Studying Ethnic Identification. Tools and Theories

As integration theories shape the lens of many scholars and politicians, this theoretical chapter begins with a brief description of the straight-line assimilation model and segmented assimilation theory. However, as the frame of ‘integration’ and ‘assimilation’ is too narrow to capture the experiences of minority climbers, the focus shifts to literature on ‘ethnic options’, which is better-suited to understanding the individual level. In addition, Bourdieu concepts (such as habitus, field, and symbolic capital) enable us to describe the relation between agency and structure, negotiation of belonging, and the contextual and temporal aspects of individuals’ experiences. Although this study departs from a constructivist perspective, upholding such a perspective in an empirical study is complicated due to the trap of essentialism, on the one hand, and the trap of ambiguity on the other. To avoid these traps, an analytical toolkit is assembled, consisting of five tools: (1) focus on practices; (2) separation of self-identification and external identification; (3) distinction between category and group, the latter implying some level of ‘groupism’; (4) distinction between label and content, which refers to sociocultural practices; (5) and intersectionality. The last section of this chapter critically reflects on the use of common terms such as integration.

. The basic assumptions of this theory resound in the assimilationist discourse that has gained in strength in the last decennia, as we will see in Chap. 4. Straight-line assimilation assumes that immigrants eventually will adapt to their new country. They will become increasingly 'similar' and will eventually be seamlessly incorporated into mainstream society (Alba and Nee 1997, p. 835). Inspired by the famous scheme of Gordon (1964), different domains of assimilation are distinguished, including a structural and a cultural dimension. Even though assimilation in one domain can precede assimilation in another, straight-line assimilation assumes that sooner or later assimilation into the society of residence (read: into the middle class) occurs in all domains. An increased 'national' orientation (an orientation to the society of residence) is considered an unavoidable outcome of immigrant incorporation over time. This is presumed to be accompanied by a gradual loss of 'ethnic' orientations (orientations towards the heritage culture of the immigrants, towards the country of origin and towards coethnics) (Alba and Nee 1997).
I identify two lines of reactions to this straight-line model. The first line challenges the zero-sum assumption that an increasing national identification coincides with a weakening ethnic-minority orientation. Instead, incorporation processes are argued to be bi-dimensional, which means that the ethnic-minority orientation is independent of the national orientation (Hutnik 1991;Berry 1997Berry , 2005. Studies even show that a combination of an ethnic and a national orientation is most beneficial for a person's wellbeing (Berry 1997(Berry , 2005Phinney et al. 2001). Nevertheless, the option for minority groups and individuals to retain their ethnic culture and identity is strongly influenced by the dominant discourses (Berry 1997;Phinney et al. 2001). As we will also see in this study, when immigrants are not allowed to retain their ethnic cultures and identifications while integrating into society, they can feel forced to choose between completely adapting to the society of residence and purely dissociating themselves from society.
The second line of reactions challenges the idea that immigrants necessarily incorporate into the middle class segment of the society of residence (which is the implicit assumption of the straight-line model). Society also has lower class segments. Lower class immigrants, living in lower class neighborhoods especially are prone to integrate into an underclass (Gans 1992;Portes and Zhou 1993) and adopt a rebellious identity that rejects the desirability of schooling and a professional career (Ogbu and Simons 1998). People with a rebellious identity assume that for them school achievement does not lead to upward mobility, and they consider high achievers to be sellouts to oppressive authority (Zhou 1997, p. 987). Such oppositional stances have drastic negative impacts on school performance and socioeconomic status and are likely to result in downward mobility.
These two critiques are combined in the famous model of 'segmented assimilation' developed by Portes and Zhou (1993, see also Portes et al. 2009). Ethnic-minority identification is not seen as a liability for integration but as a resource for upward mobility for many second-generation youth. Their parents' culture and the coethnic community provide access to valuable forms of coethnic capital and protect from discrimination. This theory makes a key contribution to the models of integration by acknowledging the (socioeconomic or structural) value of a coethnic orientation and by debunking the assumption that complete adaptation to the society of residence can be only beneficial.
These integration theories focus on the group level and discuss societal processes in objectivist ways, detached from individual actions and interpretations. This focus on the group level fails to do justice to reality. When groups are taken as units of analysis, variations over time and between individuals are ignored. Crul and Vermeulen (2003) warn of the risk of being too deterministic with premature classification, as adaptation processes can change over time. This change is illustrated by the case of Moroccan immigrants and their offspring: 'The Moroccan community (…) once seemed headed for downward assimilation, but now seems to be rising ' (ibid., p. 983). Additionally, the group approach does not do justice to intragroup differences. Portes et al., take ethnic groups as levels of analysis and in the first place use segmented assimilation theory to explain differences between ethnic groups. Challenges and resources are in the first place treated as group characteristics. Large differences that exist within ethnic groups are therefore largely neglected, and factors that possibly play a role in processes of incorporation, such as gender, class, profession, religion, and local context, are overlooked. For example, daughters of immigrants reach higher levels of education than sons-at least in the United States (Stepick andStepick 2010, p. 1153), but also sometimes encounter lower parental expectations (Thomson andCrul 2007, p. 1034) and more stringent demands with regard to modest behavior (Song 2003, p. 47). As Crul and Vermeulen emphasize, 'different segments of the same group may follow different paths ' (2003, p. 975) (see also for example Zhou and Xiong 2005). It is even possible for an individual's acculturation mode to vary per context and per life phase (Crul and Schneider 2010). The polarization between individuals who are successful and those who lag behind within ethnic groups illustrates that groups are not uniform and exist in various segments (Crul and Doomernik 2003;Gijsberts and Dagevos 2009). By focusing only on Moroccan Dutch and Turkish Dutch with higher education levels, I selected a subsegment of ethnic categories with a particular incorporation characteristic (higher education level). In other words, I build upon this intragroup variation as a given. Yet, I also look beyond the influence of education level. Throughout my study, I also remain open to other variations within the two ethnic groups, within the higher-educated samples, for example in relation to gender.

Ethnic Identification at the Individual Level. Ethnic Options
If we want to attend to processes at the individual level, we can better turn to literature that looks at ethnic identity from another angle. The literature on 'ethnic options' deals with the workings of ethnicity and ethnic identification at the individual level. Instead of assuming that ethnic identification necessarily reflects an encompassing cultural orientation or social cohesion, it seeks other explanations for and roles of ethnic identification. It shifts the focus away from ethnic groups to individuals, and from external structures to the interaction between personal agency and external structures.
Gans developed the idea that persisting ethnic identification does not necessarily reflect an orientation towards coethnics or the 'old ethnic cultures ' (1979, p. 6). He argues that ethnic identification among the third and successive generations-such as the third generation Jews in the United States-does not require cohesive ethnic networks and practiced cultures. This kind of ethnic identification, which Gans calls 'symbolic ethnicity', is not anchored in groups and roles. It is voluntary, without consequential behavioral expectations, and primarily expressive, relying on the use of symbols. In the words of Cornell and Hartmann (1998), we can call 'symbolic ethnicity' a 'thin' identity because it 'organizes relatively little of social life and action' (p. 73). Waters (1990) further illustrated this 'symbolic ethnicity' in her book Ethnic Options. She describes the 'symbolic ethnicity' of descendants of white European Catholic immigrants, which is indeed costless, voluntary, and individualistic. Many of the 'white ethnics' in her study identify in ethnic terms (only) at the moments they wish to; they choose 'to turn their ethnicity on and off at will ' (1996). They are not labeled by others in ethnic terms, and their ethnic background only influences their lives when they want it to.
In later work, Waters (1996) argues that this 'optional ethnicity' is not available for visible minorities that have a socially enforced or imposed identity and are confined to a minority status. She concludes that many ethnic (and racial) minorities do not have these 'ethnic options'. Rumbaut likewise explains that those labeled as 'non-white' face an entirely different situation than descendants of white European immigrants, whose ethnic identifications have gradually become individualized and voluntary (2008). When ethnic differences are socially relevant-for example in the context of the prejudice and discrimination that 'non-white' minorities encounter-this makes individuals self-conscious of their ethnic backgrounds. A likely response is for them to strengthen their ethnic identifications, leading to a 'reactive ethnicity'. Like Waters, Rumbaut argues that it is unlikely that the ethnic identity of the successive generations of 'non-white' ethnic minorities will become optional, voluntary and 'symbolic'.
Song wants to shift from a victimizing perspective on minorities to the acknowledgement of individual agency. She counters the proposition that stigmatized ethnicminority individuals have few or no 'ethnic options ' (2001, 2003). She shows that although the freedom to assert their preferred identity labels wherever or whenever they wish is limited, they have power to influence connotations and meanings associated with their identities. Even though structural forces can be very influential, ethnic minorities are not powerless and do not lack agency in asserting their ethnic identities. They do have ethnic options. In this study I respond to Song's call to acknowledge the agency of individuals with stigmatized minority identities and enhance our understanding of their ethnic options. Later in the book, however, I warn against overestimating this individual agency.

