Research on a Socially , Ethically , and Legally Complex Phenomenon : Women Convicted of Filicide in Malaysia 1

Little is known about filicide from the perspective of women convicted of the offence. The lack of research is partly attributable to the many difficulties entailed in researching socially marginalised and incarcerated people. Research on filicide engages with socially, culturally, and politically sensitive matters, including gendered social structures and behaviours, legal and ethical complexity, emotionally arousing topics, a rare phenomenon, and hard‐to‐reach participants. In countries where there is poor surveillance, limited local information, and few resources or experts in filicide, researchers must find innovative ways of overcoming these problems. Here we describe the particular challenges in conducting research on women convicted of filicide in Malaysia, a predominantly Muslim country, when the researchers are based at an Australian university. The persistence, resilience, and creativity required to overcome each problem were justified by the achievement of research that contributes to knowledge and has implications for change in policy and practice.


Background to researching filicide in Malaysia
When parents kill their own children, it is known as filicide (Resnick 1969).Parents are expected to love and protect their children and those who violate this expectation are judged harshly by society.Judgement is most harsh when the killer is the child's mother; attempts may be made to understand how such an event could occur by characterising the mother as 'bad, sad, or mad' (Cavaglion 2008;Weare 2013).The current global discourse of filicide often emphasises women as perpetrators because there is evidence that more women than men have killed newborns (known as neonaticide) and children aged less than 12 months (infanticide) (Friedman, Horwitz and Resnick 2005;Porter and Gavin 2010).
The first author, a Malaysian woman who is a psychiatrist, became concerned about sensational reports in the Malaysian news media of women arrested for filicide; she wanted to find out what was happening and why.The Malaysian scholarly literature revealed no published research on baby abandonment or filicide in Malaysia or neighbouring countries.Nearly all relevant data came from English-speaking countries that have excellent economic, human, and gender development; the remainder came from China and India.All these countries had very different social and cultural backgrounds from Malaysia.Although Malaysia is classified as an uppermiddle-income country, its wealth is unevenly distributed, with evidence of economic and gendered inequality.Its economy ranks 68th in the world, measured by gross national per capita income for 2013, the most recent data available (World Bank 2014).The country is ranked 62nd on the Human Development Index and 42nd on the Gender Inequality Index (United Nations Development Programme 2015).
After being unable to find appropriate research supervision in Malaysia, the first author secured a competitive international research scholarship from the Government of Malaysia and was accepted as a candidate for the research degree of Doctor of Philosophy in an Australian university, supervised by the second and third authors.Both supervisors are women trained within the discipline of psychology.
Current understanding of filicide arises from research restricted by several limitations.The relative rarity of the phenomenon means that most research must perforce include small sample sizes, but there are also problems with inconsistent descriptions, lack of control groups where these might be appropriate, and varied scientific quality.Evidence is usually derived from secondary analyses of data rather than from engagement with the person identified as the perpetrator (Friedman, Horwitz and Resnick 2005;Porter and Gavin 2010).Filicide has been defined, discussed, classified and reclassified (Bourget and Bradford 1990;d'Orban 1979;Resnick 1969) but there are still major gaps in knowledge.We need evidence from low-and middle-income countries and from women charged with or convicted of filicide.Much that is said about these women is inferred from aggregate data; their voices have rarely been heard and it is uncommon for their opinions to be sought.
Our goal was to gain greater understanding of maternal filicide in Malaysia.To do so, we approached the phenomenon from three perspectives.The most broad perspective was at the national level, where we sought to establish rates of infanticide and characteristics of perpetrators and victims.For the next, more closely focused, perspective we sought to comprehend the experiences and opinions of professionals whose work is associated with filicide or women at risk.The most intimate research perspective focused on women convicted of filicide with the goal of understanding what the experience and its antecedents meant to them.Each perspective sought different kinds of knowledge and demanded a different research method.
