Same same but different. How the play of difference intersects female refugee support workers’ constructions of a common identity

ABSTRACT As in most parts of Europe, the influx of a large number of refugees from the summer of 2015 on has greatly stirred German society and politics and deeply divided the population on how to deal with asylum seekers and immigration. This article enters this controversy by examining supporters committed to the pro-refugee side of the debate – female volunteers in refugee support work. It investigates how the volunteers interpret their relationship with refugees and adopts the social-psychological hypothesis claiming that the construction of a common identity helps decrease prejudice and facilitate intergroup relations. However, I go on to argue that manifestations of difference continuously interfere with these conceptions of ‘sameness’, as difference is fundamental for the construction of identities and meaning in general. Drawing on 22 in-depth interviews with female refugee support workers in Germany, this paper then traces how female volunteers imagine and locate ‘sameness’ and ‘difference’ when working with refugees. The article shows that while outwardly, participants are keen to develop a sense of a shared identity with the refugees, distinctions through power hierarchies or cultural or gender identities disrupt their experience in ambivalent, complex and covert ways.


Introduction
In the last years, cultural difference has become an ever more pervasive theme in the German public and political discourse. Particularly the so-called 'refugee crisis', 1 when vast numbers of refugees arrived in Europe in 2015 (Karakayali, 2018), has confronted Germany with issues of migration, integration and national identity that it sought to evade for so long (Holzberg, Kolbe, & Zaborowski, 2018). Before long, the right-wing party Alternative for Germany capitalised on latent xenophobic attitudes resulting from growing unease with Chancellor Angela Merkel's refugee policy and became the strongest opposition party in 2017. In contrast, an unprecedented wave of empathy with refugees in the summer of 2015 initiated a strong and vocal volunteering movement that set out to counter anti-refugee sentiments and advocates for solidarity (Hamann & Karakayali, 2016).
Thus, German society is currently marked by a deep split between an enthusiastic Willkommenskultur ('welcome culture') and hostile anti-immigration protests, humanitarian empathy and threatening securitisation responses, pointedly formulated: a clash between recognition and rejection of difference (Hamann & Karakayali, 2016;Holzberg et al., 2018). This paper enters this debate by investigating the most crucial incidences of difference that challenge refugee support volunteers, possibly the most ardent supports of pro-refugee solidarity, and thus must be addressed urgently to facilitate harmonious interactions in superdiverse societies. Indeed, Todd May (1997, p. 9) remarked that in times of racism, religious fundamentalism and the rejection of multiculturalism, 'the question of difference and of differences, of how to understand them and of how to respect them, needs to occupy us much more than it has'.
As the overwhelming majority of volunteers in refugee support work in Germany is female (Karakayali & Kleist, 2016) and gender and sexuality often dominated the media discourse (Dietze, 2017), this paper focuses on female volunteers only. More specifically, I aim to identify the contexts and situations where difference influences female volunteers' experience of working with refugees. Conversely, I will examine the discursive devices that volunteers use to establish a sense of sameness. In summary, I will investigate how the dynamics of bonding and boundary construction are expressed and how notions of sameness and difference interact in the everyday experiences and perceptions of the female volunteers.
This engagement with sameness and difference will be underpinned by a range of theoretical approaches, namely social psychology, and poststructuralist, feminist and postcolonial thought. While this method might seem overly eclectic to some, I argue that the complexities of integration and migration research demand a decidedly interdisciplinary research programme. Equally, Stephen Castles (2010Castles ( , p. 1569 insisted that although it is an interdisciplinary field, 'migration research is compartmentalised, with little analytical and methodological collaboration across boundaries'. As a consequence, this 'disciplinary bias' inhibits cross-fertilisation between different theoretical approaches to migration and integration. This paper aims to contribute to a more fruitful dialogue between socialpsychological and sociological approaches of intergroup conflict and cooperation. Based on social psychologists Dovidio, Kawakami, and Gaertner's (2002) 'Common Ingroup Identity Theory', I start from the premise that to reduce prejudice and facilitate the helping relationship, female volunteers construct a common identity with refugees based on commonalities and sameness. Gaertner and Dovidio essentially showed that perceptions of a common identity are integral for harmonious intergroup relations. Notwithstanding, I then argue that difference continually cuts through the research subjects' discursive constructions of sameness, as it is necessary for the production of meaning and identity. Conceptually, poststructuralism highlighted the importance of difference by demonstrating that meaning is constructed relationally and moved by the 'play' of differences (Derrida, 1978). Investigating how difference operates in the social field, feminism and postcolonialism offer enriching perspectives on processes of 'othering' and power relations between dominant and oppressed members of society. Moreover, gender and race/ethnicity represent two axes of differentiation that are foundational for this research and essential for feminism and postcolonialsm.
