Hidden Dialogicality among Eritrean Refugees in the Civic Integration Process in The Netherlands

Abstract

In the debate about the integration of refugees in European nation-states, several scholars have recently noted a so-called ‘agency paradox’ (Klaver et al., 2018), which draws a contrast between, on the one hand, the initiative and inventiveness of refugees during their flight and, on the other hand, an allegedly lethargic attitude upon arrival in a new country. In this paper, we will unpack this paradox in a critical analysis of civic integration in the Netherlands. We shall write ‘against integration’ (cf. Rytter, 2019) by recovering the muted voices of a group of Eritrean refugees whose future depends on the obligation to complete civic integration exams. Using Dialogical Self Theory (DST), we will demonstrate that a seemingly passive attitude among migrants must not be interpreted as a lack of agency. Instead, we will show that their agency is silenced and subordinated by the integration process, but that they are simultaneously involved in a hidden dialogue with the new society in order to negotiate a future in which their multiple identifications as both Eritrean and Dutch are recognized. By facilitating the recovery and recognition of the muted voices of Eritrean refugees, Dialogical Self Theory contributes to their empowerment.

Introduction

‘Integration’ has become a vexed issue in European nation-states since the beginning of the new millennium. It is a response to rising immigration that according to some has become unsustainable, especially because it is believed to threaten the national identity of European countries. International developments, such as the 9/11 attacks in 2001 in the United States reinvigorated such beliefs. As a consequence, the political discourse about immigration and integration has changed radically across the continent. In the Netherlands, a series of proposals and measures have been introduced to significantly diminish immigration figures, while mandatory forms of integration for newcomers and oldcomers alike were introduced. In these measures, the emphasis is more on the cultural adaptation of immigrants to Dutch society, on their duty to learn Dutch and to embrace Dutch values, in spite of widespread confusion and discussion about what these are or should be. Although this policy is discussed under the rubric of ‘integration’, in practice it has effectively been narrowed down to complete assimilation.

The problematic premises of integration policies have been criticized at great length in the social sciences (e.g. Loch, 2014; Crul, 2016; Rytter, 2019; Schinkel, 2017; Vertovec, 2020). The notion of ‘integration’ is considered part of nationalist discourses that evoke a dichotomy between ‘us’ and ‘them’ since it is assumed that we know who we are and what constitutes our sameness, because we also think that we know who we are not and what constitutes our difference from others (Gullestad, 2006). Steven Vertovec (2011) has drawn attention to the ‘imaginary’ dimension of this taken for granted kind of ‘sameness’, which in his view is founded upon a conceptual triad consisting of national identities, borders and political orders. Not only is some sense of cultural identity presumed to characterize a nation, but this identity is believed to be contiguous with a territory, demarcated by a border, within which laws and a moral economy underpin a specific social, cultural and political order (ibid., 245). This order is conceived to be different from other orders outside the border, and it is built on a sense of collective identity that is constructed and reproduced through an arrangement of cultural representations and narratives. Thus, nation, culture and identity are woven together in complex ways in a range of national traditions and give rise to different anxieties about so-called ‘strangers’ who in this type of thinking are categorically excluded from standard constructions of the nation-state. Indeed, in nationalist discourses that are based on bounded models of the state, immigrants are principally considered as alien disturbers of an orderly working of society (Wimmer & Schiller, 2002).

This deconstruction of new nationalist discourses that emerged in recent decades has simultaneously triggered a fundamental critique of the notion of ‘integration’ as outdated, modernist, neocolonial and oppressive. The Dutch sociologist Willem Schinkel (2017, 2018, p. 5) especially has offered one of the most compelling criticisms of the concept by reviewing the organicist representation of society in integration discourses as ‘pure’ and ‘pristine’, while it is also coded as uniquely ‘white’. Since society in practice is not such an ‘unscathed whole’ (ibid., p. 2) as integration policies assume, it remains fundamentally unclear into what exactly immigrants should integrate (Swinkels, 2019). For that reason, too, the notion of ‘integration’ may be seen as a survival of late nineteenth century nationalism, which in the 21st century has become an ‘absurd anachronism’ because in current circumstances that are characterized by globalization and unprecedented migration it can never be substantiated empirically (Favell, 2019, p. 3). In order to move beyond the normative ideal of integration, Schinkel (2018) proposes a radical review of bounded notions of society by paying more attention to human commonalities across different socio-cultural ecologies.

Against the backdrop of the critique of integration in nationalist discourses, in this article we also aim at shifting the focus from differences to similarities between nationals and immigrants. We shall do so by recovering the voice of a group of Eritrean refugees, who like all immigrants are effectively voiceless in the integration debate in which multiplicity is not tolerated. The absence of immigrant input in the integration debate has some scholars even led to draw attention to a marked contrast between the pro-active and inventive attitude of migrants, refugees especially, and their seemingly passive, if not lethargic attitude upon arrival in their new host country. Klaver et al. (2018, p. 39) have described this apparent contradiction as the ‘agency paradox’, which they explain with reference to the mandatory civic integration procedures that are so demanding and stressful that they leave little to no room to reach the required goal of self-reliance.

