Content uploaded by Wilfreda E Thurston
Author content
All content in this area was uploaded by Wilfreda E Thurston
Content may be subject to copyright.
Understanding ‘Passion Killings’ in Botswana:
An Investigation of Media Framing
By Deinera Exner1 and Wilfreda E. Thurston2
Abstract
In Botswana, local news media outlets have documented the prevalence of so-
called ‘passion killings’; however, no published studies have been conducted that
examine these intimate partner homicides. Using ethnographic content analysis informed
by a theory of framing, this study investigated the characteristics of these crimes, and
societal attitudes, myths and stereotypes regarding intimate partner homicides and
passion killings. Articles from four Batswana newspapers were analyzed. The
information derived from this analysis is used to develop future directions for the study of
intimate partner violence and homicides in Botswana.
Keywords: intimate partner homicide; passion killings; Botswana; media framing
Introduction
Despite increasing attention from the international community, violence against
women (VAW) remains a global public health and human rights issue, and continues to
impact the physical, mental, sexual and reproductive health of women (Green, 1999;
World Health Organization, 2005). According to the World Health Organization (2005),
one of the most common forms of VAW is that committed by a male husband or partner.
This intimate partner violence (IPV) is not limited to violent acts or threats—intimate
partner homicide (IPH) comprises a subset of this violence. In Botswana, a country in
sub-Saharan Africa, these homicides are locally referred to as ‘passion killings,’ and are
increasingly prevalent (Alao, 2006). Despite their brutal nature, little attention has been
paid to the so-called passion killings in the research literature; however, media reports
provide information about, and insight into, these crimes. This study endeavored to
investigate the portrayal of IPH and passion killings in Batswana3 news media, in order
to expand understanding of IPV in Botswana.
Intimate Partner Violence in Botswana
Though limited, the published research literature provides some discourse around
IPV in Botswana (Becker, 2003; Macdonald, 1996; Maundeni, 2002; McCall & Resick,
2003; Wilson & Daly, 1993). According to this literature, IPV is widespread, and in part
1 Deinera Exner, BSc, completed her bachelor's degree in microbiology at the University of Calgary,
Alberta, Canada, in April 2007. She worked on this study as part of an Undergraduate Directed Study in
Gender-based Analysis and Health Policy. She has continued her education, doing a Master of Public
Health degree at Boston University.
2 Wilfreda E. Thurston, PhD, is a Professor in the Department of Community Health Sciences, Faculty of
Medicine, and Director, Institute for Gender Research, at the University of Calgary. Her program of
research and training includes prevention of gender based interpersonal violence against women; the
interplay of gender, culture and socioeconomics as determinants of health, development and evaluation of
health promotion programs; and public participation as a key tenet of population health.
3 Batswana refers to the people of Botswana.
Journal of International Women’s Studies Vol. 10 #4 May 2009 1
stems from the patriarchal gender role system of traditional Tswana culture,4 where this
violence was considered acceptable and commonplace (Macdonald, 1996; Maundeni,
2002). Phaladze and Tlou (2006) identify the important role of past Batswana cultural
norms in understanding the current situation of women in Botswana. For example, under
Batswana customary law,5 women were traditionally considered a minor and under their
husband’s sole guardianship. Though this law no longer stands, it has consequently
continued to “entrench women’s subordination to men” (Phaldze & Tlou, 2006, p. 27),
and must be considered when investigating the prevalence of IPV. Maundeni (2002)
continues this discourse in his study of wife abuse in Botswana. In his discussion of the
social stigma attached to women experiencing domestic violence, he postulates that
among other things, cultural factors not only play a key role in ongoing IPV, but are also
primary reasons why women stay in abusive relationships; that is, women are socialized
to accept their inferior status in society and their subordination to men. Several authors
emphasize the importance of recognizing the impact of conservative gender norms on
cultural acceptability of partner abuse in patriarchal societies; in Botswana, as in other
sub-Saharan African countries, women’s violation of these norms is seen as justification
for the perpetration of IPV (Maundeni, 2002; Rani, Bonu, & Diop-Sidibe, 2004).
Only one published study was found to discuss the trend of passion killings in
Botswana. According to Alao (2006), passion killings are “viewed as a sign of patriarchal
crisis” (p. 341), and are “directed at females, where either a husband or boyfriend decides
to kill the female partner” (p. 341). Again, the role of patriarchy is emphasized. From this
definition, it appears that the importance of the term ‘passion killings’ may lie in its
social, as opposed to definitional, significance. In order to explore this, it was decided to
examine the media framing of passion killings and IPH in Botswana.
Media Framing of Intimate Partner Violence
Several authors have discussed the importance of news media in bringing
awareness to social issues, and in forming the social construction of a problem (Bullock
& Cubert, 2002; Hubbard, DeFleur, & DeFleur, 1975; Meyers, 1997). According to Pan
& Kosicki (1993), news media frame stories so as to provide central organizing thoughts;
Entman (1993) adds that these central thoughts serve to “promote a particular problem,
definition, causal interpretation, moral evaluation, and/or treatment recommendation” (p.
52), and that framing involves “select[ing] some aspects of perceived reality and
mak[ing] them more salient in a communicating text” (p. 52). Through this process of
framing, the media influence the issues citizens are concerned about, and how urgent they
view the problem to be (Sacco, 1995; Terkildsne & Schnell, 1997).
Berns (2001) discusses the importance of analyzing media framing because
“individuals draw on these sources when constructing their understanding of issues” (p.
