‘People try and police your behaviour’: The impact of surveillance on mothers and grandmothers’ perceptions and experiences of infant feeding

1

S e e h t t p://o r c a .cf. a c. u k/ p olici e s. h t ml fo r u s a g e p olici e s.Co py ri g h t a n d m o r al ri g h t s fo r p u blic a tio n s m a d e a v ail a bl e in ORCA a r e r e t ai n e d by t h e c o py ri g h t h ol d e r s .

Introduction
Routine surveillance by health professionals has been a longstanding feature of healthcare (Foucault, 1963), and increased visibility, both actual and the potential for, results in self-regulation (Foucault, 1977).Despite a lack of evidence supporting the effectiveness of individual programmes, many governments have invested heavily in health promotion activities (Speller et al, 1997).Within the UK, these interventionssuch as Health Action Zones (1998Zones ( -2003) ) and Healthy Towns (2008-present) and in Wales, Communities First (2001-present) -have been targeted at areas of economic deprivation.However, there has been insufficient robust evaluation to determine if these policies result in a reduction in health inequalities (Judge and Bauld, 2006).
Pregnancy is a key time when normal populations become subjected to increased medical surveillance, and the UK government has prioritised reducing health inequalities during this period (DH, 2010).Pregnancy-specific interventions include increased surveillance by midwives and health visitors targeted at women from deprived areas or young mothers, through Flying Start in Wales, Starting Well in Scotland and the Family Nurse Partnership in England.Evidence for beneficial health impacts is mixed (Judge, 2005;Robling et al, 2016), and health promotion interventions may actually disempower some participants (Grant et al, 2014).It is not only professionals who hold the authority to judge, as in contemporary society power is everywhere and held by everyone (Foucault, 1972). For Foucault (1977: 271), 'power produces knowledge and power and knowledge directly impact one another', and thus health promotion has both altered society's accepted 'normal' behaviours and, through societal power and surveillance, stigmatised those who do not conform to the public health ideal (Hacking, 1986;Wigginton and Lee, 2014).Accordingly, individual self-regulation of health behaviour is fundamental in surveillance and care of the self (Foucault, 1977(Foucault, , 1997)), which occurs in response to perceived scrutiny by others.This is particularly apparent in neoliberal Britain, as policy seeks to devolve responsibility for health and wellness (and other fundamental areas of social policy) from the government to the individual, and part of being a good citizen is accepting this responsibility through displaying appropriate behaviours (Dean, 2015).Within this article we use the term 'surveillance' in a broad sense, ranging from the potential to be observed by others, for example, in public spaces, through to more active interventions from others, including comments or questions to the individual, which can be interpreted as value judgements.
Within contemporary society, attention has been directed towards pregnant women who come to be seen as a vessel protecting the health of the foetus, where it is viewed as having a higher importance than the woman herself (Lupton, 2012).Appropriate health behaviours include being a healthy weight, not smoking or drinking alcohol and avoiding foods associated with increased risks to the foetus (HSCIC, 2012).Interventions may also attempt to deliver class-based values systems, including a focus on the competence of mothers (Sorhaindo et al, 2016).Within this context, surveillance is prima facie directed towards unborn babies and children (Armstrong, 1983).However, it is ultimately used to make moral judgements regarding a woman's suitability as the carrier of a foetus and/or the carer of babies and infants (Hacking, 1986).Intergenerational research with mother-grandmother dyads suggests that the intensity of the gaze directed towards mothers has increased in recent decades (Fox et al, 2009).
Mothers living in poverty are the most likely to undertake health behaviours that are currently recognised as undesirable during pregnancy (HSCIC, 2012).These behaviours often engender emotive responses -research examining the public's views about smoking during pregnancy has highlighted negative and judgemental discourses, which Wigginton and Lee refer to as 'good mothers don't smoke' (Wigginton and Lee, 2014: 267).Moreover, the rejection of working-class value systems in notions of acceptable pregnancies mean that working-class women are habitually represented as lacking 'moral value' (Skeggs, 2005: 965).These representations often fail to acknowledge the lived reality of life on a low income (Hanley, 2007), and, as Wenham (2015) argues, pregnancy, particularly for young parents residing in marginalised areas, can be characterised by vulnerability, uncertainty and a 'fragile' self-identity.This may be compounded by forms of external regulation from family and friends, who reinforce discourses of inadequate mothering.
