These types of explanations go some way to understanding individual differences in self-reports and the discrepancy between self-report and objective tests. Are there other reasons for these, and why does research with women generally nd decits in cognition whereas studies with other 44 animals generally nd improvements? Striving to understand cognitive skills, particularly memory, through carefully controlled, objective tests, psychologists and other researchers seldom study the kind of memory failures that are regarded as important by lay people. Although attempts are made to make ndings relevant and applicable to everyday life, psychologists are still more likely to be found testing memory for lists of words than the ability to remember an antenatal appointment. The development of questionnaires such as the Cognitive Failure Questionnaire, and the use of cognitive diaries, was in part an attempt to tap the kind of cognitive impairment that has an impact of people's everyday lives. So perhaps tests of cognition such as explicit memory tests are not close enough to real tasks to reveal decits of which women are aware.

The belief that women experience cognitive impairment during pregnancy has a real effect on women's lives. As we and others have shown, such beliefs affect the way that women are treated by employers and fellow workers (Pattison and Gross, 1996; Pattison, Gross and Cast, 1997).

Jackson et al. (1996) found that up to 30 per cent of registered midwives have negative att.i.tudes to women's ability to learn during pregnancy.

Several of the studies we review here found no evidence of cognitive decits during pregnancy. The following quotations are taken from two such studies: Pregnant women and new mothers generally should be condent of performing to their normal cognitive capabilities, but may be more affected than usual by a high cognitive load.

(Casey et al., 1999: 158) It is argued that the cognitive efciency of workers is not compromised by pregnancy but steps should be taken to ensure that work load is adjusted to take account of the self-reported reduced arousal that may arise.

(Morris et al., 1998: 377) As the above quotations ill.u.s.trate, even when researchers nd no evidence of cognitive decits, their conclusions or recommendations may suggest that women should a.s.sume that they have. The strength of the expectation of an a.s.sociation between pregnancy and cognitive debilitation seems to override research ndings to the contrary, and leads the researchers quoted here to undermine their own research.

4.BEING PREGNANT AT WORK.

Pregnancy and employment.

Work is one of the predominant components of everybody's daily life.

Employment is not only a means of achieving economic security or independence; it has psychological importance. At its best, work can provide opportunities for self-fullment, self-ident.i.ty, creativity, social engagement and the success of shared goals, skills or activities. At its worst, the money helps. Statistics show that the majority of women are in employment and women of childbearing age are therefore likely to be working when they become pregnant for the rst time (Eurostat, 2004). Immediately, however, the conjunction of pregnancy and work (by which we mean paid employment) invokes a variety of potential discourses concerning the role of women in the public domain. Thus, the topic of work, and the research on women, work and pregnancy, refer directly to the debates we have alluded to in the discussions of the medicalisation of pregnancy and childbirth and the normalisation of pregnancy through the predominance of the biomedical tradition. Furthermore, the issues of cognition and performance we discussed in the previous chapter can be regarded as integral to perceptions of women when they are pregnant in the workplace.

To examine these issues, we review some of the work on the a.s.sociations identied between pregnancy, work and health, particularly the health of the baby, and we examine aspects of women's experiences at work when they are pregnant, with a view to explaining how pregnancy is perceived in the workplace. In so doing, we will raise a number of issues which characterise the discourses of pregnancy and pregnancy research and which are revisited through the book. First we look at the impact of pregnancy on the workplace by reviewing evidence describing women's experiences of announcing their pregnancy and being pregnant at work. We go on to explore how these experiences might be explained. We then consider whether or how employment affects pregnancy, and the relationship between pregnancy, employment and pregnancy or birth outcome. Finally, we examine the products of research described, in terms of the advice and information about working that women may encounter and their behaviour in response to it. In the process of looking at how pregnancy affects 46 work, we make use of a range of research arising from a perceived need to address social policy and employment practices as well as psychological or health concerns addressed through academic studies of the topic. In considering the way that work may affect pregnancy, we draw on the epidemiological and physiological literature on factors inuencing pregnancy outcome as well as that examining the relationship between work and health.

Information on working during pregnancy comes from qualitative surveys of women and their employers investigating the implementation and experience of maternity rights, such as those by Rodmell and Smart (1982), O'Grady and Wakeeld (1989), McRae (1991, 1996) and a report from the National a.s.sociation of Citizens Advice Bureaux (NACAB, 1992). There are some experimental studies of pregnancy and employment, including our own (Corse, 1990; Halpert et al., 1993; Pattison et al., 1997), together with some more anecdotal work, for example Baildam (1991). Popular women's magazines, and publications aimed at pregnant women, also address these issues (Gross and Pattison, 2001). In other work of our own we have interviewed pregnant women about their experience at work and we will give some examples from these interviews. The largely quant.i.tative literature on outcomes is extensive and could form the substance of an entire book; only a sample of this is discussed here. There are a number of other texts that address similar issues from different perspectives, which are not reviewed here but provide fascinating additional material on the tricky work/life balance that has come to represent modern working life in the twenty-rst century (Brannen and Moss, 1988; Devlin, 1995; Humphries and Gordon, 1993). Our particular interest is on the period of pregnancy itself rather than on the related concerns of managing childcare and returning to work. These are signicant concerns in their own right which naturally follow from pregnancy but are not the focus of this book. Nevertheless, it is likely that pregnancy irretrievably affects women's relationship with paid employment.

Announcing a pregnancy at work Given women's current employment partic.i.p.ation rates, it might be a.s.sumed that pregnancy would be unremarkable in the workplace. But this is not necessarily the case: in 2003, a news item covered the outcome of an industrial tribunal that had found in favour of a lawyer who had been dismissed from her job while pregnant (BBC Radio News, 2003).

McGlynn's (1996) concern appeared to hold true: advances so far gained in relation to pregnancy dismissals do not yet reect a cultural shift in att.i.tudes and consequently vigilance and continued campaigning is required to improve the real situation of women in the workplace' (McGlynn, op.

cit.: 229). Recent statistics conrm this (Dunstan, 2002; EOC, 2005; James, 47 2004). Of the 440,000 working women who become pregnant in Great Britain each year, almost half can expect to experience some form of disadvantage at work and as many as 30,000 will be forced out of their jobs while they are pregnant. Such gures clearly demonstrate that for a signicant number of women being pregnant at work is seen as unacceptable in some way, prompting commentators to liken such treatment to outdated Victorian values rather than suited to the twenty-rst century (The Guardian, 2 May 2003). Women's experiences of responses to pregnancy announcements catalogued by research ndings show that these can take several forms.

