‘I've never ever let anyone hold the kids while they've got ciggies’: moral tales of maternal smoking practices

Authors


Address for correspondence: Clare Holdsworth, Department of Geography, University of Liverpool, Roxby Building, Liverpool, L69 7ZT e-mail: clare.holdsworth@liv.ac.uk

Abstract

Smoking in the home is, potentially, the next frontier in tobacco control in the developed world. As smoking regulations in public space are extended, attention is turning to private spaces and the contribution of parental, particularly maternal, smoking to children's health and socio-economic inequalities in family health. Yet relatively little is known about mothers’ smoking practices within the home and the social meanings of smoking that are constructed by these practices. In this paper we explore how mothers who smoke construct moralities of their smoking behaviour, particularly in relation to where and with whom they smoke. Drawing on in-depth Biographic Narrative Interpretative Method, in interviews with 12 smoking mothers, and their partners, we consider how these moral tales involve comparisons with other smokers and the importance of community endorsement of smoking practices, particularly around children. We also consider the role of children in the home and how children are actively involved in the regulation of smoking behaviours. Finally, we consider the implications of these moral tales for interventions around smoke-free homes.

Introduction

Smoking is recognised as one of the main causes of morbidity and premature mortality in the developed world, and a major factor in inequalities in health. Smoking is not just injurious to smokers, but to those exposed to environmental tobacco smoke (ETS). The harmful effects of children's exposure to ETS are now well established and include an increased risk for sudden infant death syndrome (SIDS), acute respiratory infections, ear problems, and more severe asthma (Cook and Strachan 1999, Hofhuis et al. 2003). Moreover, maternal smoking during pregnancy is associated with low birth weight and increased risk of SIDS and children whose parents smoke are themselves more likely to smoke (WHO 1999). As the WHO (2000) identifies in Tobacco and the Rights of the Child:

The influence of tobacco smoking is felt in three major ways: i) at the beginning of life through maternal smoking ii) through ETS and iii) through role modelling by smoking parents (2000: 21).

Children's exposure to ETS in the home has come under political scrutiny in developed countries in recent years associated with growing evidence of the relationship between ETS and inequalities in health. In the UK, the relationship between poverty and smoking is well established (Graham 1993, 1994, Harmana et al. 2006, Richardson 2001, Wanless 2004). The Acheson report on inequalities in health highlighted the fact that while one-third of children in the UK lived with at least one adult smoker, among low-income families the figure was 57 per cent (Acheson 1998).

As the medical evidence about the impact of smoking on both smokers and those exposed to ETS has grown in the last quarter of a century, the place of and attitudes towards smoking have changed (Bayer and Colgrove 2002). Smoking is becoming a more marginalised activity, both in terms of who smokes, where smokers are permitted to smoke and prevailing attitudes to smoking. In recent years smoking controls in public places have been introduced throughout Europe and North America, and these initiatives have also focused attention on the level of risk of exposure to ETS in the home. On the one hand, smoking bans are associated with an overall reduction in the prevalence of adult smoking (Gallus et al. 2006) which could lead to a reduction in the number of children exposed to ETS in the home. Prior to the introduction of smoking bans, reduction in parental smoking has been identified as the main behavioural change credited with recent declines in children's exposure to ETS (Jarvis et al. 2000). It is anticipated that a ban on smoking in public places will intensify this trend. There is also, however, the potential for further entrenchment of inequalities in health among children, as adults from disadvantaged socio-economic backgrounds, who find it harder to quit smoking (Graham 1993, Robinson and Kirkcaldy 2007a), have fewer alternative places in which to smoke outside the home. This view, which was proposed by members of the UK Government before the introduction of a smoking ban in public places in England in 2007, is, though, opposed by the medical establishment (Royal College of Physicians 2005). Some commentators have suggested that tobacco controls should be introduced for the home, and parental smoking has been referred to in child residency and adoption cases in the US (Hovell et al. 2000) and in recent guidelines for foster carers in the UK (The Fostering Network 2007).

