The low status of mothers in the labour market has therefore been closely related to the limitations of social provisions such as child care. How has child-care provision changed during the twentieth century?

Proponents of the degradation-of-mothering thesis have often argued that improvements that have facilitated child care—from antibiotics to plumbing—have been offset by rising standards of mothering, including greater stress on the emotional aspects of caring, and that the effective burden of child care has been continuously increasing. How accurate is this assumption?

At the turn of the century, unlike the upper classes who had their children fully mothered by ‘nurses’ and the middle classes who employed nursemaids to do child-minding (Horn 1990:76-7), working-class child-care arrangements and provision involved kin, neighbours and schools. Grandmothers and other kin were sometimes paid for caring for children, in the same way as child-minders. Some free care in creches was provided by state elementary schools to prevent absence from work of girls who would otherwise have needed to stay at home to look after babies. Women seemed to prefer child care through their own networks. Lewis (1984:56) notes that, in the early twentieth century, creches were abandoned partly because of economic reasons and partly as a result of pressures by the infant welfare movement on mothers to stay at home with their children.

The infant welfare movement emerged as a result of eugenic concern about the quality of the race: an imperial power needed physically strong and visible rulers (Davin 1978). The emergence of this movement did not change the prescribed role of wife and mother but, at least in theory, recognized the importance of the work performed by mothers (Lewis 1984:82; Ross 1993). Adequate training for women in housework and mothercraft was emphasized. Many middle-class women found a purpose in teaching social maternalism to the poor or worked as health visitors, rejecting older working – class styles and practices of mothering in favour of a new ideal. The prevalent middle-class ideal mother was a clock-watcher, establishing perfect regularity of habits to achieve character formation. Babies were to be toughened up and made independent of their mothers as soon as possible (Humphries and Gordon 1993:52). However, Ross (1993) asserts that the new ideal of motherhood had a limited impact on working-class mothers both because of their more relaxed attitude to baby care and because of the constraints of their day-to-day lives which led to babies being more integrated into daily living.12

The conservative and patronizing crusade to improve baby care in poor households and the movement to make childbirth safer may have had, however, some positive practical impacts: by the late 1930s a baby’s chance of survival had increased four times compared with the rate at the turn of the century (Humphries and Gordon 1993:55). Improvements in housing and health and social care were also responsible for better survival rates. Clearly most working-class mothers resented the patronizing intervention in their mothering but wanted to be helped to better their family’s life conditions (Ross 1993).

It has been argued that the atmosphere of regimentation and increasing control during the Second World War led many mothers, of all social classes, to follow dutifully the prescriptions of the mothercraft methods (Humphries and Gordon 1993:55-6; Riley 1983). Mothers were increasingly doing as they were told by doctors, nurses, health visitors, and mothercraft manuals.13 Once the war was over, a more relaxed and permissive approach came into vogue. Mothers were told to enjoy their babies and their families and to form close, warm and loving relationships.

Pronatalist concerns following the Second World War focused on the need for ‘adequate’ mothering as a means to secure social stability. One key to achieve this was, as it had been at the beginning of the century, full-time motherhood (Lewis 1992a:11, 21; Riley 1983), but these ideas were now also backed up by psychoanalytic (and feminist) thinking concerning child care.14 Ross (1993) argues that the rights of mothers to receive public resources, advocated in the late 1910s by the Infant Welfare Movement, was a fading idea by the 1930s. Public discourse came to disregard mothers’ work, intelligence and effort as a social contribution. By the 1940s, mothering appears as an aspect of feminine self-fulfilment rather than a social function. The dominant discourse did not account for mothers’ budgeting, feeding, organizing, cleaning, earning money, and structuring neighbourhood and community life’ (p. 223).

During the inter-war years, attention to the psychological needs of the child had become part of the duties of mothers. Freud’s thinking reached Britain: women who did not find satisfaction in motherhood were in some way abnormal. Bowlby had also developed his theory of maternal deprivation: a child deprived of his or her mother would develop antisocial tendencies.15 But theories on child development contained conflicting views, particularly between Klein, Bowlby and Winnicott, as to the essential role of mothers. Winnicott (1953) for instance, while agreeing with the idea that maternal deprivation existed, criticized Bowlby for not taking into account the resources that normal children have to cope with the temporary loss of a parent. In Winnicott’s work and elsewhere, space existed within the psychoanalytic discourse for an argument against the need for full-time motherhood. However, this discourse did not prevail.

