In a study of Nazi concentration camp survivors, the people who experienced most problems recovering from their ordeal were those who were moved from camp to camp during and after the war; by comparison, those survivors who had stayed in one camp fared better (Davidson 1979). In Alameda County, California, people who in 1965 belonged to a church or temple were two and a half times more likely to be alive nine years later than people without such an affiliation (Berkman and Syme 1979). A 1960s study of mice revealed that mice placed in a conflict situation were more likely to develop hypertension when with strange mice rather than with their litter mates (Henry et al. 1967). An investigation of the growth of barley seeds demonstrated that seeds watered from a beaker held for fifteen minutes in a healer's hand grew faster and taller and had a greater yield than seeds not so watered (Grad 1963). One hundred and eight people in Czechoslovakia suffering from a variety of psychological and physical disorders were given thirteen types of non-active (placebo) pills over two years; 51 per cent got better; twenty-five patients experienced side-effects, including somnolence, insomnia, palpitations, irritability, low blood pressure, headache, diarrhoea and collapse; such side-effects were more likely when the pills were reddish-grey, or greyish-green-yellow, whereas yellow-white or plain white capsules were least likely to evoke unpleasant reactions (Honzak et al. 1972).

Among the social origins of the Social Support and Pregnancy Outcome study, as we saw in Chapter 1, was the idea that social support is good for health. What follows is not intended to be a comprehensive review. Such reviews exist (see e.g. Berkman 1984; Cohen and Syme 1985; Cohen and Wills 1985; Dean 1986; Gottlieb 1981a; Leavy 1983; Oakley 1985; Oakley 1988; Schaefer et al. 1981). Instead, this chapter is an attempt to draw together ideas, insights and problems from disparate areas of sociology, psychology, psychiatry, history, epidemiology and medicine to address the question: why, in the first place, should anyone suppose that social support can be helpful to childbearing women and their families?