Copyright Sociological Research Online, 1998

 

Alison Bowes and Teresa Meehan Domokos (1998) 'Negotiating Breast-Feeding: Pakistani Women, White Women and their Experiences in Hospital and at Home'
Sociological Research Online, vol. 3, no. 3, <http://www.socresonline.org.uk/3/3/5.html>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 20/2/98      Accepted: 4/9/98      Published: 30/9/98

Abstract

Breast-feeding is a socially constructed and socially controlled practice, and the social factors which influence it have been much debated. This paper argues that a model of breast-feeding as a socially negotiated project can develop fresh insights into how women breast-feed, through focusing on women's own points of view, and on the role of health professionals. Data from a research study of women from diverse ethnic and class backgrounds are discussed. It is argued that breast-feeding women in hospital are generally subordinate to professionals, but remain active in the negotiation of breast-feeding. At home, health visitor support is especially significant. Breast-feeding appears to be a lonely struggle, and the end of breast-feeding is felt to be stimulated by outside influences. Successful breast-feeding projects are most likely for white, middle class women who have effective stocks of knowledge, and can negotiate concerted action with health professionals. Women belonging to socially excluded groups have greater difficulty in the negotiation process, and their breast-feeding projects are less likely to be successful. Whilst influenced by patterns of constraint, breast-feeding projects also show marked individuality. In conclusion, it is argued that the conceptualisation of breast-feeding as a negotiated project promises to improve sociological understanding, but that further research is needed.

Keywords:
Breast-Feeding; Health Services; Maternity Care; Negotiated Order; Pakistani Women; White Women

Background

1.1
Social scientists have recently been debating the issue of how far the social construction and social control of breast-feeding are dominated by medical models, by women's roles, or by cultural predeliction. At the same time, health promoters urge women to make a choice to breast-feed, suggesting that women's own actions can determine the practice of breast-feeding, and assuming they are free to choose. In this exploratory paper, which raises questions as well as attempting explanation, we conceptualise breast-feeding as a negotiated process, constrained by wider social forces (Strauss, 1978). We suggest that women's interaction with health services is a key site at which the interplay of choices and constraints occurs, and in discussing the process of negotiation, we adopt a perspective on breast-feeding which centres on women's own points of view.

1.2
Breast-feeding is often presented as a natural practice, sadly corrupted and curtailed by cultures (eg. Ebrahim 1978), or by economic interests (eg. Palmer, 1988). Breast- feeding promoters urge women to get back in touch with their 'natural' selves, and breast-feed their babies. The Innocenti Declaration of 1990 presents breast-feeding as best practice for mothers and babies:

As a global goal for optimal maternal and child health and nutrition, all women should be enabled to practise exclusive breast-feeding, and all infants should be fed exclusively on breast-milk, from birth to 4-6 months of age (Palmer, 1988: p. 319).

1.3
Much of the research on breast- feeding tends to support this point of view, and to promote breast-feeding as natural and right, often in a somewhat uncritical way (cf. Carter, 1995). But there is much evidence to show that, like other so-called 'natural' practices, breast-feeding and other forms of infant-feeding are socially constructed, varying historically and cross-culturally (Berger and Luckmann, 1971). Thus, in non-industrial societies, breast-feeding was not and is not universally practised (Hastrup, 1992; Fildes, 1986; Lithell, 1981), and the health of infants has not universally been seen as linked with breast- feeding. In industrial societies, other forms of feeding have been, from time to time, promoted as healthier for infants (Fildes, 1986; Carter, 1995). The paper will operate on the basis of the assumption of the social construction of breast-feeding: more could be said on this, but it is not our key concern here (Maher, 1992 gives a useful discussion).

1.4
The sociological literature on breast-feeding shows little consensus, offering varied, even contradictory analyses. There is debate about the many influences on women feeding their infants in Britain and other western industrial societies. Gordon (1989) for example sees an effective role for the medical professions in promoting breast- feeding, and, with it, a particular (rather subordinate) role for women, whereas Jones and Belsey (1977) argue that breast-feeding information given to women in hospital has little effect, thus suggesting a lesser role for medical professions. Bryant (1982) emphasises the role of social network and lay knowledge, whereas for more affluent North American women, Blum and Vanderwalter (1993) argue that La Leche League, an organisation formed in the 1950s, has been significant in promoting and supporting breast-feeding. They emphasise an association between breast-feeding and being a good mother, linking the practice with wider sets of social values. Salt et al (1994), unable to account for the variation in rates of breast-feeding between different areas of Britain, conclude that cultural factors must be decisive, but do not investigate what these might be, or how they might operate. Carter (1995), by contrast, warns against a focus on culture in the case of Asian and other minority ethnic women, arguing that this may lead to assumptions about culturally based choices, which may in fact be influenced by racism. Amid this disagreement, there is a tendency to look for deterministic explanations of breast-feeding, that is, to ascertain whether its particular form or rate is influenced primarily by medical power, lay knowledge and social relationships, health promotion, ideology, culture or ethnicity. Deterministic arguments can also be discerned in health service practice: for example, health promoters tend to present breast-feeding as an unmitigated 'good thing', which can be influenced by their activities (cf. Morse, 1990). Feminists too have made this assumption (Van Esterick, 1989), or else have linked the practice with the subordination of women, arguing that bottle feeding can liberate women from medical control (cf. Gordon, 1989; Carter, 1995; who observe the same trends), or from the necessity to be a child's sole feeder.

1.5
Examining breast-feeding from the point of view of women who do it is unusual, and many of the studies which have done so have some limitations. MacLean's (1988) Canadian study for example concentrates only on affluent, well-educated women. Carter's (1995) excellent work uses retrospective data, consisting of women's memories of feeding babies between 1920 and 1980 in a working class area of Newcastle. There remains a need for contemporary data from a range of women.

1.6
Increasingly, like MacLean (1988) and Carter (1995), researchers are arguing that there is a need for breast-feeding research to centre on women's own point of view, to facilitate questioning the naturalist model, and the other deterministic analyses, and to explore the process of breast-feeding in social life. In this paper, we follow MacLean (1988) and Carter (1995) in focusing on women's points of view, and examine how women breast-feed, using a modified version of Strauss et al's (1973) and Strauss' (1978) negotiated order perspective.

1.7
Strauss (1978) sees negotiation as a process of social interaction involving bargaining or discussion, which produces a dynamic social order, or 'concerted action' (Strauss, 1978: p. 304). He insists that analysts should examine negotiations in their wider structural context, which includes institutional arrangements and power relationships. He sees negotiation as central to the production of order, though it is entwined with other processes, which may produce other forms of social order, such as 'coerced order', or 'manipulated order': Strauss does not discuss these in detail. Negotiated order can be produced where actors are not equal in terms of power, and when they are antagonistic: there is no assumption of equality or consensus, and the concept covers a wide range of interactions.

1.8
Strauss uses a range of examples to develop his arguments, but much of his most compelling and influential writing, and that of his pupils (Strauss and Corbin, 1997) deals with health contexts: recent work in the sociology of health and illness has reopened debates about negotiated order, especially regarding relationships between staff in health care settings (Svensson, 1996; Allen, 1997). Strauss (1973: pp. 313 - 314) himself discusses the parts played by patients in the negotiated order of a hospital, emphasising the importance of paying attention to their active participation, and moving away from perspectives which can over-emphasise regulation. Strauss's discussion can be linked conceptually with Stacey's (1976, 1988) view of patients as co-workers in the production of health-care, and with recent discussions of co-production in other personal care services (Wilson, 1994; Baldock, 1997).

