Note: This is the postprint of the following paper; publisher version available here. Unfortunately, I cannot locate the tables, so you will need the publisher version to see those. Please email me for a copy.
Rennison, Callie, Scott Jacques, and Mark Berg. 2010. Weapon Lethality and Social Distance: A National Test of a Social Structural Theory. Justice Quarterly 28:576-605.
Abstract: Three paradigms can be used to explain weapon lethality: rational choice and deterrence theory; social learning and cultural theory; and opportunity and prevalence theory. Each makes distinct predictions regarding the economic, psychological, and environmental factors that affect the use of weapons. Despite their merits, the sum of knowledge about violence and weapons may be increased by exploring the influence of variables derived from another paradigm: pure sociology. Black’s theory of retaliation and Cooney’s principle of predation provide the underpinning for a social structural‐based theoretical principle of weapon lethality. Building on those ideas, we propose that the lethality of weapons involved in interpersonal violence increases as the offenders and victims become less intimate and less alike culturally. Using National Crime Victimization Survey data, we test two hypotheses derived from this principle and primarily find support of the proposed social structural principle.
Weapons and violence go hand-in-hand. Tools of damage include spears, guns, and bombs, but attacks may also involve “body objects” such as fist, feet, claws, and jaws (see Cook, 1991; Kleck & McElrath, 1991; Pruetz & Bertolani, 2007). Although any item can conceivably be a weapon, not all objects have the same potential to inflict bodily harm. There is a continuum of weapon lethality defined by the amount of physical damage that potentially can result from the use of a weapon (Black, 2004a). A greater potential for damage is equivalent to greater weapon lethality. For example, a fist tends to be less lethal than a knife, which tends to be less lethal than a firearm.
An important question is why violent altercations differ in the lethality of weapons involved? In other words, under what conditions are guns versus knives, bats, or fists more likely to be used in acts of interpersonal violence? There are many conceivable explanations of this variability, all of which have their own unique merits (see, e.g., Copes, Kovandzic, Miller & Williamson, forthcoming; Felson & Messner, 1996; Heide & Petee, 2007; Ludwig & Cook, 2000; Rosenfeld, 2000; Phillips & Maume, 2007). More recently, another possible explanation of weapon lethality has begun to emerge from the paradigm of pure sociology (see Black, 1998; 2004a; Cooney, 2006; 2009; Phillips, 2003; Phillips & Cooney, 2005).
We propose that when combined, two distinct ideas of pure sociology – Black’s (1998, 2004a) theory of violent retaliation and Cooney’s (2006) theoretical principle of violent predation – suggest that greater social distance between offenders and their victims leads to the use of more lethal weapons in interpersonal violence (also see Phillips, 2003; Cooney & Phillips, 2002). Although extant qualitative studies provide empirical support for this idea (Black, 2004a: 148-151), we are unaware of any quantitative research that tests this sociological explanation of weapon lethality.
Given this gap in the literature, the present research has two goals. One is to synthesize the theorizing of Black (2004a) and Cooney (2006) to propose a social structural-based theoretical principle of weapon lethality. Our second goal is to test this theoretical principle using 1993 to 2004 National Crime Victimization Survey (NCVS) data. To this end, the remainder of this article is organized in the following manner. The first section summarizes theories of weapon lethality. Next, we describe the framework of pure sociology and related theories (Black, 1990; 2004a; Cooney, 2006). Third, we propose a theoretical principle of weapon lethality and social distance from which two hypotheses are derived. The following section then outlines the data, methods and analytic technique used to test these hypotheses. Finally, findings are presented and discussed, including limitations, future research directions, and the implications for criminology, criminal justice, and reducing weapon violence.
Weapon lethality is a quantitative variable measurable by the amount of bodily harm that can be potentially achieved through a given number of applications of a weapon (e.g., one gun shot or ten stab wounds), to a particular area of a body (e.g., foot, stomach, and neck). Holding constant how many times a weapon is used and the body parts that are ultimately struck, weapons are said to be more lethal as greater bodily damage results from their use in violent incidents. According to this logic, weapon lethality is conceptually arrayed on a continuum.
There are several scientific paradigms with the potential to explain variability in the lethality of weapons used in violent events (see e.g., Braga, 2003; Fagan & Wilkinson, 1998; Nielsen, Martinez & Rosenfeld, 2005; Rosenfeld, Baumer & Messner, 2007; Wells & Horney, 2002). Three perspectives dominate this topic of study including the micro-macro twins of (1) rational choice and deterrence theory, (2) social learning and cultural theory, and (3) opportunity and prevalence theory. We next briefly describe these theories as they apply to weapon lethality.
Rational Choice & Deterrence
Rational choice and deterrence theories posit that the commission of criminal acts depends on their benefits and costs relative to other lines of action (Bentham, 1988 ; Cook, 1980; Stafford & Warr, 1993; Zimring & Hawkins, 1973). A choice holds greater utility as its benefits increase and costs decrease. This means crime should become more probable as it becomes more beneficial for or less costly to the perpetrator. However, when a criminal action holds fewer potential benefits relative to conventional alternatives, crime is said to be deterred.
Thus, a rational choice theory of weapon lethality examines how the frequency of weapon use in violent altercations varies as a result of a cost-reward system (e.g., Decker & Rosenfeld, 2004; Kleck, 1997; Ludwig, 2005; Ludwig & Cook, 2000; Rosenfeld, Fornango & and Baumer, 2005; Zimring, 1968). The use of any particular weapon should, according to this perspective, increase as it becomes more beneficial and less costly to the offender. For instance, as the likelihood or length of imprisonment for using a firearm increases, this behavior should decrease. When there are multiple weapons that may be used, the one chosen is the one having the greatest rewards and smallest punishments, at least in theory. For example, if using a gun is punished more harshly by the government than using a knife, than the latter behavior should occur more often the former, all else equal.
Using this logic, others theorize that victimization represents an informal sanction which also figures into the cost-benefit logic of offenders. For instance, using a fist to carry out a crime may make one seem weak which could invite future attacks and come at a cost to one’s personal safety (Felson & Steadman, 1983). Within this context, the application of more lethal weapon may translate into a lower risk for being preyed upon in the future (Felson, 2009).
Social learning and cultural theories are logical compliments; they argue that the actions of individuals and groups are conditioned by the ideas, values, and attitudes of persons (Akers & Jensen, 2006; Anderson, 1999; Sutherland, 1937). In other words, people learn how to behave. Whether people behave criminally depends in large part on the degree to which normative support for criminal behavior is salient in their community and among their peers.
Explanations of weapon lethality nested in the social learning and cultural paradigms explore how ideas about the appropriateness of weapons in interpersonal conflicts affect their use by persons or groups (see, e.g., Copes et al., forthcoming; Dixon & Lizotte, 1987; Felson & Messner, 1996; Felson & Pare, forthcoming-a; forthcoming-b; Tedeschi & Felson, 1994). In short, the use of one weapon, such as a gun, over another weapon, such as a knife or bat, is determined by the weight of ideas, values and attitudes about each. If a person or group has more positive ideas and attitudes about guns than knives, then a gun is more likely to be employed in any given violent altercation.
