Exploring the significance of relationality, care and governmentality in families, for understanding women's classed alcohol drinking practices.

In this paper we explore the importance of relationality and care for understanding women's alcohol use, using a theoretical framework comprising concepts from feminist ethics of care, the sociology of personal life, and feminist approaches to governmentality. A key focus is how care giving responsibilities and expectations in families appear to be particularly significant for creating or constraining possibilities for drinking practices. We draw on findings from a qualitative study about alcohol use and stress with 26 women, aged 24-67 years, in the North East of England, UK. We consider how care practices in families feature in the accounts of alcohol use by women with and without children, and how the symbolic and material aspects of social class interact with care to alter the drinking practices women engage in. The interpretation extends scholarship on women's drinking, by adopting a relational approach to identity and linking private care practices and alcohol use to social and political structures. Public health approaches for preventing or reducing heavy drinking practices are predominantly situated within biomedical or psychological paradigms. Intervention approaches to reduce women's drinking that draw on our theoretical framework could offer potential for reducing harmful alcohol use in a more meaningful way.

Introduction

Patterns of heavy alcohol consumption which have the potential to cause health harms are more common in current generations of women in the UK, and in many countries worldwide, than they were four decades ago (Slade et al. 2016). The trend is concerning, partly because women commonly experience health harms associated with drinking at a lower volume of alcohol than men (Mulia and Bensley 2020). In the UK, women from lower socio-economic or class groups experience the most alcohol-related morbidity and mortality (Smith and Foster 2014; Office for National Statistics 2019). Theoretical interpretations of women’s alcohol drinking practices grounded in their lived experience are needed to further understand how gender, social class and other social factors shape this phenomena (Staddon 2015). Such work can help build the case for, and aid the development of, interventions to prevent or reduce heavy alcohol consumption which engage closely with the social contexts of women’s lives (Fang et al. 2014; Staddon 2015).

Accordingly, in this paper, we aim to advance scholarship that highlights the importance of relationality, social class and care in families, for understanding women’s alcohol drinking practices. We have developed a theoretical framework, which incorporates ideas from feminist ethics of care, the sociology of personal life and feminist extensions of Foucault’s work on governmentality. This is applied to an empirical study of 26 women’s accounts of their alcohol use and stress. The paper begins by briefly situating the study alongside other empirical research from the last 20 years that has considered care, relationality and governance in women’s drinking. We then introduce the framework and explain the method and methodology before discussing the findings.

Care, relationality and governance in existing research on women’s drinking

The association of care giving, with women’s social roles and identities, is widely acknowledged as being important for explaining the historically lower rates of alcohol consumption in most groups of women compared to men, and why women who drink heavily are judged more harshly in many societies (Staddon 2015). As women’s alcohol consumption has increased, at the same time as more engage in paid work and choose not to have children (Office for National Statistics 2017), some in the media and researchers have attributed the increase to women’s perceived growing ‘independence’ from their familial care giving roles (Smith and Foxcroft 2009). This suggests gender and care may now be less important for women’s identities and understanding their drinking practices than in the past. However, we seek to counter this way of framing women’s contemporary drinking, adding to conceptual and empirical research that stresses the ongoing importance of the multifaceted interaction of gender and care in contemporary adult women’s lives and their alcohol use (For a review see Muhlack et al. 2018).

Studies that have compared the drinking of men and women have reported that parenthood limits women’s drinking practices more than men’s (Bullers 2012; Emslie et al. 2015). Parenthood also affects women’s drinking differently at different stages of the life course (Emslie et al. 2015; Waitt and Clement 2016; Dare et al. 2020). In their pivotal study of Scottish men and women’s alcohol use in midlife, Emslie et al. (2015) illustrate the importance of gender and care for understanding women’s alcohol use. They observe that childcare limited their female participants with dependent children the most, though women with older children still needed to be available to help their children and did not want their children to see them drunk. Similarly, in a study of Australian and Danish women aged between 50 and 70 years, Dare et al. (2020) show how some participants were keen to emphasise that they would not drink alcohol if they had responsibility for the care of their grandchildren.

