Intimate Partner Femicide–Suicides in Ghana

Abstract

The present study investigated the scope, nature, and determinants of intimate partner femicide–suicides (IPFS) that occurred in Ghana during 1990 to 2009. All 35 reported cases of intimate partner homicide–suicides with female homicide victims that occurred during the study period were extracted from a major Ghanaian daily newspaper. Findings indicate that offenders were of lower socioeconomic background and tended to be older than their victims. The results further show that shooting with a firearm and hacking with a machete were the primary homicide methods, whereas self-inflicted gunshots and hanging were the dominant suicide methods. Results showed that suspicion of infidelity and sexual jealousy were core contributing factors in arguments, disputes, and altercations that preceded the femicide–suicides. Furthermore, estrangement and threatened divorce or separation by the female intimate partner was a major precipitant of femicide–suicides.

Keywords

femicide Ghana homicide–suicide intimate partner violence suicide

Introduction

Homicide–suicide refers to a category of lethal violence in which an assailant who perpetrates one or more homicides subsequently commits suicide, either immediately or shortly thereafter. According to the homicide–suicide literature, intimate partner homicide–suicide involving a female homicide victim and the suicide of a male assailant is the most common victim–perpetrator relational category (Banks, Crandall, Sklar, & Bauer, 2008; Koziol-McLain et al., 2006; Morton, Runyan, Moracco, & Butts, 1998; Warren-Gordon, Byers, Brodt, Wartak, & Biskupski, 2010; Websdale, 2010). Despite this widely documented finding, intimate partner homicide–suicides with female homicide victims—hereafter referred to as intimate partner femicide–suicides (IPFS)—are rarely the focus of systematic empirical studies. To date, only a few studies have actually focused on IPFS as a subject, and of those studies, the majority have been conducted in the United States (Banks et al., 2008; Koziol-McLain et al., 2006; Krulewitch, 2009; Lund & Smorodinsky, 2001; Morton et al., 1998), Canada (Dawson, 2005), and other Western societies (Easteal, 1994). Very few studies have examined the incidence and patterns of IPFS in non-Western societies (Mathews et al., 2008), although a survey of the scholarly literature and popular press reveals that the phenomenon is not uncommon in many of these societies (Adinkrah, 2003, 2008a, 2008b; Jankey, 2009; Rude, 1999). Consequently, the literature currently reveals very little about the phenomenon in non-Western societies.

This paucity of research attention to IPFS in non-Western societies is lamentable given that it leaves IPFS uninterrogated in Africa, Asia, Latin America, the Middle East, and Oceania where the bulk of the world’s population currently lives. Knowledge of the prevalence, patterns, and etiology of IPFS can be significantly improved through the incorporation of this population into studies of this form of lethal aggression. To help address this imbalance in the literature and enhance current understanding of IPFS, the present study provides a descriptive analysis of all reported 35 IPFS incidents that occurred in Ghana, West Africa, during 1990 to 2009. Specifically, the study examined the sociodemographic characteristics of victims and offenders, victim–offender relationships, spatial aspects, modus operandi, and types of weapons used in the perpetration of these crimes, the situational contexts and triggering events, as well as police-assigned motives for the lethal events. The impetus for the present study includes the supposition that understanding IPFS will more positively impact stakeholders’ ability to design programs to prevent and, therefore, reduce this form of familial and societal tragedy.

Past Research

Extant studies demonstrate that homicide–suicides occur primarily in intimate partner relationships and are overwhelmingly perpetrated by husbands, boyfriends, or former lovers against female partners (Adler, 1999; Banks et al., 2008; Dawson, 2005; M. Liem, 2010; Lund & Smorodinsky, 2001; Morton et al., 1998). According to existing research, women rarely take their own lives after murdering their spouses or lovers (Adler, 1999; Dawson, 2005; Lund & Smorodinsky, 2001). Regarding the scope and prevalence of IPFS, Mathews and associates (2008) reported that 18% to 40% of femicides that occur internationally are followed by perpetrator suicide.

Existing studies suggest that IPFS are typically premeditated events that involve considerable and methodical planning (Adler, 1999; Banks et al., 2008; Dawson, 2005; Koziol-McLain et al., 2006; Mathews et al., 2008). For example, using Canadian data, Dawson (2005) compared intimate femicide–suicides with intimate femicides not followed by perpetrator suicide and found the former category to involve greater degrees of premeditation. Also, in an early era study of homicide–suicides in Chicago from 1875 to 1910, Adler (1999) observed that men who killed their lovers and subsequently themselves frequently contemplated their actions, threatened their victims, and expended the time needed to procure weapons and pen suicide notes; several of them also announced their intentions and plans “far and wide” and long before they executed their deadly deeds (p. 11).

Another consistent and widely reported finding within IPFS literature is that the principal method of homicide perpetration is shooting with a firearm (Adler, 1999; Banks et al., 2008; Koziol-McLain et al., 2006; Lund & Smorodinsky, 2001; Morton et al., 1998; Sillito & Salari, 2011), while the suicidal acts following the homicide are also typically perpetrated with a firearm. From their California (United States) study, Lund and Smorodinsky (2001) observed that intimate partner homicide perpetrators who committed homicide via a firearm were 13 times more likely to subsequently commit suicide than intimate partner homicide perpetrators who employed other types of weaponry. In a recent study on intimate partner femicides conducted in New Mexico (United States), Banks and associates (2008) observed a strong association between the employment of firearms to commit a femicide and the subsequent suicide of the perpetrator, reporting that 89.1% of the IPFS incidents they studied involved firearm use.

Studies examining IPFS often explore the precursors or precipitating factors that contributed to the event. A considerable number of studies report that a major precipitant of IPFS is female partner-initiated divorce, estrangement, or the threat of separation. Men kill partners who leave or threaten to leave them (Adler, 1999; Brookman, 2005; Dawson, 2005). Several studies also report that male proprietariness—a concept which refers to extreme possessiveness relating to power and control over the female partner—triggered some IPFS. Another motive often found in the literature concerns issues of sexual jealousy and allegations of sexual infidelity against the female partner. Several studies report that the male assailants’ perception that their intimate partners had been sexually unfaithful to them ignited a large proportion of IPFS (Dawson, 2005; West, 1965).

