Journal of Police and Criminal Psychology

, Volume 25, Issue 2, pp 59–66

The Nature of Mass Murder and Autogenic Massacre


    • The Pennsylvania State University Harrisburg
  • Eric S. Holmes
    • The Pennsylvania State University Harrisburg
  • Ashley Rhom
    • The Pennsylvania State University Harrisburg

DOI: 10.1007/s11896-009-9059-6

Cite this article as:
Bowers, T.G., Holmes, E.S. & Rhom, A. J Police Crim Psych (2010) 25: 59. doi:10.1007/s11896-009-9059-6


Incidents of mass murder have gained considerable media attention, but are not well understood in behavioral sciences. Current definitions are weak, and may include politically or ideological motivated phenomenon. Our current understanding of the phenomenon indicates these incidents are not peculiar to only western cultures, and appear to be increasing. Methods most prominently used include firearms by males who have experienced challenging setbacks in important social, familial and vocational domains. There often appears to be important autogenic components (Mullen Behavioral Sciences and the Law (22)3, 2004), including dysthymic reactions and similar antecedents. There have been observations of possible seasonal variations in mass murders, but research to date is inadequate to establish this relationship. It is recommended behavioral sciences and mental health researchers increase research efforts on understanding mass killings, as the current socioeconomic climate may increase vulnerability to this phenomenon, and the incidents are not well understood despite their notoriety.


Mass murderSpree killingsMultiple homocides

Mass Murder Research

To date, there is a limited amount of research literature on mass murder and/or mass killings. Previous studies have been limited to generally descriptive accounts and typologies of offenders found in well publicized cases of mass homicide from 1900 to the beginning of the 21st century (Duwe 2004). In addition, Duwe (2005) has noted that information on mass killings has been almost entirely provided by the news media. When one looks at the exciting commonalities and differences between the several types of mass murder (i.e. serial, mass, spree), it is not difficult to understand why research on mass murder, particularly, is limited in the literature. When compared to serial murder, Fox and Levin (2003) point out that mass killings: a.) do not present a great challenge to law enforcement officials because the perpetrator is more often than not, found dead at the crime scene, either from a self-inflicted injury or action taken by the police; b.) do not produce the same level of terror and worry, usually due to a single tragic event, leaving individuals in horror, but not concerned with their lives; c.) are restricted geographically, in that the murder takes place in a single neighborhood or location, keeping the crime generally localized; d.) are limited in data, thus, if the offender does not survive, motivation is hard to distinguish, often leaving the matter ignored by researchers; and e.) do not have the same sexual and sadistic characteristics as serial murders, also giving rise to why mass murder research is often overlooked. As a result, there has been relatively limited interest in mass or spree types of killings, despite the dramatic nature of those events. However, the nature of mass murders is of great interest to behavioral sciences and mental health fields, as understanding this phenomenon could lead to great prevention or protective interventions.

Multiple Homicide Definitions

From a legal point of view, homicide and murder are two distinct crimes. Palermo (2007) clarifies the distinction by defining homicide as “the killing of one human being by another” and murder as “the crime of unlawfully killing a person, especially with malice or aforethought” (p. 3–4). In addition, Palermo (2007) further differentiates the two concepts into single or multiple homicides, in that multiple murders would be classified as mass, spree, and serial murder. Due to the lack of research in the area of multiple murders, it is difficult to pin point an exact definition as to what constitutes and separates mass murder from spree murder or spree murder from serial murder. With this in mind, spree killing generally takes place within a given period of time, such as hours or days, with some kind of interval or time break between the killings (Palermo 2007, p. 15). On the other hand, serial murder, the most commonly researched class of multiple murders, is characterized by its string of victims killed one at a time over a period of weeks, months, or even years (Fox and Levin 2003). That is, the murders occur in separate incidences, sometimes separated by long time intervals and great distances (Dietz 1986). Holmes and Holmes (1998) explain that serial murder, when compared to mass murder, brings with it a sense of community terror and social paranoia because the perpetrator is at large, striking without warning, operating normally in society, whereas mass murder creates a direct and severe sense of panic that is typically short lived (i.e., usually due to the perpetrators on scene execution, either by police or his own hand). Therefore, timing may be viewed as a key characteristic when distinguishing serial, spree, and mass murder (Fox and Levin 2003).

