Care arrangements between family and state – developing hybrid scripts of ageing in a context of migration

Abstract

Due to demographic developments, the number of immigrant families with older members is growing across Western Europe. Poised between the different approaches to care in their countries of origin and destination, such families are treading new ground as they negotiate support for older family members. Based on semi-structured interviews with 26 older and younger family members from 20 Turkish immigrant families in Denmark, this article investigates the broad variation in how these families approach caring for older people. At one end of the scale, some families (re)create a traditional Turkish approach to care in which three generations live together and the daughter-in-law is the main care provider. At the other end of the scale, some families follow the Danish approach in which families rely on public care provision. In between these two poles, many families devise a variety of care solutions for their older members that often rely on considerable support by family. Utilising Cati Coe's concept of care-scription, the analysis shows how families negotiate – and sometimes struggle over – who will shoulder which tasks. In the post-migration context, daughters-in-law are often able to retreat from their traditional role of care provider, which in turn increases expectations of daughters' care provisions. In Denmark, which is a relatively gender-equal society, Turkish daughters may seek to make their brothers take on a greater role in the provision of care for their parents, and in some families, the children also turn towards public care provisions to share the burden of parental care. Resistance from older parents may, however, short-circuit this strategy. Overall, the study points both to the difficulties faced by older immigrants with limited host-country language proficiency when utilising public European care provisions and to the complex and unsettled nature of care provisions in immigrant families in Europe today.

Introduction

In the late 1960s and early 1970s, a wave of labour migrants from a number of countries in northern Africa, the Middle East and southern Asia moved to north-western Europe. The top-sending country of labour migrants was Turkey (van Mol and de Valk, 2016). Today, half a century later, most of these initial labour migrants have passed retirement age.

In conjunction with this demographic development, the literature investigating the intersection between ageing and migration has been expanding, particularly within the last decade (see e.g. Karl and Torres, 2016; Ciobanu et al., 2017; Brandhorst et al., 2021). A central question in the research on ageing immigrants is unpacking what Oxlund (2018) has called ‘hybrid scripts of ageing’. Such hybrid scripts are bound to arise, as families with older immigrant members find themselves poised between the approaches to care prevalent in their countries of origin and of destination. This article adds to this research agenda with an analysis based on interview data from Turkish migrant families living in Denmark. Informants include both older Turkish migrants and the adult children of these migrants, who often hold a major responsibility when it comes to care provisions. The analysis utilises the concepts of care-script and care-scription (Coe, 2017) in an exploration of family processes in situations where migration brings together very different norms and practices regarding care of older people.

The two implicated countries – Denmark and Turkey – are very far apart when it comes to care of older people. In the sending country, Turkey, norms regarding who is responsible for caring for older parents are strong (Çelik et al., 2018), and 40 per cent of the population ‘strongly agree’ with the statement that ‘it is children's duty to take care of ill parents’ (World Value Survey, 2022). The cultural ideal of younger family members caring for older family members may be facilitated through patrilocal extended household living. In this tradition, which is particularly common in rural parts of Turkey, one son (most often the youngest) holds the primary responsibility for taking care of older parents, but his wife – the daughter-in-law – is generally the primary care provider (Aytaç, 1998; Adana et al., 2022). In tandem with social and economic developments, however, the importance of such living arrangements has been decreasing (Saka and Varol, 2007; Akkan, 2018), even though women still carry the primary burden of caring for older family members (Çindoglu et al., 2011; Akkan, 2018). This burden is further bolstered by the relatively low rate of employment (26%) among women (Turkstat, 2022), facilitating women's possibilities for taking on this role of carer. Moreover, when it comes to public care institutions, provisions are limited, and placing one's parents in a nursing home is typically frowned upon (Şenol and Erdem, 2017; Ar and Karanci, 2019).

In the receiving country, Denmark, the state, rather than adult children, is considered responsible for providing care for older people (Marckmann, 2021). Based on survey data, Brandt and Deindl (2013) report that while adult children in Denmark do often help their older parents, they do so more by choice and at their pleasure, rather than out of a sense of obligation. Furthermore, the number of hours where help is provided is lower than in other European countries such as Germany, Belgium, France and Austria. While intergenerational relations in Denmark thus generally remain close, just 5 per cent of the population ‘strongly agree’ with the statement that ‘it is children's duty to take care of ill parents’ (World Value Survey, 2022), making the country an outlier in European comparisons (Haberkern et al., 2015). These limited expectations of family with regard to providing care for older parents must be seen in a context where the employment rate for women, at 73 per cent, is almost as high as that for men (DST, 2022), and where publicly financed care services are universally provided based on an assessment of needs (Kjær and Siren, 2020). In practice, public care is primarily provided in the form of home-helper visits and can be both practical (e.g. cleaning and shopping) and intimate (e.g. bathing and washing). If the individual's need for care grows beyond a given level, older individuals may be offered a place in a nursing home with round-the-clock staff.