1.9
Our chief concern in this paper is to focus on women's perceptions and actions in negotiating breast-feeding, especially with health personnel. This necessitates the adoption of a concept of 'project' (in the phenomenological sense Ů Schutz, 1970: p. 26), since we are attempting to understand a process of negotiation with particular reference to the aims, strategies, achievements and disappointments of one set of (diverse) participants. Their 'definition of the situation' (Thomas and Thomas, 1970: p. 154) is our starting point[1]. The 'project' gives an additional, complementary dimension to the concept of negotiation: thus, we define negotiation firstly in the Straussian sense as a process of bargaining or discussion, and secondly, as a process of finding a way, or pursuing a project. Using the term in both these senses, we therefore examine how women negotiate a way to start breast-feeding, to maintain it and then to stop. As Strauss and Schutz both insist, we look at negotiation in its wider social context, with special reference to issues of power, but also to class, race, stocks of knowledge and institutional arrangements. As our research questions focus on women and health personnel, we do not examine the participation of women's full range of significant others in this paper: this would necessitate further research[2]. We see women as active agents in the process of negotiating breast-feeding, and as constrained by wider social factors, including medical power and differential access to resources, especially resources of social knowledge.

The Research Study

2.1
The data on which our argument is based were collected in 1994-5 as part of a larger study of communication between Pakistani women, white women, health visitors and general practitioners (Bowes and Domokos, 1996; 1997)[3]. In total, the data consist of semi-structured interviews with 62 women of Pakistani heritage, who were mothers of one or more children under five, 68 white mothers living in similar areas, 50 health visitors and 25 general practitioners[4]. All the samples were random[5], and the samples of women were weighted to ensure representation of a range of socio-economic groups, using housing area type as a proxy. The assignation of women to one or other ethnic category was done using the 1991 census categories of ethnic groups, and respondents were asked to which of these categories, or which other categories they belonged. Only women who chose 'Pakistani' or 'white' were included in the sample. Our use of these terms implies no necessary commonality of culture within either category: this was treated as a matter for investigation, and, indeed, it was clear from the wider findings of the research that the women interviewed in both categories were differentiated not only in terms of socio-economic group, but also in terms of cultural practices, levels of education, and lifecourse experiences.

2.2
Despite this internal differentiation, there were systematic differences between the two samples of women. The median age of the Pakistani women was 30, and the white women, 33. Pakistani women had a median of three children, and white women, two. Pakistani women were more likely to live in an extended family (27 per cent, as compared with 4 per cent), less likely to have a job outside the home (5 per cent, as compared with 38 per cent), and somewhat more likely to have relatives living nearby (85 per cent, as compared with 73 per cent).

2.3
Interviews with women covered their contacts with and views about health visitors and general practitioners, especially in relation to their role as mothers, and included their experiences of maternity care. Interviews with health visitors covered their experiences of work with Pakistani and white women, their training, their evaluation of services, and cooperation with other professionals, especially general practitioners. The data on infant feeding were extracted from the main data set for further analysis. Not all the women interviewed spoke about infant feeding: where the issue was not discussed, this was generally because some other overriding experience was discussed by the women as central to their relationships with health visitors and GPs, or, on a few occasions, where women preferred not to discuss the topic[6]. Data on infant feeding were collected from 35 Pakistani women, and 44 white women[7]. In the paper, we concentrate mainly on those women who reported breast-feeding[8], ie. 23 Pakistani women and 33 white women (a total of 56).

2.4
All the interviews were transcribed, and indexed using categories derived from the interviews themselves, thus preserving the views of respondents, and attempting to avoid the imposition on the data of preconceived categories: the derivation of index categories involved extensive cross-researcher checking, and constant reference back to the transcripts themselves. The scrutiny of data was then completed with the aid of the NUD*IST software[9]. In presenting data, we refer to patterns established through systematic analysis, and include representative quotations. Interpretation of quotations has been done in the light of the whole transcript, to ensure quotation is not made out of context. Despite similarities in women's experiences, we maintain the ethnic identification of quotations[10] to ensure transparency of presentation, and to help demonstrate where ethnic differences occurred (in fact this was rather infrequently), and where they were cross cut by other differences.

Breast-Feeding Rates

3.1
Health Board statistics (GGHB, 1994), and English national statistics (Thomas and Avery, 1997) suggested that there would be systematic differences between Pakistani women and white women, with Pakistani women more likely to breast-feed than white women, and also quicker to stop breast- feeding. Among the women interviewed there was a rather different pattern, in that about two thirds of Pakistani women reported breast-feeding their babies, whereas about three quarters of the white women did so. This difference in overall rates was not statistically significant[11]. Several health visitors commented on what, in their experience, were low rates of breast-feeding among Pakistani women: they saw these as reflecting local values (across Glasgow, not only in the areas sampled). It is worth noting that in parts of Glasgow (which were not included in the study, as they are outside the main Pakistani areas), the general rate for breast-feeding is as low as nine per cent (Campbell, 1994).

3.2
The lowest rates of breast- feeding in Glasgow are found in poorer areas, with bad housing and high unemployment. Such a pattern is reflected nationally (Foster et al, 1997) for the general population, but not for Pakistanis and Bangladeshis (Thomas and Avery, 1997), for whom socio-economic group is not a predictor of breast-feeding behaviour. The women interviewed showed a similar pattern, in that the rate of breast- feeding was similar throughout the Pakistani category, whereas white women in higher socio-economic groups were more likely to breast-feed than white women in lower socio-economic groups[12]. We examined the data for other factors which might predict breast-feeding for Pakistani women. Age, place of birth, fluency in English, proximity of relatives were not predictors. It is tempting to conclude that Pakistani culture therefore promotes breast-feeding. But we will argue that 'Pakistani culture', as represented by the world views of the women interviewed, was differentiated: uniform views of breast-feeding could not be identified, as they could not among the white women.

Beginning Breast-Feeding

4.1
All the women interviewed had given birth to all their babies in hospital, and all who had started breast-feeding had done so in hospital. The context of birth and infant feeding are seen by those adopting a holistic approach to the study of breast-feeding (eg. Maher, 1992; Carter, 1995; Morse, 1989) as basic.

4.2
Prior to their hospital experiences, women had negotiated their initial decisions and intentions regarding breast-feeding. This reflects reports in the literature (eg. Bryant 1982) that decisions about whether to attempt breast-feeding are negotiated outside the context of contact with health services. In outlining their breast-feeding projects, women varied in their approach, and in the stocks of knowledge they used. Thus, some were determined to start breast-feeding, and spoke of strong and reasoned commitment to the practice from the start. Others felt it was best for their babies, and several referred explicitly to health promotion messages which reflected the naturalistic model of breast-feeding. Women who did not breast feed[13] either spoke of having positively chosen to bottle feed, or spoke of rejecting breast-feeding for reasons which included not liking it, being embarrassed about it, or problems with the baby, especially prematurity. All these views were expressed by both categories of women, and all represented women as exercising choices in relation to breast-feeding.