The exceptions to this limitation seem to exist in widely publicized specific examples of mass murder, such as the Columbine and Virginia Tech school shootings of 1999 and 2007 respectively. In addition, it has been suggested that what “events” the media or similar public vehicles select (or perhaps do not select) as those to publicize or otherwise broadcast to society may affect how that society perceives and understands the prevalence, history, and definition of “mass murder” as a construct (Duwe 2005). This suggests that this stereotypical or publicly-held representation of mass murder is perhaps one that has been reinforced through such vehicles as the mainstream news media and popular criminal investigation television shows. This limited definition of mass murder such as proposed by Duwe (2005) could indicate that society as a whole may have a misinformed representation of these events and the offenders who orchestrate them, as they exist presently.

Mass Murder Definitions

In using an appropriate definition of mass murder, one must not be too rigid, as research does not appear to have an exact operational definition of what comprises mass murder. Often, the definition depends on the total amount of time over which the murders took place, along with the number of persons killed (Duwe 2004). Holmes and Holmes (1998) suggested that when creating a definition of mass murder, one needs to take into consideration the number of victims slain, the location of the murders, and timing of the killings. Mass murder is generally limited to incidents that took place within a 24-hour period, or single event (Aitken et al. 2008; Dietz 1986; Duwe 2004; Duwe 2005). Although mass murders usually occur at one time, one also finds incidents that occur at different times, minutes or hours apart, and in different locations, as in a few blocks away, which would still fall under the mass murder umbrella (Holmes and Holmes 1998). Similarly, Dietz (1986) noted that the number of victims slain is arbitrary, but the type of victim is important, such as a family member, co-worker, or stranger. In general, mass murder appears to have a 3 or 4 fatality criterion (Aitken et al. 2008; Dietz 1986; Duwe 2004), although multiple victims may be as few as two to some investigators (Palermo 2007; Levin and Madfis 2009).

Levin and Madfis (2009) have defined mass murder as the “antisocial and non-state-sponsored killing of multiple victims during a single episode at one or more closely related locations” (p. 1227). Similarly, Aitken et al. (2008), cited the concept of “mass murder by intention,” referring to those incidences where fewer than the required number of people are killed, despite a clear intention and attempt to murder more people, such as firing multiple rounds of ammunition into a crowd of people (p. 262). Mullen (2004) has also created the term autogenic massacre to refer to mass killings, those that involve individuals who kill in pursuit of personal goals emerging from their particular psychopathology. From this perspective, the term autogenic massacre captures the essential feature of mass murder, that is, the killing being generated primarily out of the perpetrators own problems and personal attitudes (p. 312). Mullen (2004) pointed out that due to the relationship found between killers’ intentions and the victims, mass killings may further be divided into 3 different forms: victim specific mass killings, instrumental mass killings, and massacres (p. 313). Victim specific mass killings are incidents where death of a particular victim(s) is the intended outcome of the killings, such as in family slayings, revenge killings, cult killings, and gang killings. Instrumental mass killings are those occurrences’ in which murder is a means to an end and the intention of the perpetrator is to advance his or her objectives, which is generally criminal in nature. That is, the killer and victim(s) are connected via their membership of a very broad group, such as in terrorist killings and felony related killings. Finally, there are massacres involving indiscriminate killings in which killing people is the prime aim and victims are chosen by chance from some broad social grouping. Mullen (2004) further separates massacres into two broad types. The first would include massacres due to social conflict, where violence erupts between two antagonizing groups, while the second form of massacres would involve those that resulted from the highly personal agenda coming from the perpetrators own social situation and psychopathologies (i.e., autogenic). Therefore, Mullen (2004) concluded that autogenic massacres can overlap with other types of mass killings, in that, an individual may begin killing a selected victim, or victims, for the purpose of revenge, before proceeding to random killings to express his or her overall rage at the world (p. 314).