Consequently, I will first flesh out Gaertner and Dovidio's rationale to posit 'common identity' or sameness as essential for reducing prejudice. However, after criticising the shortcomings of their model I will engage with poststructuralist, feminist and postcolonial accounts of difference and show their relevance for volunteering in refugee support work. After a brief description of the methods used in this study, each part of the article's argument will be underpinned with empirical examples -firstly constructions of sameness, then differentiation mechanisms. The article also relates these results to similar findings by other researchers in the field. I will then conclude with a brief discussion drawing together all strands of the research and suggesting future academic and political pursuits.
Same samethe 'common ingroup identity model' To investigate the cognitive roots of difference and the importance of commonality in social relations, social-psychological concepts prove particularly useful. They mostly centre on the study of prejudice, which Gordon Allport (1954, p. 9) defined as an 'antipathy based upon a faulty and inflexible generalisation […] directed toward a group as a whole, or toward an individual because he is a member of that group'. From the beginning, prejudice was seen as a dangerous social problem to eradicate (Dovidio, 2001). However, Henri Tajfel (1981) showed that categorisation and the resulting generalisation is necessary for cognitive functioning and for establishing a sense of belonging to a group. Together with his student James Turner, he developed Self-Categorization Theory (SCT), which demonstrates that individuals affiliate with groups on different levels of abstraction to sustain their self-concept (Tajfel & Turner, 1979). Depersonalisation is one consequence of this process, which increases the perception of similarities within the ingroup and differences with the outgroup resulting in stereotyping and group polarisation (Turner, 1987). This can lead to social antagonism and ethnocentrism, where individuals upvalue their social group in comparison to others when certain conditions are given, e.g. strong identification with the ingroup and perceived relevance of the outgroup (Tajfel & Turner, 1979). Consequently, social identity theory (SIT) shows that a person's sense of self is influenced by the different groups they are a member of and that to increase self-esteem, a person will try to enhance their group's status by comparing and discriminating against outgroups (Turner & Oakes, 1986). As a result, stereotyping involves considerable emotional engagement and is rather difficult to combat (Tajfel, 1981). Prejudice often accommodates itself in more hidden negative attitudes even when blatant, overt discrimination seems eliminated (Pettigrew & Meertens, 1995). Likewise, many manifestations of difference in the empirical discussion here might emanate from more subtle, covert and unconscious forms of stereotyping.
One of the first and most influential models for prejudice reduction was set forth in Gordon Allport's (1954) 'Intergroup Contact Hypothesis'. To reduce prejudice in intergroup contact, he claimed that four conditions have to be met: equal status within the situation; intergroup cooperation; common goals; and support of authorities, law or custom. Subsequently, Gaertner and Dovidio (2005) introduced the 'Common Ingroup Identity Model' to study new forms of prejudice they termed 'aversive racism'. They argued that a single, more inclusive group identity can recategorise cognitive schemes to extend positive ingroup bias to former outgroup members so that behaviour becomes more favourable. In other words, the essence of Gaertner and Dovidio's argument is that perceptions of 'sameness' through a common identity reduce prejudice and establish peaceful intergroup relations most effectively. This idea rings true with SIT which predicted that identification with a higher-level social identity leads to more inclusivity (Turner & Oakes, 1986). Regarding this article's research environment, a common identity could extend the willingness to help to outgroup members, as other studies have found that commonly, people prefer helping ingroup members (Nadler, 2016). In addition, Kunst, Thomsen, Sam, and Berry (2015) claimed that a common, superordinate group identity also increases integration efforts by the majority group. In my study, I thus hypothesise that constructions of shared identity could be instrumental for facilitating the helping relationship between female volunteers and refugees.
However, I maintain that the construction of a common identity in Gaertner and Dovidio's model cannot proceed undisputedly and smoothly, as manifestations of difference will continuously interfere. Indeed, SIT showed that persons are always members of various social groups that will become salient in different situations and contexts (Ellemers, Spears, & Doosje, 2002;Turner, 1987). Eventually, Dovidio et al. (2002) also conceded that racial and ethnic identities are so strongly rooted in people's self-image that they cannot be easily discarded. Hence, they developed the concept of 'dual identity' that allows group members to hold more than one identity and even generalises beneficial effects to the whole racial or ethnic group. I argue that Gaertner and Dovidio overlooked the substantial power of multiple group identities and how they obstruct the construction and maintenance of a common identity repeatedly. Similarly, Kunst et al. (2015) pointed towards the danger that majority members control the common identity categorisation process and demand assimilation from minority members, to ultimately enhance their own group status. Processes of power and societal discourse are generally disregarded in much of Gaertner and Dovidio's work, and social psychology in general (Branney, 2008;Hook & Howarth, 2005). Mostly relying on experimental settings, social psychology runs the risk to simplify real-life encounters that are influenced by a myriad of social aspects. For these reasons, this article maintains that the Common Ingroup Identity Model, insisting on unconditional 'sameness', is deeply problematic, as intergroup relations are relentlessly cross-cut by difference.

