OUP user menu

The epidemiological patterns of honour killing of women in Pakistan

Muazzam Nasrullah, Sobia Haqqi, Kristin J. Cummings
DOI: http://dx.doi.org/10.1093/eurpub/ckp021 ckp021 First published online: 13 March 2009


Background: Honour killing (HK) is a problem of public health concern but published data on the phenomenon are limited and many cases likely go unrecognized. Our study focuses on the epidemiological patterns of HK of women in Pakistan, where domestic violence is common and HK occurs but is poorly described. Methods: Human Rights Commission of Pakistan (HRCP) systematically collected data on HK of women using newspaper reports from January 2004 till December 2007. We analysed the aggregated data on HK through December 2007 and estimated the rates of HK. Results: A total of 1957 HK events occurred from 2004 to 2007; complete data was not available for all variables. Adults (≥18 years) constituted 82% (803/978) of death toll with 88% (1257/1435) being married. Alleged extramarital relation was the major reason for the killing (92%; 1759/1902). Husbands (43%; 749/1739), brothers (24%; 421/1739) and ‘other’ close relatives (12%; 200/1739) were the perpetrators in known HK events. Among the weapons/methods used for killing, firearms (61%; 1071/1768), stabbing (4%; 65/1768), use of axe (12%; 220/1768), edged tool (8%;136/1768) and strangulation (9%; 167/1768) were the main means of execution. The mean annual rate of HK in females (age 15–64 years) was found to be 15.0 per million. Conclusions: Newspaper reports are good source of surveillance when information is limited. We found that adult married women constituted the majority of victims of HK. Ongoing surveillance would serve to better characterize HK in Pakistan and assess the effectiveness of preventive strategies.

  • cultural practices
  • gender
  • honour killing
  • Karo Kari
  • Pakistan
  • violence


Violence against women is a problem of public health concern. In 48 population-based surveys from around the world, 10–69% of women reported being physically assaulted by an intimate male partner at some point in their lives.1 A multi-country study by World Health Organization (WHO) on women's health and domestic violence showed that the proportion of ever-partnered women who had ever experienced physical or sexual violence, or both, by an intimate partner in their lifetime, ranged from 15% to 71%, with most sites falling between 29% and 62%.2 Women in Japan were the least likely to have ever experienced physical or sexual violence, or both, by an intimate partner, while the greatest amount of violence was reported by women living in provincial (for the most part rural) settings in Bangladesh, Ethiopia, Peru and the United Republic of Tanzania.2

Gender-based violence, only recently emerging as a pervasive global issue, contributes significantly to preventable morbidity and mortality for women across diverse cultures. Existing documentation suggests that profound physical and psychological sequelae result from intimate partner violence. The clinical manifestations of domestic violence are often culture-specific. Depression, stress-related syndromes, chemical dependency and substance abuse and suicide are consequences observed in the context of violence in women's lives.3 Physical abuse contributes significantly to the victim's lifetime risk of diagnoses including major depression, dysthymia, conduct disorder and drug abuse or dependence.4 In addition to the impact on abused women themselves, depression and behavioural problems have also been repeatedly reported in children whose parents were experiencing violent marital discord.5

A United Nations study6 found that 50% of married women in Pakistan are physically battered and 90% are emotionally and verbally abused by their husbands. A study by the Pakistan National Women's Division on ‘Battered Housewives in Pakistan’7 confirmed these results, estimating that domestic violence takes place in ∼80% of households. Domestic violence is not only prevalent in rural areas of Pakistan but also in developed cities like Karachi. A study in Karachi showed that 34% of the interviewed women when asked about domestic violence reported physical abuse.8 Human Rights Commission of Pakistan (HRCP) estimated the prevalence of domestic physical violence in Pakistan as 65% (physical violence), and almost one-third (30.4%) of those reported sexual violence; both forms of violence lead to serious injuries requiring emergency medical attention.9 A study in neighbouring Bangladesh found that young age (20–29 years), illiteracy and poverty increased a married women's risk of being sexually abused.10