A key assumption of many cultural theories of crime is that broader structural configurations determine the content of what is taught and learned (see, e.g., Anderson, 1999; Wilson, 1987; 1996). For example gun related violence is spatially concentrated in urban communities marked by overlapping socioeconomic adversities including poverty, joblessness, and isolation from conventional institutions. To account for this pattern of violence, cultural theories propose that such structural conditions give rise to an alternative cultural orientation which places greater sanction on the acquisition of firearms and their use during the commission of crime (see Baumer, Horney, Felson & Lauritsen, 2003; Liberman, 2007).
The opportunity perspective is grounded in the realization that behaviors, such as crime, cannot occur unless key minimal elements come together (Cohen & Felson, 1979). For example, the routine activity theory of crime tells us that offending is improbable unless a motivated offender and target converge in time and space when and where a capable guardian is absent (ibid; Reynald, forthcoming). Thus, crime is impossible when offenders and targets stay separated from each other. Hence, when suitable control agents are not present, crime should increase as offenders and targets increasingly come together temporally and spatially.
One factor that affects the rate at which minimal elements converge is the prevalence of those elements within a given spatial environment (see Cook, 1983; Killias & Haas, 2002). For instance, as the number of offenders or targets increases in a neighborhood, those factors may come together on a more frequent basis and, in turn, crime may increase. In terms of explaining weapon lethality, this perspective suggests that the frequency that a weapon is used depends on how many of those weapons exist (Cook & Ludwig, 2004). If there are significantly more guns than knives, the use of guns should be more common than the use of knives, all else equal.
The opportunity perspective makes clear that the number of weapons must also be considered alongside the number of situations where offenders and targets converge in time and space. If an armed robber does not come into contact with a potential victim, then armed robbery is impossible despite the presence of a weapon (e.g., see Wright & Decker, 1997). Therefore, variability in the use of weapons with different lethality is affected by the prevalence of each weapon (e.g., number of guns versus knives) and the rate at which offenders with each kind of weapon find themselves next to potential victims. For example, although there may be more knives than guns within a community, it may be true that offenders more often carry guns with them than knives. In short, these theoretical perspectives suggest both the prevalence of weapons and the opportunity to use them is central to explaining variation in weapon lethality across violent events (see Phillips & Maume, 2007).
In short, the three dominant paradigms of weapon lethality – (1) rational choice and deterrence theory, (2) social learning and cultural theory, and (3) opportunity and prevalence theory – make distinct predictions regarding the economic, psychological, and environmental factors that affect whether acts of interpersonal violence involve fists, bats, knives, or firearms. Despite the merits of those theoretical contributions, the sum of knowledge about violence and weapons may be increased by exploring the social structural factors that potentially influence weapon lethality.
Pure sociology is a scientific paradigm focused on conceptualizing and theorizing social behavior (Black, 1976; 1995; 1998). It has been used to explain a range of social behaviors, including therapy, (Horwitz, 1982), ideas (Black, 2000a), welfare (Michalski, 2003), medicine (Black, 1998), art (Black, 1998), and method (Jacques and Wright, 2008a, 2010a, 2010b). As relates to crime and its control, work within this paradigm has theorized and researched vigilantism (Black, 1983), predation (Cooney, 2006; Cooney and Phillips, 2002), lynching, rioting, terrorism (Senechal de la Roche, 1996; Black, 2004a, 2004b), genocide (Campbell, 2010), and law (Black, 1976, 1998), including the death penalty (Phillips, 2009). Pure sociology explains social behavior as the result of situational changes in social structure, or social geometry (see Black, 1976; 1998). Social structure is defined by the relative social statuses of and social distances between actors involved in a particular social interaction, or case.
The social status of an actor (i.e., person or group) increases in tandem with that actor’s position in various social hierarchies (Black, 1976, 1998; Cooney, 2009; Phillips, 2009). There are five forms of status. Vertical status refers to relative wealth; for example, the higher one’s income then the greater that person’s vertical status. Radial status is integration in social life; a potential measure of this status is employment or marriage, with involvement in either equating to higher status. Corporate status is the capacity for collective action; more memberships are one way to operationalize this kind of status, for instance. Symbolic status is comprised of conventionality and knowledge; a common indicator of this particular status is one’s level of formal education. Normative status, lastly, refers to subjection to social control; for example, a person who has been arrested or imprisoned has lower normative status than someone without a criminal record.
The social distance between actors decreases as they become more intimate (relational distance) and culturally similar (cultural distance). Relational distance is the degree of intimacy between actors. “It is possible to measure relational distance in many ways, including the scope, frequency, and length of interaction between people, the age of their relationship, and the nature and number of links between them in a social network” (Black, 1976: 41). The relational distance between two persons decreases the more often they interact; for example, two persons who together trade, raise children, work, organize, communicate, or apply social control are closer in relational distance than two actors who have not interacted in any manner. Cultural distance is the degree of cultural similarities and differences between actors. “Every difference [in culture] is a distance, so that people may have religious distance between them, or ideological, moral, linguistic, or aesthetic distance of one kind or another, and all of these together also define cultural distance” (Black, 1976: 74). Persons become closer in cultural distance as cultural characteristics become more similar.
For any given situation, the combination of statuses and distances of involved persons constitutes the social structure. Differences in social structure are thought to be a source of variability in the quality and quantity of social behavior across situations. The goal of pure sociology is to determine how social structure affects social behaviors and specify these relationships in the form of testable propositions that may be falsified through empirical research (see Black, 1995; 1998; Cooney, 1998; Phillips, 2003; Phillips & Cooney, 2005; Popper, 2002).
The goal of the present research is to contribute to the field’s knowledge of violent interactions and weapon lethality. This is accomplished by (1) combining the logic of two existing ideas nested in the paradigm of pure sociology – Black’s (1998; 2004a) theory of violent retaliation and Cooney’s (2006) theoretical principle of violent predation – to form a synthesized, social structural-based principle of weapon lethality, and (2) testing hypotheses derived from the principle with representative data of violent victimizations in the United States.
Social control is defined as a response to conflict (Black, 1976; 1998; 2000b). Violent retaliation is one way of handling conflict (Cooney, 1998; 2009; Horwitz, 1990: 127). Examples of violent retaliation includes gang members shooting rivals who cross into their territory (Shakur, 1993; Wright & Decker, 1997), Crow Indians killing Blackfoot Indians in retaliation for raiding their camp (Ewers, 1967), and drug dealers harming those who victimize them (Jacobs & Wright, 2006; Jacques & Wright, 2008b; Phillips, 2003).
Black’s (1990, 1998) theory proposes in part that violent retaliation is influenced by the social distance between disputants. Specifically, this theory proposes that violence should become more common and serious as disputants become further in relational distance or cultural distance. This means that for any given conflict retaliation should be less frequent and involve less violence among disputants who (1) have spent more time with each other and done more activities together, and (2) are more alike in their ideas and modes of expression.
More recently, Black (2004a) has added a greater level of specification to his theory of retaliation by theorizing the effect of social distance on weapon lethality. Specifically, Black theorizes that “[t]he lethality of weapons [involved in retaliation] is a direct function of social distance” (p. 148). Empirically speaking, this theory predicts that violent retaliation should involve less lethal weapons among disputants who (1) have interacted more often, and (2) are more alike in their ideas on and expressions of what is good, true and beautiful.