The literature distinguishes between two major forms of homicide–suicide based on whether the case exhibits more homicide components or suicide components. In one form of homicide–suicide, perpetrators of homicide subsequently take their own lives out of fear of the potential penal and social sanctions of their criminal actions, or are overcome by remorse for having just committed the homicide (Dawson, 2005; Selkin, 1976). In the other type, homicide–suicide perpetrators originally intended to commit suicide but the homicide victim was killed because the suicidal person did not wish to exit this world without the homicide victim. In these cases, homicide–suicide may be considered “extended suicide” (Dawson, 2005; Palermo, 1994).

Some past studies (e.g., Sillito & Salari, 2011; Steeves & Parker, 2007; Steeves, Parker, Laughon, Knopp, & Thompson, 2011) have addressed the impact of intimate partner femicides and femicide–suicides on children. Studies suggest that children of the victims or perpetrators involved in the dyadic deaths who witness these lethal tragedies are at risk for tremendous physical, social, and psychological sequelae (Sillito & Salari, 2011; Steeves & Parker, 2007; Steeves et al., 2011). Indeed, some of the children caught in the crossfire of lethal parental encounters themselves become fatal casualties, either as deliberate or accidental targets (Sillito & Salari, 2011).

In the only publication on IPFS in Africa in the literature, Mathews et al. (2008) reported that 19.4% of perpetrators of intimate partner femicides in South Africa in 1999 subsequently committed suicide within a week of the murder. The authors also found higher rates of femicide–suicides among South African Whites and Indians, white-collar and professional workers, and legal owners of firearms. Furthermore, the victims of IPFS were younger than their assailants. In addition, IPFS were more likely to be perpetrated with a gun than by any other method. Comparing IPFS with intimate partner femicide nonsuicides, the authors observed that victims in intimate partner femicide–suicides in South Africa were more likely to have ended or initiated the termination of the relationship (Mathews et al., 2008).

In sum, these investigations have offered valuable insights about femicide–suicides in Western industrialized societies. Missing is comparable research that focuses on IPFS in non-Western, nonindustrialized societies.

Ghana: The Research Setting

Situated on the Western coastal region of Africa, Ghana is a country slightly smaller than the U.S. state of Oregon. With an estimated population of 24 million in 2010, Ghana is a multiethnic, multireligious, and multilingual society. The Akans are the most numerically dominant ethnic group, constituting 45.3% of the total population. Other major ethnic groups in the society include Mole Dagbon (15.2%), Ewe (11.7%), Ga-Dangme (7.3%), Guan (4%), Gurma (3.6%), and Grusi (2.6%). In conjunction with the ethnic heterogeneity of the population is its multilingualism, although English is the official language. The country is also multireligious with Christians comprising 68.8%, Muslims 15.9%, and adherents of traditional religion constituting 8.5% of the population, respectively (Ghana, 2011; “World Factbook,” 2008).

In 2010, life expectancy at birth for the Ghanaian population was estimated to be 59.4 years for males and 61.8 years for females. According to the 2000 country census, roughly 57.9% of the population aged 15 years and above was literate. The literacy rate for males was 66.5%, whereas that for females was 49.8%. In 2008, about one half of the population lived in rural communities, generally characterized by the absence of or limited access to potable water, electricity, medical facilities, tertiary institutions, and other social and infrastructural resources. Per capita Gross Domestic Income was US$1,600 in 2010. The unemployment rate in 2000 was about 11%, whereas in 2007 about 28.5% of the population was estimated to be living below the national poverty line (Ghana, 2011; “World Factbook,” 2008).

According to the Ghana Criminal Code (1960, Act 29), nonfatal suicidal behavior (attempted suicide) is a crime. Section 57 of the code reads “whoever attempts to commit suicide shall be guilty of a misdemeanor.” Thus, persons who engage in abortive suicidal acts are liable to arrest and criminal prosecution, and upon conviction are subject to severe penalties, including custodial sanctions (Adinkrah, 2011b, 2011c, 2012, 2013). Indeed, the strong legal response to suicide is compounded by an equally strong negative social reaction. Suicide is regarded by all ethnic groups in the country as a repugnant and immoral act (Greene, 2002; Kuada & Chachah, 1999; Nukunya, 2004). Perhaps due to the criminal status of nonfatal suicidal behavior and an intense social and cultural prohibition against suicide, rates of fatal and nonfatal suicidal behavior are exceedingly low, with male rates substantially higher than that of females (Adinkrah, 2011b, 2011c, 2012). Regarding rates, there were 0.24 suicides per 100,000 resident population in 2006, 0.40 per 100,000 population in 2007, and 0.41 per 100,000 population in 2008 (Adinkrah, 2011b). Official police data covering the past decade suggest rising levels of violent crime in the country (Appiahene-Gyamfi, 2009). For homicide, police-collated data show that the number of murder cases increased from 275 in 1990 to 430 in 2008 (Ghana Police Service, 2010).

There are stringent official restrictions on the number and types of firearms Ghanaian citizens can possess. Accurate figures on firearm ownership and availability in Ghana are difficult to obtain because official data on the issue are considered sensitive and classified for reasons of national security. Unofficial estimates suggest that firearm ownership and possession rates are among the lowest in the world. One independent report estimated that there were 80,000 guns in civilian hands in 2007, a rate of 0.4 firearms per 100 people. The same report ranked Ghana 174th out of 179 countries in the rate of private gun ownership. However, illegal gun possession is believed to equal, or even exceed, official firearm registrations. In 2011, one report estimated that about 130,000 unregistered guns were unlawfully held by Ghanaian civilians, a rate of 0.6 illicit firearms per 100 people (Alpers & Wilson, 2012).

The Status of Women in Ghana

Although a few women currently occupy high-level political and administrative positions, signifying advances in socioeconomic status gained in recent years, Ghanaian women are generally subordinate to their male counterparts in nearly every social domain in the society. Presently, females lag behind males in literacy rates, completed years of schooling, enrollment in tertiary institutions, and in elections and appointments to political office. Other signs of gender inequity are evident in the workforce, where women’s access to highly remunerative jobs is limited and where executive and managerial positions are overwhelmingly occupied by males.