Rodmell and Smart (1982) interviewed 30 pregnant women at work in London about their experiences. The women represented a range of non-professional work where women were and still are commonly employed: as carers, in catering or in clerical jobs. Although women found work physically tiring, most felt able to cope well and said that they preferred to be at work rather than sitting at home. This was despite the fact that when asked how their managers had responded to the news of their pregnancy, the women indicated that att.i.tudes had varied from positive, through indifferent, to some cases where they were described as contemptuous. The women also talked about the responses from their direct colleagues; this was reported as more positive from female colleagues than male ones, with women saying that by and large their female colleagues were really pleased for them. In contrast, several women reported that men used the fact that pregnancy was not an illness but a normal event as an excuse not to be helpful or even pleasant, and were resentful that the women were going to be paid to take time off. Interestingly, some partic.i.p.ants were sympathetic to their colleagues' feelings of resentment and acknowledged the difculties that follow for all workers when someone is going to be off work for a length of time.

In their survey published seven years later, O'Grady and Wakeeld (1989) highlight similar ndings. Their survey involved 250 women and they summarise the responses of managers and colleagues to the news of a woman's pregnancy as generally positive. Women also reported a neutral response to the news from their boss as a good response, perhaps suggesting that negative comments are so obviously unpleasant that anything less than overt hostility is seen as positive. Overall, these women too reported that their female colleagues were more interested in their pregnancy announcement than their male colleagues. However, as in Rodmell and Smart's (1982) study, men's negative comments and responses often centred around the rights of pregnant women to pay and particularly to leave, on the grounds that maternity pay was a woman's benet not available to them and that maternity leave was seen as basically a paid holiday. Some older women were resentful that they had not had the chance to benet from newer maternity provisions themselves. Hence, it is not only the pregnancy 48 itself but also the consequences of the pregnancy for the rest of the employees that may create tensions in a workplace, and both their own current and future state have to be managed by pregnant women while they are working.

In Europe there are formal health and safety requirements, involving risk a.s.sessments, to move women to less dangerous working environments when they are pregnant. There are also inst.i.tutional policies with regard to pregnancy and alterations to working conditions, as well as rights for pay and reinstatement following leave. Rodmell and Smart's (1982) ndings suggest that arrangements to reorganise work or to change jobs to take account of more risky activities, for example using heavier cleaning equipment, or to cover during maternity leave, were frequently the result of informal arrangements with fellow workers rather than with their managers. Informal arrangements can break down through no fault of the individuals concerned and any arrangements can mean, of course, that women are not necessarily taken out of dangerous or physically demanding jobs, even when they should be. A further problem with informal arrangements was that the women concerned felt that their colleagues were doing them favours and this in itself caused the women anxieties about not letting fellow workers down, not being fair to them or getting them into trouble. Even when formal arrangements are made in an organisation, they are not always helpful. The relationship between the rhetoric of the legal requirements and the practice is exemplied by the case of a nurse, considered to be at risk working within a radiology department when pregnant, who was moved to a job involving anaesthetic gases, even though these carry as great a risk as radiation. It is also the case that changes in the employing organisation can effectively be used to ease women out of jobs while on maternity leave. Though this is against the law, there are many cases in literature where employers' responses have been aggressive and discriminatory, seeking to prevent women from working long enough to claim their rights to maternity benets or to create grounds for dismissal. For example, 'When the client discovered she was pregnant, her employer indicated his ''moral outrage'' at her unmarried state . . . and told her that he will reduce her salary by 40% on the grounds that she is not doing her job satisfactorily' (NACAB, 1992: 9).

Taking such an unhelpful line in response to a pregnancy announcement is not the only way that the experience at work can be affected. The women in Rodmell and Smart's (1982) survey said that personal comments increased: comments about their appearance or their size that would not otherwise have been made. mean, I got a lot of comments about my b.u.m or my body looking like some revolting blob' (op. cit.: 104). They also had to cope with some people's embarra.s.sment, particularly younger people. In O'Grady and Wakeeld's (1989) survey, they reported that men who were not directly unpleasant nevertheless behaved in ways that would normally 49 be considered unacceptable or even as hara.s.sment, such as patting a woman's stomach and making comments on fertility, on size and shape.

Comments on appearance are a common nding, even a ground for dismissal, as in the case of a care attendant dismissed by her proprietor who told her t isn't very nice having someone with a big belly working here'

(NACAB, 1992: 9). O'Grady and Wakeeld (1989) showed that marital status was another factor that affected how women were treated at work.

Like the client in the quote above, some single women reported that their colleagues wereshocked' that they were pregnant and not planning to marry: all my difculties stemmed from being unmarried but my pregnancy was deliberate. I was expected to marry straightaway. In fact I only returned to my job through lying and saying I am now married' (ibid.: 7). In our own interviews with pregnant women there were similar comments about marital status, especially addressed to the youngest women: 'people at work said I was too young' (Kirsty, age 17). The most recent Equal Opportunities Commission (EOC) survey (EOC, 2005) found that disabled women who became pregnant were dismayed to be asked questions that colleagues had no right to ask, such as how they became pregnant and whether they should have.

Rodmell and Smart (1982) summarise their ndings as demonstrating that the responses of co-workers and management to the presence of a pregnant woman in the workplace were experienced by the women in the study as implicitly dismissive, even where ridicule was not intended. In order to explain away what they obviously found upsetting or discon-certing responses from people they regarded as friends or colleagues, some women resorted to blaming themselves. They suggested that they were being oversensitive and therefore reacting to situations in an atypical way.

Indeed, in an attempt to distance themselves further from the personal criticism they felt they had attracted, they accounted for their own sensitivities as the effect of their hormones. Pregnancy can hardly be considered routine in the workplace, when women have to negotiate this complex positioning of both their colleagues and themselves.