Heightened surveillance continues in the postnatal period, and current concerns focus on infant feeding, with breast milk recommended as the exclusive food for the first six months.Within this context, how good a mother is has come to be measured by whether she breastfeeds, and the use of infant formula has been reconciled as 'somehow symptomatic of a woman's failure as a mother' (Lee, 2007(Lee, : 1088)).In the UK, less than 1% of mothers feed their infants in line with official guidance (HSCIC, 2012), often leading mothers to experience a feeling of failure (Brown et al, 2011).Alongside this, breastfeeding for longer than a few months (Stearns, 2011), or in public, tends to be perceived negatively, including claims that women are 'making a scene' (Boyer, 2011) and undertaking a disgusting, even sexual act (Acker, 2009;Grant, 2016).Reactions of disgust have long been associated with working-class motherhood, and are typically shown in the terminology used in media discourses (Skeggs, 1997;Tyler, 2008), creating and stigmatising an undesirable 'other' (Katz, 1986).
Parenting increasingly occurs in the public arena in contemporary society (Boyer and Spinney, 2016), and breastfeeding women manage their behaviour with reference to views of what is acceptable feeding practice, while utilising shared public space in which they will necessarily be visible (Battersby, 2007;Brown, 2016).Mothers who nurse in public have been labelled 'nasty', 'offensive', 'rude' and 'distasteful', and have been subject to the negative criticism that they are 'exposing themselves', leading to breastfeeding women adopting a range of strategies to minimise their visibility (Guttman and Zimmerman, 2000;Noble-Carr and Bell, 2012;Grant, 2015Grant, , 2016)).More than 40% of breastfeeding women have received negative reactions from members of the public (HSCIC, 2012), and this negative view of public breastfeeding has been linked with the increasing hypersexualisation of the breast in Western societies (Acker, 2009).Combined with class-based othering practices, breastfeeding in public may be particularly challenging for working-class women, which may lead to the rejection of breastfeeding.Consequently, infant feeding idealism and infant feeding realism become juxtaposed because infant feeding in the real world sits within multiple values that compete with discourses of the optimum health ideal (Hoddinott et al, 2012).The increased political prioritisation of breastfeeding alongside the objectification and commodification of women and their breasts has rendered infant feeding a site of moral and interactional 'trouble' (Lomax, 2013).

Methodology
Our research, funded by the Children and Young Person's Research Network (CYPRN), engaged with intergenerational accounts of infant feeding in areas of poverty within Wales, UK.Areas of poverty were principally chosen because of low rates of breastfeeding (HSCIC, 2012) and high rates of public health intervention.We aimed to understand influences on infant feeding decisions, and experiences of infant feeding, alongside how women navigated the dominant moral frameworks surrounding infant feeding and how they engaged in discursive work to justify their status as 'good mothers'.To explore continuities and changes over time, we undertook dyad interviews, with mother-grandmother pairs.A participatory approach was adopted, with participants asked to bring artefacts; these shaped the direction of the interviews.

Dyad interviewing
Intergenerational work is 'central to the project of new motherhood', which conceptualises the development of maternal identities (Thomson et al, 2011: 119) and provides opportunities to explore 'partial identifications' and moments of connection and disconnection (Mannay, 2013).Furthermore, the increasing focus on pregnancy and infant feeding as a time to avoid risk makes comparisons between generations an area of empirical interest (Fox et al, 2009).Therefore, the purposive sampling targeted new mothers to be interviewed with their own mothers, the new grandmothers.
Intergenerational dyad interviews have been adopted in earlier research on motherhood.Through the co-construction of events and the naturalness of communication between the dyad pair, they can provide increased depth and valuable insights into continuity and change, which might not be found by interviewing the participants alone (Clendon, 2006).However, dyad interviews also allow pre-existing power relations between mothers and grandmothers to be brought to the interview, which can prove problematic where grandmothers supervise mothers' behaviours (Caldwell et al, 1998;Sciarra and Ponterotto, 1998).Accordingly, it is important to acknowledge that pre-existing power relations between dyad participants were brought to the interview, and that these influenced the data produced.Five of the mother/ grandmother pairs chose to participate in a dyad interview, and one intergenerational pair preferred to be interviewed separately.

Sample
Cardiff University's School of Medicine Ethics Committee provided ethical approval, and participants were invited to contribute through the researchers' contacts and snowballing.Fourteen mothers of infants aged under 30 months who lived in urban Communities First 2 areas of South Wales, UK were invited to participate.Of these, nine agreed and were requested by the researcher to invite their own mothers (the grandmothers) to participate in the study.Seven mother/grandmother pairs agreed to participate in an interview and six pairs, 12 participants, went on to undertake an interview during July and August 2014.The mothers' youngest children were aged between six weeks and 25 months.Mothers themselves were aged between 22 and 43, and grandmothers between 42 and 74.There was intergenerational continuity in employment type among four pairs.All participants were white, with five mothergrandmother pairs identifying as white Welsh.Five of the mothers and four of the grandmothers had some experience of breastfeeding.