4.3
From both white and Pakistani women, there were suggestions that bottle feeding was seen as usual, and the rather short duration (see below) of much of the breast-feeding discussed supports this interpretation. A few Pakistani women commented that bottle feeding is a Western practice and a modern practice, and that this had influenced their wish to pursue it: sometimes, there was pressure on them from older people in their families to bottle feed for these reasons (cf Mull, 1992, who describes similar pressures in Pakistan). It is important to note that Pakistani views were not universally presented as against breast-feeding. One Pakistani woman identified community pressure against breast-feeding:

Asian women here gossip and think it's a bad thing to breast- feed. (PW138)

She had preferred to breast-feed, but did not do so, following pressure from her in-laws. A contrasting view was presented by PW105:

I like breast-feeding, because in Pakistan, all ladies give two years. Two years feeding is [a] must. (PW105)

This woman, whose case is referred to later, was a determined breast-feeder.

4.4
White women were less likely to discuss family influences on their feeding decisions, and were more likely to express their choices as having been made autonomously. However, some of the health visitors felt that older white women were inclined to prefer bottle feeding because of its popularity among a former generation of mothers, the mothers of the generation interviewed for the study, who were likely to have influenced their daughters' feeding decisions (cf Blaxter and Paterson, 1982). And there was also certainly a contrasting, mainly white, middle-class culture which was committed to breast-feeding, and worked hard at developing expertise. Thus, although women presented their decisions as autonomous, they had made them in the context of negotiations with significant others.

4.5
These data on deciding whether to breast-feed begin to reveal some of the variations in knowledge at hand between and within the categories of women. Bottle-feeding as usual or normal was a view common in both categories, and for some Pakistani women, this was linked with particular views about the nature of modernity, though the opposite point of view, that breast-feeding was traditionally Pakistani, also existed. The committed white middle class drew their knowledge of breast-feeding from books, and (less often) from formally organised support groups such as the National Childbirth Trust. No Pakistani women mentioned such an organisation: committed Pakistani women were more likely to refer to general ideas about the benefits of breast-feeding, and less inclined to produce technical accounts of its mechanisms.

4.6
Generally, and perhaps unexpectedly, the women interviewed expressed few links between the experiences of infant feeding and their ideas about their bodies and, particularly their breasts[14]. Some commentators (eg. Morse, 1989; Carter, 1995) have seen such ideas as a contextual influence on breast-feeding behaviour. Embarrassment was not a common reason for avoiding breast-feeding, being mentioned by only two women (one white, one Pakistani). Similarly, returning to work (also seen as important in other literature eg. Carter, 1995: pp. 121 - 127) was not a large consideration: it was mentioned by only one woman.

4.7
For some women who wanted to breast-feed, a decision not to do so had been imposed: two white women and one Pakistani woman stated that they had not been asked their preferences in hospital, and that hospital staff had assumed they would bottle feed. These experiences look like Strauss's (1978) 'coerced order', in which the actions and intentions of the women had no influence on the outcome. Two of these women had disabilities: it is likely that their experiences relate to constructions of disabled people as unable, and their resulting apparently total disempowerment.

4.8
Once breast-feeding had been tried, there was about a one third 'drop out' rate after six weeks for both groups of women - they appear to have been more persistent than some statistics suggest (Thomas and Avery, 1997: pp. 5 - 6). Some of the 'drop outs' felt that they had breast-fed for long enough, and stopped for that reason: they presented themselves as having completed their project as they wished. Other women had experienced various problems (see below). The longest period of breast-feeding among the whole group of women was 11 months for a white woman, and 16 months for a Pakistani woman (born in Glasgow): the majority of women (53 per cent) breast- fed for between 7 weeks and three months (less than the WHO minimum).

4.9
Some women had given up breast-feeding apparently easily and quickly after their early attempts. According to their accounts, they had thought of breast-feeding, but had not received the advice or help they wanted when in hospital after the birth. They had then resorted to bottles, or allowed hospital staff to resort to bottles, perceiving this as 'easier', ie. it did not involve women in learning the skill of breast-feeding, and it did not take up hospital staff time.

They are so busy, they don't have the time to sit and help you to do it. They really don't. They are rushed off their feet, and are quite harassed, and I was quite willing to give up....That was just agreed on all sides, just to go for feeds [i. e. bottle-feeding]. (WW248)

4.10
Interestingly, this woman herself presented the decision in terms of a process of negotiation, and her perception of the hospital staff as being too busy to help her, linked with her own willingness to stop breast-feeding indicate the potential influence of the hospital environment. The same woman subsequently did not attempt to breast-feed her second child, but, regretting that she had not done so, planned to try with her third:

And then it turned out it was twins, and I thought I'd still try, but I don't think I was really a hundred per cent behind it. (WW248)

She gave up very quickly.

4.11
For many women, hospital staff were their only source of help at this stage. For example, PW123 had relied completely on hospital advice, explaining that, as no-one in her family, to whom she would look for advice in other situations, had breast-fed. Thus the hospital staff could be key players in the process of negotiation, here in closing off women's action by declining the support they sought, and (above) in assuming women's decisions:

No-one asked me what I wanted. (WW209)
All I needed was a bit of encouragement, which I didn't get. (WW253).
They didn't take the time to sit with you. (PW134)[15]

4.12
Negotiation involving hospital staff produced a range of perceived outcomes, as women sought support in their projects, experiencing uncertainty and lack of expertise. Some women felt that they had not received enough support, and that this had led to the failure of their attempts to breast-feed:

From day one, I thought I would breast-feed, but when I went into the hospital, and I wasn't getting much help, I just thought stuff it....I didn't even know how to start myself, and the nurse showed me once, but after that I still couldn't do it....and I started getting myself depressed and anxious, and I thought 'No. I won't be able to cope'. (PW130)

The hospital staff had, she felt, been too rushed to help her:

It's not a five minute job, you know. I'd rather have somebody spending more time. (PW130)

WW202 had also wanted to breast-feed:

I wanted to breast-feed in the hospital, and I couldn't. I wasn't managing on my own, and the nurses were never there when I wanted to feed the baby. They were always busy with someone else. So [I tried] it myself, and I made a total mess of it. I got [so] I couldn't breast-feed then in the end. (WW202)

4.13
This woman spoke of a struggle with herself, and with the hospital routines in her efforts to breast-feed, and described losing the struggle. Women who reported lack of support in hospital described themselves as feeling unable to ask for more support, and referred to staff lack of time, or what they saw as their own inadequacies. Their ability to seek more support was thus inhibited.

4.14
Whilst there were women whose efforts to get help had failed, there were others who used terms such as 'absolutely wonderful' (PW113) to describe midwives who had sat with them through feeds, and advised them how to deal with problems such as pain, engorgement, cracks, babies who did not latch on easily and so on. For example, PW113 described her experience of support in hospital:

I can't say enough about them. They were really great. If they hadn't supported me, I don't think I would have persevered with my breast- feeding. They used to come in and spend hours. You really felt at ease with them. (PW113)

The hospital midwives had helped the baby latch on, and had left the woman a bell to call them if she had problems. Another woman described a similar experience:

I couldn't sit properly because of my stitches. They told me to feed him lying down. So I would lie down, and they'd put him next to me, and then they'd put my buzzer next to me, so if there was a problem, just to call them. ÷ At times, it was sore ÷ but they quite encouraged you to breast-feed, which was good. (PW115)

In these cases, concerted action, in the form of the co-production of breast-feeding support had enabled women to succeed in their aims.