In recognition of the difficulty of quitting among socio-economically disadvantaged smokers, public health interventions have begun to focus attention on the effectiveness of smoking bans in the home to reduce children's exposure to ETS (Robinson and Kirkcaldy 2007b). The prevalence of some form of smoking restriction among families varies considerably by geography, class and ethnicity (see for example Gonzales et al. 2006, Pizacani et al. 2003, and Spencer et al. 2005). Parents believe that smoking controls within the home can be effective. Blackburn et al.'s (2003) study of smoking parents in the West Midlands reported that 86 per cent of parents were aware of the risks of ETS and 90 per cent believed that infants could be protected from ETS in the home, with 65 per cent of parents using some preventive strategy, though only 18 per cent reported not allowing smoking in the home. Medical research to date shows that complete smoking bans in the home are associated with a reduction in children's exposure to ETS as measured by urinary cotinine/creatinine ratios or nicotine in hair, though partial bans are ineffective (Blackburn 2003, Johansson et al. 2004). Moreover, reviews of interventions for smoking controls in the home provide very little support for the effectiveness of these interventions, suggesting that the means for promoting changes in parental smoking behaviour are not easy to establish (Spencer et al. 2005).

Social and moral dimensions

This emergent focus on family dimensions of children's exposure to ETS needs also to consider the wider social contexts of smoking behaviour (Laurier et al. 2000). As Poland et al. (2006: 59) have recently argued tobacco control has been dominated by addiction and lifestyle models, and ‘as a result the social meaning of smoking in the context of people's everyday lives is underplayed’. In the context of maternal smoking this includes interactions with other smokers, particularly friends and family members, and shared views on the risks of exposure to ETS and how young children's exposure to ETS may be restricted. As Bottorff et al.'s (2005) findings for smoking couples illustrate, smoking is implicit in relationship dynamics both as a source of conflict and mutual support, and we argue that maternal smoking also needs to be understood in the context of family relationships and obligations.

Moreover, our focus on family obligations foregrounds the inherent moral character of public health discourses on smoking, and how social meanings of smoking are shaped by moral discourses, as well as social and environmental contexts. In public health discourses, prevalent norms make very stark distinctions between smokers and non-smokers and produce stigmatised identities for the former. As smoking practices become more entrenched among less-advantaged groups, smoking becomes a marker of class background and behaviour through which class positions are not just produced and reproduced, but also stigmatised (Bayer and Colgrove 2002). Yet prevailing moral distinctions do not consider the perspectives of smokers and that, while smoking may be castigated as amoral, the same cannot be said of smokers, who will create their own moralities of smoking practices. Yet these subjective accounts are rarely given voice, as Lupton (1995) argues in taking account of the reasons why people start and continue to smoke ‘the use of these commodities in the construction of subjectivity is rarely acknowledged’ (1995: 150). While the emergent focus on the social meaning of smoking directs attention to smoker's economic and cultural contexts, the connections between smoking and the self is not adequately addressed, especially with regard to prevailing public health moralities about smoking.

One account of this moral dimension is Poland's (2000) discussion of the strategies adopted by smokers in public places in order to navigate the complex morality of acceptable smoking behaviour. Poland claims that the self-regulation of smoking behaviour is:

as much to do with conveying the appropriate social graces of appearing to take steps to manage risks imposed on others in a responsible way, than it perhaps does with the ultimate efficacy of such measures. It is, in effect, about social competence (2000: 12).

Moreover, he argues, drawing on Bourdieu's analysis of social distinction, that the negotiation of social competence is best understood from the perspective of class-based legitimisations of appropriate behaviour. Poland's account is based on the experiences of smokers in public places, and his discussion of the moral discourses inherent in the idea of consideration is contingent on the anonymity of smokers in public places. Yet our focus is on smoking within the home where the condition of anonymity is not met. The idea of consideration does not adequately capture the complexity of smoking mothers’ moral negotiations which involve family, friends, and members of local communities as well as health and social service practitioners. In particular we argue that a class-based approach is not sufficient to understand negotiations within the home. While we recognise the potential for class-based stigmatisations of smoking to structure smoking practices within the home as well as public places, our interest is in how smokers regulate and support each other's smoking in disadvantaged communities where smoking remains the norm (see Robinson and Holdsworth 2007 for further discussion of the dynamics of smoking neighbourhoods).