Although attitudes and practices are changing due to influences of formal education and globalization, the typical Ghanaian family is a highly patriarchal institution. Ghanaian males tend to subscribe to a premarital and postmarital sexual double standard. For example, despite a decline in practice over the decades, polygyny, or marriage of a man to two or more wives simultaneously, remains a legal and culturally permitted practice among all ethnic groups in the country; for many men, having multiple wives or girlfriends is denotative of male power, prestige, and virility. Although romantic or love marriages in which prospective brides and grooms choose their own partners are the dominant marital arrangement, arranged marriages are normative in some regions of the country. Here, the selection of the spouse and other marital arrangements are made by the parents, family, or other relatives, with the bride and groom having a minimal or no decision-making role. In some ethnic communities, child-betrothal and girl-child marriages, in which prepubescent girls are married to significantly older men, are known to occur (Selby, 2008). Gerontogamy, or marriage between younger women and substantially older men, is commonplace. Although divorce is socially permitted, it is generally frowned upon. Divorce and separation are considered humiliating for men when initiated by women. Remarriage following divorce is more common and less stigmatized for men than women. A woman is expected to stay in an unhappy, unfulfilling marriage for the children’s sake, rather than risk the social disapprobation that accompanies divorcee status.

In the domestic realm, there is a sexual division of labor in which males are considered the primary providers, even if the wife earns a higher income. Traditionally, women are responsible for maintaining the home, including cooking and cleaning, and are the primary caregivers of children. In marriage and dating relationships, women are expected to be compliant and respectful toward male partners. Husbands expect their wives to obtain their permission before traveling to funerals, weddings, church conventions, and other social gatherings. Intimate partner violence against women occurs frequently, as many men consider it their marital right to discipline and control their wives (Amoakohene, 2004; Takyi & Mann, 2006). Spousal rape has yet to be criminalized in the society; many men and women hold the view that providing sex is a wife’s marital duty, even if it means submitting to sexual coercion (Adinkrah, 2011a). Although there are laws that sanction domestic violence, many cases go unreported to law enforcement authorities (Adinkrah, 2008a, 2008b; Amoakohene, 2004). Incidents of lethal and nonlethal physical violence against women, including sexual violence and domestic abuse (e.g., rape, defilement, forced sex in marriage, uxoricide) are widespread (Adinkrah, 2008a, 2008b; Amoakohene, 2004). According to one estimate, one out of every three women is a victim of male-perpetrated nonfatal assaultive violence. Emergency shelters for victims of domestic abuse are lacking in the country. This makes women in violent dating or marital relationships particularly vulnerable to repeated victimization.

Official police reports over the past decade also suggest growing rates of violent crime against women and girls in the society (Appiahene-Gyamfi, 2009). Media reports of sexual abuse of girls and women are legion, as are official police reports including rape, defilement, and intimate partner violence. Other pernicious forms of gender violence against females in the society include female genital mutilation, oppressive widowhood rites, leviratic marriages, childhood betrothal, child marriages, forced marriages, and trokosi in which young virgin girls are placed in servitude to fetish shrines to atone for the deviant deeds of male relatives (Afrifa, 1994; Duodu, 2008). Sexual harassment of females in schools and in the workplace is rife. Abortion is illegal and carries a long custodial sentence upon conviction. The scapegoating of elderly women who are repeated targets of witchcraft accusations is another expression of gender violence in Ghanaian society; many such women are banished to squalid witch sanctuaries (Adinkrah, 2004; Berg, 2005).

Method

The data analyzed in the present study were collected in Ghana during May to August 2011. Given the absence of comprehensive and reliable homicide or suicide data in the country, the current study relied on a content analysis of femicide–suicide media reports in the most reputable Ghanaian daily newspaper, the Daily Graphic, as the primary source of information for femicide–suicide. Founded in 1950, the Daily Graphic is one of only two national dailies in the country and has the largest circulation of any newspaper in Ghana. An independent readership survey conducted in 2001 revealed that the Daily Graphic was the most widely read newspaper in the country (“Graphic Still No. 1,” 2001).¹ The paper’s high reputation is attributed to its accuracy of reporting and depth of news coverage. Trained investigative journalists from the newspaper are often at crime scenes and are usually in attendance at criminal trials, which gives them access to reliable crime data in the reportage of crime stories.

For this research, in the absence of an electronic database covering the entire Daily Graphic newspaper, a careful manual search was conducted of every issue of the newspaper, from January 1, 1990, to December 31, 2009. The purpose was to identify all IPFS cases that were reported in the newspaper. “Intimate partner” was defined for this study as two people in a connubial or romantic relationship. These include current and former spouses and dating partners.² A case was considered a femicide–suicide incident as long as the femicide perpetrator subsequently committed suicide, regardless of the amount of time that elapsed between the femicide and the suicide.³

In all, the study identified 50 homicide–suicide incidents during the study period. Of these, 36 (72%) were intimate partner homicide–suicide events. Thirty-five (97.2%) out of the 36 intimate partner homicide–suicides were cases with female homicide victims, which is the focus of the present study. Thus, only one (2.8%) case involved an intimate partner homicide–suicide with a male victim. All information pertaining to each case was photocopied. All case data were carefully read and the relevant information extracted and recorded into a database. The data were then sorted into victim and offender attributes (e.g., sex, age, occupation), place and time of the femicide and suicide, modus operandi, and the assigned motives for the femicide and suicide.

Supplementary information was obtained from unstructured interviews conducted with three Ghanaian professionals—a forensic pathologist, a psychiatrist, and a senior law enforcement officer, all working in the country’s health care and criminal justice systems. The interviews served to clarify issues concerning official policies regarding determination of the cause of death, police investigations into homicide and suicide, pathologists’ autopsies, coroners’ inquests, and psychiatric evaluations of criminal suspects with perceived psychiatric disorders. The research protocols were approved by the Institutional Review Board (IRB) of the author’s university.

Information gathered from these data sources was conjoined to construct case histories for each of the 35 partner femicide–suicide events. These were systematically analyzed to identify the sociodemographic characteristics of victims and assailants, spatial patterns, precipitating circumstances, and methods of homicide and suicide commission.