4.15
For some women, such co- production had allowed them to make difficult decisions. WW221 was anxious about breast-feeding, and found it painful. She described the support she had received from a midwife:

She sat with me every night she was on night shift, and I sat bubbling my heart out, and she sat cuddling me. (WW221)

4.16
She had worried that her baby son (described affectionately as 'this wee bugger') was not getting enough milk from her, and had become increasingly upset. Eventually, the midwife persuaded her to phone her husband (in the middle of the night), and discuss her problems with him. Thus, she was able to allow herself a supported decision to stop breast-feeding.

4.17
Staff attention could also be experienced negatively: one white woman felt support had not been sensitively given:

Everyone was in, manhandling me. (WW238).

4.18
A Pakistani woman felt that staff had nagged her rather than supported her, and described how she had lied to a staff member about having tried to breast-feed 'to get her off my back' (PW130). The 'manhandled' woman had continued to breast-feed, but saw leaving hospital as a welcome release, which then allowed her to pursue her own strategy. The second woman subverted what seems to have been a staff member's attempt to help her breast-feed, which she experienced as interference. Both women described active participation in their negotiations with hospital staff. Their strategies can be compared with some of the patterns of resistance (escape and concealment) discussed by Bloor and McIntosh (1990).

4.19
All these examples portray hospital staff as influential in breast-feeding practices, and the women attributing the results of their negotiation to staff influence. So it appears that staff could prevent, stop or promote and support breast-feeding. The outcome of negotiation appeared somewhat fragile and contingent, in that for example the availability of staff at crucial times could tip matters one way or the other. There is a comparison here with other negotiation processes, such as the negotiation of sexual encounters discussed by Holland et al (1990), where apparently small factors or incidents could affect outcomes, whatever the projects of negotiators, and however determined their action plans. The fragility of the process could also be seen in the 'feeding stories' (Carter, 1995: p. 200) of individual women. For example, PW134 talked of her experience with all her children. For the first, she gave up breast-feeding after three days, having received no help in hospital. She fed the second for two months, but, 'cracked and sore', she then gave up. For the third child, she did not attempt breast-feeding:

I didn't bother, because I was only going to have sore breasts and everything. (PW134)

At the time of interview, she was happily breast-feeding her fourth child (then aged two and a half months). She explained that she had been helped with latching-on in hospital, and felt that this had made all the difference:

If I would have had that kind of support the first time you know, I would have been able to breast-feed all my children. (PW134)

In this case, the role of hospital staff in the negotiation process was especially clear. For her fourth child, her own wishes, the knowledge and the time given by hospital staff had combined to enable her to breast-feed.

4.20
A more critical perspective on hospital staff was evident from those white, middle-class women with confident knowledge of breast-feeding. Such women described themselves as particularly active in the negotiation process. Several reported that hospital staff had pressured them into using supplementary bottles during their hospital stay, a practice to which they strongly objected. One woman speaking of staff who had given bottles to breast- feeding babies, argued that this was detrimental to mothers:

They needed the stimulation [of the baby feeding]. These were midwives that are supposed to know things like that. (WW244)

4.21
She explained that she had found it hard not to contribute her own knowledge and experience of breast-feeding (she was discussing her second child) to help women in hospital, but she had not felt this was the right thing to do. In contrast, WW265, who had previously breast-fed her first baby, and also had rather a different, but still apparently effective, stock of knowledge, had been reprimanded in hospital for giving her second child a supplementary bottle. She related how a health visitor who had told her 'Do what's best for you' had provided her with the kind of support she felt she needed. These cases illustrate the difficulty many women experienced in contesting a hospital regime, even where they were confident in their own expertise. They also illustrate that there were varying kinds of expertise, different stocks of knowledge which women could use in finding a way to breast-feed. These stocks of knowledge could be seen as resources for breast-feeding negotiations, but it was not always possible for them to be fully deployed.

4.22
There was discussion by the women of the nature of hospital regimes, and their influence on infant feeding. Some women felt that bottle-feeding mothers received more support, in that their babies were cared for overnight by midwives. WW263 for example, felt that her baby wouldn't stop crying, and that she needed a break, but that she did not get the help she would have done had she been bottle-feeding:

I think it must be different for people that bottle-feed their babies. They [the babies] get taken away [to the nursery on the ward]. But the breast-feeding mothers sometimes feel that they're left to get on with it. (WW263)

4.23
Some women felt that the emphasis on breast-feeding was too strong:

I certainly felt it was just ridiculous, the amount of pressure that was put on you to do it.....I don't think [women] should be made to feel somehow or other that you've failed if you can't manage it. (WW201)

WW211 felt that mothers who breast-fed were believed incapable of doing wrong:

I think they have a very naive attitude to mothers that breast- feed, and a very ignorant one to mothers that don't. But that's the government. They are pushing that forward. (WW211)

Thus, hospital practices linked with infant feeding were a topic of live debate for these white, middle-class women, who looked at them in a context wider than their own feeding decisions. The contrast between the views that, on the one hand, bottle- feeding, and on the other hand, breast-feeding, received more support illustrates that breast-feeding was not universally promoted by the hospitals or hospital staff, and that a monolithic view of their approach would be mistaken[16].

4.24
Hospital advice, though crucial for many women, was not essential to all who negotiated breast-feeding. WW219, for example, described herself as not really committed to breast-feeding, and said she did not receive help in hospital:

I hadn't a clue. I just did it. (WW219)

But she was very positive about her experiences of breast-feeding:

I love breast-feeding÷..I learned to just love it. (WW219)

Negotiation and Constraint in Hospital

5.1
When the women interviewed were in hospital, the power of the hospital regime appeared central, and women's decisions about feeding could be decisively affected at this stage. The medicalised dimension of breast-feeding thus has to be seen as part of the context of negotiation, and the institutional structure and organisational arrangements of the hospitals are an important part of this context. Ball (1994: pp. 118 - 119) points out that maternity care is an emergency-driven service, and that at times, attention to individual breast-feeding mothers is simply not possible, because of the needs of women in labour and delivering, whose needs are prioritised by the service. Furthermore, as currently constructed, with on-demand feeding, breast-feeding is also a kind of 'emergency', requiring attention unpredictably. Thus, women may be placed in the position of competing for the scarce resource of midwives' time, and a more articulate and committed breast-feeder may well succeed in getting help, where a less decided woman may not. Many of the women interviewed had perceptions of what was 'easier' for hospital staff (bottle-feeding), and saw staff as too busy to help them. Women who had more expert knowledge also deferred to the hospital model: they did not feel it right to offer their own expertise, and found questioning hospital practice difficult.

5.2
Though further work is needed on the activites and views of hospital staff, other research suggests that hospital staff may approach women differentially, and that these differential approaches may be a significant contextual factor. Bowler (1993) and Torkington (1995) discuss the use of exclusionary stereotyping of Asian and black women health service users, and Jeffrey (1979) demonstrates other stereotyping processes. Our data did not show negative stereotyping to the extent that Bowler and Torkington identify, but did demonstrate white health service staff (health visitors and general practitioners) constructing Pakistani women in general, and white working class women as 'other'. They did not differentiate Pakistani women in class terms. Though unable to develop the discussion fully here, we have presented some evidence of health visitors believing that white working class women and Pakistani women generally are less inclined to breast-feed. If hospital staff were operating with the same view, and their expectations affected the level of help they offered, this might contribute both to the somewhat lower rate of breast-feeding among Pakistani women, and to the lack of class differentiation in breast-feeding patterns in this group. It might also be a factor in the low rates of breast-feeding among white women in lower socio- economic groups.