In order to capture the complexity of mothers’ experiences, our analysis draws on recent discussions of an ‘ethical turn’ in the social sciences (Holdsworth and Morgan 2007, Sayer 2005). These developments claim a reconnection with everyday and habitual moralities, or as Sayer (2005) terms ‘lay normativity’. Moreover these developments may be shaped by comparisons with and judgements of others, rather than dictated by absolute moral codes. The idea of the moral tale (Ribbens McCarthy et al. 2003) is to explore how families create and portray everyday family life and how a moral sense of responsibility (for both the self and other family members) is interwoven into these accounts and practices. Studies of contemporary family life and morality reveal how family members negotiate moralities and ethics of care rather than adhere to normative codes (Finch and Mason 1993, Smart and Neale 1998). Moreover, the negotiation of family obligations shapes the moral identities of family members and their social competences as mothers, fathers, sons and daughters. In the case of smoking, while public health discourses clearly promote the benefits of not smoking, and as such the context is one of moral imperative not to smoke, the potential for negotiation still exists. In particular, smoking behaviours, such as how many, what type of cigarette, where and with whom one smokes, allow for distinctions to be made between smokers. Everyday moralities about smoking are not, therefore, necessarily reduced to a distinction between smoking and not-smoking, but may invoke observed differences in smoking behaviours.

Our analysis focuses primarily on the experiences of mothers, partly due to the structure of the sample which generated relatively little empirical data on fathers, but also because the risks of children's exposure to ETS are primarily associated with maternal smoking. This is not only because of exposure in utero, but is also associated with a gendered ethic of care whereby mothers take most of the daily responsibilities for their children. We explore how the mothers in our study recognise the risks of children's exposure to ETS and seek to regulate their own smoking practices, as well as those of other family members and adult friends, to reduce perceived risks of ETS. We consider how these mothers compare their own behaviour against others, particularly other smokers, in constructing moral tales of smoking behaviour, and how they negotiate the options open to them, which include not just quitting smoking but restricting where they smoke, as well as restricting other adults’ smoking in the home and their children's exposure to ETS in other places. We also explore how mothers seek to prevent their children from smoking themselves. We are interested not just in how individuals adopt smoking controls in the home, but also how these are shared among a community of families who are linked by friendship, geography and through smoking. We argue that a moral duality between smoker and non-smoker is too simplistic and that the complexity of smoking identities and practices does not equate with this binary distinction which may be implied in discourses of risks of smoking and/or exposure to ETS. Treating smoking mothers as a unified group (and their children at the same risk of exposure to ETS) ignores the differences that smokers themselves recognise in where, when and with whom they smoke.

Methodology

The data for this paper were collected as part of a small-scale qualitative study with 12 mothers and their partners (5 partners participated). The mothers were recruited through a local Sure Start centre to take part in a project on ‘understanding smoking in the home’ and the criteria for taking part were that all families had at least one smoking parent and one child under five (though the age range of the children in the families studied ranged from 0 to 19). Eleven of the mothers were smokers, and one was married to a smoker. When designing the methodology for the study we were aware of the potential problems in talking to parents about the sensitive subject of being a smoking parent, and the need to capture parents’ own ‘private’ accounts of smoking (West 1990). We used Biographic Narrative Interpretative Method (BNIM) for the interviews (Chamberlayne et al. 2004, Wengraf 2001). This technique involved capturing an individual's own account of becoming and being a smoker, and enabled them to contextualise their smoking by presenting other aspects of their lived life. After a brief explanation of the format of the interview the researcher asked a question, designed to elicit a narrative about smoking as the parent chose to tell it (known as Subsession 1). For these interviews, respondents were asked to tell the story of their life, starting wherever they liked, and were reminded that the interviewer [JR] was particularly interested in hearing about their experiences of smoking, becoming and being a parent. The advantage of this approach is that it gives the interviewee the opportunity to construct their own account of their life, starting from wherever they like, without further prompting from the interviewer. After the initial narrative, the researcher asked questions about the topics mentioned, in the order in which they were mentioned, using the words and terms expressed by the respondent (Subsession 2). A third subsession was carried out with respondents a few months later to explore if respondents’ smoking behaviour remained the same. The interviews were supplemented by the researcher's field notes and observations.

The respondents were recruited in a disadvantaged community in inner-city Liverpool. The ward in which they all lived returned an adult workless rate of 44 per cent in 2005, with 18 per cent of working age adults claiming Incapacity or Severe Disablement Allowance (Liverpool City Council 2006). Respondents came from a range of educational (including university graduates) and occupational backgrounds, and as such, cannot be classified by a unique class position. What was most striking about respondents’ lives was the fluidity of their residential and occupational histories, with frequent moves both to and from Liverpool as well as within the City, and lack of structured occupational careers. Most respondents were from Liverpool, and all families had a strong connection with Liverpool. All respondents knew at least one other participant in the study, though no one knew everyone else and the sample was not restricted to a unique friendship group.