Studies that have considered social class differences in alcohol use between mothers of young children point to how the material aspects of social class (i.e. personal resources such as money and availability of support) and subjective aspects of social class (i.e. the meaning and value given to different practices) can also alter the extent to which women engage in certain drinking practices (Brown and Gregg 2012; Baker 2017). Drawing on the accounts of young working-class women’s drinking in the North East of England and Australia, Brown and Gregg (2012) argue that drinking with friends outside the home offered space for relief and respite for young working-class mothers. Yet, opportunities to do this were limited by their caring responsibilities, and when they did go out they were judged more for their drinking than young middle-class women.

Studies also indicate that women can position their own heavy drinking as respectable by comparing themselves to other women who are perceived not to be meeting their care responsibilities (Killingsworth 2006; Emslie et al. 2015; Dare et al. 2020). Here not having care responsibilities can legitimise people’s own heavy drinking by setting up others’ drinking, or sometimes their younger selves drinking, as disrespectable. In an ethnographic study of playgroups in Melbourne, Australia, Killingsworth (2006) found that mothers used talking about previous drinking practices as a way to display their pre-motherhood ‘self’, but also as a way of situating themselves as ‘good mothers’ and ‘in control’. Women who mentioned that they ‘liked a drink’ were careful to reveal that they had not drunk during pregnancy and were not drinking heavily at the present, as these practices might be seen as incompatible with ‘good motherhood’. Likewise, Dare et al. (2020) show how their participants normalise their heavier drinking at the present stage in their life course by stating they had drank less when they had the primary care giving responsibilities for their dependent children.

Many empirical qualitative studies also identify that drinking alcohol can be a way women (and men) create and maintain social connections (e.g. Nicholls 2019; Thurnell-Read 2021). Not all these studies use the language of care, but arguably they illustrate that care is being given and received in certain drinking practices. Drinking alcohol can facilitate emotional support, a form of care, outside the family home when it may be missing in more intimate relationships (Jackson 2018; Barnes and Ward 2015).

These studies, and others, situate women’s drinking in gendered, social and cultural contexts and highlight the importance of care giving for how women with children, in particular, govern their drinking. They also indicate some dimensions of how social class may alter the extent to which women’s care giving roles may constrain their drinking practices. Yet, there is still scope for consideration of how other dimensions of relationality and care practices in families, limit all women, not just women with children, throughout the life course, and how these care dynamics can affect women unequally through their class positions. The theoretical framework we apply here provides a lens to develop sociological understanding of these often-hidden aspects of women’s private drinking. It helps to convey how women’s relationality and their roles as people who are positioned to give care, while also expecting and needing care in family relationships, are important for all women, and are shaped by social, economic and political structures.

Feminist approaches to care and governmentality

Ethics of care ideas are an important area of feminist theorising, exploring varied ways we can think about the importance of care to both women’s lives and social relations. There are different approaches within this body of scholarship, this paper draws on sociological approaches that emphasise the socio-economic and political contexts within which care practices exist (Barnes et al. 2015).

Feminist ethics of care scholars such as Tronto (1993; 2013) and Sevenhuijsen (1998) and Williams (2001) have connected interactional care practices to wider structures of inequality and their reproduction. Tronto (1993) has argued that all people are interdependent, needing care throughout their lives. Therefore, care should be recognised as a central concern and as a matter of social justice for men and women. Ethics of care work emphasise the relational nature of identity: highlighting how people gain and maintain their moral identities within and through their interactions with others (Sevenhuijsen 1998; Barnes et al. 2015). Autonomy is also framed as relational, stressing the importance of the social and relational contexts people are embedded in for making their lives possible (Barnes et al. 2015). For Tronto (1993; 2013) a key concern in Western societies is the political privatisation of care, that is the way political systems and government policies in areas such as social care continue to frame private households as primarily responsible for care giving. In practice this has meant that adult women often take most responsibility for care in families, while not being provided with the care that would support their relational autonomy. Tronto (1993) and Williams (2001) illustrate how the public value given by political systems in Western Societies to paid work above unpaid care, unequally effects women’s everyday lives.

