Multiple sets of empirical research findings on guilt are reviewed to evaluate the view that guilt should be understood as an essentially social phenomenon that happens between people as much as it happens inside them. Guilt appears to arise from interpersonal transactions (including transgressions and positive inequities) and to vary significantly with the interpersonal context. In particular, guilt patterns appear to be strongest, most common, and most consistent in the context of communal relationships, which are characterized by expectations of mutual concern. Guilt serves various relationship-enhancing functions, including motivating people to treat partners well and avoid transgressions, minimizing inequities and enabling less powerful partners to get their way, and redistributing emotional distress.

Subjects furnished autobiographical accounts of being angered (victim narratives) and of angering someone else (perpetrator narratives). The provoking behavior was generally portrayed by the perpetrator as meaningful and comprehensible, whereas the victim tended to depict it as arbitrary, gratuitous, or incomprehensible. Victim accounts portrayed the incident in a long-term context that carried lasting implications, especially of continuing harm, loss, and grievance. Perpetrator accounts tended to cast the incident as a closed, isolated incident that did not have lasting implications. Several findings fit a hypothesis that interpersonal conflicts may arise when a victim initially stifles anger and then finally responds to an accumulated series of provocations, whereas the perpetrator perceives only the single incident and regards the angry response as an unjustified overreaction. Victim and perpetrator roles are associated with different subjective interpretations.

Unreciprocated romantic attraction was explored by comparing narrative accounts. Unrequited love emerged as a bilaterally distressing experience marked by mutual incomprehension and emotional interdependence. Would-be lovers looked back with both positive and intensely negative emotions, whereas rejectors were more uniformly negative in their accounts. Unlike rejectors, would-be lovers believed that the attraction had been mutual, that they had been led on, and that the rejection had never been communicated definitely. Rejectors depicted themselves as morally innocent but still felt guilty about hurting someone; many rejectors depicted the would-be lover's persistent efforts as intrusive and annoying. Rejectors constructed accounts to reduce guilt, whereas disappointed lovers constructed them to rebuild self-esteem. Rejectors saw would-be lovers as self-deceptive and unreasonable; would-be lovers saw rejectors as inconsistent and mysterious.

Although most interpersonal interactions take place between people who know each other, most selfpresentation research has focused on self-presentation to strangers. Five studies showed that selfpresentational favorability differed as a function of whether the interaction partner was a friend or a stranger. Studies 1 and 2 found that self-presentations to friends were consistently more modest than self-presentations to strangers. In Studies 3 and 4, self-presentations were manipulated by instructing participants to present themselves in either a self-enhancing or modest manner. Modesty with strangers and self-enhancement with friends both resulted in impaired recall for the interaction, consistent with the view that those strategies contradict familiar, overlearned patterns. Study 5 distinguished self-deprecation from modesty. Taken together, the results indicate that people habitually use different self-presentation strategies with different audiences, relying on favorable self-enhancement with strangers but shifting toward modesty when among friends.Self-presentation is one of the fundamental and important processes by which people negotiate identities for themselves in their social worlds. In the privacy of one's own mind, perhaps, one may be relatively free to imagine oneself having any sort of identity, but serious identity claims generally require social validation by other people, and so the construction of identity requires persuading others to see one as having desired traits and qualities (e.g., Baumeister, 1982;Schlenker, 1980;Wicklund & Gollwitzer, 1982).The importance of self-presentation is reflected in how pervasive the motivations are. It appears that there is a strong and pervasive desire to make a positive impression on others (e.g., Jones & Wortman, 1973; see also Leary & Kowalski, 1990;Sedikides, 1993). However, how one creates a positive impression may vary depending on the audience one is trying to impress. As Tedeschi (1986) asserted, the distinction between behavior that occurs in intimate relationships and behavior that occurs in impersonal interactions has been largely neglected by selfpresentation researchers. The purpose of the present studies was to demonstrate that self-presentation differs for these different audiences. Specifically, the studies attempted to demonstrate that people self-present in a much more self-promoting manner with strangers and in a more modest manner with friends and acquaintances. Moreover, we suggest that in terms of Paulhus's

No abstract