The use of newspaper surveillance methodology to study homicide–suicide is not a novel practice. Indeed, it has been an accepted methodology given the scarcity or unavailability of national data sets on the phenomenon in several countries or jurisdictions around the world (Warren-Gordon et al., 2010). Numerous researchers from several countries including Australia (Pirkis, Burgess, Francis, Blood, & Jolley, 2006), Finland (Nikunen, 2006), the Netherlands (M. C. A. Liem & Koenraadt, 2007), and the United States (Malphurs & Cohen, 2002; Warren-Gordon et al., 2010) have utilized data obtained from newspaper surveillance to examine homicide followed by perpetrator suicide, particularly in instances where reliable homicide–suicide data sets were unavailable. The use of newspaper surveillance for acquiring homicide–suicide and other crime data is particularly critical in non-Western, developing countries such as Ghana where police records on crime are incomplete, unreliable, or plainly nonexistent. Moreover, in countries such as Ghana, tensions in police–civilian relations have created conditions whereby eyewitnesses to criminal events are often more willing to share information with crime reporters than law enforcement personnel. This author’s own experience with working with both police-compiled data and newspaper data on crime in Ghana shows that newspaper information on crime is often a richer, more detailed source of crime data than police records.

There are several additional factors that make newspapers a particularly relevant data source on homicide–suicides in Ghana. First, unlike other categories of homicide, court records are nonexistent as a data source in studies of homicide–suicide because both the assailant and victim are deceased (Nielssen, Large, Westmore, & Lackersteen, 2009). In Ghana, the processing of homicide–suicide cases is limited to police investigations, autopsies by pathologists, and crime scene analyses by other criminalists. Second, homicide is relatively rare in Ghana. Even more rare is homicide coupled with the suicide of the assailant, as is the case with femicide–suicides. In Ghana, nonfatal suicidal behavior and homicide are both criminal offenses and homicide, suicide, and homicide–suicide are rare phenomena (Adinkrah, 2011b, 2012, 2013). Given the exceptional rarity of such incidents and the vortex of violence associated with them, homicide–suicides are regarded as particularly newsworthy, with major media outlets in the country devoting extensive, front-page coverage to the crime.

Given the high level of public interest in the homicide–suicide, myriad media agencies go to tremendous lengths to report all the details of the crime. Masses of people typically congregate on or around crime scenes. Reporters conduct extensive investigations, often interviewing relatives, family members, friends, acquaintances, neighbors, landlords, employers, and other associates of the decedents as well as law enforcement officers and such medical and public health practitioners as pathologists, psychiatrists, psychologists, and other practitioners. These efforts almost invariably culminate in the publication of significant details of the crime. News reports of homicide–suicides also generate feature-length articles in local newspapers and Internet websites (e.g., Ghanaweb.com, Ghanamma.com).

Results

Extent of Intimate Femicide–Suicide

The results showed that there were 35 cases in which men who killed their intimate female partners subsequently committed suicide. The data show that 32 (91.4%) out of the 35 assailants committed suicide within 24 hr of perpetrating femicide. For the remaining 3 cases, the time interval between the femicide and suicide was 48 to 72 hr in 2 (5.7%) cases and 60 days later 1 (2.9%) case. Because homicide and suicide are highly publicized incidents in Ghana, it is fair to conjecture that this number closely reflects the actual number of such incidents. In Ghana, homicide–suicide cases do not escape the attention of the mass media. Falling into the category of high amplitude cases, each of the 35 cases profiled in this study received front-page coverage in the Daily Graphic. In addition, several of the cases were given coverage at one or more of the numerous Ghana Internet websites.

Victim–Offender Relationship

This study explored the nature of the relationship between victims and offenders in homicide–suicide events. In 29 (82.9%) out of the 35 cases, husbands killed their wives. In the remaining 6 (17.1%) cases, men killed girlfriends.

Sociodemographic Characteristics of Victims and Offenders

This study examined the demographic characteristics of offenders and victims. Femicide victims ranged in age from 18 to 60 years old. The assailants ranged in age from 20 to 70 years old. The analysis of data further revealed substantial age differences between the assailants and victims. In every case examined, the male assailant was older than the femicide victim. The age gap between couples ranged from 1 to 35 years. The mean age gap was 6.7 years. In one case, the uxoricidal husband was 35 years older than his wife-victim.

The results show that the majority of the assailants were of lower socioeconomic background, often employed as peasant farmers (48%), unskilled artisans (20%), low-level security personnel (8%), laborers (6%), hunters (4%), or unemployed (4%) at the time of the crime. Femicide victims were also primarily dependent on husbands or from the lower socioeconomic brackets of society, oftentimes described as peasant farmers (48%), petty traders (16%), homemakers (9%), or unemployed (6%). Among the victims were a schoolteacher, a bank teller, and a university student.

Spatial Aspects of Femicides and Suicides

This study examined the locations or physical settings where the homicides and suicides occurred. The data show that 16 (45.7%) out of the 35 femicides occurred in the shared dwelling of the couple, while 7 (20%) occurred in the victim’s natal home. In 4 (11.4%) other instances involving estrangement, the perpetrator tracked the victim to her farm, ambushed her, and then killed her; two victims of femicide were killed while at the farm working with the husbands who killed them; 2 (5.7%) women who had separated from their husbands and were living in their own new homes were killed at this location. One (2.9%) killing occurred inside a church where the victim was attending a service. A woman who had separated from her husband, but who returned to collect her children’s laundry for washing, was killed at the man’s house.

Twenty-two or 62.9% out of the 35 postfemicidal suicides occurred at the same location as the homicide. The remaining 13 or 37.1% occurred at other locations, including at a public cemetery (1 or 2.9%), at the assailant’s farm (1 or 2.9%), at the assailant’s house (2 or 5.7%), at a rented room in a hotel (1 or 2.9%), or in a bush at a remote location (8 or 22.9%).

Modus Operandi

This study examined patterns of weapon use and modus operandi in the femicides and the suicides that followed. The overwhelming majority of homicide victims (22 or 62.9%) were killed with a firearm. The second most common method for femicide was hacking or cutting with a machete (10 or 28.9%). Two (5.7%) femicide incidents involved strangulation, while one (2.9%) involved setting the victim ablaze.

