Introduction

In this article, we use institutional ethnography (IE) to analyse when and how family relations are addressed in social service encounters between social workers and migrant background service users. The ‘migrant family’ became a new service user category in social work in Finland during the 1990s when immigration to Finland grew. In social services, this was largely followed by an increased interest in ‘migrant service users’ and requirements for specialised services and expertise to encounter them (Buchert, 2015). In the process of integration, migrants are expected to learn the local language (Finnish or Swedish), familiarise themselves with the Finnish welfare system and its institutions and find employment. Previous research has suggested that welfare service work with migrant background service users has been defined by perspectives of cultural differences and how to become aware of, understand, and overcome these (Hiitola and Peltola, 2018).

More recently, the ‘migrant family’ has turned into an increasingly politicised category both in Finland (Jäppinen et al., 2020), in the contemporary Nordic welfare states (Nordberg 2015; Hiitola et al., 2020), and beyond. Analyses from Scandinavian countries show that family migration has gradually become more conditional (Borevi, 2015; Cochran Bech et al., 2017). In European migration policies, migrant families are increasingly viewed through lenses of ‘failed integration’, ‘deviant’ family norms and practices, and otherness, and the legitimacy of their claim to live their family life on national territory is often contested (Bonjour and Kraler, 2015). In Finland, several concrete legislative changes have been introduced to tighten the family reunification processes of migrants since 2010, including income requirements, apart from a limited three-month exemption period for applicants who have been granted asylum. Family reunification has subsequently become extremely difficult, even impossible, for many forced migrants residing in Finland (also Nordberg et al., 2024). Moreover, as Nordberg (2015) has noted, migrant background ‘stay-at-home mothers’ have become a politicised social category of women who are labelled as passive, socially disengaged, and not fulfilling the expectations of productivity and labour market participation.

In what follows, we will analyse how family is defined and made sense of in institutional encounters between social work professionals working in street-level welfare bureaucracies (Lipsky, 1980) and migrant service users. The aim of our inquiry is to make visible the extra-local factors influencing the ways in which family is (or is not) addressed in service encounters.

The data of the study have been produced between 2018 and 2020, and it consists of observed service encounters in immigrant services, adult social services and child welfare services, and interviews with migrant service users and social work professionals. Drawing on Dorothy Smith’s (2005) IE (see also DeVault and McCoy, 2006; Kuronen, 2020; Jäppinen, 2015), we suggest that social work as professional practice extends beyond the dominant frameworks of inter-individual interactions. Thus, our analysis is led by the idea of mapping the extra-local relations of ruling (Smith, 2005) which coordinate the local encounters and the sense-making of family in them.

Sense-making of ‘family’ in institutional encounters

‘Family’ is an object of ongoing definitions. On the one hand, it is an intimate relation, embedded in the everyday lives of individuals, and on the other, an institution regulated by cultural and societal norms (Peltola, 2014). For example, the definition of family used by Statistics Finland (2023) consists of a maximum of two generations of family members: a married or cohabiting couple, or parent(s) and their under-aged children. In this definition, grandparents or other relatives do not belong to the family even if they live in the same household, let alone if they live separately or even in another country. Indeed, in their subjective definitions people tend to view their families broadly and diversely (Holstein and Gubrium, 1999; Castrén et al., 2017), and institutional factors, such as marriage and household composition, provide only limited information on family life. Previous research has looked at, for example, transnational care and family ties (e.g. Baldassar et al., 2007; Baldassar and Merla, 2014; Kara and Wrede, 2022), family reunification policies and the right to family life (e.g. Pajnik and Bajt, 2012; Kontos, 2013; Mustasaari, 2017; Hiitola, 2019; Nordberg et al., 2024), and the practices and categorisations of parenthood and parenting (e.g. Peltola, 2014; Berg and Peltola, 2015; Hiitola, 2015; Turtiainen and Hiitola, 2019; Walsh, 2018). An emerging perspective is to focus on everyday citizenship (e.g. Heino, 2018) or the citizenisation of family members (e.g. Nordberg, 2015; Nordberg and Wrede, 2015).

