Male Perpetrators of Intimate Partner Homicide: A Review and Proposed Typology

Demographics

Men who kill their intimate partners tend to be in their mid- to late-30s and, on average, older than those who perpetrate nonfatal IPV, those who kill other family members, and those who kill nonfamily members.^{6–7,16,–,18} In a pretrial forensic sample of 213 male homicide perpetrators, Daniel and Holcomb¹⁷ found that those charged with domestic homicide were almost a decade older than those charged with nonfamilial homicide (34.05 versus 26.89). Oram and colleagues,⁶ in a large population-based study, compared those who perpetrated IPH to those who killed another adult family member and found that men who killed their partners were significantly older than those who killed other family members (39.6 versus 32.2). Campbell and colleagues⁷ found that men who perpetrated fatal IPV were older than men who perpetrated nonfatal spousal abuse (34.2 versus 21.2), and Rosenbaum¹⁹ found that IPH perpetrators who committed suicide were older than those who did not (33 versus 42).

Early Experiences

Approximately one-fourth of male IPH perpetrators were abused as children. In a sample of incarcerated IPH perpetrators, Stout²⁰ found that 17 and 9 percent endorsed experiences of childhood physical or sexual abuse, respectively. These childhood victimization rates are similar to those identified in an American sample of pretrial men charged with IPH (23%)¹⁷ and a large sample of Dutch uxoricide suspects (25%).²¹ Stout reported that a substantial minority (39%) of his sample witnessed parental domestic violence, generally perpetrated by the father against the mother. Nearly all of the sample of IPH perpetrators studied by Daniel and Holcomb¹⁷ had significant behavioral problems beginning in childhood, whereas nondomestic homicide perpetrators, by comparison, evidenced less severe childhood behavior problems.¹⁷ Nearly two-thirds (65%) of the small and relatively well-adjusted sample reported by Stout²⁰ grew up in an intact family. Daniel and Holcomb¹⁷ found that 50 percent of those in their larger pretrial sample of men charged with IPH grew up without a father in the home and 12 percent without a mother.

Approximately half of all IPH perpetrators do not graduate from high school. Campbell et al.⁷ found that 49 percent of their sample of IPH perpetrators did not graduate from high school, 33 percent earned their high school diploma, 12 percent completed some college or trade school, and 6 percent completed college or trade school. Koziol-McLain and colleagues¹⁰ found that 55 percent of femicide perpetrators who did not commit suicide did not graduate from high school and that IPH-suicide perpetrators completed slightly higher levels of education. Whereas 65 percent of the sample of male IPH defendants studied by Barnard and colleagues²² did not complete high school, only 39 percent of the sample reported by Goetting²³ did not.

Employment Status

Unemployment rates among IPH perpetrators vary widely across samples, with estimates ranging from 13 to 58 percent.^20,21 Two large-scale studies provided what may be the most reliable estimates of employment rates among male IPH perpetrators and suggested that IPH perpetrators are more likely to be employed than men who kill other family members, but are less likely to be employed than nonfatal IPV perpetrators. Oram and colleagues⁶ found that 39 percent of IPH perpetrators were unemployed, compared with an unemployment rate of 55 percent among men who killed other adult family members. Campbell and colleagues,⁷ in a comparison between fatal and nonfatal IPV perpetrators, found that nearly half (49%) of all male IPH perpetrators were unemployed, significantly more than nonfatal abusers (20%). Similarly, 45 percent of all IPH perpetrators in Massachusetts from 2005 through 2007 were unemployed at the time of the homicide.¹⁸ At the extremes, Stout²⁰ reported that only 13 percent of a sample of 23 men convicted of IPH in Missouri were unemployed at the time of the offense, whereas fully 58 percent of the sample of 380 Dutch uxoricide suspects studied by Liem and Koenraadt²¹ were unemployed.

