The Midwife as Container

This is not an academic work. It is what occurred to me when I considered the concept of the midwife-mother relationship. I am writing as a retired midwife and psychotherapist. I am also the mother of two sons and have experience as member and chair of my local Maternity Services Liaison Committee. I have been retired now for many years and am essentially housebound with multiple sclerosis but I have maintained an interest through membership of the ukmidwifery egroup; I have a close friend who is a practising independent midwife and I have a carer who had her first baby in November 2007. This gives me a particular perspective. I have some theoretical knowledge and I think that my distance from current practice might give me an advantageous perspective.

Between 1982 and 1987 I worked as a community midwife in a district in inner London. My work involved all aspects of midwifery but in distinctly unequal proportions. I had two antenatal clinics a week in a local health centre and I made occasional antenatal home visits; I did very little intrapartum work because the part of London where I was working lacked that specific stratum of middle-class women who wanted home births, and domino births were only just beginning to be established there. There had not been any history locally of GPs providing cover for home births, unlike other pockets in north London. At that time it was still the case that certain traditional GPs who were interested in obstetrics continued to do this kind of work, but it was becoming uncommon and it was around this time that midwives who wanted to provide home births began to do so under the aegis of the local hospital maternity department rather than general practitioners. The greatest bulk of my work was providing postnatal care. There was very little continuity of care because, for geographical reasons, most women gave birth in a hospital outside my specific district. I might see the same women antenatally and postnatally but I missed a very important part of the process. I was frustrated about the lack of intrapartum work but the area, like many in inner London, consisted of an ethnically diverse population comprising very affluent families, a well-established white working class community with smaller subgroups of people from Caribbean and South Asian backgrounds and some very poor people, including the homeless. While this cultural mix implies a lack of interest in home births (the affluent women had private obstetric care and the poorer women were quite happy to accept what was on offer) it certainly provided a lot of socio- political interest. While I would have liked more continuity and intrapartum work, especially home births, I appreciated the fact that the balance of my work was similar to that of childbearing women. We are pregnant for around 40 weeks, in labour for maybe 24 hours, but mothers for the rest of our lives.

I believed, and still believe, that postnatal work is very important. At the time the practice was that if any woman was at home within the first three days after giving birth she was visited twice a day, and then every day until the baby was ten days old. Most mothers and babies were discharged then into the care of the health visitor and GP but we could continue to visit up to 28 days if we considered either the mother or the baby needed it. Before I started working on the community I had worked in a central London hospital where, although I considered the labour ward procedures abysmal, the postnatal care was, I think, about as good as hospital postnatal care could be.

I was, and remain, a passionate supporter of breast-feeding and I devoted whatever time I felt necessary to supporting women in this. Breast-feeding produces nutrition which is ideally suited to a human child: its nature changes as the child grows older; it also changes during the course of a feed. The immunological qualities are well-known; what is less well-known is that any administration of formula causes a breach in the gastrointestinal tract which together with the lack of maternal immunity is one of the major factors contributing to the lower health status of bottle-fed babies. Mothers who are HIV-positive in societies were clean water is not easily available are advised to breast-feed exclusively for this reason. (1, 2.) (I maintain that HIV-positive mothers in more affluent countries may also legitimately make this choice.) A breast-fed child initiates and maintains the process of feeding and will terminate it when he or she feels they are full. Breast-fed infants learn how to regulate their appetite and are therefore much less prone to later obesity. Breast milk contains essential fatty acids which are necessary for healthy neurological development in physiological terms. Because of the inevitable physical closeness of mother and child, because of the way in which the child initiates the feed and responds to the mother's let down reflex, the whole process is much more physically intimate and conversational than that of being bottle-fed. It is the quality of this communication which, it is thought, contributes on a level other than merely the physiological to the child's subsequent neurological development. Recent research on neurological development has shown that the quality of a baby's contact and communication from the first days onward affects the quality and quantity of cortical synaptic connections. (3.) This is one factor amongst numerous others affecting the better quality of health of breast-fed babies; an improvement which lasts far longer than childhood

The World Health Organisation recommends that babies be given nothing but their own mother's milk for the first six months of life with a gradual introduction of other foods during the second six months and that breast milk remain a significant part of the diet until the child is at least two years old. In the context of a country where most babies are being fed formula by six weeks this advice may seem extreme but in the world's population as a whole most babies are breast-fed for between two and five years.

Given these advantages if as part of postnatal care a midwife's work is to help the foundation and maintenance of breast-feeding this, in itself, is profoundly important work. Because we live in a society where most babies are bottle-fed the advantages of breast-feeding are never adequately described in order to avoid alienating most mothers but I maintain that it is not only important physiologically but also important because it enhances the quality of attachment, the quality of security which a child feels as its foundation psychologically and socially. What follows will suggest the profound importance of the quality of this primary attachment as the basis of subsequent mental health.

