Emotions, Attributions, and Policy Endorsement in Response to the September 11th Terrorist Attacks EMOTIONS, ATTRIBUTIONS, AND POLICY ENDORSEMENT SADLER ET AL. Melody S. Sadler, Megan Lineberger, Joshua Correll, and Bernadette Park University of Colorado U.S. citizens’ reactions to the terrorist attacks of September 11, 2001, were examined, includ- ing emotional responses, attributions for the cause of the attacks, and policy recommendations. Participants whose emotional reactions were dominated by anger attributed the attacks to the fanaticism of the terrorists and to poor U.S. security and rejected the idea that U.S. foreign pol- icy played a role in the attacks. They endorsed an aggressive military response and rejected hu- manitarian efforts. Participants whose emotional reactions were dominated by sadness, in con- trast, denied that fanaticism and security lapses were to blame, and both sad and fearful participants expressed reservations about a strong military reaction. Path models suggest that the relationship between emotional reactions and policy endorsement was partially mediated by attributions. In the wake of the September 11, 2001, terrorist attacks on the World Trade Center and the Pentagon, Americans ap- peared overwhelmed by emotions, by a need to understand what had happened, and by a struggle to decide how to re- spond. In observing both public figures and the average citi- zen, the relations among these three constructs seemed not to be haphazard but rather systematic and predictable. Individ- uals who primarily experienced anger seemed confident that terrorist extremists were responsible for the attacks and that the United States should react to the unprecedented trans- gression with intense military action. In contrast, individuals overwhelmed with sadness and fear seemed reluctant to pin- point a simple cause, much less an immediate course of ac- tion, although these individuals did not appear to favor a mili- tary response. The relationships among the emotions people experienced in reaction to the attacks, their opinions of who was responsi- ble, and how the United States should respond are consistent with empirical research in psychology focused on how mood and emotion influence judgments. Cognitive appraisal theo- ries of emotion posit that emotions differ from one another on the basis of the assessments perceivers make of a situation (Ar- nold, 1960; Lazarus, Averill, & Opton, 1970). Although early cognitive appraisal theories proposed a unidirectional rela- tionship in which the unique appraisals of a situation cause the specific emotion experienced (Arnold, 1960; Lazarus, 1991), recent research suggests that cognitive appraisals can be a con- sequence as well as a cause of emotional experience (Mikula, Scherer, & Athenstaedt, 1998; Roseman, 1991; Smith & Laza- rus, 1993). The primary appraisals differentiating anger, sadness, and fear involve assessments of control, responsibility, and cer- tainty (Smith & Ellsworth, 1985; Smith & Lazarus, 1990). In this research, angry respondents recalling past events tend to believe responsibility for the event belonged to an individual other than the self and to feel certain about what had hap- pened and what would happen. In contrast, sad and fearful re- spondents tended to focus on aspects of the situation as re- sponsible for producing an event and to feel uncertain they comprehended what transpired. Consistent with differences in the appraisals of responsibility and control, Keltner, Ellsworth, and Edwards (1993) presented evidence that an- gry perceivers are more likely to make dispositional attribu- tions to the actor in an ambiguously caused event, whereas sad and fearful perceivers make more external or situational attributions for the event (see also Lerner & Keltner, 2000, 2001; Niedenthal & Setterlund, 1994; Siemer, 2001; Tiedens & Linton, 2001). People experiencing anger are more likely than people ex- periencing sadness or fear to judge a defendant guilty of an offense and to seek stiffer penalties. Bodenhausen, Sheppard, and Kramer (1994) showed that angry perceivers were more likely than sad or neutral perceivers to find stereotypic de- fenders guilty of an alleged crime. Similarly, Lerner, Goldberg, and Tetlock (1998) found that angry participants were more likely than neutral participants to think an individ- ual’s actions led to another’s injuries and to call for punish- ment of the offender. The effect of anger on punitiveness can BASIC AND APPLIED SOCIAL PSYCHOLOGY, 27(3), 249–258 Copyright © 2005, Lawrence Erlbaum Associates, Inc. Requests for reprints should be sent to Melody Sadler, Department of Psychology, University of Colorado, Boulder, CO 80309–0345. E-mail: msadler@psych.colorado.edu