There is a widespread view in academic and, particularly, feminist literature that the psychoanalytic discourse homogeneously favoured full-time motherhood (Riley 1983 is an important exception). But this fails to consider the disagreements within psychoanalysis regarding the influence of mother-love, hatred, the possibilities of substitute mothering and the role of the environment. The dominance of one line of thought at any one time does not indicate that only one development was possible. Moreover, mothers have also been agents, albeit subordinate ones, in the dominance of the ideal of full-time motherhood. Nor did mothers accept all the prescriptions against maternal deprivation. Some reinterpreted them (Ross 1993; Tracey 1993); others followed the prescriptions of constant maternal care for just a short period until they became exhausted or gained confidence in their own ways of doing things (Humphries and Gordon 1993).

Women’s participation in the labour market was not supported by policies to ease the burden of domestic labour and child care. Summerfield (1984) argues that some policy makers believed that it was possible for women to combine some paid work with housewifery and motherhood with no detriment to standards. Lewis cites various government documents from 1947 to 1951 in which part-time work was thought of as an ideal means to ensure the worker-wife-mother role combination for women. Yet the central government soon closed down the nurseries opened during the war. Nationally there were only 700 nurseries in 1953 as compared to 4,000 in 1943 (Woman Health Officer, October 1953, in Tracey 1994:19).

Riley (1983) argues that the closure of nurseries following the Second World War was not mainly driven by theories of maternal deprivation but rather by central government’s attempts to transfer the cost of running nurseries to local authorities. But the financing of nurseries has not been a very popular matter with local tax payers.

Currently, the United Kingdom offers virtually no public provision for child care to employed parents.16 Employers offer very little support (Brannen and Moss 1991). Mothers rely mostly on their social networks, on their local child-minding system and on the private market. This makes flexible working hours a necessity. Many women work evening or night shifts when child care can be taken over by partners or relatives (Martin and Roberts 1984).

Many middle-class professional households prefer to employ nannies for child care. Gregson and Lowe (1994) associate the employment of nannies with an acceptance of a dominant ideology that children are best cared for in the home by a mother or maternal substitute, and to the needs for help beyond 9 a. m. to 5 p. m. that dualearner households have. However, cost may be a more important consideration, particularly when more than one child is cared for. A nanny’s pay is less than half one full-time private nursery school fee and her working hours are longer or more flexible than those of a nursery. While the job description for nannies (virtually all of whom are women) fits with the ‘nature’ of female jobs and the availability of women in the informal employment sector, the cost of private nurseries is a reflection of the scarcity of such provision and of the lack of public funding and support for child care.

For many lone-mother households, the non-availability of child care is particularly distressing, affecting the mother’s employment opportunities and earning capacity and offering no escape from full­time mothering (see Edwards and Duncan, Chapter 6 in this volume). As lone motherhood increases, tensions between the definitions of women as wives, mothers and workers increase (see Millar, Chapter 5 in this volume).

Are women currently more overburdened with child care than they were at the beginning of the century? As mothering was very different, so were children and what they required. Childhood, for instance, was a much shorter period of life; families had a different division of labour and living space. Comparability is therefore very difficult. However, no specific policies to ease the burden to mothers have been implemented in the twentieth century. Changes in mothering activities have basically happened as a consequence of technical improvements in housing, in science, and in other social constructions such as childhood itself (see Smart, Chapter 2 in this volume).

The experience of the transformation of mothering has been different for the working and the middle classes. But there are great continuities relating to both classes. While working-class mothers have done most of their own mothering, middle-class mothers have historically suffered from the disappearance of servants, and therefore have had to do the mothering that nurses and nannies had formerly been paid to do for them. Yet it is still the middle classes who can most afford to buy mothering. Substitute mothering can be hired for activities ranging from the bearing of offspring, wet – nursing,17 holding, touching, and body maintenance to psychological support and affection provision. Rothman (1994) refers to this process as the ‘commodification of children’ and the ‘proletarianization of motherhood’. These labels imply a deterioration of mothering. Regardless of whether this is correct, these practices involve only a small proportion of the experiences of mothering in contemporary societies.

Apart from the physical work of mothering, however, one has to consider the emotional job. It has been argued that this has been expanded (Ehrenreich and English 1979; Cowan 1983; Ferguson 1983; Rothman 1994). Ross (1986, 1993) has explored this thesis in her study of London’s working-class mothers at the turn of the century. She argues that it is wrong to imply that the ‘service aspects’ of mothering in the past meant that it involved less emotion than it does nowadays. Her thesis is that caring services to provide for sewing, cleaning, nursing and feeding carried in the past more emotional resonance than they do today. This argument raises the importance of an analysis of the transformation of caring for an assessment of the transformation of mothering.