263); in her study, she demonstrated that the media served to resist attempts to place
domestic violence within a patriarchal framework by degendering the problem.
Furthermore, the construction of a problem can affect whether or not the public believes
it is a social problem worthy of change (Loseke, 1999). Several studies have examined
4 According to Maundeni (2002), traditional Tswana culture refers to the “period before missionaries and
other white people came to Botswana” (p. 273), which is the period prior to 1802.
5 Customary law is based on the particular beliefs and customs of a certain tribe (Maundeni, 2002), and in
this case refers to the laws of the Tswana tribes in traditional Tswana culture.
Journal of International Women’s Studies Vol. 10 #4 May 2009 2
media framing of domestic violence, and all generally agree that women, the probable
victims, are portrayed as being at fault for the violence (Berns, 1999; Bullock & Cubert,
2002; Consalvo, 1998; Maxwell et al., 2000; Meyers, 1994). Meyers (1997) suggests that
the media encourage the continuation of IPV by maintaining the idea of male supremacy.
Given the role media framing plays in how particular issues are constructed and
understood, it is critical to investigate how the local media portray IPH and passion
killings to the Batswana public. Specifically, this study set out the following research
objectives: 1) determine how IPH are characterized by the Batswana media (e.g., what
cases were presented? who was involved?); 2) explore the media framing of these
homicides by analyzing articles discussing specific cases of IPH; and 3) explore the label
‘passion killings,’ in order to determine when and how it is used, and if these killings
differ from IPH in any important way.
Methods
The research team, not being located in Botswana, was reliant on Batswana
newspapers that were electronically available and that had electronically available
archives. As such, four newspapers – Botswana Guardian, Daily News (Botswana Press
Agency (BOPA)), Midweek Sun, and Mmegi – were selected because they had electronic
archives accessible within the public domain.6 A six month sampling timeframe (August
1, 2006 to January 31, 2007) was selected as a half year was considered sufficient for the
purposes of exploration.
Criteria were developed to aid in the selection of media articles from within the
determined timeframe for inclusion in the study. Articles were selected if they discussed
a specific case of IPH (Objectives 1, 2 and 3), or were an opinion or general information
article that specifically discussed passion killings (Objective 3). To be included as a case
specific article, the article had to discuss a death, an attempted murder or a reported
threat-to-kill (included as it appeared to be a precursor to cases of IPH), and the persons
involved in the violence had to be current or former romantic partners, with the
relationship specified in the article. The homicide could have involved a male perpetrator
and female victim; a female perpetrator and male victim; a male perpetrator and male
victim; or a female perpetrator and female victim. However, it was expected that no
articles would be found on same-sex violence, as homosexuality is illegal in Botswana
(Baffour-Awuah, 2004). Cases that fit the above criteria were selected whether or not the
article contained the phrase passion killing(s). Including cases that fit the criteria of IPH
but were not labeled as a passion killing allowed for a determination of any critical
differences between the two. To be included as an opinion article, the article had to
contain the phrase passion killing(s) or a clear reference to these killings (alternate
labeling such as love killings), and the primary focus of the article had to be on this
subject. General information articles also had to contain the phrase passion killing(s), or a
clear reference, but the article did not have to focus on passion killings specifically (e.g.,
could briefly mention prevalence).
Each edition of the four selected newspapers published within the sampling
timeframe was reviewed for articles that fit the study criteria. Some newspapers
6 At the time of the study, two other Botswana papers, The Voice and The Botswana Gazette, were also
electronically available but neither included archives, and thus could not be used for this study.
Journal of International Women’s Studies Vol. 10 #4 May 2009 3
published daily and some weekly. For case specific articles, both initial reports on the
case, as well as information about the trial, if applicable, were included.
To supplement the research and ensure that Objective 3 was fully explored, an
additional search was performed for historical opinion/general information articles
written between January 1, 2003 and July 31, 2006 and containing the phrase passion
killing(s). An academic paper on passion killings in Botswana suggested there was an
increase in the annual number of killings beginning in 2004 (Alao, 2006), so the start date
for the ‘historical’ search was chosen to collect information from just before this period
through to the start date of the ‘current’ search, to potentially allow an exploration of
reasons for the increase in these crimes. To conduct the historical search, the term
passion killing* was entered into Google™ Botswana,7 and articles returned in this online
search that fell within the desired date range and fit the criteria for an opinion/general
information article were selected. Although this search was initially intended to find
opinion/general information articles, it was noted that several of the articles also
mentioned specific cases. The research team decided that additional information about
these cases might provide data relevant to the study, and thus allow a more thorough
response to the research objectives. Therefore, the name(s) of victims and/or perpetrators
of a passion killing mentioned in the historical opinion/general information articles were
also put into a search performed in Google™ Botswana, and articles returned that fit the
criteria for either a case specific article or an opinion/general information article, and fit
within the historical search dates, were selected. This iterative search process was
intended to identify all relevant articles.
From the current search (i.e., articles written between August 1, 2006 and January
31, 2007), 32 articles were found: 18 case specific articles and 14 opinion/general
information articles. The historical search (i.e., articles written between January 1, 2003
and July 31, 2006) returned an additional 17 articles that fit the study criteria: 6 case
specific articles and 11 opinion/general information articles.8 In total then, 49 articles
(case specific (n=24) and opinion/general information (n=25)) were included in the
sample for this study. The majority of articles (n=45) came from two of the newspapers,
Daily News and Mmegi. All the articles were imported into the QSR N6© computer
program for qualitative data analysis and coding was done by one of the research team
members. Each article was coded as a whole (i.e., titles and subheadings were not coded
separately), and a sentence was designated a text unit. In total, 993 text units were coded.