Clinical Features

Research examining psychopathology and personality disturbance in IPH perpetrators tends to place them in the middle of the psychopathological continuum between killers of other family members, who tend to show more pathology, and general homicide perpetrators, who tend to show less.^6–7,21 Understanding the prevalence and nature of the psychopathology in IPH perpetrators requires both stable population-based prevalence rates and in-depth psychiatric data collection to account for methodological biases.^21,24,25 The role of completed suicide in the context of IPH also limits data derived from certain research designs, resulting in the omission of a distinct but not uncommon type of IPH perpetrator: the IPH-suicide perpetrator. Consistent with the observations of Liem and Koenraadt,²¹ it is reasonable to conceptualize data generated by IPH perpetrators posthomicide as reflecting the specific population of perpetrators who survive the event.

History of Domestic Violence

Two distinct conceptualizations of IPH permeate the literature in both explicit and implicit ways. One perspective views IPH as the extreme end of a continuum of domestic violence, generally with the implication that the homicide is the end point of a period of escalating partner violence.³⁹ Efforts to apply nonfatal IPV typologies to IPH perpetrators reflect this perspective.⁴⁰ An alternative view suggests that IPH may best be conceptualized as a qualitatively distinct behavior. Gelles has been a proponent of this perspective, writing: “… homicide is not simply an ‘extreme form of interpersonal violence’ or a form of behavior that, because of its seriousness, provides a valid ‘assay’ of interpersonal conflict…. Rather, homicide is a distinct form of behavior that requires a distinct explanation” (Ref. 41, p 69).

Research examining the role of domestic violence in IPH suggests that neither perspective, independently, is tenable across all cases. Rather, the wide range in prevalence rates of previous partner violence across samples of IPH perpetrators highlights the necessity of a case-by-case analysis to determine its relative fit within each of these models. Estimates of previous domestic violence by male IPH perpetrators vary widely, from approximately 22 to 77 percent (Table 2).^22,43 There are several probable reasons for this range. First, it is likely that IPH perpetrators with a history of domestic violence would be over-represented in certain settings. For example, Dutton and Kerry³⁰ reported that two-thirds of the sample of generally personality-disordered incarcerated men convicted of IPH in their study either admitted to prior domestic violence or there was evidence of such incidents in their institutional record. In contrast, Belfrage and Rying²⁶ found that only 36 percent of the forensic psychiatric patients in their Swedish sample had evidence of prior domestic violence. Even within correctional settings, pretrial defendants may be more likely to minimize previous domestic violence than those already sentenced. This may partially account for the relatively low rates of self-reported domestic violence in the pretrial sample in Barnard et al.²² as opposed, for example, to the convicted sample in Dutton and Kerry.³⁰ The operationalization of domestic violence is also a likely influence of the observed rates of previous domestic violence. For example, the relatively stringent criteria established by Campbell et al.,⁷ who asked proxy informants specifically about prior arrests for domestic violence, showed that only 26 percent of IPH perpetrators met the threshold. In contrast, if proxy informants are simply asked about previous domestic violence, not necessarily resulting in arrests, the number rises to 67 percent.⁵

Despite identifying relatively high rates of domestic violence in their sample of incarcerated male IPH perpetrators, Dutton and Kerry³⁰ highlighted an essential implication of this line of research, in concluding that there appeared to be a distinct type of IPH perpetrator, differentiated in part by the absence of partner violence before the homicide. For these individuals, the homicide appeared to be out of character and is anything but an extension and escalation of an ongoing pattern of partner abuse. Instead, it is an aberration with distinctive psychodynamic features, what Dutton and Kerry³⁰ describe as a “catathymic crisis” characterized by the three stages of incubation, violent outburst, and relief.^44,45 That multiple samples have consistently identified a subset of IPH perpetrators whose homicide was not preceded by prior domestic violence supports the understanding that this represents a distinct group rather than a methodological artifact.