Honour killing (HK) is a form of domestic violence that has been described as a custom in which mostly women and sometimes men are murdered after accusations of sexual infidelity.11 The killers, as reported, seek to avenge the shame that victims are accused of bringing to their families.12

HK is known by different names depending on the region in Pakistan in which it is practiced. In Sindh province it is referred to as Karo Kari, where Karo refers to the ‘blackened’ or dishonoured man and Kari to the ‘blackened’ woman.13

The extent and nature of HK in Pakistan have been difficult to estimate as information is reported through media but not systematically collected by any health agency. However, clear knowledge about the extent and the brutal consequences of HK may serve to alter traditional practices. Our study focuses on the epidemiological patterns of HK of women using data systematically collected by HRCP through newspaper reports in Pakistan.

Methods and materials

Human Rights Commission of Pakistan

HRCP is an independent, voluntary, non-profit organization which is not associated or affiliated with the government or any political party. It is committed to act with impartiality and objectivity in all matters. Among its main aims are spreading awareness of human rights among the people, mobilizing public opinion, collecting information and disseminating knowledge about human rights abuses.

Data sources and data collection

The National daily newspapers DAWN, NEWS, NATION and JANG, NAWA-E-WAQAT, KHABRAIN are Pakistan's largest English and Urdu language broadsheets, respectively. These newspapers regularly report the occurrence of HK from different parts of the country. The reports vary in their details from very brief, giving the gender, age and method used, to extensive details of the circumstances of the act. To standardize all the available information of HK in the newspapers, a ‘Report Form’ was developed by the HRCP. The report form abstracted the victim's age group, reason for the killing, occurrences before the event, relationship of the victim and the perpetrator, victim's social status and marital status, method/weapon used for killing and medical aid provided to the victim after the event.

Local HRCP sub-centres in all over country reviewed the above mentioned newspapers along with the local newspapers in their local languages for HK reports, checked for duplication, completed the standard ‘Report Form’ and mailed it to the HRCP head office in Lahore, a capital city of Punjab province. In the head office, staff reviewed the reports for quality. The data manager entered these reports into the final dataset that was used for analysis. See figure 1.

Figure 1

Flow diagram of data collection by HRCP.

Data management

The HRCP provided aggregated data on HK events occurring from 1 January 2004 till 31 December 2007. We present descriptive statistics using as the denominator the number of events with available information for a particular variable. Thus the denominators vary on the basis of completeness of reporting. We used the variable Karo Kari (a term specific to Sindh province) to make estimates of provincial level HK for Sindh province. An ‘event’ is defined as any happening that caused the death of one or more females in the name of honour. The information was compiled using Microsoft Excel.

We estimated the rates of honour killing for each year using estimates of Pakistan's female population age 15–64 years for each year from 2004 to 2007.14 This age group was selected because it was considered vulnerable for the HK. Total number of HK for each year was divided by the respective estimated female populations to determine the rates of HK per year for three provinces (Punjab, Balochistan, North-West Frontier province) of the country.

Sindh is the second most populous province of Pakistan with a population of 43 million, or ∼26% of the total population of Pakistan.15 We estimated Sindh province's female population of age 15–64 years for each year from 2004 to 2007 by taking 26% of the respective national population figures.


A total of 1957 events of HK were recorded during the period of 4 years. Age was available for 978 events (50% of 1957): 803 of these (82%) were adults (≥18 years), 175 (18%) were minors (<18 years). Minorities were victims of HK in 14 instances (13 Christians, 1 Hindu). Most of the HK events (92%, n = 1759/1902) occurred because of alleged extramarital relations and 116/1902 (6%) events because women married by their own choice. The distribution of known events by perpetrators and different methods used in HK can be seen in Table 1.

