Intergenerational care in corona times: Practices of care in Swedish families during the pandemic

Objective: This paper analyses intergenerational relationships in Sweden during the corona pandemic, with a special focus on practices of care. The research question is: How is care between generations – between grandparents, adult children and grandchildren – done during pandemic conditions? Background: In Sweden, where an extensive welfare state provides affordable childand eldercare, the corona strategy of generational separation has still affected family practices of care between generations. In this article we analyse narratives of intergenerational care, taking our point of departure in theories of personal life (Smart 2007), relationality (Mason 2004), and care as sentient activity (Mason 1996). Method: The paper draws on a qualitative interview study with grandparents (n=30), adult children (n=12) and grandchildren (n=12), with data collection taking place shortly before and during the coronavirus pandemic. Results: The study detects the reciprocal and complex ways in which care between generations takes place. When people relate their experiences, strategies for new ways of doing care are at the centre, involving creative ways of negotiating distance and risk, all marked by both worry and relief. Conclusion: The pandemic condition becomes a ‘filter’ affecting and leading to a reformulation of practices of care, from taken-for-granted co-presence narratives, into narratives of relational participation resulting in an overall heightened awareness of the importance and difficulties of intergenerational care practices. The study concludes that a strong welfare state does not translate into complete autonomy or independence; rather, people continue to live ‘linked lives’.


Introduction
The COVID-19 (or corona) pandemic has brought intergenerational relationships into focus in new ways. In many national contexts, lockdowns and restrictions have led to forced generational separation in families (Prime et al. 2020;Cantillon et al. 2021). In Sweden, the corona strategy stated that all citizens over the age of 70 were to avoid contact with others, and, especially, to stay away from interactions with grandchildren. This affected practices of care and emotional support in personal and kin relationships, and especially between generations (Iversen et al. 2020;Kulin et al. 2021). The crisis has brought new attention to issues of care in general (Daly 2021), and the significance of care between generations in Sweden in particular, questioning sociological theories depicting the country as the 'most individualized' society in the world where the expansive welfare state is seen as having led to a 'defamilialization' (Esping-Andersen 1999) resulting in weakening intergenerational ties (Bauman 2003;Berggren & Trägårdh 2006). Previous research shows that the involvement of Swedish grandparents in the everyday care of their adult children's families has increased in recent decades (Hank & Buber 2009), as has the engagement of adult children in the everyday care of grandparents (Ulmanen 2015). These engagements have been challenged during the pandemic, with multifaceted consequences (Kivi et al. 2020;Iversen et al. 2020;Cantillon et al. 2021).
This paper emerges from a qualitative study of intergenerational care, consisting of interviews with grandparents, adult children and grandchildren (in all, 54 participants), where data gathering took place right before and during the pandemic. The initial aim of the research project was to examine care biographies of different generations within families, related to developments in the post-war Swedish welfare state, as well as care practices today. Two months into data collection, the pandemic struck, and interviews inevitably started to centre on the possibilities of intergenerational care under the new conditions. The aim of this article is to analyse these narratives, with the overarching research question: How is care between generations -between grandparents, adult children and grandchildren -in Sweden done during pandemic conditions? Analysing narratives of care from a theoretical understanding of care as a reciprocal sentient activity (Mason 1996), and of narratives of family and kin relationships as negotiated and relational (Finch & Mason 1993, 2000Mason 2004;May & Nordqvist 2019), we argue that the corona pandemic becomes a 'filter' over practices of care, resulting in strain and changed strategies of care, as well as worry and relief for the actors involved in intergenerational relationships. In its capacity as filter, the corona pandemic enables the continued significance of intergenerational care relations in Sweden to emerge in new ways, adding to a more general questioning of assumptions of 'detraditionalization', and pointing towards the dangers of ageism (Ayalon et al. 2020) and adult-centrism in attempts to tackle the crisis.

The Swedish corona strategy and intergenerational relations
The Swedish strategy of tackling the coronavirus pandemic has been widely debated, both within the country and internationally. While often portrayed in the media as weak and passive, researchers have argued that the strategy is in Sweden's well-established tradition of governance, of placing authority and trust in independent agencies (Olofsson et al. 2020). Unlike other countries, including its closest neighbours Norway and Denmark, Sweden could not at the time of the outbreak, enforce lockdown-style measures, as the constitution prohibits this unless the country is at war. Thus, the strategy outlined by the Public Health Agency focused on individual responsibility and encouraging individuals to follow public health recommendations. 1 The strategy has been described as a balancing approach, reflecting a wider view on public health, where the consequences of epidemic control measures are weighed against the 'risk of excessive social and economic disruption and to protect public health and well-being in general' (Olofsson et al. 2020; see also Corona Commission Report 2020:146).
Of particular importance are the special recommendations for people over the age of 70: from 15 March to 22 October 2020, this group was strongly recommended to minimize social contacts, and families were advised to not visit elderly relatives. When the special recommendations were revoked in October, it was declared that all citizens, regardless of age, were equally responsible for protecting themselves and others. The reason for the change was the reported negative health effects on the elderly (Public Health Agency 2020) and thus reflected the aforementioned balancing approach. Other pandemic measures also affected the doing of intergenerational care, such as travel bans and restrictions on visits to eldercare homes (Szebehely 2020), but, as the analysis of our data will show, the most significant for intergenerational care practices was the recommendation to minimize contact with the elderly.
