Thomas A. Petee, Department of Sociology, Auburn University

Several high profile cases of mass murder have raised public consciousness on this form of multiple homicide. Incidents on a commuter train in New York, in a restaurant in Kileen, Texas, in schools in Jonesboro, Arkansas, and Littleton, Colorado, an office building in Atlanta, Georgia, and the numerous post office shootings have garnered a great deal of scrutiny over the past several years. However, despite the media focus, very little attention has been given to mass murder in the research literature.

For our purposes, mass murder is defined as the killing of three or more people in one place, at one time. This definition has some flexibility so long as the incident occurs in a limited area over a limited time period (i.e., generally less than one day). Although the victim criterion has varied in some studies (see Levin & Fox, 1985; 1996), this number seems sufficient because it eliminates many of the more common type of multiple homicides – especially lover's triangles – without being overly exclusive.

In the little existing literature on this topic, the concern has primarily been with delineating between different forms of multiple murder, or attempting to create a typology for mass murder (Busch & Cavanaugh, 1986; Dietz, 1986; Gresswell & Hollin, 1994; Holmes & Holmes, 1994; Kelleher, 1997; Levin & Fox, 1996; Rappaport, 1988; Rowlands, 1990). However, many of these studies were based on either a single case study, or a very limited number of cases. Despite the work that has been done to date, research on mass murder is clearly in its infancy (see Levin & Fox, 1996).

Prior Attempts to Categorize Mass Murder

Dietz (1986) provided one of the early attempts to classify incidents of mass murder. According to him, mass murder offenders typically fall into one of three categories: family annilhilators – who target family members, usually in a home setting; pseudocommandos – who fancy themselves as military types, have a fascination with firearms, and frequently plan their offense with some detail; and, set-and-run killers – who murder more remotely (i.e., with bombs or poison) so as to remove themselves from the crime scene before the murders occur, and allow for the possibility of escape.

Holmes and Holmes (1994) built upon Dietz's typology, and provided two additional categories: disciples – who are unduly influenced by a charismatic leader to kill; and, disgruntled employees – who retaliate for what they feel was bad treatment from their employer.

As can been seen from the expanded typology above, it makes use of several bases for classification, mixing together motivation, the specific relationship between the victim and the offender, and technique for killing. Unfortunately, this leaves a system that is neither mutually exclusive nor exhaustive in describing cases of mass murder. There are several offender, for example, who would fit in more than one of the categories. On the other hand, there are a number of offenders who do not fit into any of these offender types.

By contrast, Levin and Fox (1996) use a single criterion– motivation – in creating their own typology. According to them, most mass murder cases are motivated by vengeance. These revenge-based homicides have three subtypes: individual-specific – where the offender targets particular people; category-specific – where particular groups of people are targeted; and, nonspecific – where the murders are precipitated by the offender's paranoia, and where the offender does not have specific targets. Additionally, mass murders may also be classified as occurring out of a warped sense of love – as would be the case with some family massacres, or oriented toward profit – as with felony murder, or result from acts of terrorism. Even here, however, the categorization is not exhaustive – failing to take into account a significant amount of mass murder.

Kelleher's classification system (1997) suffers from some of the same limitations. He identified seven groupings for mass murder cases: revenge – which is essentially the same as that found in Levin and Fox, although without the targeting; perverted love – which again, is similar to Levin and Fox's "warped sense of love" category, involving the killing of family or loved ones; politics and hate – which is ideologically motivated, and usually involves acts of terrorism; sexual homicide – which is typically the motivation in serial murder, and which actually applies to only two cases; mass execution – which involves the contract murder of several people in a single incident; the insane – for whom Kelleher can find no other motivation than mental illness; and, those cases for which he can find no motivation. While some of these categories are substantive, a few of them are questionable. The "insanity" category is especially problematic, underminding the potential of this classification system to at least have mutually exclusive offender groupings.

In summary, all of the above classification systems are lacking in one way or another in describing mass murder cases. Most suffer from trying to use too many criterion as the basis of categorizing offenders. Ultimately, a more comprehensive classification has to be devised which does not suffer from these categorization problems.

Building a New Typology

A more comprehensive typology can be constructed using data from mass murder incidents occurring between 1965 and the present. There are two criteria for the categories in this typology – offender motivation and target selection (see Petee, Padgett and York, 1997). This new classification system consists of eight substantive categories, and an additional grouping which is residual in nature. As with Levin and Fox (1996), the new typology recognizes that anger and/or revenge is one of the major motivating factors for mass murder. Three of the categories in this classification scheme focus on anger/revenge as the primary motivation, with all cases involving a long standing conflict situation, and considerable deliberation on the part of the offender. The distinguishing characteristic of each anger/revenge type is the type of target selected.

