lar categories of relationship are crucial to well-being. Our most recent study of mothers of infants suggests that maternal postpartum affect is related to support from and satisfaction with the person designated by the mother as closest to her, regardless of the individual's role category ( Coffman et al., 1989). What is suggested is that personal well-being across the life span, as well as in infancy, is dependent on the presence of at least one close relationship.
This hypothesis is difficult to test empirically, because individuals devoid of close ties are quite rare in the general population. In the national sample obtained by Kahn and Antonucci, only six percent of the respondents lacked an inner circle relationship. In our high-risk sample of South Beach residents, however, fifteen percent were found to have no one in the inner circle, allowing a comparison of individuals lacking close ties with those reporting only one versus those reporting more than one such tie. The affect of those lacking a close relationship was found to be significantly more negative than those with one such relationship, and the latter did not differ from those with more than one close relationship ( Levitt et al., 1987). These results suggest that the presence of one close relationship may be sufficient to foster well-being, consistent with Bronfenbrenner's ( 1979) conclusions about the ameliorative effects of a close relationship for high-risk children.
A primary thrust of this discussion has been to suggest that close relationships across the life span can be conceptualized as attachment relationships; that they are governed by similar processes and serve similar functions. Related views have been forwarded recently by Ainsworth ( 1989) and Shaver and Hazan