Despite the growing body of literature on migrant families and how migrancy transforms family relations, surprisingly little scholarly interest has been paid to how the ‘migrant family’ takes shape in institutional encounters. Turtiainen and Hiitola (2019) have utilised Sara Ahmed’s (2000) notion of strangers being already recognised to be others to analyse the encounters between migrant families and Finnish family service professionals who categorise these families and their functions based on their perceived ethnicity and culture. Accordingly, Ylvisaker and Rugkåsa (2022) have shown how hegemonic values and discourses such as norms on gender and gender equality influence social work encounters with migrant families as well as practitioners’ interpretations and actions in these encounters. Turtiainen and Hiitola (2019) have also analysed how migrant parents ‘talk back’ to the stigmatising stereotypes of their family relationships and parenthood.

Walsh et al. (2022) have applied the concept of everyday bordering (Yuval-Davis et al., 2017) which refers to how state bordering practices extend increasingly into the everyday life of migrants and their families, in the context of social work practice. They scrutinise how everyday bordering practices regulate migrants’ everyday lives, and how a person’s immigration status directly impacts the services and welfare provisions to which the family is entitled as well as who is considered to be a family member. Homanen and Alastalo (2018) have conducted ethnography on how kinship and family relations are constructed in the statistics compilation and the registering practices of migrants in Finland. They employ the concept of ‘paper families’ to point to family relations being strongly defined by an idea of the Finnish, heteronormative nuclear family that becomes visible in the official documents and conclude that this narrow family model is deeply rooted in institutional encounters.

IE and the problematic

Theoretically and methodologically, we draw from IE developed by Dorothy E. Smith (2005; see also Campbell and Gregor, 2004). IE’s main commitment is that sociological inquiry should start from a standpoint in the everyday world instead of abstract scientific concepts and theories. Local actualities of people’s lives and their material conditions serve as ‘a point of entry’ to larger social and economic processes. The starting point of each study should be a locally identified ‘problematic’, and the focus of the study should be on exploring and solving it together with the participants of the study. In this study, the ‘problematic’ are the puzzling ways in which family relations and family members become included in/excluded from the agenda of the encounters between social work professionals and migrant background service users.

Although the starting point of IE is in the everyday lives of the study participants, the analytical focus is on the relations of ruling organising these actualities (Smith, 2005; DeVault and McCoy, 2006; Kuronen, 2020). When macro-level phenomena and social structures are understood as relations of ruling which coordinate the human action around and above the local, the local can be seen as a site to analyse broader social relations (Smith, 2002; 2005). By emphasising the interconnectivity between local encounters, attitudes and actions, and extra-local relations of ruling, Smith shows that people and actions at the local level are factually tied to each other in ways that can be grasped and conceptualised through abstractions, such as power, knowledge, capitalism, patriarchy, economy, state, politics, culture and so on (Campbell and Gregor, 2004).

Applied to this study, we understand the local encounters between social service workers and migrant service users as actively regulated by several extra-local factors, that is, relations of ruling. The concept of the relations of ruling helps us to map what lies behind the street-level sense-makings and negotiations of ‘family’. Here the relations of ruling may include, for example, cultural norms and scripts, legislation, institutional practices and professional practices and ideals. The way in which extra-local relations of ruling transmit to these encounters is often textual (Smith, 1990; Satka, 1995).

Thus, inspired by IE and puzzled by the ways in which family relations take shape in institutional encounters, we ask: When and how does family become addressed in institutional encounters between Finnish social work professionals and migrant service users? What kinds of relations of ruling coordinate this?

Empirical data and analysis

The ethnographic data in this article have been produced in four social work agencies in the Helsinki capital region. Two of the agencies provide services mainly for persons granted asylum during the approximately three-year ‘period of integration’, the third one is an adult social work agency, and the fourth provides child welfare and family support services. The research has been carried out following the guidelines of the Finnish National Board on Research Integrity. We have discussed the ethnographic fieldwork focussing on the issues of access and consent in our earlier article (Kara et al., 2023).