The anger/revenge - specific person(s) target category involves a situation where an offender seeks revenge against particular people. The targets in this type of an offense are known to the offender, and who were perceived to have wronged the offender in some way. There are precise targets for killing, and the offender will sometimes avoid injuring other innocent bystanders unless they interfere with the offender in some way. The case of Gang Lu, who killed five people at the University of Iowa in 1991, best exemplifies this type. Lu was angry over not having been nominated for a prestigious award for his doctoral dissertation, and he blamed faculty members in the Department of Physics. Lu shot and killed two faculty members from his dissertation committee, the chair of the Department of Physics, a university administrator, and the student who had actually won the dissertation award that Lu coveted. All of these victims were specifically targeted because they were the focus of his anger. One additional victim, the secretary for the administrator who was killed, was shot because she attempted to stop Lu, although she managed to survive the assault.

The anger/revenge – specific place target category involves the targeting of a determinate location for the offense. The site that is targeted for the mass murder usually has some sort of symbolic significance for the offender, and is the focus of the anger. Frequently, the targeted location is connected to an agency or organization that has some authority or control over the offender. The case of Edwin Grace illustrates this form of mass murder. Grace murdered six people in New Jersey in 1972 in an office building that housed an employment agency that was unable to help Grace secure a job. In this case, Grace apparently targeted the location, not specific people, and in fact killed people not even associated with the employment agency.

Some offenders, while motivated by anger and the need for vengeance, have a victimization pattern that is either unclear or indirect. This offender, the anger/revenge – diffuse target type, actually follows one of two different victimization patterns. In the first type of pattern, the offender targets specific groups or categories of people, although specific persons are not targeted. In this pattern, the offender usually has a deep hatred for the group in question. There are a number of high profile cases which fit in this category. Colin Ferguson, who killed six passengers on a Long Island commuter train in 1993, apparently had a hatred of White people, and his victims were simply easily accessible targets who fit the criteria for his rage. In 1989, Marc Lepine killed fourteen women at an engineering school in Montreal because he blamed "feminists" for ruining his life. Also in 1989, Patrick Purdy shot and killed five and injured twenty nine at a schoolyard in Stockton, California. Purdy was an avowed racist who had a hatred of people from Southeast Asia, and many of his victims were Asian.

The second pattern for this category involves an offender who simply lashes out at whoever is available as a target. Such was the case with William Cruse, who killed six people and wounded fourteen others in Palm Bay, Florida in 1987. Cruse had a history of run-ins with neighbors and other members of the community, and was often described as being "filled with rage". He opened fire in his neighborhood after some kids had run across his property. He then took his car to a local shopping center, and approached a major grocery store, only to be deterred because he attempted to enter through an exit door. Frustrated, he got back in his car, drove across the street and entered another store, killing several customers in the process. After his arrest, Cruse admitted that he was angry at everybody. The diffuse nature of this offense can be seen in the lack of connection between the offender and his victims, which often leads to a depiction of this kind of a crime as being random.

The domestic/romantic category, also involves two different types of victims, although the underlying pattern is primarily the same. The domestic cases involve family members in some form of conflict. The typical victimization pattern here is for a husband to kill his wife and children. This represents the most frequent form of mass murder in the United States. Usually this subtype occurs in a household setting, although occasionally they have been known to occur in a public setting. The murders are precipitated by stress or conflict, and is usually cumulative in nature. John List murdered his wife, mother and three children in New Jersey in 1972, in an apparent attempt to have a fresh start in his life. He shot them each in the back of the head, execution-style, before laying out the bodies in the living room and fleeing the crime scene. He was arrested seventeen years later, having assumed a new life in Virginia.

Other cases in this category involve a situation where an offender has been rebuffed by a romantic interest and retaliates. In fact, this type of mass murder case often involves stalking behavior prior to the actual murder. The offender in these cases often exhibits destructive possessive behavior toward their romantic interest that parallels behavior seen in the domestic cases. This is a type of "fatal attraction" that goes beyond the specific individual and spills over to friends, relatives, or co-workers of the intended victim. The case of Richard Farley best illustrates this kind of offender. Farley had an obsession with Laura Black, a co-worker at a defense contractor in Sunnyvale, California. Black put up with Farley's unwanted attention for two years before she filed a sexual harassment complaint with their employer, resulting in the company firing Farley in 1986. Undaunted, Farley continued to harass Black, finally forcing her to obtain a temporary restraining order in 1988. Two weeks afterward, and only one day prior to a scheduled court hearing, Farley showed up at the contracting office, shooting and killing seven employees of the defense firm, as well as wounding three others, including Black.