Influenced by both of these national contexts, families with older migrant members organise care in a variety of ways, from the full reproduction of traditional Turkish family support to the full integration into the public Danish system of care. Such migrant families may thus be used to explore empirically Oxlund's (2018: 75) argument that ‘the nexus of ageing and migration … brings about new imaginaries of care through human contact in concrete care arrangements’. In a world where international migration as well as population ageing are prominent features, understanding such care arrangements in immigrant populations is essential from both a scientific and a policy viewpoint (Albertini et al., 2019).

Conceptual approach

Conceptually, the analysis is inspired by Cati Coe's work on care of older people in Ghana (Coe, 2017, 2018; Thelen and Coe, 2019). Similar to Turkey, Ghana has a strong discourse of family care. However, in part due to processes of migration, care is increasingly provided by relatives other than children, as well as by neighbours or by foster children (Coe, 2017). While state provision of care is very limited, Coe observes that older Ghanaians discuss an imagined provision of institutionalised care to criticise subtly developments in a society that have made it more difficult for adult children to care for older parents (Coe, 2018).

Coe draws on Stack and Burton's (1993) terms kin-script and kin-scription in her investigation of shifting practices of care in the face of changing social and economic realities. Coining the terms care-script and care-scription, Coe argues that the strong discourse on family care glosses over the realities of care provision in individual families. The term care-script thus denotes the expectations, norms and values of care provision in a given group, with family responsibilities often being tied to gender and age. Care-scription denotes how particular family members become conscripted into undertaking particular kinds of care work in given families (Coe, 2017, 2020). Such family processes may include contestations and conflict.

In migrant families, it is likely that the care-script of the country of origin becomes challenged by the post-migration realities. At the same time, the care-script of the receiving country may seem too alien to follow. Inspired by Oxlund (2018), this situation leads me to pose the following research question:

• • How do migrant families negotiate norms, values and practical realities as members devise new and hybrid types of care arrangements?

Similar to Connidis and Kemp (2008), I not only analyse existing care arrangements, I also attend to anticipated care needs and the sometimes creative imaginaries of how future care arrangements will possibly unfold. In the analysis, I first present four cases that cover the breadth of care arrangements. Second, I use the broader interview material to explore negotiations of care-scription based on factors such as family position and gender.

Methods and data

This analysis draws on interviews with 26 individuals from 20 Turkish migrant families.¹ I carried out the interviews between 2019 and 2022 and recruited informants through a variety of network contacts as well as through snowballing. Some informants were recruited during a week-long stay in a mid-size town in western Denmark, where a well-connected Turkish woman (who had worked as a translator for decades) helped me gain access to a variety of families. Other informants were recruited in the eastern part of Denmark, as I reached out to my own network for contact to relevant informants.

All interviews, except one, were carried out face to face in a location chosen by the respondents. In most cases, I spoke with people in their homes, but two interviews were carried out in cafés and one interview was done in my office. I deliberately included families that included older family members of different ages (ranging from being in their mid-sixties to their eighties), as well as families both with and without present care needs. This broad approach was also deemed necessary as gaining access to immigrant families with high levels of care needs is often difficult (Bjerke, 2020).

I conducted the interviews in either Turkish or in Danish (or a mixture of the two languages), depending on the informant's preference. Taking a semi-structured approach, I started out by inquiring about the size of the family and where members were living. I then asked about which family members were taking care of what tasks. Depending on the context, some interviews turned into detailed narratives about present care needs, while others turned into reflections about possible future care needs in the family. The interviews mostly lasted between 45 and 60 minutes.

With the informants' permission, all interviews were recorded digitally and subsequently transcribed by the author. To protect anonymity, the informants have all been given pseudonyms and particular details in their narratives, such as a medical diagnosis or the number of siblings in a given family, may have been altered.

The 26 informants were distributed as follows: 14 were older parents (aged 63–80), ten were adult children (aged 32–50) and two were grandchildren (age 24–26, interviewed with reference to one set of grandparents). In 14 families, I interviewed one informant, e.g. a mother or a daughter. In six of the families, I interviewed two informants, e.g. an older father and his adult son, a grandmother and her adult grandchild, or an older couple together. Table 1 contains an overview of the informants.

Table 1. Overview of interview participants

Notes: T: Turkey. DK: Denmark.

Analysis

The analysis was conducted in two stages. First, I alternated between reading the interview transcripts and reading a number of scholarly publications on the topics of ageing and/or migration. After settling on using Cati Coe's concept of care-scription to focus the analysis, I conducted a directed content analysis of the interviews (Hsieh and Shannon, 2005). The analysis was done by coding the interviews for a variety of topics. One topic was present care arrangements, where I paid particular attention to narratives of how such arrangements were negotiated and potentially disputed, and how they developed over time. Another set of codes were related to the topic of expectations about future care arrangements – e.g. parents' expectations of children (and children-in-law), and expectations between siblings. I then organised the 20 families into a continuum that ran between a ‘Turkish’ and a ‘Danish’ approach to care. The analysis was supported with the use of NVivo 11 software.

Findings

The interviews reveal substantial variation, both in concrete care provisions and in imaginaries of future care arrangements. The arrangements span from practices of patrilocal extended household living akin to the traditional Turkish cultural ideal to imaginaries of using Danish care homes, should the need arise.