From November 2020 onwards, new and stronger recommendations for all continued to affect contacts between generations. When recommendations were issued, the Swedish prime minister, Stefan Löfvén, declared in an address to the nation that social contacts were to be limited to the same household, and underscored the continued threat that the disease posed to the older generation: You should only meet with people in your own household. If you live alone, choose one or a maximum of two friends to spend time with. But continue to keep a distance [...] A grandparent you infect may die. It is for these people that you will make sacrifices.
The speech reflects the fear that has characterized intergenerational contacts during the last year. As our analysis will show, spending time together has been experienced as something that potentially makes you responsible for your elderly relatives' survival.

1
In early 2021, a new and temporary 'pandemic law' was put in place, enabling more lockdown-style measures (Bill 2021:4).
Moreover, the prime minister's recommendation of only meeting people within one's household, or two selected friends, opens up conscious negotiations of who belongs to your 'corona bubble', as it has popularly been labelled, and who is excluded. Critics have, from the very start, pointed out that the Swedish corona strategy does not take into consideration that possibilities of being a 'responsible individual' can be unequal depending on class and ethnicity. To be 'responsible' and stay safe, e.g., by working from home, is a privilege for some (Swedish Trade Union Confederation 2021). Studies have shown that citizens with recent immigrant backgrounds are over-represented in reported deaths (Hansson et al. 2020: 4). Several international studies have also warned of the risks of ageism and intergenerational divisions following the corona response, and of 'an increasing portrayal of those over the age of 70 as being all alike with regard to being helpless, frail, and unable to contribute to society' (Ayalon et al. 2020: 49;Cantillon et al. 2021). The Swedish strategy of recommending isolation for the 70+ group can indeed be seen as an expression of ageism. As Olofsson et al. (2020) argue, while the strategy in general has placed strong trust in citizens, the elderly are treated differently and are 'largely framed as victims to be protected from the spread, rather than active agents to be involved in implementing the strategy.' A need for attention to 'the complex intersectionalities that may put some people at a greater risk for the direct and indirect negative consequences of COVID-19' (Ayalon et al. 2020: 51) is reflected in the few empirical studies that have so far been carried out. These show that while age is certainly a determining risk factor for becoming severely ill with corona, other groups have been more affected in other ways. For instance, younger adults have proven more at risk of psychological distress and loneliness caused by lockdowns than older adults (Losada-Baltar et al. 2021), and a longitudinal quantitative study of life satisfaction, financial satisfaction, self-rated health and loneliness in Sweden surprisingly showed that in March 2020, early in the pandemic, Swedish older adults rated their wellbeing as high as, or even higher than, previous years (Kivi et al. 2020: 7). However, a qualitative study of Swedish phone support lines for the elderly (Iversen et al. 2020 shows that personal communities, referring to intense and close relationships in the everyday, have been greatly affected and reductions in these are a major cause for worry. Most studies relating to corona so far have focused on its health aspects, and generally call attention to the negative effects on individuals, often related to psychological wellbeing. But, as  show, changes in social and personal relationships have mattered immensely (see also Magnusson et al. 2021). This is brought to our attention by Cantillon et al. (2021) in their argument for the need to focus on the effects of corona recommendations on grandparenting practices. While recognizing differences in the organization of childcare across national contexts, the authors argue that the 'stay-athome' message for the pandemic greatly impacted on grandparents' support to adult children's families. By temporarily suspending this support, this messaging 'demonstrated that grandparents, many of whom may be older people, are far from being simply "recipients of care" but rather provide a crucial and active contribution to society as part of intergenerational family support' (Prime et al. 2020:8;Cantillon et al. 2021: 11).
Swedish families are, in international comparisons, less dependent upon everyday practical grandparental support for work/family balance, due to well-established day-care facilities which, along with schools for younger children (up to age 15), remained open during the pandemic. Nevertheless, even in Sweden, the social and relational consequences of the pandemic for intergenerational care practices have been significant.

Theoretical points of departure
When welfare states such as Sweden provide for basic care of the elderly and children, this changes relationships within families, but not necessarily in ways that make people less connected, less involved or more 'independent' of each other. To understand this, and to capture the doings of care between generations, a theoretical point of departure is needed that puts relationality at the centre. As Twamley et al. (2021) argue, 'family practice' (Morgan 2011) and 'personal life' (Smart 2007;May & Nordqvist 2019) perspectives are key to the 'relational turn' in social sciences on families and relationships, and should be seen in part as 'a repudiation of the extreme ends of the detraditionalization argument, which fails to take into account empirical work demonstrating the enduring importance of kith and kin in our lives' (Twamley et al. 2021:4;May 2015). By conceptualizing family as something constituted in and through 'doings', rather than defined through the noun of The Family, the family practice perspective enables analysis of everyday activities, of that which 'appears to be trivial', and which are 'given meaning through being grouped together under one single label, that of family' (Morgan 2011: 6). Building on this, Smart (2007) suggests the concept of 'personal life', to open up further the possibilities of a sociological study of relationships of importance to people, beyond pre-defined categories. Key elements in these lines of theorizing are their focus on activity and process (Roseneil & Ketokivi 2016: 44-45), and, in contrast to individualization theories (Giddens 1992;Beck & Beck-Gernsheim 2002), not adhering to notions of an autonomous individual. People live 'linked' or 'relational lives', Smart argues, and to understand 'individual life' we need to grasp 'the lives that it crosses over with, impedes or is impeded by, or run in parallel to' (May & Nordquist 2019: 9). This, in turn, is done within, and fundamentally relates to, the social context: how we relate to one another is 'shaped through layers of meaning, history, biography, and emotionality that resides within and between people' (May & Nordquist 2019: 9).