Identity and Social Mobility. Bourdieu's Lens
Bourdieu's academic legacy provides a vocabulary for analyzing the contextual, relational, and temporal nature of identification and linking identification with social mobility. The concepts that form the core of his thinking-habitus, field, capital, and symbolic power-help interpret the empirical results. Bourdieu's theory forms an analytical lens for describing and understanding the dialectic relation between agency and structure; feelings of belonging and discomfort, and strategies of negotiation; and, the contextual, relational, and temporal aspects of these experiences.

Habitus, Practices and 'Objective' Structures
Bourdieu's theory-what he calls a 'theory of practice'-connects the individual's agency with the overarching structures, which Bourdieu sees as originating in each other (see e.g . Bourdieu 1990;Wacquant 2008). At the core of Bourdieu's thinking is that the individual dispositions through which we perceive, judge, and act in the world-what he calls habitus-are formed through the conditions of one's life that shape possibilities and impossibilities. These conditions include profession, income, education level, gender, ancestry, and religion, but also more subjective properties such as feelings of belonging (Bourdieu 1992, p. 225). Habitus shapes-primarily unconsciously-how one thinks, walks, eats, laughs, what one aspires to and estimates as attainable 'for people like us', and what one views as just. In the first instance, habitus is shaped through the primary socialization at home, but formal education also strongly influences the habitus. In short, habitus is formed though societal structures that are coercive but not deterministic. These dispositions form the basis for people's actions, or practices, which in turn continuously (re-)generate the societal structures. These practices do not necessarily only reproduce existing structures; they can also change these social structures.
This interaction between agency and structure is reflected in the stories in this book. These stories delineate how individuals are partly predisposed by their upbringing and other societal forces, but at the same time show that these forces do not necessarily render them powerless and do not make them mechanically reproduce the existing structures and hierarchies. The self-identifications, the articulation of identity labels by the individuals who are central to this book, can be regarded as practices.

Field and Capital
Bourdieu's concepts 'field' and 'capital' also prove useful for describing and interpreting the experiences and practices of this study's participants. Society consists of various spheres, fields, with their own structures, rules, regularities and forms of authority, which are continuously recreated. Examples of fields are art, science, economy, law, and politics (Wacquant 2008). As we will see throughout the book, smaller, more personal social spheres, such as peer networks, can also function as fields. When the rules of the game in a particular field are deeply internalized in the habitus, an individual feels like a fish in the water. Bourdieu also speaks of 'belonging to a field ' (1990, p. 68).
Different fields require different resources, capital, to obtain a certain position or status. Bourdieu distinguishes between various kinds of resources: economic capital (material and financial assets), social capital (direct and indirect social support), cultural capital (including skills, knowledge, and behavioral styles), and symbolic capital (such as prestige and reputation, which have a more implicit value) (Wacquant 2008, p. 268). These resources make those individuals function in particular ways in particular fields, and sometimes change their position in a field or even enter a new field.

Class, Identity, and Belonging
People holding similar positions in certain fields live under similar conditions, have similar experiences, and similar political interests, and possess similar amounts of capital. They have a similar (homologous) habitus. They are also more likely to feel affiliated with each other and have feelings of mutual understanding. In other words, people in similar positions have the same 'social identities' and form 'classes'. They are more likely-but not predestined-to form alliances and groups (Bourdieu 1985).
Although socioeconomic class in Bourdieu's theory functions as the primary determinant of habitus (Reay 2004), Bourdieu's theory is inherently intersectional. As Bourdieu argues, one's habitus and one's position are shaped by a combination of socially-relevant characteristics, including gender and ethnic background (see also Friedman 2016;Reay 2004;Silva 2016).
Bourdieu opposes the idea that individuals are rational actors who consciously pursue economic gains (Bourdieu 1990, p. 50). He considers dignity, or recognition, to be the primary motivation of individuals. The judgement of others impacts feelings of uncertainty, certainty, insecurity and assurance (1977( , p. 238 in Wacquant 2008. '[B]eing granted a name, a place, a function within a group or institution' gives meaning to one's life (Wacquant 2008, p. 265). Apparently, as we will also see in the participants' stories, belonging in a field does not only depend on the habitus, but also on recognition by others.

Symbolic Power, Classification and Struggle
Social recognition is influenced by the dominant worldviews. All people continuously construct views of the social world, including classifications and hierarchies. They try to impose their view of the world and their own place in the world (Bourdieu 1985, p. 727). When a certain view of the social world is internalized, people accept their position and the associated limitations as natural and self-evident ('that's not for the likes of us'), and they respect and strengthen the existing classifications (Bourdieu 1985). This is how ideas about groups contribute to the real existence of groups and their identity.
Although multiple views on the world exist, some views have more legitimacy than others. Symbolic power is the power to make other people adopt a certain worldview. It refers to the ability to influence the rules of the game, establish authority, and make people perceive existing classifications and hierarchies as legitimate and fair. Symbolic power is based on status and prestige, on symbolic capital. Credentials such as certificates, diplomas, and titles provide individuals with certain amounts of symbolic power and are institutionalized instruments in this system (Bourdieu 1985(Bourdieu , 1990. People and institutions with more symbolic power have 'the power to name and to make-exist by virtue of naming' (Bourdieu 1985, p. 729;italics in original). They have the power to assign people to certain categories, telling them who they are, what they have to be and what they have done (Bourdieu 1989, p. 22). According to Bourdieu, the state holds 'the monopoly of legitimate symbolic violence', exerted through official discourse, institutionalized classifications, administrative taxonomies, and professional titles and diplomas (1985, p. 732). This is also the case in relation to ethnic categories and ethnic hierarchies, which are created and strengthened by immigration policies, census-taking, redistribution of resources, affirmative action, and rules for political access (Brubaker and Cooper 2000;Nagel 1994;Wimmer 2008). This is exactly what Dutch politicians, media, and governmental institutions do when they consistently label citizens with immigrant backgrounds as 'allochthonous' (literally 'not from this soil'; implying a lesser belonging) and base reports about society on the statistical categories 'allochtonous' and 'authochtonous'. As we will see in Chap. 4, this contrived distinction between those who fully belong and those who do not is legitimized through the idea that geographical rootedness forms the ground for entitlements, and through a persistent portrayal of 'Others' as traditional, orthodox, and backward, and inherently different from the 'real' Dutch.
The existence of differing worldviews produces an ongoing struggle between those who want to preserve the status quo and those who pursue change (Bourdieu 1989, p. 21). This struggle contains individual struggles in everyday life as well as collective struggles in the political domain. Bourdieu speaks of the incessant work of categorization, which is performed 'at every moment of ordinary existence, in the struggles in which agents clash over the meaning of the social world and of their position within it, the meaning of their social identity ' (1985, p. 729). According to Bourdieu, one of the aims of scientific work is to understand the principles of the classification strategies though which individuals conserve or modify the world and their own position in this world (1985, p. 734). This is exactly what this research aims to do. I study the assertion of self-identification (practices) as expression and negotiation of categorizations, as it takes place in relation to mechanisms of inclusion and exclusion, and feelings of belonging.