To cover this varied terrain, I present the care arrangements of four families, exemplifying the breadth of the material. The cases are the following, with the individuals interviewed provided in parentheses.

• • A patrilocal extended household (the father Volkan and the son Mehmet).

• • Shared care and support in a family with ample family resources (the mother Defne and the daughter Zeynep).

• • Shared care and support in a family with strained family resources (the mother Fatma and the son Yusuf).

• • Limited family support and imaginaries of public care (the mother Hülya).

Case 1: Support through patrilocal extended household living

The 76-year-old father Volkan and his 49-year-old son Mehmet lived in the same household. Mehmet's wife Banu (i.e. Volkan's daughter-in-law) was also intermittently present during the interview. The family exemplified the Turkish ‘cultural ideal’ of a patrilocal extended household (Aytaç, 1998). Mehmet described how the living arrangement had come about six years earlier:

We had plenty of space, so instead of my mum and dad living alone, I proposed we moved together. Initially, they turned down the offer. ‘We don't want to burden you’, they said. But then my wife said – sagolsun [a Turkish expression of gratitude] – ‘you won't – just come. We have plenty of space’ … And luckily they accepted, and since then, we have been together.

A few years later, when Mehmet's mother became terminally ill, she was able to remain in their home until her death. While the Danish health system supplied medical care, the family – predominantly Banu – provided non-medical care. About this arrangement, Mehmet said:

It was really good that we could all be together. Had we lived apart, my mother would have had to have been hospitalised. And she would not have felt well there.

After his wife died, Volkan remained in his son's household. In response to my question about the arrangement, he chuckled and said: ‘I am very happy with living here. They cook, they wash the clothes.’

In line with the strong filial norms typically prevalent in Turkish migrant families (de Valk and Schans, 2008; Schans and Komter, 2010), Mehmet explained that he attributed great value to upholding a Turkish family care-script:

I think it is a good thing to take care of your parents. And I tell that to my children, too: that if you do it, it is also an investment in the future … We saw our grandfather and grandmother [being cared for by family in Turkey]. And now we take care of my father and mother in the same way. And, Inshallah [God willing], our children will do the same. It becomes a healthy tradition, which we hope will continue.

Regarding processes of care-scription, Mehmet explained that it was he – as a filial son – who had invited his parents to live with him, rather than them having pushed for the arrangement. When doing so, however, Mehmet had effectively care-scribed his wife into being the major care-provider (Lorenz-Meyer and Grotheer, 2000). In the first years, Mehmet's parents helped out with the grandchildren, but this ended when Mehmet's mother fell ill. The daughter-in-law Banu, who had paid work as a cleaner, thus found herself with a heavy workload. Towards the end of the old mother's life, Banu's ability to provide care was facilitated by a Danish option for paid leave when attending to a terminally ill relation. This shows how upholding the Turkish care-script may be aided by Danish social policy provisions (for a related example, see Næss and Vabø, 2014).

Case 2: Family support with considerable resources

The 72-year-old widow Defne also received considerable family support. Akin to settlement patterns commonly found in Turkey, Defne lived in a ‘functional extended family’ (Aytaç, 1998: 242): she and three of her children all had flats in the same public housing estate. Most of her remaining six children had settled in the vicinity. Such proximate residence patterns are also seen in Asian societies, where they enable considerable degrees of intergenerational support to co-exist with the unfolding processes of modernisation and urbanisation (Croll, 2006).

Defne had mobility issues, and her children helped her regularly with cleaning, shopping and transportation. Because Defne was both illiterate and unable to speak Danish, her children also helped her with all (verbal and written) interaction with Danish authorities, including with the health system. Many older Turkish migrants need such linguistic support (Liversage, 2009; Liversage and Jakobsen, 2016; Pot et al., 2020); this is one of most common areas in which their needs differ from the situation of their majority peers.

Regarding who provided what kind of help, the 42-year-old daughter Zeynep described the siblings' situation in the following manner:

The oldest [sibling] works in cleaning, so she can't make it. And number two is taking a course, so she can't either. Then there is number three and me, we have more time, so we do it. We just come over, both of us, and then it is quickly done … We have been pushing a bit to some of the others. Some can't because they have young children, but there are others with more time on their hands.

Zeynep's description of the division of chores shows both an extensive provision of family support and the children's negotiations of who does what. These negotiations hinge on what Finch and Mason (1993) have termed ‘legitimate excuses’: that certain circumstances are seen as socially acceptable reasons for not fulfilling what could otherwise be considered a family responsibility (Connidis and Kemp, 2008).

Defne and her daughter described their family as one that has an excess of family care resources. The daughter recounted how one of her brothers, who had built himself a large house, had said: ‘Look mum, I made extra rooms. If you want to move in now, we have the space for you.’ The fact that Defne lived in her own flat, rather than in an extended household, was thus a deliberate choice. In fact, after a life with many children, Defne enjoyed being on her own. This solution was viable as she – in the functional extended family – still had help close by:

If I look one way out the window, I can see one daughter. And if I look the other way, I can see another. I can just open the window and shout, and then they will come (laughs).