Empirical studies from within the sociology of personal life signal the complexities of how care is negotiated in women’s daily lives and personal networks. Finch and Mason’s (1993) seminal UK-based work, although now over thirty years old is still relevant. They investigated how adults in families negotiate their care responsibilities to each other. A key finding was that people’s decisions about assistance in families were made based on both material and subjective considerations. In making decisions about assistance, people were observed to balance their own, and their family members’ dependence and independence. Yet, Finch and Mason point out that in these care practices ‘People’s identities are being constructed, confirmed and reconstructed – identities as a reliable son, a generous mother, a caring sister or whatever it might be’ (1993, p. 163). This work on care emphasises that in any relationship where care is not offered, but is expected, people’s sense of self and identity can be damaged (Lawler 2000; Smart 2007). More recent work on relational care in families illustrates how the way it shapes identity is influenced by class. In a study of intergenerational familial care from the perspective of Irish women who both give and receive care, Conlon et al. (2014) show that care is important for all women, but direct care giving appears to constrain working-class women’s lives the most, while middle-class women seem to have more freedom to negotiate ‘caring about’ instead of doing the direct care giving.

Another way to think about identity dynamics embedded in care comes from Foucault’s work on governmentality, which has been used to understand how power works to govern people through their identity positions in the social practices and intimate aspects of their lives (Foucault 1976, 1977). He argues that governmentality works through normalised ideas or discourses about the ways people should behave and ‘pathologizes’ people who are different. People internalise these ideas, such as notions that unpaid care is less valuable than paid work, which become part of their subjectivities, the way they act and how they think about their actions (O’Grady 2004). Self-surveillance works through self-policing where deviation from normalised practices leads to private recrimination.

Some feminist scholars have adapted Foucault’s work on governmentality by bringing gender to the fore of these dynamics of self-regulation. They emphasise that self-surveillance and self-policing are more debilitating for women than men because of the emphasis women place on putting care for others first (Ussher 2004). Various scholars also argue that governmentality does not affect women equally, and there are differences between women in terms of their levels of self-surveillance and self-policing (Skeggs 1997; Lawler 2000; O’Grady 2004). Skeggs’ (1997) work in particular has highlighted the importance of social class for understanding different women’s self-surveillance in relation to care practices. She argues working-class women can engage in harsh self-surveillance in relation to caring and familial relationships, especially care giving in motherhood, as a way to claim value and maintain a moral identity, partially because they can be seen to lack value in wider society. This work points to how middle-class practices that are valued by wider society can alter the practices working-class women believe they can, or want to engage in (Skeggs 1997; Lawler 2000).

Together, the concepts we are drawing on—interdependence, relational autonomy, self-surveillance, and self-policing—help us to explore the usefulness of care for understanding women’s alcohol use. They enable us to consider how different material and subjective aspects of care giving and care receiving are important contexts shaping the drinking practices of women with and without children. In particular, how women connect self-regulating expectations of care in family relationships to their alcohol use. Ultimately, our focus is how the social, economic and political shaping of relationality and care come to impact women’s private drinking practices and are linked to the creation and maintenance of different women’s gendered moral identities.

Method and methodology

The paper derives from a study which aimed to understand women’s alcohol use in relation to everyday stress, from the perspectives of women with a wide range of drinking practices, but not currently seeking treatment or with recognised alcohol dependence. We adopted a social constructionist approach to stress, taking the position that the experiences women feel are stress emerge from socially created discourses about how they should be (Wiklund et al. 2010). We used qualitative, one-to-one face-to-face semi-structured interviews to explore the aspects of the phenomena of drinking and stress that may be of particular concern to women themselves (DeVault and Gross 2007).