Jennifer Mason (2004) offers helpful theoretical tools for detecting the nuances of relationality in people's narratives. Peoples' biographies always demonstrate 'the significance of context, contingency, constraint and opportunity' (Mason 2004: 166), over abstract strategies for how and why one's life has turned out the way it did. When asked about reasons for making a particular decision in life, people are 'usually unable or unwilling to offer one reason' and instead tell 'complex and lengthy stories', within which four different 'relational layers' can be identified (Mason 2004: 166). In narratives of relational inclusion and co-presence, there is a taken-for-granted co-presence and inclusion of kin, to the extent that 'the narratives tell a consensual "family story"' (Mason 2004: 168).
In narratives of relational participation, the common theme is 'explicit reference to open discussion and negotiation' and making decisions with key others, with 'full consideration of all the circumstances as seen by each partner' (Mason 2004: 171). Here, the narratives are often 'characterised by "we-speak", in which mutual co-operation and participation between partners' are 'displayed as moral virtues' (Mason 2004: 169). While Mason is careful to observe that all layers of relational narratives entail both positive and negative elements, she singles out one layer where the negative is at the centre: narratives of relational constraint and conflict. In these, relationships are portrayed as destructive, and negotiations are fraught with conflict, making the very relationality of people's practices a major source of resentment (2004: 172). Finally, in narratives of relational individualism, a strong 'self' is put at the centre, and people emphasize individual control and agency, but even here relational practices emerge, as the choices made are still related to others (Mason 2004: 176-177).
A central feature in narratives of relationality is care. In line with our understanding of family practices, we depart from the notion of care that emphasizes activities and processes. While the feminist analytical division of care into labour and emotion (Finch & Groves 1983;Graham, 1983) served an important purpose when first articulated, it is insufficient to capture the full complexity of care (Tronto 1993;Eldén 2013;Eldén & Anving 2019). Conceptualizing care as a sentient activity (Mason 1996) enables to make visible the 'thinking' and 'knowing' activities that actors need to engage in when in care situations: [A]ttending to, noticing, hearing, being attuned to, seeing, constructing, interpreting, studying, exercising an interest in [the] needs, health, wellbeing, behaviours, likes and dislikes, moods, individuality, character, relationships [...] of specific others [or] thinking through, working out, organizing, planning, orchestrating relationships between oneself and others [...and] relationships between others. (Mason 1996: 27) Emotions in care are not 'states-of-being', but activities, Mason argues; to worry for someone (1996: 29), for example, entails an array of the activities listed in the quote above, or to engage in play with a child can, under some circumstances, be seen as a caring activity, especially from the perspective of a child (Eldén & Anving 2019: 101;Schwartzman 1978). Sentient activities are a significant part of doing care. They are often demanding and gruelling, but rarely acknowledged and often underestimated, even by the people doing them (comp. DeVault 1991; Anving 2012).
Mason further argues that a commitment to care is always developed out of processes of negotiation in specific relationships (Mason 1996: 24;Finch and Mason 1993). This is not to say that caring is unaffected by moral and cultural norms of what good care should be, nor of social structures such as gender or genealogy, but the influences of these are not straightforward (Mason 1996: 25). Seeing care as a relational activity allows a focus on the ways in which care is done reciprocally, avoiding too singular a focus on one party as a 'giver' (active) and the other as a 'receiver' (passive) of care (Waerness 1984;Eldén 2016). This motivates our inclusion and conceptualization of all three generations: we are interested in doings of care in and between all parties in all directions: grandparents, adult children and grandchildren. While cultural norms and structural positions, such as being a woman or a grandmother, might depict one as 'more caring' in a particular context and time, and while they might very well affect how one does and narrates care (in relation to, for example, a grandchild), the specific contours and circumstances of a particular care situation will always involve reciprocal negotiations with specific others (Finch & Mason, 1993).
The perspectives of family practices/personal life, relationality and care as a sentient activity allow us to zoom in on the complex ways in which intergenerational care is done and narrated in Sweden today. Taking as the point of departure our overarching research question of how care is done between generations during the pandemic, our theoretical tools allow us to focus specifically on care as emotional and reciprocal activity, and to capture the ways in which the condition of the corona pandemic affects, and necessitates re-formulations of relationality in the lives of grandparents, adult children and grandchildren.