During the fieldwork, we understood ‘family’ broadly and our approach was intended to include research participants not only from families with under-aged children sharing the same household but also people from different personal and family situations. The service user participants had migrated to Finland for different reasons and from different parts of the world. The majority of the observed service encounters and research interviews with the service user participants were facilitated by language interpreters. The practitioner participants were mostly social workers and social advisors. In some of the observed service encounters, other social and healthcare professionals were also present. All research participants have given their permission to be part of the study based on informed consent.

The ethnographic data analysed in this article consist of thirty-nine observed service encounters in immigrant services, adult social services and child welfare services, thirty-two brief reflective discussions with professionals directly after the encounters, and fifty-one thematic interviews conducted separately with both migrant service users and professionals. Most of the observed encounters were audio-recorded and transcribed. In some cases, the researcher only took notes instead of audio-recording, according to the participants’ wishes. The opportunity to participate in an interview was provided to all service user participants, but some preferred to participate only by consenting to the observation of the service encounter. All practitioner participants were interviewed.

The data has been transcribed and coded by research assistants. In the analysis of this article, we employ the data coded by the following codes: ‘family relations’, ‘family and close relationships present in the service encounters’, ‘family definitions’ and ‘family and relatives in the country of origin and transnational family relations’. We began the analysis by reading the dataset carefully, making notes, and paying special attention to those parts of the data that revealed or touched upon certain tension in whether family becomes addressed or not, reasons presented in the data for why it is important/useful/unnecessary/harmful to talk about family, and how family was understood and defined. We present the analysis and its results based on case examples instead of thematic categorisations.

After locating the key parts of the data and, within them, the two cases that served as a starting point for the analysis, we continued by close reading to track the relations of ruling based on a comparison of the data excerpts with texts identified in them in order to find the relations of ruling and make them visible (see also Campbell and Rankin, 2017). In this study, these texts include, for example, legislative acts, service descriptions, professional guidelines and protocols. For example, when a social worker explained that the social security legislation defines whose expenses can be considered in an income support decision, we looked at the law. Accordingly, when a social advisor referred to city-level directions on income support to baby carriages, we asked to see these instructions. These texts are not employed here systematically as a predefined data set, but according to the ideas of IE, we have followed up on each ‘clue’ given to us by the data (Campbell and Rankin, 2017) and consequently created a dialogue between the ethnographic data and the texts. Some of the relations of ruling in this study, such as the professional ideals about how to perform social work assessments, do not relate to any particular text body (e.g. Virokannas, 2017).

Case 1: Social security legislation limiting the presence of family in the service encounters

We start the analysis with a quotation from an observed service encounter which took place in adult social services between a service user, Jasmin, a mother of five, and a social advisor, Sandra (names changed to protect the participants’ privacy). It was their second meeting. After informing them of the research and signing the informed consent forms, Sandra opened the discussion by suggesting that they address family:

Last time, during the whole meeting we spoke about your financial situation, there were some problems with it. And we used all the time on… Well, we… I don’t yet have much information about your family and about other aspects of your life. You told me last time that you live with three children. Just now [when discussing the participation in the study] you mentioned that you have five children, some of them are adults, and then… Last time you just briefly mentioned that you’ve moved [a couple of months ago]. Would it be ok to discuss your family situation now a bit more? Because I’m of course interested in everyone’s well-being and situation in your family. So, you can tell me with your own words who belongs to your family, and how they are doing. (Audio-recorded observation, translated in English by the authors)

A moment earlier, when we were presenting the research project, Jasmin spontaneously mentioned that in addition to her three minor children already known to the social advisor from the earlier meeting, she also had two older children who had recently left home and disappeared after their residence permit applications were rejected. In a research, interview organised a few days later, Jasmin said that she would not have brought the issue of her two adult children into the discussion with Sandra had it not been for the presence of the researcher and her expressed research interest in broader family relationships. According to Jasmin’s understanding, encounters in social services were meant to focus on the needs and situations of family members living in the same household with her.

Jasmin’s solution to limit the family relationships she would take up in the service encounters draws a contrast to the above quotation where the social advisor speaks of her professional interest in all family relationships relevant and meaningful to the service user. Generally, social work practitioners who participated in the study considered family and close relationships of the service users highly relevant psychosocially. Similarly, the template used for writing an assessment of service needs in adult social services guided the professionals to ask broad questions about family and other close relations. Moreover, the general ideals of social work as a service-user-oriented practice, guide the practitioners’ interest towards the service users’ family and close relationships as extensively as possible.