Black illustrates these hypotheses with an array of anthropological evidence. The effect of intimacy, for instance, is seen among the Nuer of Sudan, Ik of Uganda, and Mbuti pygmies of the Congo; all would typically use wooden clubs when fighting persons of the same village but used spears when fighting persons outside their village (p. 148). The influence of cultural distance is evident among Irishmen; they usually keep to their fists when squaring off against each other but employ sharp objects when fighting Britons (p. 149; for additional examples, see Anderson, 1999; Black, 2004a: 148-150; Phillips, 2003; Shakur, 1993).
Although pure sociology has been primarily used to explain social control, including retaliation, the paradigm is applicable to other forms of behavior as well – such as predation. Like retaliation, predation is associated with violence (Anderson, 1999; Wright & Decker, 1997). However, retaliation is conceptually distinct from predation.
By definition, the former focuses on social control and the latter focuses on resource exchange (Black, 1983; Cooney & Phillips, 2002; Jacques & Wright, 2008b). “If altruism is giving without taking, and reciprocity is a mutual give and take, predation is taking without giving. In predatory violence, the taking is accomplished coercively, through force” (Cooney & Phillips, 2002: 81). Examples of predatory violence include a street robber “pistol whipping” a victim until the money is forfeited (Jacobs, 2000; Wright & Decker, 1997), and pirates raiding the cargos of commercial sea vessels (Sherry, 1986).
Building on Black’s ideas, Cooney argues that predatory violence can also be explained by the social distance between potential predators and prey and theorizes that predation increases with social distance (2006: 59; see also Cooney, 1997; Cooney & Phillips, 2002: 86). “The proposition implies, then, that as the relational and cultural distance between actors increases so does the probability and severity of predatory behavior” (Cooney, 2006: 59).
Cooney’s (2006) theoretical principle of predation suggests that greater social distance between predators and prey may lead to the use of more lethal weapons in violent predatory incidents. Empirically speaking, this principle predicts that violent predation should involve more lethal weapons among predators and prey who (1) have spent less time with each other and done fewer activities together, and (2) are less alike in their ideas and modes of expression.
A truism of the philosophy of science is that the value of theory increases concomitantly with its “empirical diversity” (see Kuhn, 1977; Popper, 2002). “Science craves generality. The greatest glory is enjoyed by theoretical scientists whose formulations reach previously unattained levels of generality” (Black, 1995: 833-834). As described above, the consistencies between the theorizing of Black (1998; 2004a) on violent retaliation and Cooney (2006) on violent predation suggest it may be possible to synthesize them into a more general form capable of explaining weapon lethality in both retaliatory and predatory incidents.
By combining these theories a more favorable degree of theoretical generality is possible with regard to the behavior of weapon lethality. We build on these theories by simplifying them into a single theoretical principle:
Weapon lethality increases as the social distance between offender and victim increases.
Empirically speaking, this principle predicts that acts of interpersonal violence should involve more lethal weapons among offenders and victims who (1) are less acquainted, and (2) are less similar in their ideas and forms of expression. We next turn to testing these hypotheses.
This research employs data from the NCVS derived from a large, household sample that is representative of non-institutionalized people age 12 or older in the United States. In recent years, approximately 80,000 persons in 40,000 households are interviewed every six months for a total of seven interviews conducted both in person and over the telephone. Through a series of screening questions, it is determined whether a respondent was a victim of a threatened, attempted or completed crime during the preceding six months. Additional questions gather detailed information about the nature of victimizations including the victim characteristics, offender characteristics, their relationship, and characteristics of the incident.
The NCVS has several key strengths for the present research. First, it offers a large, nationally representative sample (approximately 200,000 interviews annually). Second, in contrast to many other crime surveys, the NCVS collects detailed incident-level information. And third, the NCVS data contain victimizations regardless of whether or not they were reported to the police.
Limitations of the data include limited population coverage since one must be 12 years of age or older and live in a housing unit or group quarter to be eligible for this survey. A second limitation is limited coverage of violence since the NCVS does not collect data on all forms of violence. A third limitation is that it does not provide measures of all aspects of social structure as defined by pure sociology. Another limitation is that the NCVS does not to distinguish between retaliatory and predatory acts of the offender. A final limitation is the NCVS does not provide the necessary measures to test or control for the effect of alternative theoretical perspectives, namely rational choice, learning/culture, and opportunity theories.
Despite these limitations, the NCVS provides an opportunity to examine the relationship between weapon lethality and relational and cultural distance. The analyses that follow are based on a sample of 8,293 violent victimizations occurring from 1993 to 2004 The data are aggregated to maximize the number of available cases. Aggregation is especially important given that non-fatal violence involving certain weapons (e.g., firearms) is a relatively rare event.
The dependent variable in this research is weapon lethality. Weapon lethality is conceptualized as the amount of physical damage that can potentially result from the use of an object in violent altercations.  There is a continuum of weapon lethality:
At the lower lethality end are cases in which the weapon used to threaten or attack the victim is a body part such as the offender’s feet and hands.
More lethality is attributed to violent incidents in which a victim was threatened or attacked with a blunt object, such as a rock or bat.
More potentially lethal are cases in which the victim was threatened or attacked with a knife or other cutting implement.
And finally, weapon lethality is greatest in cases where the offender threatens or attacks the victim with a firearm.
Thus, body parts are less lethal than blunt objects, which are less lethal than sharp objects, which are less lethal than firearms.
This research models the degree to which weapon lethality is a function of social distance. Two independent variables are used as measures of social distance: relational distance and cultural distance.
Relational distance is the aspect of social life defined by actors’ interaction with each other and is measured as a dichotomous variable: 0 = known participants and 1 = strangers.
Cultural distance is measured as the similarity between victim and offender’s race/ethnicity. Victims and offenders characterized by the same race/ethnicity are considered closer in cultural distance and are measured via a dichotomous variable: 0 = same race/ethnicity and 1 = differing race/ethnicity.
All models control for a number of variables that existing theory and research suggest may influence violence and weapons. These include measures to account for the social status (i.e., social disadvantage) of victims, including their wealth (vertical status), employment and marital status (radial status), and level of formal education (symbolic status) (see Black, 1976; 1995; 1998). In addition, the models control for factors that alternative perspectives and prior research demonstrate are relevant, such as the age and gender of victims and offenders (see Gottfredson & Hirschi, 1990; Heimer & DeCoster, 1999; Rennison & Rand, 2003; Rennison, 2002), and the location of the violent incident (Wilson, 1996). Details about all variables are found in the Appendix.
Before presenting the results, additional issues require attention: the analytic technique used and variance estimation. In the following analyses, binary logistic regression is utilized. Binary logistic regression models the classification of cases into one of the categories of the dependent variable. In this case, models demonstrate the likelihood of weapon lethality based on social distance. We operationalize weapon lethality in three ways using dichotomous variables:
Personal weapon lethality (i.e., fist or feet attacks coded as 0) vs. non-personal (i.e., rock, knife or firearm attacks coded as 1).
Low weapon lethality (coded as 0 and including body blows, strikes with objects) vs. high weapon lethality (coded as 1 and including knives and firearms).
Physical force of the offender weapon lethality (body blows, strikes with objects, slashing of cutting implements, coded as 0) compared to the mechanical deliverance of lethality provided by a firearm (coded as 1).
Assessing the influence of social distance on each measure of weapon lethality offers a more robust understanding of their theorized relationship.
A second analytic issue stems from the fact that the NCVS is collected using stratified, multistage cluster sample methodology meaning that the assumption of independence between observations is violated. Failure to account for this complex sample design in analyses leads to biased standard errors and inflated t-statistics. To account for the sample design, the following analyses use Taylor series linearization to allow for accurate estimation and significance testing (Levy & Lemeshow, 1999). In addition, the analyses utilize the appropriate weights available on the file to account for nonresponse and undercoverage.