I soon noticed that a substantial part of my postnatal work was not the detection and treatment of pathology. It is true that there was quite a bit of this: particularly dealing with perineal breakdown. But in addition to supporting breastfeeding I found I was listening to women talking about the pattern of their days and nights with a new baby and I came to believe that this simple act of listening was in itself very important. For first-time mothers the newness of this task is evident but even for mothers of subsequent babies it seemed necessary. The intensity of their involvement with their existing children seemed to mean that they had forgotten the reality of a newborn.

What was this listening providing for these mothers? Why did I come to feel it was so important?

I was practising midwifery in a context where many first-time mothers had never held or seen a newborn baby. Much of my postnatal work was to reassure them that their experiences were normal. New mothers did not need to be poorly educated to be naive about this. I visited one couple where the father was a paediatric consultant and the mother an experienced paediatric nurse. Both were totally inexperienced in the ways of the normal newborn. Both needed the same kind and level of support and reassurance.

Before I became a midwife I had studied psychology for five years and had a first-degree in social psychology and a masters in psychopathology. As a student of psychology I was told that postnatally new mothers were in a state unlike any other. Some authors referred to it as a critical period; others as a sensitive period. But whether critical or sensitive the theorising was that there was a specific window of opportunity for mother and baby to relate to each other with a strength and quality unique to this particular time. Midwifery tutors referred to this as' bonding'. It was even occasionally referred to in the notes but I was critical that this concept had not been sufficiently well understood and that a particular behavioural style was expected; this misunderstanding allowed opportunities for judgment and condemnation. A mother may be described for example as 'not bonding properly' but there was seldom any recommendation about how this situation, if indeed it was the case, may be improved. And it seemed that much maternity care fostered judgment and condemnation while precluding the conditions for autonomous relationships, either between professional and client or between mother and baby. Ironically, although the psychological theorising was postulating an instinctive response it seemed to me that the instinct must be based on autonomy.

In psychotherapy the word 'holding' is often used, as is the word 'container'. Both refer to something psychotherapists do for their clients: they apprehend and accept what is often feared and denigrated and the simple fact of acceptance enables the client to live with what was previously felt to be intolerable. In this particular context the word 'holding' is particularly pertinent. It might be said that the midwife metaphorically holds the mother so that she can both literally and metaphorically hold her baby. It is obvious that when women are in labour they need a high level of care and attention, but I think that a particular quality of attention continues to be required in the postnatal period. I think that when I was listening to the stories mothers were telling about their new experiences with their babies I was providing this kind of holding.

I feel it is necessary to state something about the academic status of psychotherapeutic, and particularly psychodynamic and psychoanalytic, concepts. Within academic psychology such theories are frequently derided because they are considered unscientific. I agree that they are unscientific but I do not think that this means that they are useless. Not all phenomena are amenable to scientific analysis and quantification. Some, such as the phenomenon of labour, are simply too complex and multifarious to be subdivided into the easily quantifiable. Others are simply too subjective. An entomologist is clearly very different from his subject matter. This is not true of psychologists, sociologists or anthropologists. I think that psychodynamic concepts are useful because they offer an imaginative explanation of subjectivity and particularly irrationality. In the context of childbirth I think they are particularly relevant because they emphasise the strength and importance of early childhood experience and therefore imply a depth of importance to the mother-child relationship. They also therefore imply a commensurate importance to the work of the midwife. They also include the concept of the parallel process and I shall be suggesting that the midwife-mother relationship is in some ways parallel to that of the mother and child. The particular psychodynamic concepts I shall be focusing on here are those of attachment theory and primary maternal preoccupation.

When I was an undergraduate we learnt the famous 1958 experiment of Harlow's monkeys (4). The rationale for this experiment was to test the behaviourist tenet, which now seems unbelievably naive to me, that the only positive reinforcements or rewards experienced by any animals, including humans, are those which meet very basic drives. In other words we are only motivated by the most basic of instincts. Harlow disagreed with this and theorised that if he could demonstrate that animals other than humans can be motivated by needs other than the most basic this must apply to human beings also. He took baby rhesus monkeys and separated them from their mothers and gave them two options. One was of a wire contraption which provided milk and the other was one which consisted of soft terry towelling. Both contraptions had pretend and very unconvincing faces with eyes. Behaviourist theory would predict that the baby monkeys would prefer the option which provided food. We were taught that the baby monkeys, while they would go to the milk producing contraption when they were hungry, much preferred the terry towelling one, in other words they preferred the sense of tactile quality to that of being fed. I learnt this theoretically from a textbook but later actually saw a film of the experiment and wept when I saw the terrified little monkeys desperately clinging to this lifeless and unresponsive pseudo-mother. When these baby monkeys grew up they were unable to rear their own young because they themselves had not been adequately mothered. They had been removed from their kinship group and therefore not seen other adults mothering and, possibly more significantly, had not experienced being mothered subjectively themselves.