Women's experiences at work when they are pregnant are not solely the result of personal interactions. In a study of 2,250 pregnant women, Cherry (1987) examined the physical demands of work during the last three months of pregnancy. She found that 20 per cent of the women studied were required to stand for four hours or more every day, some with their back bent over work or handling loads with a twisted position. In most cases women could take a break when they felt it necessary but 8 per cent (still nearly 200 women) had to sit or stand for two or more hours at a stretch. The incidence of fatigue, varicose veins, back problems and breathlessness were all found to be related to general or specic job demands like these. Very few women in Rodmell and Smart's survey of pregnant women had regular and permitted access to rest facilities; many 50 were not allowed to sit down during the working day, and this refusal has been the basis of employment tribunal cases for constructive dismissal over a number of years. Women in the 2005 EOC survey also reported that although they were told they were allowed to sit down regularly by the health and safety a.s.sessor, managers overrode this permission because by sitting down 'they didn't look busy'. Furthermore, for some women who were expected to wear a uniform or protective clothing at work, there was not always an alternative version available for pregnant women. Indeed, many workplaces, workstations and much equipment and protective clothing are adaptable for overweight people but not for pregnant women.

Not having appropriate work clothes made the women feel self-conscious and drew attention to their state, making comments more likely.

The emphasis so far has been on negative experiences, since it is these that give such cause for concern. However, it is important not to automatically a.s.sume that all women will be treated the same way, since all the studies mentioned also found that women had had a positive experience, with colleagues and managers being extremely supportive and thoughtful.

A small survey of maternity policy in practice in Eire found that during their pregnancy women were treated well; the difculties arose when they returned to work (Brady et al., 1999). Importantly, however, the dif- culties that women encounter in the workplace do not necessarily detract from the benets that women gain in working. One woman in Rodmell and Smart's (1982) survey commented that she found her work easier, because she felt 'centred' and 'calm'. The emotional high of pregnancy can thus carry women through the more negative aspects of their working experiences during this 'special' time.

Nonetheless, an extremely important issue raised by women in several of the studies was that they no longer felt that they had the same employment status as they had done before they were pregnant (and this has implications for return to work too). Women found themselves excluded from training courses or promotion-linked activities, and this they felt was 'not exactly direct discrimination' but a form of exclusion. One woman quoted in the summary report of the EOC survey (Greater Expectations, EOC, 2005: 2) said that she felt treated like an utsider' the minute that she found out she was pregnant. O'Grady and Wakeeld's (1989) earlier study reported very similar comments together with some incidents of demotion or problems with promotion. There seemed to be an expectation that, once pregnant, women would no longer be as interested in work as before, and would probably not return to their job after maternity leave. In some cases this was taken to the extreme of appointing a permanent replacement before any nal decision had been made about a woman's return to work following her leave. Kate Figes, in her book Because of Her s.e.x, also gives examples of harsh treatment, for example the dismissal of a woman, who had taken a week's annual leave following the announcement 51 of her pregnancy, because the employer did not like the woman's att.i.tude (Figes, 1994).

It is usually the case that by looking at a series of ndings over time, it is possible to demonstrate a change for the better. This is not so in the treatment of pregnant women at work. Depressingly, the experiences described by Rodmell and Smart's study published in 1982 and the EOC surveys of 2005 are almost interchangeable, whether this is the negative responses to pregnancy announcement, the feelings of exclusion and lack of opportunities or the intrusive personal comments. While many women do report a positive experience, given the extent of possible negative treatment, it is perhaps not entirely unexpected that pregnancy also provides some women with an opportunity to give up work that they have disliked (Harris and Campbell, 1999). The consistency of the comments and ndings over time suggests that there is nevertheless a persisting resistance to pregnancy in the workplace.

The frequency of pregnancy discrimination highlighted by these studies draws attention to the different ways in which pregnancy continues to be represented in different domains, as we alluded to in Chapter 2. As we have discussed, pregnancy can be seen as a normal event that is treated in many ways as an abnormal one, and women are treated accordingly. The explanations appear to draw on not only social and cultural norms and expectations of women and of women when they are pregnant but to incorporate knowledge which arises in part from research on pregnancy.

Such research, in turn, depending on its rationale or theoretical position, both colludes with and challenges these representations. As with the other topics we are addressing in this book, the area of pregnancy and employment is a good example of how conicting and confusing evidence can be and how entrenched att.i.tudes and beliefs about pregnancy remain.

Why is pregnancy still so stigmatised within the workplace?

There are a number of explanations that might provide pointers to the way that pregnancy is situated within the public perception and why it continues to provoke such negative responses in the workplace. As the evidence suggests, treatment may reect current socio-political and cultural att.i.tudes to women in the workplace more generally. Discourses of femininity brought into play by the crossing of private and public boundaries through the visibility of pregnancy may present difculties for women wishing to maintain their role as worker, and for employers and colleagues (Gross and Pattison, 2001). Alternatively, the persistence of negative treatment may result from the cultural perception of pregnancy as an ambiguous health state, falling between a natural healthy experience and a health state requiring medical attention. This is emphasised by the legislation which allows time off for visits to hospital and clinics, as well as maternity leave 52 itself. The a.s.sociation of pregnancy with sickness can lead to problems (Hanlon, 1995), since the ideology surrounding pregnancy as a healthy or sick state may affect women's own views of themselves as competent workers or as responsible mothers and their colleagues' expectations of pregnant women's performance (Pattison et al., 1997). The various explanations proposed are not mutually exclusive but represent a means by which the complexity of the issues, which are not exclusive to employment, can be explored.

Private life/public activity boundaries The rst of these concerns the overlap between the private and the public.

At a fundamental level, pregnancy very obviously brings into the workplace aspects of a person's private life that are normally restricted to conversations between close friends or family (though they may create the subject for jokes and gossip at work). The working environment comprises a mix of managers, subordinates, colleagues and friends, all of whom may have different responses to this visible transgression of privacy. The visibility of pregnancy as the representation of reproductive fertility and of s.e.xuality can transgress boundaries, here between home and work, which may normally be carefully policed. For example, workers may wish to display their commitment and adherence to the hegemonic male model of work.