Breast-Feeding at Home

6.1
The negotiation of breast-feeding continued once women had left hospital. Most of them spent only a few days there (a shorter stay was general for second and subsequent children). The early days at home, it should be noted, were a period in which many interviewees had stopped breast- feeding, following wider trends. According to Raphael and Davis' (1985) cross-cultural study, support for the newly delivered mother is a central factor in ensuring that breast-feeding continues. But breast-feeding at home was described generally by women as a lonely pursuit, fraught with difficulties which they faced alone. Many accounts described a process of overcoming problems, especially physical problems, and persisting with breast-feeding in the face of difficulties.

6.2
One of the strongest themes in the accounts at this stage was physical pain associated with breast-feeding. PW105 (referred to earlier), saw two years' breast-feeding as essential. She described her nipples as 'swelling and running with blood', but had been given a nipple shield by her GP and a cream (calendula) by her health visitor. She had persisted in her efforts to breast-feed, which had proved successful. Similarly, PW113 carried on feeding for three months, with health visitor support, despite problems

My nipples were so sore, they were cracked. And the midwife and the health visitor said how brave I was, trying to feed. I tried, because I wanted to feed them very much. (PW113)

6.3
This emphasis on the physical problems and pain of breast-feeding went alongside a lack of discussion of other embodied aspects of breast-feeding, including for example Morse's (1989) association between perspectives on breast-feeding and sexuality. It is possible that this association is not relevant for women who have decided and started to breast-feed. An alternative explanation, which would also explain the emphasis on pain itself may lie in the constructions and experiences of pain associated with childbirth and its aftermath, which appear central to the experience of breast-feeding, but which do not appear to be linked with sexuality. Both the medical model of childbirth and its alternative, 'natural birth' (Annandale and Clark, 1996: p. 31) disassociate pain and childbirth, the former by the use of drugs, and the latter by seeing pain as medically constructed (constructing itself on the basis of the model it opposes - Annandale and Clark, 1996)[17]. But the women interviewed in common with those in many other studies, experienced pain in childbirth, and afterwards: it became a defining aspect of the experience for them, and entered their stock of knowledge at hand. The experience of pain conflicts with the notion of breast-feeding as a natural practice, which might be expected to 'come naturally' (simply, easily, effortlessly). It is a reminder for women of their skill and effort, and of their success or failure in their breast-feeding project. It also emphasises at once the loneliness of breast-feeding (only one body can do it), and its reliance on sociability, the need for help and support from others: the others who can help do not appear to be sexual partners.

6.4
After the tenth day following the birth[18], women's main source of support for breast-feeding seems to have been health visitors, especially where family members and friends were ignorant of, indifferent to or even hostile towards breast-feeding. For example, PW123 explained that none of her female relatives had breast-fed their babies, and that, though she had had help whilst in hospital, she had found latching her baby on difficult, and no-one at home had been able to help her, 'because they never did it themselves' (PW123).

6.5
Women in pain who looked to health visitors for support were generally positive about their experiences. For example, WW247 had had some difficulties feeding her baby:

I said I'd had enough. I was giving up breast-feeding. I was just going to give her bottles because it would be so much easier....And she said how about trying one of these, and she gave me a nipple shield, and that was fine. (WW247)

Another woman had also received what she described as good, no nonsense advice, which had helped alleviate her anxiety about feeding her baby:

She came in 'Oh, you're not eating enough to keep a sparrow, never mind feed your baby' and told me what to do and all that. She was excellent. (WW262)

For some women, health visitor support had been decisive:

I would have given up without her. (WW215)

6.6
There were some accounts of very considerable efforts by health visitors ('fringe work' in Delacuesta's (1993) terms) to provide effective support: particularly striking was an account given by WW239 of a health visitor who had come to the house during the evening, and stayed with a baby throughout its characteristic fretful period, had then offered appropriate advice to the mother, and left a home telephone number for use if problems continued. In these accounts, health visitors are presented as allies in the struggle to find a way to breast-feed, and as active participants in breast-feeding projects.

6.7
There appeared to be a link between some of the physical problems described by the women and a lack of effective stocks of knowledge. When women spoke of having too little milk, there were indications that they did not know how to increase their supply. For example, PW102 explained that her baby had not been putting on weight while she breast fed him, and the 'health ladies' (health visitors) had said he was not getting enough milk. So she began to give him what she described as 'normal milk'[19] (ie. infant formula in bottles), and he began to gain weight. But she still felt she had not received enough support:

I would have liked to continue giving breast feeds....I would have liked more advice....Sometimes you don't know what to do....So I just kept quiet, and kept giving him bottles. (PW102)

6.8
She had been disappointed that she had given up breast-feeding, and it was apparent that she had not been supported in her attempts to continue. There was, for example, no indication that she had been given advice on how to increase her milk supply. In this case, the health visitors had served rather to discourage her efforts to breast-feed.

6.9
Similarly, WW221 did not recognise that she needed GP attention for her mastitis, and had to be advised to visit the GP by her health visitor. And when women spoke of damaged nipples, it appeared that they did not know ways of dealing with this problem. The accounts already noted show previously unknown remedies and advice being dispensed by health visitors and GPs, to women who lacked not only sources of support other than health professionals, but also effective knowledge of how to deal with problems.

6.10
Challenge to health service regimes continued as women breast-fed at home, where less covert and possibly more effective challenges seem to have been possible, away from the institutional constraints of the hospital regime. Some of the women who were very committed to and articulate about breast-feeding criticised advice they had received from health visitors, such as to limit the duration of feeds, or to introduce solids very early[20]. They were contesting the professional stocks of knowledge which the health visitors were using. These women felt that some of the health visitors' knowledge of breast-feeding was old-fashioned or limited.

I would have appreciated someone who knew something about breast-feeding.....I don't know if they [health visitors] go for refresher courses or not, but I think they should. (WW244)

WW210 described being threatened by a health visitor, when she was experiencing breast-feeding problems:

She said 'Well, that's two weeks, and she hasn't made up her birth-weight. It just means we'll have to take the child to care if you're going to persist with this breast-feeding.' I thought that was a terrible thing to say. (WW210)

Another woman was advised to give up breast-feeding when her baby developed a cold. All these women actively and critically resisted the advice given to them.

6.11
Resistance was also expressed by women whose stock of effective knowledge was small, but who held views which conflicted with those of professionals: two believed that if they did not drink milk themselves, they would not produce enough for their babies, another stopped breast- feeding because she thought the extra food she felt she wanted would make her fat.