Attachment theory is based on the work of John Bowlby who, starting in the 1950s, wrote three books entitled Attachment, Separation and Loss (5, 6, 7). As his theorising became more focused on the concept of attachment in 1988 he published a book called A Secure Base (8). His work was descriptive, essentially ethological. He maintained that babies need a single, primary carer with whom they develop a particularly strong relationship and that it is only after the development of such a relationship that they are able to tolerate separation without finding it traumatic. Of course, degrees of separation must be age appropriate: a separation of a few minutes is all that a newborn can tolerate; when children can walk they can mitigate separation by following their primary carer: it is noticeable how children can appear to be playing quite separately while their mother is in the room but as soon as she leaves, they will notice and follow her. As children grow older the length of separation which can be tolerated grows commensurately. Loss, bereavement, is unusual in childhood in affluent societies; but while it is statistically unusual it is not abnormal in the sense of pathological.

For most children their primary carer is their mother. There will be important significant others in the immediate and extended family. In the west where small nuclear families are the norm it is probable that the numerical extent of significant others is underestimated.

Second wave feminists of the 1970s strongly disapproved of Bowlby's theories because they thought he was trying to rationalise a post-war discouragement of women in the workplace; indeed, I heard a similarly disparaging assessment of attachment theory at a conference organised by the Psychology of Women Section of the British Psychological Society in the year 2000. However I always thought this was unfair criticism. Firstly, it seemed to me that all Bowlby was doing was describing an inevitable reality. It is only women who can gestate and give birth. Secondly, while other mothers can nurse an infant, and this is quite common in some societies, all mothers of newborns will produce milk and as an advocate of breast-feeding it therefore seemed appropriate that a child's primary carer should be his or her mother.

But while attachment theory describes the setting up of secure foundations, it does not describe the parallel process that a midwife provides to a mother. For that I should like to evoke the work of Winnicott and the concept of primary maternal preoccupation (9).

Winnicott was a consultant paediatrician at Paddington Green Children's Hospital in central London. He used this work as the basis of his theory and practice as a psychoanalyst. Like Bowlby his work was descriptive and ethological. (It may be true to say that this is the way in which most psychotherapists work and that it is an appropriate discipline: one which is scientific because it can generate hypotheses which may be tested although it is possible that much human experience cannot be tested experimentally because, as stated above, human experience cannot be quantified and controlled for.) The fact that Winnicott was a paediatrician, unlike most psychoanalysts who were either psychiatrists or neurologists, I consider very important. He was not dealing with children who were psychiatrically ill or whose families were particularly dysfunctional and he therefore experienced a range of normal family environments unusual for a psychoanalyst. He is famous for his concept of 'good enough mothering', for his realistic description of a family environment where children experience their parents' human failings and, indeed, where such experiences are necessary for a child to develop a realistic sense of self and others.

When Winnicott described his concept of primary maternal preoccupation he was referring not only to the critical or sensitive period immediately after childbirth but to a longer period which starts during pregnancy when a mother begins to relate to her unborn baby. This relationship will involve fantasy to some extent and each woman's personal history and will alter irrevocably when the child is born. Sigmund Freud described the act of childbirth as 'an impressive caesura' (10). The child becomes visible, becomes an air breathing creature. But this physical act of birth he maintained did not coincide with the psychological act of birth. As a psychology student we were taught that children develop a sense of self only after they have developed a concept of the other. I was studying psychology in the early seventies and, at the time, we were taught that this occurred at about eight months. Like the behaviourist tenet described above this timescale seems based on a naive adherence to theory without any realistic observation. It seems obvious to me that eight months is far too old. Psychoanalytic theory focuses on the imagined psychic reality of much younger babies.

Psychoanalysts, such as Winnicott, who focus on early experience and relationship are referred to by the ugly terms 'object-relations theorists' and their focus was on the psychic reality of much younger babies but this reality does not include a concept of self separate from that of the other. The birth of the psyche could be considered to occur when the child is able, in Bowlby's terms, to tolerate separation. The length of this time will vary according to each individual relationship.

When Winnicott was writing about primary maternal preoccupation in his book From Paediatrics to Psychoanalysis he said that in any other context this degree of focus would be described as an illness. He did not elaborate because his concern was to demonstrate that primary maternal preoccupation was a normal part of new motherhood, but the psychiatric terms which occur to me are obsession, monomania and psychosis in the sense of having loose boundaries to the self. (11) Primary maternal preoccupation is not only normal in the sense of being statistically usual but is positively healthy for the baby's subsequent development. It provides the maternal template on which the secure base is founded.

