Back to XIV Abstract Summary

Back to Symposium Presentations Summary

S-2.-Social representations and beliefs about aggression

Organizer:Archer, John
Department of Psychology, University of Central Lancashire, Preston, Lancashire, UK.

Symposium Abstract

Since Campbell & Muncer (1987) introduced the concept of  social representations of aggression, there have been a number of studies documenting instrumental and expressive beliefs or social representations that people hold, principally about their own aggression. Women tend to hold more expressive and men more instrumental beliefs, although in the case of instrumental beliefs this finding only holds for same-sex physical aggression. The four papers in this symposium illustrate the different ways in which social representations and beliefs have been applied to more general issues concerning aggression in recent studies. They involve the following topics: their relation to classic measures of attitudes and to normative beliefs; whether instrumental beliefs apply to the supposed upsurge in masculine behaviour by young women; investigations of the dynamic nature of  “social representations” of aggression; the relation between beliefs about aggression and the extent of physical aggression to partners.

Reference: Campbell, A. & Muncer, S. (1987). Models of anger and aggression in the social talk of women and men. Journal of the Theory for Social Behavior, 17, 489-511.


Archer, J.
Department of Psychology, University of Central Lancashire, Preston, Lancashire, UK.

The association between instrumental (I) and expressive (E) beliefs about aggression and (1) attitudes to aggression (measured by semantic differentials), and (2) normative beliefs about aggression, was investigated among a sample of 165 British students, aged between 18 and 52 years. I and E beliefs were only weakly associated with these attitudinal measures. The extent to which they predicted self-reported aggression was also assessed, showing that physical aggression was predicted mainly by instrumental beliefs about aggression. The strongest predictors of verbal aggression were higher instrumental and lower expressive beliefs. Anger was associated with instrumental beliefs. Hostility was predicted by holding both instrumental and expressive beliefs. Men and women showed some differences in the attitudinal predictors of self-reported aggression. Men also showed more instrumental and less expressive beliefs than women, they viewed aggression and fighting more positively, and reported more physical aggression. All aggression measures showed a slight decline with age, and men viewed aggression as more negative, less potent, less instrumental, and more expressive with age.


Muncer, S. and Campbell, A.
Department of Psychology, University of Durham, Durham, UK.

In 1997, women committed only 17% of all violent crime in the United Kingdom. However, in the ten year period between 1984 and 1994, arrests for violent crime (as a percentage of all female arrests of young women under 21) rose from 11.2% to 20.1%. The rise in youthful female violence gave rise to a flurry of media speculation that "laddism" (the adoption of anti-social male attitudes) was to blame. This kind of attribution is not new--twenty years earlier in the United States a similar outcry about the “masculinising" effects of women's economic and social "liberation" was also prominent. Despite this, little research has empirically addressed this relationship. We developed a scale designed to measure endorsement of "laddish" behaviour by women.  We predicted that laddism among females would be positively associated with higher self-reported aggression and with the holding of a more instrumental representation of aggression. In males, we predicted that laddism (support for women's masculine behaviour) would be negatively correlated with self-reported aggression (since previous literature suggests that "macho" attitudes supporting violence are associated with a traditional stance toward the female role) and negatively correlated with instrumental orientations to aggression. The resulting data will be presented and discussed.


Richardson, D. S. and Latané, B.
Department of Psychology, Florida Atlantic University, Boca Raton, Florida, USA. .

This paper will review evidence of the existence of social representations of aggression and then focus on the group-level processes by which such representations are likely to develop. The existence of social representations (SRs) of aggression suggests that individuals’ concepts of aggression develop through social influence processes and that ideas about aggression vary as a function of group membership. Supporting the notion that representations differ by group membership, Richardson and Huguet (1999) found that instrumental social representations of aggression were more strong and more organised among males and disadvantaged young people who had more experience with aggression, either as aggressors or as victims, than among college students.  Dynamic Social Impact Theory (DSIT) predicts group-level consequences of ongoing communication among individuals whose propinquity affords reciprocal interactions (Latané, 1996; Latané, et al., 1994). As the minority is exposed to contrary pressures, the distribution of opinions among neighbours in social space becomes less varied and more organised through consolidation and clustering. As a social system develops such organisation, people are more likely to perceive regularities and generate SRs that in turn promote higher levels of organisation. Walker (1999) examined the development of social representations of direct and indirect aggression, arguing that indirect aggression, because the concept is relatively “unorganised” among laypeople, should be a fertile topic for examination of the development of social representations related to aggression.  Consistent with the predictions of DSIT, exchanging messages in electronic space about indirect aggression resulted in clustering and consolidation of opinions about the aggressiveness of such behaviour, suggesting the development of social representations. Through interpersonal communication and the principles of dynamic social influence, indirect aggression is given meaning and is transmitted over time to people who share physical-space.  Extending the study of such processes to face-to-face communication, Richardson and colleagues (1999) also found evidence of regional clusters of attitudes about aggression among individuals who engaged discussion about justifications for aggressive behaviour.



Graham-Kevan, N. and Archer, J.
Department of Psychology, University of Central Lancashire, Preston, Lancashire, UK.

This study concerns the association between beliefs about aggression and measures of physical aggression and injuries sustained by partners. From a sample comprising students (N=113), women from a domestic violence refuge (N=44), and male prisoners (N=108), those reporting that they had used physical force at some time completed the following: an adapted version of the EXPAGG (Archer and Haigh, 1997), the Conflict Tactics Scale, CTS, (Straus, 1979) for themselves and their partner, and items regarding fear experienced by themselves during conflicts and injuries sustained to both themselves and their partner. Over all these respondents (N=113), the instrumental scale was significantly positively correlated with self-reported use of physical aggression,  and six of the eight individual acts which comprise the physical aggression scale of the CTS. Injuries to partners were positively correlated with instrumental beliefs. Correlations for the students (N=38) and prisoner (N=46) samples were similar to the whole sample, but there were differences for the refuge sample. Regression analysis showed that, overall, instrumental beliefs predicted participants hitting their partners, and injuries sustained by partners. The expressive scale showed lower but significant positive correlations with only one item of physical aggression, that of beating one’s partner. The male prisoner sample showed positive correlations between the expressive scale and self-reported physical aggression, both minor and severe, and the use of four of the eight individual acts. Regression analysis revealed the holding of expressive beliefs about aggression predicted only beating one’s partner for the aggregated sample.