Contributing to the excess of care resources in the family arrangement was also the fact that Defne's care needs were relatively low, as she could carry out many tasks herself.

Defne's family thus followed a Turkish care-script of family care provision. Regarding care-scription processes, when Defne turned down her son's offer of extended household living, her daughter-in-law was not care-scribed into providing in-house care. Instead, Defne's children – particularly her seven daughters – helped her to varying degrees depending on competing obligations in their specific life situations. Proximity had little relevance as almost all nine children lived near their mother.

Case 3: Family where resources are strained

In other families, the balance between needs and resources was less favourable. This was the case in relation to 79-year-old Aslan, who had suffered both a stroke and the onset of dementia. His health gradually deteriorating, Aslan needed help with tasks such as dressing, bathing and moving about. His main carer was his 69-year-old wife Fatma, whom I interviewed. The couple had three children, and I also interviewed the middle child – their 48-year-old son, Yusuf.

While Aslan's wife provided round-the-clock care for her husband, she got substantial support from her oldest child – a daughter – who lived nearby. Unwed and childless, this daughter largely divided her time between a demanding (but well-paid) job and supporting her parents. Besides doing tasks such as cleaning and translating, she also helped her parents financially, as they – as is often the case for older Turkish migrants – had limited financial means (Jakobsen and Pedersen, 2017). Several nights a week, the daughter slept at her parents' home. The son Yusuf also provided substantial help.

Regarding formal care, the municipality had offered the couple home help, but as Aslan resisted intimate care from anybody but his wife, the family had turned the offer down. About the present situation, 69-year-old Fatma said the following:

I have managed until now, but it is getting so difficult for me to dress him and bathe him. He is big and it is straining my body… My back and my shoulders are starting to hurt. It didn't use to be like that.

Due to Aslan's increasing need for help and Fatma's waning resources, the pressure on the children was mounting. Yusuf described how exchanges such as the following had recently begun occurring:

Sometimes, I just say [to my sister]: ‘Now you just have to take over – I just can't.’ And then my sister says: ‘You have to.’ And I say: ‘Maybe you can't leave your job, but I can't either.’

Strained by mounting care needs, Yusuf and his sister seemed to be negotiating who had the most ‘legitimate excuse’ (Finch and Mason, 1993) for not providing support at a given juncture. While both the mother and adult children wanted to include public home-helper support, the father's resistance had hitherto hindered them from doing so. The family thus still adhered – albeit with growing difficulties – to a Turkish care-script.

Case 4: Parents expecting to utilise public care provisions

The last of the four cases is 70-year-old Hülya. She was living with her 81-year-old husband, and both spouses were in quite good health. The couple had three children: two lived in a distant part of Denmark and one lived elsewhere in Europe. In contrast to the three previous families (cases 1–3) who all lived close to one another, Hülya did not have much physical interaction with her children on a day-to-day basis. Hülya also underscored that she had raised her children differently than most of her Turkish migrant peers:

I did not want to pressure my children the way the Turks living here did – controlling them all the time … My husband was more old-fashioned, so we argued quite a bit over that.

Hülya's children were thus brought up with more individual freedom than most Turkish migrant children at the time. This freedom entailed the oldest daughter being able to take a work-stay abroad to learn English. During the stay, the daughter met and married a Scotsman and had not returned to Denmark. The other two children had also married spouses of different ethnicities. Their choice of spouses thus deviated considerably from the general pattern in Turkish migrant families in Denmark, where children have predominantly married spouses of Turkish origin, either raised in Turkey or – especially in the past decade – raised in Denmark.

While Hülya was proud of her children and the good lives they were living, she did not have anyone who could stop by to provide help on a daily or weekly basis. Worn down by the unskilled factory work she had done in the 1970s, Hülya was gradually beginning to need help:

My arms hurt – especially my shoulders. So it's tough for me to look after the home. But I don't know if they [the municipality] will give [home] help … If someone came every other week and vacuumed, it would be great … I would like to talk with them about it – but I don't know who I'm supposed to talk to.

Hülya thus expressed no reservations about following the Danish care-script and utilising public home-help provision. Her access to such help was, however, impeded by her limited Danish skills and her unfamiliarity with Danish care provisions. Regarding possible future care needs, Hülya said that two of her children had repeatedly suggested that she and her husband moved closer to them. The couple were contemplating doing so in the foreseeable future. This would, however, not be done with an expectation that their children would provide care for them. As Hülya explained:

I might live near them [her children] – but never with them. And if need be, I'll move to a nursing home. It is not an option that they look after you. That is not a possibility for them. They have their own life.

Hülya's emphasis on the fact that employment makes it untenable for children to care for parents in a European context is found across studies on older labour immigrants (Naldemirci, 2013; Næss and Vabø, 2014; Palmberger, 2019). Regarding adherence to either the Turkish (family) care-script or the Danish (public) care-script, Hülya thus expressed her intention to follow the latter.