I should like to make some comments on Winnicott's use of the word 'illness'. He was clearly referring to psychiatric illness. When I was working as a midwife I had a very generous manager who, knowing that my background was in psychology, would arrange for me to be a delegate to conferences of the Marce Society and even, on one occasion, to be the Royal College of Midwives' delegate to the Seventh International Congress of Psychosomatic Obstetrics and Gynaecology at Trinity College, Dublin. The Marce Society was overtly concerned with what it described as postnatal illness, specifically postnatal depression and puerperal psychosis. I am critical of the medical model of both mental illness and childbirth. I do not believe that postnatal 'illness' can be reduced to the hormonal. The fact of new motherhood is no doubt crucial but new motherhood involves more than just the hormonal; it involves the social, political and spiritual to say nothing of each woman's own personal history. I believe, on the basis of my own experience as a mother and a psychotherapist, that when mothers are dealing with children too young to speak they need to empathise on a non-verbal level and that by doing so they relive on an unconscious level their own childhoods and when these have been problematic or traumatic they are more vulnerable to postnatal 'illness'.

There has been a recent edition of the AIMS Journal (12) in which mothers were saying that they felt they could not speak freely about their mental state to health professionals in case their children were taken into care. I think it is statistically normal for new mothers to have 'mad' thoughts. These thoughts are specific to new motherhood but are rendered pathological by the fact that our society is not accepting of the state of primary maternal preoccupation. The mothers of new babies are socially isolated at a time when they are having to come to terms with this unique condition.

I should like to describe some of the mad thoughts I had when I first became a mother to illustrate which ones I think are intrinsic to the condition of motherhood and which ones I think are distinctly personal to me. It is a cliché of feminism that the personal is political and by being autobiographical here I would like to express the normality of these supposedly mad thoughts. By including the personal I think it should be possible to make similar distinctions for others.

When I gave birth to my older son I was a midwife; I was used to the phenomenon of birth; I understood newborn babies. None of this theoretical and practical knowledge prevented me from being shocked by the sight of my son. I was so used to experiencing him as a mobile being inside me and through palpitating myself that the distinction between the visual and tactile had never occurred to me. I had not been able to imagine him as separate from myself. He was evidently normal and healthy but was so distinctly an individual: skinny, redheaded and half bald -- and male. I had been convinced I was going to give birth to a dark haired girl. Nowadays most women have scans which indicate the sex of their babies and it is my impression that most mothers are keen to know this in advance. I had not wanted this. I had not wanted a professional to tell me about my baby; I had wanted the uncertainty which would only be resolved at the time of birth but, even so, I was not able to apprehend the reality at all.

Dealing with him initially was easy. I knew how to handle newborn babies. When he had problems latching on I took the advice I would have given any other new mother and it worked. Because I had mild relaxing remitting MS and knew that mothers with MS were most at risk of acute episodes postnatally my partner had taken four weeks off work to support me and everything was happening within an intense bubble as I got to know the reality of living with a newborn. As soon as my partner went back to work the bubble broke and I became terrified. I was terrified that I had now become totally vulnerable: the most precious thing in the world was outside me and it was my duty to protect him and the world seemed a terrifying place. He was born the day after the ship the Herald of Free Enterprise had sunk off Zeebrugge harbour with massive loss of life -- in fact my waters had broken as I was watching the news bulletin. I found watching the television news intolerable. I found the world outside my house very frightening. When I went out I used to carry him in a sling and I had fantasies of somebody running towards me and smashing his skull with some sharp object. My partner used to attend meetings of a particular political group and I was convinced he was going to be arrested under the Prevention of Terrorism Act, that he would go out and I would never see him again, that he would go out and I would not know when or whether he would return.

I think the shock of seeing your baby for the first time is very common, in spite of the way in which ultrasound tries to make visible what is internal and tactile. I think most new mothers are in a state of shock, both in the colloquial sense and in the psychological sense. Psychotherapists use the word shock to describe a state different from but analogous to physical shock: with physical shock the body shuts down peripheral systems to maintain the viability of the vital core. With psychological shock the mind shuts off thoughts and perceptions which are threatening or irrelevant to the matter in hand and in this case the matter in hand is learning about one's new baby.

I think the degree of shock may relate to some extent to factors intrinsic to labour. My son was born two and half weeks early; maybe if my pregnancy had lasted forty or forty two weeks I might have been better prepared psychologically for the sight of him. And I know that women who experience rapid labour also experience shock which, in that case, probably has a considerable physiological component. (I am not concerned here with psychological aspects of what is unusual or abnormal. Everything I write is intended to relate to what is within the bounds of the physiologically normal, in other words the province of midwifery expertise.)