Articles were analyzed through ethnographic content analysis (ECA) informed by
a theory of framing. As described by Altheide (1987), ECA is a “reflexive analysis of
documents” (p. 65) that “supports a theoretically informed account of media content” (p.
74). According to Entman (1993), content analysis informed by framing “avoid[s]
treating all negative or positive terms…as equally salient and influential” (p. 57). For this
study, it was important to determine the framing of the articles, as content analysis
without framing may have missed key media messages that are perceived by the
audience. ECA was performed on all selected articles to determine what words and
themes were common to the discussion of IPH and/or passion killings. Prior to coding,
two templates, a facts template and a frames template, were created by the research team
7 http://www.google.co.bw
8 The 17 articles found in the historical search were from four sources: Daily News, Mmegi, News from
Africa and The BBC (Gaborone).
Journal of International Women’s Studies Vol. 10 #4 May 2009 4
based on previous research in media framing, and applied to both sets of articles (case
specific and opinion/general information). During coding, the facts template remained as
initially developed, while the frames template developed through an iterative process.
Results
Objective 1: The Characterization of Intimate Partner Homicides
Characteristics. In the 24 case specific articles analyzed, a total of 23 cases of
IPH were reported (see Table 1). All cases except two involved a single male perpetrator
and a single female victim. In the cases that deviated, one included male and female
perpetrators and female and male victims (i.e., the perpetrators killed their ex-intimate
partners – his girlfriend and her boyfriend). In the other, the female victim committed
suicide, as opposed to being murdered by the male perpetrator. While suicide was not
originally included in the criteria for IPH, this suicide appeared to be the result of
ongoing IPV by the perpetrator. Furthermore, while the article described the death as a
suicide, it was not entirely clear that this was actually the cause of death; it may have
been a homicide reported as a suicide. Therefore, it was decided that this death would be
included as a case of IPH. In 4 of the 23 cases, the perpetrator committed suicide
following the murder.
Table 1: Characteristics of 23 cases of IPH reported in Batswana news media,
January 1, 2003 to January 31, 2007
Perpetrator Victim Relationship a Crime
Age Sex Age Sex Location Date Type
b
NL M 18 F BG Lobatse 1997 Murder and
hanging
NL M 17 F BG Gaborone 2003 NL
28 M 25 F NL NL 2003 Stabbing (of
victim)/
suicide (of
perpetrator)
NL M&F NL F&M Ex-boyfriend of
the female
perpetrator and
ex-girlfriend of
the male
perpetrator
Francistown 2003 Burned to
death
(victims)/
suicide (of
male
perpetrator)
NL M NL F E Kumakwane 2003 NL
NL M NL F BG estranged NL Dec
2005 Burned to
death
54 M NL F BG - LT - CH Francistown Jan
2006 TTK
21 M 18 F BG - LT Francistown Aug
2006 Stabbing
NL M NL F BG - CH Francistown Aug
2006 TTK
Journal of International Women’s Studies Vol. 10 #4 May 2009 5
NL M NL F BG Gaborone Aug
2006 Killed with
axe
35 M NL F BG Francistown Oct
2006 TTK
32 M 21 F BG - LT Moshupa Nov
2006 Stabbing
29 M 24 F L Gaborone Nov
2006 Stabbing (A)
NL M NL F BG - LT Nkoyaphiri Dec
2006 Victim
suicide
(burned to
death) c
50 M NL F MM Thangane Dec
2006 TTK
NL M NL F MM Kweneng Dec
2006 Shooting (of
victim)/
suicide (of
perpetrator)
50 M NL NLd L Mokgenene Dec
2006 Stabbing
NL M NL F MM (for 10
days) Malhalapye 2006 Murder (of
victim)/
suicide (of
perpetrator)
33 M NL F BG Francistown 2006 Stabbing (A)
29 M NL F BG - LT Gaborone Jan
2007 Stabbing
26 M NL F BG NL Jan
2007 Stabbing (A)
NL M 29 F Former BG Gaborone Jan
2007 Stabbing
Legend
NL = not listed; M = male; F = female; BG = boyfriend/girlfriend; LT = lived together;
E = engaged; MM = married; L = lovers; CH = had a child together; TTK = threat to
kill; A = attempted murder
Notes
(a) Relationship between perpetrator and victim as identified in article; (b) Type of crime
refers to how victim was murdered (unless otherwise specified); (c) In this case of victim
suicide, the perpetrator was not a perpetrator of IPH, but rather a perpetrator of IPV
(the article discussed that IPV occurred prior to the victim’s suicide);(d) In this case, the
article did not specify the sex of the victim. However, as it stated that that the perpetrator
and victim were lovers, and since homosexuality is illegal in Botswana (Baffour-Awuah,
2004), the victim is presumed to be female.
Journal of International Women’s Studies Vol. 10 #4 May 2009 6
Reasons for lack of intervention prior to IPH. From the analysis of these articles,
it became clear that there existed a general lack of intervention to protect these victims of
violence either prior to or during an IPH. Reasons for this lack of intervention were
interpreted as stigma, social silence and fear. Twenty-four of the 49 total articles
contained some allusion to stigma, either surrounding the crimes themselves (e.g., do not
occur in Botswana) or around the perpetrators and victims. Victims were the most heavily
stigmatized, and allusions to victim provocation (e.g., she terminated the relationship, she
spent too much money) were found in 12 of the 24 case specific articles. The theme of
social silence, or unwillingness to discuss or acknowledge the existence of IPH, was
found in 5 of the 49 total articles. Examples of social silence included the beliefs that
“Botswana culture…does not permit the type of killings that have been witnessed here”
(Owino, 2005) and that Botswana is a “nation of peace and tolerance” (Letlhogile, 2005).