Despite a shortage of current quantitative studies regarding mass murder, there does appear to be a selection of qualitative literature that spans several decades. For example, Levin and Fox (1996) provided analysis and data on all of the massacres (operationalized as the murder of 4 or more people in one location and at the same time) reported in the U.S. between 1976 and 1989. Shortly thereafter, Petee et al. (1997), released a study in which they investigated mass murders (defined as the murder of 3 or more people) occurring in public places and at the same time in the U. S. between 1965 and 1996, and presented a typology for mass murder based on the examples and data they presented. There also exists numerous examples of rich case studies to compliment the qualitative chronological timeline provided by Levin and Fox (1996) and Petee et al. (1997) mentioned above (see Mullen 2004; Palermo 1997).

In a recent case history of mass murders, Mullen interviewed five mass murderers who survived their attacks, which is often not the case in mass killings. Mullen (2004) indicated the need to differentiate whether an event is to be considered a “mass murder” (also called “mass killing”) or a “massacre,” which is proposed as a specific sub-category of mass killing. This specification appears to be necessary not only due to the existing definitional ambiguity of what “mass murder” should be as a construct, but also in that “mass murder” and “massacre” may not be constructs that are completely interchangeable. For example, Fox and Levin (2005) and Petee et al.’s (1997) chronological studies use similar definitions for their constructs of mass murder (defined as four and three victims respectively at the same time and in the same location), yet Levin and Fox (1996) refer to their definitional act as a “massacre” whereas Petee et al. (1997) refer to their construct as a “mass murder.”

Mullen (2004) proposes that the difference between these two constructs lies not just in quantifying a body count or time frame of the act, but also in regards to the target population and the motive of the offender. Mullen (2004) posits that mass killings have historically been depicted as occurring as a result of ideological principles such as war, religious differences, or political propaganda. The author then asserts that the term “massacre” is also considered a type of mass murder, but one that implies “indiscriminate killing” on behalf of the offender, where the victims may not represent a specific cohort and therefore could be chosen at random (Mullen, 2004). Mullen (2004) recounts five case histories that are utilized as examples of a new concept of “massacre” termed “autogenic massacre,” by which the goal of the offender is simply to kill multiple people, chosen either at random or happening by chance at a specific location as a result of a deep personal agenda or psychopathology. This definition suggests the offender acting out from a primarily internal motivation versus being prone to certain external stressors or triggers observed in other categories of “mass murder.”

Prevalence of Mass Murder

Researchers have held the notion that mass killings were a rare occurrence in the United States prior to the 1960’s (Duwe 2004; Duwe 2005; Fox and Levin 2003). In reality, through selected research, it is apparent that this notion may not stand true. For example, Duwe (2004, 2005) looked at 909 mass killings in the U.S. between 1900 and 1999. He found that the first mass murder wave occurred during the 1920’s and 1930’s, with familicides accounting for nearly two-thirds of the incidents, and making up 52% of the mass killings between 1900 and 1975. Duwe (2004) suggests this was due to the agricultural depression and consequences of WWI that were taking place at the time. In addition, he cites the 71% increase in the divorce rate during the 1920’s as creating a motive for revenge.