Although we concur with the view that the impact of civic integration processes hampers the neoliberal aim to make migrants independent, we argue that it would be wrong to conclude that immigrants are passive, let alone submissive to the goal of ‘integration’ as understood and implemented in the Netherlands. Instead, using Dialogical Self Theory (DST) we shall show that immigrants are deeply involved in internal negotiations between their cultural identity and the prescribed national identity in the integration process, but that the dialogicality between old and new socio-cultural positions remains hidden. The voice of immigrants is subdued in public discourses as a result of the possible sanctions not to receive the desired status of permanent resident, but their voice is not missing. It may be muted, but it does whisper in safe and secure social situations. By examining the perspective of a group of Eritrean refugees on integration policies in the Netherlands from the perspective of Dialogical Self Theory, we aim at retrieving and reconstructing their voices. It is important to obtain better insight in the views and voices of immigrants because ‘integration’ cannot be considered a one-sided encompassment of alien elements in an organicist whole, but in practice it is invariably a two-way exchange between nationals and immigrants (Penninx, 2013; Klarenbeek, 2021). Thus, we are ‘writing against integration’ (cf. Rytter, 2019) in order to reestablish the difference between emic and etic discourses of what integration entails or should entail.

As mentioned before, we focus on a group of Eritrean refugees that applied for asylum in the Netherlands at the height of the European-wide ‘refugee crisis’ in October 2015. A group of approximately 100 young Eritrean refugees, between the ages of 18–22 (all male), was transferred from asylum shelters for minors to a block of university housing, even though the impact of living together in student accommodation on the civic integration process was controversial from the outset. While it is unclear what exactly the consequences of the collective residence were for the integration process, after two years it appeared that many were facing tremendous challenges in reaching the targets of civic integration upon which their status as citizens of the Netherlands depended.

Under the auspices of the local Radboud University, a working group of volunteers made up of (retired) staff was subsequently set up to support a group of some 25 Eritrean refugees in their struggle to pass the civic integration exams. It raised funds to finance extra lessons and extra opportunities for exams, and also to employ an Eritrean academic, Dawit Tesfay Haile, one of the authors of this paper, to mediate between the Eritrean refugees and the integration industry in the Netherlands. He worked under the supervision of Toon van Meijl, the second author of this paper. Both collaborated with the working group to secure the status of the Eritrean refugees as citizens of the Netherlands.

During the last months of the project in 2019, we conducted some 15 semi-structured in-depth interviews to hear from the Eritrean refugees how they experienced the integration process, what their views were of the Netherlands and how their future aspirations reflected the inevitable exchange between Eritrean and Dutch cultural practices. Ethical approval for these interviews was waived by the ethics committee of the Faculty of Social Sciences at Radboud University, and all participants were adequately informed about the purpose of the interview to which they consented. The interviews were conducted in Tigrigna, one of the two official languages in Eritrea, recorded and later translated into English by Dawit Tesfay Haile. The interviews were coded in AtlasiTi, first openly, subsequently axially and, finally, thematically (Beuving & De Vries, 2015). The sensitizing concept that we applied in the analysis was dialogue.

From our analysis of these interviews, but also from our frequent and intense interactions with the refugees over a period of two years, it appeared that all participants were involved in a deep dialogue between their home country and their host country. Although they rarely shared their qualms about the goal of civic integration during the lessons, in small talk after the training it was obvious that they struggled to strike a balance between potential differences or even contradictions between Eritrea and the Netherlands. For that reason, too, we should like to emphasize that this paper is not only based on an analysis of the interviews, but also on participant observation and small talk (Driessen & Jansen, 2013) that drew attention to their dilemmas around integration. The most suitable concept to describe their struggle to meet Dutch requirements without abandoning central aspects of their Eritrean identity is dialogue, which motivates our use of Dialogical Self Theory for the analysis of the cultural and political dialogues within their self.

After this introduction and brief outline of our methodology, we proceed with a sketch of the research setting, to be followed by an explanation of integration practices in the Netherlands. Subsequently, we highlight the identity issues that emerged among Eritrean refugees during the integration process, after which we zoom in on their muted voices to recover their cultural perspective on integration in the Netherlands. Dialogical Self Theory will prove to be very useful to analyze the dilemmas and contradictory demands upon which they stumbled in the context of civic integration. We begin, however, with a sketch of the country context of the research group and their shared background in Eritrea before they requested asylum in the Netherlands.

Research setting and context of civic integration

The state of Eritrea is a country of approximately 3.5 million people located in the Horn of Africa that belongs to the poorest countries in the world, ranking 179th out of 188 countries included in the Human Development Index (Ferrier et al., 2017, p. 6). It seceded from Ethiopia after a war lasting 30 years and gained independence in 1993. After independence, the rivalry between Eritrea and Ethiopian continued, which in the late 1990s let to an outburst of hostilities again. The border conflict caused the president of Eritrea to mobilize all adults between 18 and 40 years of age for the army. Although conscription into the army is officially for 18 months, in practice it is indefinite (Sterckx & Fessehazion, 2018, p. 43). During this war, Eritrea also became effectively a one party state, where an armed independence movement that liberated the country from Ethiopia transformed itself into a political organization under one president. Elections have never been held.

The micro-control of Eritrean citizens by the state was also reflected in the initial reticence of our research participants who are among the 100 young men accommodated in student housing at Nijmegen-Lent. When they applied for asylum in the Netherlands upon their arrival in 2015, they were granted the status of refugee, formally described as ‘status holder’ (statushouder in Dutch), meaning they had been recognized as refugees who are not able to return to their country of origin because their lives, freedom and/or livelihood are in danger upon return. All participants in our research, who settled down in Nijmegen after a long journey, had been born in rural or semi-urban parts of Eritrea, where they had received education until middle school before they left the country. They were raised among their extended families and had barely lived on their own. The values that characterized their cultural background revolved around social obligations to family and solidarity with the village community, which contrasts markedly with the central value of autonomy and self-sufficiency among adolescents and emerging adults in the Netherlands (Ferrier & Massink, 2016, pp. 13–20). At the same time, self-reliance is also a key characteristic of the civic integration trajectory that came to dominate their lives after they were recognized as forced migrant.