Temporality
Although Bourdieu's thinking is extremely useful for understanding continuity and social reproduction, contrary to what is often assumed, his theory also leaves room for change, improvisation, struggle, and individual agency (see also Friedman 2016;Jo 2013;Reay 2004;Sweetman 2003). Bourdieu focuses on practices and how they enfold in time and thus does not regard the world as static; he sees practice as 'inseparable from temporality ' (1990, p. 81). He emphasizes that the habitus functions in a 'non-mechanical' way (1990, p. 55). When the habitus is not aligned with the field-because of the individual's (social) mobility into a new field or new position in the same field, or because of a change of the field itself-a change in the habitus is a near-inevitable consequence. His assertion that people constantly try to impose their view on the world in an ongoing struggle for symbolic power further illustrates the dynamic character of his theory (1985, p. 727). Wacquant even concludes: 'Struggle, not "reproduction", is the master metaphor at the core of his thought ' (2008, p. 264).

Impact of Social Mobility: Cleft Habitus, Navigating Multiple Fields
Bourdieu's concepts have been widely applied in studies about the impact of social mobility on individuals (see for example Abrahams and Ingram 2013;Byrom and Lightfoot 2012;Carter 2003;Friedman 2016;Horvat 2003;Jo 2013;Lee and Kramer 2013;Reay et al. 2009Reay et al. , 2010Schneider and Lang 2014;Sweetman 2003). Bourdieu argues that, in the case of social mobility, the habitus is no longer completely aligned with the new position in the field and the individual can feel like a fish out of water, both in the 'old' lower-class field of the home and in the new field of higher education or middle class. The mismatch between the new position and the old habitus can lead to discomfort and feelings of insecurity and alienation because the sense of self is 'torn by contradiction and internal division' (Bourdieu 2004, p. 109, in Friedman 2016. Bourdieu calls this a 'habitus clivé', or 'cleft habitus'. This situation often leads to a 'painful and disorienting struggle to reconstruct one's sense of place within social space' (Bourdieu 1999in Friedman 2016 and to changes in the habitus. Empirical studies illustrate the challenges, struggles, and negotiations that result from social mobility (see for example Friedman 2016).
Other authors nuance this problematic picture. Although none of them claim that dealing with social mobility is easy and painless, their articles focus on the agency of the social climbers and explore how social climbers deal with the challenges of social mobility (Byrom and Lightfoot 2012;Lee and Kramer 2013;Reay et al. 2010). Some social climbers resolve the tension between the two fields by only choosing for one field as the primary anchor of identification and social belonging. However, many climbers try to reconcile the two fields and in both fields negotiate their belonging (Abrahams and Ingram 2013). In the home field, for example, they use specific language to avoid coming across as a snob or to express skepticism about their new advantages (Lee and Kramer 2013). They try to maintain a personal identity that is in line with their lower class background (Lee and Kramer 2013).
These processes and strategies are influenced by ethnicity. Carter's research (2003) shows that African American climbers negotiate their belonging in the home field by showing 'ethnic authenticity' (comparable to what we maybe can call 'working class authenticity'). Social recognition in the home field is related to the use of 'black cultural capital'. In Chap. 7 we will see that this 'ethnic authenticity' intersects with class.
Some authors assert that this switching between positions and fields results in a habitus with a characteristic element: reflexivity. This 'habitual reflexivity' is a particular type of habitus (Sweetman 2003). The idea of the reflexive habitus resolves the dilemma of how habitus-presented as striving for confirmation and continuity-can be connected with flexibility. This reflexive habitus develops especially when individuals have dealt with mismatch since early childhood. Many social climbers had already, in early childhood, been forced to deal with a mismatch between their own individual habitus-characterized by curiosity, ambition and discipline-and their low-class environment. These early experiences of discomfort, of feeling like a fish out of water and of knowing that nothing can be taken for granted, can lead to flexible and reflexive dispositions (Reay et al. 2009), a 'chameleon habitus' (Abrahams and Ingram 2013).
This chameleon habitus is more than a habitus that consists of separate parts that are aligned with separate fields. Inspired by Bhabha (1994), Abrahams and Ingram call this chameleon habitus a 'third space'. This third space refers to a separate place, separate from the two fields, 'from which to navigate and reconcile the apparent incommensurability of the two fields' (Abrahams and Ingram 2013, par. 4.21). The description of Reay et al. elucidates how the reflexive habitus functions as a third space (2009). They describe that their informants have a critically reflective stance on the academic field and its hegemony, combined with a strong commitment to this field. The reflexive habitus functions as an overarching, binding layer, a third space, which-through reflexivity, awareness and constant deliberation-helps them navigate the multiple different fields.

Ethnicity as Social Construction
The various academic fields, which build on divergent views on ethnicity and identity, are based on different assumptions about the inevitable or 'substantial' character of ethnicity and ethnic or national identification. These ontological perspectives influence how, as researchers, we observe the world. In this section, I discuss the potential and drawbacks of a constructivist and an objectivist stance on ethnicity and on identity in a broader sense. I explain my preference for a constructivist perspective.

The Academic Consensus: Ethnicity as Social Construction
In everyday life, there is little recognition of the dynamic aspects of social identities, and of ethnic identity in particular. It is commonly assumed that people with the same 'ethnicity' are highly similar to one another and are bound together-that they have shared behaviors, emotions, morals, skills, and so forth-solely because they share a certain characteristic such as (some part of) their descent. This is also a common view in the Netherlands, as I will describe in Chap. 4. Ethnic identity is seen as an indisputable, primordial characteristic, something a person or a group 'just has' by nature, and which is unchangeable. This objectivist or essentialist view is based on the assumption that all ethnic groups have static cultures that are inherently different from each other. Groups and cultures are seen as monolithic, meaning they are taken to be 'internally homogeneous and externally bounded' (Brubaker 2002, p. 164).
In academia, however, it is common to consider social identities, including ethnicity, as being continuously created through people's actions. Rather than viewing ethnic identities as self-evident products of naturally-existing ethnic groups, ethnic identities are seen as emerging from boundaries that are constructed between (imagined) social groups. These constructed boundaries make people see themselves as members of groups and are recognized as such by others. These ethnic groups are then tagged with ethnic labels and defined in cultural terms. Particular cultural elements are selected to demarcate the ethnic boundaries, which are consequently defined in terms of language, religion, customs, rituals, moralities, or ideologies; or, more specifically, in terms of dress, food, gestures, space, or gender roles (Jenkins 2008a, p. 79, 111, Nagel 1994. In summary, 'culture' provides the meaning and content of ethnicity in society (Nagel 1994, p. 162). In the social sciences, this constructivist view, which is traced back to a paper of Norwegian anthropologist Fredrik Barth in (1969), is the dominant perspective (Baumann and Sunier 1995;Jenkins 2008a). Bader states: 'We are all constructivists now ' (2001, p. 251).
A constructivist view enables us to identify variations in the meaning of a specific concept; to unravel the mechanisms that bring about these meanings; and to identify power imbalances between different stakeholders that factor into the process of meaning-making. (See for example Stuart Hall's description of the evolving meaning of the category 'black' [1991].) This view can be 'liberating', as it provides a tool for unmasking power inequalities that underlie the roles that are attributed to people. Societal roles and positions are often regarded as inescapable because they are based on classifications and stereotypes that are seen as natural. An example is the presumed 'natural' tendency of 'the woman' to take care of the children that 'inevitably' leads to underrepresentation of women in high-profile public functions. It can be liberating to understand how such stereotypical ideas emerge, how and why these images are fed and spread, and how individuals deal with these images. In the words of Schulz: viewing phenomena as social constructions and unmasking authority 'contributes to our understanding of social and political processes through which individuals and groups locate themselves in relation to others, understand themselves, and define their possibilities ' (1998' ( , p. 336 in Song 2003. A constructivist view allows us to examine how identities in general, and ethnic identities in particular, are constructed and reconstructed over time.