Study, methods and analysis
This article emerges from an ongoing qualitative study of intergenerational care, with data gathering taking place right before and during the pandemic. The data consist of interviews with grandparents, adult children, and grandchildren (including step, bonus 2 and family-of-choice relationships); in all, 54 participants (see Table 1). 3 The grandparents' interviews were the point of departure for data collection, with the chosen cohort of participants mainly born between 1945 and 1955. Thirty grandparents of different class and ethnic backgrounds participated, 20 women and 10 men. The grandparents were recruited through associations and activity centres for pensioners, posters, online and social media advertisements, and through snowballing. Because we started recruitment with grandparents, and because data collection is still ongoing, this group's narratives are predominant in the analysis.
Contact with adult children (36-53 years old; nine women, three men) and grandchildren (5-19 years old; eight girls, four boys) was enabled through the grandparent participants, but as some were hesitant to participate, and some in the grandparent generation did not want us to contact their relatives, we have also recruited participants from these generations independently, primarily through advertisements on websites and in social media, and through snowball techniques. In all, there are 30 related relations to another generation in the study. 4 The interviews with grandparents (of which four were couple interviews) and adult children focused on individual biographies and care practices and relationships today. 5  The analysis is not dependent upon respondents being related, as narratives of care from different generational positions are of primary interest. While family cases can allow for more interconnected analysis, reporting family cases is also ethically problematic (Gabb 2010), not least since it may skew the sample (only families with low levels of conflict are likely to participate) and jeopardize the right to informed consent for all parties. 5 The interviews were semi-structured and departed from themes related to care at significant points in time (e.g., childhood, becoming a parent/grandparent). In this article, we focus on narratives relating care practices during corona, which were more present in some respondents' narratives than in others, but emerged as a theme in all interviews taking place after the corona outbreak.
technique, Sheridan et al. 2011; concentric circles of closeness, Smart et al. 2001;Mason & Tipper 2008). A diary interview method (Zimmerman & Wieder 1977;Bolger et al. 2003) chosen by 20 adult participants proved particularly helpful in capturing the nuanced ways in which the pandemic situation changed everyday practices. Twenty grandparents and six adult children were interviewed twice or more, according to their chosen level of participation in the study. In the interviews with grandchildren (12 participants, one sibling interview) we used a 'draw-your-day' method and concentric circles of closeness (Eldén 2013). Some interviews were carried out face-to-face (either pre-corona, or outdoors during the calm phase of the pandemic in the summer/autumn of 2020), while others were conducted using digital devices. As a consequence of the pandemic, interviews are spread out in time to encompass the preferences, needs and safety of participants. 6  (2017) and is approved by the Swedish Ethics Review Authority (ID 2019-03890). We pay attention to the particular ethical dilemmas that emerge in family studies in general, and in studies engaging respondents from the same family in particular (Gabb 2010;Eldén & Anving 2019). Recruitment from the same family has only been pursued with the approval of the already interviewed party; the level of involvement (e.g., methods, tools, repeated interviews) have always been the choice of the respondent to ensure full informed consent; to protect the intra-family anonymity we make use of multiple techniques for anonymization (Eldén 2020); and we refrained from systematically reporting identifying demographic data on respondents, including internal connections between respondents in the dataset. 7 One participant in a couple interview was born in 1960 and was aged below 65 at the time of the data collection; this participant is not included in the analysis for this article.
All interviews were recorded and transcribed in full, and all data were anonymized for this article. In the process of analysing the data, we collectively and carefully delineated the overarching tendencies in each participant's life narrative, specifically focusing on narratives of care practices and relationships to kin in everyday life before and during the pandemic. Taking an hermeneutic interpretative understanding of analysis, we used different tools: repeated readings of field notes, transcripts and visual material and coding, then reducing, the data into the overarching themes that emerged. The visual material of life-lines, drawings and diaries were analysed 'in context', as part of the entire research encounter of transcribed interview data and researcher field notes (Smart 2009;Eldén 2013).

Result: Intergenerational care practices during corona
When discussing care between generations, the participants of our study confirm the general image conveyed by the quantitative studies quoted above: intergenerational involvements are generally highly valued in Swedish families, and there are indications that they are becoming increasingly important. This does not mean that they are conflictfree; on the contrary, the narratives are full of complex and often difficult negotiations. A recurring narrative in the older generation is to contrast the relationship one has with one's own children and grandchildren to those of previous generations, stressing the greater closeness today, both practical and emotional. This is also visible in the narratives of different practices aimed at increasing involvement in everyday life: the decisions of different generations to move closer to one another, weekly schedules for grandparents to pick up grandchildren from day-care, or grandchildren talking about grandparents helping with schoolwork.
According to participants from all generations, the pandemic has affected all of these practices and relationships, mainly through forced generational separation. In doing so, it also seems to have highlighted different meanings of intergenerational care practices and relationships to the participants. In the following we will show how the pandemic situation serves as an inevitable context for the narratives in all interviews after March 2020. However, how the pandemic is experienced is relational. Doings of care are filtered through the pandemic situation, but people do not offer one single reason for handling a specific situation in a particular way, but rather tell stories of interrelated 'sets of considerations, constraints, opportunities coincidences and serendipity' (Mason 2004: 166), where relationships with others are at the centre. In this study we consider three themes that emerge from the data: strategies for doing care during corona; narratives that focus specifically on taking risks in order to care for one another when relationships seem to outweigh corona; and, finally, the worry but also occasional relief that emerges in the wake of forced physical distancing.