To determine prevalence and trends over time and to describe the recorded characteristics of suspected perpetrators and victims, we undertook quantitative analysis of existing Malaysian national data.We requested data on infant abandonment and infanticide for the years 1999 to 2011 from the Criminal Records Unit and data on live births 1999 to 2011 from the Births, Deaths and Adoptions Unit.Both sets of data are confidential; access is severely restricted.Once permission had been obtained from the most senior authorities in each organisation, raw data from police records and national registries were analysed by assistants appointed by each department and only the summary data were made available to us.(Further details of the statistical data and analysis have been published in Razali et al. 2014.)We sought professionals who might have privileged insights into the position of at-risk girls and women in Malaysian society and who would be able to tell us about public policies and practices for prevention of maternal filicide.The first author undertook lengthy negotiations with senior managers of prison headquarters, police headquarters, psychiatric institutions, a public hospital, a university faculty of medicine, a baby hatch unit, and one government and one private agency providing shelters for at-risk women.Once permission had been granted, she contacted potential participants to explain the research.The first author also inquired through her professional networks and searched the Internet for professionals such as lawyers and teachers responsible for student discipline.Fifteen volunteers from diverse professions and places of work were recruited.They were interviewed (in person, by telephone, or by email) using a semi-structured, study-specific interview guide that sought opinions on what led women to commit or be convicted of filicide; effects of current policies, systems, and services; any barriers to optimal management of women who may have committed or be vulnerable to committing filicide; and whether these women have their needs met.At the end of the interview, key informants were also invited to comment on anything of importance to them about maternal filicide that had not arisen during the interview.Transcripts of interviews were analysed thematically (see Razali 2015).
It was particularly important to us to listen to women's own accounts of their lives and what they understood had led to their conviction and incarceration for filicide.It was most appropriate to use in-depth interviews.The first author sought and eventually achieved permission for the research from senior staff at the Prison Headquarters, responsible for the five prisons in Malaysia, and directors of the two psychiatric institutions for criminal patients.She then visited (when in Malaysia) or telephoned and emailed (from Australia) the seven institutions several times to persuade senior staff of the value of the research, ask them to appoint a reliable staff member to assist with recruitment, and train recruiters in ethical practice.After many weeks, 10 eligible women were identified and informed about the research.Nine consented to participate; one refused because she denied that she was guilty of filicide.The first author conducted the interviews in person.Following formalities and attempts to put women at ease, each was asked, 'Please talk about whatever you would like to tell me about your life.You may begin wherever you choose'.Gentle prompts encouraged her to describe her early life, the events surrounding the infant or child death, and her reflections on her circumstances.Women's accounts were translated into English, analysed thematically, and interpreted within narrative theory (Kirkman 2002) (see Razali 2015).
Each component of the research generated its own challenges.Researching filicide, especially in a country like Malaysia, involves examining socially, culturally and politically sensitive matters (which are legally and ethically complex); managing emotionally and psychologically delicate situations; exploring gender-related concerns; and investigating a rare phenomenon with hardto-reach research participants.There is also a need to overcome limited resources and logistical problems.By sharing our experiences, we hope that other researchers can benefit, especially in research on filicide and its links to women in resource-poor countries with a patriarchal social structure.In this endeavour we follow others who have described the research process (see, for example, Westmarland 2011), especially when qualitative methods have been used (see for example, Bartels and Richards 2011).

A socially, culturally, and politically sensitive phenomenon
Research about filicide is especially challenging because it involves multiple sensitivities.For example, researchers in China investigated filicide associated with its controversial one-child policy (Fuse and Crenshaw 2006).In India, researching filicide entailed exploration of cultural practices such as dowry: gifts the bride's family gives to the bridegroom and his family (Diamond-Smith, Luke and McGarvey 2008).The researcher must understand filicide in its social context, which can include the stigma of illegitimacy, as is the case in Indonesia (Spinelli 2005).