Data production
Visual methods can 'allow for the explorations of greater complexities in everyday life and give greater prominence to the ways in which people are active everyday theorists' (Mannay and Morgan, 2015: 482), and have been applied successfully in a wide range of studies (Rose, 2012;Lomax, 2015;Sorensen and Poland, 2015).More specifically, visual data production has been introduced and generated in studies with mothers to engender more complex data than interviews alone, to facilitate more equal power relations within interviews and to prompt participants to tell the stories that they felt were most relevant (Clendon, 2006;Rose, 2012;Mannay, 2015).However, given the nature of the sample, it was important to avoid time-consuming pre-interview tasks, so participants were invited to select everyday artefacts, prompting narratives of associated experiences (Hurdley, 2006).Specifically, mothers and grandmothers were asked to select everyday objects that they associated with infant feeding and their experiences of motherhood more generally.
Four mothers and three grandmothers brought a range of artefacts including bottles, breast pumps, infant clothing, photographs and books.Participants were asked to talk about the artefacts that they had brought along, with the interviewer prompting for further details if it did not emerge through conversation between the two participants.A brief topic guide was created and used in interviews, focusing on views and experiences of infant feeding, including interaction with partners, family and health professionals.Where topics did not arise naturally, they were introduced as part of open-ended questions towards the end of the interview.
Where artefacts were not presented, participants were asked to recount their experience of infant feeding in a free-form participant-led discussion, supported with the same open-ended questions.However, the interviews that were based around artefacts were longer and tended to be directed more easily by the participants, with the interviewer taking the position of actively listening to the accounts of participants, rather than raising a series of questions to be answered.All of the interviews were conducted by the same member of the research team, Ruby, who did not have any children, or direct experience of infant feeding.

Analysis
Audio recordings of the interviews were transcribed verbatim, with interviewer notes added to make it clear which artefact was being described and to record other points of interest.Data production and analysis were conducted concurrently, with emergent themes being explored in future interviews.In all, 24 artefacts were brought to the interviews and there were six hours of recorded discussions that generated 83,236 transcribed words.The visual productions, which were photographed at the point of data production with participant consent, acted as tools of elicitation, rather than objects of analysis per se; however, they were considered in the analysis to clarify and extend the associated interview transcripts (Mannay, 2016).Inductive thematic analysis was facilitated by NVivo 10, allowing codes, categories and themes to be generated from the empirical data produced with the participants (Braun and Clarke, 2006).The theme of surveillance was further explored across the interviews as a whole to appreciate its situatedness.

Findings
The participants' accounts covered a range of themes including family relationships, changes in identity, health and wellbeing, and childcare practices and routines.However, the findings presented here focus on participants' subjective experiences of surveillance by family, friends and strangers.Feelings of being watched, evaluated and judged, with some direct experiences of being questioned by strangers, were centralised in many of the participants' reflections, and the following sections explore these in relation to pregnancy and infant feeding.Participants articulated two key narratives around 'good motherhood', which sometimes conflicted: first, devoted selfless mothering, largely through following public health guidance; and second, chaste or invisible mothering, through hiding the maternal breast.One participant also highlighted a third discourse of freedom and maternal choice.