An interesting anthropological study of a North American working environment, referred to as The Laboratory, demonstrates the way that home and work boundaries can be created and maintained by different types of workers and artefacts (Nippert-Eng, 1996). Nippert-Eng points to the variety of ways that such boundaries are represented, for example by the restriction of personal telephone calls, the availability of places to put personal effects and the degree of freedom to carry work equipment between home and workplace, as well as the integration of domestic and work items, such as keys or addresses, onto combined keyrings or home address les, and socialising with work colleagues out of work time. She describes both individuals and workplaces as integrative or segmenting, that is, those that encourage or allow a crossing over of the private into the public arena and those that do not. In this context she says: t is hard to imagine anything more fundamentally, undeniably integrating than a visibly pregnant worker. A pregnancy is a powerful souvenir of home life.

It brings the very essence of home into the workplace in its most sacred form' (Nippert-Eng, 1996: 213).

In an integrative workplace, pregnancy may enhance a family-friendly environment and Nippert-Eng describes the experience of three women who became pregnant around the same time within The Laboratory's personnel section. For them, pregnancy was a highly positive experience, 53 with the women sharing notes and discussing their progress with each other and other members of their group. In addition, they became more friendly outside work than they had been before their pregnancies. However, signicantly, such harmony was not the case elsewhere in the same organisation and by contrast t [pregnancy] seriously, undeniably challenges . . .

more segmentist groups. More than anything else, the varying treatment of pregnant women in the workplace shows that ''pollution'' is in the eye of the beholder' (op. cit.: 214). In clarifying this position, Nippert-Eng cites the case of a worker who found that on announcing her pregnancy not a single departmental member commented on it throughout the ante- and postnatal period and most avoided her altogether. This extreme response, which might in other circ.u.mstances be described as discrimination, chimes with reports in the EOC study (2005) of the ways that women were treated, which included not speaking to them and ignoring their presence.

Clearly, pregnancy can really create problems in maintaining boundaries for others and calls into question the way that women are perceived and how they are able to function in their work roles when at work.

One reason that this public/private distinction is so powerful may be because of continuing strongly held beliefs that it is a woman's primary duty to care for her family. Working during pregnancy may be seen as simply inconsistent with that belief. Furthermore, the boundary shifts between the personal/private and the workplace, and the a.s.sociated links with taboo images of women, can mean that it is seen as inappropriate, offensive or embarra.s.sing to have pregnant women in the workplace at all.

In part this may arise from the physical changes in body size and weight in pregnancy, which become increasingly visible as the pregnancy progresses and which may be accentuated through current fashions in maternity clothing (see also Chapter 7). This is borne out by evidence from the surveys described above, whereby women reported that they had had to change their role from receptionist to ofce work behind the scenes, and of having to give up working at the counter in a bank because of the potential embarra.s.sment to customers and fellow workers. The physical changes in women's appearance can provoke a range of responses from employers and colleagues, which may affect women's experiences of working at this time, and lead to the perpetration of discriminatory behaviours by employers and colleagues.

Workplace cultures An alternative approach to explain the ndings relates to the culture of organisations. The corollary of falling birthrates is that pregnancy is less common in the workplace than it used to be. Faludi's (1992) discussion of the nfertility epidemic' among 30-something middle-cla.s.s career women suggests that the proportion of working women becoming pregnant may be 54 very small in some types of workplace. There is some evidence from our own study of the perceptions and beliefs of fellow workers (Pattison et al., 1997) that negative att.i.tudes are more likely to be expressed by those who have no direct experience of working with or alongside pregnant women and that there are gender differences in att.i.tudes. The ndings do suggest that women are generally more positive than men about women's rights at work and pregnant women in the workplace. Even in these circ.u.mstances, however, if pregnancy is uncommon then the likelihood of poor treatment may be greater.

The relevance of the workplace culture is ill.u.s.trated by ndings reinforced by the EOC survey (2005) that the women most affected by discriminatory acts at work were those in lower-paid, less secure jobs, in routine occupations such as sales or customer services. More importantly, they very often worked in smaller workplaces (less than ten employees) but this does not prevent individuals in large organisations also feeling dismayed by their treatment or under pressure to perform once they have announced their pregnancy, and professional women are by no means exempt from discrimination. In Rodmell and Smart's (1982) survey there appeared to be no difference between the type of employer and the nature of their response, though the number of women in the study and therefore of employers was small. McRae (1991) suggests that there are differences whereby employers in the public sector tend to be better informed about maternity policy and rights, which have in the past caused problems for some employers, for example replacement of staff while on maternity leave or holding jobs open (Callender et al., 1997). Other factors contribute to poor experiences, as research shows that refusals from employers to grant favourable arrangements during pregnancy were actually more frequent when working conditions were tiring and that taking sick leave was more common when working conditions were hard and job adjustment was not made by employers to accommodate pregnant workers (Saurel-Cubizolles et al., 1991; Strand et al., 1997).

Employers' att.i.tudes are by their own admission in part driven by the need to balance the requirements of legislation and their costs with running a compet.i.tive business: f you have a list of people with equal merits, I might be more inclined to employ a bloke as he is not going to get pregnant' (EOC, 2004: 22). This is despite ndings that where employers and organisations provide a supportive environment, they appear to take their women workers seriously and regard good support as a sound business investment in their workforce. This is especially important given that women who feel badly treated during their pregnancy are less likely to return to work and those who feel supported are more likely to return to work (Killien, 2005; Lyness et al., 1999). Some very small businesses have trained every member of staff to do another person's job so that one person's absence does not cause disruption to the business. Evidently, it is 55 possible to create a supportive culture and it makes economic sense to do so. It requires a willingness to see women as equal partners in the workplace and pregnancy as a normal and manageable event.

The difculties highlighted by workplace culture may reside in the gap between policy and practice. The danger would seem to lie where women lose the condence of their manager or colleagues if they become pregnant.