Sampling concentrated on ensuring diversity of participants across the sensitising concepts of paid work, motherhood and social class. The lead author carried out the data collection between May 2014 and June 2015. Participants were recruited in-person in community groups, or virtually through the social media site Facebook. Interviews were held face-to-face in participants’ homes, workplaces or at community venues. To be eligible to participate the women had to have consumed alcohol in the past year. The interview topic guide focussed on exploring women’s experiences of stress in their everyday lives, their experiences of alcohol use through their lives, and times when they related their drinking to stress, if at all. Each interview lasted around one hour. The interviews were audio-recorded with a digital recording device and transcribed verbatim. Women’s names and other possibly identifying details have been changed to preserve their anonymity.

The initial stage of analysis used inductive techniques of analysis (Rapley 2011). We compared and contrasted the accounts, and identified a preliminary coding framework, and from here we identified three overarching themes in the data: Care, Self and Control. Data collection was carried out alongside this stage of analysis. We stopped at 26 interviews as it was considered there was enough diversity in the sample and sufficiently rich data to meet the aims of the study (Sandelowski 1995). In the next stage of analysis using techniques of abduction (Timmermans and Tavory 2012), we came to the theories and concepts introduced above to develop the interpretation here.

The study was given ethical approval by our University Ethics Committee. The researcher was a middle-class woman in her late thirties, with young children, who also drank alcohol. She has reflected on her positionality, and some of the relational ethical issues raised during the fieldwork elsewhere (Jackson 2021).

The final sample of 26 women was varied in terms of their social class, age, parenting and paid work status (see Table 1). Their drinking practices were explored qualitatively, and they self-identified as drinking at a range of levels from irregular consumption to regular heavy consumption. We have used an area deprivation score as an indication of class position.1 When we refer to women as working class, we are referring to women who live in the most deprived areas, and when we refer to middle-class women, we are referring to women who live in the least deprived areas. The women were less diverse in terms of ethnicity, as all were white. However, we felt the women were similar in demographics to other women in the North East of England where the research was carried out. Our interpretation focuses on the following dynamics: 1.) Negotiating care giving responsibilities in women’s accounts of their drinking practices 2.) Care giving expectations and boundaries in women’s accounts of their drinking practices.

Table 1

Sample Characteristics

Sample characteristic	Characteristic Value	Number
Social class based on index of multiple deprivation score	Working class (most deprived 1–5)	13
	Middle class (least deprived 6–10)	13
Age group	20–29 years	6
	30–39 years	9
	40–49 years	8
	50–59 years	2
	60 + years	1
Employment status	In full-time paid employment	10
	In part-time paid employment	6
	Not in paid employment	8
	Full-time student	2
Parenting status	Woman without children	10
	Mother of dependent children only	11
	Mother of adult children only	3
	Mother of dependent and adult children	2

Negotiating care giving responsibilities in women’s accounts of their drinking practices

When the women spoke about their alcohol use their interdependencies and relational autonomy were tangible. Drinking alcohol with others out of the home, or at home, was often presented as an opportunity for relational support (Jackson 2018). However, these practices were navigated around their familial care responsibilities. This was most evident in the women with dependent children but was also present in the accounts of the women with adult children and the women without children.

The women who were mothers of dependent children drank at different frequencies and quantities to each other, but almost without exception their accounts suggested they engaged in self-surveillance and self-policing, to ensure that their drinking practices did not affect their performance of their roles as responsible mothers who put their children first.