Is There Really a Constructivist Consensus? Two Traps
This post-Barth constructivist consensus is in reality not beyond dispute. Critique is voiced on the constructivist stance itself, on the easy dismissal of the relevance of objectivist perspectives, and on the actual application of the constructivist perspective in much of the scholarly literature.

Constructivism Versus Objectivism
Objectivism 'asserts that social phenomena and their meanings have an existence that is independent of social actors' (Bryman 2001, p. 17). This implies the existence of a reality that is external to the people involved. Instead of perceiving culture and cultural meaning as shaped and reshaped by people, the objectivist view regards culture in a reified way: as existing 'repositories of widely shared values and customs into which people are socialized', existing independently of these people with an 'almost tangible reality of its own' (ibid., p. 17). Perspectives like primordialism (the idea that a phenomenon is a primal given) and essentialism (the idea that a phenomenon has a real and static 'essence', independent of people and contexts) are related to objectivist thinking. In the social sciences, objectivist perspectives are often dismissed as 'essentialist', which has strong normative connotations. Nobody proudly claims to be an 'essentialist' (Phillips 2010). 'Most people who use it use it as a slur word, intended to put down the opposition' (Hacking 1999, p. 17). Essentialism is associated with racism and is 'increasingly employed as a term of criticism' (Verkuyten 2005, p. 125).
However, simply dismissing essentialism makes it easy to overlook the possible value and relevance of essentialist and objectivist thinking and the possible downsides of a constructivist perspective. Firstly, according to some, constructivism does not lead to adequate descriptions of social phenomena. As constructivism tends to emphasize processual, unstable, instrumental, and political aspects, and tries to explain fluctuations in definitions of ethnicity and ethnic identification, it can fall short in accounting for the social relevance and tangible consequences of certain phenomena (Liebkind 1992, p. 154). Brubaker and Cooper suggest that a constructivist conception of 'identity' makes it hard to account for crystallized self-understandings, the sometimes coercive force of external identifications, the singular understandings, unitary groups, and the power of identity politics (2000, p. 1). For example, when we emphasize the fluid character of ethnicity, we risk underestimating the inevitability of ethnic classification and its consequences for certain individuals or certain groups, as well as possible practical, social, and mental benefits of ethnic identification. Bader even accuses constructivists of seeing phenomena as not 'real', as only 'abstract and purely analytical notions', as 'fictions ' (2001, p. 254).
I disagree with this simplified presentation of constructivism. A constructivist perspective does not necessarily imply that a phenomenon or concept is indefinitely or individually malleable. As explained by Bourdieu: that the dynamics of everyday life are a consequence of human action does not make social structures less 'real', rigid and durable. A constructivist view does not deny the concreteness of situations, but sees it as a consequence of human action rather than as an external, lawful given.
Secondly, some argue that essentialist perspectives are portrayed too negatively. Verkuyten criticizes the widespread idea that essentialist views on culture are generally oppressive (Verkuyten 2005, Chap. 5). He argues that essentialist reasoning, which presumes the inevitable incompatibility of two cultures, is not only used by ethnic majorities to oppress minorities, but is also used by ethnic minorities in emancipatory ways; for example when used in protests against assimilationist demands and in claims for recognition of cultural differences and identities-as 'strategic essentialism' (Spivak 1988).
Thirdly, some argue that essentialist views are evident in many political, social, and psychological processes. Brubaker explains that essentialism forms the base of politics: 'Reifying groups is precisely what ethnopolitical entrepreneurs are in the business of doing', as in politics, the 'political fiction of the unified groups' is important, and these unified groups are partly evoked by talking as if they exist (2002, p. 167). In similar vein, Phillips (2010) argues that essentialism is a common way of thinking in many social and political contexts. Furthermore, both Phillips and Verkuyten explain that essentialist thinking is a key psychological mechanism because it helps people process complex information by providing a firm understanding of the world (Medin 1989in Verkuyten 2005Phillips 2010). An essentialist perspective can also contribute to a secure sense of identity as people who strongly identify with a group are generally more inclined to see their group as essentially homogenous and distinctive (Verkuyten 2005, p. 142).
Even though Verkuyten and Phillips convincingly show the political, social, and psychological importance of essentialist reasoning, their argument is not necessarily convincing from an ontological perspective. The argument that essentialist thinking is very common in practice does not prove that a social phenomenon is essentialist in its character. When a phenomenon-such as a specific ethnic group-is dynamic over time in shape and meaning, and varies per context, it is still possible for people to view it in a reified way, perceiving it to be static, with a natural essence. Brubaker emphasizes the important distinction between the realm of practice and the realm of analysis. He argues that a social phenomenon, such as a reified idea about ethnic identity or an ethnic group, is 'a key part of what we want to explain, not what we want to explain things with; it belongs to our empirical data, not to our analytical toolkit' (2002, p. 165). We should carefully distinguish between a 'category of practice', which refers to categories as used in everyday social experience by 'ordinary social actors', and a 'category of analysis' as used by the analyst (Brubaker and Cooper 2000, p. 4). It is up to us, researchers, to study why ethnicity is presented as a reified given, rather than adopting this view as our own.