I also think it is very common for new mothers to find the outside world threatening and to find themselves particularly sensitive to others. When I attended conferences of the Marce Society it was a truism that postnatal depression was different from other sorts of depression because it included aspects of agoraphobia. And I think the way in which the boundaries of the self become fluid in pregnancy and new motherhood can often lead to unbearable empathy with others. In pregnancy one's body is accommodating a completely different human being in development yet in the early stages this completely different person is utterly physically dependent upon one. I remember as a student reading that psychoanalytic descriptions of pregnancy stress that the psychological task for the mother is first to accept the dependence and indivisibility of the relationship with the baby and then to conceive of the baby as a separate being. This indivisibility is the basis of primary maternal preoccupation, attachment. If attachment is a state of specific hormonal as well as psychological characteristics then I think the state of heightened empathy has a physical basis. (Maybe midwifery knowledge is that which integrates the hormonal and the emotional. Tricia Anderson was researching the relationship between oxytocin and trust.). I remember my partner delightedly telling me how much he was enjoying watching me fall in love with my baby: this is what the 'babymoon' is all about -- and it is not specific to humans. It is well known that any mammal with a new offspring can be lethally dangerous. I think that my fantasy of a lethal attack upon my baby was possibly a reflection of my own capacity to defend him lethally if necessary.

'Mad' thoughts are not limited to the postnatal period. An article on antenatal depression in the G2 section of the Guardian on 29th January 2008 (13) quoted a number of women who considered their feelings and responses in pregnancy abnormal. One was so concerned about causing damage to her unborn child that she went from ensuring that she used no artificial chemicals in toiletries and make up to feeling a need to go home and wash after a stranger had accidentally brushed against her while out shopping. This woman's previous pregnancy had ended in miscarriage. She went on insightfully to say: "I knew I was being irrational, but the behaviour also seems logical to me. The fears were layered onto one another -- the fear of harming the baby, the fear of the baby dying, the fear that I was going mad." Another said, "I have read that being pregnant is the closest you will ever get to the other side... with the soul inside of you straddling the worlds of darkness and light."

Winnicott says that primary maternal preoccupation begins in pregnancy. I find both of the women quoted above perfectly comprehensible. The first's response was exaggerated terror, a concern which had been coloured by previous loss. And much compulsive behaviour is to do with cleanliness. It would be interesting to know whether this behaviour was cured by the birth of a healthy baby. The second is speaking from the position of 'psychosis' as psychoanalysts use the term, from the state where the boundaries of the self are loosened.

I think all the above is generic to primary maternal preoccupation. Adam Phillips in his book on Winnicott wrote: "it was no longer sexuality or the death instinct that constituted the unacceptable in Winnicott's version of psychoanalysis, it was the early dependence and the terrors involved both its full acknowledgement and in its possible insufficiency." (14). Early psychoanalytic theory articulated and made conscious what, at the time, was considered unacceptable. The reference to sexuality as something taboo is obvious; the reference to the death instinct is somewhat arcane and I would consider the word 'aggression' to be more appropriate. What early psychoanalytic theory did was to make clear that we all of us embody sexuality and aggression, whether we choose to express this or not. Phillips here is emphasising the total acknowledgement of dependency and its insufficiency. Much of the above can be seen in this light. Many new mothers' experience of shock is concerned with the realisation that this baby needs looking after 24 hours a day, 7 days a week, 52 weeks a year. When I was a midwife I used to have discussions with health visitors about whether it was possible to tell pregnant women the reality of new motherhood. I was never sure and used to wonder whether the health visitors' certainty that it was impossible was because they didn't want to make the attempt. But at the time I wasn't a mother. Recently one of my carers has had a baby. During her pregnancy I said that with a very small baby I had sometimes found it difficult to make time to brush my hair. She couldn't imagine that until after her baby was born.

But Phillips is also describing the totality of this dependence as something which is somewhat taboo. I agree with this but I also think there may be something culturally specific. I remember once going to pick up my older son from nursery school with his baby brother in a sling on my front and one of the other mothers said that she thought carrying a baby like that 'made them dependent'. I wondered what level of independence she thought a baby only weeks old was capable of. But we live in a culture where, if there is sufficient space, it is considered normal for new babies to sleep in separate rooms. There are other cultures where children will always be carried until they are capable of independent movement.

Winnicott is also famous for saying 'there is no such thing as a baby'. What he means is that a baby outside a caring relationship will die. This is the other part of the taboo Phillips is referring to. Mothers need to experience primary maternal preoccupation, to fall in love with their babies, in order to provide the constancy and intensity of this care without resentment.

What is specific to me concerns my experience of my father's death at the age of eight. He had left one day and I had never seen him again. When I became a mother my partner became a father and on an irrational level I had learnt that beloved fathers die. These irrational beliefs are held by all of us but they are unique to each of us and, I believe, they influence the nature and course of each new mother's postnatal experience.

Adam Phillips referred to the insufficiency of the dependence. My individual experience involved my father dying when I was eight. But I think that new parenthood generally evokes mortality: the realisation of the vulnerability of the infant; the heightened empathy; the fact that the world is in many parts a dangerous place. What is specific to me gave the experience a certain flavour but did not, I think, alter the content.