To illustrate, Penny, a middle-class, married women, who worked full-time, like many of the other women with dependent children emphasised that she had reduced her drinking when she had children to be able to perform responsibilities expected of her. She mentioned she had not wanted her drinking to affect her ability to ‘make (the children’s school packed) lunches the next day’. However, she also said that when her children were younger, she would drink at home in the evening when she came home from her paid work, and while her husband was at work:

The kids were young, the pressure of getting them to bed, getting their homework done, getting them turned around and bathed and in bed at a reasonable time... I probably did have a glass of wine when I was cooking the tea, because it might well have been that I was on my own more.

Penny’s account suggested her gendered circumstances contributed to her drinking alone, but if she could meet her responsibilities, she felt it was legitimate to drink alcohol when she was at home with her children to help make herself feel better.

In contrast, some working-class women in the study appeared to engage in harsher-self-surveillance and self-policing of their drinking than the middle-class women, particularly in relation to their care giving roles (Schmidt 2014). This was most visible when they spoke about drinking at home alone. Indeed, some of the working-class mothers said drinking at home with no one else there was not something they felt they wanted to do. To illustrate, Laura a single mother with four dependent children said she did not want to drink at home alone (without another adult) in case her children became ill. Notably, Laura compared her options to drink at home alone as a full-time parent, to her sister who drank heavily when she came home from doing two paid jobs:

When she comes in, obviously she has a bath, cracks open a can, and that’s her de-stressing. But I mean, maybe if I was to work two jobs, I might be the same, but I can’t, I can’t get in, have a drink and say, “Well, I’m stressed, I’ve been at work all day”, because I haven’t. And you can’t say, “Well, I’ve had the kids all day”, because they’re yours anyway.

Here, subjective notions of the value of paid work compared to the lack of social value attributed to being a working-class single parent appeared to affect what Laura felt she could do for herself when she is alone. Although Laura does not say whether her sister had children, her account suggests she felt it was more respectable for her sister to drink because of paid work, while her own care responsibilities were not a reason to say that she drank because of stress, or to feel that she could drink on her own. Laura also explained how the cost of childcare prevented her from going out to drink with her sister. Overall, her account suggested her gender and class, meant she did not drink at home and her ability to access particular self-care practices to manage stress may have been more limited than Penny and other middle-class participants.

There were three women—Pauline, Sharon and Diane—who had solely adult children in the sample. Compared to the women with dependent children, all these women appeared to have more time for drinking with others outside the home. Nonetheless, their care responsibilities for their adult children, and sometimes other family members, still appeared to influence their drinking practices. Both as something that was appropriate, but also the opportunities they had to consume alcohol as part of their own self-care. All three of the women’s accounts indicated how they were navigating their paid work and care giving roles.

Sharon, a middle-class woman who worked full-time said a recent period of caring for her adult daughter and grandson, when her daughter was in hospital, had restricted her from going out to meals and for drinks with her partner. This period of her life, which she described as stressful, was explained in the following way:

There was obviously no time to do... the normal release mechanisms... I needed to take my daughter back into the hospital at seven in the morning or something. And then she didn’t want me to leave her, so I was tied there!

Sharon said that during this phase of intense care giving she had needed to navigate care alongside her paid work responsibilities, where she was also concerned about letting her colleagues down.

Sharon’s account suggested that while she prioritised caring for her daughter and grandson, she could sometimes find this hard. Yet, not taking responsibility for her daughter and grandson’s care here would have had tangible material consequences. Moreover, choosing to drink rather than give care in this situation had the potential to damage Sharon’s identity as a mother and grandmother who cared for her family at times of crisis. Governmentality is suggested here because as a woman in a family, she was allocated the private responsibility for care. Sharon did not have the option to go out to drink as she would normally, if she wanted to maintain a moral identity in this important family relationship.

Pauline the only working-class women with solely adult children in the sample did not speak about paid work and care limiting her drinking. She did not drink at home ‘I don’t like sit and drink wine, one after the other’ and rarely went out to drink. However, in contrast to Sharon her account illustrated the increased challenges and constraints of balancing care giving and paid work with limited financial resources. She mentioned a stressful period in her life when she had not been able to take time off from paid work to care for her sister, when her sister was terminally ill. Thus, comparing Pauline to Sharon we see what while Sharon was constrained by her gender, her financial circumstances and flexibility in her paid work role, meant she was able to take time off work to care for her daughter.