Constructivism Implemented: Traps of Essentialism and Ambiguity
This brings us to a discussion about how constructivism is applied or 'implemented' in the realm of analysis. How is a constructivist perspective applied in academic studies in our 'categories of analysis'? It appears that constructivism is easier said than done. I identify two traps: the 'essentialist trap' and the 'ambiguity trap'.
In 1999, Hacking already complained that social construction was frayed. In his view, the numerous studies tagged as the 'social construction of…' were more cases of 'bandwagon jumping' than anything actually related to social construction (1999, p. 35). Correspondingly, Brubaker and Cooper argue that the academic consensus has turned into 'clichéd constructivism' (2000, p. 11), as they 'often find an uneasy amalgam of constructivist language and essentialist argumentation' (ibid., p. 6). This is not equally the case for all categorizations. For example, in the case of class, there has been a remarked change. Nowadays, the term 'working class' can hardly be used without quotation marks and 'the working class' is seldom regarded as a homogenous entity and an autonomous actor (Brubaker 2002). By contrast, ethnicity is often considered in 'groupist' terms as Brubaker et al. explain: Despite the constructivist stance that has come to prevail in sophisticated studies of ethnicity, everyday talk, policy analysis, media reporting, and even much ostensibly constructivist academic writing about ethnicity remain informed by 'groupism': by the tendency to take discrete, sharply differentiated, internally homogeneous, and externally bounded groups as basic constituents of social life, chief protagonists of social conflicts and fundamental units of social analysis. Ethnic groups, races, and nations continue to be treated as things-in-theworld, as real, substantial entities with their own cultures, their own identities and their own interests. (…) the social and cultural world is represented in groupist terms as a multichrome mosaic of monochrome racial, ethnic, or cultural blocks. (Brubaker et al. 2004, p. 45) The groupist perspective is illustrated by the fact that ethnic categories ('classes' in Bourdieu's terms) are often called ethnic 'groups', which suggests a certain level of uniformity and/or cohesion and interaction that is not necessarily present (Goffman 1990(Goffman [1963Brubaker 2002). Also the widespread use of the terms 'ingroup ' and 'outgroup', and 'bonding' and 'bridging' in reference to entire ethnic categories reflects the prevalence of groupist thinking. In academic literature as well as in common integration discourses, the social context of ethnic minorities is commonly discussed in dichotomous terms. The social context of ethnic minorities is divided into a so-called ethnic ingroup and a so-called ethnic outgroup. These terms are derived from social identity theory, developed by Tajfel and Turner (1979). This theory postulates that the act of self-categorization in itself leads to 'ingroup' favoritism (with an emphasis on sameness, belonging, and consonance) and 'outgroup' derogation (with an emphasis on distinction, non-belonging, and dissonance). 1 This is even the case when this categorization is totally arbitrary and, for example, based on the toss of a coin. Connection of the terms 'ingroup' and 'outgroup' to ethnic categories ignores the condition of self-categorization and imposes the idea that ethnic minorities inevitably have a coethnic favoritism and always distinguish themselves from people with a different ethnic background. Applying the basic idea of social identity theory to entire ethnic categories blindly presupposes a self-categorization in solely ethnic-minority terms. A similar faltering line of thought underlies the common use of the concepts 'bonding' and 'bridging' (like for example by Putnam 2000). These terms are often used to denote coethnic and interethnic relations. This implies that ethnic groups are necessarily cohesive, that people with the same ethnic background are naturally more similar than people with different ethnic backgrounds, and that coethnics are naturally drawn towards each other; it presupposes groupness.
'Groupist' scholars do not deliberately take a 'groupist' (or objectivist or essentialist) stance, but apparently it is difficult to avoid such ways of thinking. The line between concepts as they are used as 'categories of practice' and as 'categories of analysis' is often blurred (Brubaker 2013, p. 5). This confusion of categories of analysis and categories of practice in empirical studies leads scholars to speak as if such internally homogeneous, externally bounded groups exist (ibid., p. 5). Thus, instead of studying how meanings originate and shift in practice, they often contribute to reifications of categories such as 'Moroccans', 'Turks' or Muslims. Apparently, there is what I call an 'essentialist trap'. This is primarily a problem in empirical studies, which explains the gap between 'the grand theoretical work that asks us to rethink everything on the basis of no serious empirical data and the empirical work that keep churning out the same banalities as it did twenty years ago' (Baumann 1999, p. 143). A groupist stance easily leads to an 'ethnic lens', to an overestimation of the relevance of ethnicity resulting from the narrow focus of the researcher. This ethnic 1 The terms consonant and dissonant are also employed in segmented assimilation theory but with a different meaning. There, consonance and dissonance specifically refer to how acculturation processes of children relate to the acculturation processes of the parents. The proposition is that the social mobility of the children is hampered in a situation of dissonance, i.e., when the acculturation process of the parents severely lags behind the acculturation of the children, which is supposed to often be the case in low-capital minority groups (Portes and Rumbaut 2001;Portes et al. 2009). lens prevails in the now-common approach of taking the 'ethnic group' as a unit of analysis (Glick Schiller 2008;Glick Schiller et al. 2006). Baumann explains this relation: (…) yet, when it comes to empirical studies of ethnicity, most students are still given topics such as 'The Turks in Berlin', 'The Berbers in Paris', or 'The Sikhs in New York'. The focus is on a national, ethnic, or religious minority as if anyone could know in advance how this minority is bounded and which processes proceed inside and which outside that assumed community. We have, in effect, created a little island; we study this island, and we usually conclude that the island is, in so many ways, an island. (Baumann 1999, p. 145-146) Taking 'ethnic groups' as units of analysis tends to contribute to reification of the ethnic categories because 'ethnicity' becomes the primary lens of observation and interpretation. Apparently, this often occurs in the social sciences. Carter and Fenton (2009) even speak about a broad 'ethnicization of sociology' in which ethnic and national identities dominate our thinking. Fox and Jones argue that this preoccupation with ethnicity, particularly in the scholarly field of migration, has given ethnicity 'a fixity in both popular and scholarly imagination that is at odds with its contingent and socially constructed nature' (2013, p. 385). This preoccupation not only leads to an overstating of the concreteness of ethnic groups, but also to the mobilization of an ethnic explanatory framework at the expense of alternative and possibly more relevant explanations for social phenomena, such as for example poverty (Brubaker 2013). Such an ethnic lens can obscure underlying mechanisms, such as educational values and social support (Carter and Fenton 2009). Furthermore, one risks overlooking external mechanisms, which can lead to blaming the victim. For example, social processes governing the socioeconomic status of immigrants often are more strongly influenced by immigrant status, social origins and education, and market dynamics, than by culture and identity (Brubaker 2013, p. 5). In similar vein, by referring to people by their ethnic background (for example as 'Moroccans') and using ethnicity as a central analytical term, their ethnic background is emphasized, together with their supposed cultural Otherness. Although my study is yet another study that focuses on ethnic identity, in which I selected the participants based on their ethnic-minority background, I try to avoid the ethnic lens and examine rather than assume the relevance of ethnicity and ethnic identification.
A second criticism of the work of constructivist scholars is that their concepts are often vague and ambiguous. This point of critique is roughly the opposite of the previous accusation of essentialism (even though the critics are the same). Hacking argues that social constructivist claims are often confusing because the phenomena studied are multifaceted and therefore complex (1999). This makes it hard to use concepts in clear and unambiguous ways, and leads to what I call the 'trap of ambiguity' of constructivist thinking. As an example, Hacking raises the issue of the construction of gender: does the social construction of gender refer to the idea that gendered people exist, to the gendered people themselves, to the language, institutions, human bodies or perhaps to 'the experience' of being female?
Similarly, Brubaker and Cooper argue that the analytical use of 'identity' is often characterized by ambiguity (2000). 'Identity', just like 'ethnicity', is used in divergent ways. The terms refer to both structural characteristics and individual affiliations, and to both external labeling and self-understandings. They have contradictory connotations, as they sometimes imply stability and fundamental sameness, and sometimes seem to reject notions of basic sameness (ibid., p. 10); sometimes they refer to tight 'groupness' and sometimes to loose affiliations. A term cannot be used to distinguish between different phenomena and variations if these are all captured by the same term. In other words, the language of identity and ethnicity 'blurs what needs to be kept distinct', making these concepts ill-suited to do the analytical work (ibid., p. 27). Apparently, the complexity of phenomena that we label 'identity' and 'ethnicity' leads to the overuse and dilution of these analytical concepts. These concepts then become unfit for analytical purposes, despite their importance for everyday politics-as categories of practice.
In summary, across the social sciences there is a broad preference for constructivist thinking. Nevertheless, a few scholars highlight the practical importance of objectivist and essentialist thinking, pointing to its prominence in political, social, and psychological practices. However, the practical relevance of objectivism does not mean that we need to adopt this as our analytical perspective. For analytical purposes, a constructivist approach might still be preferred. Yet, adopting a constructivist approach appears to be easier said than done, particularly in empirical studies. On the one hand, we find the essentialist trap. Scholars often unintentionally end up reinforcing essentialist notions of ethnicity and ethnic groups, particularly when they take ethnic categories as units of analysis. On the other hand, we find the trap of ambiguity. The multifaceted character of social phenomena makes it hard to analyze these phenomena in unambiguous ways. In the following section, I explain how I try to avoid these two traps and discuss how I employ (ethnic and national) identity as an analytical concept.
My description of my research theme exposes my ontological position. The use of phrases like 'what ethnicity means for the higher-educated second generation' or 'practices of identification' show that my point of departure is constructivist. However, in response to the argument that a constructivist perspective risks overlooking the social relevance of a phenomenon, I argue that starting from a constructivist perspective does not preclude finding that a phenomenon is rather static, rigid, and uniform. This perspective does not predetermine that the phenomenon as it emerges from the empirical study is entirely dynamic, malleable, or social in character. At the same time, I believe that starting from an objectivist position presents the considerable risk of overlooking existing variations and dynamics that could (then falsely) lead to the conclusion that a phenomenon is objectivist in character. As I see it, the risk of drawing wrong conclusions about the character of a phenomenon is smaller when we start from a constructivist perspective. 2