In researching this topic in Malaysia, we had to deal with matters that are especially sensitive to Muslims and traditional Malaysian society.Gender inequality (Othman 2006), family planning and abortion (Tong et al. 2012), sex and reproduction (Mutalip and Mohamed 2012), and mental illness (Mas and Hatim 2002), all of which may be associated with maternal filicide, are highly stigmatised taboos in this patriarchal society.These, especially gender inequality, can create unease among Malaysian people to the point of triggering political turmoil.Malaysia accepted the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW) on 5 July 1995 (United Nations Treaty Collection 2014).However, the requirement to accede to several articles in the convention, particularly those related to gender equality, has led to intense arguments between religious fundamentalists and the few liberal leaders.To maintain stability, it is argued, some prefer to remain silent (Aziz 2008).The Government of Malaysia has also set aside some articles promoting gender equality because they are not in keeping with Syariah Laws and the Federal Constitution of Malaysia (Saleh 2012).
In Malaysia, it is difficult to consider matters associated with children born out of wedlock, a potential risk for neonaticide, because of the association in the public consciousness with illicit sexual relationships.Following the media focus on filicide, Islamic fundamentalists have urged the enactment and implementation of Hudud Laws as part of current Syariah Laws (Zurairi 2014).Hudud Laws provide severe punishment for 'fornication and adultery'; a married Muslim found to have committed adultery is to be punished by stoning to death and an unmarried person so convicted is to be punished by flogging (a hundred lashes) (Mohamed and Ali 2012).Current Syariah Laws in Malaysia are not as extreme.For example, Section 23 (Sexual intercourse out of wedlock) of Act 559, Syariah Criminal Offences (Federal Territories) Act 1997, states that: (1) Any man who performs sexual intercourse with a woman who is not his lawful wife shall be guilty of an offence and shall on conviction be liable to a fine not exceeding five thousand ringgit or to imprisonment for a term not exceeding three years or to whipping not exceeding six strokes or to any combination thereof.
(2) Any woman who performs sexual intercourse with a man who is not her lawful husband shall be guilty of an offence and shall on conviction be liable to a fine not exceeding five thousand ringgit or to imprisonment for a term not exceeding three years or to whipping not exceeding six strokes or to any combination thereof.
While this research was being conducted (2012-2015), there was great social pressure, especially from religious fundamentalists, to implement Hudud Laws in one of the states in Malaysia.It created heated debates between Muslims and non-Muslims (Yunus 2015).
It has been suggested that researching sensitive matters might be threatening, or even damaging, to both the researcher and the researched (Johnson and Clarke 2003).Researching filicide in Malaysia required us, therefore, to understand, to be aware of, and to be sensitive to these social, political and legal contexts.We also had to anticipate that the results of our research might be controversial, perceived as flouting the social, cultural and political standards of Malaysian society, and take this into account in the production of all outputs from the research.

A legally and ethically complex phenomenon
The legal and ethical complexities of filicide meant that, as researchers, we had to be aware that the validity of voluntary participation and informed consent could be questioned; that we might be asked to breach confidentiality to satisfy or protect public welfare; that research data might be used by the justice authorities to trace offenders; and that there could be difficulty publishing the results should they be ethically or legally challenging (Gadd, Karstedt and Messner 2011).We endeavoured to pre-empt these possibilities, especially to protect vulnerable women and limit the potential for the research to be susceptible to legal challenge.The most significant decision was to invite as potential participants only women already convicted and sentenced, with no appeals pending, so that there were no legal reasons for requiring confidential information.
Further, the topic of filicide is extremely sensitive for people who are being investigated and stressful to the investigators.It concerns sexual relationships, reproduction, parenthood and (often violent) death, subjects that arouse powerful emotions, stigma, fear, controversy and social conflict (McCosker, Barnard and Gerber 2001).We considered these sensitivities at every stage of the research.The assessment of risk, both physical and psychological, was thoroughly canvassed.Potential physical threats to the first author during fieldwork and steps to handle such situations were discussed in detail among the research team and with the prison and hospital authorities and the ethics committees.The potential for criminal justice agencies to seek confidential information or for the first author to be subpoenaed as an expert psychiatric witness in a participant's trial would constitute a serious conflict of interest and might jeopardise the research.Clear guidelines and information about confidentiality were therefore developed by the research team and discussed with senior authorities in the institutions, the ethics committees, and the research participants.For example, recruiters nominated to approach eligible incarcerated women were given a script to ensure that women were not coerced into participation and that they understood what participation entailed.The script also made it clear that women could use real names or pseudonyms for themselves and anyone else they mentioned, and that strict confidentiality was assured.It was made clear to recruiters that they must not reveal to anyone who had consented to participation and who had not.Given that incarcerated women are under constant observation, the researchers planned how to minimise the likelihood of being overheard during interviews, including speaking in whispers and being aware of any observers and their proximity.