Strategies for doing care in pandemic times
As corona has halted taken-for-granted ways of interacting and doing care in everyday life, people have developed strategies for adjusting to, and trying to compensate for, the new situation, with the goal of staying in touch and continuing as much involvement in each other's lives as possible. In all the interviews we carried out, narratives of 'staying-closewhile-keeping-a-distance' strategies were discussed, and almost all participants talked about turning to digital resources in order to stay in contact.
Carina, for example, a grandmother who lives close to her grandchildren and is usually very involved in their everyday lives, tries to play with her six-year-old grandchild using video calls. This has worked out surprisingly well, Carina tells us; she and the grandchild show each other things that they have crafted, talk about and play with them on camera, and check in regularly so that they follow each other as the play develops. Other participants tell about digital devices enabling new -and even increased -ways of participating in each other's everyday life. The grandparents Henry and Harriet, for example, have a daughter who has lived with her family abroad for many years. All visits are now impossible, and while, under normal circumstances, they are used to not seeing each other for long periods of time, the pandemic situation seems to have heightened the importance of staying in touch and presented new possibilities for doing so. They have started to participate in each other's dinner times through video calls. Harriet says: It has been very fun actually, with all these video calls that we've had with [daughter], we've called each other almost every day. Often, when she is cooking food or something, she's just there. So, it becomes like a live TV cooking show. It has felt really good.
The frequent video calls are a new feature in their relationship, even though they have lived apart for several years. The grandchildren are there too; they talk and laugh together, and both Henry and Harriet express great satisfaction with this. As Henry says, 'Actually, we are more present and know more about them now than we usually do.' However, many participants, and especially grandparents, express frustration with having to interact through digital devices. Both grandparents and grandchildren talk about having different views on how best to stay in touch. For example, grandparents prefer to talk on the phone, while grandchildren find this awkward and wish to communicate through text messages instead. Several grandparents tell about the difficulty of staying in contact with younger grandchildren who quickly lose patience and want to leave the screen or phone and do something else. Digital devices also seem to more often require focused interaction, such as asking and answering questions, or relating one's activities in a declarative manner. This works badly for the more tactile and implicit sentient activities of care that many grandparents and children experience as key if a care situation is to be experienced as 'good' (Mason 1996;Davies 2011;Eldén & Anving 2019). The grandmother Agneta, for example, stresses the importance of having time to just 'be there' together, without pressures of expectations of having fun or feeling happy, something that she sees as especially important for her relationship with her teenage grandson.
I just sit down next to [grandson] and then I feel that he, well he knows that I'm his grandmother [laughs] [...] you have to try to feel the bond, I reckon.
This 'being there' dimension of doing care forms the basis for a care relationship for both parties, and is, inevitably, impossible in the pandemic situation. The importance and difficulty of keeping a distance, even when possibilities emerge to actually see each other face-to-face, and often meeting outdoors, is repeatedly talked about in the interviews, especially by grandchildren. The children are well aware of the importance of distance, as expressed by the grandchild Freja: Grandma comes to visit when she can. Now, during corona, we would rather wish for her to stay at home [...] When someone is celebrating a birthday, she comes to visit. But then we make sure to sit apart from each other.
Special occasions, such as birthday and Christmas celebrations, are singled out by both adults and children as times when the grandparents' presence is important, but these have obviously changed: either the celebrations are suspended during the pandemic, or special strategies are needed to keep them going. The children are as aware as the older participants of the importance of maintaining a distance and the risks involved in not doing so.
Grandparents, adult children and grandchildren alike narrate their experience 'before' and 'after' the corona outbreak, but in contrast to adult interviewees who talk about this period in time as exceptional and as something that will soon be over, younger grandchildren more often describe the new ways of interacting and doing care as a normal part of everyday life. Sometimes even extreme measures, such as when ten-year-old Diana describes visiting her grandmother at an eldercare home, outdoors and with a plexiglass sheet between them, are related in a matter-of-fact way. For the older generation, time seems more limited, with less time left to 'put on hold', as the grandmother Nancy puts it, pointing to a stronger need to define the current conditions as 'exceptional' and soon to be over. The grandmother Vera speaks longingly about when she can meet her grandchildren again, and in the meantime, they frequently exchange text messages where they dream about the 'marvellous fika' [traditional Swedish coffee and cake] they are going to have when they meet again. The adult child Jenny imagines a time after corona, when she will value more the simple ways of just spending time with her parents. I think that somehow you will appreciate your freedom more and maybe meet with them [referring to the older generation] more frequently, because we didn't just drop by each other's houses before, except in relation to the children, but maybe in the future, it doesn't have to be so organized, with nice dinners and such, but rather that you can drop by for a cup of coffee in a messy kitchen, that you will appreciate this in a different way.