In an organisation where a time limit operates on work completion, so that a project ends after the individual running it is on maternity leave, it is not difcult to see how misunderstandings can arise. The resolution of such issues is undoubtedly one that causes problems even in the most positive working environments and means that even within them the experience of pregnancy is less than ideal. Figes (1994) suggests that because of the hostility expressed in the context of work, women are likely to become secretive and guilt-ridden about what should be a signicant and positive life event. However, as we have seen, this may be more prevalent in certain types of organisations or sections of organisations than others. The integrating environment is able to absorb the violation of the boundaries whereas the segmenting one nds difculty in adjusting to the visibility of such an intrusion.

Reinforcing stereotypes An additional explanation for the treatment of women once they become pregnant could be straightforward stereotyping. Not only might the publicly held beliefs about a woman's public role as family carer impinge on the perception of her suitability in the workplace, but also on her recently acquired status as a worker. By this we mean that the working pregnant woman seems to provide for some colleagues and employers the opportunity to compound negative att.i.tudes held more generally about working women. These att.i.tudes typically characterise female workers as suited only to particular types of work (nurturing roles), as being interested in jobs rather than having careers, as unreliable workers because they have more time off than men to have children or look after children and as displaying certain types of behaviour, such as showing more emotion, being more conciliatory and submissive at work (e.g. Rosen and Jerdee, 1978; Wilson, 1995). While we can dismiss these stereotypes, pregnancy remains as a visible marker for the perceived lack of commitment: that women are not, and will not be, as committed to their jobs as they were before their pregnancy, in particular that they are less hard-working, (Bistline, 1985) integral to the stereotype. Added to this, the pregnancy stereotype itself also carries with it some a.s.sumptions about women's behaviour (see Chapter 2), in particular that they are more emotional and that their brains shrink, and hence may be less effective at work. These myths are supported not least by pregnant women themselves and by 56 anecdotal reports. However, studies investigating cognitive processing, in particular cognitive failure, do not generally endorse this, as we have already discussed in Chapter 3. A study by Nicholls and Grieve (1992), investigating the relationship between performance and posture, found that although women typists in the third trimester felt that their physical discomfort at work was affecting their work efciency, the measures of their performance, in terms of speed and accuracy, showed this was not the case. Even in the face of contrary evidence, pregnancy is aligned with the female stereotype, making it even more difcult for women to be regarded as effective workers and increasing the expectations of others that women will behave in a stereotypically feminine way (Taylor and Langer, 1977). If businesses are using economic arguments to justify their att.i.tudes and behaviour, investing in women's skills and careers of women who are regarded as likely to leave is certainly bad economics and adherence to such a stereotypical representation allows unfair treatment to be maintained. The signicance of the stereotype in determining responses is examined by some experimental work by Halpert et al. (1993) and by Corse (1990).

In some of the limited psychological research on this topic, Halpert and her colleagues (1993) used an att.i.tude questionnaire which covered issues of pregnant women as employees, the treatment of pregnant employees, emotional stereotyping, physical limitations of pregnancy and choosing career or family. The ndings of this part of their study were that, particularly among the men who responded, there was substantial negative stereotyping in att.i.tudes to pregnant employees. A second part of the study asked the partic.i.p.ants to view a videotape of a pregnant or a non-pregnant woman carrying out a.s.sessment centre tasks and to rate her performance.

The pregnant employee was consistently rated lower compared to the nonpregnant employee, and the a.n.a.lysis of the interaction showed that men a.s.signed lower ratings than women and were more negatively affected by the pregnancy condition. There are some problems with this study, the main one being that the people doing the rating were undergraduate students. But, a worrying aspect of Halpert et al.'s ndings is the implication for appraisal of pregnant women's performance at work and the role of young male managers in this activity. Bragger and colleagues (2002) found a bias against pregnant job applicants and, as recommended in any strategy to minimise discrimination, suggest that it is possible to reduce the impact of negative bias against certain groups of applicants for posts by using structured interviewing and attention to equal opportunities policies.

Corse (1990) used postgraduate MBA students in a role play exercise with two women managers, one of whom was apparently pregnant and the other who was not. Partic.i.p.ants had more negative impressions of and lower satisfaction with the pregnant manager than with the non-pregnant 57 manager. Interestingly, they (the subordinates) initiated more conversation with the pregnant manager than with the non-pregnant manager, that is, they were more social in their comments. The data from the interviews with the partic.i.p.ants suggest that the reasons for the negative impressions of the pregnant manager were because they had expected her to bepa.s.sive, nice and giving' and were surprised by her authoritative behaviour. The author relates these ndings to work gender expectations and s.e.x-stereotyping, for example Taylor and Langer's (1977) nding that people like pregnant women better when they behave pa.s.sively and Butensky's (1984) nding that they are evaluated more favourably when they occupy a stereotypically feminine rather than masculine work role. However, the issue remains a complex one.

In attempting to explain why pregnant managers are viewed so negatively, Corse (op. cit.) suggests that they may evoke images of the Great Mother, one of the three archetypes of femininity (Neumann, 1955), the other two being the Good Mother, representing our culture's view of desirable femininity (and which we also discuss in its various forms in Chapter 7), and the Terrible Mother, who is devouring and aggressive. The Great Mother combines aspects of both the Good and Terrible Mother, having creative and destructive powers. The female stereotype conveyed by pregnancy combines these elemental aspects of femininity and the challenge it presents, via the Terrible Mother, to masculinity. Thus, the image of the pregnant manager creates discomfort, confusion and hostility in subordinates whose cultural stereotypes and feelings of vulnerability lead them to want her to behave like the Good Mother nurturing, giving and caring.

Pregnancy is uniquely a time when women can demonstrate visibly the evidence of their creative power; and coupled with an inst.i.tutional sanction of power in the managerial context, the image evoked may be too threatening. One way of dealing with this is for subordinates to reframe their position by using social contact rather than work contact. Corse suggests that there may be dangers in the relationships between a pregnant manager and subordinates or other workers who may limit their style or content of communications so that pregnant managers may nd themselves excluded from certain types of information, something that has been raised by a number of women interviewed in studies. Corse raises, too, the general issue of diversity in the workplace, of which pregnancy is one aspect, and which cannot just be ignored or regarded as a female-only issue.