But differentthe unavoidable intrusion of difference
This article defends the view that difference is inevitably bound up with sameness. As Mark Currie (2004, p. 4) remarks, 'a category, a generalisation, a potential collective identity and a potential stereotype […] involve this double process of saming and differentiating, of positing a common essence between members of the set and at the same time marking the differences that give the set its identity.
In the following, I will review the importance of difference in three disciplines the concept most profoundly influenced, specifically poststructuralism, feminism and postcolonialism.
Among the first to elevate the concept of difference to the heart of theory and thus initiating structuralism was the Swiss linguist Ferdinand Saussure (1949). He suggested that the meaning of a sign is constituted by its relation to another sign, often its binary opposition (Currie, 2004). In other words, Saussure showed that 'reality is carved up in various ways according to the manifold patterns of sameness and difference which various languages provide' (Norris, 1982, p. 5). Particularly relevant here, Claude Lévi-Strauss (1970) extended the structure of difference to cultural codes. In this view, binary classifications into insiders and outsiders symbolically construct social order (Norris, 1982;Woodward, 1997), thus laying the foundation for feminist and postcolonial engagements with difference (Currie, 2004).
However, the poststructural movement soon contested Saussure's strong emphasis on a reductionist, abstract and static structure, particularly through Jacques Derrida (Currie, 2004). He insisted that the 'play' of difference continuously moves meaning, which is never fixed and stable (Derrida, 1978). Derrida thus helps understand the manifold and complex ways difference interacts with common identity constructs. This study thus draws on Derrida's neologism différance, alluding to the verbs 'to defer' and 'to differ' (Culler, 1983). Différance captures how meaning is continuously delayed and put off in a concealed 'weave of differences' (Derrida, 1982). Consequently, relations to differing elements, which need not be present, create meaning and refer to past or future 'traces' in a complex chain of differences (Derrida 2002(Derrida /1972. Crucially, Derrida (2002Derrida ( / 1972 showed how binary oppositions form relations of power, where one pole usually dominates the other. The sameness the sign always postulates profoundly represses and excludes the actual difference it depends upon (Currie, 2004), an idea of great significance for this article's contestation of the 'Common Ingroup Identity Theory'. Ultimately, Derrida (2002Derrida ( /1972 suggested that all identities in Western culture are shaped by hierarchical binary oppositions, such as nature/culture or man/woman (Rattansi, 1994). This linguistic operation is 'laden with social and political significance' (Currie, 2004, p. 17), opening up potentials for feminist and postcolonial critique of repressive constructions of the 'other' in social identity processes.
As gender constitutes one of the main axes of difference in this study, feminist engagements with the question of difference are particularly worth looking at. Feminism has been vacillating between the ideas of equality, or sameness, and difference since its inception (Crosby, 1992). On the one hand, feminist activism relies on limitation of sexual difference and solidarity among a unified category of the 'woman' (Rhode, 1990;Offen, 1990). As I will show below, the research subjects use a similar strategy to counter right-wing arguments by stressing the common category 'human'. On the other hand, the difference between male and female, or woman as the 'other' to man, constitutes the basis of feminist theory (Crosby, 1992).
To a large extent, feminism understood the power of difference as processes of 'othering', which are at work here concerning refugees. Simone de Beauvoir (2011Beauvoir ( /1949 pioneered the view that the category 'woman' has been constructed through biology, historical conventions and internalised cultural myths in relation to the male norm, as its 'other', its inessential negative, and is thus culturally produced (Kaufmann, 1986). Later, poststructuralist feminists such as Julia Kristeva and Luce Irigaray criticised this rigid dualism of the sexes by recognising the non-binary, fluid multiplicity of women's identities (Currie, 2004;Offen, 1990). Particularly Irigaray (1994) stressed the positive potential of difference to break through the oppressive structure of patriarchal culture. When analysing female volunteers' engagement with difference, Irigaray's argument helps resist the urge to frame all instances of difference as binary, negative and destructive. Julia Kristeva (1991), then, demonstrated that the 'other' is instrumental in processes of self-construction. As a result, to live peacefully with the difference we must address the stranger within ourselves (Kristeva, 1991). In line with this paper's argument, Kristeva thus revealed that identity and difference are deeply intertwined.
In recent years, the question of difference again unsettled the feminist movement, as women of colour started to question the assumed universal female identity posited by white feminists and identity politics (Crosby, 1992). Black feminists such as Audre Lorde (1984) exposed how white women obscured the experiences of black women and overlooked the interplay of racism and sexism. As a consequence, Kimberle Crenshaw's (1989) concept of 'intersectionality' became a fundamental principle in contemporary feminism. Intersectional approaches pay attention to the complex ways in which different positionalities such as gender, race and class intersect in the construction of identity and discrimination of minorities (Crenshaw, 1989;Lorde, 1984). This article finds particularly valuable the observation that common identity inevitably suppresses other differences, even if this identity may be instrumental for achieving certain (political) goals.