At the same time, Jasmin visited social services with a concrete mission: she wanted to apply for financial support for some very specific and practical needs of her family. This was common in our data: the discussions in the service encounters often focussed on social security benefits, as had also been the case in the previous encounter between Jasmin and Sandra. It is logical that service users limit the discussions to issues that have a concrete influence on the goal they set to the service meeting instead of building a holistic relationship with the professional. In the case of Jasmin, social security legislation steers the conversation. According to the Social Assistance Act (1412/1997), ‘“family” is used to mean parents, a parent’s minor children or adopted children, married couples, and men and women living in circumstances similar to marriage who live in a common household’. Therefore, social security benefits can be issued only to those family members who enter the narrower definition of family as parents and their minor children sharing the same household. When social security benefits were discussed, they were followed by a narrow understanding of the family on which the social security legislation is based.

When social security benefits were at the forefront of the encounters, there was also less time to talk about other issues. Thus, service users or practitioners might have little motivation to bring family relations into the discussion when these are not directly relevant to the decisions on services and benefits. The example of Jasmin’s and Sandra’s encounter reveals a general lack of time for discussing family relationships at the meetings when there are several practical matters to be dealt with. In the conditions of time shortage, the encounters often revolved around current issues and discussions on family relations in a broader sense were left out. In the above example, Sandra is interested in hearing the story of the older children, as she evaluates the stress caused by the situation to be an important factor in potentially influencing the well-being of all family members. Finally, however, she was not able to offer any concrete assistance to Jasmin in this situation.

The above example reveals a network of extra-local relations of ruling, such as laws regulating who is a family member in the eyes of the official social benefits system, as well as an understanding of the issues that regularly are or should be addressed in social service meetings in adult social work. In the research interviews, many professionals reflected upon their own expectations and preconceptions concerning family and the way they address family in the service encounters or whilst filling different service-related forms, such as the service needs assessment forms, with the service users.

Well, the instruction is maybe not so defined, but family and personal network… the idea is to always ask about relatives and friends. […] When thinking about service users who are Finnish or have been born in Finland it’s more about the nuclear family or those who live in the same apartment, but with migrant background service users somehow my assumption is that the concept of family is broader. I wouldn’t say that it necessarily has an influence on anything, I ask everyone the same questions, but maybe I think that, for example, in many cases the family relations of migrant background service users are more binding, and they have more obligations, as a whole, than generally is the case here in Finland. (Interview)

The way in which this social worker reflects upon her assumptions and preconceptions concerning family and the nature of family relations gives us an understanding of the relations of ruling that were present also in the encounter between Jasmin and Sandra. First, the account shows that the encounters between social workers and service users are strongly coordinated by the official forms that need to be filled, as the social worker states that she basically always asks everyone the same questions. She therefore believes that what she might assume beforehand on the matters tackled in the meeting is irrelevant. She nevertheless also produces a distinction between the culturally ‘Finnish family’ that is generally small, consisting of the nuclear family, and in addition to its small size, is also less binding with fewer obligations. She contrasts this to an idea of a ‘larger family’ that she posits the migrant background service users have that seems to her also less flexible and more binding.

At the same time, it was not uncommon that, for example, rent debts had interconnections to care responsibilities that exceeded the boundaries of the family definition of the Finnish social security legislation. For example, sending money transnationally to the members of the ‘larger family’ is a potentially complicated topic in the service encounters, and social workers assumed that many service users avoid this topic in the encounters. Some of the professionals evaluated that service users do not want to talk about the transnational family responsibilities for financial support, but some of them said that sending money is discussed often.

Social advisor: There is quite a lot of such that money is sent to them [relatives abroad] from Finland, as they are kind of family and even aunts and cousins belong to it, and uhm…

Researcher: Does it come up often, sending money to relatives in the country of origin?

S: Yes, often.

R: What do you think about it?