The psychological impact of the research on the researchers was also addressed in regular debriefing sessions within the research team and by consulting colleagues who are mental health professionals.These steps were successful in avoiding psychological distress and vicarious trauma from hearing about such troubling experiences that included not only the deaths of infants and children but persistent marginalisation and cruelty experienced by the women.
The process of seeking permission from the authorities to conduct each aspect of the research involved layers of formal and informal approval, from the state level (ministry and headquarters), through administering departments and, eventually, to ground level (the head or director of each police department, prison and hospital).The vital steps before starting research in these settings was to determine the feasibility of the study and to gain the confidence of the highest-level authorities that the research would yield more benefit than risk to participants, the authorities and society.Gaining full permission in Malaysia took more than a year.Maintaining communication with all parties, making appointments and arranging schedules of fieldwork with each police department, prison and hospital required meticulous attention to detail and to institutional protocols and sensitivities.Universiti Teknologi MARA, Malaysia; and Monash University Human Research Ethics Committee, Australia.Given the originality and sensitivity of the project, the researchers were required to attend Committee meetings to demonstrate that the rights and welfare of the participants and researchers were protected and that the research was scientifically rigorous and of public benefit.In Malaysia, the first author also had to sit for a formal examination on Malaysian Good Clinical Practice Guidelines, a compulsory hurdle for all researchers engaging with patients or hospital staff.

Emotionally and psychologically complex research
Those who research filicide can expect to encounter intense emotions and, possibly, psychological disturbance.This is especially true when the research uses qualitative methods, entailing close interactions the researcher and the research participants.Research on any emotionally sensitive matter requires researchers to deal with their own preconceptions, moral values and prejudices (Sollund 2008).Researchers should also consider their own susceptibility to distress and the adequacy of their interpersonal skills in dealing with other people's emotions (Seal 2012).
The first author's experience as a psychiatrist, supported by the psychological skills of the coauthors, contributed to our capacity to conduct the research under these demanding circumstances.The women who were interviewed, all of whom were incarcerated after being convicted of filicide, were living in pitiable conditions and gave accounts of extremely distressing life experiences.As they told their stories, the women demonstrated the effects of rekindling past events and memories: shock, anger, grief, despair, fear and, occasionally, hope.A few women appeared to be overwhelmed with sadness, crying throughout the interviews.Each woman was given as much time as she needed to recover from each episode of distress before being asked if she wanted to continue.When one woman wept uncontrollably, the researcher terminated the interview and sought specialist psychological assistance for her.
The first author, as interviewer, had to take care that her role as researcher was not overtaken by her role as psychiatrist.A researcher must be independent and occupy a neutral position, including when asked to intervene (Davies, Francis and Jupp 2011: 229).This is the opposite of her duty as a psychiatrist who must assess, intervene and treat.The interviewer had to listen to each woman's story and to convey her non-judgemental interest, not be the therapist who discussed solutions to the woman's problems.When interviewing vulnerable participants, such as women convicted of filicide and incarcerated, it was essential to prioritise their rights and dignity and to act on the ethical principles of justice, beneficence, autonomy and integrity (Ward and Willis 2010).
As a qualitative researcher, the first author had to overcome years of training as a psychiatrist.In her routine life in the clinic, she was trained to deal formally and consistently with her patients.

Gender-related concerns
The majority of crimes, including homicide, are committed by men.Most theories, research and conclusions on criminalities are thus based on men's behaviours (Ogle, Maier-Katkin and Bernard 1995).Nevertheless, it cannot be assumed that women's pathways to criminality, including filicide, are the same as men's.Researching filicide by women, therefore, requires investigation of their social contexts, living arrangements and interpersonal relationships, including with the child, intimate partner, parents, friends and the community.