Another mechanism of differentiation between German volunteers and refugees is race, culture and ethnicity. This article thus takes inspiration from postcolonial concerns with the construction of cultural and racial otherness through the dominant 'West' (Hall, 1992). At its heart, postcolonialism considers the 'boundary marking' between self and other as integral to define and reassert personal and collective identities (Rutherford, 1990). Western nations like Germany rely on oftentimes binary classificatory systems (Woodward, 1997) that assign minority groups such as refugees the status of 'constitutive outsides' to mainstream society (Hall, 1996(Hall, , 1997. As Stuart Hall (1992, p. 189) maintains, Western discourse uses 'crude and simplistic distinctions' such as civilisationsavagery, reasonemotion, and culturenature to construct the racial 'other' and assert its own superiority (Hall, 1997). In his study of 'Orientalism', Edward Said (1978) uncovered how colonialists and ethnologists historically codified these discursive strategies in networks of knowledge and social, political and cultural institutions that still operate in the research participants' social environment today. Essentially, Europe still exerts authority over a supposedly 'underdeveloped' Orient that is a source of both fear and pleasure, familiarity and strangeness (Bhabha, 1983). Particularly relevant for my purposes is Nira Yuval-Davis (1997) assertion that women play an essential role in reproducing the nation and its cultural identity as symbolic border guards. Besides, they represent the endangered collectivity in need of defence against the Orientalist 'rapist' (Rattansi, 1994). Lastly, inspired by Derrida's 'play' of difference, postcolonialism also sheds light on the highly ambiguous, fragmented and disordered processes of real-life othering (Bhabha, 1983;Hall, 1996).
In the current German context, a new culturalist racism again lays emphasis on difference to react against the fears and transformations the refugee influx triggered. The 'refugee crisis' only further intensified the fundamental anxieties accompanying inevitable societal change in Germany. Refugees, in particular, rupture our sense of belonging and remind us of the experience of change, loss and disruption in modernity (Rutherford, 2007). An illusion of stable, homogeneous and absolute sameness then seems to ban the threat of difference posed by refugees (Gilroy, 1997). Therefore, 'calculating the relationship between identity and difference, sameness and otherness is an intrinsically political operation' (Gilroy, 2000, p. 99).
Across all these theoretical fields, it becomes clear that difference is inescapable even within constructions of sameness. The empirical discussion below will be guided by a poststructuralist emphasis on the fluid, relational and ambivalent character of difference; feminist challenges to unified common identities that risk overlooking differences within; and postcolonial descriptions of power processes in racialised, culturalised and ethnicised 'othering'.

Methods
This article is based on 22 semi-structured face-to-face interviews with female refugee support work volunteers in four different locations in Germany. To account for differences between urban and rural areas, two fieldwork sites were bigger cities -Berlin and Hamburgand two sites were smaller towns. Moreover, one of the towns is located in East Germany and one in West Germany, as attitudes towards refugees are often considered to be more hostile in East Germany (Karakayali & Kleist, 2016). I obtained access to potential interviewees through contacting refugee support work initiatives, where usually a coordinator distributed the request and organised a number of interviews. In all cases, I did not know the participants before the interviews. Some interviews, however, were organised through common acquaintances. Beforehand, the research design was examined for the need for ethical approval by the institutional ethics commission, but as it did not include vulnerable populations, incentives or overly sensitive topics, this has not been the case. At almost all research sites, I had the opportunity to observe volunteering situations and sites first hand, which allowed me to collect additional data on the volunteering environment. The interviews were carried out between May and July 2018 and lasted one hour on average. Participants were between 20 and 70 years old and came from a range of occupational backgrounds, from students to working professionals and pensioners. Their relationships with refugees varied in depth and length. While some research subjects dealt with a larger number of refugees by providing assistance in a certain domain or looking after groups, others developed close links to particular refugees or refugee families that they assisted in all matters of life. The interview guide included questions on motivations for volunteering and prior expectations, narrative descriptions of the voluntary work including any potential challenges or rewards, and the impact of the voluntary work on the volunteers themselves. It concluded with a section on broader topics such as integration, the current right-wing discourse or gender. While relying on the same interview guide in all interviews, I used a relatively open and flexible interviewing method inspired by Lofland's (1971) 'guided conversations'. This approach allowed me to focus in-depth on aspects important to the participants, or skip or change the order of elements. After obtaining informed consent, all interviews except for one were digitally recorded and transcribed verbatim. Following principles of qualitative content analysis (Mayring, 2000), a code frame based on the overarching categories sameness and difference that link to the research questions was developed. The analysis of the first set of interviews further refined these categories and added a number of subcategories of the main themes that emerged. I then added or removed categories successively in feedback loops when analysing the transcripts of the remaining interviews in the order they were conducted. To ensure the privacy and safety of the participants, this article omits names and other identifying information.

