According To Social Psychologists, How Do Victim, Offender And Th

  • Category: Psychology
  • Words: 4106
  • Grade: 90

During the late 1940s, Sutherland (1947) advanced that explanations of crime and deviance are of either a situational or a dispositional nature. Additionally, he argued that of the two explanations, situational ones might be of the most importance. Hirschi & Gottfredson (1986) made a critical distinction in light of this issue, the distinction was between the terms crime and criminality. Crime, they proposed refers to "˜events that presuppose a set of necessary conditions'. Criminality on the other hand refers to "˜stable differences across individuals in the propensity to commit criminal acts' (Hirschi & Gottfredson, 1986: 58). They went on to point out that criminality is necessary, but is not a sufficient condition for crime to occur, since crime requires important situational inducements.

Despite these propositions, social psychologists in the following decades tended to focus on dispositional theories of crime and deviance, that is, focusing on individual differences. There is a wealth of literature focusing on motivations and characteristics of criminal offenders (e.g. Cohen, 1955,as cited in Birkbeck & LaFree, 1993; Cloward & Ohlin, 1960), and a modest amount attending to the victims of crime (Cohen, Kleugel, & Land, 1981). However the suggestion is well documented
(e.g. Hepburn, 1973; Athens, 1985; Luckenbill, 1977) that there is a need for research to focus on the sequential development and interactional dynamics of criminally violent situations. This is based on the notion that violence is, at least in part, situationally determined (Felson & Steadman, 1983). Symbolic interactionism is such a guiding approach in this field, so it is important to clarify what sets it apart from others in the area; there are two main important such points. Firstly , social interactionist theory focuses on the objective fact of situations (as overlooked by criminologists), and secondly their subjective definition by actors (as overlooked by both opportunity and experimental psychologists).

It was Goffman (1967) who set the ball rolling as it were for symbolic interactionism. He uniquely emphasized the nature of the violent criminal act as important, instead of just the criminal actor. It was his notion of a "˜character contest' that inadvertently proposed one of the first violent criminal behaviour theories of its kind. An individual is said to demonstrate "˜strong' character when they stand "˜correct' and "˜steady' in the face of adversity, and "˜weak' character when he or she does not. He posited that a character contest is a special form of interpersonal action, since the confrontation involves (usually) both parties linked together in the common pursuit of demonstrating whose respective character is the strongest. Whoever wins the contest maintains honour and face and therefore positive self-image of self, whilst the weak character loses honour and face and is cast into a negative self-identity.
Luckenbill (1977) examined "˜situated transactions' as set out by Goffman (1967) in an attempt to specify the processes involved. Luckenbill (1977) examined 70 transactions leading to criminal homicide; he reconstructed these transactions using all attainable accounts as resources. The analysis revealed that the transactions took a processual form consisting of six stages. The fourth of these stages involves a "˜working agreement' being forged between the major participants, favouring violent means by which to settle the dispute. However since Goffman's (1967) theory and Luckenbill's (1977) follow-up work provided such a major source of ideas for explaining interpersonal violence, it has been subject to much critical examination. The notion of the "˜working agreement' has proved the most controversial since participants in violent criminal situations do not typically agree that violence should be used in the resolution of a dispute (Athens, 1985). Even Luckenbill (1977) himself found that 14% of the offenders in his sample killed before the fourth stage i.e. before a working agreement could be forged between the two parties. Also Felson & Steadman (1983) declared that their findings on assaultive violent criminal acts cast serious doubt on Luckenbill's (1977) major conclusion that there is a working among participants appropriating the use of violence.
Another fundamental criticism of character contests is that in reality, the meanings with which most violent criminal acts are instilled are very different from those of a character contest. Although the purpose of this essay is not to unduly dwell on the motivations of the offender, in the case of a character contest, displaying strong character maintains honour and face, that is, pride and no shame. This is not however the meaning with which criminal perpetrators often attribute their violent actions e.g. jealousy, hate, disgust (Athens, 1985).

Despite these major criticisms, Goffman's (1967) work provided a basis for a theory explaining both the nature of the violent criminal action and the actors. It is from this work that a more inclusive, complex theory of violent criminal behaviour was developed; the foundations of such a theory can be found in the Impression Management Approach (IMA).

