Legitimate Violence Spillover Theory

LEGITIMATE VIOLENCE AND CRIMINAL VIOLENCE: A MULTI-NATION TEST OF THE SPILLOVER THEORY*

Abstract

This article reintroduces the debate and research into the Legitimate Violence Spillover (LVS) theory of violent crime and provides empirical tests of the theory. The LVS theory asserts that the prevalence of socially legitimate violence ranging from corporal punishment of children to executions is part of the explanation for the prevalence of criminal violence. The theory was tested at both the individual and national level with scales to measure level of LV. Results at both the micro- and macro-levels found the hypothesized link between LV and criminal violence. The results suggest that reducing LV can make an important contribution to reducing criminal violence.

Key Words: legitimate and criminal violence, spillover theory, gender, cross-national

Norbert Elias (1978), Manuel Eisner (2003) and Steven Pinker (2011) argue that there has been a centuries long "civilizing process" which has resulted in a major reduction in interpersonal violence since the late middle ages (see also Clark (2012)). This article is about one aspect of the civilizing process, which we suggest is implicit in the analyses just cited, but not given sufficient attention: what we call the Legitimate Violence Spillover (LVS) theory of criminal violence. The LVS theory argues that the prevalence of socially legitimate violence ranging from spanking children to executing murderers is part of the explanation for criminal violence. The LVS was originally introduced to as part of the explanation for the large differences between the states of the USA in rape and homicide rates (Baron and Straus 1988). These studies found that the higher the score of a state on an index to measure the prevalence of LV, the higher the rates of rape and homicide.

The underlying idea of spillover from legitimate and culturally approved violence to criminal violence is present in theories of violence such as the Southern Culture of Violence theory and the Brutalization effects of the death penalty theory. Despite a prominent place of these theories in criminology in the 1980s and 1990s, the number of studies examining the relation of legitimate violence to criminal violence has been minimal in recent decades. We hope that results of this study and instruments it makes available will reinvigorate debate and research into LVS theory. It presents tests of LVS theory at cross-national and individual personal levels. However, both the individual-level and the macro level data are cross sectional. Tests of a theory using cross-sectional data cannot prove the theory but if the hypothesized relationships are not found, the results would raise serious questions about the validity of the theory and suggest a need for either abandoning or further developing theory. On the other hand, if our tests of the theory using cross-sectional analyses are consistent with the LVS theory, it suggests the value of investing resources in more definitive research.

Legitimate Violence

Legitimate Violence (LV) consists of acts of physical violence carried out to secure or maintain socially approved ends. Examples of government-level LV include violence by police to maintain social order, corporal punishment to punish criminals and to maintain discipline in the armed forces, and of course, war as seen by each side. For individuals LV includes what Black (1983) characterizes as self-help justice such as feuds between families, and apprehension and killing of criminals, and of course, duels to settle insults. Within the family, there was the common-law right of husbands to “physically chastise an errant wife,” and the right, indeed in most societies the obligation, to use corporal punishment to correct persistent misbehavior. Many forms of LV have been decreasing. Corporal punishment in the armed forces and the right of husbands to use corporal punishment on wives ended in Europe in the late 19th century, and the process is continuing in other nations (Pierotti 2013). The most severe LV in child rearing has been declining for a long time (DeMause 1984). In respect to the less severe form of violent child-rearing called “smacking” and “spanking,” the European Union and the United Nations have requested all member countries to ban corporal punishment by parents, and 46 nations have done so in 2015 (Zolotor and Puzia 2010). A recent and still ongoing reduction in LV is the prohibition of capital punishment in most Western nations. This decrease in LV might be part of the explanation for the decrease in violent crime that has occurred in Europe and North America in recent years (LaFree et al. 2015).

The Legitimate Violence Spillover Theory

One of the roots of our focus on spillover from LV to criminal violence originated in analytical traditions based on the principle that society constitutes a social system in which each part of the system tends to influence the other parts, including general systems theory (Buckley 1967). The applicability of systems theory to understanding violence in the family was suggested by (Straus) (1973). Spillover in cultural norms and beliefs concerning violence is an important aspect of LVS theory. Specifically the more a society tends to endorse the use of physical force to attain socially approved ends (such as order in the schools, crime control, and international dominance), the greater the likelihood that this legitimization of force will be generalized to other spheres of life where force is less socially approved, such as the family and relations between the sexes (Baron and Straus 1989: 147).

