to exaggerate the conditionality of ethnic difference. For example, respondent Ella brings up her own Jewish heritage when she reflects on anxiety for Muslim women in public transport because of wearing visible religious symbols (a headscarf): “I can imagine that for Muslim Dutch this is also very annoying.
That you are looked [down] upon because of that. Just as annoying when people ask me if I’m pro Israël. I get to thinking ‘I don’t have anything to do with that!’. So I know how that feels like”.Her argument
decontextualizes ethnic background and downplays different notions of ethnic prejudice; it implies ‘being considered Jewish is just as bad as being considered Muslim’. Symbolic markers of both religions fall outside of the mainstream norm and are therefore observed with suspicion and hostility22. However, remarkably, the respondent indicated she is not an active religious
practitioner, nor does she not wear any visible religious symbols. As a matter of fact, she rather resembles a mainstream ethnic Dutch stereotype: she has blond hair, white skin, blue eyes and wears mainstream fashionable clothing. In order to show emphatic support, she compares different forms of ethnic prejudice, which conflates structural with conditional inequalities between the ethnic mainstream and ethnic minority group. This mechanism has a tendency to overstress the contingency of ethnic bias between conditional and structural disparity, in order to emphasize support for ethnic egalitarianism. When doing so, respondents normatively emphasize the need for ethnic egalitarianism. Dirk, for example, chose statement 723 as least preferred. Upon prompting to reflect on this, he noted:
“Yes, to be honest I don’t know that many examples [of mixed couples], actually. At times, I sit with… with
non-allochthones – autochthones that’s called – so, I don’t know that many examples. […] What I do see, however, are people from America or Canada, who are here for their PhD, or something, and then have a relationship… Is that an allochthone…? […] No, exactly. We also have a Chinese in the family. And my son-in-law is a Spaniard.
22 Unfortunately, hostility towards both outlets of religious affiliation is still very much present in Amsterdam. Jewish
businesses and institutions, for example, are currently under heavy private and police surveillance.
27
Hmm… They are allochthones… But I’m not bothered by it. Although… I don’t see cultural differences with my daughter. […] I think it would be more difficult for me if he was a fundamental protestant from Zeeland, than an integrated Spaniard.”
First, we see that the respondent diligently problematizes the public use of ‘allochthtones’, by stressing that immigrants do not have to be, in essence, culturally different from the ‘us’, natives. However, upon explaining this, he notes that ethnic differences are less important than other structural differences (such as religious), by taking culturally near ethnic Others as examples of groups he identifies more with than culturally distance natives. Overstressing this principle might result in a tendency to take ill regard of structural inequalities and to neglect empirical differences between ethnic minorities and the majority – most eminently when empirical facts disfavor ethnic minority background. It should be noted of course that these respondents do not ignore or dismiss empirical differences as a whole (which is out of the scope of this study). The distinct discursive logic is that empirical differences and facts are cast aside here to argue against ethnic essentialism.
Repertoire B: Ethnicity as (de)emphasizing class culture
The statements of composite factor 3 (see below) are characterized by an emphasis on civic and ethnic egalitarianism, and opposition to nativist and racial exclusion. Figure 2 in the appendix shows that three respondents load significant on this factor. In total, the factor accounts for 11% of the explained variance (see table 1 above). On the positive side of the composite factor we see that statements number 23 (z-score 1,83, ‘Ethnic background gets too much associated with criminality in
the media, which leads to discrimination and racism’) and number 13 (z-score 1,68, ‘Decency does not have anything to do with integration. Antisocial migrants deserve the same rights as decent autochthones’) underscore
the orientation towards egalitarianism. On the negative side, the discursive orientation is
exemplified by statements number 10 (z-score -2,19, ‘An allochthone, in my opinion, does less, or not at
all, belong to the Dutch culture’) and number 5 (z-score -1,35, ‘I do not associate someone with dark skin-color with being Dutch’). Thematically, this illustrated by the extensive focus on race & aestheticism,