S: Well, I am obliged to say that the family needs to take care of their own nourishment and cannot think that they get support from somewhere while they send money abroad. That it just does not fit in the Finnish way of thinking, I understand it very well, and I always say to them that in a way it is good [that you take care of your relatives] but in this situation when you don’t have money, you can’t. But they send it anyway no matter what I say.

(Interview)

As the social advisor explains, the role of professionals was to explain the institutional border according to which social security benefits are meant to fulfil only the needs of those family members sharing the same household, and other expenses cannot be taken into consideration. Thus, the professionals felt their position to be uneasy in situations in which they knew about money being sent abroad and at the same time having rents left unpaid in Finland. Some of them assumed that migrant service users may be reluctant to address these issues in service encounters because of the fear of being punished for using the social security money ‘incorrectly’. Here again, the social security legislation becomes a strong ruling relation that overrides the holistic professional interest towards listening to the self-definitions of service users’ family relations and considering the care responsibilities that these may include. An interesting question is whether the information on transnational responsibilities of care and money transfers as a background factor to the financial troubles and rent debts influences the practitioners’ professional discretion and how.

Yet some of the professionals noted that the transfer of the decision-making on basic income support from municipal social work agencies to the National Insurance Institution has brought new flexibility in listening to the service users’ self-definitions on who is a family member.

(—) now I practically don’t need to care about whether there is someone living in the apartment informally or not […] because I don’t need to make the decision of income support every month in which I should take a stand on that. So maybe now they can tell me that there’s another person living here too but it’s not official. It has perhaps brought that freedom…[…] and that’s good, because families are not the same as how it’s written in the income support legislation. (Interview)

Here the social worker describes that the removal of income support decision-making power and responsibility has resulted in increased freedom in her encounters with the service users to discuss different family and household situations. In other words, she describes a situation of a diminished influence of social security legislation as a relation of ruling on her work. The social worker expresses this change as increased freedom for service users to express their situation, but arguably what also has been altered is the freedom of the social worker to listen to these situations without this turning into an obligation to ‘take a stand’ in terms of what effect these might carry on the service users’ income support applications and their handling.

To sum up, the case of Jasmin revealed certain tensions that take place in the negotiations about ‘family’ in social service encounters. First, it made visible the tension between a client-centred idea of giving space to everyone’s own family definitions and the narrower understanding of family as a nuclear family of parents and their under-aged children which is deeply rooted in the social service legislation. Secondly, it showed a tension between the professionals’ will to provide psychosocial support and the service users’ rationale of coming to the meeting with a certain concrete (often financial) need, which can be solved. Both service users and professionals had to balance these tensions, as was further exemplified by the negotiations over transnational family obligations of financial support.

Case 2: On how transnational families become visible through negotiations about family reunification

Whilst we have discussed mainly hidden, implicit or limited talk of family in the service encounters above, the negotiations between different definitions of family were at times explicitly present in the service encounters, specifically in the discussions on family reunification, which is our next topic. Especially in the immigrant services, these were the moments in which the negotiations between different kinds of family definitions intensified and became more explicit in the service encounters.

Well, we can have such discussions with really many [service users] that why the [residence permit] application based on family reunification was negative, and why an elderly mother is not regarded as a nuclear family member, and you are an adult, and how the Finnish society understands it [adulthood and familyhood]. And many grieve their own parents, especially their mother. (Interview)

In the above extract, the professional defines their role as one to explain the narrow understanding of family in the legislation concerning family reunification. The professionals also emphasised their limited authority in family reunification matters, even though in immigrant services they, in practice, were often involved in the application processes, communicated between the Finnish Immigration Service and the service user, and showed interest in and enquired after the advancement of the process. The Finnish Immigration Service explicitly produces a dichotomy between the narrow ‘Finnish concept of family’ and that ‘in many other countries’ the concept of family is different (Migri, 2023a). However, the nuclear family consisting of an adult (often also married) couple and their biological under-aged children is prevalent in most European countries as the basis of family reunification legislation (Strasser et al., 2009). In the case of broader families, such as adult siblings or grandparents, dependency and ‘active family ties’, meaning that the family members have not been separated for ‘too long’, or other humanitarian causes are required (also ibid.). Generally, family reunification processes were associated in the data with varying degrees of ambiguity and uncertainty as to the definitions and grounds for who is eligible and who is not. Even if the professionals at times questioned and criticised the decisions made by the Finnish Immigration Service, they largely felt unable to influence them.

