Tasja Sophie Mueller

In this dissertation we aim to examine when third-party observers support violent retaliatory action on behalf of someone else, when they withdraw support, and when they reinterpret violence as justified or motivated by seemingly less harmful motives such as protection or helping. We aim at addressing a central theoretical puzzle: why do individuals who generally reject harm nevertheless endorse violence in some contexts, but withdraw support in others? While classic strategic models in international relations, such as deterrence theory and spiral conflict models, explain violent escalation at the level of states, this dissertation shifts the focus to the psychological judgments of third-party observers of the public. Across three empirical Chapters and a general theoretical integration, we advance the central claim that third-party responses to violence are best understood as reciprocity-guided social judgments, rather than as fixed attitudes toward violence or as simple expressions of generalized aggression. At the same time, we examine how motivations that are typically associated with prosociality, such as empathy and concern for others, intersect with these judgments and may shape how violence is interpreted. Violence occupies a paradoxical position in social life. It is widely condemned as immoral and socially destructive, yet in some contexts it can also be framed as protection or defense (e.g., Fotion, 2004; Ginges, 2019; Orford, 2024). We argue that this tension cannot be fully captured by explanations that focus mainly on trait aggression, hostility, or dehumanization (e.g., Anderson & Bushman, 2002; Haslam & Stratemeyer, 2016). Instead, we conceptualize third-party evaluations of violence as dynamic judgments structured by normative expectations, especially the norm of reciprocity and its implicit standard of proportionality (Gouldner, 1960; Fehr & Gächter, 2000; Gollwitzer et al., 2024). Observers assess whether violent action appears warranted in light of prior harm, whether it remains directed at the original perpetrator, and by that whether it affirms or violates shared norms. In this sense, support for violence is treated as an ordinary, norm-governed form of social judgment. This approach also allows us to examine whether prosocial orientations, such as empathy, agreeableness, and prosocial behavioral intentions, necessarily inhibit support for violence, or whether they may under certain normative conditions coexist with or shape third-party support for violent action framed as helping. In Chapter 2, we investigate third-party support for lethal military intervention in the context of the Russia-Ukraine war. Drawing on reciprocity theory and research on relational closeness, we test whether support for lethal military action by one’s own country on behalf of Ukraine increases when violent action can be construed as a norm-affirming response to severe wrongdoing (Gouldner, 1960; Fehr & Gächter, 2000). Across samples from Germany, the United States, and Poland, support for lethal military action increased stepwise as scenarios reflected increasingly severe Russian escalation. Relational closeness to Ukrainians reliably predicted higher support across escalation levels. By contrast, dispositional prosocial traits such as empathy, agreeableness, and prosocial behavioral intentions did not emerge as robust independent predictors. These findings suggest that third-party support for lethal military action is shaped primarily by situational escalation cues and perceived relational obligation, rather than by generalized prosocial dispositions. In Chapter 3, we test the complementary prediction that reciprocity norms not only legitimize retaliation but also constrain it. Using escalation scenarios in the context of the 2023 Israel-Hamas war, we examine whether support declines when retaliation becomes displaced from the original perpetrator, either through horizontal escalation or vertical escalation (Pabón & Duyvesteyn, 2023). Support for lethal military intervention declined as in scenarios retaliatory action became increasingly displaced, indicating that proportionality functions as a boundary condition for third-party endorsement. Relational closeness to Israelis predicted higher support when retaliation remained directed at Hamas, but it did not shield from normative backlash once actions were perceived as more and more displaced. These findings show that reciprocity-based expectations can both facilitate and constrain third-party support for violence. In Chapter 4, we extend our analysis to an act of extreme individual violence, namely the assassination of a health insurance executive in the United States. Here we distinguish between evaluations of the violent act itself and inferences about the perpetrator’s motives. Building on attribution theory (Weiner, 1995; Hewstone, 1990) and prior findings on diffusion of responsibility (Chapter 3 covariate), we examine whether observers attribute prosocial intent to the perpetrator when responsibility is partially shifted from the individual to broader systemic forces. Very few participants judged the act itself as helpful or prosocial. Nevertheless, a subset of observers were more likely to infer prosocial motives of the perpetrator when responsibility was perceived as diffused to external forces. Some limited evidence also pointed at the potential for empathy to have the same effects on judging the act as prosocially motivated. These findings indicate that third-party evaluations of violence are multidimensional: observers may reject an act as impermissible while simultaneously construing the actor’s motives in partially prosocial terms. Across chapters, we show that third-party responses to violence vary systematically with situational escalation, relational closeness, attributional processes and limited also individual prosocial orientations. Reciprocity-based expectations structure judgments of whether violent action appears warranted or disproportional, while relational closeness increases perceived obligation and shapes sensitivity to norm violations. Attributional processes such as diffusion of responsibility further shape how violent acts are interpreted. In the General Discussion, we integrate these findings into a norm-based social psychological account of third-party violence evaluation. We argue that third party support for violence is a dynamic and reversible process that updates in response to changes in perceived proportionality and responsibility. By focusing on the evaluative processes through which violence is interpreted as legitimate or misdirected, this dissertation contributes to social psychological theories of moral judgment, reciprocity, and prosociality, and clarifies the conditions under which violence may be construed as protection or helping, as well as the boundaries at which such interpretations erode.