Although the original meaning of integration refers to the process in which newcomers become part of the society to which they migrated, it has obtained a rather specific meaning in postcolonial Europe in recent decades (Swinkels, 2019). In the 1990s, the first integration policies aimed primarily at repairing the so-called mismatch between unskilled ‘guest workers’ with limited command of the Dutch language and the post-industrial labor market. In 2007, however, the obligation to learn Dutch was replaced with the requirement to pass a more comprehensive civic integration exam, consisting not only of Dutch language, but including also knowledge of Dutch society (Kennis van de Nederlandse Maatschappij, KNM) and an additional module that would offer an orientation on the Dutch labor market (Oriëntatie op de Nederlandse Arbeidsmarkt, ONA). All these exams are to be completed within three years. In case migrants have not been successful in completing their civic integration trajectory, they may be sanctioned through a fine or, in final instance, may even be withheld a more permanent residence permit or nationality.¹

In this context, it is important to clarify that implicitly integration has come to be connected to a different notion of citizenship. It is no longer framed around legal rights and duties as citizens, but it increasingly requires cultural adaptation, if not assimilation. As such, citizenship entails an extra-legal dimension of normativity as a formal citizen is not necessarily also a good citizen. The new conception of citizenship is not only descriptive, but it is also prescriptive, demanding people who are considered to be culturally and morally different to adjust to mainstream thought and action (Schinkel, 2010; Schinkel & Van Houdt, 2010). This new framing of citizenship has been branded as a form of culturalization and is intertwined with a widespread belief that migrants, including second generation migrants, are insufficiently ‘integrated’ by still being culturally different (Duyvendak et al., 2016). The priority that is now given to cultural citizenship enables the state to address all migrants as people who are still lacking citizenship in cultural or moral terms. As a consequence, citizenship has been transformed from a right to be different to a duty to be similar, that is ‘assimilated’, both culturally and morally (Schinkel & Van Houdt, 2010, p. 204).

Paradoxically, however, the Dutch state passed on the responsibility for the intensified integration process to the status holders from the moment they increased the demands that newcomers have to meet in order to become ‘integrated’ and to qualify for citizenship. This shift in civic integration procedures has been informed by neoliberal policies, which have been translated in terms of a responsibilization of immigrants and the marketization of civic integration lessons and exams (Suvarierol & Kirk, 2015). Migrants have not only been made responsible for successfully completing the civic integration programmes that they are obliged to follow, but since the organization of the course modules is based on market freedom they have also been made responsible for choosing and financing courses offered by recognized providers in order to pass the standardized and computerized exams. This neoliberal approach of integration has transformed citizenship from a status with entitlements to a contract between the state and its agencies with newcomers who aspire to become a full-fledged citizen. The contractualization of citizenship (cf. Suvarierol & Kirk, 2015, p. 3) converts citizenship into a commodity that needs to be earned, which in turn makes integration into a one-way process. Civic integration contracts apply ‘technologies of self’ (ibid.) by producing new citizens who will govern themselves within the ambit of social norms and cultural values that the state has declared to be dominant. It may also be described as a form of ‘repressive responsibilization’, particularly designed for migrants who are classified as a ‘risk’ and a potential threat to the social order because in principle they are deemed unable to assume responsibility (Schinkel & Van Houdt, 2010, pp. 708–709). Needless to say, the Eritrean refugees in Nijmegen were also subjected to these policies.

Performing integration

When asylum seekers are recognized as refugees and become status holders they are entitled to a loan of up to €10,000 for three years so that they can independently pursue the obligatory civic integration lessons and exams. The loan is administered by the Education Executive Agency (Dienst Uitvoering Onderwijs, DUO),² which also offers the exams. As mentioned above, refugees are expected to organize their own programme by registering with one of the accredited training providers. In view of their background in rural Eritrea and the limited education they have had, this proved to be a real challenge for most of them.

The Netherlands is a highly individualized and digital society, where communication with government agencies and other organizations takes place increasingly in electronic form. It implies that refugees can usually only register for training in an online environment, but since most of them did not have a computer this was often done on a mobile phone. As a consequence, they were rarely informed adequately of the conditions, one of which was regular attendance. The experience of settling down in the Netherlands, however, not infrequently after a traumatic journey, in combination with the need to build up a new life independently, far away from home and family, made it difficult to go to school several days a week. Simple logistics of catching a bus and transferring onto another bus line in order to reach the school, having to check in and out of each bus with an electronic travel pass, uploading credit onto that pass, monitoring the balance of their bank account, it was all very complicated for a significant number of Eritrean youngsters. Hence, too, they missed many lessons and had to be reminded regularly to go school.

Initially, the young men from Eritrea residing in a block of university housing at Nijmegen received support from a large number of volunteers in the neighborhood, volunteers who were affiliated to the official refugee support organization VluchtelingenWerk as well as from people working for local government agencies. Interestingly, however, the collaboration between the volunteers and the agencies that were involved in the integration of the Eritrean refugees was not ideal since it was unclear who coordinated the process and on what grounds, while formally the refugees were themselves responsible for their ‘integration’. This relative chaos among the supporters of the Eritrean young men did not immediately help them to gain the necessary skills to pass the integration exams and become eligible for permanent residency.