As in the case described above, the negotiations between different understandings of who belongs to the family were typically activated in our data when pondering the possibility for family reunification or when handling the disappointment of a negative decision. As above, the role of professionals was to explain the narrow family definition on which the legislation on family reunifications is built, and this is something that repeats in our data. The Aliens Act (301/2004) and its regulations concerning the conditions for family reunification became a strong, yet ambiguous and troublesome, relation of ruling coordinating these street-level negotiations. Importantly, this also restricted the assistance offered to the service users by the professionals of the immigrant services for filling in the family reunification applications. The assistance was mostly limited to applications evaluated as most likely to succeed: applying for a residence permit for a spouse or an under-aged child. This was justified by targeting the limited resources: assisting in applications for other relatives was sometimes regarded to be a loss of time and resources. According to the Aliens Act, a family reunification issued to other relatives is possible only in exceptional situations.

In adult social work and in child welfare services, family reunification issues were, in general, discussed less than in immigrant services. Sometimes the professionals quite explicitly avoided this topic (also Nordberg et al., 2024). Starting from 2016, all migrants applying for family reunification must show ‘secure means of support’, meaning that the applicant must have sufficient income to provide for themselves and for their family members to live in Finland without social security benefits. The required amount of income depends on the size of the family and is relatively high. For example, a family of two adults and two children under eighteen years of age needs a total of 2,600 euros net income per month to be able to show secure means of support (Migri, 2023b). As most of the service users in adult social work are there to apply for financial support, they very seldom reach these income requirements. Thus, the income requirement became a relation of ruling that coordinated whose family is worth talking about. Both migrant service users and social work professionals might avoid speaking about family also because it was too painful or even traumatising to address family relations without hope of changing the situation of living apart.

Nevertheless, the majority of the service user participants mentioned persons not residing in Finland as their family members in the interviews. The professionals working on each of the sites knew that most migrant background service users do have transnational family and close relations. Social workers’ will to pay attention to the migrant service user’s life situation more holistically is reflected in their considerations about how transnational family relations should be discussed actively. Even though, at the same time, many of the service users had adopted a strong institutional script of transnational family relations not belonging to the service encounters, as described below by a service user whose own parents continue living in a conflict area:

Adam: […] When it comes to the fact that the family is there [in a conflict area] and they are in trouble—no, we have not received any help.

Researcher: Well ok. Do I understand right that you do not even take up these family relations except those here in Finland in the service encounters in adult social work?

Adam: Mostly we talk about our family that is here. Because in the beginning we were told that in Finland they help only those who are in Finland, and it doesn’t make any sense to start talking again and again, if we know how this system goes and that it helps only those residing in Finland. (Interview)

Some of the professionals also felt uneasy asking questions about transnational family ties or regarded such discussions as not belonging to the services encounters. In the following interview excerpt, a social worker from adult social services concludes that asking about transnational family relations feels a bit intrusive.

[Transnational family relations] do not easily come up [in service encounters] and I have often thought about how they keep in touch […] It is complicated to find this out without being intrusive. […] Of course, I should just ask if they keep contact and so on. But it does not come up automatically. (Interview)

Our interpretation is that this feeling of unease makes visible the tensions referred to above. First, it points once again to the tension between the narrow definition of family rooted in the social security system and the professional ideal of holistic practice including a service-user-based understanding of the family. Secondly, it connects to the tension that despite a psychosocial interest in broader family relations, talking about left-behind family members and hopes for family reunification without realistic possibilities to assist in the process may cause extra pain for the service user or evoke false hopes (also Nordberg et al., 2024). Given the limitations made by the narrow family definition in both social service and family reunification legislations, professionals recognise that discussions of family relations would not have any concrete consequences on the decisions about social security benefits or prospects for family reunification. Thus, for a service user, talking about transnational family relations may show up as painful, useless, or harmful if information about sending money to family members abroad would affect the issuing of social security benefits in Finland negatively.