Pregnancy and surveillance
Within the interviews, participants were not asked directly about their experiences of being pregnant, but pregnancy often featured in the discussions between mothers and grandmothers, and this, in turn, provided the context for their experiences of being subjects of surveillance when feeding their infants.For example, in the dyad interview with Tanya 3 (mother) and Diane (grandmother), Diane described several instances when the pair had been out in public and strangers had suggested what Tanya should have been doing, or sometimes controlled her behaviour through restricting access to 'harmful' goods, with the explicit rationale of protecting her unborn child.In this way, participants become subject to the power of strangers, who legitimise their control by evoking a discourse of the 'good' maternal subject who follows health guidance.In one recollection, which was represented by a coffee mug, Diane recalled how a waiter acted "like the kinda food police" and refused to serve them the afternoon tea they had been expecting, because of Tanya's "big belly".When Diane began this account, Tanya's first contribution was to roll her eyes ("oh yeah [rolls eyes], yeah, yeah, yeah"), suggesting a level of frustration at the way she was treated in the story her mother is sharing.Tanya goes on to explain: 'Yeah, he said: "You can't have this", "You can't have that."He didn't ask us what we wanted.He said: "The only things you can have are, um, cheese and pickle."He brought things [he decided] we could have, but he didn't ask if we liked 'em, he didn't, he didn't ask if we wanted them … he even went to the chef about what I can have instead of asking me: "What has your health visitor said you can have?"…'Diane explained that the waiter was "adamant" that he would not serve them other types of food, as if he had a legitimate right to dictate what kinds of food both Tanya, who was pregnant, and Diane, who was not pregnant, could eat, recalling "he didn't, like, give us any choice, did he?It was just, like: 'We can't serve you this'….It was like he was the pregnancy expert, wasn't it?"Later in the interview Tanya reflected on this as something that made her feel "quite inferior" and how she felt like saying, "yeah, you haven't got the right to tell me what I can and can't eat!" Diane also commented, "and I weren't impressed."Mother and grandmother did not appear to be reticent in the interview setting; however, they did not actively resist the waiter's restrictions, highlighting the power of societal expectations and actors over maternal subjects.This lack of resistance could suggest a normalisation, and tacit acceptance, of surveillance and intervention in the lives of pregnant women by non-professional strangers.At this time, Tanya reported that she felt as though she, or at least her bump, was "everyone's property", which was "a pain" for her to navigate, and unlike any previous situation she had encountered.
Diane questioned the legitimacy of the increased public gaze and intervention in the antenatal period: "I mean, it's strange isn't it, like, why do they care?I can't see why there's like a fixation: it's like everybody … someone's always got [to intervene]."Tanya also reported that she felt a strong desire to reject any future interventions directed towards her by strangers focused around maternal health risk behaviours: "on my next one I'm gonna be, like, a right cow to everyone about everything and I'll say: 'No!I do want prawns and will have prawns … [laughter]….Let me speak to the manager.'"Thisrejection of external control suggests a future where Tanya is able to revise the role that she will play, refusing the scrutiny directed towards pregnant women, and performing as she would in everyday (non-pregnant) life, where she has freedom of choice.However, as representations of acceptable maternal subjects are constructed by societal surveillance directed towards "big bellies", Tanya is unlikely be afforded such freedom in subsequent pregnancies.This is particularly relevant in contemporary society where outside others take on the role of assuring compliance to their understandings of appropriate behaviour in relation to the public health ideal.Tanya's account was in stark contrast to Diane's own experiences of being pregnant, when strangers undertook less surveillance of maternal health risks on behalf of the unborn child, and she commented: '… it was really weird….I couldn't believe how interfering, like, from when I was pregnant … when I was pregnant no one cared [laughter] you could say: "Oh I'll have a double vodka and coke with my fag" [laughter] and it was like: "Yeah, no problem" [laughter].So it's just totally different … people try and police your behaviour … when I was pregnant when no one cared … you could have asked for any food or drink: I never got stopped having anything.'Diane's comments support previous research findings of a significant intergenerational shift where pregnancy has become (experienced as) a public (as opposed to private) phenomenon (Fox et al, 2009).In this instance, Tanya's "big belly" marks her out as someone to pay additional attention to, in her maternal embodiment.
For Erica (mother), surveillance in pregnancy was specifically linked to infant feeding, and the concept of whether to choose breast milk or infant formula, and she explained: 'People generally go "Oh [infant formula's] gonna be really expensive."Y'know they think that when you're pregnant they can ask you anything, and it would be: "Are you gunna breastfeed or are you gunna bottle feed?"And "Oh well, y'know, breastfeeding now it'll be cheaper.Think of all the money you'll spend on formula."'Diane, Tanya's mother, also reported that she had mentioned to Tanya in the antenatal period that: "[breastfeeding] was a good thing and I hope you try it", but that her friends also asked her about Tanya's infant feeding intentions from early in the pregnancy: 'Yeah my friends are really pro-breastfeeding, so as soon as Tanya said, like, "I'm pregnant" [my friend's] words were, like: "Are you breastfeeding?"' Again, Erica's comment, "they think that when you're pregnant they can ask you anything", is symbolic of the increased community surveillance on the maternal body.Erica reported that she planned to, and tried to, initiate breastfeeding, but was physically unable to because she could not produce milk.The questions Erica received in the antenatal period were value judgements around quality of mothering behaviour (Lee, 2007), which contributed to producing her view of breastfeeding as the ideal method of infant feeding.Erica performed considerable identity work throughout the interview to highlight that some women could not breastfeed, the positives of formula feeding, and that her baby had not been harmed because of not receiving breast milk.Despite lengthy accounts of surveillance from family, friends and strangers during pregnancy and early motherhood, participants rarely discussed the input of health professionals, perhaps showing that this surveillance was expected in the antenatal period, or, in Erica's case, the intervention from health professionals was minimal: "I thought the midwives in the hospital would push [breastfeeding] but they didn't."

