one or more of their parents or siblings, even when separate rooms are available (Caudill and Plath, 1966; Konner and Worthman, 1980; Shweder et al., 1995). In two-thirds of the cultures surveyed in one international study, mothers routinely slept in the same bed with their infants, and the percentage was even higher when sleeping in the same room was included (Barry and Paxon, 1971; Burton and Whiting, 1961). This pattern was found not only in developing nations, but also in highly industrialized societies such as Japan, where children typically sleep with their parents until age 5 or 6 years (Caudill and Plath, 1966).
Despite the relatively unusual nature of typical U.S. sleeping practices compared with the rest of the world, there is also considerable subgroup variability within the country. In a Cleveland, Ohio, study, only 3 percent of babies in white, urban, middle-class, two-parent families slept in their parents' bedroom under one year of age, and only 1 percent did so in the second year. In contrast, parent-child cosleeping was reported for black children in the same urban area at a rate of 55 percent for children in the first year and 25 percent between 1 and 5 years of age (Litt, 1981). A subsequent study in a predominantly white, blue-collar community in Appalachian Kentucky found cosleeping among 71 percent of children between 2 months and 2 years of age, and 47 percent between 2 and 4 years (Abbott, 1992). Lozoff et al. (1984) found that babies in white, middleclass families are more likely to sleep with their parents when they are ill, when the family moves, or when there is marital conflict.
Both cosleeping with a parent or sleeping alone appear to be adaptive in a variety of cultural contexts. For example, !Kung children have been observed to be more independent than their counterparts in the United States, notwithstanding their early cosleeping experiences (Klein, 1995; Konner, 1982). It is of interest to note, however, that parental concerns about sleep problems in young children are common in the United States, less frequent in Japan (Nugent, 1994), and nonexistent in Kenya (Super and Harkness, 1982). It is not clear whether these differences reflect parent perceptions or actual sleep disturbances. Moreover, although the Japanese mother-child relationship remains relatively strong into adulthood, the husband-wife bond is typically less close than in the United States (Lebra, 1994). In short, cultural differences in early childhood sleeping arrangements are neither better nor worse; they simply reflect contrasting preferences and differential trade-offs.
Differences in early caregiver-child interaction patterns and communication styles further illustrate alternative childrearing strategies, as well as the futility of searching for universally normative or optimal practices. Mothers of Gusii toddlers in Africa and Zinacantecan toddlers in Central America, who use a great deal of imperatives when they speak to their offspring, generally have children who grow up to be relatively obedient