Finally, 3 of the 49 total articles discussed fear as a reason why a person who either
witnessed or was told about the violence, or the victim herself, did not intervene. Persons
outside the violence discussed fear of repercussion as a reason for lack of intervention
(i.e., killer would come back and kill them if they intervened), whereas the victim’s fear
of repercussion resulted in withdrawal of cases reported to the police.
Objective 2: Media Framing of Intimate Partner Homicides
Blame Frame. One broad frame, referred to here as the blame frame, was
consistently used to convey the problem of IPH in Botswana. This frame, which served to
assign responsibility for the existence of these crimes, and in so doing diagnose the cause
of the problem, was used in 44 of the 49 total articles. Within this overarching frame,
several underlying sub-frames were found: a frame that blamed gender or gendered
structures9; a frame that blamed broader social structures; and a frame that blamed
outside factors without implicating the victim or perpetrator specifically. The number of
articles assigned to each sub-frame is listed in Table 2.
Table 2: Number of articles using the Blame Frame, by sub-frame and type
SUB-FRAME TYPE OF ARTICLE TOTAL
Case Specific Opinion /
General Information
Gender 13 10
23
Social 1 3
4
Other 1 1
2
Gender & Social 2 8 10
Gender & Other 1 1 2
Social & Other 0 0 0
Gender, Social & Other 0 3 3
TOTAL 18 26 44
9 For the purposes of this study, gender is defined as both a “culturally specific set of characteristics that
identifies the social behavior of women and men and the relationship between them…and the way [the
relationship] is socially constructed” (Status of Women Canada, 2005, p. 57) and a social institution that
“establishes patterns of expectations for individuals, orders the social processes of everyday life, [and] is
built into the major social organizations of society” (Lorber, 1994, p. 1).
Journal of International Women’s Studies Vol. 10 #4 May 2009 7
As stated previously, all of the cases involved female victims (one by suicide),
and all but one perpetrator was male; thus, victim and perpetrator blaming appeared to
represent blaming of females and males, respectively. As such, it was felt that the victim
and perpetrator blaming most appropriately fell within the gender frame. Of the three
sub-frames, the gender frame was used most often, and was found almost equally in both
sets of articles (Table 2).
Victim blaming occurred not only in the majority of articles using the gender
frame, but also in just over half of the total articles (n=25). Victims were blamed
explicitly for terminating the relationship, or ‘ditching’ their partners, and implicitly for
causing an argument or misunderstanding that led to the murder. In one case, the article
discussed an argument the victim had supposedly caused, as the perpetrator “had seen his
girlfriend talking to another man and this could have stirred the trouble” (BOPA, 2006
Nov 22), and in another, the victim’s “uncontrolled expenses” had led to a “running
misunderstanding with her boyfriend” (Ditshipi, 2006). In a case involving arson of the
victim’s home, where the victim survived but five other family members were killed, the
victim was blamed by her relatives for having a relationship with a “criminal” (Piet,
2005). These articles also tended to exonerate the perpetrator, by blaming victim
provocation for the perpetrator’s actions. As stated by one article, “[the perpetrators]
often make the decisions because they fail to handle rejection. Sometimes it is not all
about the fear of rejection but they wonder how they would cope after being left by their
lover” (Setsiba, 2005).
Several articles using the gender frame did blame the perpetrator, either in
combination with the victim, or on his/her own. However, articles that seemingly blamed
the perpetrator generally spent more time discussing why perpetrators committed these
crimes (e.g., due to anger management issues, impulse control, emotional stress) then
what he or she did. In this way, articles that ‘blamed’ the perpetrator represented another
way to excuse the perpetrator for his or her actions. Articles using this frame also blamed
gendered social institutions, such as the family unit (e.g., not counseling children well,
encouraging children to live together prior to marriage) and traditional culture, as well as
power relations between the perpetrator and victim.
Articles using the social structure frame (n=4) primarily blamed civil society as a
whole for the existence of this violence; for example, for not “tak[ing] the issue of
passion killings seriously” and for not coming up with strategies to “eliminate all these
gruesome killings” (Modiega, 2005; BOPA, 2006 Nov 21). The police and justice
systems were also blamed for the continuation of this violence, primarily due to lack of
protection of victims. Two of the articles discussed cases in which either the victim or her
family had tried to protect themselves by reporting the perpetrator to the police; however,
in both cases, the perpetrator was released and the victim killed shortly thereafter. In
these cases, the police were blamed for not preventing the violence. In addition to the
police and justice system, other social structures blamed for both the violence and the
lack of protection of victims included the government, the West, economics and
development, and the media.
Finally, several articles used a frame that blamed outside factors, and not any
particular person, group, institution or system, for the occurrence of IPH (Table 2). These
factors included witchcraft, alcohol and mental illness. Of these, alcohol was blamed
most often.