Among the research participants, care giving responsibilities appeared to constrain the women without children the least but care was still conspicuous in their accounts. It was notable that within their discussion of heavy drinking two women without children, Nina and Hannah described that meeting their responsibilities to extended family could limit their drinking at times. Nina, a professional middle-class woman, said a regular practice of drinking heavily with friends at the weekend was an important way she gained support and relaxed at the end of her week at work. However, she mentioned she could still meet her responsibilities to her family if she needed to:

My family live away, my sister and my parents and things, so if I was going down [to visit them] on the Saturday, I would still go out on the Friday, but I would probably leave at like 9 pm.... I know that if I need to be sensible, I’m sensible, but if I’ve got no need to be, then I’m not!

Likewise, Hannah, also a middle-class woman, said that although she drank regularly and heavily most evenings after work, on the days she needed to visit her sister and nieces her drinking practices were shorter:

I suppose other nights you’ve got to go and see nieces, you’ve got to go and see your parents,... things like that.

Their accounts suggest that though the women without children were not limited as much by their care responsibilities as the women who were mothers, showing they cared about family members was still important for how they came to present themselves as a responsible woman and a ‘sensible’ moderate drinker. Class differences in how the women with adult children spoke about care were less evident than they were with the women who were mothers.

Overall, the material and moral dimensions of care giving or ‘caring about’ in families were present in all the women’s accounts and appeared to regulate their alcohol use, or modified their drinking practices, in different ways. The findings challenge notions that some women can act independently of their familial roles and suggest care remains important because of continued interdependences within the relationships women are embedded in. Care is also important for how women present and maintain a moral and gendered identity.

Care giving expectations and boundaries in women’s accounts of their drinking practices

A key tenet of ethics of care approaches is that as well as being positioned as care givers women themselves need care throughout their lives (Barnes 2012). The women’s accounts illuminate this as their own needs for care and their expectations of care receiving in families was a prominent theme in their discussions of stress and alcohol use. This also further illustrates their relational autonomy and is another dimension of how the self-regulating expectations of care in families shape drinking practices.

In their accounts of stress several women spoke unprompted about feeling let down when family had not offered the care they expected. These unmet expectations often featured in the examples women gave of drinking very heavily in relation to periods of significant stress. Indeed, this perceived lack of care was often privately presented as more important than the original concern or stressor. For example, Helen a middle-class woman without children, explained a heavy drinking occasion when she had been made redundant as related to her father not offering the support she expected rather than the redundancy itself:

My Dad had actually visited on that day, and he hadn’t been very sympathetic to me about my situation, and kind of just told me to get my act together, so I was feeling really sorry for myself and em obviously like really stressed out em, so yeah... I drank em a lot of wine.

Similarly, Tara, a working-class woman with dependent children, discussed a one-off heavy drinking occasion when she had been planning to move to a new house because her children were being bullied by neighbours. She explained how disappointed she had been when her mother changed her mind about helping her move because she was offered the opportunity to go on holiday:

I felt so upset that she was going to let us stay there longer... because she would rather go to Menorca, than see everything her daughter, and her kids, her grandkids were going through... I went home and I got really, really, really drunk, on my own, listening to music, getting really drunk.

The example points to Tara’s governmentality in her family, and her relationality, as she was constrained both by her care giving and her care receiving roles. It is possible that with few financial resources she relied particularly heavily on her family for support, while a middle-class woman with more financial resources may have had more options. It may also be that she was hurt and felt let down by her mother because this relationship was important to her own identity as a daughter who was cared for in a family relationship.