Time will tell if these dreams of simpler and more frequent interaction and engagement in each other's lives will actually be realized when pandemic restrictions are lifted. Nevertheless, it is evident that the new strategies for everyday care doings between generations necessitated by the corona pandemic have affected the ways in which the participants in our study look upon, narrate and do their relationships and understandings of care in everyday life, most significantly how they turn narratives of relational co-presence, where care doings are taken for granted, into narratives of relational participation where discussions and negotiations become necessary (Mason 2004).

Taking care -taking risks: When relationships outweigh corona
Sometimes the corona filter -both the actual recommendations, and the more general discourse of distancing -becomes so impervious that everyday life as it used to be is completely interrupted and put on hold. At other times and in other cases, doings of care between generations are allowed to continue in more recognizable ways. As the previous section has shown, following the public health recommendations has significantly affected people's practices, but also, it has become a strong moral narrative on a discursive level: to be a 'good' citizen is to take individual responsibility and do as the authorities ask of you (Kvarnlöf 2021). This means that not following the recommendations necessitates legitimizing accounts (Scott & Lyman 1968). While the majority of the respondents talk about following the recommendations, sometimes this is just not possible. In this section, we will look more closely at instances where different needs for care seem to outweigh corona restrictions, examining both 'extraordinary' circumstances and more mundane reasons for bending the recommendations to make one's everyday life work.
The grandmother Jea, for example, tells how her adult daughter went through a divorce just after the corona outbreak. She and the daughter are very close, and when the younger woman had gone through difficulties before, Jea had been there for her. The divorce was a rough period for her daughter, and Jea worried a lot. In the beginning of the pandemic, Jea followed the recommendations and isolated herself, and only met her daughter for walks outside. But when the divorce came through, she saw no other choice but to end her isolation. As she explains: You need to take it all into consideration, as I did then, it's a matter of priority, it had to be like that, there was no other choice. She was very sad and slept really badly at night, so she would need, she felt this need to talk [...] So that became more important than corona.
These special circumstances led Jea to the conclusion that she had no choice but to be with her adult daughter. She felt forced to break the recommendations as her daughter's needs outweigh the risk of getting infected by the virus. Although one might have an initial desire to follow the recommendations, things can happen that force one to reconsider.
Another example of overcoming the recommendations is related by the grandfather Ola, in describing the plan for his daughter's upcoming childbirth. The daughter was worried that her partner could not come to the hospital when the baby was due, as the slightest symptom of a cold would mean that he would not be allowed in. Furthermore, she needed Ola and his wife to care for her older child when she was giving birth.
Previously, the grandparents had not fully isolated themselves, despite being in the risk group, and continued to, for example, collect their son's children from school. To be able to help their daughter, they decided to be more careful: they stopped seeing their son's family and generally started to isolate themselves. When the baby arrives, the plan is for them to form 'a bubble' with the daughter's family, an action that follows some of the public health recommendations, but not all. This decision also means that they are acting differently in relation to their different children and grandchildren. Like Jea, Ola's decision came out of a particular need in a specific, extraordinary situation, and resulted in prioritizing and acting differently with his two adult children and their families.
These acts of reconsideration, as formulated by Jea and Ola, are in a very direct sense relational and done in response to different needs. They attest to how the corona recommendations can be overlooked when other needs are more acute, but also how adherence to the recommendations can differ even within the same family over time, in relation to changing needs and special circumstances. Different needs and circumstances can outweigh the recommendations, but when doing so, they require that the person 'account' for this decision in terms that make it apparent to the listener that there are other and more important moral considerations.
However, sometimes less 'exceptional' circumstances can also lead to corona recommendations being outweigh. In several instances, people told us how they have created a 'corona bubble', consisting of a small, selected group of people with whom they spend un-distanced time. This is the case for the adult child Isak who, together with his family, has decided to include his mother and bonus father in their closest circle, to be able to continue to see them as usual. He is well aware that this is not the correct thing to do if one follows the recommendations, especially considering the age of the grandparents, but, as he says, it is a risk they are willing to take. A couple of years ago, Isak and his partner moved to live closer to his mother and bonus father, and since then, they have helped each other out a lot in their daily lives: having the grandparents pick up grandchildren from school is crucial to the functioning of the adult children's work and family situation, not least since Isak and his partner have a long commute, and the grandchildren have an important role in the lives of their grandparents. When the pandemic started, they jointly decided that they would continue seeing each other, despite the recommendations. 'We risk a little', Isak says, as solving the challenges of organizing everyday care, and maintaining the close relationships between grandparents and grandchildren, is worth the risk of infection.
At the same time, Isak also talks about other relationships that are organized very differently. His biological father, who also lives nearby with his wife, is not included in the bubble, partly because the wife belongs to a medically high-risk group, but also because they, from the outset, have not played an important role in the everyday care arrangements of Isak's family. Isak's case shows that the decision of whom to include in your corona bubble must be understood in relation to different care needs prior to and during the pandemic.