Toward the end of 2017, some 35 Eritrean refugees were identified with sincere difficulties to meet the integration requirements. The fees for their training had been automatically collected from their loan, leaving them with an almost empty account, whereas in practice they had attended very few lessons and were far from ready to go up for the exams. This caused a group of staff members from Radboud University to become involved and offer support. Funds were raised from private agencies to finance extra lessons, while voluntary training was also offered, all in order to prepare the refugees for the necessary exams. Since the pressure was mounting on the young Eritrean refugees to complete their civic integration, their daily activities were shaped around these obligations to follow lessons and to study for the exams. They were guided intensively by professionals and volunteers who were dedicated to their future as a Dutch citizen.

Ironically, however, the volunteers or professionals who made an effort to salvage the Eritrean refugees from the consequences of not meeting their integration obligations hardly ever questioned the desired future of the young men. They shared the “will to improve” (Li, 2007) the performance of the boys in the integration process, but the ultimate goal “to become Dutch” was rarely reflected upon. Everyone was aware of the colossal cultural differences, but an alternative to simply becoming Dutch, whatever that implied for different categories of people, was beyond question. Nobody even asked what the aspirations of the Eritrean refugees were? How did they imagine their future? In the Netherlands or perhaps also in some other country? And in case they were adamant to stay in the Netherlands, what were their views on the country, its people, their culture and their values? In our intensive social interaction with the Eritrean young men, however, it clearly transpired that they were struggling with cultural dilemmas surrounding their identity as Eritrean refugee in the Netherlands. Dialogical Self Theory offered us the analytical tools to address the hidden dialogue about these quandaries within the self of the Eritrean young men.

Imaginaries of integration

The lack of interest in the future aspirations of the Eritrean status holders and their views of the Netherlands is puzzling in light of the so-called ‘agency paradox’ that draws attention to a striking contrast between, on the one hand, the initiative and vigor to escape from oppression in Eritrea and, on the other hand, an alleged lack of agency among refugees upon arrival in the Netherlands (Klaver et al., 2018, p. 39). The perceived ‘lack of agency’, however, tends to ignore the cultural baggage of Eritreans, which they could not simply abandon, and which evoked a continuous dialogue between past and future. This appeared, among other things, from the usually numb response to the question that they were constantly asked: ‘what would you like to do?’, a future-oriented question that invariably neglected their past. Therefore, the alleged lack of agency must be interpreted, first, in view of the pressure to complete civic integration exams within a specific time frame, leaving refugees with little leeway to show initiative. Second, it should be realized that the Eritrean refugees grew up in a country where the government took complete responsibility of the lives of citizens, whereas in the Netherlands they were as individuals expected to take responsibility for their own integration.

Eritrean society is not only organized top-down with the government in full control, but the rural background of the country was also in stark contrast with the complexity of public life in the Netherlands, where a proactive attitude is required of citizens as formal interactions with government authorities are highly bureaucratic and increasingly also virtual. Indeed, this contrast involves a clash between different conceptions of ‘responsibility’. At the same time, it must be made clear that the integration process leaves little room for the right to interpret the world differently than what is exhaustively laid out in the integration program. As Schinkel (2019, p. 1) has pointed out, the integration process is foremost an encounter between “us the civilized, secular, liberal, liberated who have the courtesy to take up the burden of bringing ‘them’ up to speed, of including ‘them’ in what is inevitable anyway”. During the interviews that we conducted with a selection of Eritrean refugees, the tension between the expectation to take responsibility for their civic integration and the requirements of the mandatory exams was brought forward repeatedly:

Eritreans in general are not open people. We don’t tell our stories and challenges and we don’t ask for help. We keep our problems to ourselves… We came here with fear deep inside us. [8]

It all has to do with the way we grew up in a very oppressive system and that cannot change overnight. We feel scared to say or do what we want to do. If you see Dutch people saying what they want and do what they feel like doing, that is a big challenge for us.[6]

These expressions evidently blur the boundaries between culture, classically understood as long-lived customs, beliefs and traditions, and totalitarian political systems, relatively short-lived and unstable, that informed and structured the lives of the Eritrean refugees in the Netherlands. Despite this complexity, refugees’ experiences are usually viewed as radical ruptures with the past and their specific circumstances in order to fit them into a bureaucratized understanding of immediate harm or persecution and refugeedom (Treiber & Hepner, 2021). In this context, it is important to clarify that the interviews were held approximately three years after they had arrived, and one year after the volunteers from the university had stepped in to support and facilitate the civic integration proceedings. At that time, the Eritrean refugees had obtained an adequate understanding of the integration trajectory, which they also endorsed, at least formally:

I think they mean to say that we should be part of society as long as we live here. We have to forget our culture and get used to the culture of Dutch society [6].

Interestingly, however, this person added that in his view it is not feasible to abandon the culture in which they were brought up:

Your culture is your culture, you cannot completely change that. But you have to learn how to function in society, the way they do things. [6]

Statements such as this clearly indicated that Eritrean refugees, reproducing the messaging of integration, believed that the goal of civic integration was strictly unattainable simply because nobody can abandon her/his cultural upbringing.