Countries in Western Europe are affluent and therefore have low maternal, neonatal and infant mortality rates within the global context. But mothers and babies do still die and a zero mortality rate is unachievable. I do not think any woman undergoes pregnancy and childbirth without considering the possibility that either she or her baby might die. The quotation above from the Guardian where the pregnant mother is describing her condition as straddling both worlds is, I think, a metaphorical acknowledgement of this. And I think this is an appropriate thought which gives due regard to the enormity of the phenomenon: the creation of a genetically unique human being. If this implies that I believe the context of childbirth renders 'mad' thoughts normal then that is correct and I think Winnicott's reference to illness supports me here.

I believe that the end of primary maternal preoccupation occurs when the baby, who is no longer so young, can tolerate separation. But that this is also influenced by the mother's ability to tolerate separation and that sometimes this can be pathologically difficult depending on her own past experiences.

I realise that I am conflating Winnicott's description of primary maternal preoccupation with attachment theory, but I think they are both describing the same phenomenon from different perspectives and I think this conflation is therefore intellectually economical.

Winnicott is here describing pregnancy, childbirth and the postnatal period from the mother's perspective. He also described birth from an imagined perspective of the baby. He said that in normal childbirth babies feel as though they are initiating the process. Nobody knows exactly how labour is triggered but it is thought that some mechanism, probably hormonal, within the feto-placental unit is responsible and so there is a sense in which the baby really does trigger the process; but I think Winnicott was talking about something different. I think he was contrasting his imagined perspective with one more normally described whereby the baby is overwhelmed and endangered by the process of birth. It is an obstetric cliché that the journey down the birth canal is the most dangerous journey one can take. I think that midwives should be critical of obstetric clichés. I maintain that while obstetricians may believe that labour is only normal in retrospect this is not a suitable outlook for midwives. Labour is a normal physiological process unless pathology can be demonstrated. Winnicott was suggesting the possibility that since in normal labour the contractions are rhythmic and regular the baby may come to anticipate and respond to the regularity, may even find the process exhilarating rather than threatening, may even find the regular pressure of the birth canal and the subsequent release pleasurable. Just as this is imaginary and unprovable, so my interpretation is also personal. But the rhetoric of childbirth which is framed in terms of damage and danger is also politically driven and emotive rather than scientific.

The word 'regression' can be used in two senses. Janet Balaskas (15) means it in one sense when she writes:

"in the hours of labour you will want to withdraw from the normal day-to-day level of things and your attention will naturally turn inwards, as if the whole world contracts to what is happening in your body. In your mind, time takes on a fresh dimension. Hours can pass in what seems like minutes. It is like being in another world... this great opening of the womb happens only a few times in your life. It is a very deep emotional experience which involves a regression to your most basic and primitive feelings."

What she means here is leaving behind the normal state of everyday consciousness. When she is writing about active birth she is describing a state of being in labour which will enable the woman to respond instinctively to her body's promptings such that she is more easily able to avoid pharmacological analgesia. If women in labour are to be able to do this rational thought is an impediment. It is the function of the midwife to take on that aspect for the woman. Regression in this sense means regression from an ostensible state of civilisation to a return to the animal. In these circumstances the midwife can be said in psychotherapeutic terms to be holding the boundaries, to be creating a safe space in which the mother can trust that someone else can deal with the practicalities of everyday life. Mothers and other carers do this for young children because they are physically and intellectually incapable of doing it for themselves. In this sense, although the labouring woman is performing a profoundly adult act, the regression she is experiencing is similar to that most often meant when the word is used to mean a return to the infantile. If it is the case that the mother of a new baby relives her own babyhood in order to empathise with and understand her child and if it is one function of primary maternal preoccupation to be in this state then on one level all of a midwife's clients will be regressed to some extent.

Bowlby wrote of attachment, which is quite a cold and mechanical word. Winnicott wrote more emotionally from both the mother's and the baby's point of view. Sheila Ernst (16) also wrote from the point of view of both the mother and the baby but she used quite an extreme word: 'merging'. Psychotherapists may refer to clients being still in an unhealthy state of merger with the mother. When Sheila Ernst was writing in this way about merging she was deliberately giving a feminist interpretation to what was generally perceived as something pathological. She was implying that the intense closeness of the mother-newborn relationship was not unhealthy but is in fact the foundation for later mental health. She says that if a child is to achieve a healthy state of separation there needs to have been an initial total merger. If a mother is merged with her newborn she must, by definition, be in a regressed state. It is possible that the metaphorical holding by the midwife of the mother in labour can make it easier for her to achieve and feel comfortable in this state afterwards.

