On the Variability of Anger Cross-Culturally: An Assessment of General Strain Theory's Primary Mediator

Abstract

This article assesses general strain theory's (GST) primary mediator, anger, as a process that exhibits important variability across cultures. It presents data from structured interviews and fieldwork in India and the United States that suggest variations in the understanding and experience of anger across three samples: Americans, lay Tibetans, and Tibetan Buddhist clergy. Findings suggest cultural differences in normative social approval of anger, perceived effects for self of becoming angry, reaction tendencies, and emotional memory. Future research should test and map variations in anger across and within populations and explore implications for macro-level GST in explaining cross-cultural differences in crime.

Acknowledgments

Support for this research was provided by the US Fulbright Foundation, the Woodrow Wilson Foundation, the University of Chicago's Committee on South Asian Studies, and the College of Arts and Sciences of Seattle University.

Notes

¹Under the terms of a body of emerging emotion research called cognitive appraisal theory, one way of linking dispositional/trait-based anger to heightened vulnerability to state anger is to note that individuals show a systematic bias toward making certain types of attributions across a range of situations. Dodge (1993) previously found that aggressive children tend to appraise other children's actions to be intentional and undertaken with a desire to cause harm in ambiguous situations. The aggressive child's habitual appraisals (i.e., “he's trying to humiliate me”) trigger state anger, with aggression ensuing (Dodge 1993).

²Much like this section focuses on the distinction between dispositional and situational negative emotions within the context of GST, recent research has extended this line of exploration into low self-control with substantively similar results (Mischel et al. 1989; Piquero and Bouffard 2007).

³Almost all respondents were willing and able to recall such a situation. Occasionally, elicitation required patience and a repetition of the prompt. Several of the findings described hereafter involved analyzing the characteristics of recalled Tibetan and American anger incidents having parsed them into several component parts.

For main test, ***p < .001.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .001 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

For main test statistics, **p < .01, ***p < .001.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .01 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

⁴Because individual respondents could give answers that reflected more than one derived response category, the total percentages in the vertical column for a single sample group can total more than 100%. The n reported in each group is the number of individuals who offered valid responses to a given question. The n values may vary considerably from question to question, sometimes because some individuals chose not to answer it, and on certain questions, only a subset of respondents from each group received the follow-up question. In this case, only individuals who said they thought anger was a bad or ambiguous thing and who then went on to offer an open-ended justification for that view in response to the follow-up question are reflected in the group's n.

⁵These doctrinal judgments “rlung langs is a sin/root afflictive state/non-virtuous act …” are then also connected causally with harms that will accrue to the person who commits them.

For main test, *p < .05, **p < .01, ***p < .001.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .001 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

For main test statistics, *p < .05, ***p < .001.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .05 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

⁶This category is an interesting one to compare across the two cultures. The Tibetan language lexicalizes a concept quite similar to “justice” (dang bden), but Tibetans do not typically invoke this concept in the context of offering justifications for anger or for harm toward others, either on the personal or the political level. Instead, much of the work that American concepts of “justice” and “injustice” perform, expressing a judgment of the extent to which particular events or acts are morally sanctionable, tends instead in Tibetan to be addressed through the rhetorical idiom of truth (bden pa). Instead of invoking the need for a remedy to injustice, Tibetans will often assert the need to demonstrate or make the truth of a particular matter known, with procedures for restitution and reconciliation then to follow.

For main test statistics, *p < .05.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .05 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

For main test, ***p < .001.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .001 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

For main test statistics, ***p < .001.

Across a given row, groups whose % values are marked with different subscripts differ significantly from one another at p < .001 in paired comparison tests. Thus, if the frequency statistics of the three groups are marked “a,” “b,” “b,” then group 1 differed significantly from group 2, and group 1 differed significantly from group 3, but group 2 and group 3 did not differ significantly.

⁷The Tibetan settlements of India are governed under an arrangement of limited autonomy. While ultimately remaining under the formal legal jurisdiction of the Government of India (GOI), informally, the Central Tibetan Administration's Settlement Offices respond to many crimes and offenses that occur within the settlements. In keeping with traditional Tibetan social moirés (French 1995), the process of remediation and/punishment is often informal and is kept off the GOI record. In interviews, many Tibetans expressed limited confidence in the even-handedness of the GOI law enforcement and judicial process. In assessing the usefulness of criminological data from Tibet proper, judicial process inside the Tibetan Autonomous Region continues to be characterized as intensely politicized (see Human Rights Watch 2010), with charges, for instance, that the PRC confounds nonviolent political dissent with violent dissent.

Additional information

Notes on contributors

Randall Horton

RANDALL HORTON is an Assistant Professor in the Department of Psychology at Seattle University. His research includes studies of the cultural psychology of emotion, the psychology of ethno-political violence, and studies of the clinical impact of violence in the lives of immigrants and refugees. He has been a fellow of the US Fulbright Foundation, the Woodrow Wilson Foundation, and the Social Science Research Council.

Stephen K. Rice

STEPHEN K. RICE is an Assistant Professor in the Department of Criminal Justice at Seattle University. His research focuses on deference and defiance in individuals' interactions with the justice system (e.g., procedural and restorative justice, racial profiling, radicalization, final statements of the condemned and criminological theories that inform these processes). His publications have appeared in outlets to include Criminology, Justice Quarterly, the Journal of Quantitative Criminology, and NYU Press.

Nicole Leeper Piquero

NICOLE LEEPER PIQUERO is a Professor in the Criminology Program at the University of Texas at Dallas and currently serves as a Trustee-at-Large on the ACJS Executive Board. Her research interests include white-collar and corporate crime, criminological theory, as well as gender and crime.

Alex R. Piquero

ALEX R. PIQUERO is Ashbel Smith Professor of Criminology at the University of Texas at Dallas, Adjunct Professor Key Centre for Ethics, Law, Justice, and Governance, Griffith University Australia, and Co-Editor, Journal of Quantitative Criminology. His research interests include criminological theory, criminal careers, and quantitative research methods. He has received several research, teaching, and mentoring awards.