Hidden dialogicality

When we were conducting the interviews we were not surprised to record such views. Based on participant observation and small talk in the years before we sat down to interview a selection of Eritreans, our research was guided by the assumption derived from Dialogical Self Theory that the self is not bounded and coherent, but instead consists of a dynamic multiplicity of I-positions located in social and societal fields of tension, especially in societies that are culturally heterogeneous (Hermans et al., 1992; Hermans & Kempen, 1998). Accordingly, the self fluctuates among different and even opposed positions, between being Dutch and being Eritrean, and it has the capacity to endow each position with a voice so that dialogical relations between positions can be established. Thus, Eritrean refugees were as we expected involved in a dialogue between different conceptions of their cultural identity in their new country: “you have to live the lifestyle of the local people and accept their manners, but we cannot completely be a Dutch person”. The voices speaking in the self of Eritrean refugees functioned like interacting characters in a story, involved in a process of question and answer, agreement and disagreement. Each of them had a story to tell about his own experiences from his own stance, from the viewpoint of Eritrean culture and from the viewpoint of Eritrean refugee in the Netherlands who was trying to strike a new balance between Eritrea and the Netherlands. As different voices, these characters exchanged information about their respective positions, resulting in a complex, narratively structured self: “If you want to live and function in this country, you have to understand society better, but it feels a bit strange. It feels like I do not belong here”.

In this light, the so-called ‘agency paradox’ is blind for the hidden dialogicality that characterizes the self of Eritrean refugees. They cannot be considered as passive or lacking in agency, but their dialogue with their new country is necessarily hidden behind the formal mantra that it is compulsory to complete civic integration. In this sense their dialogical self is a ‘society of the mind’ (Hermans, 2002), because there is no essential difference between the positions a person takes as part of the self and the positions people take as members of society, in particular a highly heterogeneous and multicultural society. Both self and society function as a polyphony of consonant and dissonant voices. Hermans (2018) has demonstrated this in his book Society in the self, in which he shows that the self is not only in society, but that the self even functions as a society.

Society as a metaphor for the self does not only refer to internal I-positions, such as ‘I as Eritrean refugee in the Netherlands’, but also to external I-positions, such as ‘my Dutch integration coach’. The latter are located outside of the person, but they form nevertheless an intrinsic part of the self. In other words, the self of Eritrean refugees extends to volunteers and integration coaches as essential parts of Dutch society who are active within the broadened boundaries of their selves. The question is whether representatives of the Netherlands are welcomed in the extended self or that they are considered as unwelcome intruders.

In our analysis of that dilemma, we wish to highlight two distinctive features of the dialogical self that are crucial for understanding its dynamics. First, in a multi-voiced self, there is constant dialogical interchange, while, second, the relationships between the several positions are characterized by relative dominance. Regarding the first feature, it may be pointed out that the most important characteristic of the dialogical perspective is that the self is considered a relational phenomenon that transcends the boundaries between the inside and outside, between self and society. Methodologically, this feature of the dialogical self has been elaborated by relating the spatialization of dialogical relations to the simultaneity of voices within the self’s discourse. In Bakhtin’s view, individual speakers are not simply talking as individuals, but in the utterances of individual people the voices of their surrounding groups and institutions are heard, just as the Eritrean refugees initially simply reproduced the mantra of their integration coaches that “we have to forget our culture”. In the course of the interviews, however, it usually appeared that most Eritrean refugees had serious qualms about the possibility to abandon their own culture. Indeed, they were relieved to be able to speak with a double voice: “we have to forget our culture” versus “your culture is your culture, you cannot completely change that”; and “you have to live the lifestyle of the local people and accept their manners” versus “we cannot completely be a Dutch person”. Bakhtin (1984, p. 197) described this dilemma or contradiction as a form of ‘hidden dialogicality’. An attempt is made to internalize the other’s discourse of integration, causing them to provide the model answer of which they thought it was correct, but which they themselves have not - yet - integrated within their self. Dutch integration coaches are invisible and their words are absent, but profound traces of their mantra that integration is mandatory determine all their answers. Thus, the notion of dialogue is not necessarily equivalent with explicit conversation, but still it is at the heart of all thought. The ‘other’ representatives of Dutch society were absent during the interviews, but still they influenced all answers that the refugees thought were correct from the perspective of civic integration. After all, their very existence in Dutch society depended on it.

The absence of the other in dialogue is closely connected with the second distinctive feature of dialogical relationships, namely relative dominance between self and society (Hermans & Kempen, 1993, p. 73). Since the self is defined as a multiplicity of different identifications, society is not only able to address the self in a variety of identities, but also to let the self know which identities are approved. As a consequence, civic integration proceedings have the capacity to make some constructions of identity more dominant than others. And the dominance of civic integration in the lives of status holders in the Netherlands not only organizes but also restricts the multiplicity of possible identities in the public arena of Dutch society. Eritrean refugees, however, cannot deny their cultural background, which is an intrinsic part of their self, let alone they can abandon it: “you cannot erase the culture that you grew up with because it is the foundation that makes you who you are”. Experience with integration proceedings and negative experiences with discrimination in the Netherlands (see below) may cause the active suppression of unwanted dimensions of their Eritrean identity, which may slow down dialogical interaction between different cultural positions. The dominance of societal discourses of integration, however, does not exclude dialogical exchange, that is only submerged: “I would like to be part of Dutch society, but… I [also] want to keep the culture and tradition that I grew up with.”

Thus, Dialogical Self Theory helps us to draw attention to the hidden dialogicality in the self of Eritrean refugees and to avoid seeing them as passive. At the same time, it is inevitable to refer to an additional factor that explains the ongoing dialogue between Eritrean and Dutch culture, which is related to the racial dimension of the cultural encounter: “I don’t feel good here, but is it because I am an outsider or is it because I am black? I don’t know, but it doesn’t feel nice”. This person clearly indicates that the encounter of Eritreans with Dutch society also involves a confrontation of long-lived, exhaustively studied conceptions of Whiteness and Blackness (e.g. Du Bois, 1903). W. E. B. du Bois was among the first pioneers investigating the encounter between black and white America, and his early reflections continue to be relevant in contemporary debates about migration and citizenship in Europe.