Marital status was available in 1435 events (73% of 1957): 1257 of these (88%) were married, 131 (9%) unmarried, 20 (1%) widowed and 27 (2%) divorced. Accused were arrested in 39% (516/1316) of cases where 32% (641/1957) were not known. Whether medical attention was given to the victims was unknown in 98% (n = 1927) of events respectively.

View this table:
Table 1

The Distribution of perpetrators and weapon used for Honour Killing in Pakistan, 2004–07

PerpetratorsN (%)Method used for honour killingN (%)
Husbands749 (38)Firearms1071 (55)
Brothers421 (22)Axe220 (11)
Close relativesa200 (10)Strangulation167 (9)
In-lawsb132 (7)Edged tool136 (7)
Fathers121 (6)Stabbing65 (3)
Sons50 (3)Stick24 (1)
Local residents27 (1)Hanging20 (1
Neighbours9 (0.5)Fire16 (0.8)
Employers3 (0.2)Poisoning14 (0.7)
Othersc27 (1)Torture13 (0.7)
Not known218 (11)Electrocution5 (0.3)
Othersc17 (0.9)
Not known189 (10)
  • a: Close relatives: Include cousins, maternal and paternal uncles, grandparents and nephews.

  • b: In laws: Include father-, mother-, brother-, sister- and son-in-law.

  • c: Others: Include cases which do not fall into any of the category ‘Perpetrators’ and ‘Method used for Honour Killing’.

A total of 1032 (53%) HK events occur in Sindh province. Use of firearms was more common in HK in Sindh province (72%; n = 665/924) than in HK in the rest of the country (48%; n = 406/844). The proportions of HK in which an axe was used were approximately similar for Sindh province and the rest of the country (14%; n = 127/924 vs. 11%; n = 93/844) respectively.


The rates of HK in females (age 15–64 years) for years of 2004–07 in three provinces were found to be 9.4, 5.3, 7.9 and 7.7 per million, respectively. On the other hand, the rates of HK in specific to Sindh province (Karo Kari) in females for same age group were calculated to be 24.6, 9.7, 27.5 and 28.0 per million, respectively. See figure 2 for comparison.

Figure 2

Comparison of rate of honour killings in Pakistan: Sindh province vs. all other provinces 2004–07.


To our knowledge, this study is the first to document the epidemiology of HK in Pakistan. We found that the total of 1957 events of HK occurred during four years with an average annual rate of 15.0 per million. The majority of known HK events were of adults and occurred because of alleged extramarital relations, with the major perpetrators being husbands.

When we look at the methods used for the crimes, use of fire-arms had been reported in the majority of the cases. This implies a serious issue of easy access to firearms by the public. Thus regulation of firearms access might be one way to reduce these crimes of HK. While overall, males are more commonly killed by firearms than females in Pakistan,16>,17 there is evidence by autopsy findings that females are predominately killed by firearms in certain parts of country,18 such as in the Northern part of Pakistan where carrying firearms is more of a culture. It was expected that the majority of the HK would have been caused by firearms in this area, but the fact that firearms were used in a greater proportion of killings in Sindh province, located in the south, than elsewhere was surprising.

In most of the cases of HK, the perpetrator was closely related to the victim. The most common relationship was that of husband, followed by brother. All over the world, women are most often killed by their husbands, boyfriends and ex-husbands and ex-boyfriends, however what is different for HK's is that brothers are often involved in the killings.

Men tend to victimize women whom they know, who are often female family members. Society tends to blame the victim, even when she is a child as is reported in a South African study.19 Similar trends have also been reported in Jordan, where a review of all court files of women murdered during 1995 found 38 such cases (out of a total of 89) in which a male relative of the female victim, primarily the brother, committed the murder.20 This is in stark contrast to the situation for men, who in general are much more likely to be attacked by a stranger or acquaintance than by someone within their close circle of relationships.21,22

HK is most prevalent among Muslim countries20 but our study showed that Christian and Hindu minorities were also victims, suggesting that the phenomenon is not limited to a single religion but rather reflects cultural norms. Further research on this topic may help to design effective preventive strategies.