As we have relatively little empirical data from fathers, the analysis draws primarily on the interviews with mothers, who were also almost exclusively responsible for the care of their children. All the transcripts were included in a thematic analysis and we use examples from the mothers’ own accounts to illustrate the main findings. Our analysis is not intended to be representative of smoking mothers, and given the small numbers it is not possible to infer the prevalence of smoking strategies in the home. Rather, we use the data to illustrate how mothers articulate an ethic of care for their children while being smoking mothers, and how they construct a moral tale around these discussions.

While the biographical method was used to capture mothers’ individual narratives, we recognise that as the interviewer [JR] was non-smoking, these accounts of smoking behaviour are likely to have been shaped by prevailing public health discourses around the need to protect children from ETS. Furthermore, the interviewer was known by participants to be a mother of young children and so may have been excluded from the shared moralities of smoking mothers. However, respondents were prepared to challenge the public health orthodoxy on smoking. Two mothers smoked during the interviews carried out in their homes, illustrating both how adherence to smoking rules could be quite fluid, and, as discussed below, challenged the recognised risks associated with smoking and ETS.

Results

All the respondents described some attempt to reduce children's exposure to ETS in the home. Four respondents reported a smoke-free home (Pauline, Natalie, Fiona and Lorna); one respondent (Kelly) reported a smoke-free home at the second interview following the birth of her baby. In the remaining seven homes, respondents smoked in the home though attempted to confine it to certain rooms and times, e.g. Sally described her home as smoke free, but admitted to smoking inside when no one else was around.

Smoking rules

In line with earlier studies (Blackburn et al. 2003, Spencer et al. 2005), respondents recognised that keeping a smoke-free home has clear benefits. An overriding concern associated with the need to have some sort of control of smoking was a dislike of tobacco smoke and how smoke ‘polluted’ spaces and people, and not just in recognition of the need to reduce children's exposure to smoke. Smoky homes were associated with nicotine-stained walls and furnishings, and they smelt of smoke. The smell of smoke and how it lingers on clothes and furnishings in particular was mentioned not just in terms of how the women disliked the smell, but the stigma associated with it. Sally describes this:

An’ I hate going into people's houses while all the paintwork's yellow because they've done nothin’ but smoke. I was just paranoid about the way other people smell, that I'd smell like that because I smoke you know. Because you do stink an’ when I gave up I noticed people I'm like Jesus have you had a cigarette and they stink. An’ I'm thinking God I must smell like that all the time and that's a big issue for me. A big issue. Sally

What is at issue for Sally is not just being identified as a smoker, but the anticipated distaste that others have of her because of how she smells. Her interaction with non-smokers is not characterised by consideration on her part (Poland 2000), but by how the knowledge that she is a smoker could be used to stigmatise her. Her perceived stigma, however, is related to how she smells rather than wider, potentially class-based, discourses about ‘smokers’.

While families recognise a need to have some rules about where and around whom they smoke, rules about smoking in the home are not fixed, but are continually negotiated. The contingent nature of smoking rules is congruent with wider discussions of family responsibility and moral obligations (Finch and Mason 1993). For smoking in the presence of children, age of the child is a key factor. The importance of not smoking around new-born babies was recognised by all respondents, though for some mothers, as children grow up, the perceived need to restrict exposure to ETS lessens, as older children's bodies are regarded as more resilient to ETS (Robinson and Kirkcaldy 2007c), as Kate describes:

I used to like not smoke where the baby was, and now I am smoking a bit more where the baby is, I think it's because she is that bit older. Because she is like two and a bit now, so I am like she is not a new born anymore, so it doesn't harm her as much, and I know it does but I don't smoke around her, do you know what I mean, I try not to, but occasionally I will have one. Kate

Kate describes how difficult it is to maintain self-imposed smoking restrictions over time. In particular, as children become more mobile, it becomes much harder not to smoke around them, as this necessitates leaving children on their own, albeit for a short time as Kelly recounts:

So we used to come in the kitchen an’ shut that door an’ have a ciggy out here by the window an’ then go back in. An’ every now an’ again we'd hear her cry to come out . . . Erm Mike [partner] said to me that's out of order. We’re punishing her ‘cos we want a cigarette. An’ it made me feel really bad so. And he is right, you know but I need a cigarette. If I need one I've got to have a ciggy I mean all good an’ well if I could pack in or if I could hold on but sometimes you just can't do that you've got to have one. Kelly

Mothers therefore have to negotiate their need for a cigarette with their own children's needs, and while not smoking around children is recognised by the mothers as the ‘right thing to do’, in practice the demands of children and the circumstances in which mothers are caring for them make smoking restrictions difficult to sustain over time. Mothers are negotiating two competing discourses of mothering: not to expose their children to smoke versus the need for constant physical co-presence and the fear of leaving their children alone, if only for a short period of time, to have a cigarette.