Du Bois developed the concept of ‘double consciousness’ to understand the internal conflicts experienced by subordinated groups in oppressive societies by making the racial aspect of that confrontation explicit. The concept of double consciousness, in that sense, not only draws attention to the hidden dialogue between Eritrean and Dutch culture, but it also helps to uncover the racialized dimension of that cultural encounter. Du Bois (1903) used double consciousness to describe “an internal conflict [cf. ‘hidden dialogicality’, DTH & TvM] in the African American individual between what was ‘African’ and what was ‘American’”, which he interpreted in analogy with a confrontation between Black and White (Bruce Jr., 1992, p. 301). This viewpoint resonates with the numerous statements of our research participants about racist and discriminatory encounters in the Netherlands. They conveyed the strong impression that Eritrean culture is not accepted or recognized in the Netherlands, and that the reasons behind this exclusion are likely related to them being black. For that reason, too, it is necessary to acknowledge the racial dimension of the cultural encounter and read racial sentiments in statements such as ‘my culture is not accepted therefore I am not accepted as an individual’. Indeed, our Eritrean respondents were not only not accepted because of their different cultural background, but also because they were Black in a predominantly White country.

Negotiating integration and navigating discrimination

Since Dialogical Self Theory has helped to uncover the hidden dialogicality of Eritrean refugees in relation to civic integration, and now also mindful of the racial dimension of that encounter, it becomes easier to obtain a better understanding of the perception of Dutch society and the future aspirations of the Eritrean refugees. As indicated above, they reproduced the goal of integration without hesitation as they were keen to get a job, earn a salary to build up a new life in a new country and they did not question the need to speak the language and acquire better insight in Dutch customs and values. At the same time, they were unable to envisage how they possibly could abandon their own cultural background, although they were rather reticent in sharing that view because their legal status was still insecure. This belief was compounded by their first experiences with Dutch bureaucracy and the civic integration trajectory being far from positive, which naturally influenced their imaginaries of a future in the Netherlands.

Most refugees did acknowledge that the volunteers, language buddies and integration coaches were all very friendly and supportive, but when they found out that their loan had almost run out whereas they still had not even passed one out of six compulsory exams they came to realize that in first instance they had been misinformed and perhaps even manipulated by the private language schools. In any case, they felt that they had not been informed adequately about the school schedules, the obligation to attend daily, how to get there and many other practicalities. During the first couple of days they were dropped off at school, but in the afternoon they did not have a clue how to get back home. And how their finances had been arranged was also unclear to them. The details and the conditions became only clear to them after some time, when it appeared that in some cases 80% of their loan had been transferred with no progress in their integration trajectory. In this context, one refugee drew a comparison with Eritrea and argued that the education system there was in fact much better since the teacher would check students’ work daily, while in the Netherlands the schools did not do anything:

In Eritrea, a teacher would explain everything for the students and he would speak all the time trying to make you understand the topic. Here, however, they put the screen on for and you would learn by yourself. So if you see the difference, the teachers in Eritrea would spend time and energy to explain the topic to the students. Here they would just put the computer on and the rest is up to you…

The negative interaction with the bureaucracy and agencies operating on behalf of the government such as language schools also affected their sense of belonging in the Netherlands at large. They never failed to acknowledge the support they received from volunteers, but after a while many became also frustrated with the difficulties to find a job. They understood the need to speak Dutch, but at the same time they felt discriminated because they argued to be good and hard workers. In some cases they were even expected to work three months for free to test their capabilities, but when the probation ended they were told to leave. When they did manage to land a paid job, they would receive a lower salary for exactly the same type of work, even though they claimed to work harder than their Dutch colleagues.

Experiences with discrimination also extended beyond the labor market. Although some people expressed genuine interest in their well-being, they were sometimes also sworn at, even at one occasion when someone was simply waiting for a traffic light and a Dutch passerby told him he must have stolen his bike because “you are a refugee and cannot afford that”. One person also shared that he and a friend were refused entry to a night club for no apparent reason, which made them feel depressed: “it makes you feel bad and unwelcome”. It also affected their motivation to take initiative and develop other activities. A relatively biding attitude, however, must not be mistaken for a lack of agency. Their agency, however, was focused more on the dilemmas they were facing and the need to negotiate the tension between their struggle to adjust to their new country and the necessary reflection on their Eritrean cultural background this entailed. Naturally, they wanted the best of both worlds and, in the words of Du Bois (1903, p. 11), “merge [their] double self into a better and truer self”, but it was not always easy to deal with the challenges and to make choices they never even had to consider before. One refugee, for example, mentioned that he enjoyed the company of some student volunteers, but he added that “they drink, take drugs and don’t believe in God”. Since the orthodox christian church offered him strong support in the new situation with all its uncertainties, which also implied abstinence from alcohol and drugs, the dilemma this entailed was new and at times confusing as well.

Against this background, Eritrean refugees gradually realized they had to navigate the situation and creatively build a new identity as resident in a country with numerous unattainable requirements for immigrants. First, it became apparent that their transnational identity and embodied subjectivity as Black Africans exceeds the bounded conception of Dutch society and the bounderies of the nation-state at large. Furthermore, refugees and other immigrants may successfully pass civic integration exams, but that does not necesarily transform them into citizens who like ‘natives’ are accepted by society (Ghorashi, 2017; Slootman & Duyvendak, 2015). Cultural encounters in the Netherlands echo continuously a dichotomy between ‘us’ and ‘them’, although more and more immigrants are situated within transnational spaces and draw from a multitude of intercultural experiences. In that sense, the political discourse of integration, which in practice is effectively translated into assimilation, is lagging behind global, increasingly multicultural practices. It leaves immigrants with no other option but to navigate their multiple identifications and strike a balance, in this case, between Eritrean culture and Dutch culture, that will shift according to context, sometimes primarily Eritrean, sometimes also a little Dutch. The interchangeable self-identification of a combined future as being both Eritrean and Dutch, of being outsider as well as insider, illustrates the dynamics of their belonging. It is a form of belonging that cannot be contained within the confines of the matrix of civic integration. As one of the refugees confided to us: “I think I consider myself an Eritrean, a refugee, an immigrant, an expat and a global citizen, just not one specific identity.”