A noteworthy feature of this case is that governmentality seems absent in Tara’s mother’s decision to go on holiday instead of helping her family (though we do not have the account her mother would have given). Suggesting families do not always enact governmentality in the way that is expected, but women themselves continue to be led by normative expectations of the care in these relationships (Smart 2007). The account also illustrates that while some of the working-class women explained they did not drink much at home, a few women’s accounts suggested that their material circumstances and a lack of relational support had contributed to heavy drinking at home at times.

Although there were many examples of care in nonfamilial relations across the women’s accounts, our interpretation points to the boundaries around expectations of care receiving in both non-family and family relationships (Weeks et al. 2001). Feeling that they could not ask for care in certain relationships could sometimes limit women from getting the relational support they needed. Indeed, a few women’s accounts suggested this relational context had led to them feeling that alcohol was the only thing they had to make them feel better. Dawn described a period of heavy drinking alone soon after she discovered that her husband had been having an affair. She mentioned that her parents lived abroad and could not easily offer support. She was pregnant and had a young child, these responsibilities meant she could not leave the house:

I had one friend who knew (about separation from partner), erm so I relied on her quite a lot. Which probably she had her own issues going on,... So, I was like, "I can't go and throw more stuff at her I've already thrown so much at her." So, it was the bottle of wine that came sort of came to my rescue, of sorts.

Here, Dawn conveys that she could not turn to one close friend, who she was able to talk to about her situation, because she felt she had already overstepped the expectations of this relationship. She suggests that feeling that she could not gain relational support, because of the boundaries around care receiving, contributed to her feeling that drinking alcohol alone was the only thing she could do to make herself feel temporarily better. Dawn later indicated the relational support she received when her father immediately travelled from abroad to help her when he heard about her situation. Although the governmentality in both her care giving and care receiving roles is present here, to some degree, Dawn’s account suggested that her material circumstances offered a level of protection because her father had been able to travel a long distance to be with her. She said, she felt this support had contributed to her reducing her drinking during this period.

In comparison, Abbey, a working-class woman with adult and dependent children also described a period of heavy drinking at home in the evenings, while caring for her dependent children, after her partner had suddenly left her. A dominant topic in Abbey’s interview was that she had little contact with her own parents, indeed she said she felt she had ‘nobody to turn to’. Her discussion of a conversation she had with her daughter, who was living away from home with her boyfriend during this period, suggested her expectations of who she can legitimately receive care from, contributed to her managing alone:

“Mam, you’ve lost a hell of a lot of weight.” She says, “I’m worried about you.” I gans (said) “I’m fine.” I says..., “ I’m okay, as long as you and Paul’s fine, that’s the main thing... I’m going to bed now”, but I didn’t, I stopped up (stayed up) even more, drinking, and drinking till I finished the… I had to drink the full bottle.

Lawler (2000) notes that while care giving is part of the mothering role, asking for care is not. In her family Abbey’s daughter was one possible means of support, Abbey conveyed that she did not want to accept help or even talk to her adult daughter about her own problems, perhaps because it was not compatible with the identity she wanted to maintain in this relationship. It could also be related to other factors such as the social stigma of heavy drinking (Staddon 2015). This not feeling that asking for help was consistent with presenting her gendered moral identity contributed to Abbey trying hard to self-manage alone. At the time of the interview Abbey said she had reduced her alcohol use and was trying to manage it alone by herself.

As with other women in the sample, self-regulation, self-policing and the political privatisation of care are suggested in Abbey and Dawn’s accounts; the expectation that women should continue to engage in self-regulation in their familial roles, regardless of their material circumstances and relational support. They also point to women’s own needs for care and how difficult it can be to get this support. Like other women in the study, Dawn and Abbey had people they could have approached for support, but they did not always ask for it because it was not what was expected in these relationships. Indeed, here we see how socially constructed relational circumstances, which unequally constrain what women can legitimately do, particularly when alone and unsupported, might contribute to some women feeling that alcohol is the only thing they have in these contexts to make them feel better.