Infant feeding and surveillance
Participants reported a range of pressures to feed their babies in particular ways, including a general pressure to breastfeed, as opposed to using infant formula, that came from societal knowledge and repetition of the mantra 'breast is best'.They also discussed an awareness of their visibility when feeding infants in public, which related to breasts as sexual, or around other mothers, that could result in a sense of competition to be the most selfless maternal subject.Participants reported various strategies to respond to these pressures, including politely responding to direct questions about their feeding methods from strangers, disengaging in friendships with those who forcefully promoted infant feeding, and quiet resistance to pressure from family.Alongside this, some participants attempted to perform a "good mother" role when feeding infants in public, where they were aware of having an audience, by simultaneously demonstrating that they were breastfeeding while showing as little as possible of their body, or by using expressed breast milk.
Among the mothers, an explicit rationale for feeding choice was absent for all but one.Tanya reported that she made the decision to breastfeed autonomously, based on convenience, rejecting the notion that discourses around appropriate feeding influenced her: "my decision wasn't based on what anyone said I just.…I didn't care if anyone said breastfeed: if I wanted to bottle feed, I would have bottle fed."However, this articulation of autonomy and freedom of choice in the maternal period was contradicted twice later in the interview.Tanya reflected on the pressures on mothers to breastfeed: "I don't think there should be so much pressure for mums to breastfeed and then feel guilty about it", suggesting that she had felt the force of opinions from others and recognised that mothers are subject to societal pressure in relation to infant feeding.Tanya suggested that partners, and possibly grandmothers, should be the only people with legitimate power in the decision-making process.However, Tanya's mother was present, which may have influenced this comment, particularly as Diane mentioned that she wanted Tanya to breastfeed several times during the interview.In contrast, Diane, who exclusively formula fed her children, reported that she had experienced much less scrutiny regarding infant feeding.This included contrasting the detailed guidance on how to prepare infant formula that Tanya had received with the do 'whatever' approach that was commonplace when her own children were infants.
Similarly, Sharon (grandmother), reported that her daughter, Kayleigh, "has been absolutely marvellous to be perfectly honest" in her infant feeding behaviour, which included considerable "research" about best practice.Although Sharon had breastfed herself, she felt that she had been "quite naïve … I probably hadn't even thought about it", and stated that she felt that "nobody influenced me or anything", in contrast to the plethora of information that Kayleigh had accessed.Arguably, Sharon's decisions would have been influenced by some form of social power.However, her articulation of decision-making highlighted a relative lack of societal concern regarding infant feeding at that time (Hacking, 1990).The other four grandmothers did not explicitly state their infant feeding preferences for their daughters during the interviews, perhaps showing an acknowledgement of the potential for maternal guilt and the high levels of scrutiny already focused on this area.
Grandmothers reported a range of infant feeding experiences, although none recalled a strong pressure to breastfeed or instances in which they had encountered surveillance of their infant feeding.For example, Debbie (Erica's mother) reported a strong desire to breastfeed, but stated that this was based on very limited information, including minimal guidance from health professionals, and reported the normalisation of formula feeding at the time: "I know people of my generation … some of them wouldn't [breastfeed] they'd go on bottle straight away."In contrast, Geraldine (Lauren's mother) reported that she had breastfed one of her children for seven months; however, she expressed disgust toward a friend who had breastfed long term at the time: "my friend breastfed until [her daughter] was five [years old] … that's disgusting … there's no need."This suggested that extended breastfeeding was found to breach discourses of appropriate use of the (sexual) breast in contemporary society (Stearns, 2011).