Before presenting our findings of hypotheses tests, Table I describes the sample used in our analyses. Most (73%) non-fatal violent altercations involve a victim who is physically threatened or attacked by an offender with no tangible weapon (i.e., a stick, knife, or firearm). About 15% of incidents involve a threat or an attack with an object such as a bat or stick. Violent acts rarely involved a threat or use of a knife or similar cutting implement (7%) or firearm (5%). Almost half of the violence was committed by a stranger (47%). Most victims were male (56%), never married (60%), and employed (61%). About half of victims live in the suburbs (47%). The mean age of victims was 28 years old, and victims had completed a mean of 11.8 years of education. Most violent victimizations occurred away from the home of the victim or an acquaintance of the victim (65%), involved one offender (77%), and no bystanders (68%).
- - TABLE I ABOUT HERE - -
Results from regression models evaluating the influence of social distance on weapon lethality are presented in Table II. In each of these models, it is hypothesized that closer social distance – in terms of relational and cultural distance – is related to less weapon lethality.
Panel A presents findings from a binary logistic regression that investigates the influence of social distance on weapon lethality measured as threat or attacks using body parts (coded 0) versus threats or attacks involving objects, knives or firearms (coded as 1). Results presented in Panel A offer support for the social distance—weapon lethality hypothesis. Relational distance between victim and offender is positively related to weapon lethality. That is, the greater the relational distance between the victim and offender, the more likely that the victim will be threatened or attacked with an object, knife or gun than with a “body” weapon such as a fist or foot (b = 0.33, p < .001, AOR = 1.39). As relational distance increases, the odds of threat or attack with an object, knife or firearm versus a bodily attack increases 39 percent.
As hypothesized, cultural distance between victim and offender is positively related to weapon lethality. Violent acts involving victims and offenders with the same race/ethnicity (i.e., less cultural distance) are more likely to involve threats, attempts or attacks with body parts compared to objects, knives or firearms (i.e., less weapon lethality) (b = 0.17, p = .01, AOR = 1.18). As cultural distance increases, the odds of attack with an object, knife or firearm attack (versus a physical attack) increase 18 percent. Thus, in this model both aspects of social distance are positively and significantly related to weapon lethality.
- - TABLE II ABOUT HERE - -
Panel B presents findings from a model assessing the influence of social distance on weapon lethality, with weapon lethality measured as low (i.e., body or object attack, coded as 0) compared to high weapon lethality (i.e., knife or gun attack, coded as 1). Again, it is hypothesized that closer social distance – in terms of relational and cultural distance – is related to less weapon lethality. Findings presented in Panel B support the hypothesis. Findings indicate that greater relational distance is associated with greater weapon lethality (b= 0.32, p<.001, AOR= 1.38). The adjusted odds ratio indicates that as relational distance increases, the odds of a higher weapon lethality violent victimization increase 38 percent.
Findings in Panel B also indicate that the less the cultural distance (i.e., the more similar the victim and offender in terms of race/ethnicity), the less likely the victim will be threatened or attacked with a knife or firearm (compared to a physical attack or one involving an object) (b = 0.28, p = .001, AOR = 1.32). The associated adjusted odds ratio indicates that the odds of more lethal weapons being involved increases 32 percent when the victim and offender’s race/ethnicity differ. Therefore, as was found in the previous model, social distance is positively and significantly related to weapon lethality.
The final model considers weapon lethality measured as physical force of the offender (i.e., body blows, strikes with objects, slashing of knives, coded as 0) compared to the mechanical delivery of violence provided by pulling the trigger of a firearm (coded as 1). Findings offer mixed support for the hypothesis that greater social distance is related to greater weapon lethality. In support of the hypothesis, findings show that relational distance is positively related to weapon lethality (b = 0.54, p<.001, AOR = 1.71). As relational distance increases, the odds of a firearm threat or attack compared to an attack by any other means increase 71 percent. In contrast, no relationship between cultural distance and weapon lethality is observed in Panel C. That is, cultural distance is not a predictor of weapon lethality when conceptualized in a manual versus mechanic fashion (p=0.31).
A weapon is any object used to inflict physical damage. Weapons may be arranged along a continuum of lethality based on their ability to damage objects. Using the ideas of Black (1998; 2004a) and Cooney (2006), this research proposes a social structural-based theoretical principle: The lethality of weapons involved in acts of interpersonal violence increases as the social distance between offenders and victims increases. This principle predicts that more lethal weapons are used in violent acts when the victim and offender are less familiar with each other or are more culturally distinct.
This research tested two hypotheses using data from the NCVS. There was support for the relationship between weapon lethality and relational distance in each model. As relational distance between a victim and offender increases, more lethal weapons are involved in their violent interaction. Results are mixed regarding the hypothesized relationship between weapon lethality and cultural distance. In two of the three models, results suggest that violence between individuals of similar race/ethnicity is less likely to involve more lethal weapons than is violence between culturally dissimilar individuals.
The present research has limitations and future research should attempt to resolve them. Future research should determine whether social distance explains weapon lethality in (1) both predatory and retaliatory events, (2) violence between groups, (3) changes within relationships, (4) homicides, or (5) when using nuanced aspects of culture. In addition, research should (6) explore whether social distance affects the other key components of violent acts, including the number of times a weapon is used, to which body parts, and the quantity and quality of the resultant injury.
For one, it is important to test for difference in the relationship between social distance and weapon lethality in predatory versus retaliatory violence. NCVS data do not allow for this kind of delineation because it is unknown whether reported victimizations are predatory or retaliatory. Although the purpose of this paper is to provide a general principle of weapon lethality and social distance by combining Black’s (2004a) theory of retaliation and weapon lethality and Cooney’s (2006) principle of predation, it is possible that the effect of social distance differs depending on the kind of violence that occurs.
Second, future research should investigate the effect of social distance on weapon lethality in violent interactions involving groups. For example, the social distance between warring nations may have an influence on the lethality of weapons used in battle. Or, fights between gangs may involve more lethal weapons than fights between individuals within a gang (see, e.g., Shakur, 1993).
Third, the theoretical principle presented in this research should be examined with homicide incidents not available in the NCVS. Determining whether and why there are differences between lethal and non-lethal violent altercations in the lethality of weapons involved and how this is affected by social distance is an important endeavor.
Another conceivable line of research would be to examine variation in weapon lethality within relationships. Social distance is constantly in flux for any given relationship. For example, intimates become more intimate as they spend more time together, acquaintances lose relational distance as they spend less time together, and strangers can evolve into intimates over time. The theoretical principle proposed in this research predicts that as two persons become more integrated in each other’s lives, the lethality of weapons involved in their violent altercations should diminish.
Fifth, culture is a multifaceted concept. A limitation of this paper it is uses an indirect measure of it, namely differences in race/ethnicity. Yet culture also varies within race/ethnicity (Stewart and Simons, 2010). Consider Anderson's (1999) work in the inner-city, which shows that “decent” kids who do not conform to street culture are relatively likely to be victimized by “street” kids who do conform to the “honor culture.” Thus, while street and decent kids are often from the same neighborhood and of the same race/ethnicity, a different orientation toward life – i.e., greater cultural distance – can increase the odds of violent victimization. Because cultural distance is such an encompassing term it is important that future studies examine how other manifestations of culture, say “street” versus “decent” orientations, contribute to variation in weapon lethality (and when holding constant the intimacy between offenders and victims).