This analysis of the dialogical interaction between Eritrean refugees and civic integration in the Netherlands also draws attention to the self-reflexivity and agency of the young men. They were involved in constant negotiations to resolve the tension they experienced and to shape their muted resistance to the unachievable demands to ‘become Dutch’, which made it necessary to be creative in enacting their multiple identifications in everyday practices. As a consequence, they cannot be described as passive or lacking in agency. Instead, they attempted to navigate the new situation in which societal demands came to dominate their self.

Dialogical self theory and ambiguity within a multicultural self

In his most recent book, Hubert Hermans (2018, p. 60) argued that the self is not only social, but also societal. And from this theoretical angle, the I-positions of people located in fields of tension created by cultural differences, conflicts and oppositions constitute a particularly fertile field of study. Indeed, Dialogical Self Theory is particularly fruitful for the analysis of the people with multiple identifations living in diverse sociocultural circumstances and who are therefore faced with ambiguity and contradictions. Yet few people have used it for analyzing cultural and political dilemmas of migrants, with Sunil Bhatia (2007) being a notable exception. He studied a group of Indian Americans who functioned as respected members of American society, but at the same time felt they were seen as radically different and not as ‘real Americans’. Bhatia (2007) criticized mainstream theories of acculturation that are based on the questionable assumption of mutually exclusive strategies suggesting that when people integrate, they are not marginalized, and when they are marginalized they are not integrated. In contrast, he argued that integration and marginalization may well co-exist in the life of one person and function as mutually complementing components of a relatively successful cultural adaptation. He substantiated his view with reference to a dialogical perspective, claiming that conflicting or contradicting voices in a multiple self may form a useful combination of positions that is helpful to give an adequate response to cultural diversity that is characterized by dynamics, discontinuity, ambiguity and contradictions.

An important question in this context concerns the strategies that people may apply to deal with ambiguity and contradictions in a multicultural self. In the metaphorical representation of the self as a society, Hermans (2018, p. 65) distinguishes two types of I-positions that may play a central role in the coordination of conflicting position within a multicultural self: meta positions and promotor positions. Since these concepts may also clarify the multicultural identity of Eritrean refugees, we briefly elaborate on their meaning.

A meta position permits a broader view on the increasing diversity of specific I-positions and works as a centralizing counterforce against the decentralization of the self. This is necessary for long-term planning and well-balanced decision-making. It goes without saying that during the first couple of years after their flight from Eritrea, the refugees were still in search of a stable identity. As a corollary, they were not yet able to develop a powerful meta-position to strike a balance between the contradictory demands they were facing during the integration trajectory.

A promotor position, on the other hand, might also help to prevent the self from becoming fragmented or decentralized by giving a developmental impetus to one or more specific I-positions so that the self remains sufficiently centralized and organized. A promotor position guards the continuity of the self, but at the same time it offers room for discontinuity. Continuity is guaranteed by their ability to link the past, present and future of the self. A certain degree of discontinuity results from the fact that they function as a source of new positions. In this sense, promotor positions function as innovator of the self (Hermans, 2018, p. 71). In the case of the Eritrean refugees, a promotor position would entail a successful combination of multiple identifications based on cultural orientations originating both in Eritrea and the Netherlands. In the Netherlands, however, there is no political space to develop a so-called hybrid identity as Eritrean-Dutch, recognizing and respecting both dimensions of their multicultural self (see also Slootman & Duyvendak, 2015). Refugees and migrants have no option but to become Dutch, full stop, and discard their ethnic identity. Still, in practice many people with a migration background do identify, of course, in terms of both their country of origin and their new home country, e.g. as Turkish-Dutch or Moroccan-Dutch. This model offers also a serious possibility for Eritrean refugees, but during their civic integration trajectory they were, as indicated above, not yet capable of developing a promotor position to organize and give direction to a multiplicity of cultural I-positions.

In sum, then, we may conclude that the decentralizing diversity of cultural positions held by Eritrean refugees in the Netherlands (I as Eritrean versus I as Dutch), while the integration discourse is struggling for dominance, is not immediately complemented with a centralizing tendency in the form of a meta-position or a promotor position. Since a centralizing position seems as yet to be missing in the self of Eritrean refugees, the question emerges how their different I-positions as Eritreans and refugees in the Netherlands are related. And also whether they perhaps do have a position that is coordinating their self. These questions are crucial since we have demonstrated that civic integration policy institutionalizes differences between locals and immigrants by requiring the latter to meet demands that are principally unachievable, namely to abandon the cultural identity derived from their country of origin and to replace it with a Dutch identity. As a consequence, immigrants and refugees are involved in a dialogical process of negotiations between old and new identities, between the past and the present.