Journal of International Women’s Studies Vol. 10 #4 May 2009 8
Upon analysis of the different frames used by the articles, it was noted that
several articles used a combination of the gender, social or outside sub-frames when
assigning blame. Primarily, both gendered and social structures were blamed (e.g., the
victim and the police and justice system) (Table 2). In one article, perpetrators were
blamed for letting their anger reach the “boiling point,” while the media were blamed for
portraying the “perpetrators of violence as heroes” (BOPA, 2006 Nov 21). Overall
however, the gender frame alone was used in the overwhelming majority of articles
included in this study.
Police Frame. The five articles not using the blame frame used what Bullock and
Cubert (2002) refer to as the “police frame” (p. 490). As they describe it, the police frame
conveys a “just the facts approach” (p. 490), where there is “no speculation about the
nature of the relationship between the perpetrator and the victim and no indication that
the killing or its attempt was part of a larger problem between those involved” (p. 490).
This frame was only used in articles discussing specific cases of IPH, and blame was not
placed on either party.
Stigma Frame. Finally, a stigma frame, that is, the perception that this violence
only happens to certain people, was evident in several of the articles (n=2). Rather than
explicitly specifying what type of people would be expected to experience this violence,
however, the articles using this frame presented two homicides as shocking or surprising
because of who was involved. In one, the perpetrator was a respected Botswana
Democratic Party (BDP) councilor, described as a “humble person who was slow to
anger,” and the victim his wife (Setsiba, 2007). In the other, the victim was portrayed as a
promising young university student who was engaged to her perpetrator. By their focus
on the shock that these women could be murdered, the articles suggested that IPH is
unexpected among those who are successful, or those in a committed relationship.
Objective 3: Passion Killings
In the case specific articles (n=24), the phrase passion killing(s) was not used in
any of the current articles, but was used in 4 of the historical articles. In these 4 articles
(number of text units=98), the phrase passion killing(s) was found in 11.2% of text units
(n=11). This finding is important as neither the current nor historical case specific search
criteria required that the case be described as a passion killing, so any use (or non-use) of
this phrase in these articles represents a choice by the author, and not a result biased by
search methods. The percentage of text units containing the phrase passion killing(s) in
opinion or general information articles was not determined, as, to be included in the
study, these articles had to contain either this or a similar phrase (e.g., love-related
killings).
Common Beliefs. Common beliefs about the nature of passion killings were
identified by analyzing the 25 opinion/general information articles. In these articles,
passion killings were often associated with love, and were referred to as “love-inspired”
(Owino, 2005), “love-related” (Letlhogile, 2006), “love killings” (Dube, 2003), and as
“love turning sour” (BOPA, 2006 Oct 9). Only one article discussed the discord of the
association of these crimes with love, and with characterizing them as a crime of passion
(Dingake, 2006). Despite the strong association of these crimes with love, they were also
described negatively throughout the articles, using terms such as brutal, femicide,
Journal of International Women’s Studies Vol. 10 #4 May 2009 9
disturbing, inhumane, cynical, barbaric, horrific, gruesome, cruel, ghastly, sadistic,
outrageous, fiendish, painful and abominable.
Eighteen of the 25 opinion/general information articles acknowledged the
increasing prevalence or occurrence of passion killings in Botswana. Two of the 25
articles compared the severity of passion killings to the HIV/AIDS pandemic that is
currently affecting Botswana. Though statistics were not available from any national
agency, statistics about the number of passion killings per year were reported in news
articles (via a member of the local police and a member of the Batswana government’s
Men’s Sector) and in a recently published academic paper (Alao, 2006) (Table 3).
Finally, all articles that provided a definition for a passion killing characterized the crime
as being perpetrated by a male, where the victim was almost always female, and where
the killer was likely to commit suicide following the crime.
Table 3: Number of Reported Passion Killings (January 2003 – Nov 2006)
Year Number of Reported Murders
Total Women Men
2003 54 46 8
2004 56 54 2
2005 85 74 11
2006 (Jan 1-Nov 19) 53 51 2
TOTAL (Jan 2003- Nov 2006) 248 225 23
Notes: Number of reported murders are from Alao (2006), Chwaane (2006), BOPA (2005
Feb 25) and BOPA (2006 Nov 28). All articles reported numbers of “passion killings,”
except BOPA (2006 Nov 28), which reported the number of “relationship murder cases.”
Discussion
The analysis of the selected articles provided a consistent pattern of IPH in
Botswana, and aided in clarifying the label ‘passion killing.’ In this study, no
characteristic was found that distinguished a passion killing from an IPH; therefore, this
label appears to represent a particular social construction.10 This is discussed in further
detail below.
In terms of the characteristics of IPH in Botswana, the analysis of case specific
articles suggested that they generally occur between current, or recently separated, non-
married intimate partners, and are perpetrated by the male. Many of the partners were
listed as living together at the time of the killing, but further study would be required to
determine if this is a significant correlation. The finding that most IPH occur between
non-married partners was also interesting, but again, cannot be determined as significant
here as it is possible that a husband killing a wife (or vice versa) is less likely to gain
media attention due to stigma. In the future, police records of all reported crimes could be
looked at to determine if either of these findings is significant. Contrary to the common
and strongly held Batswana social belief that perpetrators of IPH commit suicide
following the murder, the majority of perpetrators in the case specific articles were not
reported to have committed suicide. Due to the social stigma surrounding suicide
10 Since the term ‘passion killing’ was found to be a unique social, and not definitional, label for IPH, the
remainder of the discussion will use the term IPH to describe the killing of an intimate partner, unless the
label ‘passion killing’ is specifically being explored.