A broad variety of approaches to care

The above four cases show different ways in which Turkish migrant families handle – or foresee handling – the care needs of older family members. The cases show that needs as well as resources may vary substantially both between different families and within the same family over time. The cases also show both different attitudes towards the Turkish and the Danish care-scripts, and the unsettled and negotiated nature of care provisions in individual families.

While family care in many cases is accomplished without overt negotiations between family members (Connidis and Kemp, 2008), the interviews contained numerous references to explicit negotiations over care obligations. This may be because the hybrid care arrangements were under construction (Oxlund, 2018) in the case families. In the next sections, I explore negotiations over obligations and commitments in the family care arrangements, as different members sought to care-scribe (Coe, 2017) each other into greater participation, or as family members resisted processes of (attempted) care-scription. I structure the analysis along different relational lines.

I look, first, at negotiations between children and children-in-law. Second, I attend to negotiations between male and female adult children and grandchildren. Third, I turn to negotiations relevant to the division of work between the family and the state. I end by discussing different imaginaries of future care in Turkish migrant families in Denmark today.

Theme 1: Division of work between children and children-in-law

As mentioned earlier, the primary care provider in a traditional Turkish care-script is a co-resident daughter-in-law. In some Turkish villages, daughters-in-law may have few resources to resist such care-scription, even if they wanted to (Liversage, 2019: 912). In contexts of migration, daughters-in-law from countries such as Turkey and Morocco may also provide substantial care for older migrant family members with high support needs (van Wezel et al., 2016; Berdai Chaouni and De Donder, 2019).

Case 1, where Banu cared for her parents-in-law, exemplifies that such family arrangements exist in Denmark. In fact, Danish register data analysis shows that 8 per cent of all 65–74-year-old Turkish migrants live in a household with a son who is married and/or has children. This type of living arrangement thus holds some significance in Denmark today (Liversage and Rytter, 2021).

The interviews showed, however, that support provided by a daughter-in-law could not be taken for granted. Banu's willingness to provide care for her father-in-law may be due to two facts: first, she is a marriage migrant (hence she had her primary socialisation in a Turkish village); second, her father-in-law is also her uncle, as Banu's husband is also her cousin. When Banu and Mehmet got married in 1989, most spouses in Turkish migrant families were marriage migrants to Europe (Lievens, 1999; Liversage and Jakobsen, 2010), and a substantial minority of marriages was contracted between relatives such as cousins (Reniers, 2001; Suren et al., 2007). Since then, however, ever-stricter immigration legislation has contributed to decreasing the number of marriage migrant spouses (Carol et al., 2014). Furthermore, Danish legislation has made it next to impossible to obtain a visa for a spouse who is also a cousin (Bech and Mouritsen, 2013; Liversage and Rytter, 2015).

Several informants explained that this shift away from marriage migrant spouses had contributed to making extended household living unfeasible in Denmark. As 65-year-old Sezen put it:

Some [young women] are used to [living in extended families] – the ones who were ordered from Turkey. They still have the same culture. But not the ones who grew up here [in Denmark], and educated themselves. And the girls did that a lot.

Sometimes older parents themselves are also reluctant to enter into an extended-household living arrangement, as the adult daughter Leyla explained:

The older generation aims less and less for living together. Because you hear from others that it just doesn't work. It creates so many conflicts. And when you've finally managed to get your son married to a nice girl, you don't want to ruin it.

The ‘ruin’ in this quote points to a son's wife possibly divorcing him, rather than living with her parents-in-law against her will (Liversage, 2019). As wives in Turkish families generally have easier access to leaving their husbands in Denmark than in Turkey (Liversage, 2012), this dynamic may have contributed to both decreasing shares of extended household living and to reducing the expectations that daughters-in-law should provide care.

An example of the negotiated dynamic of this process is seen in Dilara. She had married a man who, like herself, was born to Turkish migrant parents in Denmark. In contrast to marriage migrant wives, whose parents live in Turkey (Liversage, 2017), Dilara thus had both her own parents and her parents-in-law close by. Considerably involved in supporting her own parents, Dilara came to meet ‘unexpected’ and ‘fierce’ expectations that she should also support her parents-in-law. Juggling both a career and children of her own, the situation developed as follows:

I wasn't able to handle it in the end … So I decided to withdraw from some tasks, so now I don't do as much as I used to [for the parents-in-law] … For example, I don't help them make phone calls [in Danish] anymore, because to my mind, my husband can do that. We discussed that – just the two of us … Before, he could just say: ‘Dilara can do that’ – on my behalf. But not anymore.

Similar to Banu's husband Mehmet (in case 1), Dilara's husband seemingly also felt entitled to make dispositions on his wife's behalf. Dilara began actively resisting this care-scription, however, and was indeed able to push some of the tasks away from herself and on to her husband. Her doing so points out how the agency of (often self-confident and well-educated) women raised in Denmark may contribute to changing patterns of care in migrant families. Such actions could also have a domino effect; with Dilara no longer supporting her in-laws much, not only Dilara's husband, but also his two sisters, had to shoulder more work. That brings us to the next topic regarding changing patterns of care-scription: negotiations regarding the division of work between siblings, particularly the gendered negotiations between sisters and brothers.