In accordance with Goffman's (1967) theory, IMA upholds that violent escalation occurs when an individual is cast into a negative situational identity and retaliates in order to save face. In addition, the basic determinant of aggression is believed to be perceived intentional attack (i.e. Luckenbill's 1977 stage two). An insult is said toaltercast (or place) the target into a negative situational identity and it is through retaliation that an actor attempts to nullify that identity and altercast the initial aggressor into an unfavourable identity.

A variety of research methods have been employed in order to investigate symbolic interactionism. Evidence from observational studies of aggression on children (Rausch, 1965, as cited in Birkbeck & LaFree, 1993), participant observation (Horowitz & Schwartz, 1974) , and the examination of homicide police reports (Hepburn, 1973; Athens, 1978; Luckenbill, 1977) have all demonstrated that retaliation manifests face-saving concerns.

In his study, Felson (1982) relies on three different sample groups namely: ex-mental patients, ex-criminal offenders, and a third group thought to be representative of the general population. One of his hypotheses predicted that participants would be more likely to verbally attack an antagonist during a conflict if they had been insulted themselves. Respondents were interviewed about incidents in which they had been involved at differing levels of severity. The self-reports were analysed and in support of the afore mentioned hypothesis, it was found that respondents were more likely to engage in verbal dispute (as opposed to be angry but do nothing) when they had been insulted.

In a similar study Felson & Steadman (1983) found that victims of homicides were more aggressive than those of assaults i.e. the victims of homicides engaged in more identity attacks (p<.06), physical attacks (p<.05), and threats (p<.05) than did the victims of assaults. These results were further examined to reveal that antagonists appeared to respond in kind for particular types of aggressive acts. Felson & Steadman (1983) found that identity attacks from one party were likely to lead to identity counterattacks from the other party (r=.38), this is in accordance with Felson's (1982) finding. Similarly physical attacks from one antagonist often lead to physical counterattacks from the other antagonist (r=.30) (see also Fritzon & Ridgway, 2001). These results are crucial to IMA in that they strongly suggest the successive behaviours of a participant are more a function of the antagonists behaviour than they are of their own earlier action, thus demonstrating the importance of interaction in these incidents.

In Felson & Steadman's (1983) study, they analysed the characteristics of offenders and victims and their interaction. Through examination of 500 case files of disputes leading to criminal violence they concluded that victims of homicide and assault were of a similar age and sex. There were however significant difference in the variables associated with the victim's behaviour during the incident, finding that individuals were more likely to be killed if they displayed a weapon or were intoxicated with drugs or alcohol. This is consistent with the interactionist approach in that the behaviour of the victim seems to have affected the behaviour of the offender.

Other social psychology studies have also investigated the interaction between participant variables, especially that of gender. A further hypothesis in Felson's (1982) research predicted that males were more likely than females to attack an antagonist verbally when they had been insulted. This was based on the assumption (strongly related to IMA) that males may be more concerned with their identities in conflict situations than females. This assumption is justified in terms of traditional gender roles; for masculine identity, retaliation is more normative (Frodi, Macaulay & Thome, 1977, as cited in Felson, 1982). IMA takes this to explain that negative identities are more likely attributed to males who backdown when they have been attacked (insulted) than to females.

Felson's (1982) examination of self-reports revealed some interesting and somewhat unpredicted results. In support of his hypothesis, it was found that an insult increases the odds of a verbal dispute 3.2 times when the respondent is male, and 2.2 when female. Furthermore, a highly significant three-way interaction was found involving severity of outcome, sex of respondent and sex of antagonist. This showed that hostilities are more likely to involve physical violence when males conflict with other males. The odds of a physical dispute versus a verbal dispute increases 2.1 times when both parties are male, as opposed to both female. The likelihood of physical versus verbal dispute for two males is 4.3 times more likely than for cross-sex conflicts. These significant results support the view that individuals retaliate when cast into negative identities, particularly when these identities are important to them. This in turn implies that males are more concerned with their identities in aggressive conflicts than females.