Journal of International Women’s Studies Vol. 10 #4 May 2009 10
(Jenkins, 2002), it is possible that cases involving a suicide are less likely to be reported,
or that media reports would mask this trend within their reporting of cases. This
represents an important topic for future study, as the occurrence of perpetrator suicide
may be a defining characteristic of a ‘passion killing’ and a reason for the use of this
label, while also a characteristic that may keep passion killings out of public documents.
Finally, no articles discussed a previous history of physical violence between the two
partners prior to the IPH, an exclusion that likely serves to represent these cases as
isolated incidents of ‘passion.’ As discussed by Bullock and Cubert (2002), making
incidents isolated removes association of these crimes with a larger social problem.
Based on statistics found in this study, one to two killings occur each week;
however, the majority of these cases are not reported in the media. The under-reporting of
IPH in Batswana newspapers is significant. Due to the lack of information typically given
about perpetrators and victims in the articles, it cannot be validly discerned why some
cases are reported, while others are not; however, based on common themes found in the
analysis, it is likely that the overall lack of reporting revolves around stigma, patriarchy
and social silence. Stigma is entrenched in social myths and stereotypes, including the
belief that IPV is the fault of the victim (Meyers, 1997). Because myths and stereotypes
blame the victim for her death, they in turn make IPH less newsworthy, and hence less
likely to be reported (Maxwell et al., 2000). In addition to victim blaming, the patriarchal
gender role system found in Botswana likely represents a key reason why these crimes
are so rarely covered. Globally, the majority of victims of IPH are women, and the
perpetrators men (World Health Organization, 2005); as was shown by cases of IPH in
the articles analyzed, this trend is no different in Botswana. Since, in traditional
patriarchal Botswana culture, men are the dominant sex, violence against women may not
be considered especially newsworthy (Denbow & Thebe, 2006). Indeed, two of the most
heavily reported cases involved violence against a man. In one, a mob beating of the
perpetrator occurred following the killing of his pregnant ex-girlfriend (Baputaki, 2007
Jan 11; Baputaki, 2007 Jan 18; BOPA, 2007 Jan 18), and in the other, a female and male
had killed their former partners (BOPA, 2005 Jan 13; BOPA, 2005 Feb 1; Tutwane,
2005). Finally, social silence is likely implicated in the lack of reporting. Several of the
articles discussed portraying Botswana as “a nation that other nations would envy”
(BOPA, 2005 Oct 26) and one that is “healthy and responsible” (BOPA, 2006 Dec 1). If
IPH were widely publicized, this would disrupt the desire to portray Botswana as a nation
of “peace and tolerance” (Letlhogile, 2006). These possibilities for the under-reporting of
IPH deserve the attention of future research, as reasons for lack of reporting may also be
implicated in lack of protection of victims.
As shown in the results of the article analysis, victims, who are primarily female,
are most commonly blamed for this violence. This finding was not surprising, and is
corroborated by previous studies (Berns, 1999; Bullock and Cubert, 2002; Consalvo,
1998). However, the emphasis on gendered blaming further affirmed the key role of
gender in understanding these killings. Several articles cited changing gender roles,
stemming from an increased emphasis on gender equality, a shift from patriarchal beliefs,
and increased socio-economic freedom and opportunities for women, as reason for the
increase in IPH in Botswana. Within this context, Denbow and Thebe (2006) discuss how
Batswana men still consider themselves the head of the household, and “remain
uncomfortable if their wives have more education, earn more money, or hold more
Journal of International Women’s Studies Vol. 10 #4 May 2009 11
prominent positions in the workforce than they do” (p. 153). From this theory, it can be
hypothesized that changing gender roles, and women becoming more central in society,
may leave men feeling that their authority is threatened; in response to this threat,
violence is used to regain power and control. While this idea represents an overly
simplified explanation for killings that are the result of complex and interacting issues, it
should nonetheless be explored in future research as a key component of IPH in
Botswana.
Finally, this study attempted to understand the unique label ‘passion killings.’ In
general, there appears to be a persistent association of these crimes with love. By
constructing these crimes as love-related, the media not only remove culpability from the
perpetrator, as the crime is committed out of his great passion for the victim, but also may
affect how the public views their role in the prevention of these crimes. If these crimes
are the result of passion and love, positive and private emotions, what place does the
public have to intervene? Furthermore, the media present these crimes as occurring with
no warning signs of prior violence, further constructing passion killings as a problem
outside the reach of public intervention because there is no signal of need or opportunity.
While the media do use negative adjectives to describe these crimes, their underlying
connection with love and passion underwhelm these descriptors. Since the use of the
label ‘passion killing’ by the media provides a particularly detrimental construction of
IPH to the public, it is of utmost importance that these killings are reframed as crimes of
violence and control.
Given the media’s construction of IPH as a love-based crime, the lack of use of
the term ‘passion killing’ when describing specific cases is noteworthy. Despite the fact
that cases of IPH as described in current case specific articles fit the definition of a
‘passion killing’ (as presented in opinion/general information articles and given by Alao
(2006)), none of these articles used this term. A discord exists, then, between willingness
to discuss passion killings and willingness to apply this label to actual murders; the same
newspapers that wrote opinion and general information articles about passion killings
never labeled specific IPH cases as such. Why this discord exists, and what it represents,
remains to be explored. As several historical case-specific articles did use the term
passion killing to describe cases of IPH, it may be that a shift in the use of this term has
occurred; however, more historical case articles would need to be examined before any
definitive conclusions can be drawn.