6.12
There is a reminder in the potentiality for health visitors to give or withhold effective advice of the continuing exercise of professional control, despite the changed context between hospital and home. The comment about taking a child into care highlights the continuing ambiguity of the role of the health visitor, between support and surveillance (cf Bloor and McIntosh, 1990). At home, health visitors apparently had an important support role, as women continued to negotiate breast- feeding. This finding contrasts somewhat with Carter's (1995) work, which suggested that, once they were at home, women rejected the work of health professionals as intrusive. The difference may relate to changes in health visiting towards more enabling practice (see Bowes and Domokos, 1998), as our data come from the late 1980s (when the respondents' first children were born) to mid-1990s (the time of the study), whereas Carter's deal mainly with 1920-1980. More enabling practice might, to follow Wilson's (1994) argument, facilitate more effective co- production. Certainly, the health visitors we interviewed saw breast-feeding support as an important part of their job, and women generally appreciated the service they had received: in general, women who criticised health visitors were mostly white women[21].

6.13
It was interesting to note that, as the women explained it, the end of breast-feeding seemed very frequently to have been brought about by something other than their own wishes, sometimes other social actors. It was common for women to report that they had stopped breast-feeding because of a lack of milk:

I lasted ten weeks with both children, then after that, I felt they needed something more than myself. (WW245)

6.14
Giving 'something more' did not necessarily involve weaning: usually, breast milk was first supplemented, and then replaced, by formula milk. Physical difficulties were also presented as a common trigger for stopping breast-feeding. For example, WW221 described the pressure she felt, after an attack of mastitis:

I just felt dreadful after that. I just felt I had absolutely no time whatever on my own÷..I didn't even know if he was getting enough. (WW221)

She had stopped breast-feeding, but felt that, with support, she might not have done so. Several women had been persuaded by others to stop breast-feeding: sometimes, a health professional had advised them to stop, whilst in other cases, families had put pressure on them to do so. For some women, their babies were a cause of stopping breast-feeding: 'greedy babies' (eg. PW113) or babies that 'would not feed' (eg. WW221) were described as influencing decisions. PW161 linked her own feelings and her baby's character:

Six weeks I breast-fed her, but I didn't enjoy it. Apart from that, she wasn't taking it very much, and she was not a very happy child, so I took her off it. (PW161)

6.15
Other commentators have also noticed the tendency to seek external explanations for ending breast-feeding, which has been interpreted (eg. by Carter, 1995) as a way of dealing with the guilt which ending breast-feeding may provoke, as it may suggest resistance to the wishes of professionals, or question women's perfection as mothers (Blum and Vanderwalter, 1993). Similarly, when women talked about their problems breast-feeding, they spoke of their need for and appreciation of support. These areas of data showed women defending and explaining their feeding experiences often as things which happened to them, rather than events over which they had control. Here, the role of constraint, and the difficulties of finding a way to breast-feed are particularly apparent.

Conclusions

7.1
From a social constructionist viewpoint, the assumption that breast-feeding is socially negotiated is, perhaps, not surprising. It is striking however, that from the point of view of the women whose accounts have been discussed in this paper, a socially negotiated process appears so lonely and so difficult. Repeatedly women, perhaps reflecting an expectation that a 'natural' process should 'come naturally' to them, presented accounts emphasising both their own responsibility, and their own failure or success in different stages of their breast-feeding projects. What from an analytical perspective was concerted action, or negotiated order, was presented by women as individual action, with some support and some obstruction from others.

7.2
Previous research on breast- feeding has been preoccupied with the search for single factors (less often, combinations of factors) promoting or discouraging it, generally in an attempt to discover why rates of breast-feeding are low or less low in different areas, social classes or ethnic categories. In arguing against the search for determining factors, we suggested that a sociological understanding of breast-feeding experiences could be developed by conceptualising breast-feeding as a project pursued by women through a process of social negotiation which would result in concerted action which might facilitate the original project, modify it, or bring it to an end. Thus, moving away from the preoccupation with what makes women breast-feed or not, we wanted to examine and explain how they find a way to breast-feed.

7.3
In deciding on the aims of their projects, many women presented themselves as departing from the 'normal' bottle- feeding in wanting to breast-feed. It was clear that breast-feeding was seen as involving deliberate choice, and that a wide range of factors, some of which were health promotion messages, were considered when the choice was made. Strong commitment to breast-feeding could involve study, the deliberate acquisition of new stocks of knowledge.

7.4
Once a woman set out on a breast-feeding project, our discussion showed that a whole range of factors could influence both her success or failure, and the continuing process of decision making that she faced, concerning whether to continue, how to deal with problems, whether to supplement, when to stop and so on. Whilst these factors influencing success or failure can be identified however, the particular history of each breast-feeding project showed its own trajectory, and though we can highlight some regularities, there is a strong sense in which the projects were indeed individual and particular to the women pursuing them.

7.5
Breast-feeding projects were constrained by hospital regimes, in that the institutional set-up and organisational norms made it difficult for women to receive the staff support they wanted. Furthermore, the hospital was perceived by women as a context in which it was difficult to ask for what they wanted, and difficult to contest and challenge. They saw hospital staff as a key source of help, but also as too busy to provide such help. For Pakistani women and white women in lower socio-economic groups, it is likely that wider factors of social exclusion, in the form of stereotypical views of their wants, needs and experiences, further restricted their possibilities of getting help with their breast-feeding projects (though this does need further research). Many women had limited or ineffective stocks of knowledge about breast-feeding, and needed to acquire relevant knowledge and skills. The fund of breast-feeding lore which exists, for example in some Muslim societies and elsewhere (Khatib-Chahidi, 1992) appeared absent among them. Gaining access to knowledge and skill involved effective negotiation. Lack of knowledge and skill, combined with a childbirth and puerperium in which pain was denied but experienced could increase the significance of pain, sharpening women's focus on their own bodily experiences, and promoting the sensation of the lonely struggle. Women's dependence on health personnel for support in their breast- feeding projects was striking: health personnel appeared in women's eyes to possess the knowledge and expertise they sought. Dependence was reinforced by the limitations of wider stocks of knowledge. All these were patterns of constraint, important dimensions of the wider context in which negotiation took place: each breast-feeding project might be restricted by any or all of these factors, and for some, there were serendipitous elements. So, for example, a woman seeking help in hospital might look for it at the same time as other, more assertive women, and therefore lose out through contingency. Or a woman at home experiencing mastitis might not be visiting a clinic that day, and not receive the necessary advice to visit her GP. Chains and coincidences of events could thus be of significance in constraining breast- feeding projects.

7.6
Breast-feeding projects were facilitated by resources which women could bring into action. An effective stock of knowledge of breast-feeding, which could include experience of having done it before, was an important asset, reducing women's need to seek out help and advice for every problem, and hence their dependence on midwives and health visitors. Being white and middle class, ie. belonging neither to the working class nor the Pakistani category (both potentially subject to social exclusion), also appeared helpful, and was linked with more assertiveness in seeking help and contesting health personnel's views. A further factor was staff ways of working. In hospitals, despite the constraints of regimes and routines, and in the community, especially among health visitors, it appeared that a more enabling approach to practice was more effective in co- production of breast-feeding projects. And, as with the constraints, particular chains of events in particular cases were strongly influential in project outcomes: encountering the right health visitor at a crucial time could result in markedly different concerted action.

7.7
This detailed examination of the social processes of women's breast-feeding projects thus leads to a rather depressing conclusion. We have demonstrated how it is that white, middle class women breast- feed with the greatest success, how even they do so with considerable difficulty, and using complex social skills. Other women face powerful constraints, which appear to overwhelm their attempts to negotiate breast-feeding, and their tendency to give up breast-feeding, even when they are determined to do it, is not surprising.