Some women's experience of labour may be a precipitant of later problems, or a contributor to them. For example, a sense of having been traumatised in labour may exacerbate subsequent problems; a sense of having lost control may be particularly relevant in cases of previous physical or sexual abuse. But it is also the case that experience of labour can be healing. This is very individual. What may seem objectively intolerable to me may be experienced by the woman as helpful. It may be the mere fact of having been the focus of caring attention for a number of hours, as the mother must focus many subsequent hours of caring attention on her baby: or labour may be experienced as physically cathartic.

Labour tends not to be seen in these terms: it is usually described in terms of the biomedical and psychotherapists do not have the experience of being with women in labour to inform their discourse. The power of labour to influence the quality of the mother's subsequent relationship with her baby is, I believe, underestimated for these reasons and this underestimation to some extent determines how resources and care are delivered.

I feel that psychology, when it applies itself to childbirth, does not take certain factors into account. I have been a member of the Society for Reproductive and Infant Psychology for twenty-eight years and have seen every one of their journals during that time. Since I qualified as a midwife in 1980 I have done some work in every aspect of women's reproductive life; I have been a nurse in a daycare abortion clinic and a gynaecology clinic and trained as a family planning nurse. So I have worked with women who wanted to get pregnant but were unable to, women who were pregnant and wanted the pregnancy terminated, women who wanted to avoid pregnancy as well as childbearing women throughout their experience from conception to the postnatal period. On the whole when I read journal articles looking at the psychology of any aspect of this it seemed to me that two things were seriously underestimated: the power of the phenomenon of labour and the power of the institution. I will deal with the institutional power below.

I think that some midwives take refuge in fragmentation in order specifically to avoid the power of labour. The midwife-mother relationship is therefore attenuated, and it is possible that the mother-baby relationship is also reduced. Certainly Michel Odent thinks that this is the case in many cultures worldwide including those of the west when it comes to practices in the third stage of labour which involve separating the mother and baby or hastening the process. He believes that this reduces the level of oxytocin output and therefore the intensity of the immediate attachment (17).

It is obvious how a midwife can provide a powerful quasi-psychotherapeutic holding for a woman in labour, but I maintain also that when, as a community midwife, I was listening to mothers tell their stories of the early days, I was also metaphorically holding them in this state of merger, assuring them that it was normal and healthy and that it was not going to last forever.

So, I do think that a midwife can be a container in the sense of holding the space for a woman in labour and providing metaphorical holding through the intensity of primary maternal preoccupation, or at least its most intense initial phase. But I think there are powerful institutional pressures militating against midwives adopting this role.

When I was working as a community midwife that level of postnatal support was taken completely for granted; there were many midwives who believed that they were under a legal obligation to visit a new mother daily for ten days. But although this was taken for granted in Britain it was unusual internationally. I remember wondering with some incredulity about the situation in the United States where there was no state provided postnatal support. There was also a completely different attitude to the medical profession. I was under the impression that general practice was rare. Affluent women would take their baby to a paediatrician and I was concerned that this meant that normal postnatal blips would be subject to overmedicalisation. I had seen this myself particularly with blocked milk ducts treated unnecessarily with antibiotics and mothers discouraged from breast-feeding. It was my impression that few doctors understood or cared about breast-feeding. It was quite normal at the time for radical midwives in this country to look with some envy at the Netherlands where midwives had higher status and were self-employed and there was a home birth rate, at that time, of well over 50%. At that time in the Netherlands the normal pattern of postnatal support was for a maternity nurse to live in the home for some time and for midwives to take no further part in postnatal care. This system was not as advantageous as it seemed because the maternity nurses, in some cases, supported the mother by bottlefeeding the baby. But postnatal support in this country was taken so seriously that, when I started training in 1978, I was taught that local authorities were under a legal obligation to provide a home help for the first ten postnatal days if the mother was not adequately supported by her family. I must say that I never knew this to occur and the legal obligation soon evaporated under the pressure of financial constraints.

All this seems very alien now. It is my impression that, in some areas, postnatal visits are limited to one on day 7 to check the baby and take the Guthrie test and that this visit may be made by a maternity care assistant. Maternity care seems very fragmented to me. There are not enough midwives. It is my opinion that best practice, as judged on almost every criterion, is provided by independent midwives and case holding practices and that continuity of carer is intrinsic to this. If midwives are to provide the parallel process of containment this continuity allows them to be there at the beginning of primary maternal preoccupation in pregnancy and through the intense climax of labour and birth. And if midwives are to be able to provide this they need to be given their due reward both financially and in terms of status as independent practitioners equivalent to but very different from obstetric consultants.