Theme 2: Negotiating the division of work between siblings

The division of work between siblings is centrally influenced by how many siblings make up the sibling flock. In the families included in this study, the older migrants on average had 4.5 children, and two families had as many as nine or ten children. The number of children was highest in families with parents older than 75. For comparison, in Turkey, the average number of children per family was six in 1965, around the time where the labour migration to Denmark began (World Bank, 2022).

Siblings took on different roles depending on their age, gender and other factors. Generally, sisters spent more time than brothers supporting parents, in line with a gendered pattern documented in a variety of countries across the world (Hequembourg and Brallier, 2005; Connidis and Kemp, 2008; Haberkern et al., 2015).

In some families – as in Defne's (case 2) – negotiations about who was to do what were primarily between sisters. In other families, siblings – and particularly sisters – found the division of work unfair, and they objected to the perceived free-riding of some of their siblings (typically brothers). An example of this is seen in the following comments by the gender-conscious Dilara:

It's always been us [eight] girls who've cared for our parents. But conflicts are growing now because my younger brother was never given any responsibility. You hardly dared ask him. But then, you know: people work, and he is a student. He could very well drive [parents] some place once in a while. After all, he has a car because they gave him one.

Dilara recounted that, most of the time, she and her sisters would not confront the brother with their resentment. However, at times

…we also say it to his face, and then he gets so defensive. He feels that he has no time and no energy to do anything besides his exams. And then we think: ‘We both studied and worked, and did all this practical stuff for our parents anyway. And we have children to attend to, too.’

In a life with many competing demands, Dilara and her sisters wanted their brother to provide what they considered his fair share of help as a sibling. The brother, however, seemingly tried to hold on to his privileges. This situation led to arguments over what should count as ‘legitimate excuses’ (Finch and Mason, 1993), and whether a perceived double standard for sons compared with daughters was acceptable or not.

The sisters' pressure had, in fact, begun to pay off. Dilara explained that ‘my little brother has begun to participate more. Even though he calls us up all the time, asking if it was right, what he just did (laughs)’. While the brother's calls could be both genuine requests for guidance and demonstrations of him being a poor care provider, he had nevertheless – albeit with some effort – been successfully care-scribed into providing parental support. Also Yusuf (case 3) explained that he and his unmarried sister were presently putting pressure on their younger brother to make him shoulder more of the growing care load. Sisters' demands for a more equal division of care work can also be linked to the Danish context where – in line with the high female employment rate – gender differences in family care provision are among the smallest in Europe (Haberkern et al., 2015).

While Yusuf's and Dilara's brothers were both being positioned to engage more in future support for their ailing parents, the situation was different for Ipek. When her mother fell ill some years ago, Ipek and her three sisters and two brothers had prioritised keeping their mother in her home. While the health system contributed medical support, Ipek's father together with Ipek and her siblings shouldered the mother's personal care, with especially Ipek and one other sister pulling a heavy care load both day and night. Ipek described her brothers' help in the following manner:

They are there, but it's like: ‘As long as it's not my business, I don't do anything.’ That's their mentality … My mother needed so much help. She was partially paralysed … And the two brothers, they were more of a burden. It does something to you, emotionally. You have two brothers, and they ought to be there. And then there is disappointment. And anger.

Such stories of brothers not pulling their weight, while other siblings – mainly sisters – try to make them engage more, point to a situation where the family disagrees about who is responsible for providing care for family members. The situation is likely affected both by the high female labour force participation rate in Denmark and by the country's higher level of gender equality as compared with Turkey (World Economic Forum, 2021). The existence of such arguments thus yields insights into how siblings may try to affect each other's care-giving – an issue about which our present research knowledge is limited (Tolkacheva et al., 2010).

Theme 3: Negotiating the division between family and public care provisions

In most of the 20 families, family members talked about handling care needs within the family. This aligns with research that shows that older immigrants often make less use of public care provisions than their peers from the majority population (Denktas, 2011; Hansen, 2014).

In families who were not yet eligible for public care, some of the older migrants expressed that they expected to make use of home-help provisions should the need arise. In other cases, parents (as the old couple Aslan and Fatma in case 3) had been offered help by the authorities, but had turned the offer down, sometimes to the dismay of their adult children. Or they (as Hülya in case 4) found it difficult to access public care provisions, in part due to difficulties speaking the Danish language. Of the few interviewees who were presently receiving home help, some were relatively satisfied, while others voiced grievances over the visits being too few and too short, as well as irritation over the constant helper turnover and the ‘inadequate’ standard of cleaning. Such complaints reflect the fact that the level of municipal home help in Denmark has declined substantially in recent years (Fredslund, 2021; Rostgaard et al., 2022).

