Family systems models have often been invoked in order to understand a variety of adjustment problems (Bowen, 1960; Minuchin, 1974), such as anorexia (Minuchin, Rosman, & Baker, 1978), and adolescent substance abuse (e.g., Pardeck, 1991). A significant body of empirical literature suggests that suicidal behavior in the teenage years is associated with family processes (Gispert, Wheeler, Marsh, & Davis, 1985; Herman & Hirshman, 1981; Humphrey, French, Niswander, & Casey, 1974; Kosby, Silburn, & Zubrick, 1986; Kreider & Motto, 1974; Molin, 1986; Paluszny, Davenport, & Kim, 1991; Pfeffer, Adams, Weiner, & Rosenberg, 1989; Rosenkrantz, 1978; Schaffer, 1974; Smith & Smith, 1976; Wenz, 1979; William & Lyons, 1976; Wright, 1982), but the question of whether the family systems framework provides a plausible explanation of teenage suicidal behavior has not been explored in great depth.
Resolving this question is of interest because it would enable us to specify how suicidal behavior in adolescence is linked to family contingencies. Moreover, it would enable is to further specify the range of applicability of family systems models. This paper seeks to determine whether existing hypotheses in the family systems literature on the relationship between family processes and mental health in general may be applicable to the suicidal behavior of teenagers in particular.
This analysis focuses on two aspects of the family that have been identified in the literature as dysfunctional processes and which have been linked to the adjustment problems of family members. First, Minuchin and Fishman's (1981) contention that the transgression of boundaries in the family has strong affective responses such as quilt and anxiety is considered. The question to be explored is whether teenage suicidal behavior can also be seen as such an affective response. Second, the applicability of Bateson, Jackson, Haley and Weakland's (1956) double bind hypothesis to teenage suicidal behavior is explored to assess whether teenage suicide can be seen as an indication that the youngster is caught up in contradictory interactive patterns in the family.
To address these questions, the connections between the two concepts within the family systems framework are assessed, and the case of one youngster's suicide attempts are discussed and placed in context to illustrate how double bind contingencies and boundary transgression may contribute to our understanding of how suicidal behavior occurs in adolescence.
Functional and Dysfunctional Family Processes
A systems approach assumes that within any given social unit, relatively autonomous subsystems are formed to optimize adaptive capacity of a larger system (Parsons, Bales, & Shils, 1953; Minuchin, 1974; Minuchin & Fishman, 1981). In families, subsystems typically are formed which define the family structure according to gender (e.g., an alliance which includes grandfather, father, and son), generation (grandparents vs. parents vs. children), and kinship (related vs. not related) (Minuchin & Fishman, 1981, Chap. 2). Family systems can be seen, then, as an overlapping set of alliances which each include certain individual members, and exclude certain others. These subsystems delineate the behavior of particular family members toward particular other family members. For example, a parent and child do not interact in the same way as a husband and wife. Patterns of inclusion and exclusion in the family do not exist in their own right, however, but are actively reinforced by the members of the family system through interaction; the existence of a spousal subsystem, for example, depends on whether two people in a family system actually behave as such. In part, then, the behavior of individuals is assumed to be determined by the structure of the family, while this structure, in turn, is partly determined by the behavior of the individuals who are part of it (Koopmans, 1993). …