Shooting with a firearm was the most common suicide method (18 or 51.4%). The second most common suicide method was ingesting poison, usually Dichloro Diphenyl Trichloroethane (DDT) or some other agrochemical (7 or 20%). Hanging occurred 5 times (14.3%), while 2 (5.7%) femicide perpetrators killed themselves through self-inflicted knife wounds to the neck and abdomen. Assailants who fatally shot their victims were more likely to use shooting as a suicide method as well. In 17 (48.6%) incidents, the assailant fatally shot his intimate partner before lethally shooting himself.

Premeditation

The premeditative elements of a homicide–suicide event typically include an assailant threatening to kill the victim if she follows through with a separation or divorce, or refuses reconciliation following a separation; stalking the victim; purchasing a firearm shortly or immediately before the homicidal event; lying in wait for or ambushing the victim (Dawson, 2005). The data show that 26 (74.3%) of the 35 cases examined in this study evinced elements of premeditation or prior planning. Cases involving estrangement and the threat of divorce or separation were more likely to exhibit premeditative elements. Femicide victims in these cases were frequently harassed, routinely stalked, and recurrently threatened with violence prior to the homicidal event, and ultimately they were killed when they rejected the assailant’s demands for reconciliation. Before arriving at the crime scene, the assailants typically had armed themselves with firearms, machetes, knives, ropes, insecticides, and other lethal weapons for carrying out the femicides and suicides. To illustrate, in 5 of the cases where the assailant and the victim were separated, the assailant ambushed, then murdered the victim on her way to the farm. In 3 other cases, the assailant tracked the victim to her farm and killed her as she worked. In 5 other cases, also involving estrangement, the assailant traced the victim to her natal home or new abode and murdered her. In 1 case, a man purchased a gun just days before the femicide–suicide. In another case, a gun-bearing husband stormed his wife’s church during a service, shot and killed the preaching pastor whom he accused of having an affair with his wife, then murdered his wife, before turning the gun on himself.

Incident Witnessed by Children

Four (11.4%) of the intimate partner femicides occurred in the presence of children of the couple involved. One femicide was witnessed by the couple’s four children, ranging in age from 9 to 17 years. One femicide was witnessed by the couple’s two children aged 8 and 5 years old; one femicide was witnessed by the couple’s 6-year-old son, whereas another one occurred in the presence of the couple’s 3-year-old daughter.

Motivation and Contexts

Sexual jealousy, suspicion of infidelity, and relationship termination (e.g., divorce, estrangement) constituted the most common precipitating factors in IPFS. In 13 (37.1%) out of the 35 cases, assailants killed female intimate partners that they suspected had been sexually unfaithful to them. Estrangement, divorce, and the victim’s threat to end the relationship were the motivating factor in another 13 (37.1%) out of the 35 cases. Marital quarrels of variable sources were factors in another 6 (17.1%) femicide–suicides. The motivations for the remaining 3 femicide–suicide cases were reported as unknown or not yet established. In none of the cases studied was it reported that the assailant was mentally ill or exhibited symptoms of psychopathology. However, this may be due to the fact that in Ghana there are currently very limited mental health services. In addition, there is a major stigma associated with receiving any form of psychiatric treatment. Thus, a large proportion of mental disorders and illnesses go undiagnosed and untreated (Barke, Nyarko, & Klecha, 2011).

Descriptive accounts of a random selection of cases are provided below to illustrate the three major categories of motivations and situational contexts for IPFS.

Sexual Jealousy and Suspicion of Infidelity

The couple in Case 3 left behind 6 children, aged 4 to 17 years of age. The facts indicate that the assailant suffered a mild stroke, which left him with erectile dysfunction. He incessantly quarreled with his wife whom he accused of having sexual relations with several young men in the community. In a prelude to the lethal assault, the husband started an altercation pertaining to the same accusation. He fatally shot his wife before turning the weapon on himself.

In Case 6, a husband stormed the church where his wife had gone to worship one evening. He shot the preaching pastor to death before fatally shooting his wife and then himself on the premises of the church. According to reports, he suspected her of having a sexual relationship with the pastor.

When the 25-year-old victim in Case 9 told her 60-year-old husband that she was pregnant, little did she know that the news would lead to her own murder and the subsequent suicide of her husband. Her husband responded angrily to the news, denying responsibility for the pregnancy and accusing her of sexual infidelity. He fatally stabbed her before hanging himself in the family dwelling.

Case 11 involved a 70-year-old man and his 60-year-old wife, who had 12 children together. The husband accused his wife of being a “flirt” and could not withstand the thought of other men making sexual advances toward her. Following an incident in which the wife reportedly returned from a funeral and teased him about receiving a handkerchief from a potential lover, he hacked her to death with a machete on their jointly owned farm before drinking a lethal dose of a poisonous chemical.

In Case 14, a man who suspected his wife of carrying on a sexual affair with another man, shot her to death before fatally shooting himself. Moments before the wife’s murder, she had received a phone call on a cellular phone, but would not reveal to him the identity of the caller.

In Case 23, a man killed his wife to whom he had been married for 7 years, following suspicion that the numerous phone calls she had been receiving on her cellular phone came from other men with whom she was having extramarital sexual affairs. On the day of the murder, he secretly took her cellular phone with him to the family farm to monitor incoming calls and to retrieve messages and phone numbers stored on the phone. When the wife noticed the phone missing, she left for the farm, accompanied by the couple’s 3-year-old daughter, to reclaim her phone. There, he hacked her to death with a machete and subsequently hanged himself.

In Case 34, a man procured a gun barely a month prior to the femicide–suicide. Three months prior to the lethal events, he had accused his wife of having an affair with her ex-husband. Following one such quarrel where the husband leveled charges of infidelity, the woman moved out of the residence, returning after a formal reconciliation through their families. On the night of the murder, the husband raised the accusation of infidelity again, this time fatally shooting his wife before killing himself with the same gun.