Studying Ethnic Identification: Analytical Toolkit
One of the major challenges in studying second-generation Moroccan-Dutch and Turkish-Dutch individuals, whom I selected because of their ethnic background (not for their presumed evident ethnic identification), is avoiding an essentialist and groupist perspective and an ethnic lens. Based on various suggestions of migration and identity scholars, Fox and Jones propose a three-part approach for avoiding this trap (2013). Two of these solutions are methodological. The first solution is to avoid sampling on an ethnic dependent variable. For example, if the study only sampled individuals with strong ethnic identifications, we would overlook the existence of a potentially broad range of ethnic identifications. The second suggestion is to start from 'the everyday' as a means for observing practices beyond 'ethnic practices'. By expanding our focus beyond these practices, we can observe the possible relevance of non-ethnic dimensions and specify rather than infer the relevance of ethnicity. In Chap. 3, I explain how my research design complies with both of these suggestions. I explain how I try to avoid selecting participants based on their ethnic identification and avoid centering the interviews on the theme of ethnicity through an initial focus on the participants' trajectories of social mobility and their relationships with various social others. The third solution echoes Brubaker's warning about not conflating the category of analysis with the category of practice. We should not use 'ethnicity' as it is used in practice to analytically explain what 'ethnicity' means in everyday life. Instead, we should try to explain how ethnicity becomes socially meaningful. This is the challenge I take up here. In order to avoid this conflation of 'ethnicity' as a category of analysis and 'ethnicity' as a category of practice, as well as avoid the trap of ambiguity, I assembled an analytical toolkit consisting of five conceptual tools, which I present here.
The first analytical tool I use is to think in practices. Combining the lines of Bourdieu and Barth, instead of thinking in terms of fixed notions of 'identity' and 'ethnicity', the analytical focus is on practices of identification. This is an oft-used way of avoiding the essentialist trap, although most scholars speak of 'processes' of identification (see for example Baumann 1999;Brubaker and Cooper 2000;Giddens 1991;Hall 1991;Jenkins 2008b). Identification then, also in terms of ethnicity, is viewed as something that is not necessarily static over time and over situations but is 'done' in situations in which people concretely act and interact with each other. I specifically study the labels that individuals use in reference to themselves. I research the articulation of labels such as 'Moroccan', 'Turkish' or 'Dutch' and the reasons and mechanisms behind it. Thinking in practices enables us to recognize that individual identifications do not simply 'exist', but come into being and are asserted in various ways. Instead of assuming that people with ethnic-minority backgrounds identify in ethnic terms simply because they have certain ancestors and 'have' a specific 'culture', we can study what makes individuals emphasize a certain aspect of their identity, whether it be in ethnic or national terms or any other way. This way of thinking enables us to study whether and how various dimensions of identification vary in salience and meaning by context and over time. It enables us to analyze how and to what extent identifications are contextual and dynamic.
Because the aim is to research what 'ethnicity' means for the research participants, I distinguish between 'ethnic background' and 'ethnicity'. 'Ethnic background' exclusively refers to the fact that the participant's parents are born in Morocco, Turkey, or the Netherlands. 'Ethnicity' refers to the meaning that ethnic identity, or the ethnic label, has for individuals. It is studied through the participants' experiences in order to learn about the value and meaning that ethnicity has in their lives. This distinction is not applied to other terms, such as coethnic (referring to people with the same ethnic background) or ethnic categories (referring to people with a certain ethnic background). When I speak about Moroccan-Dutch or Turkish-Dutch people, these labels refer to their ethnic background and not to their self-identification.
The focus on practices of identification enables us to recognize the interactional aspect of identifications. According to Barth, the idea that ethnicity depends both on how people see themselves as groups and how others see them is central to the emergence of ethnic boundaries. We also saw that social belonging is an important aspect of Bourdieu's framework. Before Barth, Cooley had already introduced his metaphor of the 'looking glass self' to describe the social nature of one's self-perception as an interaction between how one sees oneself and how (he thinks) others see him ( in Jenkins 2008b. Hence, the second tool in the toolkit is the distinction between self -identification and external identification, which refers, respectively, to the self-ascription of identity and identity-ascription by others (see for example Jenkins 2008a, b;Penninx 1988;Song 2003;Verkuyten 2005). I use 'categorization' or 'labeling' as synonyms for external identification.
When we attend to the influence of external ascription on one's self-identification, we can unmask power relations and the effects of external identification on individuals. The influence of external factors and underlying power hierarchies are easily overlooked when the focus is exclusively on processes of self-identification, as is often the case in the anthropological literature (Jenkins 2008a, pp. 57-58). In studies on integration and assimilation, it is often the other way around, as-because of a focus on the group level-these often focus primarily on structural factors and thereby overlook individual self-determination or individual agency (Song 2003, p. 8). By distinguishing self-identification from external identification, and by focusing on the interaction between both mechanisms, we can research the influence of structural factors on individual agency.
The third tool in the analytical toolkit is the distinction between category and group; or 'class' and group, as Bourdieu would say. This is a way to avoid groupist thinking, which is based on reifying assumptions about categories (Brubaker and Cooper 2000;Goffman 1990Goffman [1963Jenkins 2008a, b;Verkuyten 2005, p. 56). As we saw, groupist views assume that (ethnic) categories are highly homogeneous and cohesive. The term 'group' in itself elicits groupist thinking because even in its most minimal definition, 'group' implies a sense of affiliation, 'a capacity for collective action', and 'a stable and embracing pattern of mutual interaction' (Goffman 1990(Goffman [1963, p. 36, see also Carter and Fenton 2009). Groupness is something we should study instead of presuppose (Brubaker 2002;Brubaker and Cooper 2000).
The analytical distinction between category and group is required for analyzing the ways in which ethnicity can exist and 'work' without the existence of ethnic groups as substantial groups or entities. The concept of category 'can help us envision ethnicity without groups' (Brubaker 2002, p. 170) and enable us to analyze the relations between categories and groups (and categories and identifications, and identifications and groups). Category refers to an individual's characteristics that determine their position in a system of classification. Having at least one parent who is born in Morocco means that one's 'ethnicity' as a category is 'Moroccan'; I use ethnic background. I do not automatically assume anything about one's self-identification, external identification, one's social network or behavior solely on the basis of one's ethnic background. In the remainder of the book, I have often used 'category' where 'group' is common, except where this leads to an overly abstract reference to concrete people (citizens, neighborhood residents, or respondents). Here, I use 'group'.
The fourth analytical tool is the analytical distinction between label and content. A few authors elaborate on this distinction. For example, Verkuyten talks about label and the 'cultural component ' (2005, p. 46). Jenkins uses the terms 'nominal identity ' and 'virtual identity' (2008a, p. 76). 3 The distinction of label and content enables us to study what identification with a certain label means for an individual. It enables us to bring the 'cultural stuff' into the analysis, just as Jenkins pleads for, but as a topic of analysis and not as a self-evident aspect of one's self-identification (ibid., p. 172). The term 'identification' in this book solely refers to one's self-identification with a label-specifically to the practice of articulating a label in reference to oneself-without any broader connotations or automatic assumptions. Chap. 5 is built on this distinction and analyses the association between identification with the ethnic label and content.
The last tool to help avoid treating ethnicity in an essentialist way is the idea of intersectionality. Intersectional thinking is based on the idea that the various dimensions of a person's identity do not work separately but shape one another. So, how a female Muslim experiences her gender is not similar to the experiences of all women, and how she experiences being a Muslim is not similar to the experiences of all Muslims. Rather, her experiences as a woman are shaped by the fact that she is a Muslim, and her experiences as a Muslim are influenced by the fact that she is a woman. This idea that social divisions are interconnected has existed for a long time, but it was not labeled until Crenshaw (1989) coined the term 'intersectionality' to direct attention to the specific experiences of black women. Their experiences were misrecognized because gender inequality and racial inequality were only recognized as separate forms of oppression. Although the idea of intersectionality has become a central philosophy in (black) women's studies-in critique of the presentation of the experiences of white middle class feminists as the female experience (McCall 2005)-it has hardly extended beyond women's studies and 'black' women (Nash 2008, p. 4). It has not been employed to correct essentializing tendencies in studies on ethnic groups.
For example, in the comprehensive overview works on (ethnic) identity of Jenkins (2008a, b) and Verkuyten (2005), and in the critical articles of Brubaker discussed above, this view is not discussed, let alone promoted for its de-essentializing merits. These authors only discuss the multifaceted (or 'hyphenated') character of identity in the context of a combination of ethnic and national dimensions. I consider this a missed opportunity. The acknowledgement of identity as a complex phenomenon and the decomposition of binary ways of thinking make intersectional thinking highly effective in avoiding groupist thinking (see also Anthias 2013). Nash places the call to broaden the application of intersectional thinking: 'If (…) intersectionality purports to provide a general tool that enables scholars to uncover the workings of identity, intersectionality scholarship must begin to broaden its reach to theorize an array of subject experience(s). ' (2008, p. 10). However, we must be careful not to slip into new forms of essentialism by replacing larger homogenizing categories with slightly smaller homogenizing categories and by looking at an 'intersection' as two unproblematic social sections coming together, creating a new 'groupist' category (Desmond andEmirbayer 2009 in Fox andJones 2013, p. 390;Anthias 2013). Rather, we should acknowledge the socially and historically constructed character of the relevant social categories. Subsequently, we should go beyond noting a dependency, and dissect the process of intersecting. My study responds to these calls by focusing on the higher-educated men and women with an ethnic-minority background, trying to disclose how their experiences in relation to their ethnic-minority background are shaped by their class position.