Being a smoking mother

In terms of how successful these smoking restrictions were for children's health, most mothers did not report any health problems among their children. As Blackburn and colleagues’ (2003) study of Coventry families describes, parents do believe that they can protect their children from ETS. For mothers of children with health problems, any suggested link with their smoking may be challenged. For example, Natalie's daughter Rose has chronic asthma and Natalie describes how careful she is not to expose her daughter to ETS, so much so that she claims her daughter does not even know she smokes.

Given the care she claims she takes, Natalie describes her anger on reading Rose's case notes:

I went to the doctors for Rose's asthma check-up and you know it's all on screen now on the computer, well the screen was facing me so I could see what she was taking and it said on Rose's notes passive smoking. I said why does it say that and she said well she is and I said but she's not because I don't smoke around her, I never have done since she was born because I didn't agree with that anyway around children and stuff. I said she doesn't even know that I smoke anyway and if I do smoke it's when she's either in school or she's upstairs playing and I’ll go to the back of the garden or whatever, I said I don't smoke around her and she said well it's on your clothes and stuff and . . . do you know what I mean. Natalie

Natalie's experience illustrates the difficulties of health professionals using a binary distinction between smoker and non-smoker, which mothers themselves do not necessarily acknowledge. For Natalie what matters is that she does not smoke around her children. That her smoking is attributed by her doctor as a causal factor for her daughter's health, does not recognise her attempts to minimise her daughter's exposure to ETS.

Yet while the mothers who we talked to were all taking some measure to reduce children's exposure to ETS, this can be compared with maternal smoking during pregnancy. This is more common than smoking around children: three of the 11 smoking mothers reported giving up smoking when they were pregnant (though not necessarily for all of their pregnancies) and one reported giving up when breast feeding. Hence, while mothers did try to reduce children's exposure to ETS, they were less successful in reducing in utero exposure. Some mothers recognised the inconsistency in their smoking practices:

But it's just something that I do because I don't believe, again hypocritically, a new-born baby should be around smoke when I'm pregnant with this new born baby in me body an’ yet I'm polluting it with smoke so I don't really know why. I argue with meself with this one in me head. Liz

Liz's description of arguing with herself about continuing to smoke during pregnancy illustrates how she is constantly having a dialogue with herself about her smoking. While she recognises the hypocrisy of smoking while pregnant (and not smoking around newborn babies), it clearly is not something she is doing out of ignorance. For other respondents the issue of smoking during pregnancy was not just about an internal dialogue, but was directed towards how others judged them if they were observed smoking while pregnant. Sally, whose dislike of smelling like a smoker we discussed above, describes the guilt that she feels:

Erm well the smoking thing I suppose really I was I felt really guilty because I smoked while I was pregnant and that made me feel extremely guilty. It really did. I used to be highly embarrassed of it. You know like I wouldn't smoke out in the street . . . Because when people see a pregnant woman smoking everyone stares at ya and I know why they’re staring because you know it isn't that well I don't think it's right but I couldn't find the will power to give up. If anything I thought I craved more when I was pregnant. Sally

Obviously, mothers need to make more radical changes to their smoking behaviour to reduce in utero exposure to ETS. For some mothers, dismissing the risks of in utero exposure justifies their inability to give up smoking. One common way of doing this is by reference to older generations’ smoking behaviour and the outcomes of this on respondents’ own health. For example, Kelly is able to draw on shared familial experiences to contest medical knowledge about the risks of smoking:

But they are right what they say about small babies. But I don't know me mum had eight children an’ me step mum had two. An’ they smoked through pregnancy an’ me mum actually drank as well. Don't know if me step mum did I can't remember don't know if it was in moderation or what . . . But none of us have got asthma an’ we’re all over five foot nine. So it's stumpin’ your growth an’ it causes asthma don't know. But the small baby thing that's true. Kelly