The murder wave in the 1920’s and 30’s was then followed by one that began in the mid 1960’s, with serial killings being increasing common from 1900–1940 and becoming more prevalent beginning in the 1960’s. Felony-related massacres, those committed in connection with criminal activities, such as robbery or burglary, were infrequent until the 1960’s when they began to increase (p. 754). Thus, Duwe (2004) concluded that researchers were correct in noting that a mass murder wave began in the 1960’s, but the increase wasn’t unprecedented nor did it continue to escalate with each passing year. On the contrary, there appeared to be an increase in high body-count killings over the last few decades, with 25 of the deadliest mass murders in his study taking place since 1980 (p. 752). Moreover, following 1966, Duwe (2004) explained that there was an increase in the frequency of mass public shootings. From 1900 to 1965, there were only 21 incidents of mass public shootings, whereas since 1966, there have been 95 more, with the most apparent since 1980 (p. 755). In general, 1986 marked the beginning of the rise in workplace violence, and 1997 began a string of school shootings. Prior to 1997, mass murder had taken place in schools, but none were committed by juveniles. In his 2005 study, Duwe found that of the 828 mass killings between 1900 and 1996, 47 incidents (6%) involved juveniles as offenders, but not one was a mass public shooting (Duwe 2005).

These findings are supported by Aitken et al.’s (2008) research that showed 14 of the deadliest school attacks occurred in the last 8 years and span 8 different countries (i.e., 6 in the U.S., 2 in China, and 1 each in Canada, Britain, Japan, Yemen, Germany, and Finland—p. 265). In addition, Hempel et al. (1999) identified 30 mass murders in the United States and Canada from 1949 to 1998 and found that the majority of them (21 out of 30) occurred after 1985. In terms of exact numbers, Fox and Levin (2003) identified 497,030 murdered individuals in the U.S. from 1976 to 1999 and found that of these, 3,956 were killed in incidents claiming four or more victims. More specifically, there were 599 mass killings involving 2,800 victims and 826 killers (p. 49). Thus, there are roughly 2 incidents of mass murder per month in the United States claiming more than 100 victims annually (Fox and Levin 2003). Therefore, Duwe (2004) appears to be correct in stating that the recent mass murder wave is not unprecedented, but that it merely has attracted newsworthy attention because of the significantly larger number of public shootings (p. 755).

Mass Murder Patterns

Research literature on mass murder entails the use of five key areas to describe its current pattern, and those are: location, victim characteristics, motive, weapon, and end result.


Research is contradictory when it comes to identifying where exactly mass murders usually take place. Palermo (2007) claims that mass murders frequently occur in a public setting (i.e., restaurant, post office, school) or anywhere people are assembled together, making it a locally limited and non-repetitive crime (pp. 16–17). Similarly, Hempel et al. (1999) found in their study that 70% of mass murders occurred in the workplace, a public street, or school during weekday morning hours (p. 222). On the contrary, Duwe (2004) argued that mass murders seldom take place in public locations; rather, most occur in residential settings, such as in familicides. Fox and Levin (2003) also believe that mass murders do no tend to happen in large cities, but rather, in small towns or rural settings.

Victim Characteristics

It is generally accepted that mass murderers infrequently attack strangers who are in the wrong place at the wrong time. Fox and Levin (2003) found that 40% of the mass murders were committed against family members, and almost as many involved acquaintances (co-workers). Victims were generally young, white, and female, making it a largely interracial offense due to the large proportion of familicides (Duwe 2004; Fox and Levin 2003). Thus, researchers maintain there is a greater representation of females among mass murder victims because familicides usually involve the breadwinning husband killing his wife and children.


The majority of mass murderers have a clear-cut motive, often revenge, and victims are chosen because of what they have done or what they represent (Aitken et al. 2008; Fox and Levin 2003). Fox and Levin (2003) argued that the more specific and focused the element of revenge, the more planned and methodical the murder will be, rather than spontaneous and random. Thus, the more specific the target, the less likely the rage stems from psychosis or some other brain altering condition. Furthermore, motivation to commit mass murder may be broken down into five separate categories of investigation: revenge, power and dominance, loyalty, profit, and terroristic (Fox and Levin 2003). Revenge is the most common type of motive, against a specific individual, particular categories or groups of individuals, or society. Often times the offender is rejected and alienated by peers, which only breeds feelings of inadequacy, fear, rejection, and hostility. Power and dominance is most often seen with the pseudocommando killers, which regularly entails a military obsessed narcissist attempting to gain attention for his cause or just teaching the world a lesson (Dietz 1986). Another form of motivation is loyalty; an unemployed father kills his entire family to protect them from pain and suffering, or out of jealously kills his wife and then his children to protect them from becoming orphans. The last two types of motivation for mass murder are profit (i.e., robbery or burglary) and terroristic endeavors (Fox and Levin 2003). In addition, Fox and Levin (2003) use the aforementioned categories of motivation as the basis for their typology of mass murderers (i.e., in their model, there are five different types of mass murderers: revenge, power and dominance, loyalty, profit, and terror killers) (p. 56–61).