Finally, it is important to make clear that weapon lethality is not the only measure of violence (Black, 2004a; Phillips, 2003). Future research should test for the effect of social distance on other key parts of violent events, including (1) how many times a weapon is used (2) striking of different body parts, and (3) the seriousness of resulting injury. For example, some violent altercations involve one gunshot but others involve ten; some acts involve a gunshot to the head whereas others involve a shot to the foot; and, some gunshots result in death whereas others involve minimal injury or none at all. The theorizing of Black (1998) and Cooney (2006) suggest that as social distance between people increases then acts of violence may, for example, involve more applications of weapons, to more integral parts of the body, and more injury.
To conclude and summarize, there are several different paradigms that imply different explanations of weapon lethality, including the rationality/deterrence, social learning/culture, and opportunity/prevalence perspectives. All of these contribute to our understanding of violence generally and weapon use specifically (see, e.g., Copes et al., forthcoming; Felson & Messner, 1996; Ludwig & Cook, 2000; Phillips & Maume, 2007; Rosenfeld, 2000). This paper, however, has increased the sum of knowledge about violence and weapons by constructing, testing, and finding largely supportive results for a new theoretical principle that uses social distance between offenders and victims to explain weapon lethality (also see Black, 2004a; Cooney, 2006).
The beauty of criminology is its interdisciplinary nature. It is open to different “scientific cultures”, so to speak; a very admirable trait. Yet this trait is admirable for more than academic reasons. An interdisciplinary orientation to crime, violence, and weapon use is useful because it holds promise for increasing – indeed maximizing – our understanding of the distal and proximal causes of social harms. Obviously, some causal factors are easier to manipulate than others (e.g., the amount of CCTV versus the amount of intimacy between people or their cultural differences). Yet until we know the full cast of variables that do and do not affect social harms – including the use of more lethal weapons – our knowledge of any given effect will be opaque. In turn, our understanding of how best to reduce such social harms will be limited. For these reasons, the present paper has contributed to the criminology and criminal justice literature by revealing one more factor affecting victimization involving weapons: the social distance between victims and offenders involved in violent incidents.
Akers, R. L, & Jensen, G. F. (2006). The Empirical Status of Social Learning Theory. In F. T. Cullen, J. P. Wright, and K. R. Blevins (Eds.), Taking Stock: The Status of Criminological Theory (37-76). New Brunswick, NJ: Transaction Publishers.
Anderson, E. (1999). Code of the Street: Decency, Violence, and the Moral Life of the Inner City. New York: W.W. Norton & Company.
Baumer, E., Horney, J., Felson, R., & Lauritsen, J. L. (2003). Neighborhood Disadvantage and the Nature of Violence. Criminology, 41, 39-71.
Bentham, J. (1988 ). The Principles of Morals and Legislation. Amherst, NY: Prometheus Books.
Black, D. (1976). The Behavior of Law. New York: Academic Press.
Black, D. (1979). Common Sense in the Sociology of Law. American Sociological Review, 44, 18-27.
Black, D. (1983). Crime as Social Control. American Sociological Review, 48, 34-45.
Black, D. (1990). The Elementary Forms of Conflict Management. In School of Justice Studies, Arizona State University (Eds.), New Directions in the Study of Justice, Law, and Social Control (43-69). New York: Plenum Press.
Black, D. (1995). The Epistemology of Pure Sociology. Law and Social Inquiry, 20, 829-870.
Black, D. (1998). The Social Structure of Right and Wrong, revised edition, San Diego, CA: Academic Press.
Black, D. (2000a). Dreams of Pure Sociology. Sociological Theory, 18, 343-367.
Black, D. (2000b). On the Origin of Morality. In L. D. Katz (Ed.), Evolutionary Origins of Morality: Cross-Disciplinary Perspectives (107-118). Exeter, UK: Imprint Academic.
Black, D. (2004a). Violent Structures. In M. A. Zahn, H. H. Brownstein, & S. L. Jackson (Eds.), Violence: From Theory to Research (145-158). Cincinnati, OH: LexisNexis/Anderson Publishing.
Black, D. (2004b). The Geometry of Terrorism. Sociological Theory, 22, 14-25.
Black, D. (2004c). Terrorism as Social Control. In M. Deflem (Ed.), Terrorism and Counter-Terrorism: Criminological Perspectives (9-18). New York: Elsevier.
Braga, A. A. (2003). Serious Youth Gun Offenders and the Epidemic of Youth Violence in Boston. Journal of Quantitative Criminology, 19, 33-54.
Bureau of Justice Statistics. (2006). Homicide Trends in the U.S. Washington, DC: US Bureau of Justice Statistics. http://ojp.usdoj.gov/bjs/homicide/weapons.htm.
Campbell, B. (2010). Contradictory Behavior During Genocides. Sociological Forum, 25, 296-314.
Cohen, L, & Marcus, F. (1979). Social Change and Crime Rate Trends: A Routine Activity Approach. American Sociological Review, 44, 588-608.
Cook, P. (1980). Research in Criminal Deterrence: Laying the Groundwork for the Second Decade. In N. Morris and M. Tonry (Eds.), Crime and Justice: An Annual Review of Research, vol. 2 (211-268). Chicago, IL: University of Chicago Press.
Cook, P. J. (1983). The Influence of Gun Availability on Violent Crime Patterns. In M. Tonry and N. Morris (Eds.), Crime and Justice: An Annual Review of Research, vol. 4. Chicago, IL: University of Chicago Press.
Cook, P. J. (1991). The Technology of Personal Violence. In M. Tonry (Ed.), Crime and Justice: An Annual Review of Research, vol. 14 (1-71). Chicago, IL: University of Chicago Press.
Cook, P. J., & Ludwig, J. (2004). Does Gun Prevalence Affect Teen Gun Carrying After All? Criminology, 42, 27-54.
Cooney, M. (1997). Hunting among police and predators: The enforcement of traffic law. Studies in Law, Politics, and Society, 16, 165-188.
Cooney, M. (1998). Warriors and Peacemakers: How Third Parties Shape Violence. New York: New York University Press.
Cooney, M. (2006). The Criminological Potential of Pure Sociology. Crime, Law, and Social Change, 46, 51-63.
Cooney, M. (2009). Is Killing Wrong? A Study in Pure Sociology. Charlottesville, VA: University of Virginia Press.
Cooney, M., & Phillips, S. (2002). Typologizing Violence: A Blackian Perspective. International Journal of Sociology and Social Policy, 22, 75-108.
Copes, H., Kovandzic, T. V., Miller, J. M., & Williamson, L. (Forthcoming). The Lost Cause? Examining the Southern Culture of Honor Through Defensive Gun Use. Crime & Delinquency.
Decker, S. (1993). Exploring the Victim-Offender Relationship in Homicide: The Role of Individual and Event Characteristics. Justice Quarterly, 10, 585-612.
Decker, S. H., & Rosenfeld, R. (2004). Reducing Gun Violence: The St. Louis Consent-to-Search Program. NIJ Research Report. Washington, DC: U.S. Department of Justice.
Dixon, J., & Lizotte, A. J. (1987). Gun Ownership and the Southern Subculture of Violence. American Journal of Sociology, 93, 383-405.