7.8
The study also has some important limitations, and suggests areas of potentially fruitful further research. Firstly, and in common with much research on breast-feeding, the interview data were retrospective, even though many of the women were still feeding a baby. This presented problems of interpretation, especially where women were recounting decisions they had made with which they were not comfortable. In analysing the data, it was important to try and disentangle women's accounts of their experiences from their justifications of them: as the interviews were very extensive, this was possible, but not, we are sure, infallible. Thus, an examination of prospective data, which captures negotiation as it occurs, is merited. Secondly, the study did not include the perspectives of hospital and community midwives, the other group of health personnel with whom breast-feeding women have close contact as they negotiate breast-feeding. Research is needed to ascertain the perspective of these personnel, and their styles of working. Thirdly, the conceptual model suggests that women's significant others are involved in their negotiation of breast-feeding. Interestingly, the women interviewed here rarely mentioned their partners, friends, or family in discussing breast-feeding, giving the health personnel a much more prominent role. This may partly be due to the focus of the interviews on their interaction with health personnel, but the interviews ranged very widely, and significant others were discussed in other contexts. The issue of why breast-feeding seemed to be such a lonely pursuit, and such a personal struggle for many women, may be further illuminated by work on the role of significant others.

Notes

1 Allen (1997) has argued that only observational data will allow the examination of negotiated order, implying that this can be grasped objectively, and that the objective account is the only valid one. She is critical of Svensson (1996) for his use of interview data. Our work is based on interview data, which is used to examine negotiations from the point of view of the women interviewed, on the basis that 'situations defined as real are real in their consequences' (Thomas and Thomas, 1970: p. 154).

2 Currently in the planning process.

3 The research was funded by the Chief Scientist Office, Health Department, the Scottish Office.

4 This paper focuses mainly on the data from the women. A detailed discussion of the health visitor data is published elsewhere (Bowes and Domokos, 1998).

5 Sampling strategies were as follows:

  1. For the sample of Pakistani women, census data were used to identify areas of Glasgow and environs where more than one percent of the population was of Pakistani origin. To maximise the chances of finding households where there were mothers of children under five, areas of known high student and older persons population were excluded. Polling districts corresponding with the postcode areas were identified using detailed maps, and all Muslim households identified from the electoral roll ( it was known that nearly all Muslims in the selected areas are of Pakistani origin). A random sample of 382 households was drawn from this list, weighted so that 287 were from more disadvantaged areas, and 97 from more affluent areas. A sifting survey was then conducted, which identified 80 Pakistani mothers of at least one child under five years of age within the larger sample. 62 of these women were subsequently interviewed (response rate 78 per cent). Census data had suggested that about one in four Glasgow Pakistani households would include at least one child under five, thus the response rate for the sifting survey is considered good. It is suspected that the women who did not agree to be interviewed, or who were not identified by the sifting survey, were those most isolated and less confident. It should be noted that only one in four of those interviewed were in contact with community groups: this sampling method therefore calls into question sampling strategies which use such groups for snowballing.
  2. The sample of white women was intended to provide a comparator group for the Pakistani women, in similar circumstances. A sample of 380 white households was generated by moving ten addresses up from each of the Muslim addresses. The addresses thus generated were checked against the directory of post codes to ascertain, as far as possible, if the address existed. If it did not, counting returned to the start of the street, on the same side. The addresses were also checked against the full list of Muslim addresses drawn from the electoral roll to ensure that the address was not a Muslim one. If it was, the next address in the street was listed. White households were much less likely to include a child under five, so, for the screening survey, interviewers were instructed to treat each address as the centre of a cluster (cf Brown 1984). If there was no household in the target group at the address, they called at the two neighbouring addresses. There was still some uncertainty, once the list of 380 addresses had been generated, as to whether they all existed: it was not always possible to identify gap sites, or to ascertain the number of addresses in a street. Interviewers were instructed that if an address could not be found, they should go to the nearest house. The sifting survey identified 80 white women with at least one pre-school child, and 68 of these women were interviewed for the study (response rate 78 per cent).
  3. A random sample of fifty General Practitioners was drawn from the Health Board lists covering the areas in which the women lived. 25 of these agreed to be interviewed (50 per cent response). This response rate is considered good for this extremely difficult group.
  4. All but one of the local health visitors agreed to be interviewed (98 per cent response).

6 It is possible that women who did not discuss infant feeding might have had different experiences from those who did. But the extracted data reveals a range of experiences and strategies. Throughout, we emphasise this range, and avoid quantitative discussion of the more detailed findings.

7 The characteristics of this sub-sample were broadly similar to those of the full sample, ie.the median age of the Pakistani women was 30, and the white women, 33. Pakistani women had a median of three children, and white women, two. Pakistani women were more likely to live in an extended family (27 per cent, as compared with 3per cent), less likely to have a job outside the home (5 per cent, as compared with 38 per cent), and somewhat more likely to have relatives living nearby (85 per cent, as compared with 80 per cent).

8 Breast-feeding is here defined as any attempt to do so.

9 A set of tools for organising data.

10 PW denotes a Pakistani woman, WW a white woman.

11 Chi square = 1.745 (1d.f. p=.186)

12 For Pakistani women, the relationship between socio-economic group and breast-feeding was not significant (Chi-square = 3.407 (5d.f. p=.637)). For white women, breast-feeding was strongly associated with higher socio-economic group (Chi-square = 17.097 (5d.f. p=.004)).

13 These women would require a separate discussion.

14 Later in the paper, we discuss some of the bodily experience of breast-feeding, and suggest that experiences of pain over-ride other bodily considerations.

15 A recent (January 1998) BBC 'Watchdog' campaign about breast-feeding also emphasised women's reports that they had not received the necessary support for breast-feeding whilst in hospital.

16 The Glasgow Women's Health Book (Glasgow Healthy City Project, 1993) states that women are given information about both breast and bottle feeding, and emphasises choice for women.

17 It is worth noting here that reports in the literature suggest that Pakistani women may experience denial of their pain on the basis of exclusionary stereotyping: Bowler (1993: p. 166) reports midwives insisting that Asian women (including Pakistani women) were inclined to 'make a fuss about nothing' during childbirth, ie. to cry out in pain unnecessarily, and Martin (1989: p. 153) makes a comparative point about black women in the U.S.A.

18 After the tenth day, the health visitor takes over from the hospital and community midwives in post-natal care.

19 Echoing the 'normality' of bottle-feeding discussed earlier.

20 Interestingly Blaxter and Paterson (1982: pp. 134 - 135) report that the working class women they interviewed were very keen to start their babies on solid food at a few weeks old, to the consternation of health visitors and clinic doctors.

21 See Bowes and Domokos (1998) for further discussion.

References

ALLEN, D. (1997) 'The Nursing-Medical Boundary: A Negotiated Order?', Sociology of Health and Illness, vol. 19, no. 4, pp. 498 - 520

ANNANDALE, E. and CLARK, J. (1996) 'What is Gender? Feminist Theory and the Sociology of Human Reproduction', Sociology of Health and Illness, vol. 18, no. 1, pp.17 - 44

BALL, J. (1994) Reactions to Motherhood: The Role of Postnatal Care. Hale: Books for Midwives.