The results of this study must be considered in light of its limitations. Firstly, the
inability to access Batswana newspapers that were not electronic, or that did not have
electronic archives, may have led to a biased sample. Several newspapers, inaccessible at
the time of the study, now have electronic archives (The Voice, The Ngami Times and The
Sunday Standard), and so should be included in future studies. Secondly, because a fairly
short, convenience sampling timeframe was used, important articles may have been
missed. However, due to the iterative nature of the search, as well as the inclusion of
historical articles, it is felt that the sample was robust for that particular time period, and
that the sampling frame did not negatively impact the results.
In spite of the aforementioned limitations, these finding have implications for
both future studies and prevention programming. Firstly, further investigation into social
beliefs surrounding this violence is warranted, and, if gender is found to be central as
predicted by this study, it must be incorporated into programming. Specifically,
Journal of International Women’s Studies Vol. 10 #4 May 2009 12
programming focused on gender roles, both traditional and changing, should be
considered potentially important to the prevention of this violence. Secondly, the
construction of IPH as a love-based crime, and specifically the use of the label ‘passion
killings,’ must be halted, so that a proper construction of these crimes, specifically one
that presents them as the violent acts they are, is presented to the public.
Acknowledgements
The authors would like to thank Josephine Mazonde who provided insight into the
culture of Botswana and the current situation for women experiencing intimate partner
violence in this country. We would also extend our acknowledgement to Kathy Dirk for
her invaluable assistance in the preparation of this manuscript.
References
Alao, A. (2006). Lack of Mutual Respect in Relationship: The Endangered Partner.
Annals of the New York Academy of Sciences, 1087(1), 311-319.
Altheide, D.L. (1987). Ethnographic Content Analysis. Qualitative Sociology, 10(1), 65-
77.
Baffour-Awuah, M.E. (2004). To be legal or not to be legal: Illegality of homosexuality
in Botswana and the spread of HIV in prisons and the general community
[Abstract]. International Conference on AIDS. Jul 11-16; 15: abstract no.
MoPeE4211. Retrieved April 11, 2007 from
http://gateway.nlm.nih.gov/MeetingAbstracts/102280492.html
Baputaki, C. (2007, January 11). Passion killer gets mob justice. Mmegi, 24(5), online.
Retrieved January 12, 2007 from
http://www.mmegi.bw/2007/January/Thursday11/6988966101700.html
Baputaki, C. (2007, January 18). Police deny receiving death threat report. Mmegi, 24(9),
online. Retrieved January 19, 2007 from
http://www.mmegi.bw/2007/January/Thursday18/489798460755.html
Berns, N. (1999). “My Problem and How I Solved It”: Domestic Violence in Women’s
Magazines. The Sociological Quarterly, 40(1), 85-108.
Berns, N. (2001). Degendering the Problem and Gendering the Blame: Political
Discourse on Women and Violence. Gender and Society, 15(2), 262-281.
Becker, H. (2003). The Least Sexist Society? Perspectives on Gender, Change and
Violence among southern African San. Journal of Southern African Studies,
29(1), 5-23.
Botswana Press Agency. (2005, January 13). Murderer appeals 50-year sentence. Daily
News, online. Retrieved September 15, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20050113
Botswana Press Agency. (2005, February 1). Court reduces sentence. Daily News, online.
Retrieved September 15, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20050201
Botswana Press Agency (2005, February 25). Passion killings escalate. Daily News,
online. Retrieved September 15, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20050225
Journal of International Women’s Studies Vol. 10 #4 May 2009 13
Botswana Press Agency (2005, October 26). Youth urged to concentrate on nation
building. Daily News, online. Retrieved September 15, 2006 from
http://www.gov.bw/cgi-bin/news.cgi?d=20051026
Botswana Press Agency (2006, October 9). Malikongwa celebrates love. Daily News,
online. Retrieved October 11, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20061009
Botswana Press Agency (2006, November 21). Inmates learn to cope. Daily News,
online. Retrieved November 22, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20061121
Botswana Press Agency (2006, November 22). Moshupa man appears for murder. Daily
News, online. Retrieved November 23, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20061122
Botswana Press Agency (2006, November 28). Police committed to reduce relationship
violence. Daily News, online. Retrieved November 29, 2006 from
http://www.gov.bw/cgi-bin/news.cgi?d=20061128
Botswana Press Agency (2006, December 1). Traditional beer most dangerous. Daily
News, online. Retrieved December 4, 2006 from http://www.gov.bw/cgi-
bin/news.cgi?d=20061201
Botswana Press Agency. (2007, January 18). Suspect released from hospital. Daily News,
online. Retrieved January 19, 2007 from http://www.gov.bw/cgi-
bin/news.cgi?d=20070118
Bullock, C.F. & Cubert, J. (2002). Coverage of Domestic Violence Fatalities by
Newspapers in Washington State. Journal of Interpersonal Violence, 17(5), 475-
499.
Chwaane, T. (2006, November 2). Men’s Sector plan anti-passion killing meet. Mmegi,
23(165), online. Retrieved November 3, 2006 from
http://www.mmegi.bw/2006/November/Thursday2/10092951911028.html
Consalvo, M. (1998). “3 Shot Dead in Courthouse”: Examining News Coverage of
Domestic Violence and Mail-order Brides. Women’s Studies in Communication,
21(2), 188-211.
Denbow, J.R. & Thebe, P.C. (2006). Culture and customs of Botswana. Westport, CT:
Greenwood Press.