Mothers reported that comments and behaviours from their family could influence their feelings towards infant feeding in the postnatal period.This ranged from uncomfortable feeding experiences through to comments that the mothers perceived to be judgements of their capability to look after their children.For example, Tanya recounted feeling exposed: "it is still quite embarrassing when I'm sat on the sofa with [my] dad doing these breastfeeds … and then I'm like really covering myself … and even like under the covers it's just embarrassing 'cos dad's quite old fashioned, isn't he?" Alongside this, Kayleigh experienced direct comments regarding the frequency of her breastfeeding in the early weeks: "and you'd have some family members going, ooh, is he feeding again … is he feeding, is he feeding that often?"Kayleigh reported that these comments, which had also come from friends, made her question her ability to feed her baby successfully: "and it just makes you doubt yourself." Comments focused on adequacy or appropriateness of breastfeeding suggest a tension between discourses focused on 'good' mothering through selfless dedication to the public health ideal, and the regulation of the maternal body, through a critique of visible care work.These watchful commentaries serve to undermine breastfeeding behaviour, which may lead to a decision to stop breastfeeding, increasing maternal guilt for lack of adherence to the public health ideal.
Fearne also described how particular friends pushed her to breastfeed her baby, and she disengaged in these relationships.This demonstrates that pressure around infant feeding will not be tolerated by some mothers, where they are able to have a high degree of control in the relationship, which is in contrast to relationships with strangers, where mothers have less control over unplanned interactions and their associated, possibly negative, reactions.A further issue that was explored in Tanya's interview was the tension between being a breastfeeding mother and engaging in occasional nights out where alcohol would be consumed in line with guidance from health professionals: 'I went out and mum had [my baby] … my [relative], we said that we was going out, and she was like: "Oh, she can't go out; she's breastfeeding.She can't drink alcohol" and I was obviously gonna pump it out and then breastfeed the next day … and it's just, it's so intrusive and rude, and you get so angry about it that you think, d'you know what [laughs], it'd be easier just to give her [my baby] a bottle [of infant formula] from all these different inputs here, and left right and left.' In addition to family members, Tanya commented that friends had remarked on her plan to drink alcohol, saying that they would not judge her, but making her feel judged by doing so: 'People say things like "Oh, I won't judge you!", and it's like "Oh, well thank you!" Yeah, it's like, why would you judge me?It's nothing to do with you, so why do you, why are you even saying?I didn't even know I was in a jury, like….It's just so rude and that's, like, really close personal friends who were saying "I won't judge you".'Tanya's references to judgement reflect the continual sense of evaluation from social networks that has become attached to motherhood and mothering.Alongside comments on feeding from family and friends, Tanya and some of the other participants reported that strangers had questioned their feeding behaviour, as illustrated below: 'Me and my partner went to a coffee shop and we was heating up a bottle, um, for [our daughter] and then the cleaner, this man, was, just cleaning around the tables, who worked there, and he came up to me and said, "Are you breastfeeding?" and, like, I said: "yes" I was, and I didn't really take offense to it but if I wasn't, then I'd feel quite like … it's intrusive … like I wouldn't walk up to him and say "What did you have for your lunch today?" [laughter] like, why are you asking me what my child has for milk?But yeah, like, random things like that….' Tanya struggled to understand the rationale for such questions ("it's like, why do you wanna know?"), and reported later in the interview that the cleaner, and other strangers who questioned her infant feeding method, had "no right to tell someone to breastfeed or not."However, she noted that a negative response might occur if the cleaner didn't like her answer ("spitting in my tea"), perhaps explaining her decision to answer his question.In contrast, when Tanya's GP suggested that she may be winding her baby incorrectly, Tanya told him that he was wrong.Diane also recalled that Tanya had an "interfering" midwife for part of her antenatal care, to which Tanya responds: "but that got nipped in the bud quite early."Tanya's reasoning for being able to respond to these demonstrations of power differently is that she is confident that she is a good mum: "you're just getting teared down so much and 'cos I know I'm a good mum.I'll just say, I'll just answer them back."It may be that the professional relationships between Tanya and her midwife meant that standards of behaviour were clearly defined, ensuring that a 'wrong' answer could not have an unknown negative impact, which could occur in interactions with strangers.