Regarding nursing homes, a large share of the informants expressed the view that they would avoid using them if possible. According to Yusuf (case 3), the general view on the matter was that: ‘If you put your parents in a nursing home, then you have not done your duty as a child.’ In line with such norms, only one informant had first-hand knowledge of an older Turkish migrant who had moved to a nursing home. This informant was Dilara, who said that it had happened to a relative's father after he developed severe dementia. Dilara's mother had been very upset about it, and ‘the elders mocked [the children]. They could not understand why they could not look after their sick father. But he was difficult’. Research documents that negative views on nursing homes seem particularly entrenched among lesser-skilled immigrants from predominantly Muslim countries (Næss and Moen, 2015; Ahaddour et al., 2016; Albertini and Mantovani, 2022).

Having witnessed both the challenges of family care provisions and of turning to Danish nursing homes, some informants imagined other ways in which care could be provided. Such imaginaries – which might, and might not, become a reality – are a central way in which practices of care of older people may be transformed in response to changing political, social and economic conditions (Coe, 2017). Such imaginaries are the topic of the next section.

Imaginaries of future help provisions between the family and the state

Several of the adult children were adamant about never putting their parents in a nursing home. When asked how they would cope if one day their parents needed large amounts of care, several of them said they had no clear answer, but that they would wait and see how things turned out.

One of the adult children who expressed this ‘we'll-wait-and-see attitude’ was Dilara, who had seven sisters and one brother. She admitted: ‘We haven't really talked about it. But if it comes to that, it will probably be us alternately taking them in.’ In her imaginary about future care, Dilara points to an alternative to nursing homes, extended households and shared provisions of care in the older parents' home. This alternative is a model where children share the care load, most often caring for a bereaved and thus single parent through a rotation scheme where the parent alternates between living with each of their adult children. Such arrangements are one way of trying to retain family care, but without overburdening individual family members (Nielsen et al., 2021). This type of arrangement may, however, turn out to be unsatisfactory in practice and be replaced with other types of family care arrangements, testifying to the dynamic and contingent character of immigrant families' hybrid scripts of care (for two examples, see Ismail, 2021; Liversage and Ismail, 2022).

Another imaginary of future care came from 36-year-old Gül, who was a university graduate. Her parents were in their mid-sixties and in good health. Gül, who was raised partially in Turkey, stated:

I know for sure that my parents don't want to live in a nursing home … I also feel it deeply within myself – that if your parents end up in a nursing home instead of you looking after them, you have failed.

In her case, patrilocal household living was out of the question. While Gül did have a brother, she was sure that he and his wife – who was not a marriage migrant – would not volunteer to take in their parents. Instead, Gül (whose husband was of non-Turkish migrant origin) considered how she could provide extended family living herself, through the inclusion of outside resources:

The municipality may come and help out while I am at work. Or I might hire someone to do it – try to solve it that way. If we can afford it, I might work shorter hours. We may also buy a bigger house. There are a lot of thoughts and considerations. You don't quite know what you end up with.

Trying to imagine a solution where she would both be able to work (a prerogative in Danish society) and keep her parents out of institutional care, Gül depicts a family care imaginary that includes elements from not only the family, but also from the state and the market (Esping-Andersen, 1990). Gül's comment about subcontracting filial obligations to either hired daytime staff or publicly employed home-helpers has clear affinities with the care arrangements that Lan (2002) observed in Chinese families in the United States of America, and shows the patterned ways in which migrant families may seek to create new types of care provisions in a context of migration.

While some informants thus tried to imagine new forms of family care, others imagined Danish public care provisions that were better suited to accommodating older migrants' needs. A central element was sharing the Turkish (or Kurdish) language, not only with the staff but also with other older migrants. As Ipek, an adult daughter who also worked as a translator, imagined:

I went to translate for an old man. He lived in an assisted living facility – not a nursing home. It was really nice, and I thought: ‘If only you could make a place like this for our elders.’ You have a home. An elevator. But there's also a nurse. That would be great. Like a small village, where you could visit each other. And nurses, who could speak Turkish.

Ipek imagined that this type of accommodation could both protect the older migrants from the loneliness that many of them experience (Fokkema and Naderi, 2013; van Tilburg and Fokkema, 2021) and alleviate adult children from their care burden. Other studies document Turkish migrants elsewhere in Europe who envision similar kinds of care solutions, where individuals who share the same language and history can receive public support in old age (Naldemirci, 2013; Palmberger, 2019).

Such care provisions are not available in Denmark today. Thus, adult children who are unable to provide sufficient family care risk being stuck in an impasse. This was the case for 41-year-old Melek. She and her only sibling, a sister, had recently learned that their 79-year-old widowed mother had early stage Alzheimer's. Even though her mother, who lived alone, was still able to take care of herself in most regards except for situations that required Danish skills, Melek was unsure of what they would do if her mother's situation deteriorated, as she explained in the interview:

I think – damn! What if it goes totally downhill? … I don't know what the options are – can you get help in the home? … And if it ends with a nursing home, what are the options? Her Danish is deteriorating, and it makes my heart break to think about her left there, with nobody to communicate with … But I just won't have her live here. She would make me crazy … And we don't have the space – and we have stairs. My sister hasn't even considered the thought [of our mother living with her]: ‘No, I am certainly not going to do that!’ And my sister is right – you shouldn't. She [mother] might think: ‘You ought to.’ But really.