The relationality in the discussed cases also results in different forms of 'we-speak'. According to Mason (1996), 'we-speak' is a characteristic of relational narratives emerging from the principles of co-operation and participation which, in contrast to narratives of inclusion and co-presence, involve a more conscious consideration and negotiation of all parties' interests (Mason 2004: 169). Narratives of when corona is outweighed by other needs and moral imperatives necessitate an even stronger reformulation, from the 'takenfor-granted' to a negotiated decision, compared to strategies for everyday care. The stressing of a 'we' enhances collective responsibility in making the decision to breach public health recommendations. However, as this entails both an awareness of contravening recommendations and, often, very obvious differentiations between kin, a relational constraint is also evident in the narratives. Negotiating risk against care becomes a precarious enterprise, not only by increasing the risk of getting the virus, but also by instigating potential conflicts between different family members.

Distance and its consequences: Narratives of worry and relief
During the pandemic, worry for the older generation has been a recurrent feature in the media, and has been at the very heart of the corona strategy, both in discourse and in measures to 'protect the elderly'. This is also apparent in our data but, as in other studies (Kivi et al. 2020;, and probably related to the predominance of grandparent respondents, worries present themselves in more complex and reciprocal ways: worry for the elderly co-exists with worry for adult children and grandchildren.
The situation of grandparent Helena and her daughter Cecilia is illustrative of this. Shortly before the pandemic began, Cecilia went through a divorce, and Helena and her husband were heavily involved in helping her, both practically, with her new home, and emotionally, supporting her and her children who are very close to their grandparents. But corona has made their engagements in each other's lives marked by worry. In her interview, Cecilia retells a conversation with her teenage child, expressing the worry 'for the elderly'. I was recently ill, but then I felt a bit better and I said, 'now we can go and see grandma and grandpa', and my son just, 'How do you think it would feel to have a mother that killed your grandma and grandpa?' Kind of like, he felt like we could not risk anything.
Echoing the strong statements from the prime minister quoted at the start of this article, Cecilia's child expressed the need for his grandparents to be protected from them, and the worry of what could happen if the recommendations were breached. But the grandparents, Helena and her husband, are also worried, not so much for their own sake, but for Cecilia and the children. In the first interview -which took place before the pandemic -they talked at length about their worry, especially about their grandson, who had problems in school, and how they tried to help out in every way they could, for example, by having him to come stay with them. With corona, all of this was interrupted. Helena writes in her diary, 'It's all about the coronavirus these days. We stay at home as much as possible [...]. It's hard not to be able to help Cecilia as much as we want.' Helena expresses worry for both the daughter and the grandchildren, writing about the difficulty of seeing how tired and sad they are and not being able to be there as much as she wishes to. The worry and frustration of not being able to be physically present in their everyday life -to support them, or to just 'be there' and see for themselves how they are doing -is recurrently expressed in our interviews with grandparents, especially, but not only, when they know that their children's families are going through hardship.
In some instances, worry caused by distance can be coupled with relief. For the grandmother Marianne, whose daughter's family is also going through tough times, not being able to see them evokes mixed feelings (Luescher & Pillemer 1998: 416-17). Marianne has been deeply involved in her daughter's family for many years, ever since a grandson became chronically ill. After retiring, Marianne and her husband moved to the city where the daughter's family lives, to take part in their everyday life and help out with their often quite complicated circumstances. Marianne expresses complex feelings in general -while really wanting to help out, the situation has also put a lot of stress on her and her husband. Now, the forced distancing due to corona makes Marianne worry a lot, but she also expresses sentiments of temporary relief: not seeing her grandson means that she does not have to face her more general worry for the child's well-being all the time.
Relief can, as Marianne's narrative indicates, be felt in relation to being temporarily unburdened of emotional and practical care duties. These are, following Mason's argument (1996), often intertwined. In that sense, corona seems to offer some possibilities for women to step back from traditionally gendered care activities (DeVault 1991;Roman & Peterson 2011). Unna, for example, who is a bonus grandmother to her sister's grandchildren, lives near her niece's family, and before corona, often helped out with their everyday care. Now during corona, they still try to arrange to meet, but the character of the care doings has shifted. In the interview, Unna describes a particular event during the summer of 2020 when, in order to follow the recommendations to avoid indoor interactions, the family had gathered for a garden party and this time, she asked others to organize everything. I just kind of sat there and felt like an old grandma or grandpa. They had to take care of everything and it went very well. It was so nice. It's not that I otherwise have a hard time delegating or letting others do the dishes if they want to, and they do, sometimes. But I often feel this responsibility. 'Oh gosh, where are they going to sleep? Is that person angry or sad?' I just let go. It was great, really great.
Because of corona, Unna could, for a change, let go of all the caring responsibilities that she is usually in charge of. This feels peculiar, but very positive to her. Corona has presented a legitimate reason to step back from her usual role, prioritize herself, and reformulate her relational self into one approximating individualism (Mason 2004). At the same time, her possibilities of prioritizing herself, and her comfort in doing so, are still dependent on her relatives stepping in and taking over. In this sense, her narrative continues to be immersed in emotional and practical dependencies to significant others, while also offering a possibility for Unna to see and reflect upon the otherwise taken-forgranted and invisible care doings she used to engage in.

Discussion: Corona as a filter
In our analysis of the pandemic experiences of grandparents, adult children and grandchildren, it is clear that, when life is filtered through corona, it changes people's ways of doing care.