Ewers, J. C. (1967). Blackfoot Raiding for Horses and Scalps. In P. Bohannan (Ed.), Law and Warfare: Studies in the Anthropology of Conflict (327-344). Garden City, NJ: The Natural History Press.
Fagan, J., & Wilkinson, D. L. (1998). Guns, Youth Violence and Social Identity in Inner Cities. In M. Tonry & M. H. Moore, Crime and Justice: A Review of Research, vol. 24 (105-188). Chicago, IL: University of Chicago Press.
Felson, R. B. (2009). Violence, Crime, and Violent Crime. International Journal of Conflict and Violence, 3, 23-39.
Felson, R. B., & Messner, S. F. (1996). To Kill or Not to Kill? Lethal Outcomes in Injurious Attach. Criminology, 34, 519-545.
Felson, R. B., & Paré, P-P. (Forthcoming-a). Gun Cultures or Honor Cultures?: Explaining Regional and Race Differences in Weapon Carrying. Social Forces.
Felson, R. B., & Paré, P-P. (Forthcoming-b). Firearms and Fisticuffs: Region, Race, and Adversary Effects on Homicide and Assault. Social Science Research.
Felson, R. B., & Steadman, H. J. (1983). Situational Factors in Disputes Leading to Criminal Violence. Criminology, 21, 59-74.
Gottfredson, M. R., & Hirschi, T. (1990). A General Theory of Crime. Stanford, CA: Stanford University Press.
Heide, K. M., & Petee, T. A. (2007). Weapons Used by Juveniles and Adult Offenders in U.S. Parricide Cases. Journal of Interpersonal Violence, 22, 1400-1414.
Heimer, K., & DeCoster, S. (1999). The gendering of violent delinquency. Criminology, 37, 277-317.
Horwitz, A. V. (1982). The Social Control of Mental Illness. New York: Academic Press.
Horwitz, A.V. (1990). The Logic of Social Control. New York: Plenum Press.
Jacobs, B. A. (2000). Robbing Drug Dealers: Violence Beyond the Law. New York: Aldine de Gruyter.
Jacobs, B. A., & Wright, R. (2006). Street Justice: Retaliation in the Criminal Underworld. New York: Cambridge University Press.
Jacques, S., & Wright, R. (2008a). Intimacy with Outlaws: The Role of Relational Distance in Recruiting, Paying, and Interviewing Underworld Research Participants. Journal of Research in Crime and Delinquency, 45, 22-38.
Jacques, S. & Wright, R. (2008b). The Relevance of Peace to Studies of Drug Market Violence. Criminology, 46, 221-253.
Jacques, S. & Wright, R. (2010a). Right or Wrong? Toward a Theory of IRBs’ (Dis)Approval of Research. Journal of Criminal Justice Education 21:42-59.
Jacques, S. & Wright, R. (2010b). Criminology as Social Control: Discriminatory Research and Its Role in the Reproduction of Social Inequalities. Crime, Law, & Social Change 53(4):386-396.
Killias, M., & Haas, H. (2002). Weapons in Violent Acts: Some Results of a Swiss National Cohort Study. Journal of Interpersonal Violence, 17, 14-32.
Kleck, G. (1997). Targeting Guns: Firearms and Their Control. New York: Aldine de Gruyter.
Kleck, G., & McElrath, K. (1991). The Effects of Weaponry on Human Violence. Social Forces, 69, 669-692.
Kuhn, T. S. (1977). The Essential Tension. Chicago, IL: University of Chicago Press.
Levy, P. S., & Lemeshow, S. (1999). Sampling of Populations: Methods and Applications, 3rd ed. New York: John Wiley and Sons.
Liberman, A. (2007). Adolescents, Neighborhoods and Violence: Recent Findings from the Project on Human Development in Chicago Neighborhoods. NIJ Research in Brief. Washington, DC: U.S. Department of Justice.
Long, J. S. (1997). Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage.
Long, J. S. & Freese, J. Regression Models for Categorical Outcomes Using Stata. Second Edition. College Station, TX: Stata Press.
Ludwig, J. (2005). Better Gun Enforcement, Less Crime. Criminology & Public Policy, 4, 677-716.
Ludwig, J, & Cook, P. J. (2000). Homicide and Suicide Rates Associated with Implementation of the Brady Handgun Violence Prevention Act. Journal of the American Medical Association, 284.5, 585-591.
Michalski, J. H. (2003). Financial Altruism or Unilateral Resource Exchanges? Toward a Pure Sociology of Welfare. Sociological Theory, 21, 341-358.
Nielsen, A. L., Martinez, R., Jr., & Rosenfeld, R. (2005). Firearm Use, Injury, and Lethality in Assaultive Violence: An Examination of Ethnic Differences. Homicide Studies, 9, 83-108.
Perkins, C. 2003. Weapon use and violent crime. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. NCJ 194820.
Phillips, S. (2003). The Social Structure of Vengeance: A Test of Black’s Model. Criminology, 41, 673-708.
Phillips, S., & Cooney, M. (2005). Aiding Peace, Abetting Violence: Third Parties and the Management of Conflict. American Sociological Review, 70, 334-354.
Phillips, S. (2009). Status and disparities in the capital of capital punishment. Law and Society Review, 43, 807-837.
Phillips, S., & Maume, M. O. (2007). Have Gun Will Shoot? Homicide Studies, 11, 272-294.
Popper, K. L. (2002). The Logic of Scientific Discovery. New York: Routledge Classics.
Pruetz, J. D., & Bertolani, P. (2007). Savanna Chimpanzees, Pan troglodytes verus, Hunt with Tools. Current Biology, 17, 412-417.
Rennison, C. M. (2002). Hispanic Victims of Violent Crime, 1993-2000. Washington, DC: USGPO.
Rennison, C. M., & Rand, M. (2003). Criminal Victimization 2002. Washington, DC: USGPO.
Rennison, C. M., & Rand, M. (2007). Introduction to the National Crime Victimization Survey. In J. P. Lynch, & L. A. Addington (Eds.), Understanding Crime Statistics: Revisiting the Divergence of the NCVS and the UCR (17-54). New York: Cambridge University Press.
Reynald, D. (Forthcoming). Guardians on Guardianship: Factors Affecting the Willingness to Supervise, the Ability to Detect Potential Offenders, and the Willingness to Intervene. Journal of Research in Crime and Delinquency.
Rosenfeld, R. (2000). Tracing the Brady Act’s Connection with Homicide and Suicide Rates. Journal of the American Medical Association, 284, 616-618.
Rosenfeld, R., Baumer, E., & Messner, S. F. (2007). Social Trust, Firearm Prevalence, & Homicide. Annals of Epidemiology, 17, 119-125.
Rosenfeld, R., Fornango, R., & Baumer, E. (2005). Did Ceasefire, Compstat, and Exile Reduce Homicide? Criminology & Public Policy, 4, 467-469.
Senechal de la Roche, S. (1996). Collective Violence as Social Control. Sociological Forum, 11, 97-128.
Shakur, S. (AKA Monster Kody Scott). (1993). Monster: The Autobiography of an L.A. Gang Member. New York: Atlantic Monthly Press.
Sherry, F. (1986). Raiders and Rebels: The Golden Age of Piracy. New York: Hearst Marine Books.
Stafford, M. C., & Warr, M. (1993). A Reconceptualization of General and Specific Deterrence. Journal of Research in Crime and Delinquency, 30, 123-35.