Findings
Same sameabstract and concrete discovery of common identity Taking up Gaertner and Dovidio's (2005) 'Common Identity' theory to reduce prejudice, this section shows the argumentative strategies female volunteers employ to imagine 'sameness'. Altogether, they either construct similarity on a wider, abstract level, or find commonalities in concrete, everyday life situations.
The more theoretical, broader line of reasoning first and foremost refers to a 'common humanity' that unites all human beings and therefore the volunteers and the refugees as well. As SIT predicted, salience of this most abstract social identity can change the perception of former outgroups and induce more prosocial behaviour (Ellemers et al., 2002). On the one hand, the recognition of similarity leads to categorising others as group members, as specified in SCT (Turner, 1987). On the other, the identification as a group then often exaggerates similarities, as the idea of depersonalisation suggests (Turner & Oakes, 1986). These two processes will possibly always interact and be hard to divide empirically. Common humanity and, in connection, universal human rights, were present in all of the 22 interviews and was also mentioned most often out of all 'abstract' themes of sameness. As Dovidio, Love, Schellhaas, and Hewstone (2017) found that majority-group members accentuate commonalities more, this hardly comes as a surprise. For example, one volunteer from Baden-Württemberg believed: We are all born by a mother and at the end of life, we die. 2 Participant 5, small town West Shephard Masocha (2015, pp. 570-71) demonstrated that this linguistic strategy normalises 'asylum seekers as individual human beings […] constituted as just like us' and thus makes them worthy of help and a potential resource. Additionally, participants derived common humanity through recognising that 'they' have the same needs as 'us' be it food, shelter, or love. Ultimately, this linguistic strategy for the volunteers implied respect and tolerance, approaching refugees 'on an equal footing' and a willingness to involve them in any decisions or activities using what one Hamburg volunteer called a 'participatory approach'. Hamann and Karakayali (2016, p. 81) observed a similar 'a culture of recognition of differences' in other 'welcome initiatives' in Germany. Correspondingly, one Berlin volunteer called for a return to, somehow, the interpersonal values that are simply independent of, which religion, which skin colour, whatever […] Not paying so much attention to otherness. Participant 14, Berlin Social justice, solidarity and human rights also featured strongly in appeals for asylum seekers, based on a sense of cosmopolitan responsibility and ideas of interconnection and common identity that are commonly observed in the literature (e.g. Austin & Fozdar, 2018;Nightingale, Quayle, & Muldoon, 2017;Nowicka, Krzyzowski, & Ohm, 2019). Participants drew on notions of equal opportunities, similar values or justice, as one volunteer illustrates: They are humans like you and I and they have the same entitlement to a good and fair life. Like us, the lottery win born in Germany. Participant 18, Hamburg On this level, the participants also solidarised with refugees in a commitment to antiracism and countering (perceived) xenophobic responses by the state or German society, a stance volunteers have also adopted elsewhere (Stivens, 2018) and can be conducive to a shared identity (Drury & Reicher, 2009). This activism may represent a 'shared goal' that could lead to a greater sense of common identity according to Dovidio et al. (2017). Eventually, Hamann and Karakayali (2016) maintained that the volunteer movement played an instrumental role in countering the right-wing protests that emerged in Germany soon after the 'summer of welcome' in 2015.
A final interesting finding emerged when over half of the research subjects referred to children when framing the 'proper' way to deal with cultural difference. Supposedly, children constitute the ideal subjects of integration as they were perceived to not 'see' difference, to embody sameness, and to be more adaptive and innocent. They thus represent the ideal 'deserving migrant' (Vollmer & Karakayali, 2018). When reflecting on integration strategies, one volunteer suggested: You'd simply have to target the most unbiased and somehow toughest members of our society, children […]They don't care about all these differences at all, the only thing that counts is, I don't know, if you are good at football or something like that. Participant 14, Berlin On a more everyday, concrete level, commonalities were established through common interests or activities. Specific group similarities are particularly conducive to establishing a common identity (Gaertner, Dovidio, Anastasio, Bachman, & Rust, 1993). Hence, Whyte, Romme Larsen, and Fog Olwig (2018) suggested that local everyday practices such as football help produce pragmatic 'relations of mutuality' that encourage the incorporation of asylum seekers into the local community. In this study, particularly food and cooking played an important role, but common hobbies like poetry, music or art also served as a basis for a shared identity with particular refugees. After a while, then, close relationships and friendships between volunteers and refugees developed, even after the latter had left the camps and were not directly dependent on help any more. In one instance, a female volunteer and a refugee began a romantic relationship, which bundled a number of strategies to achieve a shared identitycommon interests, adaptation to German everyday life and respect for cultural difference. In sum, friendship and intimacy are hypothesised to provide especially conducive conditions for positive intergroup relations (Pettigrew, Tropp, Wagner, & Christ, 2011).