Research essentially agrees that the weapon of choice in mass murders is the firearm (Duwe 2004; Fox and Levin 2003), which includes handguns, rifles, and semi-automatics to name a few. In theory, it is difficult to kill many people with a knife or physical force, and it is rare to have an individual who is proficient in the use of explosive devices. Therefore, the firearm, with its large ammunition capacity, is the weapon of choice in nearly all mass murders.

End Result

Research also tends to agree that in the case of mass murder, the perpetrator often dies at the scene, either by their own hand or by the police (Duwe 2004; Fox and Levin 2003; Holmes and Holmes 1998; Palermo 2007). Additionally, Mullen (2004) concluded that perpetrators of autogenic massacres typically plan the murder-suicide, removing the impulsive outburst of violence and rash suicide factors from the mass murder equation (p. 319).

Factors Contributing to Mass Murder

Palermo (2007) described mass murder as a “culmination of a continuum of experiences, perceptions, beliefs, frustrations, disappointments, hostile fantasies, and perhaps pathology” (p. 18). Similarly, Holmes and Holmes (1998) described the phenomenon as a unique combination of biology, sociology, and personal psychology, which accounts for an individual’s personality, and thus, his or her behavior. More specifically, Fox and Levin (2003) grouped these contributing factors into three types. First, there are predisposes, or long term and stable preconditions that become integrated into the mass murderers’ personality (p. 52). These are essentially outside influences that, over time, prompt the individual to act in a violent manner. These perpetrators have been found to suffer long histories of frustration, humiliation and repeated failures (Aitken et al. 2008). As a result, there is often a diminishing ability to cope, along with an increasing negative self-image, which may develop into an overwhelming sense of depression (Fox and Levin 2003). Frequently, the offender is characterized by his strong externalization of blame, perceiving all those around him as the cause of his problems and failures (p. 53).

Fox and Levin (2003) entitled the next collection of contributing factors as precipitants. That is, short-term and acute triggers, such as certain situations or events, that may precipitate a violent act of aggression. Possible precipitants may include a sudden loss or threat of loss, such as an unwanted separation or termination from employment, financial losses, marital conflict, and/or excessive bullying (Aitken et al. 2008; Fox and Levin 2003). Duwe (2004) also believed that an individual, who was already angry, hopeless, and feeling persecuted, may view a divorce or employment termination as the “final straw” (p. 732).

Lastly, there are facilitating contributing factors, or conditions that may increase the likelihood of violent outbursts, but are not necessary in order to produce such a response (Fox and Levin 2003). For example, mass murderers often socially isolate themselves; having little to no intimate or close relationships, cut off from any source of emotional support, frequently awarding them the title of “loner” (Aitken et al. 2008; Fox and Levin 2003; Hempel et al. 1999). Although, for some offenders these details alone may serve as catastrophic, but in theory, social isolation is not enough to cause an individual to act violently, it merely increases the likelihood of such a sudden occurrence. Thus, biological and psychological factors, such as brain pathology, psychiatric illnesses, a debilitating low self-esteem, and grandiose and obsessional traits (which is often coupled with a sense of entitlement), may all influence, but not cause, the perpetrator to take steps toward a violent incident (Aitken et al. 2008, p. 263).

Similar to Fox and Levin’s (2003) three types of factors contributing to mass murder, Levin and Madfis (2009) presented a five stage model of what they call, cumulative strain, in which they applied to school massacres, specifically. The authors use the term to highlight the critical point in which causal factors intersect and build on one another in order to produce a school massacre (p. 1229). However, none of these factors are viewed by themselves as causing a massacre to occur.