What's in the Name?
The term integration is not unproblematic, as I mentioned before. Nor are the terms 'ethnic groups' and 'natives'. The term integration (and assimilation) is frequently used to denote general processes of incorporation, both in academic and practical settings, but it is hardly ever accompanied by an explicit definition. When terms such as integration or 'ethnic group' are used as concepts of analysis without explicit definitions, the distinction with concepts of practice is unclear. This is highly problematic because when these terms are used in daily practice, they are loaded with normative connotations and contribute to power inequalities. When we fail to define the terms 'integration' and 'assimilation' as concepts of analysis and/or blindly take over their usage in the practical realm, we do not unmask these underlying mechanisms and might even contribute to power inequalities. I argue here that the terms 'integration', 'assimilation', 'ethnic group', and 'natives' are unfit as concepts of analysis, and describe how I avoid the use of these terms.
In the language of politics and daily life in the Netherlands, 'integration' is seen as something inherently 'good'. When something is framed in terms of integration, it is positive and beyond dispute (Veldboer and Duyvendak 2001, p. 17). Apparently, the fact that 'integration' can be oppressive for some, and might reduce individual freedom or the freedom of minority groups, is often ignored. Furthermore, mainstream discourses ignore the wide variation in the meanings of integration, which sometimes even contradict each other. Neither is it debated whether 'integration' is beneficial in all cases. The uncritical use of the term 'assimilation' to neutrally denote processes of incorporation is even more problematic because of the strong normative and ideological usage that appears both in political discourses (at least in the Netherlands) and in (older) scholarly literature (Alba and Nee 1997, p. 827) that presents 'assimilation'-understood as complete adaptation to the society of residence and a loss of ethnic traits-as a desired outcome.
This point leads to another argument, namely that the analytical use of 'integration' without an explicit definition contributes to existing power imbalances between ethnic minorities and the ethnic majority. In scholarship, integration and assimilation are often regarded as the blurring of the boundary between an ethnic-minority category and the majority category, for example, as Alba and Nee define: 'the decline of an ethnic distinction and its corollary cultural and social differences ' (2003, p. 11). This sounds balanced, as in principle both the minority and majority can contribute to the decline of ethnic dimensions. However, when describing such boundary decline, it is easy to overlook the fact that in the current neoliberal political climate, it is often only the minority individuals who are held accountable to their integration processes, thus for this decline. Hardly any demands are placed on natives for closing the gap (Veldboer and Duyvendak 2001;Veldboer et al. 2007). For example, residential concentration of ethnic minorities is evaluated differently than the concentration of 'natives'. It is frowned upon when ethnic minorities establish their own organizations, whereas the existence of completely 'white' organizations in ethnically-diverse societies like the Netherlands are rarely problematized.
This focus on the minority individual is partly a consequence of the 'neutrality' of members of the ethnic-majority category, of the so-called mainstream. The ethnic majority is seen as 'neutral', without ethnicity. This is reflected in the use of 'ethnic groups' to refer to ethnic-minority categories, which suggests that the ethnic-majority category is not an ethnic category. This 'neutral' status means that the majority's ideas, beliefs and attitudes are taken as self-evident and therefore function as the undebated yardstick against which minorities are held. This also means that the majority identity is strongly normalizing and minority categories are usually on the 'losing side' (Liebkind 1992, p. 156). The mere differentiation between a 'minority' and a 'majority' reflects a 'normative hierarchy which combines the idea of status and legitimacy, of numbers and of deviation from the norm' (ibid., p. 156). The fact that the majority category is unlabeled and unmarked implies that the majority does not form an explicit category, thus masking its position of power and contributing to the power imbalance. After all, it is hard to make the standards and power inequalities explicit when these cannot be questioned and are taken for granted (Verkuyten 2005, p. 59;Wekker 1996, p. 73). Furthermore, when a category is unmarked, individuals are judged on their individual merit. When one belongs to a marked category, suddenly (s)he is assumed to be similar to co-categorical individuals and not similar to inter-categorical individuals (Captain and Ghorashi 2001). There is much less awareness of the multiple sides and qualities of minority individuals. For example, whereas an ethnic-Dutch individual can strive to be a talented volleyball player, a nice neighbor, or a capable mayor, a Moroccan-Dutch individual is often primarily judged as a 'Moroccan' volleyball player, a 'Moroccan' neighbor, a 'Moroccan' mayor and measured primarily against stereotypical images of 'Moroccans'.
Another consequence of the 'neutrality' of the 'mainstream' is that the yardstick for integration is undefined and unclear, and hence is not questioned. Based on Alba and Nee (2003), Lindo argues that 'the measuring stick, the point of reference, is often indicated with vague vocabulary like "the society in general", "the mainstream" or "the middle class" ' (2005, p. 12). In the Netherlands, with regard to sociocultural aspects in the last years, the character of 'the' Dutch identity has been strongly debated. This has not lead to unambiguous results, but nevertheless the proposition that such homogeneous Dutch identity does not exist has been loudly opposed. 4 In socioeconomic terms, there is an implicit demand to integrate into the 'middle class', whereby the integration of (children of) immigrants is evaluated against the yardstick of the Dutch average. This is also the most common usage of socioeconomic integration in the literature on ethnicity and immigrant incorporation, which is 'equated with attainment of average or above average socioeconomic standing', rather than compared with the current statuses of population segments with similar socioeconomic backgrounds (Alba and Nee 1997, p. 835). How appropriate is it to use the socioeconomic population average as a frame of reference to assess the 'integration' of (children of) Moroccan and Turkish immigrants who once came to the Netherlands to work in low paid jobs (as, for example, is done in the Integration Report 2009 by Gijsberts and Dagevos)? Natives of the lower classes do not have to meet such expectations, as Thomson andCrul remark (2007, p. 1026): 'We rarely, if ever, hear that sections of the indigenous population are not integrated despite their own experience of poverty and deprivation'. Conversely, why then are immigrants with lower socioeconomic statuses considered as being 'not integrated'? This is a relevant issue, as matters of 'integration' in the Netherlands are often exclusively discussed for categories with below-average socioeconomic statuses at their moments of arrival. These are primarily people from Morocco, Turkey, Suriname, and Curacao who are aggregated under the label 'non-western immigrants'. 5 Comparing children of the former 'guest workers' with the ethnic-Dutch lower classes results in a fairer assessment of processes of incorporation. It appears that immigrant children on average do not lag behind, but actually have relatively high achievements not only in the Netherlands (see e.g. Stevens et al. 2014;Gracia et al. 2014), but also elsewhere (Waldinger and Feliciano 2004;Kasinitz et al. 2008;Stepick and Stepick 2010).
These problems regarding to the term 'integration', and the portrayal of the second-generation Moroccan Dutch and Turkish Dutch as newcomers who are in a process of 'integrating' make the concept 'integration' unfit to use as a concept of analysis in this research. Instead, I prefer to use the more neutral term 'incorporation' and more explicit terms such as 'socioeconomic mobility', 'socioeconomic advancement' or 'sociocultural adaptation'. I use these in a descriptive rather than a normative sense. I try to describe processes and mechanisms that are at play (including discursive and normative mechanisms), but I refrain from taking such a normative stance myself; I am not suggesting that children of immigrants should show advancement and adaptation.
The reflection on the middle class as the implicit norm is also important because it exposes a circularity in the Dutch construction of 'ethnic-minority categories' and their evaluation as not being fully integrated, at least in socioeconomic terms. This is a consequence of the selective application of 'ethnic-minority group' to categories that in general have a lower socioeconomic status (Rath 1991). In the Netherlands, the term 'ethnic minority' primarily refers to people with a non-western background who have lower socioeconomic positions. Strangely enough, immigrants from the U.S., Germany or Japan are not generally labeled as 'ethnic minorities', and these categories are not central to integration debates and integration policies. And, whereas people from the (former) colonies in the Caribbean fall under this category of ethnic minorities, immigrants from the former Dutch-Indies/Indonesia are categorized as 'western immigrants'. This shows that-at least in Dutch society-the perceived distance to the standard of the mainstream rather than a certain ethnic background leads to categorization as 'ethnic'. The fact that 'ethnic minorities' do not reach the standard, then, is not because of their ethnic and immigrant background, but is simply because of their categorization as (ethnic) minority. In my study, I do not refrain from using the term 'ethnic minority'. However, I hope to contribute to a more nuanced use of this terminology by emphasizing the intra-categorical variation and by focusing on those in higher socioeconomic positions.
Furthermore, I use labels that do not obscure the ethnicity of the ethnic majority. I use 'ethnic-minority group/category' and 'ethnic-minority identity' where terms like 'ethnic group' and 'ethnic identity' are commonly used. I also use 'ethnic majority' or 'ethnic Dutch' to refer to people whose parents are born in the Netherlands. I refrain from using the term 'native' to refer to ethnic Dutch, as this term is part of the nativist discourse and incorrectly excludes the children of immigrants who are born in the Netherlands and who therefore are also 'native' to the Netherlands. It is important to note that thinking in 'majority' or 'established' in some cases is obsolete. In many major cities, young children of the second generation are often more established in the cities than ethnic Dutch (Crul and Schneider 2010); their parents have lived there for a long time, and they themselves are born and raised there, whereas the ethnic-Dutch children often have parents who moved there from other parts of the country more recently. Sometimes, the ethnic majority is not even a majority anymore, at least in numbers, as has recently become reality in Amsterdam.
I also refrain from the use of the term 'host country'. This visitor analogy invokes images of temporality and suggests that, as 'guests', immigrants and their offspring should be modest and grateful for the offered hospitality (Ghorashi 2014). Instead, I use society of residence.