Conclusions

In this article, we have analysed how family relations are addressed or remain unaddressed in the institutional encounters between social work professionals and migrant service users. Combining family research with a focus on service encounters and a methodological underpinning of IE provided new insights into how family takes shape in these encounters.

The negotiations on ‘family’ in the service encounters seem to become an act of balancing between a broader psychosocial understanding of family that builds on self-definition, and a stricter and narrower definition of family as a household unit of parents and under-aged children residing in Finland. The narrow definition is rooted in Finnish legislation on social security and immigration and thus becomes a strong relation of ruling that coordinates the actual encounters and what happens in them. Combined with a lack of time in the encounters, the narrow family definition often overrides the attempts to bring broader family relations into the agenda. Social workers have only little power to influence the situations of those broader family members who are not recognised by the social security system. Therefore, migrant service users may see it as useless even to touch upon these issues in the service encounters. Why talk about intimate issues if it has no consequence to the services that one gets and if the professional does not have anything else to offer than some compassionate words. On the other hand, the transfer of basic income support decisions from municipal social services to the National Insurance Institution (Kela) seems to have brought with it some space for listening to service users’ self-definitions of their families. The negotiations about who belongs to the family were activated especially when discussing service users’ hopes for family reunification. In these discussions, the narrow family definition of the Aliens Act overrules a more holistic approach stressing self-definitions.

Migrant service users may sometimes avoid talking about transnational family relations also because of painful and traumatic issues related to them. Some service user participants told us in the interviews that they want to concentrate on practical issues in the service encounters because that is all they have strength for at the moment. Touching upon transnational family relations deserves great sensitivity from the professionals: they should provide opportunities, space, and time for discussion, but at the same time avoid intrusiveness. Psychosocial processing of transnational family relations in social work encounters requires time and trust, and even if the service user was inclined to do this, it might take several meetings before they are ready for this.

In the end, the tension between the bureaucratic, narrow definition of family and the more diverse self-definitions of service users’ families is not only an issue of importance to migrant service users. Should there be more room for self-definitions in the institutional encounters as a whole? In any case, family relations are essential to human subjectivity, and if the institutional encounters focus only on what counts from a bureaucratic point of view, many important aspects of human life and subjectivity remain unseen. At times, a narrow assessment of a specific service need is enough, but if social work wants to see its strength in the view of person-in-environment (Clark et al., 2014) and claims to work with a holistic understanding, a broader and more flexible take on family relations is important. Thus, it would be important that social workers become aware and strive to change the institutional environment in which they work to create more space for individuals’ own family definitions. Moreover, power asymmetries influence what becomes seen or remains unseen in social work encounters. The social location of social work practitioners themselves further regulates their reflexive engagement and the potential for transformative resistance (Burke and Harrison, 2016). Here, the notion of anti-oppressive practice and its implementation in everyday social work practice (e.g. Danso, 2015; Morgaine and Capous-Desyllas, 2015) becomes important.

We found IE to be a useful tool for combining the local and extra-local and for analysing the connections between the street-level work with individuals and families, the organisational contexts and guidelines, and legislation (also Kuronen, 2020). In the case of a category of service users that is at the same time structurally vulnerable and politicised, these connections appear particularly important. Moreover, it was crucial that our ethnographic data included the voices of both professionals and service users, because otherwise the tensions discussed in this article would have remained invisible. Now, it was possible to map the institutional logics that guided these encounters. At the same time, the focus on the institutional dimension arguably sets some limits to the analysis. When zooming in on the ruling relations that govern the everyday knowledge and activities of the research participants, the local ethnographic description becomes less in focus (see also e.g. Malachowski et al., 2017). Moreover, the analytical approach based on cases and on following clues in the everyday world omits a broader systematic analysis of concrete cultural artefacts, such as policy documents. However, in this article, IE provided possibilities for critical analysis of institutional power beyond a narrow conceptualisation of practitioners as control agents of the state (Høgsbro, 2017; Kuronen, 2020) who may themselves have rather limited possibilities to do things differently. At its best, recognising and conceptualising the extra-local relations of the ruling may help to resist the institutional order, to promote socially just social work practices with disadvantaged service users and seek for new ways to develop the institutional practices from the inside.