BALDOCK, J. (1997) 'Social Care in Old Age: More than a funding Problem', Social Policy and Administration, vol. 31, no. 1, pp. 73 - 89.

BERGER, P. L. and LUCKMANN, T. (1971) The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Harmondsworth: Penguin.

BLAXTER, M. and PATERSON, E. (1982) Mothers and Daughters: A Three-Generational Study of Health Attitudes and Behaviour. London: Heinemann.

BLOOR, M. and McINTOSH, J. (1990) 'Surveillance and Concealment: A Comparison of Techniques of Client Resistance in Therapeutic Communities and Health Visiting' in S. Cunningham-Burley and N.P. McKeganey (editors) Readings in Medical Sociology. London: Routledge.

BLUM, L. M. and VANDEWALTER, E. A. (1993) '"Mother to Mother": A Maternalist Organization in Late Capitalist America', Social Problems, vol. 40, no. 3, pp. 285 - 300.

BOWES, A. M. and DOMOKOS, T. M. (1996) Issues of Communication between General Practitioners, Health Visitors and Women of Pakistani Heritage. Final Report to Chief Scientist Office, Scottish Office Department of Health.

BOWES, A. M. and DOMOKOS, T. M. (1998) 'Health Visitors' Work in a Multi-Ethnic Society: A Qualitative Study of Social Exclusion', Journal of Social Policy, vol. 27, no. 4, (in press).

BOWLER, I. (1993) ' "They're not the same as us": Midwives' Stereotypes of South Asian Descent Maternity Patients', Sociology of Health and Illness, vol. 15, no. 2, pp. 157 - 177.

BRYANT, C. A. (1982) 'The Impact of Kin, Friend and Neighbor Networks on Infant Feeding Practices', Social Science and Medicine, vol. 16, pp. 1757 - 1765.

CAMPBELL, H. (1994) Breast-Feeding in Scotland. Glasgow: Scottish Forum for Public Health Medicine.

CARTER, P. (1995) Feminism, Breasts and Breast-Feeding. Basingstoke and London: MacMillan.

DELACUESTA, C. (1993) 'Fringe Work: Peripheral Work in Health Visiting', Sociology of Health and Illness, vol. 15, no. 5, pp. 665 - 682.

FILDES, V. (1986) Breasts, Bottles and Babies: A History of Infant Feeding. Edinburgh: Edinburgh University Press.

GLASGOW HEALTHY CITY PROJECT, (1993) The Glasgow Women's Health Book. Glasgow: Glasgow Healthy City Project.

GORDON, J. (1989) '"Choosing" to Breastfeed: Some Feminist Questions', Resources for Feminist Research, vol. 18, no. 2, pp. 10 - 12.

GREATER GLASGOW HEALTH BOARD, (1994) Child Health Statistics 1990-1993. Glasgow: Department of Public Health.

HASTRUP, K. (1992) 'A Question of Reason: Breast-Feeding Patterns in Seventeenth- and Eighteenth-Century Iceland' in Maher V (editor) The Anthropology of Breast-Feeding: Natural Law or Social Construct. Oxford: Berg.

HOLLAND, J. RAMAZANOGLU, C. SCOTT, S. SHARPE, S. and THOMSON, R. (1990) 'Don't Die of Ignorance' - I Nearly Died of Embarrassment: Condoms in Context. London: Tufnell Press.

JEFFREY, R. (1979) 'Normal Rubbish: Deviant Patients in Casualty Departments', Sociology of Health and Illness, vol.1, no. 1, pp. 90 - 107.

JONES, R. A. K. and BELSEY, E. M. (1977) 'Breast- Feeding in an Inner London Borough - A Study of Cultural Factors', Social Science and Medicine, vol. 11, pp. 175 - 179.

KHATIB-CHAHIDI, J. (1992) 'Milk Kinship in Shi'ite Islamic Iran' in V. Maher (editor) The Anthropology of Breast- Feeding: Natural Law or Social Construct? Oxford: Berg.

LITHELL, U-B. (1981) 'Breast-Feeding Habits and their Relation to Infant Mortality and Marital Fertility', Journal of Family History, vol. 6, no. 2, pp. 182 - 194.

MACLEAN, H. M. (1989) 'Women's Experience of Breast-Feeding: A Much Needed Perspective', Health Promotion, vol. 3, no. 4, pp. 361 - 370.

MAHER, V. (editor) (1992) The Anthropology of Breast-feeding: Natural Law or Social Construct Oxford: Berg.

MARTIN, E. (1989) The Woman in the Body: A Cultural Analysis of Reproduction. Milton Keynes: Open University Press.

MORSE, J. M. (1989) '"Euch, those are for your husband!" Examination of cultural values and assumptions associated with breast- feeding', Health Care for Women International, vol. 11, pp. 223 - 232.

MULL, D. S. (1992) 'Mothers' Milk and Pseudo- Scientific Breast-Milk Testing in Pakistan', Social Science and Medicine, vol. 34, no. 11, pp. 1277 - 1290.

PALMER, G. (1988) The Politics of Breastfeeding. London: Pandora.

RAPHAEL, D. and DAVIS, F. (1985) Only Mothers Know: Patterns of Infant Feeding in Traditional Cultures. Westport, Connecticut: Greenwood Press.

SALT, M. J. LAW, C. M. BULL, A. R. and OSMOND, C. (1994) Determinants of Breast-Feeding in Salisbury and Durham', Journal of Public Health Medicine, vol. 16, no. 3, pp. 291 - 295.

SCHUTZ, A. (1970) On Phenomenology and Social Relations. Chicago and London: University of Chicago Press.

STACEY, M. (1976) 'The Health Service Consumer: A Sociological Misconception' in Sociology of the National Health Service, Sociological Review Monograph 22. Keele: University of Keele.

STACEY, M. (1988) The Sociology of Health and Healing: A Textbook. London: Unwin Hyman.

STRAUSS, A. (1978) Negotiations: Varieties, Contexts, Processes and Social Order. San Francisco: Jossey Bass.

STRAUSS, A. SCHATZMAN, L. EHRLICH, D. BUCHER, R. and SABSHIN, M. (1973) 'The hospital and its negotiated order' in SALAMAN, G. and THOMPSON, K. People and Organisations. London: Longman.

SVENSSON, R. (1996) 'The Interplay between Doctors and Nurses - A Negotiated Order Perspective', Sociology of Health and Illness, vol. 18, no. 3, pp. 379 Ů 398.

THOMAS, M. and AVERY, V. (1997) Infant Feeding in Asian Families. London: Stationery Office, for the Office of National Statistics.

THOMAS, W.I. and THOMAS D.S. (1970) 'Situations Defined as Real are Real in their Consequences' in G.P. STONE and H. A. FABERMAN (editors) Social Psychology Through Symbolic Interaction. Waltham, Mass: Ginn Blaisdell.

TORKINGTON, N. P. K. (1995) 'Black Migrant Women and Health', Women's Studies International Forum, vol. 18, no. 2, pp. 153 - 158.

VAN ESTERICK, P. (1989) Motherpower and Infant Feeding. London: Zed.

WILSON, G. (1994) 'Co-Production and Self-Care: New Approaches to Managing Community Care Services for Older People', Social Policy and Administration, vol. 28, no. 3, pp. 236 - 250.

Copyright Sociological Research Online, 1998