Stewart, E., & Simons, R. 2010. Race, Code of the Street, and Violent Delinquency: A Multilevel Investigation of Neighborhood Street culture and Individual Norms of Violence. Criminology, 48, 569-605.
Sutherland, E. H. (1937). The Professional Thief. Chicago, IL: University of Chicago Press.
Tedeschi, J., & Felson, R. B. (1994). Violence, Aggression, and Coercive Actions. Washington, DC: American Psychological Association.
Wadsworth, T., & Kubrin, C. E. (2003). Structural Factors and Black Interracial Homicide: A New Examination of Causal Processes. Criminology, 42, 647-672.
Welford, C. F., Pepper, J. V., & Petrie, C. V (Eds.). (2004). Firearms and Violence: A Critical Review. Washington, DC: National Academies Press.
Wells, W., & Horney, J. (2002). Weapon Effects and Individual Intent to Do Harm: Influences on the Escalation of Violence. Criminology, 40, 265-296.
Wilson, W. J. (1987). The Truly Disadvantaged: The Inner City, the Underclass, & Public Policy. Chicago, IL: University of Chicago Press.
Wilson, W. J. (1996). When Work Disappears: The World of the New Urban Poor. New York: Vintage Books.
Wright, R., & Decker, S. (1997). Armed Robbers in Action. Boston, MA: Northeastern University Press.
Zimring, F. (1968). Is gun control likely to reduce violent killings? University of Chicago Law Review, 35, 721-735.
Zimring, F. E., & Hawkins, G. J. (1973). Deterrence: The Legal Threat in Crime Control. Chicago, IL: University of Chicago Press.
Lethality is conceptualized as four points on a continuum: 1 = attempted or completed physical attack, 2 = attempted or completed attack with an object, 3 = attempted or completed attack with a knife or cutting implement, 4 = attempted or completed attack with a firearm. Violent acts with multiple levels of weapon lethality were coded based on the highest level of lethality present.
Three measurements of weapon lethality using dichotomous variables are utilized in the analyses. In the first model (Panel A), weapon lethality is measured as 0 = attempted or completed physical attack, and 1 = attempted or completed attack with an object, knife or firearm. The second measurement presented in Panel B operationalizes weapon lethality as 0 = attempted or completed physical attack or an attack with an object, and 1 = attempted or completed attack with a knife or firearm. The final model presented in Panel C examines weapon lethality measured as 0 = attempted or completed attack with body parts, object or knife, and 1 = attempted or completed attack with a firearm.
Social distance is defined as an actors’ social integration with others. Two measures of social distance are utilized in the analyses.
Defined as “the degree to which [actors] participate in each other’s lives, including the scope of their interaction, its frequency, duration, and their linkages in a wider network” (Black, 1998: 160), this concept is operationalized using a dichotomous measure of the victim and offender’s relationship: 0 = known offender and 1 = stranger.
Defined as the differences in aesthetics, morals, and knowledge between actors (Black, 1998: 160), cultural distance is operationalized using a dichotomous measure of the victim and offender race/ethnicity since similarities in race/ethnicity is presumed to denote greater closeness in cultural distance. Violent acts involving offenders and victims of the same race/ethnicity are coded as a zero, and those in which victim and offenders do not share race/ethnicity are coded as a one.
Pure sociology argues that behavior is governed by several social characteristics. In particular, it points to the importance of social status which is comprised of vertical status, radial status, symbolic status and corporate status. The analyses control for these correlates.
Vertical status increases as one’s income increases.
Annual HH Income
This concept is measured using 14 categories of unequal size representing the victim’s annual household income. These categories include, from lowest to highest: Less than $5,000, $5,000 to $7,499, $7,500 to $9,999, $10,000 to $12,499, $12,500 to $14,999, $15,000 to $17,499, $17,500 to $19,999, $20,000 to $24,999, $25,000 to $29,999, $30,000 to 34,999, $35,000 to $39,999, $40,000 to $49,999, $50,000 to $74,999, $75,000 and over.
Radial status increases as one’s involvement with work and family increases. Two measures describing the victim are used to capture radial status: marital status and employment.
This concept is measured using an ordinal level variable in which 1 = never married, 2 = widowed, divorced or separated, and 3 = married.
This concept is captured using a dichotomous measure in which 0 = has not worked in the past six months, and 1 = has worked in the past six months.
Corporate status increases as the number of actors with a common purpose increases. It is measured in two ways: number of offenders and the presence of bystanders.
Number of offenders
This concept is measured using a dichotomous indicator in which 0 = one offender and 1 = multiple offenders.
The presence or absence of bystanders is measured using a dichotomous variable in which 0 = no bystanders present and 1 = at least one bystander present.
Symbolic status increases as an individual’s conventionality or knowledge increases. This concept is operationalized using a measure of educational attainment.
Educational attainment is captured using an ordinal level variable based on the number of years of formal education the victim has completed. It ranges from 0 to 19 years.
Criminological research suggests additional relevant violence correlates not encompassed by pure sociology. Failure to include these correlates leads to a situation in which the influence attributed to variables included in the model actually represents the combined influence of included and excluded variables (Hanushek & Jackson, 1977).
Males are more likely to be victimized than are females with the exception of rape/sexual assault. To account for the relationship between gender and violence, a dichotomous variable where 0 = male, and 1 = female is included.
Violence and age are related. In general, younger people are victimized at higher rates than are older people. To account for this, victim’s age is measured using a continuous variable ranging from 12 years to 90 years. Persons age 90 or older are included in the “90” category.
Research shows that owning versus renting one’s residence is related to violent victimization risk. This is accounted for by a dichotomous measure in which 0 = owned, and 1 = rented.
MSA of victim
Location is also relevant to the behavior of violence. To account for this characteristic, three dichotomous measures are included: urban (reference category), suburban, and rural.
Research shows that offending and age of the offender is related. To account for this, offender’s age is controlled using a series of four dichotomous indicators: less than 18 years, 19-29 years of age, 30 years of age or older, and mixed age group of offenders. Offender age 18 or less serves as the excluded reference category.
Females are less likely to perpetrate violence. To control for this, offender’s gender is measured using three dichotomous variables: male, female, group of males and females. Male offender is the excluded category.
Location of incident
A series of three dichotomous measures account for the relationship between location of the incident and victimization risk: In or near the victim’s home, in or near the home of a victim’s friend, family, or neighbor, and some other location. In or near the victim’s home is the reference group.
 Although they do not test this particular idea, there are existing quantitative studies that suggest such a proposition may have validity. Nationally, for instance, firearms are used in approximately 14 percent of violent victimizations between strangers, but are used in fewer than 6 percent of incidents between intimate partners (Perkins, 2003). Another project found that in the city of St. Louis, three-quarters of stranger-homicides were committed with a gun, whereas that total for homicides between friends was roughly half (Decker, 1993).
 Space limitations do not allow for an extensive review of research findings related to these theories. For a comprehensive review, see Welford, Pepper, and Petite, 2004.
 Note that pure sociology’s notion of social structure is different from more common notions of the concept found in the criminal justice literature (see, e.g., Wadsworth and Kubrin, 2003). The difference is that pure sociology sees the case/situation as the unit of analysis whereas other conceptions of social structure see the person/group as the unit of analysis. It is important to analyze the characteristics of particular situations because this approach recognizes that people/groups do not always behave the same way; in other words, the behavior of people/groups varies depending on the situation they are involved in. For instance, even the most violent people are not violent in all of their interactions; this variability between situations is worthy of attention (Black, 2004a: 147).