When I became a midwife I had the idea that midwifery could be a kind of preventative psychotherapy, that by providing care of the highest standard the midwife could enhance the quality of the mother's relationship with her baby and because this relationship is the foundation of later mental health the baby's mental health would inevitably be affected for the better. This idea soon came to seem naive when I realised the extent to which I had underestimated the power of the institutions which dominate childbearing. When I was a student of social psychology our main textbook was Social Psychology by Roger Brown. Much of this was concerned with research on conformity and compliance which had been conducted in the aftermath of the Second World War in an attempt to understand how Europe had been in thrall to fascism in an attempt to prevent the horrors, particularly of nazism, ever happening again. The tenor of this research demonstrates that people are prone to obedience and compliance if the authority of the institution is sufficiently powerful. This obedience can manifest comparatively easily as behaviour which is abusive.

In 1961 Stanley Milgram devised an experiment on obedience {18}. An experimental subject was seated at a table with two stooges, one performing the role of an authoritative scientist wearing a white coat and the other a pretend experimental subject. The real, innocent subject was told that the experiment was an attempt to investigate the role of punishment on learning. The authoritative scientist would give the pretend subject questions and, when he made an error in responding, order the innocent subject to administer an electric shock. In fact, no real electric shocks were administered; both stooges were acting. A significant number of real, innocent subjects would administer electric shocks even when the pretend subject was showing a degree of pain, even when a dial on the table reached a level marked 'danger' in red. The experiment was designed explicitly to test the extent to which ordinary people would obey orders even when this involved ostensibly doing something counter to their principles. The initial experiment was carried out at a time when the Nazi war criminal Adolf Eichmann was on trial in Israel and asked whether perpetrators of the Nazi Holocaust were also merely obeying orders. This trial became generally very famous. Milgram was sued by some of the real subjects who claimed they had been traumatised by the experiment and that the experiment was unethical. In spite of the notoriety of this experiment its results have never been subjectively taken on board in real-life situations where authoritative people administer orders to those lower than them in the hierarchy.

Another example of this was provided by Philip Zimbardo in the Stamford experiment (19). Psychology students were allocated at random to one of two categories, that of prison guard or inmate. Within a week those who had been designated inmates were experiencing severe psychological distress as a result of incarceration while those designated guards had developed abusive behaviour analogous to that which is being demonstrated by Allied forces in the Iraq war.

It may seem an exaggeration to compare medical institutions with prisons but it is a comparison which has seemed blatant and legitimate to me since I was a student midwife. There is a clear hierarchy: doctor; paramedical staff such as physiotherapists and radiographers; midwives and nurses; students, although the position of medical students may be slightly higher up the hierarchy; and most definitely at the bottom is that of patient. There is an expectation that those higher in the hierarchy instruct those lower. Theoretically any procedure carried out without informed consent is assault. In practice, true informed consent is rare. I have seen cases of bullying at all levels including those of women at their most vulnerable, in labour. However much midwives are taught that they should communicate with the women they care for and that their care should be based on informed consent it seems that they prefer to concede the authority of the obstetricians and institutional procedures.

In addition to these apparently ineradicable psychological attributes there are more specific social circumstances.

In February 2008 David Cameron, the leader of the Conservative party, said that if they win the next election he will institute a system similar to that in the Netherlands of providing new mothers with maternity nurses. There are many practical problems implicit in this, not least that there is no training for this post established in this country. Midwifery training has been well-established and there are about 5000 too few midwives in post. But regardless of practicalities it implies a fragmentation of care. As things are the role of the midwife arches over the first important part of primary maternal preoccupation and this arching is important because it accommodates the critical period of labour and birth, linking it with the postnatal period.

As stated above the continuation of the midwife's role into the postnatal period has been unusual internationally. This recommendation would bring Britain more into line with other nations. This is not necessarily a good thing. Fragmentation of care is usually to the psychological benefit of the professional, not the client.

We live in a fragmented and unhealthy society. Within the last year two teenagers have been killed in my neighbourhood, one less than half a mile away, the other known to both my sons as a fellow school student. I think we are in a situation analogous to that of Harlow's monkeys who could not mother their own young because they had neither been adequately mothered themselves nor been exposed in their kinship groups to other mothers' mothering. I became a midwife because I thought that it might offer a kind of preventative psychotherapy. Then, when I was working in the field, I thought this intention very naive because I was impressed by the power of institutional structures. I do not think that the provision of good quality postnatal care in the context of continuity of care can allay social breakdown. I think that the fragmentation and undervaluing of maternity care is another symptom of this social breakdown.

However I am reluctant completely to ditch my intuition that midwifery could indeed be a preventative psychotherapy. I think the psychoanalytic theorising around this indicates that the midwife could be a container, holding the boundaries for the mother to hold her baby metaphorically as well as literally. The fragmentation of care, the low status, the inadequate numbers of midwives indicates the low status of motherhood generally. I think this low status and social pathology are not coincidental. But if it were to be the case that midwifery could properly fulfil its function by providing holistic care at this uniquely powerful time of transition a considerable level of social change would be required requiring a higher status to be accorded to both midwifery and motherhood.