Although it is clear that the results of the discussed studies generally support an IMA to aggression and violence, the relationship between insult and expression of anger in a verbal attack is far from perfect. Besides the moderate amount of measurement error anticipated in the variables, there are many other ways of eliciting aggression apart from insult. Furthermore, it may be unintelligible to retaliate in some situations, for example if an individual fears further, more severe, retaliation from the opponent. It is flaws like these that disparage inferences drawn from IMA studies

Despite their methodological and theoretical difficulties, interactionist studies have played an instrumental role in bringing the criminal situation as a unit of investigation in the study of crime and deviance. Moreover, IMA's emphasis on the subjective role of the offender serves to criticise both prior and ongoing research ignoring such factors (e.g. Cloward & Ohlin, 1960) It is clear on the basis of interactionist research that violent criminal acts are not a one-sided event with an unaware victim undertaking a passive role. Rather, they are the outcome of a dynamic interchange between an offender and victim.

Given the importance of situational variables in the determination of behaviour, it seems obvious to investigate the effect which others present in the situation may have on the criminal outcome. In line with IMA, there is experimental literature showing that the mere presence of a third-party can lead the conflict to become more severe (Borden, 1975, as cited in Felson, 1982; Felson, Ribner & Siegel, 1984). Felson (1978) suggests that third-party presence in a conflict raises the identity costs of backing down to an antagonist and therefore generally conflicts are likely to be more severe in front of an audience. He tested this hypothesis but found only partial support (Felson, 1982). It seems that simply the presence of a third-party is not sufficient to cause the conflict to escalate.

It has now been established that third-parties may play a variety of roles in criminally violent situations, namely instigating, mediating, escalating or facilitating of criminal violence (Decker, 1995). There is experimental evidence supporting the suggestion that the outcome of the violent situation can be affected by the perceived values of bystanders (Felson et al., 1984). Instigating actions by third-parties indicate to the potential offender that the audience is favourable to aggression; mediation on the otherhand conveys the opposite message.

Brown (1968) gave adolescent male participants feedback informing them they had looked "˜foolish and meek' when they backed down; these males were more likely to retaliate than those who had not received such feedback. Felson (1982) too found support for the hypothesis that conflicts are likely to be more serious if others present encourage the conflict to continue. He found third-party instigation increases the odds of a verbal dispute 1.5 times, and increases the odds of a physical dispute by double.

Similarly, the effects of third-party mediation has been investigated. Mediation may allow both parties to back down without losing face. Rubin & Brown (1975, as cited in Felson, 1982) found that in general, aggression in a conflict can be inhibited by mediation. Felson (1982) found a positive relationship between mediation by a third-party and severity of incident, implying that the more severe the incident, the more likely that third-parties will attempt to mediate the situation.

A study focusing solely on third-party effects was conducted by Felson et al (1984). Their research was a more detailed concerning others present in a violent criminal situation. Their analysis was based on descriptions of incidents for 155 males incarcerated for felonious assault, manslaughter or murder. Their examinations revealed that third-parties were far more likely to be aggressive than serve as mediators. They also found that an offender would be more violent if their significant others (i.e. within the third party) were also aggressive; mediation effects for these cases received mixed support. It was also revealed that third-parties were more supportive of violent incidents involving young offenders. The age of the offender had a strong, positive relationship with the rate of third-party mediation, and a negative relationship with aggression by significant others of the offenders.

Mediating actions were found to be fairly infrequent in many of these studies (13%, Felson et al. 1984). This can be partially explained by the fact that the samples only included incidents that resulted in criminal violence. In Felson's (1982) study where this was not the case, mediation was found to occur much more frequently (39% mediators versus 9% instigators).

Based on third-party research, it seems it would be a mistake to view criminal violence as a function of either the behaviour of one individual, or even as the interaction between only the offender and victim. IMA assumes that participants alter their behaviour in front of third-parties to manage a favourable impression for the audience. However it is also possible that the actions of third-parties influence the definition of the situation for the antagonists. More research is needed before one can conclude whether participants are actually conforming rather than seeking approval