The most common form, however, of invoking 'sameness' operated through emotions such as fun, trust or empathy. Particularly the latter often served as a bridge to connect with refugees. Indeed, Dovidio et al. (2017) assigned a critical role to affective processes in the Common Ingroup Identity Model, similarly to SIT (Drury & Reicher, 2009). Correspondingly, Holzberg et al. (2018) and Vollmer and Karakayali (2018) observed a frequent appeal to empathy in the German discourse on refugees that both justified humanitarian action and countered impersonal state responses and right-wing hostility.
Well I think that friendliness and that showing, you are not all the same to us and we understand you as far as we can, because nobody can understand how you are feeling, and we try to help you, is very important. So, I try to treat the people like I would like to be treated if I arrived in Syria now and didn't speak a word of Arabic. Participant 3, Berlin A more ambivalent strategy of constructing a common identity demanded that refugees adapt to German life. Taking up German everyday life through school, work, learning the language or developing relationships with Germans, the refugees supposedly become more 'like us'. As observed elsewhere, for women, particularly education and language appear crucial in becoming 'German' (Braun, 2017). Particularly in Germany, this attitude often prevails as integration policy still demands for immigrants to assimilate by giving up their differences (Holzberg et al., 2018). Thus, ultimately the members of the dominant group set the meaning of the common identity for everyone (Frankenberg, 1993).
Finally, a perceived shared identity also sprang from similar experiences such as having experienced migration either themselves or in their family, a hint at common fate and experiences of conflict that according to SIT increase group cohesion (Drury & Reicher, 2009;Tajfel & Turner, 1979). In six cases, participants asserted that they could empathise with refugees because they themselves had immigrated or moved and thus knew how it felt to be perceived as foreign. Interestingly, two volunteers interpreted their experience of moving from East Germany to West Germany or vice versa as a 'migration experience' as well, lending support to Daniel Kubiak and Foroutan's (2018) recent assertion that East Germans' experiences are comparable to migration experiences. So I just turned 18 when the wall came down, I actually lost my home as well and, this is not comparable, but ultimately I also freshly had to find my way in this society and I also know how hard it is. Participant 17, Hamburg Another four volunteers had family that fled from Eastern Europe to Germany after World War Two as 'Vertriebene' (Displaced Persons) and believed that this influenced their motivation and approach to volunteering. Similarly, gender identity provided a frame for a shared identity as well. Many research subjects reported that they got along better with female refugees, for example bonding over the experience of motherhood (Braun, 2017;Stivens, 2018).
But differentthe play of difference Centrally, in this paper, I argue that the above-described strategies to construct a common identity are complicated or even contradicted by the incessant materialisation of difference. The 'colourand power-evasiveness' evident in these certainly well-intended constructions of sameness still preserves power hierarchies, where the speaker can allow difference only when it is safe, nonthreatening or makes the white self feel good (Frankenberg, 1993). The following section will lay out the manifold forms in which difference interrupts feelings of sameness and establishes power hierarchies within refugee support work, be it consciously or unconsciously.

Power and privilege
In all interviews, usually at some point a 'multilayered web of power relations' (Braun, 2017, p. 40) surfaced more or less openly. Particularly common were strategies or expectations that relied on a certain power hierarchy and constructed difference through domination and subjugation. Patronising behaviour witnessed on somebody else or displayed personally constitutes a particularly suitable example, as the German refugee support movement has been criticised for its paternalist tendencies before (Omwenyeke, 2016). Occasionally, participants claimed to know better than the refugees themselves what their needs were, as frequently assumed in humanitarian aid (Kapoor, 2005). In some instances, the participants talked in a condescending tone, infantilised the refugees and adopted the role of the teacher, guide or parent making them 'fit for society'. As Braun (2017) proposed, framing refugees as uncivilised and deficient, in need of charitable European help, plainly connects to colonial discourses of Orientalist othering. In the interviews, however, a paternalist attitude not necessarily included a rejection of refugees, an attitude that Nightingale et al. (2017) called 'ambivalent paternalism'. One volunteer exhibited a very strong patronising attitude, for example claiming that But ultimately, if we want that they don't stay in a state of dependency forever, you actually have to also kick them a bit. Demand something. And also just make them come down to earth a bit. Participant 17, Hamburg Moreover, this attitude often involved misinterpreting the refugees' needs, which a number of participants critically reflected. A common disappointment for the volunteers resulted from refugees rejecting goods and services they organised. For instance, volunteers organised activities such as sports classes or women's circles which were sparsely attended because, as the volunteers speculated later, they did not take into account the actual circumstances and needs of the refugees' everyday lives. One volunteer thus pondered whether integration should also entail the obligation of the German society to assess the needs of refugees, because as she remarked, a refugee might not be too keen on learning the trumpet to join the town band.