Pregnancy as sick role Finally, as the surveys suggested, an explanation for negative att.i.tudes may stem from the linking of pregnancy with illness, something which has been strongly resisted but in which some forms of legislation have been complicit.

58.In our study of att.i.tudes (Gross et al., 1997), the most negative responses were given in the physical limitations category and the emotional stereotypes category, and these give credence to the idea that pregnant women are perceived as invalids, conforming to the most extreme feminine aspects of the female stereotype, and are not capable either emotionally or physically of fullling the demands of their employment. This aspect of pregnancy is addressed at some length in a paper by Taylor Myers and Grasmick (1990) in which they highlight the social rights and responsibilities of pregnant women (the t.i.tle of their paper). Their case is that because of the anomalous position of pregnancy as a healthy and as an illness state, the beliefs held by the general population have the effect of ascribing the sick role (described by Parsons, 1951) to pregnancy. While the sick role is presented as a set of rights and responsibilities for individuals, rather than the equation of pregnancy with illness in the legal context, it is their contention that the adoption of this role actually prevents women being accorded adequate healthcare during pregnancy, and particularly those women who are least able to access such care. This is because they are held responsible for their own condition and it is their duty to manage it. Taylor Myers and Grasmick are concerned about publicly held beliefs, which their study conrmed, and access to healthcare in the US, not specically with employment, though of course many of the same issues arise. Clearly, the issue of pregnancy as an illness has reper-cussions for both social and legal treatment (Tavris, 1992) and Hanlon (1995) explores the notion of pregnancy as sickness in relation to pregnancy discrimination in employment and considers the way that pregnancy has been treated in law.

Hanlon, like Tavris, argues that, ironically, attempts to establish equality at work had actually reinforced stereotypes of men's and women's work and created further distinctions rather than equality. His view is that pregnancy discrimination has its roots as long ago as the Industrial Revo-lution, and represents a paternalistic desire to protect women from heavy jobs in the emerging coal and steel industries. This initial notion of protection is now one of the two ways in which pregnancy has been treated in law: the comparative (same/different) or the protected status approach.

The comparative approach requires that where situations are alike they should be treated alike but that where they are different they should be treated differently. Leaving aside the extensive room for debate about sameness and difference that this implies in relation to denitions and criteria, this of course fails to recognise that some inequalities arise from gender differences and that gender is about power relations and the persistence of structural inequalities. So previously, the comparator for a pregnant woman was a sick man, thus ensuring that the same structural inequalities were reinforced. Moreover, a.n.a.logies with the treatment of sick men are misleading because pregnancy is a healthy state (Rubenstein, 59 1992). Rulings in the appeal courts and the European Court of Justice challenged this approach, and the concept of 'protected status' effectively a gender-specic approach has taken over. Its strength is that women have uniquely to be protected in order to prevent them being discriminated against in the labour market and it does not require them to be compared with men. While Hanlon is concerned that the protected status approach is not entirely free of gendered stereotypes, for example that women are responsible for parenting, he considers that having treated pregnancy as a form of sickness is tantamount to the dismissal of women's role in society.

However, the provision of this gender-specic status is not without problems when women nd themselves pregnant in the workplace. In this sense, therefore, the advances in the law and the protection for women have not necessarily overcome the persistent willingness to ascribe the sick role to pregnancy as described by Taylor Myers and Grasmick (1990). In terms of women's experiences at work, it is perhaps inevitable that the notion of protected status enshrined in law as maternity rights to time off and leave can create resentment among fellow workers. It is seen as unfair treatment, although as the legislative approaches had previously conceded, employees are ent.i.tled to sick leave, but it remains the case that a colleague who has time off because of a broken leg is regarded quite differently from a woman having time off to have a baby. How leave or time off is perceived is very much in the eye of the beholder. As Goode and Bagilhole (1998) point out in their study of university academics and gender, men who leave work early to pick up children are viewed positively while women are not. Similarly, going to test drive a car or have a game of golf may be considered acceptable or even enviable but staying at home to look after sick children is not. A pregnant woman has the responsibility of remaining a good employee at the same time as being given time off when they are healthy. Not only is this an issue of pregnancy but also of women having responsibility for children, which contributes to the perception of women as unreliable workers.

Regardless of the legal position, the equation of pregnancy with sickness is still prevalent: 'The problem is the archaic perspective of what pregnant women can do in the workplace. They see it as an illness. It wouldn't have affected the way I do my job, but I wasn't given the chance' (EOC, 2005: 2).

As we have discussed, responses in the workplace are likely to reect the culture of that workplace as well as individual internalisations of public att.i.tudes and reference to female stereotypes. Employers' responses to pregnancy may be related to beliefs about women's role at work, coupled with the beliefs of women and their fellow workers about the competence and commitment of pregnant workers. Effectively, therefore, when in the workplace pregnant women are in a double bind, if pregnancy is a normal event then they should not receive special treatment, which is regarded as unfair; if it is an illness then they should not be at work and the negative treatment is justied and can persist.

60.In summary, therefore, whether women's experience can be accounted for by the prevalence of stereotypes and the expectations of the sick role, or whether the violation of the boundaries between home and work is too explicit in pregnancy, we can conclude that the exclusively female demands of pregnancy appear to exacerbate conicts around these issues concerning the role of women and the relationships between work and home. Pregnancy could be said to provide a focus for all these concerns that at other times may remain unexpressed. The very visibility of a pregnancy makes the woman the target of resentments una.s.sociated with the specics of pregnancy itself. At this point, having looked at what pregnancy seems to accomplish in the workplace, we move on to look at the other side of the relationship between pregnancy and employment, whether and how work, and their treatment at work, affects women and their pregnancy.

The effects of discrimination: work and wellbeing We have seen that there are a number of possible ways of accounting for the negative att.i.tudes expressed towards pregnant women in the workplace. Whatever the merits of these various accounts for negative treatment, the fact of its existence has the corollary that it impacts on women's experience at work, for example making it less likely that women will return following maternity leave. As we have already discussed, extreme treatment might have a negative effect on women's future employment and also on their psychological wellbeing. The impact of such treatment and anxieties may not be psychologically trivial, as a quote from one of the women surveyed by the EOC indicates: t's appalling that someone I had worked with for so long had a good relationship with and had no complaints from should treat me like this. It was extremely traumatic. I had stress related hospital treatment' (EOC, 2005: 2). The impact of this treatment is undoubtedly compounded by anxieties about fairness, or anxieties about working conditions or the health of their baby.