Shared moralities

What emerges from the analysis described so far are the contexts in which smokers’ moral tales are situated. They are not just constructed with reference to normative rules about smoking (though the general idea about smoking round children being a ‘bad’ thing to do is relevant), but more through individuals’ dialogues with themselves, how they perceive others who smoke both in their family and community, and how they anticipate judgements are made of their behaviour. References to ‘others’ who may be generalised or refer to a specific friendship or family group are key to the narration of the moral tale (Holdsworth and Morgan 2007). Moreover, another way in which community is evoked is through a sense of shared morality, that the need to restrict where and when people smoke was common among friendship groups and, to a lesser extent, family members. Among the 12 families whom we spoke to, rules about smoking around the house were universal. Though the details of these controls and the rigour of family members’ adherence did vary between the families, everyone reported having some control on where and when people smoked in their homes and that similar practices were observed in friends’ houses:

Everyone is the same, I don't think I have ever took them [children] anywhere where there has been people smoking. Tanya

This reported universal acknowledgement of the necessity of restricting smoking around children makes maintaining these restrictions easier for the families in our sample. The mothers’ claims about the universal acceptance of smoking controls contrast with other studies (Ratner et al. 2001), though indicate the importance of community endorsements of smoking practices, particularly within a neighbourhood where the majority of adults smoke (Robinson and Holdsworth 2007). While some respondents did discuss having to ask visitors not to smoke in the home, this is not recognised as a source of tension. There is, however, more evidence of intergenerational disagreements regarding smoking restrictions. In some cases family members would not visit, or parents would not go to particular homes, because of disagreements about smoking in the home. The exceptions to the practice of not smoking in the home tend to be restricted to older generations or younger generations who do not have children. For example Lorna describes her father:

Yes, yes. It's his house, it's his lord of the manor and he smokes where ever, even though I don't like it. Lorna

We can see how the morality of being a smoking mother is negotiated, that smoking is not a clear cut issue, but rather that mothers are aware of the dangers to children's health and take appropriate action to reduce this. Comparisons with other smokers who do not take the same precautionary measures are an essential part of how mothers justify their own smoking practices: smoking is bad, but smoking around children is worse. Hence, Lorna, though herself a smoker, is able to criticise her father's smoking behaviour as he does not attempt to control others’ exposure to ETS. Sally provides a particularly evocative account when she describes her step-mother's family and their smoking practices. Sally talked in considerable detail about her disapproval of her father's new family, often in a humorous tone, and his new partner in particular. Her disapproval is not, though, restricted to smoking behaviours. For example, she describes her disgust after her step-sisters both got pregnant at the same time by the same man, and moreover that this was something that her step-mother had allowed to happen. When it comes to smoking, her disapproval of her step-mother's behaviour is important in how she distances herself from her father's new family:

I don't allow people to walk through with cigarettes. Erm I've never ever let anyone hold the kids while they've got ciggies. I absolutely have that. Can't stand it. And that's another thing that me dad's girlfriend used to do. She'd have the babies on her knee, like her daughter's babies an’ she'd be like puffin’ away and all this smoke there'd just be a cloud round the baby an’ I'm thinking aggh that poor child. Sally

Sally's reference to ‘and that's another thing’ illustrates how her judgement of her step-mother's family is not restricted to smoking but incorporates other dimensions, particularly her attitude to her daughters’ sexual activity. For Sally, smoking around young children and babies is morally reprehensible and she distances herself from the kind of people, such as her step-mother, who would do such a thing.

Children's agency

Most of the literature about children and tobacco positions children as the passive recipients of ETS (either in utero or through childhood); role modelling smoking peers and family members; and, targeted advertising campaigns that encourage smoking. There is little attempt to engage with more recent theoretical developments in the sociology of childhood that have sought to reclaim children's agency and voices (see for example James et al. 1998). The conventional view of tobacco and children's rights is that children do not have a choice about being exposed to tobacco smoke (see Royal College of Physicians 2005 and WHO 2001), and that it is the responsibility of adults alone to reduce children's exposure to ETS. While we did not talk to children in our study, parents’ accounts of their relationship with their children, and their children's views on smoking, depict a far more complex position than treating children as the ‘voiceless’ victims of passive smoking.

Not surprisingly, mothers’ accounts of children's understanding of parental smoking varies by age. Younger children are more likely to be described as copying their parents’ smoking behaviour, and this apparent role modelling is something that can cause distress for mothers who smoke:

Interviewer: Do you think Daisy knows you smoke?

Sally: Yes definitely. She does she understands something about cigarettes she really does because sometimes she’ll pick up like a pencil and she’ll say ‘ciggies ciggies’ and I’ll be like aahh and that just devastates me then. That really does. And like Dave [partner] just flashes me a look that could kill me and I'm like oh no and it is, it really is nasty.