Since social psychology research has emphasized the importance of peer influence for criminal behaviour (Rubin, 1980; Cloward & Ohlin, 1960), it seems logical to look at collective behaviour in violent situations. Clark (1995) took third-party and bystander research one step further by investigating differences in situational context for lone versus multiple offending. The secondary analysis of data on 1,748 homicides was found to include 10% multiple offender homicides. The analysis of variables found that a victim's lifestyle was the only victim-related variable distinguishing the two types of homicide; several offender-related variables on the other hand differentiated the homicide types, these included legal history, gender, age and lifestyle. Of the situational characteristics of the homicides, nine variables were found to be significantly related. Multiple offender incidents were more likely to occur at a place of employment, and with a stranger of the same sex. Clark's (1995) analysis revealed that of the three broad categories (i.e. victim, offender and situational), the situational variables were of the most importance in distinguishing lone homicides from multiple offender homicides. This finding lends great support to the situational approach in general and aids the refutation of dispositional theories. However, so far, an interactionist theory has not been advanced that can account for the results.

Interactionist theories have though been found useful for explaining the effects of account giving in conflicts. The studies so far have mainly focused on retaliatory aggression; but Felson (1981, as cited in Felson, 1982) suggests that initial attacks may be administered as a form of punishment for wrongdoing. Stokes & Hewitt (1976, as cited in Felson, Baccaglini & Ribner, 1985) believes that when an account is given for wrongdoing, the individual aligns themselves with the normative order. There is a modest amount of experimental literature to support this view. Felson (1982) found that when a respondent gives an account during a verbal dispute, the odds of the conflict resulting in physical violence are lessened 2.9 times. An impression management device seems to be at work here. It appears accounts given during the incidents enable participants to align themselves with the normative order and therefore avoid further punishment.

A well cited study by Dedrick (1978) found that sanctions given by participants were more lenient for a boy who behaved in an arrogant, unfriendly manner if he offered an account immediately afterwards as opposed to if he had not. Furthermore Blumstein, Carssow, Hall, et al., 1974, as cited in Felson & Ribner, 1981) found that the type of account given for a minor violation was more important than the nature of the infraction itself when predicting participant judgements about an offender. This has implications for official accounts given by offenders and the effects of these on the legal system.

It has been found through analysis of official accounts that offenders often excuse their actions (i.e. deny personal causation) by not mentioning their verbal actions which indicate strong intent to the victim. Additionally, they often justify (or rationalise) their behaviour by claiming the victim physically attacked them (Scott & Lyman, 1968, as cited in Felson et al., 1985). These excuses and justifications do not aid offenders in reducing their sentence, and if their claims are unsubstantiated they are likely to increase the severity of their sentence (Felson & Ribner, 1981). There is of course the issue of bias in official police reports to consider here. The accuracy of reports and self-reports in criminal incidents will always be problematic unless a purely subjective view of the offence is the unit under investigation (Felson, 1982).

Commonly, the matter of generalisability has been questioned in interactionist's findings since the majority of samples employed consist of US males incarcerated in a state prison (e.g. Felson et al., 1984; Felson et al., 1985). Even when research does adopt a broader sample (e.g. Baumer, Messner & Felson, 2000), this usually involves cross-national participants as opposed to cross-cultural ones. In a recent study by Halpern (2001), cross-national social attitude data was investigated using multivariate models. It was found that the interaction of these variables accounted for at least two-thirds of the variance in victimisation at the national level. Such a finding must surely have implications for the generalisability of research findings across States let alone across cultures.

Interactionists are however fully aware (e.g. Athens, 1977) of the disadvantages of their data collection methods, but view this as a necessary cost of assembling a comprehensive situational explanation of crime and deviance. If their methods yield corroborating evidence, then confidence can be increased in the conclusions drawn from their findings. The implications of research on crime and deviance are great. Intervention strategies can be more successfully developed regarding the differing situations of crime. Furthermore research is now gradually taking into account differences in the actual crime itself e.g. Clark's (1995) piloting study on different types of homicide which revealed some important differences in both the situation of the crime and the nature of the offender. Targeting such individuals for increased sanctions may have a positive effect on preventing future violence.

It is clear that the situational approach has a place in the explanation of crime and deviance as posited by Sutherland (1947). Impression Management Approach accounts well for results on retaliatory effects and account giving, but not so convincingly for all third-party effects and differing types of homicide. Increasing research by social psychologists is revealing that the victim, offender and third-party roles in the outcome of criminally violent incidents is perhaps more complex than previously credited. An interactional approach needs to be developed incorporating presently neglected areas such as types of crime and cross-cultural issues.