Expecting gratitude comprises another crucial mechanism at work in othering processes in refugee support work, which numerous studies observe both in direct interactions and in public discourse on refugees (Braun, 2017;Heins & Unrau, 2018;Holmes & Castañeda, 2016;Vollmer & Karakayali, 2018). Although participants often denied expecting gratitude from refugees, implicitly the theme arose in the majority of the interviews. Interestingly, Heins and Unrau (2018) discovered that Marcel Mauss (1954) thoughts on gift exchange played a critical role in Germany's 'welcome culture'. They interpreted the voluntary help as 'arrival gifts' that implicate expectations of reciprocity to restore the social order between the two groups. Furthermore, gratitude plays an instrumental role in framing the 'deservingness' of the refugee (Holmes & Castañeda, 2016), as a perceived lack of gratitude seemed to make some participants less inclined to help certain refugees. Hence it became apparent that the help the refugees receive is conditional on implicit expectations, such as conforming to the ideal of the deserving migrant, obeying certain rules and not challenging established power hierarchies.
Difference as 'culture' Frequently, participants referred to difference more directly by remarking on the other 'culture' 3 of refugees. Strikingly, apart from a general reference to values and the significance of family, religion emerged most regularly as a marker of culture. As Mahmood Mamdani (2002) illustrated, this essentialist 'culture talk' pits the modern, liberal European against the premodern, fundamentalist Muslim, ultimately extending religious convictions to the secular and political field as well.
Participants often saw the experience of different cultures as a motivation or reward of voluntary work and thus were interested in and fascinated by cultural practices. The volunteers drew parallels to getting to know other cultures on holiday and found they could experience the benefits of travelling right on their doorsteps now. They appreciated what they could learn through intercultural contact, for example on geopolitical issues, the actual experience of refugees in Germany, or religion (Hamann & Karakayali, 2016). One volunteer illustrated: Before, I studied ethnology. And I'm just also interested in how people live outside of Germany and here was the opportunity to get to know very many people from very many different nations, without having to collect money for research expeditions. Participant

21, Hamburg
This strategy of perceiving the colonial figure as the idealised 'exotic' other, however, was instrumental in European self-perceptions and cosmopolitan aspirations throughout history (Loftsdóttir, 2018).
On a wider level, occasionally participants framed refugees as an enrichment for German society because of their personal fascination with the refugees' cultures. As a consequence, they praised the creative potential of difference (Hamann & Karakayali, 2016). Indeed, Serhat Karakayali (2018) asserted that the experience of 'crisis' surrounding refugees in Germany prompted a critical questioning and transformation of established customs. Even so, Kathryn Woodward (1997) remarked that the celebration of difference and diversity still relies on binary oppositions that produce symbolic boundaries of othering. As one volunteer summarised concisely: And then you just always think, we are here on the island of the blissful, the way we do it, the way we live, that's how the world is, that's how it has to be. And that's how it's right. And when you see, other cultures, they also don't just live in misery. And even if it looks very differently there economically, maybe they are still content.
[…] But to realise, we are not the hub of the world, but there are other people as well. If you look on the map, how small Europe is. And how large Africa and how large the Near East is as well. That brings you down to earth a bit, I think. Participant 7, small town West Nonetheless, perceived different cultural practices at times also irritated participants or even were rejected. Particularly attitudes towards childrearing, religious practices, and different perceptions of politeness, reliability and tidiness volunteers claimed to be incapable to comprehend, or they just generally remarked on different 'mentalities' they perceived as difficult. As a result, occasionally some volunteers preferred one group of refugees, for example Arabs over Africans. Furthermore, cultural attitudes towards women constituted a major source of irritation as well, as will be elaborated in the next section.

Gender difference
As predicted in feminist theory, gender identities and attitudes toward the emancipation of women played a central role in most participants' perception of difference. Most commonly, the perceived dominating behaviour of male refugees irritated them profoundly. Some volunteers, for example, told stories where they felt disrespected by male refugees or observed attempts to dictate the arrangement of the relationship either with volunteers or within refugee groups.