Tanya reported that she attempted to feed expressed breast milk from a bottle whenever possible when she was in public, thus undertaking significant additional work in order to juggle competing good mother discourses around providing breast milk while maintaining the invisibility of the maternal body.Her rationale for this was her lack of confidence, a feeling of increased visibility and the potential for (negative) intervention from strangers when she was feeding directly from the breast.Although Tanya had not received any comments explicitly related to feeding from the breast, she felt that her breasts were viewed as sexual by observers, and breastfeeding was viewed as a sexual display akin to pole dancing, which made her feel very uncomfortable.In order to avoid showing her breasts when feeding in front of others, Tanya routinely used a shawl to cover her body, which she brought to the interview.Tanya described the importance of the shawl in her breastfeeding.As part of her account, Tanya reported that even though she tried to cover her body, she would defend her right to breastfeed in public if it was challenged, expressing a freedom of choice discourse alongside a discourse of motherhood centred on the public health ideal, and later mentioned that she would love to feel more confident with breastfeeding in public: In contrast, Kayleigh (mother), initially reported that she had never considered the need to cover her body when breastfeeding in public.However, she implicitly responded to an invisible maternal body discourse, in that she wore clothing that aimed to reduce possible exposure when feeding, showing an awareness that exposing part of her breast during feeding may be perceived negatively by those inadvertently observing the feed.Moreover, through media coverage of women receiving negative attention for breastfeeding in public and her participation in this research, she had begun to understand that she was aware of others viewing her breastfeeding.Kayleigh had not been conscious of receiving the negative looks that Tanya had been concerned about, but was aware of a quiet surveillance that had always made breastfeeding in public different to the relaxed experience she reported at home.Like Tanya, she reflected that when breastfeeding in public she also took a defensive stance to guard against possible negativity, recalling, "I did have, like, a very defiant look on my face … that I don't think anyone really would say anything!"Again, this suggests that Kayleigh is subject to, and responsive to, the power of surveillance in relation to her breastfeeding practices.
The point that participants felt the need to feed their infants in a particular way outside of the home can have implications for feeding choices for later children (Fox et al, 2009).Tanya reported three times during the interview that in the future, if she had another child, she would consider not breastfeeding, including this example of her conversation with her mother Diane: Diane: 'You're glad you breastfed though, aren't you, and gave her all the nutrients and the whatever, the immunities and all that, aren't you?' Tanya: ' 'Yeah, I haven't looked at it like that.' Fearne also identified that feeding her baby in public was challenging.Like Tanya, she routinely expressed her own milk to use in public, engendering the invisible maternal body.However, she felt that when feeding her baby from a bottle she was being judged by strangers as not embodying the public health ideal.These competing dominant moral discourses, emphasising the superiority of breast milk over infant formula, but the necessity of the hidden breast, impacted on Fearne's perception of her own parenting practice: 'If you do bottle people are thinking: "why is she bottle feeding?Why is she?"Even to the point I almost feel that I have to make comments that it's my own milk.' When her baby was undergoing a growth spurt, Fearne reported that she was unable to express enough of her milk to meet his needs, and that she supplemented her own milk with infant formula.In recognition of her visible failing to meet the public health ideal, signified by formula milk, Fearne felt concerned that she would face negative reactions from strangers.As such, she performed the act of mixing formula as though it were a deviant act: 'Yesterday was the first time I felt comfortable to actually get the bottle out in public and mix the bottle with people actually seeing and then to give it to him 'cos everyone's watching 'cos they think, y'know, cute baby, and that's what people do, but I felt kept thinking, oh, y'know, what are people gonna think?What are people…?And at one point, we were in one restaurant once, I was actually conscious I was hiding the powder, like I was actually doing it really secretively, mixing it.'However, Fearne was also aware that breastfeeding could be met with negative reactions, including disgust at the visibility of the maternal body, demonstrating that regardless of the method of feeding, mothers are aware of judgements that may lead to negative comments or actions from strangers: 'If I breastfeed in public [laughs] … this standard joke in my house, if I breastfeed in Starbucks the whole café's just gonna leave … and every middle-class, y'know, over 60-year-old woman is just going to be horrified.' In this extract, Fearne first refers to a general viewing public and her feelings about their anticipated reaction to her breastfeeding in public.She positions this scenario as a "standard joke", which suggests that she is not expecting breastfeeding in Starbucks to trigger a mass evacuation.However, her account resonates with accounts that mothers may be viewed with disgust (Tyler, 2008), particularly when feeding their infants (Boyer, 2011;Grant, 2015Grant, , 2016).Fearne's more specific reference to middleclass women over the age of 60 doesn't reflect the wider literature; however, the cultural location of the participants, in marginalised and stigmatised Welsh locales, may account for the reference to class as part of her experience of an undercurrent of a society divided by class (Hanley, 2007).It is important to appreciate the classed position and spatial marginalisation of the participants, and the ways in which this inflects their ideas about how they are judged and who makes these judgements.Fearne's experience stood in contrast to her mother Barbara's recollections of breastfeeding.Barbara reported that breastfeeding in Australia, where she had previously lived, was subject to less public surveillance.She rationalised this as being because the society was less conservative, so the display of the maternal breast was less provocative, which was in part due to people generally showing more of their bodies.However, she also commented that the acceptability might be down to the way in which women were breastfeeding, with public breastfeeding in Australia being described as almost invisible, thus displaying the good motherhood discourse of lack of visibility that was common in the UK: 'Y'know, the women out there [Australia] have got it down to a fine art, you can walk past a woman and not know she's breastfeeding.'