Sally's partner Dave does not smoke, and any risk of their daughters becoming smokers is clearly associated with Sally's behaviour.This is a further source of guilt for Sally as she, in common with all the smoking mothers we spoke to, does not want her children to smoke when they are older.

Older children (from age six/seven upwards) have different attitudes towards parental smoking. Here, the impact of public health campaigns in schools is recognised, as children learn about the health risks of smoking:

And Miss said Mums and Dads who smoke are going to get very sick so he's doing this now, like, beforehand he didn't really understand about smoking but now I think he's educated a bit more. Kate

Older children are described as being very anti-smoking and concerned that their parents might get ill and die prematurely. Their growing awareness of the risks of smoking for parents’ health is reported as a source of tension at home, and, according to the mothers, emerges as the main focus of children's anti-smoking concerns, that their parents will die as a result of smoking, rather than worrying about how parental smoking impacts on their own health. It is noteworthy that children's fears of parental illness and premature mortality are not discussed by the WHO when outlining the relationship between tobacco and the rights of the child. In our study, fear of orphanhood was the main way in which mothers recorded their children's awareness of the risks of smoking. One mother, Tanya, reported giving up smoking because of her son's concerns:

I packed in smoking because my nine-year-old son asked me to because he thought I was going to die of cancer. Tanya

At the time of the second interview, however, Tanya had started smoking again and reported that her son was very angry about this. A more common response than giving up smoking was that mothers felt guilty about the pressure from children and would ‘promise’ to give up:

Me son who is 13, he actually caught me having a cigarette in the garden on Monday and he absolutely went ballistic and he was shouting and screaming. . . . he really got really, really angry and I was upset and he was upset and I had to promise that I would try again, to pack them in. Natalie

Natalie's son Oliver is the only child described as worrying about how his mother's smoking impacts on his health (as well as the worries he has for his mother's health). As a result of his strong anti-smoking stance, Oliver is described as being very resolute in policing the smoking ban in the home:

Well Oliver wants to be a footballer and he is really good at football and at the moment he is dead set against smoking or anybody that smokes and he tells people as well. You know if someone comes in this house, although they know they can't smoke in this house, he’ll always remind them. Natalie

Moreover, as Natalie goes on to discuss, Oliver's views on smoking and his criticism of her behaviour negates her role as a parent and adult as Oliver takes on adult responsibilities in trying to encourage her to give up smoking:

Oh he [Oliver, son] went mental, yeah. He was like I don't believe this and I said, who's the adult here, do you know what I mean, he was really, really angry. I think he was just, he was like I'm so proud of you packing them in and stuff and I was like I am really sorry, do you know what I mean and we had a nice chat and stuff and I just said I won't give up giving up, just at the moment I'm finding it hard. Natalie

Mothers recognise the difficulty of being a smoking parent in that, as in Natalie's case, it can be difficult for their views on smoking and other health-related behaviour to be acknowledged by their children. Moreover, despite the fact that at young ages children were described as being anti-smoking, this does not necessarily rule out the potential for them to start smoking when they are older. The mechanism for inter-generational inheritance is not the same as toddlers mimicking smoking. According to the parents, older children, despite attempts by parents to minimise smoking around them, associate smoking with acquiring adult identities, as it is what their parent(s) do. For example, Tanya and Tony describe how their son Peter is very anti-smoking:

Tony: Yeah he . . . he says doesn't he, he worries you’re gonna die, you’re gonna die of cancer. Peter's [son] always been way ahead, even since he was a baby like . . .

But later in the interview they reflect on how Peter might well become a smoker himself:

Tony: I always worry about the kids copying me smoking . . . especially Peter now.

Tanya: Yeah.

Tony: Because Peter's easily led isn't he.

Discussion

The analysis of mothers’ biographical narratives of smoking illustrates the complexity of being a smoking mother and how mothers negotiate ‘moral tales’ of smoking, and in particular how they distinguish their practices from other smokers who do not take precautions to minimise the risks to non-smokers, especially young babies. The mothers’ accounts describe how within a community of smokers certain ‘rules’ and normative values about smoking are maintained. These accounts are far more complex than depicted by the WHO, for example, which situates parents as a causal agent in children's exposure to ETS and future smoking, and children as passive victims. Mothers describe how children themselves are active agents in the policing of smoking bans, though children's concerns, at least as reported by their mothers, are more towards protecting their parents’ health rather than concerns about passive smoking. Children are, therefore, potent agents in articulating prevalent public discourses about smoking in the home. As such their agency is constructed by dominant moralities of smoking behaviour.