Summary
In my attempt to understand the ethnic and national identifications of secondgeneration social climbers, I first turned to the literature on processes of immigrant incorporation. I explained that the famous models of straight-line assimilation theory and segmented assimilation theory are not fully adequate to understand ethnic identification at the individual level. Literature on ethnic options and the theory developed by Bourdieu provide better angles for exploring individual dynamics. In the coming chapters, I zoom in on the experiences of higher-educated second-generation Moroccan Dutch and Turkish Dutch. I examine how these second-generation climbers identify in ethnic terms and what this means to them (Chap. 5), why they identify in certain terms in specific social contexts (Chap. 6), and how these identifications develop over time (Chap. 7).
At the end of this book, I will be able to reflect on the meaning of ethnicity and ethnic identification in daily life for the participants based on my empirical results. As we have seen in this theoretical chapter, phenomena such as ethnicity and identity can be viewed from different ontological perspectives. In the social sciences, there is a common consensus on the idea that ethnic and national identities are social constructs. The State plays an important role in the creation of such categories and the (self-) labeling of people.
One of the problems with the constructivist perspective is that the focus on the dynamic, variable, and contextual character of ethnicity could make us underestimate the importance and substance of ethnicity in daily life. Nevertheless, the substantial effects of a phenomenon, such as ethnicity in daily life, do not mean we have to take objectivism as our analytical point of departure. A constructivist perspective does not necessarily preclude finding that a phenomenon is static, constant, and tangible. In my view, this makes a constructivist perspective the most suitable for academic study.
However, applying a constructivist perspective appears to be easier said than done. On the one hand, there is the trap of essentialism. Often, constructivist scholars unintentionally reproduce and contribute to essentialist views because they fail to distinguish their 'categories of analysis' from 'categories of practice' and unreflectively employ reified ideas of ethnicity and ethnic groups as these are used in political and general discourses. Furthermore, they often apply an 'ethnic lens' to a study, which can contribute to groupist thinking and to the (possibly inappropriate) prevalence of ethnic explanations. On the other hand we have the trap of ambiguity.
The complex character of social phenomena makes it a real challenge to employ concepts in unambiguous ways, and the use of abstract concepts (such as identity and ethnicity) often confuses or conflates aspects that need to be kept distinct.
In order to avoid these two traps, I assembled an analytical toolkit containing five tools. The first is thinking in practices of identification, instead of thinking in terms of 'identity'. The second is the distinction between self -identification and external identification (being labeled or categorized by others), which enables the exposure of power inequalities and a study of the interaction between external structures and individual agency. In order to avoid groupist assumptions, the third tool is the separation of category and group. A social category (such as Moroccan Dutch) does not necessarily comprise a group, and members of the same category are not necessarily connected by sameness, interaction, and solidarity. Levels of groupness should be studied rather than assumed. The fourth tool is the distinction between label and content. Use of an identity label does not necessarily reflect an underlying set of cultural norms and practices. Also here, the connection between label and what it means for individuals should be studied rather than assumed. The last analytical tool is the idea of intersectionality. The idea that the various social dimensions of a person shape each other helps prevent groupist thinking and makes us attentive to intracategorical variations. My focus on higher-educated members of the second generation enables me to explore the intersectionality of class (education) and ethnicity.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.