Above, we have already pointed out that the new host society and the self of refugees are positioned in an unequal power relationship, but it is important to add that there is also emotional and, to some extent, even racial distance between Dutch and Eritrean orientations. Extreme distance appeared at moments when Eritrean refugees noticed that they were discriminated, either by employers, in shops, or at the entrance of a disco. At such moments, Eritrean refugees whom we interviewed became aware that they would never be accepted as full-fledged members of Netherlands society: “I don’t feel good here… It feels like I do not belong here”. In terms of Dialogical Self Theory, it might be argued that their Eritrean identity had necessarily become a ‘shadow position’ in the Netherlands (Hermans, 2018, p. 93), since their ethnic identity was considered undesirable or even unacceptable and therefore it had to be suppressed. At the same time, they could not deny ethnic differences: “I am an outsider… because I am black.” Indeed, in view of the racial dimension of the encounter between Eritrea and the Netherlands, it cannot be surprising that the rejection of their ethnic identity caused a revaluation of their Eritrean identity. For in spite of their criticisms of the authoritarian political regime in their home country, they did regard themselves as dignified and respectable Eritreans, with a sense of pride in their cultural traditions and values. Indeed, they could and would never give up their ethnic identity as Eritreans, irrespective of the demands of civic integration policies in the Netherlands.

In this context, the concept of shadow position, as indicated above, is interesting because it draws attention to the opposite effect of what the casting of a shadow on some position aims to achieve. Rather than facilitating the integration of immigrants by stigmatizing their ethnic identity, the shadow cast on the identity of refugees leads to a closure of the boundaries with these others, resulting in sharp ingroup versus outgroup separations and oppositions. At the same time, the shadow cast on the cultural position of refugees triggers a reflection on the accessibility of the entire range of cultural positions within the self. For a dialogue between positions to be successful it is required that they are all equally accessible so that their energies can mix and mingle and each position can be empowered in different social situations. When the process of positioning moves fluently from one situation to another, I-positions have not only an entrance that permits access, but they also have an exit. This process, however, is interrupted when positions relevant in a particular context cannot be accessed or when it is impossible to leave them. Precisely this insight derived from Dialogical Self Theory clarifies the predicament of Eritrean refugees engaged in civic integration in the Netherlands: they cannot access Dutch citizenship because they cannot leave their ethnic shadow position and qualify for another cultural identity. In terms of Dialogical Self Theory, this Catch-22 is described as an ‘I-prison’, which occurs when someone is placed in a particular position from which escaping is not possible (Hermans, 2018, p. 91). In the Netherlands, the position of Eritrean refugee cannot be abandoned, not simply because they are black among a predominantly white population, but since immigrants with a non-European background are perceived as ethnic others who can never become fully Dutch. As a consequence, it may be argued that they are locked up in an I-prison. The dominant discourse of integration in the Netherlands demands identifications as non-Dutch to be suppressed and, as such, it fails to acknowledge the inescapable field of tension in multicultural identities. Dialogical Self Theory helped us to uncover this form of hidden dialogicality.

Concluding remarks

What is the relevance of Dialogical Self Theory for the analysis of the challenges and difficulties that are faced by Eritrean refugees in the process of civic integration in the Netherlands? The answer to this question is simple, yet it is not self-evident. A dialogical perspective on the performativity of civic integration by Eritrean refugees draws attention to the negotiations that are necessarily taking place between their ethnic identity and the goal of Dutch citizenship. Understanding this process begins with the acknowledgement that refugees who aim at passing civic integration exams and obtaining citizenship in a country that is culturally very different from their country of origin, are actively involved in the construction of a multiplicity of cultural identifications. In terms of Dialogical Self Theory, it may be argued that a diversity of cultural I-positions pulls the self into different or even opposite directions, while at the same time they belong to one and the same self that strives for coherence and continuity. In other words, the plural self of Eritrean refugees finds itself in a field of tension between decentralizing movements, which is not balanced with a clear centralized position, such as a meta-position as successful immigrant or a promotor position that would characterize a hybrid Eritrean-Dutch identity. Instead, a shadow is cast on their ethnic identity as Eritrean, which is locked up in an I-prison as eternal ethnic other.

The incarceration of the ethnic identity of Eritrean refugees in an I-prison, however, does not necessarily imply that they are passive or even that they lack any agency, as suggested by scholars who introduced the so-called ‘agency paradox’ described in the introduction (Klaver et al., 2018). Instead, as we have demonstrated in this paper, the dialogicality of the self of Eritrean refugees remains in first instance hidden behind the mandatory goal to pass civic integration exams that is required to obtain permanent residency, or, on the long term, Dutch citizenship. In the context in which we conducted our research for this paper, the need to pass civic integration exams dominated the lives of Eritrean refugees entirely, which also contributed to the model answers that we received initially, for example that it is important to internalize Dutch culture and its values when living in the country. In response to our questions regarding the implications of civic integration for their cultural and ethnic identity, however, they had no hesitations to express their qualms about the unrealistic goal of Dutch integration policy to abandon their Eritrean identity. The pressure to assimilate in the Netherlands had even triggered a new perspective on their Eritrean identity as a precious treasure, which they would and could never give up, in spite of the dictatorial regime that had caused them to leave. This revaluation of their ethnic and cultural identity as Eritrean made the main goal in their new life fairly simply: to engage in a balanced dialogue between their new host country and their home country. By recognizing the inevitable multiplicity of the self of Eritrean refugees and by bringing their hidden exchanges to the surface, Dialogical Self Theory contributes to the creation of conditions that are essential for their empowerment.

Notes

1 A new integration law has been introduced in 2021.

2 One of the changes in the new integration law of 2021 is that the loan is administered and can only be accessed by the municipality where the status holders reside and not by the individuals themselves as used to be the case under the previous integration law.