Estrangement, Threat of Separation, and Relationship Termination

As noted, in a large number of cases, threats of separation, estrangement, and victim-initiated divorce acted as precipitants to femicide and subsequent perpetrator suicide. In Case 1, the couple had been married for several years and had three children; the wife abandoned the marital home following a protracted family quarrel, moving with her belongings into her father’s home. When she declined persistent reconciliatory overtures from the husband to return to the marriage, he ambushed her on her farm, lethally shooting her twice in the head before fatally turning the gun on himself. Case 7 was a strikingly similar case; after the wife threatened divorce and moved her belongings to her natal home, the husband tracked her to her new farm and shot her dead before subsequently fatally shooting himself.

In Case 21, a woman and her 6-year-old son were ambushed and shot dead by the woman’s estranged husband while returning from a public toilet at 6:15 a.m. The woman had left her matrimonial home and taken residence with her family in another town, following a separation. Agitated by the separation and the impending divorce, he monitored her movements before murdering her.

In Case 26, a 30-year-old man shot and killed his 28-year-old wife for initiating divorce proceedings. The man apparently felt aggrieved by the fact that he had paid the cost of his wife’s beautician apprenticeship and financed the construction of her beauty salon.

In Case 35, a 20-year-old man killed his 18-year-old girlfriend when she broke off the relationship to devote her life in service of her newfound God. She had just become a “born again” Christian.

Miscellaneous Marital Quarrels

A number of femicide–suicides were preceded by domestic quarrels of variable sources. The couple in Case 4 had been married for 26 years. Following a series of marital quarrels spanning several weeks, the wife reportedly teased the husband over his erectile dysfunction, implying that he “[was] not a man.” The following day, while working on their farm, the husband fatally shot her before turning the gun on himself.

In Case 8, a man was irritated by his girlfriend’s importunate demands for money to pay for a new hairstyle. He retrieved a gun from an adjoining room, shot her dead and later fatally shot himself.

In Case 20, a man shot and killed his wife before killing himself when the woman spurned his 2:00 a.m. sexual advances. The wife had told him that she feared the prospect of contracting HIV–AIDS from him because she suspected him of having sexual relations outside their marriage. After fatally shooting her, he later hanged himself in a thicket nearby.

In Case 22, a woman asked permission from her husband to visit her sick mother in another town. The husband refused, saying that it was a ruse to visit another boyfriend. During the ensuing verbal altercation, he hacked her to death with a machete before fleeing the scene. His corpse was later found hanging in the public cemetery.

Discussion and Conclusion

This study contributes to the scant extant literature on IPFS in non-Western, nonindustrialized societies. By examining and describing the scope, patterns, situational contexts, and precipitating circumstances, as well as the sociodemographic characteristics of victims and perpetrators in 35 IPFS in Ghana, this study has contributed to our understanding of this tragic form of lethal aggression. Indeed, several of the findings of the current study are consistent with those of existing research. As with previous studies, the present study found that in Ghana IPFS cases were more likely to be premeditated, involve firearms, and to be ignited by victim-initiated divorce or estrangement and the male perpetrator’s suspicion of his partner’s marital infidelity.

Perhaps the most notable finding to emerge from this study is that suspicion of partner infidelity was a dominant precipitating factor in police-assigned motives for the crimes. Another noteworthy finding is that a large proportion of femicide–suicides were triggered by estrangement and victim-initiated divorce. The data showed that women who were already estranged from male partners or those who initiated or threatened divorce or separation were highly vulnerable to fatal abuse; they were repeatedly threatened with lethal violence, routinely stalked, and ultimately killed, prior to the assailant’s suicide.

The findings of the current study are also congruent with the results obtained from previous studies of IPFS regarding the predominance of firearm as the dominant modus operandi. The lethality of firearms ensures that both the homicide and suicide can be executed within a short time span. Indeed, recent analyses of Ghanaian society report noticeable increases in the quantity and quality of firearms in the society (McCaskie, 2008). Such a discernible increase in the volume of firearms and expansion in their accessibility bodes ill for the society. Research shows that when a violent interaction occurs between two persons in a domestic setting, a gun in the home immensely increases the risk of a homicide (Kellermann et al., 1993). Thus, the increasing availability of guns in Ghanaian homes is a growing risk factor for familial violence. It accentuates the prospect of a trivial or mundane marital quarrel degenerating into a fatal assault. An argument or altercation that at worst would have resulted in a physical wounding can now escalate into lethal violence.

Evidence of premeditation was apparent in many of the femicidal killings studied. In many of the femicide–suicide cases that involved estrangement, divorce, and threats of relationship termination, there were forewarnings of violence prior to the lethal incidents. Some of the women had previously been warned by the assailants that they would be killed if they made good on their threats of separation or divorce. The lethal assault constituted the execution of that threat. In some cases, the separations themselves were initiated by the victims in response to chronic physical abuse at the hands of their partners. In addition to previous histories of violence toward their victims, some men who did not previously own firearms purchased weapons as they made preparations to carry out their fatal assaults. As noted above, in one case (Case 34), a quiet unassuming husband purchased a gun just days before the femicide–suicide. Many tenants in his multifamily dwelling unit felt it was out of character and disposition for him to own and possess a gun. Little did they know that his plans were to use the weapon against his wife and then himself.

This brings us to the issue of the vulnerability of women and girls in Ghana to lethal and nonlethal assaults. The current lack of emergency shelters for Ghanaian women fleeing abuse and violence or those contemplating divorce constitutes a serious and daunting obstacle to the prevention of lethal and nonlethal abuse of women. The data show that many women were killed as they sought to separate from or divorce their abusive husbands, ex-husbands, boyfriends, or lovers. There were numerous case examples of women who separated from husbands or boyfriends, only to be stalked and killed a few hours or days later. Some had taken refuge at their natal homes, but were still vulnerable to fatal attacks. Absent emergency shelters where they could anonymously seek safety, many went to their natal homes where they could easily be tracked by their assailants.