Empirical Research on LV and Criminal Violence

Domestic Violence. Using Human Relations Area File data for 90 societies, Levinson investigated the relation found that women were more likely to be beaten, permanently injured, scarred, or killed by their husbands in societies in which military glory was a source of male pride, criminals were subjected to physical punishments, animals were treated cruelly, and enemy captives were tortured. He concluded that that wife beating was a part of a broader culture pattern of violence (Levinson 1989: 45). A study of incarcerated men and women (Robertson and Murachver 2007) found that the incarcerated group had greater acceptance of legitimate violence than the comparison group, and that this was associated with whether they had assaulted a partner. Cauffman et al. (2000) found that approval of violence to redress perceived inequity, for conformity to peers, and for self-defense, were related to self-reported aggressive behavior toward peers by men but not by women. For women the same time, these aspects of LV were associated with self-reported date violence.

Rape and Sexual Coercion. Two cross-cultural studies found that rape was more common in societies in which pubescent boys were subjected to the legitimate violence of genital mutilations such as circumcision (Minturn et al. 1969) and in societies waging war (Sanday 1981). Baron and Straus investigated differences between US states in rates of rape (Baron and Straus 1988; Baron and Straus 1989; Baron and Straus 1987). They found that the higher the score of a state on a LV Index, the higher the rate of rape. A study of approval of socially LV by university students used a scale consisting of agreement with questions such as “All states should incorporate the death penalty into their judicial systems” and “Military force is often a useful tool in achieving foreign policy goals” (Hogben et al. 2001). It found that the higher the score on this index, the higher the score on a measure of coercive sexual behavior by men, although not by women.

Homicide. Several cross-national studies of war have found results that are consistent with the LVS theory by showing that war was associated with an increase in the homicide rate (Archer and Gartner 1984). The authors suggested that wars tend to legitimate the general use of violence in domestic society via a message that killing another human being was, under certain circumstances, acceptable in the eyes of the nation’s leaders. Another cross-national study of nineteen nations provided evidence that variations in attitudes justifying killing (i.e., in defense of property, if someone kills a person who has raped their child, and the death penalty) are related to differences in national homicide rates (McAlister 2006). The acceptance of killing varied significantly among national/regional groups. Together the attitude toward killing and social inequality (GINI) explained 65% of the variance in homicide rates.

Youth Crime. Durrant (2000) studied trends in the wellbeing of Swedish youth since the passage of the 1979 act banning corporal punishment by parents. The subsequent trend in selfreported violence and young people suspected of homicide remained relatively steady between 1976 and 1994. This is contrary to the reduction predicted by LVS, and also contrary to the increase predicted by opponents of the law who believed that forbidding corporal punishment would result in an increase in crime by youth. However, consistent with LVS theory, rates of youth involvement in rape and other crime decreased since corporal punishment was banned in Sweden.

Corporal Punishment by Parents. Spanking or smacking by parents may be the most prevalent and culturally legitimate type of violence. The meta analysis of 88 studies found 93% agreement of the link between spanking, social and psychological problems, including delinquency as a child, and crime as an adult (Gershoff 2008). Among these are studies which have found that corporal punishment is associated with an increased probability of antisocial behavior as a child (Straus et al. 1997), serious adult crime (McCord 1991), physically abusing children (Straus, 2001), and assaulting non-family members (Straus et al., 2014) as well as marital partners (Straus 2001). A systematic review of longitudinal studies (Ferguson 2013) found small but generally significant relationship between corporal punishment and long-term negative outcomes, such as externalizing and internalizing symptoms and cognitive performance of a child. At the macro-level, a study of 186 societies found that frequent corporal punishment was related to higher prevalence of violence and endorsement of violence (Lansford and Dodge 2008). Many processes are involved in producing this linkage, three of which were tested in a large multi-nation study: reduced opportunity to learn non-violent methods of resolving conflict with parents, depression from being the victim of physical assault by parents, and social learning (Straus and Yodanis, 2014). Given the number and consistency of these studies, and the empirically demonstrated intervening processes, it is reasonable to conclude that smacking children is associated with an increased probability of crime.

Hypotheses

Hypothesis 1 was tested the LVS theory at the individual-level. Ho 1. The higher the score of students on a scale to measure their belief in LV, the higher the probability of having perpetrated acts of criminal violence.

Hypotheses 2 and 3 tested the LVS theory at the nation-level. Ho 2: The higher the average LV score of students in each of the 32 nations, the higher the level of criminal violence of the nation from which the student sample was drawn. Ho 3. The higher the level of LV as measured by published national statistic, the higher the level of criminal violence of the nation from which the student sample was drawn. Complete Article