There were situations where men were really cocky along the lines of 'she doesn't understand me anyway'. I simply behaved equally rude and dominant and made myself heard. […] And there were some rows showing disrespect for women that I just didn't accept, and made it known. They still don't respect me, but it can't be helped. Participant 4, Berlin Moreover, themost often only assumeddanger of sexual assaults by male refugees emerged as a marker of difference. The participants often related negative attitudes towards gender equality to an implicitly backwards culture. Accordingly, this study again underscores that gender is intrinsically linked to ethnic and racial identities as constitutive categories of social relations and processes of exclusion (Anthias & Yuval-Davis, 1992) in intersections that Joane Nagel (2003) termed 'ethnosexual'. Ruth Frankenberg, 1993 showed how discourse frames the sexuality of racialised men as threatening and enormous and thus constructs them as hypersexualised aggressors, a debate that can be traced back to colonial and orientalist legitimations of European domination (Rattansi, 1994). In fact, women are implicated in and sometimes draw on 'ethnosexual' power mechanisms as well (Dietze, 2017;Frankenberg, 1993). In Germany, these mechanisms have become ubiquitous again after the supposed attacks of Muslim refugees on German women on New Year's Eve 2015/16 in Cologne. Ethnosexist rhetoric depicted particularly the Arab male as a sexual danger to the German woman, and thus the nation, that cannot be tolerated (Dietze, 2017). Interestingly, in many cases the volunteers expected to encounter male dominance among refugees or were concerned about sexual infringements but were quick to add that they never experienced such incidences themselves: At the start I wondered, how do they actually feel about women helping them? Well, they do have a certain pride. And how do they feel, they come into a foreign country and then a woman helps them and that's totally taken for granted. Can they even accept help? And I realise, they can definitely accept it. Participant 8, small town West Thus, ethnosexism serves as a powerful public discourse to cement power relations in German society, without necessarily being grounded in actual facts and experiences. This further lends support to the argument that gender issues are among the most decisive boundary marking mechanisms today (Holzberg et al., 2018).
In another ethnosexual narrative, some volunteers took issue with the oppression of female refugees, often bound up with implicit or explicit intentions to 'free' these women from their paternalistic culture and emancipate them. The gendered and racialised contrast between 'the modern, emancipated female volunteer and the female, oppressed refugee plays a central role' in the German refugee support movement (Braun, 2017, p. 39). Therefore, as highlighted by intersectional feminism, differences between women do significantly impact on wider power relations. Indeed, Lila Abu-Lughod (2002) argued that concerns about Muslim women perceived as needing to be saved by the enlightened West serve as a crucial tool to construct difference and superiority since the 'war on terror' began.
The older boy gives her a hard time.
[…] These rascals. They are inculcated by their father, he wanted to forbid the mother to go to the swimming course. And there we said, Amri, this far and no further, that's your mother, not your wife. Well, from the cradle they are inculcated. Participant 13, small town East Surprisingly in this context and the theoretical frame, some volunteers also used these clashes around gender to question their own prejudice or cultural convictions. Forced marriage, the headscarf or the actual distribution of power within the refugees' family relationships were some examples that prompted participants to reflect on their own habits and convictions. This suggests that even such an emotionally laden and deeply ingrained discourse as gender holds the potential to critically reflect and incorporate difference.
We always say so viciously, forced marriage. […] But then some refugees also said […] our parents know us from an early age, they roughly know what's good for me. That fits. And look at your marriages, do they work better? And then you start to reflect, we say oh these bad guys, and the poor girls. Participant 7, small town West

Conclusion
This article showed that, consistent with the social-psychological Common Ingroup Identity Model, female refugee support work volunteers in Germany established a sense of 'sameness' through both abstract reasoning and everyday connections. Nevertheless, manifestations of difference along the lines of power, culture and gender repeatedly cross-cut these propositions more or less openly, as predicted by poststructuralist, feminist and postcolonial emphasis on difference. Processes of power along various axes such as gender or ethnicity regularly permeated the relationship between volunteers and refugees. Consequently, this article maintains that the construction of a common identity is much more complex and problematic than often assumed, as difference can never be eradicated completely. Thus, I argue that social psychology needs to recognise the power of discourse more firmly and incorporate inevitable expressions of difference in any model of prejudice reduction or intergroup relations.
Furthermore, the question of difference troubles Germany and many other societies to an extent that the resulting social divisions and political conflicts threaten social cohesion and democratic institutions substantially. As global societies will only become more diverse, we urgently require a practicable and sustainable model that facilitates peaceful group relations in a superdiverse world. Most approaches developed in the last decades have been questioned academically or politically, be it assimilation theory, multiculturalism or theories of integration. Hence, this article calls for renewed scholarly attention and political dedication to develop strategies that both take into account the need for a common identity, and the reality of difference and power continuously interfering in social relations. Activists in refugee support work might act as a good starting point for the analysis of everyday negotiations of difference. While being confronted with difference on a daily basis, they still manage to uphold productive and often amicable relationships with their cultural 'other' and advance the positive recognition of diversity on a societal and often political level as well. Future research could thus investigate refugee support work volunteers from the perspective of political strategies on diversity.
Notes 1. Karakayali (2018) illustrates how the state and right-wing populists frame migration movements as 'crisis' to generate a feeling of decline and advance anti-immigrant arguments. As a result, even though it is widely used, the term is avoided in this article when not absolutely necessary. 2. All quotes have been translated from German to English by the author.
3. I am aware that this bounded, territorial, static and homogenising definition of 'culture' is deeply problematic and outdated in social science. Still, this is how the term is still mostly used in public discourse and the participants' narratives.

Disclosure statement
No potential conflict of interest was reported by the author.

Funding
I acknowledge support by the Open Access Publication Fund of Humboldt-Universität zu Berlin.