The complexity of the social and political dimensions of the issue of pregnancy and employment is highlighted by Karen Messing (1999), who points to the way that women have been excluded from examination of reproductive hazards in the workplace in favour of the foetus, even by feminist-oriented research concerned to place pregnancy as a normal event and yet to identify reasons for increasing rights at work. Messing suggests that legal pressure is necessary to protect women's employment and their health at the same time. The possibility that work and aspects of women's experiences at work might cause stress-related health problems raises a related issue of how women respond to the demands of their employment during pregnancy and how they are able to combine advice and information about the progress and management of pregnancy within these demands. There is a long tradition of research investigating the impact of 61 employment on pregnancy and on pregnancy outcome specically, which pays less attention to women's experiences and the value of employment. A further issue that we have already raised concerns the research methods used in examining the impact of work on physical and psychological wellbeing as well as on pregnancy outcome. Looking at women remaining in work during their pregnancies may be sampling from a restricted population (see Chapter 2; Bramwell, 1997). This makes any conclusions about the impact of work in general without reference to other aspects of women's experience inconclusive and confusing.

Some evidence that pregnant women nd life to be more stressful during pregnancy comes from a study by Nicholls and Grieve (1992). They found that just under 50 per cent of 200 women asked directly whether they found life to be more stressful during pregnancy reported that they did and that most of this stress came from work rather than from home, though it is important not to exclude the possibility of some response bias. In particular, however, and in keeping with the ndings we have outlined above, work-related stress was a.s.sociated with negative att.i.tudes of both supervisors and colleagues, a lack of exibility in the working schedule and pressures of time. Of course, it is not necessarily the case that the presence of a stressor equates with feelings of stress (Dewe et al., 1993) or poor health outcomes. Care must be taken in interpreting ndings in this area, especially since many of the studies looking at potential stressors in the workplace have as the primary focus their impact on the baby, or birth outcome measures, rather than women's concerns about working during pregnancy. Also, the attention focused on women working during pregnancy and the outcome of pregnancy emphasises the rhetoric of risk; thus, employment is presented in research terms as a potential source of risk, as dangerous. The results of many large epidemiological and cohort studies are thus represented within the context of danger avoidance or protection.

It goes without saying that no one would advocate that women be required to work in environments which could cause direct harm either to themselves or to their baby and health and safety guidelines do exist to provide protection for pregnant women. However, the role of work in women's lives, as well as their work role, is rarely accorded the same level of attention as those potential risks.

As we shall see from our discussions below, all available evidence over the past 30 years has indicated that women in stressful occupations or carrying out low-grade jobs, for example on production lines, with poor working conditions, are more likely to suffer from pregnancy-related problems such as preterm delivery, lower birthweight and to give up work earlier. These are also the women most likely to suffer from pregnancy discrimination. In considering the impact of work on women's health, it is, of course, essential to be aware that the health impact of job tasks is not restricted to women during pregnancy but may affect all workers carrying 62 out that job. In this regard, therefore, problems and discomforts faced by pregnant women are frequently the result of poor work practices that affect the whole workforce and may reect a poor att.i.tude to work design and conditions. Typically female jobs, including clerical or shop work, childcare and a.s.sembly line work, as well as waitressing and hairdressing, are done either standing up or sitting for long periods, may be repet.i.tive and may require to be done at speed, regardless of whether the woman is pregnant. Such jobs are a.s.sociated with poorer health at all times. For example, Cherry (1987) collected data on the experience of fatigue and nervous tension in the rst and third trimester of pregnancy and found breathlessness, back problems and varicose veins increasing throughout the pregnancy. However, she also collected information on health states in the 12 months before pregnancy. Though the two time periods are not directly comparable, the number of women reporting back pain was higher in the 12 months before pregnancy than in the two trimesters of pregnancy under study and more women reported nervous tension in that earlier period than at any stage of pregnancy. This cautions us against attributing health problems during pregnancy to pregnancy per se and also highlights the prevalence of health problems among working women more generally.

It is not unreasonable that women exposed to negative att.i.tudes or doing stressful jobs might experience poorer health. The mechanisms by which this occurs and the extent of the effects are probably too involved to tell a simple tale of cause and effect, even within the positivist biomedical tradition: hence, the proliferation of studies seeking to identify which factors at work determine what outcomes. Given what we have already said about women remaining in work and their reasons for doing so, there is a high likelihood of a range of responses in health terms to their circ.u.mstances. However, it is important not to fall into the trap of considering women to be especially vulnerable, and pregnant women even more so.

As we shall see, the evidence for pregnant women suffering distress is equivocal, despite many compelling anecdotes.

If we continue to address issues of women's health and wellbeing rather than that of the unborn baby, we nd similarly ambiguous relationships though often this produces positive relationships between working and mental health, particularly for part-time workers. Comparing workers to non-workers, De Joseph (1992) found that there were no signicant differences in mental health between the groups in measures of distress, including depression, state and trait anxiety and perception of negative life events. Like Homer et al. (1990), Saurel-Cubizolles and Kaminski (1987) emphasise that it is other factors rather than working by itself which can lead to poorer mental health among women working during pregnancy, and these tend to be the factors over which women have little control, such as socioeconomic status and working conditions. De Joseph (1992) admits 63 that there were large demographic differences between her groups in that homemakers were younger, less educated and had lower incomes and the study groups were not subdivided according to whether they already had children. Once more it seems likely that working during pregnancy is confounded with other factors that are a.s.sociated with distress (Mamelle et al., 1987).