Discussion
Overall, the interviews with mothers and grandmothers produced a rich seam of data, and we have only been able to discuss some of the points they raised within this article.The use of dyad interviews and artefacts was particularly valuable in generating conversations between participants, and allowing them to reflect on their experiences of motherhood and infant feeding in preparation for the interview.As in previous research, the use of visual approaches (Hurdley, 2006;Rose, 2012;Mannay, 2016) and dyad interviews (Clendon, 2006) generated multifaceted and emotional accounts from participants.These techniques offered advantages, but it could be argued that some experiences may not have been discussed in a dyad setting that would have been raised in individual interviews.For example, the presence of grandmothers may have reduced discussion regarding the challenges of the mother-grandmother relationship in relation to infant feeding (Caldwell et al, 1998).
We found that mothers reported increased surveillance compared to grandmothers.This surveillance, and the negotiation of acceptable motherhood behaviours in relation to the intrusive policing of lifestyle choices and consumption from family, friends and strangers, began in pregnancy and then continued to have an impact on mothers' everyday lives, particularly through infant feeding.The most challenging form of surveillance to manage was that from strangers in the public realm, where women felt less in control of situations and dynamics between individuals.Participants often performed public motherhood in ways that were highly orchestrated and self-aware, as they had either experienced negative interactions with strangers regarding their feeding choice, or felt anxious that these might occur (Lee, 2007).This indicates that societal surveillance acts to shape individual mothers' behaviours and identities as maternal subjects (Foucault, 1977(Foucault, , 1997)).This appeared to be a challenge that grandmothers had not encountered, suggesting that changing societal interest in avoiding risky health behaviours, in relation to pregnancy and maternity, had affected the direction of the anonymous public gaze (Hacking, 1990).Some mothers did report that they were able to reduce or control interactions with those who attempted to regulate their feeding behaviour, but this agentic stance was usually related to friends and health professionals rather than the general public.
Although this study was not comparative and only engaged with participants from marginalised areas, previous research has explored the ways in which working-class women and mothers are subjected to stigma, moral judgements and surveillance (Skeggs, 1997;Tyler, 2008;Mannay, 2015).We were unable to fully explore the classrelated discourses within this article.However, the positioning in existing literature (see, for example, Skeggs, 1997) may go some way towards explaining participants' accounts of attempts to placate strangers who questioned their infant feeding practices, but also the latent anger that this engendered for these marginalised women.
In their reflections, mothers articulated two key discourses of 'good motherhood' that they attempted to display, but that sometimes conflicted: first, the need to conform to the public health ideal through devoted and selfless motherhood, which usually materialised as providing breast milk; and second, the desirability of chaste, invisible or disembodied motherhood, through attempts to hide the breastfeeding body.Accordingly, participants amended their infant feeding behaviours from those used at home, to attempt to perform respectable motherhood, including minimising exposure of the breast while feeding (Guttman and Zimmerman, 2000;Acker, 2009;Noble-Carr and Bell, 2012;Grant, 2016).For some mothers, this resulted in additional work to express breast milk (Stearns, 2011).Participants described some instances of undertaking both breast and formula feeding in public as though it were an illicit act (Boyer, 2011), as they felt each option placed them at risk of judgement, negativity and reprimand.These contradictory 'good motherhoods' render infant feeding a site of moral and interactional 'trouble' (Lomax, 2013).Importantly, within this moral framework, neither breast nor formula feeding could guarantee an acceptable presentation of the maternal subject in public spaces.

Notes
1 Corresponding author.
Pl e a s e n o t e: C h a n g e s m a d e a s a r e s ul t of p u blis hi n g p r o c e s s e s s u c h a s c o py-e di ti n g, fo r m a t ti n g a n d p a g e n u m b e r s m a y n o t b e r efl e c t e d in t his ve r sio n.Fo r t h e d efi nitiv e ve r sio n of t hi s p u blic a tio n, pl e a s e r ef e r t o t h e p u blis h e d s o u r c e.You a r e a d vis e d t o c o n s ul t t h e p u blis h e r's v e r sio n if yo u wi s h t o cit e t hi s p a p er. Thi s v e r sio n is b ei n g m a d e a v ail a bl e in a c c o r d a n c e wit h p u blis h e r p olici e s.