Discussion and conclusions

By focussing on care giving and receiving in families in women’s accounts of alcohol use and stress, and applying a framework including feminist ethics of care and feminist approaches to governmentality, we have illustrated the importance of these practices for the development and maintenance of women’s gendered identities. Where data were available we have also indicated how material and subjective dimensions of class can differently and unjustly effect various aspects of the phenomena.

The concept of interdependence from feminist ethics of care theory (Tronto 1993) helps to exemplify that women are embedded in different relational networks not just as care givers but also as people who need and expect to receive care. The ‘everydayness’ of care interactions in families was ever present in the participants’ accounts. This resonates with other empirical studies which indicate people negotiate their care responsibilities in families, with other areas of their lives such as paid work, to create a moral gendered identity in the location they are in (Finch and Mason 1993; Ribbens McCarthy et al. 2000; Conlon et al. 2014). Feminist ethics of care scholars emphasise how care giving roles are always important but shift and change at different times in the life course (Barnes 2012), while a strong body of contemporary sociologically informed work has illustrated that gender and care remain important for understanding women’s drinking practices (e.g. Brown and Gregg 2012; Emslie et al. 2015). A key novel contribution of our work and framework is we show how care giving in families is important for how all women, including women with and without children, govern their drinking. Moreover, our analysis indicates that even when women do drink heavily, relationality and care are part of their experiences.

The concept of relational autonomy, which recognises that people need care to support their wellbeing, and that autonomy is limited by the resources people have access to (Barnes et al. 2015), is valuable for exploring various aspects of women’s accounts of alcohol use and stress. They need relational support both in their everyday lives and at times of crisis, but their capacity to gain relational support is related to different gendered and classed dynamics of care. The political privatisation of care (Tronto 1993) and the self-regulating expectations of care in families (Ribbens McCarthy et al. 2000; Smart 2007) mean women often manage care giving alone, and are governed in where they can seek relational support. While these gendered relational contexts govern all women, the symbolic and material contexts of class appear to make it more difficult for working-class women to access care to support their autonomy, because of the material resources they have access to and because of the emphasis they put on care to claim value (Skeggs 1997).

The study was conducted prior to the COVID-19 pandemic. Survey data collected during the 2020–2021 lockdowns in the UK and other countries suggest that heavy drinking occasions that are harmful to health may have increased in women, particularly in women in midlife, during this period (Miller et al. 2021). Moreover, the expectations of women to provide unpaid care while lacking care in return, and the additional burdens experienced by working-class women, may have been exacerbated during the pandemic (Gulland 2020). Some scholars have argued that post-pandemic is an opportune time to consider how we respond to home drinking and inequalities in alcohol use (Callinan and MacLean 2020). This study and the theoretical framework could contribute to that endeavour.

The strengths of the study should be considered alongside the limitations. Social class is a complicated concept with different dimensions, and we recognise the area measure of class we used is unrefined. Nonetheless, we did observe differences between the women using the measure that was adopted. Another limitation is that the participants were white, and mostly heterosexual. Future research should focus on exploring the themes we highlight here in black and minority ethnic populations and in lesbian, gay, bisexual, trans* or queer people.

A key implication of our interpretation for policy and practice is that women’s relationality and care should be the starting point for interventions to help to reduce health inequalities in alcohol use. Using this theoretical framework highlights that interventions to prevent/reduce heavy alcohol consumption should enable women to gain care and support, and address their material circumstances (Fang et al. 2014). Because of the focus on the importance of care in all women’s lives and their own need for care, drawing on ethics of care theory would also ensure that interventions do not unequally focus on women with young children as care givers, and ignore the needs of women themselves (Bell et al. 2009). Theoretically grounded interventions, which target the social factors which lead to unequal contexts of care, ‘stress’ and alcohol use, hold promise for developing approaches to reduce heavy drinking that engage with what really matters to women.