It is also important to be aware that women may feel differently at different stages of pregnancy. In a study carried out in the Netherlands, Kleiverda et al. (1990) found positive effects of working on pregnant women's psychological wellbeing at 18 weeks of pregnancy, but at 34 weeks the relationship was more complicated: higher job satisfaction predicted an increase in psychological wellbeing and was related to lower anxiety. Leaving work earlier in pregnancy was related to a decrease in depressive symptoms, and more hours of paid employment correlated with an increase in anxiety. But, low job satisfaction, leaving work later in pregnancy and working longer hours were all a.s.sociated with low occupational level and, as the authors point out, socioeconomic variables might be more important as determinants of psychological wellbeing than work itself. The picture is complicated by the fact that unless women are in a position to give up work without nancial penalty, unemployment is likely to bring a lower standard of living, which in turn is related to an anxiety state and depression during pregnancy.

In a study of women working in law rms Schenker et al. (1997) found that women lawyers who were working long hours (more than 45 hours per week) were ve times as likely to report high stress than those working less than 35 hours per week. Those women lawyers who worked longer hours during the rst trimester of their pregnancy were also more likely to report high stress at work during pregnancy overall. This was probably not special to pregnancy per se, but to do with the nature of the job, the culture of the workplace and the expectations of work colleagues. In some cases, such expectations may be more imagined than real, and relate to their own concerns over poorer performance. In other cases, the comments and behaviour of others may add to the demands of the job. The ndings suggest that though pregnancy does create some special problems at work, results are not all explained by the pregnancy alone. Work ethos and working patterns are likely to be a factor in how levels of stress are reported. In some (mainly white collar) jobs, suggesting that a job is highly stressful may be a way of enhancing self-esteem, and being able to cope with a highly stressful job can be regarded as evidence of work commitment, despite expressing complaints about long working hours and pressures of time. It is also a way to justify time away from work when leave is due. Few people complain about the stresses of having too easy a job, for example. Nevertheless, highly stressful jobs may have a knock-on effect on other parts of people's lives, and on lifestyle. It is some elements of lifestyle and behaviours resulting 64 from stressful events, such as smoking, that are a.s.sociated with preterm delivery (Petridou et al., 2001).

The work on distress during pregnancy does not tell us whether those who did not suffer distress did not appraise pregnancy as stressful or whether they coped with the stress effectively. There has been little if no direct research on the way in which women appraise the stress of working during pregnancy or cope with work-related stress during this time. Hees Stauthamer (1985) described a process beginning in the rst trimester, whereby some of the women she studied reorientated from work to motherhood. Such women found that they were unable to integrate motherhood with their profession despite being highly committed to their careers before pregnancy. The women did not give up their long-term career goals but saw the reorientation as temporary. Rodmell and Smart (1982) reported that women relied on emotional support from friends and family and the most dramatic form of coping with stress at work was to absent oneself from the workplace and it is clearly the case that women who do not get better working conditions or better working arrangements are more likely to go on sick leave or to take up maternity leave earlier.

We can conclude from this survey of the literature, that most women adapt well to working during pregnancy inasmuch as levels of psychological distress are not found to be signicantly raised, using the kinds of measures available. There are changes in psychological state during pregnancy but working while pregnant per se does not account for signicant changes in psychological health. However, it is clear that some groups of women are more vulnerable, notably those who are unskilled, poorly educated, on low incomes and who have little job satisfaction or commitment. Working women who suffer psychological distress during pregnancy seem to differ from those who do not, mainly on variables that are outside their own control and that are not exclusively a.s.sociated with pregnancy, since these factors map closely on to those identied as causing psychological distress in non-pregnant working women (Cooper et al., 1988) and they will undoubtedly be confounded. If it is indeed the case that employment only rarely causes health problems for women, what of the other major issue, the impact on pregnancy outcome?

Work and pregnancy outcome Pregnancy is about having a baby. Whatever cultural overlay might be operating in stereotyping women as primary carers, as female workers or in discriminating against women when they are pregnant in the workplace, it is difcult to deny that women's concern during their pregnancy is that their healthy baby will arrive safely. In this context, the prolic research identifying the risks of certain jobs for pregnancy outcome cannot be ignored. This section is by no means a thorough review of this research 65 literature. It is intended to provide a avour of (a) the way that such research has been conducted, (b) the rationale for such research activity and (c) the nature of the ndings regarding pregnancy outcome and maternal health.

A cla.s.sic text, Chamberlain's book Pregnant Women at Work (1984), focuses heavily on teratogenic aspects of work without necessarily considering the role of work, the economic value of work and the women's psychological health or wellbeing. Because of the protective model operating in the workplace, the effect of the discourses of risk is not necessarily encouraging or sympathetic to women and, as we have seen, can mean that women are removed from their jobs or the workplace to make things easier for employers, rather than changing working practices. As Queneau and Marmo (2001) point out, women have to manage the tension between employment, pregnancy and health concerns for themselves and their child, and their economic position. The relative balance may mean that women are prepared to undertake risky work. In addition, research on paid work rarely takes into account the range and quant.i.ty of unpaid work done in the home or elsewhere, nor, as both Karen Messing and Patrizia Romito point out (e.g. Romito, 1989), the extreme physical demands of work that is traditionally undertaken by women, including waitressing, laundry work and cleaning. Except in cases of major illness, medical advice rarely includes suggestions that pregnant women should give up either paid or domestic work, whether paid or unpaid (Frazier et al., 2001; McKechnie, 1984), nor that teachers or nurses give up their jobs when pregnant.

In addition to the work we have already referred to in looking at the impact of pregnancy in the workplace as a social and political phenomenon, which has pointed to some of the psychological effects of working, there has been wide-ranging research investigating links between work and pregnancy outcome in particular. Like much of the work on pregnancy and employment, the research is rarely undertaken from a purely psychological perspective even if it frequently invokes psychosocial measures as a factor.

These studies are attempting to address several different questions, though one could be forgiven for not recognising the subtle differences between them. They include, for example: Is paid work itself related to poorer outcomes for women and babies?

Is it the type of job that produces an a.s.sociation with poor maternal or foetal outcome in pregnancy?

Are routine aspects of carrying out a job task, including lifting or carrying, standing for long periods or walking, related to outcome?

And nally, are the ndings on types of jobs and job tasks explained by the extent of physical activity or work effort involved in those tasks?

66.A positive answer to the rst question, that is, work has detrimental effects for women and babies, has enormous political ramications, particularly at a time of high female