The first stage, chronic strain, is characterized by unremitting frustration and a string of failures in trying to achieve ones individual objectives. This may include negative experiences or disappointing events in social relationships, either at home, school, work, or in the neighborhood in which the offender lives (Levin and Madfis 2009). Subsequently, because he or she is lacking in social constraint, having no attachments to peers or family members, the perpetrator is more likely to react by engaging in delinquent behavior. Thus, stage two, uncontrolled strain, is characterized by the lack of conventional bonds and feelings of marginalization that leave the individual less restricted to conventional behavior (p. 1233). It is at this point that the presence of some catalyst (i.e., loss of job, loss of large amounts of money, divorce, custody battle, loss of academic standing, rejection by a loved one, or a humiliating loss of face) may advance the mass murderer into stage three, acute strain. Although these events are short-lived, they appear to be very troubling, especially for an individual who is frustrated, isolated, and has lost the ability to cope with everyday stressors. Consequently, with nothing left to lose, no supportive environment and a long history of frustration and strain, the perpetrator is inspired to get even and hence enters the planning stage. Concurring with similar research, Levin and Madfis (2009) maintained that there seems to be one singular, acute episode that serves as the last straw for these offenders (p. 1237). For that reason, the mass murder often serves as the final power-asserting moment of a failed existence and as a result, stage five is crossed into: the massacre at school (Levin and Madfis 2009, p. 1239).

Offender Characteristics

Mass murder has been found to follow a comparable pattern, regardless of the victim characteristics or the specific location and circumstances surrounding the incident. Likewise, recent research has identified several offender characteristics commonly found in mass murder cases. For starters, researchers are fairly in agreement when it comes to the statement that most mass murders are of the male gender (Duwe 2004; Fox and Levin 2003; Hempel et al. 1999; Levin and Madfis 2009; Mullen 2004; Palermo 2007). Fox and Levin (2003) contend that males are more likely to suffer such catastrophic losses associated with mass murder, such as separation or divorce and unemployment. Men often define themselves in terms of their occupational role and therefore, are more likely to suffer psychologically from being without a job. In addition, Levin and Madfis (2009) in their recent research made the observation that in western culture, masculinity is frequently equated with violence and as such, mass murders are only carrying out their plan to kill in the socially approved manner for men (p. 1238).

Similarly, another common feature among mass murderers is the “warrior mentality” (i.e., fascination with weapons and war regalia) (Hempel et al. 1999; Mullen 2004). These individuals have been found to possess an arsenal of guns (Palermo 2007) and accordingly, are equipped with a certain level of competence and firearm proficiency that is necessary to carry out a successful murderous rampage (Levin and Madfis 2009). Consequently, these perpetrators are frequently found to have some kind of military training or service in their background. As a result, Hempel et al. (1999) designate military experience as a key feature that normalizes killing as a problem-solving behavior, often desensitizing the person to the act itself (p. 220).

Recent research on mass murder has also identified age as a common characteristic of the offenders. Essentially, perpetrators are older, middle-aged, or in their fourth decade of life (Duwe 2004; Fox and Levin 2003; Hempel et al. 1999; Levin and Madfis 2009; Mullen 2004). Fox and Levin (2003) reason that it takes years to accumulate such a deep sense of frustration, at work or at home, that is distinctive in mass murders. In addition, these perceived failures repeatedly become evident in the stages of life when one should be achieving occupational and personal success, creating even more intense pressure and anxiety (Levin and Madfis 2009).

Concurrently, research has also identified externalization of blame as another common characteristic of mass murderers (Duwe 2004; Fox and Levin 2003). These individuals have been found to hold others accountable, blaming spouses, relatives, co-workers, and society, for their present condition (Duwe 2004), as well as viewing the world and others as persecutory, and malevolent objects (Mullen 2004; Hempel et al. 1999). For example, in workplace massacres, perpetrators have often gone from job to job never achieving the promotions and raises they feel they are entitled and as such, believe their hard work has gone unappreciated (Levin and Madfis 2009). This cycle of unemployment or marginal work is enough to exploit the chronic frustration and disappointment already felt by these individuals.