 It is potentially true that Black (2004a) has contributed to the study of violence by explicitly bringing the concept of weapon lethality to the attention of criminologists. Although rationality/deterrence, social learning/culture, and opportunity/prevalence theories of weapons imply distinct explanations of weapon lethality (as described above) they do not directly address this concept, at least to our knowledge. Thus, Black’s theory of weapon lethality may very well be the first explicit explanation of it.
 For example, the NCVS offers information on non-fatal violence only. Adding fatal violence to the proposed research would not change the substantive findings given that homicide makes up less than 1% of all violence in the United States.
 The NCVS is not a proper tool for testing the effect of social status on violence because it does not systematically gather information on offenders’ income, employment, marital status, or formal education, although the survey does collection such information for victims. As relates to testing pure sociology theories, the absence of such data is a limitation because tests should examine the difference in status between an offender and victim and not simply one or the other’s status. However, the NCVS is useful for testing the social distance aspects of pure sociology theories because the survey does collect information on the victim-offender relationship (a direct measure of relational distance) and their race/ethnicity (an indirect measure of cultural distance).
 Since we have generalized Black’s idea to explain predation and retaliation with the same factor (social distance), this limitation does not directly apply to our results, but does preclude our ability to make claims about how retaliatory and predatory violence may behave differently. For a short discussion of differences in violent predation and retaliation with respect to relational distance, see Cooney and Phillips (2002: 86).
 The analyses are based on data beginning in quarter 3 of 1993. This marks the first quarter in which pseudostratum and secucode (standard error computation unit code) variables were included on the NCVS data file. These variables in conjunction with the ‘person weight’ on the file enables analyses that take into account the complex sample design of the NCVS (e.g., the survey weighted regression functions in STATA). The last year of data used in the analyses (2004) represents the most recent data available at the time of the analyses.
 Note that weapon lethality is not simply a measure of weapon presence though these concepts are related. Weapon presence fails to consider instances in which physical assault was attempted or completed. Because of the overlap between the two concepts however, weapon presence is not included in the models. Correlations between the two variables are as follows: no weapon = -0.7214; firearm = 0.7103; knife = 0.5126; other type weapon = 0.2378; and unknown = -0.1182.
 The measure of weapon lethality does not measure actual harm done or injury to the victim, but rather is an indicator of the potential amount of damage that can be generated by an object The correlation between injury (none, minor, serious) and the 4-point measure of weapon lethality is 0.1054. The coding of weapon lethality is not discussed here given issues related to analytic approach. Specific coding of this concept can be found in the Analytic Technique section below. In incidents where multiple levels of weapon lethality are present, incidents are coded based as the highest level of weapon lethality.
 A possible conceptualization of relational distance is a three-category variable: intimate, other known, and stranger. This conceptualization indicates that ‘intimates’ are closer relationally than are ‘other knowns.’ Unfortunately, given the way the data are collected, it is not possible to ensure this. Research routinely defines intimates as “current or former spouse, boyfriend or girlfriend”. The problem is that it is not clear that an ‘ex’ is closer relationally than some “other known” (e.g., one’s best friend) or even “stranger.” While it is possible to parse current and former spouses in the NCVS, this cannot be done with current and former boy- and girlfriends. Additional analyses were conducted using a three-category relational distance measure. Findings indicate that in the models presented in Panels B and C, weapon lethality used does not differ between “knowns” and “intimates” (findings are available upon request). Given conceptual considerations and based on a consideration of additional analyses, we opted to go with the two-category presentation. This portrayal is accurately represents the theory under consideration and does not obscure any important findings.
 The NCVS does not provide information for every aspect of social structure, such as data on offender’s wealth, employment, family, education, or data on the legal history (e.g., past arrests) of the victim or offender.
 Some frequently used correlates of victimization are not used as controls in these models. First, race/ethnicity of the victim is not included given issues of multicollinearity between white non-ethnics and cultural distance (correlation = -.2869) and between ethnics (any race) and cultural distance (correlation = 0.4774). Multicollinearity also precluded the inclusion of offender’s race in the models. Correlations between cultural distance and offender race are as follows: white = -0.5802; black = 0.2601; other = 0.4001; and unknown = 0.2474.
 Another option is the use of a proportional odds logistic regression model (Long, 1997; Hosmer & Lemeshow, 2000). This approach models the probability that weapon lethality falls in the four intervals defined by the three cut-points based on the measurement of social distance. This proportional odds model compares a single outcome response to one or more reference categories and resulting inferences from them lend themselves to a general discussion of direction of response and without a focus on specific outcome categories. We opted not to use this approach for several reasons. First, it is not possible to properly test for the assumption of parallel regression using post-estimation techniques when using the survey weight procedures required by NCVS data. As an exercise a model was run using non-survey weighted ordered logistic regression and post-estimation tests demonstrated that the assumption of parallel regression was violated. That is, these post-estimate tests indicated that should a model be run separately for each category of the dependent variable, slopes of the regressions would not be parallel – a requirement of using a proportional odds model (Long & Freese, 2005) . Second, the use of a series of survey-weighted binary logistic regressions provides more useful information to the reader. For example, with a proportional odds logistic regression, only one set of coefficients are presented for each independent and control variable. By using the multiple models approach, one can examine how coefficients vary in terms of value and sign across models.
 For more detailed information about weighting in the NCVS, see Rennison and Rand (2007).
 It is possible to examine the influence of social distance as a composite measure on weapon lethality instead of the way it is presented in our analyses. That is, rather than using two distinct measures of social distance (cultural distance and relational distance), one can create a scale reflecting the sum of known/stranger (0,1) and similar/different race/ethnicity (0,1) into a single measure. This approach enables one to test the influence of social distance in total on weapon lethality. Models using this measurement scheme were estimated and findings demonstrate no substantive difference in the coefficients of all included variables. Coefficients associated with the summated “social distance” measure were significant at the p=.0000 level for models presented in Panels A, B and C (Panel A, b=0.24, SE=0.04, p=0.0000 and AOR=1.27; Panel B, b=30, SE=0.05, p=0.0000 and AOR=1.35; Panel C, b=0.30, SE=0.08, p=0.0000 and AOR=1.35). Given that estimating the social distance measure using two variables representing relational distance and cultural distance offers findings obscured when using the composite measure, and given space limitations, only the dual-measure approach is presented. Findings from the composite measure analyses are available upon request.
 Moreover, it is also possible that for cases of retaliatory violence, the effect of social distance on weapon lethality depends on the nature of the conflict. An ideal test of pure sociology theories of conflict management – such as retaliatory violence – would control for the type of wrongdoing being responded to (e.g., robbery versus burglary; or, less money versus more money stolen); see Black (1979, 1983) for details. The NCVS does not allow for this level of specificity and so future work should explore this possibility with more suitable data sets.
 For instance, we found that high-lethality weapons rarely enter into conflicts recorded in the NCVS (11% in the case of knifes/guns, 4.5% in the case of guns compared to 76% of cases with a bodily weapon). However the results suggested that the presence of such weapons is predictable. In other words, whether a highly lethal weapon is brandished in a conflict is not a random phenomenon, but varies closely with the social distance between disputants.