Imagining the corona pandemic through this metaphor of a filter helps us to see its inevitable presence, while simultaneously recognizing that it affects us differently. Everything has to pass through the filter, which is sometimes impervious and hardly lets anything through, and sometimes it is more permeable and things are almost -although never completely -'as normal'. The strategies for doing care while trying to uphold distance and reduce risk, especially for 'the elderly', entailed innovative but also demanding ways of using digital devices, and complex negotiations of who and how to continue meeting face-to-face. In some instances, the filter had to be permeable, as the risks of not doing care outweigh the risk of getting corona, resulting in conscious breaches of the recommendations, to enable what was experienced as necessary relational engagements.
As an undercurrent of all these doings and engagements, we identified reciprocal worries on the part of everyone engaged: grandparents, adult children and grandchildren, but also streams of relief, especially for women when isolation temporarily took away taken-forgranted expectations of particular caring activities. While our frame for the central themes relating to corona partly reproduces the dichotomy of doings-emotions, we have also shown how they are, in practice, connected. Doings are inseparable from emotions in participants' experiences of care (Mason 1996) -a grandmother's worry for a child, for example, is not an emotional state of being, but is deeply intertwined with the practical doings she can or cannot engage in, in specific situations.
From the narratives it is evident that care is at the heart of kin relationships, and that the corona filter seems to make taken-for-granted care doings, and the very presence of relational layers, more visible to the parties involved. When people relate their experiences of living under corona, a need arises to reformulate practices of care as taken-for-granted co-presence narratives (Mason 2004: 168). Narratives of constrain emerge, as restrictions put new demands on engagements, sometimes necessitating prioritizing one kin over another, sometimes just halting all engagements, leaving people with difficult feelings of 'missing out'. Hints of narratives of a stronger more agentic self, of individualism, are also discernible, although presenting themselves as intertwined with the most common way of narrating in our study: through narratives of relational participation (Mason 2004). This results in a general, heightened awareness of the presence, importance and difficulties of intergenerational care doings. In these, the contours of care come through more clearly, as it necessitates negotiation, discussion and decision, in relation to both the surrounding context (the corona recommendations) but also to specific others. As Marjorie DeVault argues, care, and especially the emotional doings of care, are so takenfor-granted that they are difficult to label or pin down even for those doing it, and often are only revealed when absent (DeVault 1991: 228;Mason 1996). The result is sometimes dissatisfaction: the general feeling of missing out, of communication over digital devices not being 'the same', of things being cancelled, frustration at not having time to 'put on hold' permeates all our data. Yet, corona conditions still seem to draw people's attention to the nuances and complexities of care and, thus, ultimately have the potential to lead to a new and enhanced appreciation of the sentient activities that care entails (Mason 1996).
When care doings between kin need to be more explicitly negotiated due to the corona filter, this further highlights the fundamentally reciprocal nature of care. Care, as a sentient activity, is not done by one party to another, nor is it always coupled with particular structural categories or positions (Mason 1996). Some children in our study, contrary to common understandings of being 'passive recipients' of care, see themselves as deeply involved in doing care for others, not least for their grandparents (see also Davis 2011; Eldén 2016). Grandparents are simultaneously recipients and doers of care in relation to adult children and grandchildren but, notably, this is also subject to change and negotiation: care for one grandchild can demand halted care for another. Following the arguments of Finch and Mason (1993), while kinship in general and care doings in particular are intertwined with and needs to be understood in relation to cultural norms and social categories, it is not determined by these.

Conclusion
Corona is, in a fundamental sense, a relational disease. The virus spreads when we meet and interact, and the consequences of trying to stop it are visible in relationships, most importantly, as a filter on personal communities (Iversen et al. 2020), especially those between generations.
Taking its point of departure in perspectives on family practices, personal life and relationality, and with an understanding of intergenerational care as sentient activity, our qualitative interview-based study of grandparents, adult children and grandchildren detects the reciprocal and complex ways in which care between generations takes place. When people relate their experiences, strategies for new ways of doing care are at the centre, and involve creative ways of negotiating distance and risk, all marked by both worry and relief.
Our analysis adds to previous cautions about ageism in the wake of corona, pointing towards the fundamental problems with sweeping generalizations about who is in need or who is vulnerable, and even more about discourses and policies depicting 'the elderly' as 'helpless' or unable to contribute to society (Ayalon et al. 2020: 49). The study also points towards the risk of adult-centrism. Intergenerational care happens in all directions. Grandparents play an important caring role for adult children and grandchildren ( (Hank & Buber 2009;Cantillon et al. 2021), but even though our data are, at this point in time, dominated by grandparents' narratives, they indicate that adult children and grandchildren also 'do care' for grandparents (Eldén 2016). We also show in this article that 'defamilialization' does not translate into complete autonomy or independence. On the contrary, people continue to live 'linked lives' (May & Nordqvist 2019) and to narrate their biographies in relational terms, even in societies with strong welfare states. The contours of relating will likely change when, for example, dependency between generations is less a financial or practical concern, with access to affordable and good public child-and eldercare. However, the character of this change cannot simply be