Separation and Abandonment

Researchers over the past 35 years have consistently found a robust association between separation and IPH.^{20,22,30,46,–,51} Barnard and colleagues,²² for example, found that 57 percent of male IPH defendants in their small sample had been separated from their partners on the day of the homicide. Further, research suggests that roughly 20 to 30 percent of IPHs are precipitated by separation. Crawford and Gartner⁴⁶ concluded that 32 percent of the 896 male-perpetrated IPHs in their Ontario sample were estrangement killings; 26 percent of the Dutch forensic sample of IPH suspects studied by Liem and Koenraadt²¹ were deemed to be motivated by fears of abandonment; 20 percent of all partner homicides in Massachusetts from 2005 through 2007 were determined to have been motivated by the termination of the relationship.¹⁸

Notably, abandonment as a robust risk factor for killing one's partner appears to be relatively unique to male-perpetrated IPH, with consistent gender differences emerging in samples from the United States, Australia, and Canada.^30,51,52 Regarding the temporal association between separation and intimate homicide, approximately 50 percent of male-perpetrated IPHs that occur in the context of estrangement occur within two months of separation, and approximately 90 percent occur within the first year after separation.^20,50,51,53

Dutton⁴⁷ implicates early attachment traumas in the etiology of abandonment rage in affectively motivated male IPH perpetrators. Drawing from the literature on the neurobiological effects of early trauma—including compromised development of the orbitofrontal cortex, low levels of serotonin, and high levels of norepinephrine—he observed that these consequences of early trauma have also been found to be risk factors for violent behavior. He concluded that early attachment traumas leave an enduring neurobiological mark that establishes “neural networks containing malignant memories [that] may be the neural mechanism by which perceived abandonment generates such symbolic terror and rage” (Ref. 47, p 407).

Stalking

Various typologies of stalkers have been proposed, most of which include a category of those with a previous intimate relationship with the victim.^54,–,56 Research on the relative risk of these subtypes suggests that ex-intimate stalkers present the highest risk of engaging in fatal and nonfatal violence.^57,–,61 Although the association between stalking and nonfatal IPV has received considerable attention,^62,–,66 the role of stalking in fatal partner violence has been relatively under-researched. Preliminary evidence, however, reveals high rates of stalking preceding the killing of an intimate partner. In a large sample of female IPH victims in North Carolina, Moracco and colleagues⁴³ found that 23 percent had been stalked before the homicide. McFarlane et al.,⁵ in a 10-city multisite study of female victims of IPH (n = 141) and attempted IPH (n = 65), found even higher rates; more than three-fourths of the women who were killed or whose partner had attempted to kill them had been stalked by their partners in the past year. However, these authors' definition of stalking did not require multiple instances of stalking behavior, possibly overestimating the true prevalence. The 11-city multisite study by Campbell and colleagues,⁷ comparing 220 female IPH victims to 343 female victims of nonfatal IPV, found that stalking emerged as a significant predictor of fatal abuse. McFarlane et al.⁶⁷ found that women who had been “followed or spied on” in the past 12 months were more than twice as likely to be victims of actual or attempted IPH than were women without these stalking experiences.

History of General Violence

Approximately one-fourth to one-half of all IPH perpetrators have been arrested for a previous violent crime. Campbell and colleagues⁷ found that 22 percent of IPH perpetrators had a previous arrest for a violent crime, a rate twice that in perpetrators of nonfatal partner violence. More than one in four (27%) uxoricide suspects in the Dutch forensic psychiatric sample in Liem and Koenraadt²¹ had been convicted of a violent offense. In a multisite study of IPH perpetrators, Koziol-McLain and colleagues¹⁰ found that, among those who did not commit suicide, 23 percent had been arrested for a violent crime. This figure decreased slightly (17%) among IPH-suicide perpetrators. Forty-one percent of the pretrial sample of accused IPH perpetrators reported by Daniel and Holcomb¹⁷ had been charged with a crime against a person, and fully 45 percent of all IPH perpetrators in Massachusetts from 2005 through 2007 had been charged with a violent crime.¹⁸