The literature on homicide–suicides indicates that as a phenomenon, homicide–suicides are extremely rare events. In Ghana, the exceedingly low number of homicide–suicides and femicide–suicides that occurred over the study period may have been impacted by cultural factors. It should be noted that in Ghana, attempted suicide as well as homicide are criminal offenses. Deaths from suicidal and homicidal acts are also considered “bad deaths” with serious social and magico-religious repercussions for the victim, perpetrator, and their extended kin. Among the Akans of Ghana, for example, people who die “bad deaths” are considered atofo and their deaths as atofowuo. The decedents are denied proper burial and funerary rituals. It is believed that the spirits of such persons cannot become ancestors or be reincarnated into new births. Their spirits, known as nsamantwentwen or samantefie (roaming ghost or “ghost in the house”), will linger at their death scene, frightening innocent people, albeit unable to harm them (Adinkrah, 2012). It is likely that the opprobrium surrounding homicidal and suicidal deaths acts as a deterrent for this typically premeditated familial tragedy.

The present research suggests practices, policies and services that could be adopted to alleviate the incidence of IPFS. First, there is an urgent need to provide refuge centers or shelters for women fleeing domestic violence. Second, in Ghana, concerted efforts should be directed toward ensuring that the rural populace has access to the material assistance and counseling services extended to women facing abuse. At present, the majority of the social service agencies and counseling centers are located in the cities and major urban centers. As a result, women who live in communities distant from these metropolitan areas lack counseling and domestic violence legal advocacy services, which puts them at greater risk for nonlethal and lethal violence. Third, resources need to be directed toward more efficiently facilitating the enforcement of Ghana’s Domestic Violence Act (Act 732). Passed by the government in 2007, the Domestic Violence Act is intended “to provide protection from domestic violence particularly for women and children and for connected purposes.” Stakeholders have criticized the government for not providing the necessary money, infrastructural resources, and services in support of the law (“URBANET Chides Gov’t,” 2012).

Some limitations of this study must be recognized. First, the study focused on femicide–suicides reported in a respected Ghanaian national daily newspaper. Despite the high reliability of the data, unfortunately, as is the case with all criminological studies relying on secondary data, there may a problem with the completeness of the data. Despite the dramatic and tragic nature of femicide–suicides, the high public interest that homicidal and suicidal events generate, and the zealousness with which crime reporters pursue such stories in Ghana, it is still possible that there are some femicide–suicide cases that were unreported to the authorities and thus never reached the attention of newspaper reporters. It is also possible that some cases were overshadowed by more high-profile news events of the day, whereby editorial discretion may have been exercised not to publish the femicide–suicide story. The second shortcoming is a logistical problem that confronts all homicide studies. Despite the assiduousness of medical examiners and police homicide investigators to investigate and assign the causes of death, police and medical examiners may still misclassify the cause of death, listing a killing as a homicide–suicide when, in reality, it may have been a death resulting from a suicide pact or an accident. In conclusion, additional research on IPFS in non-Western societies will continue to enhance our knowledge of the nature, distribution of, and etiology of IPFS and the necessary steps to combat it.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Mensah Adinkrah, PhD, is professor of sociology and criminal justice at Central Michigan University, Mount Pleasant, Michigan. His current research interests focus on homicide, suicide, and witchcraft-related violence from a comparative perspective. He was a United States Senior Fulbright research scholar to Ghana during 2004-2005. Some of his previous publications have appeared in such journals as Homicide Studies, Social Science and Medicine, and Aggressive Behavior.

References

Adinkrah

(2003). Homicide-suicides in Fiji: Offense patterns, situational factors, and sociocultural contexts. Suicide and Life-Threatening Behavior, 33, 65-73.

Adinkrah

(2004). Witchcraft accusations and female homicide victimization in contemporary Ghana. Violence Against Women, 10, 325-356.

Adinkrah

(2008a). Husbands who kill their wives: An analysis of uxoricides in contemporary Ghana. International Journal of Offender Therapy and Comparative Criminology, 52, 296-310.

Adinkrah

(2008b). Spousal homicides in contemporary Ghana. Journal of Criminal Justice, 36, 209-216.

Adinkrah

(2011a). Criminalizing rape within marriage: Perspectives of Ghanaian university students. International Journal of Offender Therapy and Comparative Criminology, 55, 982-1010.

Adinkrah

(2011b). Epidemiologic characteristics of suicidal behavior in contemporary Ghana. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 32, 31-36.

Adinkrah

(2011c). Patterns of female suicidal behavior in Ghana. Psychological Reports, 109, 649-662.

Adinkrah

(2012). Better dead than dishonored: Masculinity and male suicidal behavior in contemporary Ghana. Social Science & Medicine, 74, 474-481.

Adinkrah

(2013). Criminal prosecution of suicide attempt survivors in Ghana. International Journal of Offender Therapy and Comparative Criminology, 24, 1477-1497.

10.

Adler

J. S.

(1999). “If we can’t live in peace, we might as well die”: Homicide-suicide in Chicago, 1875-1910. Journal of Urban History, 26, 3-21.

11.

Afrifa

(1994). The problem of child abuse and child neglect in Ghana. In Mensa-Bonsu

H. J. A. N.

Dowuona-Hammond

(Eds.), The rights of the child in Ghana—Perspectives (pp.10-17). Accra, Ghana: Woeli Publishing Services.

12.

Alpers

Wilson

(2012). Guns in Ghana: Facts, figures and firearm law. Sydney School of Public Health, University of Sydney. Retrieved from http://www.gun-policy.org/firearms/region/ghana

13.

Amoakohene

I. M.

(2004). Violence against women in Ghana: A look at women’s perceptions and review of policy and social responses. Social Science & Medicine, 59, 2373-2385.

14.

Appiahene-Gyamfi

(2009). Crime and punishment in the Republic of Ghana: A country profile. International Journal of Comparative and Applied Criminal Justice, 33, 309-324.

15.

Banks

Crandall

Sklar

Bauer

(2008). A comparison of intimate partner homicide to intimate partner homicide-suicide: One hundred and twenty-four New Mexico cases. Violence Against Women, 14, 1065-1078.

16.

Barke

Nyarko

Klecha

(2011). The stigma of mental illness in Southern Ghana: Attitudes of the urban population and patients’ views. Social Psychiatry & Psychiatric Epidemiology, 46, 1191-1202.

17.

Berg

(Producer). (2005). Witches in exile [Motion picture]. San Francisco: California Newsreel.

18.

Brookman

(2005). Understanding homicide. Thousand Oaks, CA: Sage.

19.