Ambivalently placed between a Turkish and a Danish care-script, Melek has considered her mother's future care needs but has not found an obvious route ahead. In contrast to Gül, Melek is unable to image her mother living with her; and in contrast to Dilara, she does not have many siblings with whom she can share the care load. Instead, Melek seems caught between the Scylla and Charybdis of maternal co-habitation and a Danish nursing home unsuited to her mother's needs. If Melek's only sibling would take in their mother, this would solve Melek's conundrum in the same way that Gül's willingness to provide family care in her home would unburden Gül's brother. Melek's sister, however, has no such intentions.

In the quote, Melek argues with the internalised voice of her mother and her Turkish background – a voice seeking to care-scribe Melek into care provisions through stating that Melek ‘ought’ to provide care (de Valk and Schans, 2008). Resisting these voices, Melek points out both the practical and emotional reasons for not doing so, conjuring up an array of what may be considered ‘legitimate excuses’ (Finch and Mason, 1993). Time has yet to tell how the care arrangement for Melek's mother will be, but at present no good solutions seem in sight.

Limitations of the study

Using qualitative methodology, this study provides insights into a variety of care arrangements among Turkish immigrants in Denmark, but yields no information about how common or rare such arrangements might be in the broader population of Turkish immigrants. Furthermore, the relatively wide variation in informants included in this study limits the detail in which individual topics can be explored. The breadth of the study (including families both with and without present care needs, and including the perspectives of different types of family members) can, however, also be considered a strength; this approach enables an exploration of the broader picture of post-migration adaptation to a new context.

Another limitation is the study's focus on one group (immigrants from Turkey) in one context (Denmark). Hence, more research could be directed towards other immigrant groups in other contexts. For example, Danish register data analysis has shown that co-habitation between older parents and adult sons with wives/children is considerably more common among Pakistani immigrants, but rarer in families from Iraq and Lebanon, for example (Rytter et al., 2021). While Pakistani immigrants also arrived as labour migrants with subsequent chain migration over several decades, the immigrants from Iraq and Lebanon arrived as refugees more recently (often in the 1990s and 2000s). Other qualitative studies could thus explore similarities and differences in care for older individuals in other immigrant groups, where both home country care preferences and trajectories of migration as well as trajectories of host-country inclusion differ from the situation among the Turkish immigrants investigated in this study.

Conclusion

This interview-based study with older migrants and their kin adds to our knowledge of care arrangements in migrant families. Overall, the study shows a wide variation of care provisions – actual as well as imagined – spanning from the (re)creation of the traditional Turkish cultural ideal of extended household living to a reliance on public Danish care provisions. Similar to the situation in Ghana described by Cati Coe (2017, 2018), older and younger family members try to devise workable solutions to providing care in a changing environment. In contrast to the Ghanaian situation, the Turkish families' migration to Denmark has radically changed the environment in which the families live, most likely speeding up processes of change. The many different solutions devised thus attest to the flexibility and dynamism in the creation of care arrangements in a context of migration.

In many families, the general Turkish care-script seems upheld at present. Underneath the narrative of family support, however, the dynamics of care provisions are shifting, as daughters-in-law are able to resist processes of care-scription and expectations to them are dwindling. Due to the gendering of care, the retreat of the daughters-in-law increases the load on daughters, who, in turn, may try to care-scribe brothers into participating more.

Families may also seek to include public care provision in their care arrangements. While some express clear reservations with regard to nursing homes – much in alignment with country-of-origin norms – others are more open to receiving this type of institutionalised care, in order not to burden their children's busy lives (Naldemirci, 2013). The latter opinion seems more often to be voiced by parents with some educational qualifications.

Overall, children (and even grandchildren) may provide substantial amounts of care. While some of the informants report being happy to do so, others experience a substantial care strain, particularly when needs increase and/or the children themselves face competing demands on their time and energy. This may lead to negotiations between parents and adult children over the acceptance – or not – of public care provision. An important impediment to using public care provision is, however, the low Danish-language proficiency of many of the older migrants.

These negotiations include adult children considering whether they indeed ‘ought’ to provide high levels of parental care (de Valk and Schans, 2008) as well as parents retreating from such expectations. These processes, as well as the ways in which increasing care demands may deplete family resources, all indicate a move towards a greater inclusion of public care provision in Turkish migrant families.

Overall, the Turkish migrant families may gradually be becoming more receptive to including public care and support in their care arrangements for older family members. One impetus for such a change is the higher level of Danish-language proficiency of the younger generation of older migrants (below 70–75 years), as compared to their old-old peers, making public care provisions more accessible for them. Furthermore, several of the adult children interviewed – some of whom were around 50 – said that they themselves had no expectations that their children should care for them, and that they would happily move to a nursing home. Over time, the negotiations and tensions over which family members should be care-scribed into doing various tasks may thus be configured in new ways as Turkish migrant families gradually reorient themselves from the Turkish care-script to the Danish instead. In some families, however – and also due to a decline in public care provision in Denmark – family care may be retained, possibly in new and creative configurations.