Dingake, M. (2006, January 17). Public Inquiry on Passion Killings. Mmegi, 23(6),
online. Retrieved September 15, 2006 from
http://www.mmegi.bw/2006/January/Tuesday17/656904332246.html
Ditshipi, T. (2006, December 6-12). Tonota woman sets herself on fire. Midweek Sun,
online. Retrieved December 8, 2006 from http://www.midweeksun.co.bw
Dube, M. (2003). Women’s organisations tackle gender-based violence. News From
Africa, online. Retrieved September 15, 2006 from
http://www.newsfromafrica.org/newsfromafrica/articles/art_2331.html
Entman, R.M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of
Communication, 43(4), 51-58.
Green, D. (1999). Gender Violence in Africa: African Women’s Responses. New York,
NY: St Martin’s Press.
Hubbard, J.C., DeFleur, M.L. & DeFleur, L.B. (1975). Mass Media Influences on Public
Conceptions of Social Problems. Social Problems, 23(1), 22-34.
Journal of International Women’s Studies Vol. 10 #4 May 2009 14
Jenkins, R. (2002). Addressing suicide as a public-health problem. The Lancet,
359(9309), 813-814.
Letlhogile, L. (2006, February 6). Botswana Floored by Passion Killings. BBC News
(Gaborone), online. Retrieved September 15, 2006 from
http://news.bbc.co.uk/2/hi/africa/4685216.stm
Lorber, J. (1994). Paradoxes of Gender. New Haven, CT: Yale University.
Loseke, D.R. (1999). Thinking about social problems: An Introduction to Constructionist
Perspectives. Hawthorne, NY: Walter de Gruyter, Inc.
Macdonald, D.S. (1996). Notes on the Socio-economic and Cultural Factors Influencing
the Transmission of HIV in Botswana. Social Science and Medicine, 42(9), 1325-
1333.
Maundeni, T. (2002). Wife Abuse Among a Sample of Divorced Women in Botswana: A
Research Note. Violence Against Women, 8(2), 257-274.
Maxwell, K.A., Huxford, J., Borum, C., & Hornik, R. (2000). Covering domestic
violence: How the O.J. Simpson case shaped reporting of domestic violence in the
news media. Journalism and Mass Communication Quarterly, 77(2), 258-272.
McCall, G.J. & Resick, P.A. (2003). A Pilot Study of PTSD Symptoms Among Kalahari
Bushmen. Journal of Traumatic Stress, 16(5), 445-450.
Meyers, M. (1994). News of Battering. Journal of Communication, 44(2), 47-63.
Meyers, M. (1997). News Coverage of Violence Against Women: Engendering Blame.
Newbury Park, CA: SAGE Publications.
Modiega, D. (2005, June 20). BCC Condemns Passion Killings. Mmegi, 22(92), online.
Retrieved September 15, 2006 from
http://www.mmegi.bw/2005/June/Monday20/876403434662.html
Owino, W. (2005, June 20). The Problem Is Not Up-bringing. Mmegi, 22(92), online.
Retrieved September 15, 2006 from
http://www.mmegi.bw/2005/June/Monday20/876403434662.html
Pan, Z. & Kosicki, G. (1993). Framing analysis: An approach to news discourse. Political
Communication, 10, 55-75.
Phaladze, N. & Tlou, S. (2006). Gender and HIV/AIDS in Botswana: a focus on
inequalities and discrimination. Gender and Development, 14(1), 23-35.
Piet, B. (2005, December 12). The Vow Of A Passion Killing Survivor. Mmegi, 48(5),
online. Retrieved September 15, 2006 from
http://www.mmegi.bw/2005/December/Monday12/7431851611870.html
Rani, M., Bonu, S., & Diop-Sidibe, N. (2004). An Empirical Investigation of Attitudes
towards Wife-Beating among Men and Women in Seven Sub-Saharan African
Countries. African Journal of Reproductive Health, 8(3), 116-136.
Sacco, V.F. (1995). Media Constructions of Crime. Annals of the American Academy of
Political and Social Science, 539, 141-154.
Setsiba, T. (2005, June 12). Movement formed to combat passion killings. Mmegi,
22(88), online. Retrieved September 15, 2006 from
http://www.mmegi.bw/2005/June/Friday10/598087027319.html
Setsiba, T. (2007, January 4). BDP cllr kills wife, commits suicide. Mmegi, (24)1, online.
Retrieved January 5, 2007 from
http://www.mmegi.bw/2007/January/Thursday4/502380764744.html
Journal of International Women’s Studies Vol. 10 #4 May 2009 15
Journal of International Women’s Studies Vol. 10 #4 May 2009 16
Status of Women Canada. (2005). Report of the Standing Committee on Status of Women.
Retrieved December 3, 2006 from http://www.parl.gc.ca
Terkildsne, N. & Schnell, F. (1997). How Media Frames Public Opinion: An Analysis of
the Women’s Movement. Political Research Quarterly, 50(4), 879-900.
Tutwane, L. (2005, January 12). Court expresses concern over lawyers’ failure to file
arguments. Mmegi, 22(5), online. Retrieved September 15, 2006 from
http://www.mmegi.bw/2005/January/Wednesday12/6827436001537.html
Wilson, M. & Daly, M. (1993). An Evolutionary Psychological Perspective on Male
Sexual Proprietariness and Violence Against Wives. Violence and Victims, 8,
271-294.
World Health Organization. (2005). Multi-Country Study on Women’s Health and
Domestic Violence Against Women. Retrieved March 2, 2007 from
http://www.who.int/gender/violence/who_multicountry_study/en/index.html