Moreover, mass murderers are frequently identified as “loners”, or social isolates (Hempel et al. 1999; Levin and Madfis 2009; Mullen 2004). Often single or divorced (Hempel et al. 1999), these offenders seem to lack any external controls (i.e. friends and family) that may have a positive influence on their behavior (Levin and Madfis 2009). As Levin and Madfis (2009) mention in their discussion of school massacres, these individuals must deal on a regular basis with the rejection from their peers, which may include being bullied, teased, and humiliated at school. In many cases, these perpetrators have little attachment to their parents and have friends who get into trouble, if they have friends at all. Thus, every day may be viewed as a constant battle with not only the influences around him, but also the negative thoughts and feelings going on inside him. From this perspective, it may be easy to understand why many mass murderers are identified as having some kind of paranoid or depressive symptoms and/or dysthymia (anxiety and depression accompanied by obsession) (Dietz 1986; Mullen 2004). As well, these individuals are often haunted by thoughts focused on emotional helplessness and cognitive hopelessness (Hempel et al. 1999), which only magnetizes the depressive symptoms.

Finally, research has suggested that those individuals who take part in mass murderers have been frequently found to possess atypical behaviors and/or some of the cluster B personality traits (Hempel et al. 1999; Holmes and Holmes 1998; Levin and Madfis 2009; Mullen 2004; Palermo 2007). These traits include: antisocial personality, narcissism, hostility, oversensitive, rigid with obsessional traits, self-righteous, grandiose traits (i.e., a sense of entitlement), and impulsivity.

Typology of Mass Murderers

Throughout the research literature, mass killers are commonly divided into three major types (Dietz 1986; Hempel et al. 1999; Holmes and Holmes 1998; Palermo 2007). Duwe (2005) mentions, in his study of mass murder in the United States, that there is not much difference between the current typologies because they all use many of the same homicide cases in order to typify mass murder, which are obviously the most heavily publicized ones (p. 67). In addition, the cases used to typify mass murder are more likely to involve high profile massacres, which would include: a larger number of fatal and wounded victims, stranger victims, public locations, assault weapons, workplace violence, and interracial victim-offender relationships (Duwe 2005). Accordingly, Dietz (1986) conveys that there are no adequate typologies and that mass murderers fit unambiguously into one of the following three types: family annihilator, pseudocommando, and set-and-run or hit-and-run killer.

Family annihilators are usually the senior man of the household, depressed, paranoid, and often intoxicated, who murders his entire family out of jealously, revenge, or loyalty (Dietz 1986). These types of offenders are usually intrinsically motivated, often carrying out their fatal plans against family members or affiliates. Pseudocommando killers are those individuals with the “warrior mentality,” often preoccupied with weaponry and firearms, and in most cases, have a deliberate plan for mass murder. These killers are frequently motivated by social components; often lashing out at society, teaching the world a lesson, or drawing attention to themselves and their cause. Charles Whitman, the ex-marine who killed 14 and wounded 30 others at the University of Texas is a good example of a pseudocommando (Fox and Levin 2003), although he also suffered from an amygdaloidal tumor. Before his clearly planned, he hauled to the top of the tower a footlocker containing a rifle, a shotgun, two pistols, a revolver, 700 rounds of ammunition, food, water, a radio, and toiletries (Dietz 1986). Lastly, there is the set-and-run killer, which is the most rarely seen of the three types. This type of mass murderer employs techniques that allow themselves the possibility of escape before the planned deaths occur. This would include those individuals who bomb buildings or vehicles, set arson fires, or tamper with food or products, as in the Tylenol poisoning. In these cases, the offender considers the random killings of bystanders an unimportant cost in comparison to the increased likelihood of escape afforded by these methods (Dietz 1986, p. 483).