The direct interpersonal conflict type contrasts with the anger/revenge forms of mass murder in that the former involves a conflict situation that is more immediate – as opposed to the long-simmering anger typical of the latter. These cases usually result from a relatively trivial dispute, and the offender is often a volatile person who reacts violently to the conflict situation. The case of Ray Ojeda, who killed three in San Antonio, Texas in 1991 is fairly characteristic of this type of mass murder. Ojeda had been involved in a minor traffic accident, and began to argue with the driver of the other vehicle. As the argument escalated, Ojeda went back to his car, pulled out a gun, and shot the other driver. When two bystanders tried to intervene, Ojeda killed them as well.

Felony related mass murder involves the killing of three or more people as the result of the commission of another felony – usually robbery. This is an often overlooked category, and represents the most frequent form of mass murder occurring in a public setting. The principal motivation is instrumental, although the murders themselves usually result from either a perceived need to eliminate witnesses, or as a consequence of the offender losing control of the situation. Robert Melson and Cuhuatemoc Peraita robbed a Popeye's Chicken in Gadsden, Alabama in 1994. After gathering all of the money, they herded the employees into a walk-in freezer and shot each of them, killing three and seriously wounding one in what amounted to an attempt to avoid detection. This case is also representative in that it involved multiple offenders – more than half of these felony related mass murders involved an accomplice.

Gang-motivated mass murder involves incidents such as drive-by shootings or gang confrontations. These murders often have a dispute element as in direct interpersonal conflict murders, except that the offender has a gang affiliation, which dramatically changes the dynamics of the offense. The gang-motivated cases are often perpetrated by multiple offenders, and have a level of organization that is typical of gang activity. The dispute between two Vietnamese gangs in New York in 1990 is characteristic of this type of offense. After an argument in a Manhattan bar, David Tai, Tommy Tam, Peter Wang, and Ywai Yip followed the members of a rival gang into a parking lot, shooting and killing all three execution style.

Politically motivated mass murder usually involves acts of terrorism. The motivation is primarily ideological, usually for some political cause – although there may be a religious connection – or for political change. The nature of this type of offense is such that typically the offender is absent from the scene of the crime when the victims are actually killed – a sharp contrast to the more intimate involvement in most other mass murder cases. The bombing of the Murrow Building in Oklahoma City in 1995 or the World Trade Center in New York in 1993 serve as examples of this form of mass murder.

Non-specific motive cases are essentially those which cannot be readily classified, usually because of ambiguity as to the offender's motivation – essentially only the offender knows the reason for the murders. Very frequently, the offender has a history of psychological problems or mental illness. The 1988 case of Clem Henderson that occurred in Chicago illustrates this problem. In a span of less than 20 minutes, murdered four people. First, walked into an auto parts store, where he shot and killed two employees. He then walked across the street to a school, where he killed a janitor and a policewoman, before being killed himself in an exchange of gunfire with the policewoman's partner. Although Henderson had a history of mental illness, no connection could be found between him and his victims or with the locations where the murders took place.


Busch, K.A., & Cavanaugh, J.L. (1986). The study of multiple murder: Preliminary examination of the interface between epistemology and methodology. Journal of Interpersonal Violence, 1, 5-23.

Dietz, P.E. (1986). Mass, serial, and sensational homicides. Bulletin of the New York Academy of Medicine, 62, 477-491.

Gresswell, D.M., & Hollin, C.R. (1994). Multiple murder: A review. The British Journal of Criminology, 34, 1-14.

Holmes, R.M., & Holmes, S.T. (1994). Murder in America. Thousand Oaks, CA: Sage Publications.

Kelleher, M.D. (1997). Flashpoint: The American mass murderer. Westport, CT: Praeger Publishing.

Petee, T.A., Padgett, K.G., & York, T.S. (1997). Debunking the stereotype: An examination of mass murder in public places. Homicide Studies, 1, 317-337.

Rappaport, R.G. (1988). The serial and mass murderer: Patterns, differentiation, pathology. American Journal of Forensic Psychiatry, 9, 39-48.

Rowlands, M. (1990). Multiple murder: A review of the international literature. Journal of the College of Prison Medicine, 1, 3-7.