The majority of the victims were married. One possible explanation might be the disturbingly high incidence of domestic violence towards married women in Pakistan.8 There is evidence where HK is associated with low level of education.23 This is beyond the scope of our study however more research is needed in this area to find the relation between HK, education level and socioeconomic status. Death certificates and medical examiner records are routinely used for mortality surveillance but may not provide sufficient information about prevention. Newspapers are an available, inexpensive source of potential information.24,25

While data are limited, one estimate of the overall homicide rate in Pakistan was 70 per million in 2000.26 Our result suggests that the honour killing of women constitutes at least 21% of all homicides (both male and female victims) in Pakistan. Given that the majority of Pakistan's homicide victims are likely to be male, HK represents a substantial proportion of female homicide in Pakistan.

Pakistan's Criminal Law (Amendment) Act 2004 provides women protection against ‘offences committed in name or on the pretext of honour’27 and its constitution enshrines the principle of equality before the law. Yet, according to the National Commission on the Status of Women (NCSW), the state is failing to punish those guilty of HK.28 Under Qisas and Diyat laws of Sharia offenses like honour crimes are compoundable (open to compromise as a private matter between two parties) by providing for Qisas (retribution) or Diyat (blood-money). The heirs of the victim can forgive the murderer in the name of God without receiving any compensation or Diyat (Section 309), or compromise after receiving Diyat (Section 310). This information is admissible in the criminal justice system and can lead to a perpetrator's being freed, despite the national criminal law.

Newspaper surveillance is useful to identify where HK are occurring most frequently, but they are likely to be underestimates of the true incidence. However, the number of incidents detected is large enough that the events detected may be a fairly representative sample. The information provided in many newspaper reports is certainly limited. The data set in many of the reports is incomplete. For example, there is no mention of the education level or social class of the subjects, nor of any psychiatric illness the perpetrator may have had in the past or was currently suffering from. Similarly, no distinction is made between urban and rural areas. This is important as 70% of the population of Pakistan lives in rural areas and it is anticipated that more HK take place in rural areas as of low literacy rate there. In addition, we did not have direct access to the victims and perpetrators. The death certificates can give information about victim and perpetrators but our study only looked at the newspapers reports. Follow up of the victims’ family with interviews may have proven useful, but it was impossible in our studies because of missing information. The denominator while calculating rates of HK constitute only females between ages 15–64 years. We might overestimate the figures by not counting the female population below and above 15 and 64 years of age respectively in our denominator. These limitations should, however, be seen in the context of the overall paucity of information on HK from Pakistan. Despite these shortcomings the study highlights important findings.

In summary, in countries where there is a dearth of official information regarding HK but where other sources of information, for example, newspaper reports, are available, these should be studied. Such reports, though of limited value, can still provide useful information on HK. The present study highlights the dire need for further urgent research and systematic data collection that might facilitate analysis for research on honour crimes in Pakistan. In particular, studies must address culture-specific factors in HK in Pakistan. This would help in identifying groups at risk as well as in formulating preventive strategies for this important public health problem, which remains largely neglected in Pakistan. In addition, there should be enforcement of the existing law on HK along with means of sentence to the perpetrators. Efforts should be made in raising public awareness against HK.


We greatly appreciate the support, guidance and data provided by the team of HRCP especially Saira Ansari, the HRCP Information Officer and Naveera Khan, the HRCP Database Officer. We would like to thank in particular Mr I. A. Rehman, country director HRCP without the support and encouragement of whom this manuscript was impossible.

Conflicts of interest: None declared.

Key points

  • In countries where there is a dearth of availability of official and authentic data regarding HK, newspaper reports could provide a useful source.

  • Such newspaper reports help in shedding light on the still underreported grievous act of HK.

  • This article highlights the dire and urgent need for further research, systematic data collection and preventive measures in this important area of public health.

  • Though law exists for punishment of culprits of HK, they need enforcement.


View Abstract