Research addressing the social causation of alcohol abuse has alluded to the tension-management or self-medication function of alcohol consumption. The use of alcohol (as well as other substances) can constitute a means to escape from, avoid, or otherwise regulate unpleasant emotions derived from undesirable social experiences (Cooper, Frone, Russell, Mudar,1995; Cooper, Russell, and George, 1988; Pohorecky, 1991; Wilsnack, 1992). This paper addresses the self-medication of distress engendered by relatively understudied forms of chronic stressors; negative interpersonal experiences in the that potentially may be defined as workplace harassment. We first present a critical review of the literature on contributors to the use of alcohol and drugs, arguing that future research should focus greater attention on interpersonal rather than task-related psychosocial stressors. We then present a conceptual framework for initiating this type of research. This framework focuses on the role of societal social constructions, interpersonal support resources, and attributional processes, which are hypothesized to function as mediators between harassment experiences and the use of substances to self-medicate distress. Last, we present case studies illustrating this model and depict an ongoing study designed to test the proposed model empirically. Workplace stressors as etiologic determinants of drinking behaviors Studies embodying the stress and alienation perspectives on alcohol consumption have examined particular attributes of work that may influence drinking for stress-reduction purposes (Fennell, Rodin and Kantor, 1981; Mandell, Eaton, Anthony and Garrison, 1992; Sonnenstuhl and Trice, 1991). Occupational conditions encompassing heavy work demands but limited control over work (Karasek, Baber, Marxer, Ahlbom and Theorell, 1981; Karasek and Theorell, 1990); Crum, Muntaner, Eaton and Anthony, 1995) or job content that is repetitive and monotonous in nature (Parker and Farmer, 1988, 1990) have received major attention. However, many studies have found limited empirical evidence for a link between stressful work conditions and drinking outcomes (Cooper, Russell and George, 1988; Mensch and Kandel, 1988; Seeman, Seeman and Budros, 1988). Other studies have found that heavy work demands coupled with little control over work (Crum, Muntaner, Eaton and Anthony, 1995) or unchallenging work (Lennon, 1987) are predictive of alcohol outcomes in males but not in females. Some researchers have therefore concluded that personality characteristics brought into work environments or problematic social experiences occurring outside of the play more salient roles in the psychosocial etiology of alcohol abuse than do stressful work conditions (Seeman, Seeman and Budros, 1988). Other researchers have suggested that occupational stress constructs have failed to tap all of the salient domains of stressors that affect female as well as male workers (Richman, Flaherty and Rospenda, 1996). Harassment as an additional domain of stress Variability in the extent to which lack of control over work is experienced as stressful may depend on the source and perceived meaning of the lack of control. Some task requirements, such as the pressure of meeting deadlines, may be viewed as inherent to the nature of given occupations. By contrast, lack of control over work linked to harassing or abusive interpersonal interactions rather than to task-related demands may have very different meanings and implications for perceived stress and the self-medication of related distress by workers in a variety of occupational statuses (Richman, Flaherty and Rospenda, 1996). In particular, abusive treatment by an employee's supervisor (or someone else who can affect an employee's job) might have greater deleterious consequences by virtue of engendering particular negative emotional reactions. Research conducted within one occupation (physicians surveyed from medical school entrance through the internship year of training) demonstrated significant links between negative interpersonal aspects of work and deleterious alcohol outcomes (Richman, Flaherty, Rospenda and Christensen, 1992; Richman, Flaherty and Rospenda, 1996). …