Dawson

(2005). Intimate femicide followed by suicide: Examining the role of premeditation. Suicide and Life-Threatening Behavior, 35, 76-90.

20.

Duodu

(2008, March 25). Reform customs affecting women, children. Daily Graphic, p. 11.

21.

Easteal

(1994). Homicide-suicides between adult sexual intimates: An Australian study. Suicide and Life-Threatening Behavior, 24, 140-151.

22.

Ghana. (2011). Retrieved from http://www.infoplease.com/ipa/A0107584.html?pageno=1

23.

Ghana Police Service. (2010). Annual report. Accra, Ghana: Author.

24.

Graphic still no. 1. (2001, June 16). Daily Graphic, p. 1.

25.

Greene

S. E.

(2002). Sacred sites and the colonial encounter: A history of meaning and memory in Ghana. Bloomington: Indiana University Press.

26.

Jankey

O. M.

(2009). “Passion killings”: A media and cultural perspective of intimate femicide in Botswana (Unpublished doctoral dissertation). University of Utah, Salt Lake City.

27.

Kellermann

A. L.

Rivara

F. P.

Rushforth

N. B.

Banton

J. G.

Reay

D. T.

Francisco

J. T.

. (1993). Gun ownership as a risk factor for homicide in the home. New England Journal of Medicine, 329, 1084-1091.

28.

Koziol-McLain

Webster

McFarlane

Block

C. R.

Ulrich

Glass

Canpbell

J. C.

(2006). Risk factors for femicide-suicide in abusive relationships: Results from a multisite case control study. Violence and Victims, 21, 3-21.

29.

Krulewitch

C. J.

(2009). Epidemiology of intimate partner homicide-suicide events among women of childbearing age in Maryland, 1994-2003. American Journal of Forensic Medicine and Pathology, 30, 362-365.

30.

Kuada

J. E.

Chachah

(1999). Ghana: Understanding the people and their culture. Accra, Ghana: Woeli Publishing Services.

31.

Liem

(2010). Homicide followed by suicide: A review. Aggression and Violent Behavior, 15, 153-161.

32.

Liem

M. C. A.

Koenraadt

(2007). Homicide-suicide in the Netherlands: A study of newspaper reports, 1992-2005. Journal of Forensic Psychiatry & Psychology, 4, 482-493.

33.

Lund

L. E.

Smorodinsky

(2001). Violent death among intimate partners: A comparison of homicide and homicide followed by suicide in California. Suicide and Life-Threatening Behavior, 31, 451-459.

34.

Malphurs

J. E.

Cohen

(2002). A newspaper surveillance study of homicide-suicide in the United States. American Journal of Forensic Medicine and Pathology, 23, 142-148.

35.

Mathews

Abrahams

Jewkes

Martin

L. J.

Lombard

Vetten

(2008). Intimate femicide-suicide in South Africa: A cross-sectional study. Bulletin of the World Health Organization, 86, 552-558.

36.

McCaskie

T. C.

(2008). Gun culture in Kumasi. Africa: The Journal of the International African Institute, 78, 433-454.

37.

Morton

Runyan

C. W.

Moracco

K. E.

Butts

(1998). Partner homicide-suicide involving female homicide victims: A population-based study in North Carolina, 1988-1992. Violence and Victims, 13, 91-106.

38.

Nielssen

O. B.

Large

M. M.

Westmore

B. D.

Lackersteen

S. M.

(2009). Child homicide in New South Wales from 1991 to 2005. Medical Journal of Australia, 190, 7-11.

39.

Nikunen

(2006). Parenthood in murder-suicide news. Idealized Fathers and murderous mums. Journal of Scandinavian Studies in Criminology and Crime Prevention, 7, 164-184.

40.

Nukunya

G. K.

(2004). Tradition and change in Ghana: An introduction to sociology. Accra: Ghana Universities Press.

41.

Palermo

(1994). Murder-suicide: An extended suicide. International Journal of Offender Therapy and Comparative Criminology, 31, 205-216.

42.

Pirkis

J. E.

Burgess

P. M.

Francis

Blood

R. W.

Jolley

D. J.

(2006). The relationship between media reporting of suicide and actual suicide in Australia. Social Science & Medicine, 62, 2874-2886.

43.

Rude

(1999). Reasonable men and provocative women: An analysis of gendered domestic homicide in Zambia. Journal of Southern African Studies, 25, 7-27.

44.

Selby

(2008). Ghana: Inhumane aspects of cultural practices. Retrieved from http://www.fgmnetwork.org/gonews.php?subaction=showfull&id=1214172902&archive

45.

Selkin

(1976). Rescue fantasies in homicide-suicide. Suicide and Life-Threatening Behavior, 6, 79-85.

46.

Sillito

C. L.

Salari

(2011). Child outcomes and risk factors in U.S. homicide-suicide cases 1999-2004. Journal of Family Violence, 26, 285-297.

47.

Steeves

R. H.

Parker

(2007). Adult perspectives on growing up following uxoricide. Journal of Interpersonal Violence, 22, 1270-1284.

48.

Steeves

R. H.

Parker

Laughon

Knopp

Thompson

M. E.

(2011). Adolescents’ experiences with uxoricide. Journal of the American Psychiatric Nurses Association, 17, 115-123.

49.

Takyi

B. K.

Mann

(2006). Intimate partner violence in Ghana: The perspectives of men regarding wife beating. International Journal of Sociology of the Family, 32, 61-78.

50.

URBANET chides gov’t over frail commitment to fight domestic violence. (2012, April 29). Retrieved from http://www.ghanaweb.com/GhanaHomePage/NewsArchive/artikel.php?ID=237380

51.

Warren-Gordon

Byers

B. D.

Brodt

S. J.

Wartak

Biskupski

B. S.

(2010). Murder followed by suicide: A newspaper surveillance study using the New York Times Index. Journal of Forensic Sciences, 55, 1592-1597.

52.

Websdale

(2010). Familicidal hearts: The emotional styles of 211 killers. New York: Oxford University Press.

53.

West

D. J.

(1965). Murder followed by suicide. Cambridge, MA: Harvard University Press.

54.

World Factbook. (2008). Washington, DC: Government Printing Office.