Likewise, in their typology of mass murder, Holmes and Holmes (1998) incorporated Dietz’s (1986) family annihilator and pseudocommando typology, but replaced the set-and-run killer with disgruntled employee as the third type. In this form of typology, the disgruntled employee is usually a former employee of the company at which the murders take place, either through dismissal or placement on medical or disability leave. This individual believes he is suffering some personal injustice beyond his control and thus, retaliates at his former place of employment. Time and again this type of perpetrator is found to have psychological problems, including being on some kind of medication and/or undergoing a particular form of counseling (often for paranoia). The disgruntled employee usually sets out with a particular individual or group of persons to kill, but once at the designated location, it is not uncommon for the offender to randomly fire at people (Holmes and Holmes 1998).

Similarly, Harper and Voigt (2007) studied 42 homicide-suicide cases in New Orleans from 1989 to 2001 and found that the data disclosed a typology of homicide-suicide offenders. In their 3-type model, the first is intimate or domestic lethal violence-suicide. This offender style is the most frequently found and the most widely discussed (thus, making it the most prevalent) type of homicide-suicide in the homicide-suicide literature. These incidents tend to be more brutal than other forms of murder, in the sense that there are usually more wounds to the victim, particularly to the head (Harper and Voigt 2007, p. 304). The second type in their model, which is analogous to Dietz’s (1986) family annihilator type, is the family annihilation-suicide. This individual is the head of the household who often can no longer care for their family, due to employment or financial losses. These types of incidences seem to be planned, evolving more gradually in the context of immense personal failures and stress. Frequently, in planning his own suicide, this type of individual concludes that his family cannot survive without him and thus, kills them all in order to save them from the suffering he believes they will endure with him gone. Lastly, Harper and Voigt (2007) established the public killing spree-suicide type, in which the offender, under extreme crisis or sudden loss, goes on a mass shooting binge, usually involving nonfamily members. Although rare, this type of phenomenon is often expressed as a form of suicide by cop (i.e., individual allows police to kill him instead of taking his own life), leaving a large number of dead victims at the scene.

In some of the most publicized mass killings and massacres, there appears to be a history of higher occurrences of these events within the springtime months, most notably in April. For example, some infamous examples of autogenic massacres occurring in April on a global scale include the Columbine and Virginia Tech massacres (April 20, 1999 and April 16, 2007 respectively), the Port Arthur massacre (April 28, 1996), the Erfurt massacre (April 26, 2002), and the Azerbaijan State Oil Academy massacre (April 30, 2009). Other notable events that may not meet the current exercise’s criteria for “autogenic massacre” but nonetheless are considered mass killings include the Oklahoma City bombing (April 19, 1995), the Waco/Branch Davidian Seige (April 19, 1993), and, perhaps more historically, the Ludlow massacre, which occurred on April 20, 1914 (see reports from Alvis-Banks 2007; Europe Intelligence Wire 2002; Lackmeyer and Zizzo 1995; Madigan 2008; Obmascik 1999; Ostrow and MacCormack 1993; Simpson and Bronwen 1996; Smith 2009). However, it cannot be stated that the distribution of these events is more heavily weighted into the spring months, because media reports may reflect sampling difficulties which may not be evident on first observation.


Mass murder in the United States may not be a regular occurrence, but recent research has indicated these events occur more often than the public would like to acknowledge. While the incidence of mass murders is of only limited interest to police and law enforcement services, because of the very high clearance rate, it is very important for behavioral sciences and mental health services to understand the phenomenon of mass killings. Family slayings and angry ex-employees taking revenge on their former co-workers are highly motivated, highly publicized episodes are also not unique to the Western culture. There are important gaps in our understanding of mass murder, including the nature of significant provocation, personal characteristics of the individual, and even the seasonal distribution of the events. Researchers urgently need to continue their pursuit in distinguishing the stressors, traits and antecedents to these events in order to prevent such episodes from happening in the future.

Copyright information

© Springer Science + Business Media, LLC 2009