Athens, L. H. (1978) A symbolic interactionist approach to violent criminal acts. Dissertation
Abstracts International, 37, 627-628. Abstract from: DIALOG File: PsycINFO item:

Athens, L. H. (1985) Character contests and violent criminal conduct: A critique.
The Sociological Quarterly, 26, 419-431.

Baumer, E. P., Messner, S. F. and Felson, R. B. (2000) The role of victim characteristics in
the disposition of murder cases. Justice Quarterly, 17, 281-307.

Birkbeck, C. and LaFree, G. (1993) The situational analysis of crime and deviance. Annual
Review of Sociology, 19, 113-137. Retrieved October 3, 2001 from the World Wide

Brown, B. R. (1968) The effects of need to maintain face in interpersonal bargaining. Journal
of Experimental Social Psychology, 4, 107-122.

Clark, R. D. (1995) Lone versus multiple offending in homicide: Differences in situational
context. Journal of Criminal Justice, 23, 451-460.

Cloward, R. A. and Ohlin, L. E. (1960) Delinquency and Opportunity: A Theory of
Delinquent Gangs. NY: Free Press. Abstract from: DIALOG File:
Bids Item: 01527-000.

Cohen, L. E., Kleugel, J. R. and Land, K. C. (1981) Social inequality: predatory criminal
victimisation: An exposition and test of a formal theory. American Sociological
Review, 46, 505-524.

Decker, S. H. (1995) Reconstructing homicide events- the role of witnesses in fatal
encounters. Journal of Criminal Justice, 23, 439-450. Abstract from DIALOG File:
Bids Item: 0047-2352.

Dedrick, D. K. (1978) Deviance and sanctioning within small groups. Social Psychology, 41,

Felson, R. B. (1978) Aggression as impression management. Social Psychology, 41, 205-213.

Felson, R. B. (1982) Impression management and the escalation of aggression and violence.
Social Psychology Quarterly, 26, 419-431. Retrieved October 3, 2001 from the
World Wide Web:

Felson, R. B., Baccaglini, W. F. and Ribner, S. A. (1985) Accounting for criminal violence: A
comparison of official and offender versions of the crime. Sociology & Social
Research, 70, 93-95.

Felson, R. B. and Ribner, S. A. (1981) An attributional approach to accounts and sanctions for
criminal violence. The Social Psychology Quarterly, 44, 137-142.

Felson, R. B., Ribner, S. A. and Siegel, M. S. (1984) Age and effect of third-parties during
criminal violence. Sociology and Social Research, 68, 452-462.

Felson, R. B. and Steadman, H. J. (1983) Situational factors in the disputes leading to
criminal violence. Criminology, 21, 59-74.

Fritzon, K. and Ridgway, J. (2001) Near-death experience. The role of victim reaction in
attempted homicide. Journal of Interpersonal Violence, 16, 679-696. Abstract from:
DIALOG File: PsycINFO Item: 0886-2605.

Goffman, E. (1967) Interaction Ritual: Essays on Face-to-Face Behaviour. NY: Doubleday.

Hepburn, J. R. (1973) Violent behaviour in interpersonal relationships. The Sociological
Quarterly, 14, 419-429.

Hirschi, T. and Gottfredson, M. (1986) The distinction between crime and criminality. In
Critique & Explanation: Essays in Honour of Gynne Nettler; ed.s T.F. Hartnagel, R.
A. Silverman. Pp55-59. New Brunswick: Transaction Books. Abstract from:
DIALOG File: Bids Item: 391598-5.

Horowitz, R. and Schwartz, G. (1974) Honour, normative ambiguity and gang violence.
American Sociological Review, 39, 238-25.

Luckenbill, D. F. (1977) Criminal homicides as a situated transaction. Social Problems, 25,

Rubin, J. R. (1980) Experimental research on third-party intervention in conflict: Toward
some generalisations. Psychological Bulletin, 87, 379-391. Abstract from: DIALOG
File: PsycINFO Item: 0033-2909.

Sutherland, E. H. (1947) Principles of Criminology (4th ed.) Philadelphia: Lippincote.

ad 4
Copyright 2011 All Rights Reserved