Associated Deaths

There is at least one additional associated death in nearly 40 percent of male-perpetrated IPHs. Bourget and Gagné¹⁶ found that 61 percent of such incidents in Quebec resulted in a single death, 32 percent included one additional death (usually suicide; less frequently, the homicide of a biological child; but in one instance, a stepchild), 4 percent resulted in three associated deaths, and 3 percent in four associated deaths; in one case, there were 7 associated deaths. Ten percent of a sample of incarcerated male IPH perpetrators studied by Dutton and Kerry³⁰ had killed at least one other person in addition to their partners. In contrast to the data in Bourget and Gagné,¹⁶ suggesting that biological children are the most common third-party victims, Dutton and Kerry found that the female victim's current partner was the most common associated death.³⁰ Liem and Koenraadt,²¹ in their Dutch sample comparing perpetrators of uxoricide, filicide, and familicide distinguished those who engaged in IPH only versus those who killed multiple family members and found that those who killed multiple family members were more likely than single-victim IPH perpetrators to be married, to have a personality disorder, and to attempt suicide after the killings. In contrast, single-victim IPH perpetrators were more likely to have committed a previous violent offense.

Jealousy and Envy

Jealousy is a frequently cited motive for IPH.^3,7,10,23,26 Belfrage and Rying,²⁶ for example, reported that 20 percent of police files specified jealousy as a motive. Campbell and colleagues⁷ found jealousy to be an even more common motive in their population-based study, concluding that 39 percent of IPH perpetrators were triggered by jealousy. These studies, like many in the IPH literature, are limited by a lack of definitional clarity and leave open the possibility that both jealousy and related affects are common in IPH.

Despite the frequency with which jealousy appears in the literature, the related construct of envy has received minimal attention. It is likely that both jealousy and envy are present in many cases of IPH, probably to varying degrees, and that these dynamically distinct motivations are frequently confounded. A sound conceptual model would allow for a better understanding of these mental states as they relate to acts of partner homicide. In a description by Richards⁶⁸ of the “anatomy of envy,” for example, he distinguishes envy and jealousy along the three dimensions of possession, possibility, and rivalry. Whereas jealousy is based on a fear of losing the good object to a real or imagined other who threatens to take it away, envy is distinct as a feeling toward others who already possess the good object. Further, jealousy entails the possibility of retaining the good object that the individual currently possesses, whereas envy reflects recognition of the impossibility of obtaining it. The conceptualization of envy set forth by Klein⁶⁹ shifts the focus of the envious person from the other who possesses the good object to a destructive impulse directed toward the good object itself.⁷⁰ That is, the impossibility of obtaining the good object characteristic of envy results in the urge to destroy the good object itself. Elaborating on Klein's distinction between envy and greed, Mitchell and Black wrote: The infant's greed is not destructive in its intentions toward the breast, but deeply resentful of receiving its precious bounty only in drips and drops…. Envy is a different response to the same situation. The envious infant no longer wants to gain access to and possess the good, but now becomes intent on spoiling it…. The infant would rather destroy the good than remain helplessly dependent on it. The very existence of goodness arouses intolerable envy, the only escape from which is the fantasied destruction of the goodness itself…. Thus, envy undoes splitting, crosses the divide separating good from bad, and contaminates the purest sources of love and refuge. Envy destroys hope [Ref. 71, p 100].

Efforts to better understand the motivations of male IPH perpetrators would benefit from conceptual clarification, such as that provided by the differentiation drawn by Richards⁶⁸ between envy and jealousy along the dimensions of possession, possibility, and rivalry, along with the emphasis by Klein⁶⁹ on envy as distinct in its destruction of the good, as opposed to a bad or devalued object. The common sentiment of male IPH perpetrators that “if I can't have you, no one can,”⁴² reflects the absence of possession and possibility along with the destructive desire to spoil the good object pathognomonic of envy. Further, the recognition by Stout²⁰ that “the perpetrators of intimate femicide need to seek help to more fully understand the lethal violence they used against a partner they almost unanimously claimed to still love” (Ref. 20, p 93) captures the destruction of the good object unique to envy.