The social meanings of smoking as discussed in this paper are not just about the contexts, in terms of class, lifecourse and wider environments, that pertain to individual smokers, but about their relationships with others, both specific and generalised, and with smokers and non-smokers. Moral tales evoke comparisons not only with other smokers (and illustrate how identities of being a ‘good’ smoker can be constructed), but also how more generalised comparisons are made, particularly by other non-smokers. While the mothers acknowledge how their smoking marginalises them, and, as such, are aware of the wider class connotations about smoking, they do not justify their smoking in the context of their social marginalisation or material deprivation. Rather, they see their smoking in the context of an ethic of care for the self (i.e. their need for a cigarette), and that this conflicts with their ethic of care for their children. This tension is found in other studies of contemporary family life, for example Ribbens McCarthy et al. (2003) argue that where an ethic of care for the self competes with that for dependent children, then the latter prevails, and that ‘people could not weave them together equally in one narrative and still present themselves as a morally responsible adult (parent) in their own terms’ (2003: 139). The mothers in our sample negotiate this seemingly amoral position in a number of ways, by linking care for themselves with that of their children (e.g. smoking calms them down and this beneficial effect is good for children), by comparing their needs and smoking behaviour with smokers who have less constraint, by denying the risks of smoking to themselves and others, or through expressions of guilt. The latter is particularly evident in our data, as mothers are concerned about doing the best for their children, and think that others, who may be family members including children, neighbours or health professionals, will judge them for this. The implications of being a smoking parent are not just about the objective health risks to children, but about how parents become maternal and paternal subjects and subject to normalising discourses of parental responsibilities (Lawler 2000). Mothers combine different strategies in order to negotiate their seemingly incompatible positions as mothers who smoke; for example, expressions of guilt may be assuaged by comparisons with other smokers. This latter strategy is central in all the mothers’ accounts, as they continually compare their behaviour with that of other smokers who are placed on the ‘unacceptable’ side of the boundary of the rights and wrongs of smoking around children. The articulation of the ‘good’ smoker can also be seen as resistance to class-based stigmatisations of smokers.

Resistance to prevalent norms about smoking is also prevalent in some accounts. For example, the relationship between maternal smoking and children's health is contested. Mothers use case studies from their own families as ‘proof’ of how the risks are less prevalent than as depicted in health promotion campaigns. Inter-generational narratives, particularly comparisons of mothers’ experiences with their own mothers’ accounts (and also sisters’ and aunts’) provide a valid (for the mothers), evidence-based account of how inconsequential smoking can be. Health is viewed more as a lottery and not as something that parents have that much control over. Moreover, mothers are aware that smoking is not the only risk that their children are exposed to, and to understand mothers’ attitudes towards smoking, it is important to consider wider social and environmental contexts. The connection with an ethic of care for the self is again relevant, as for some mothers their need for a cigarette cannot be isolated from the day-to-day reality of caring for young children. Hence, mothers themselves do not necessarily see their smoking behaviour and the impact that it has on their children's health as a clear-cut issue, despite the epidemiological evidence on the dangers of passive smoking. They might contest the impact that their smoking has on their children while at the same time attempting to have some measure (with varying success) of controlling children's exposure to ETS.

As smoking within the home becomes the next frontier in tobacco control, the seemingly imperative need, from the perspective of public health campaigners, to regulate smoking in private as well as public spaces is at odds with how mothers negotiate public health messages. In particular, we argue that fixed rules about smoking are inappropriate, and do not reflect the fluidity and dynamics of family practices. Mothers’ own subjective morality of their smoking practices as parents needs to be considered. In particular, for mothers who reconcile the conflict between ethics of care of the self and their children through guilt are unlikely to respond to interventions as they are clearly aware of the potential harm of smoking. Rather, extension of unreflexive public health interventions is more likely to be met with resistance. The findings from this small-scale study also suggest that locating such health interventions within communities is more appropriate than focusing on parents in isolation. We have shown how smokers regulate and support each other. Hence, an intervention that can tap into the dynamics of smoking communities may enable more mothers to adopt smoke-free strategies in their homes.

Acknowledgements

This study was funded by a small grant from the Wellcome VIP award to the University of Liverpool. We thank our respondents for giving so generously of their time in taking part in the research, and the three anonymous referees for their comments on the original draft. All names and personal details of respondents have been changed.

Ancillary