Researching filicide by women in Malaysia is especially challenging because it requires the 'unveiling' of the hidden story of women in a patriarchal Muslim society.Malaysia is unique in that, even though most women are taught to adhere to 'Islamic family values', they may participate actively and progress in their professions, reflecting the increasing processes of development and modernisation (Nagaraj et al. 2014;Stivens 2006).Nevertheless, from the outset, the first author encountered resistance from her (male) colleagues and superiors who believed that it was inappropriate for a woman to conduct research in prisons.The masculine construct of 'the prison' is found beyond Malaysia, however, and in less overtly patriarchal societies (Claes et al. 2013).The first author was able to overcome resistance initially by providing evidence that she had worked as an intern in forensic psychiatric wards and as a visiting doctor in a women's prison.
It was important to explore whether or not patriarchal doctrines and the oppression of women, often reported in Muslim countries, contribute to the genesis of filicide in Malaysia.We had to consider the alternative explanation, which is that the emerging emancipation of women is an enabler of filicide.We also needed to reflect on whether the historical processes documented elsewhere, where there had been an 'epidemic' of illegitimacy, infant abandonment and infanticide as part of Western industrialisation (Higginbotham 1989), were evident in contemporary Malaysia.
The choice of suitable methods, as always, was vital.In conducting this research in Malaysia, at the practical level, we had to adhere to the social rules pertaining to Muslim women's interactions with others.For example, Islam discourages a woman from having a 'private' conversation with a man unless it has a clear purpose.As a woman, the first author had an advantage in conducting interviews with religious women.However, during interviews with key informants as well as in making arrangements to conduct the research, the first author also had to interact with men.As a Muslim woman having conversations with a non-mahram (unmarriagable kin) Muslim man, she had to ensure that the sessions followed Islamic etiquette.For example, the door of the room in which encounters took place had to remain open to avoid syubhah (suspicious) behaviour.In such a patriarchal society, the first author had to be aware of remaining positive, confident and obviously assured, without being intimidated by men.In short, to carry out research on filicide by women in Malaysia, the status, role and position of women in the larger society had to be understood.

A rare phenomenon and an elusive participant group
Because filicide is a rare and sensitive phenomenon that is often denied and concealed, there can be problems in collecting and analysing data.It is difficult to conduct research directly with women convicted of filicide because the population of convicted women constitutes only a small percentage of the population as a whole.When convicted women are no longer incarcerated, the stigma associated with a conviction for filicide is a powerful disincentive to participate in research: women are unlikely to want to identify themselves as former inmates.Researchers often use secondary data sets derived from surveillance, registries and court or police records maintained over decades; they are thus reliant on vigilant surveillance, well-maintained registries, accurate record-keeping and consistency (Friedman, Horwitz and Resnick 2005; Porter and Gavin 2010).Such reliance can be especially problematic in newly industrialised countries such as Malaysia.
Until our research, no data on filicide in Malaysia had been published (Razali et al. 2014).The data to which we had access were limited.Assistants were nominated from the Police Department to retrieve data on infanticide and illegal infant abandonment; they were permitted to pass on only summary data that did not enable data linkage between characteristics of victims and of any suspected or convicted perpetrators.No data fields were fully populated and it appeared that information from each case was not systematically recorded.For example, of the 1147 suspects recorded as having been arrested 1999-2011, perpetrator sex was logged for only 248 (Razali et al. 2014).We understand our contribution to include not only publishing all available data but also indicating how data gathering and record keeping can be improved.Poor records have been described by researchers from low-and middle-income countries other than Malaysia; the available resources have to be channelled to more basic needs and requirement (Frerichs 1991).
Finding participants for qualitative interviews in filicide research is also very difficult.Because of the associated stigma, potential participants who have completed their sentences are unlikely to volunteer; those who have not been convicted of the crime are unlikely to risk capture by identifying themselves to the researcher.We concluded that the most ethical way, entailing least risk to participants, was to find volunteers among women serving sentences in prisons and psychiatric hospitals.This approach produced far fewer potential participants than we had expected.Given the total of 1147 recorded arrests and the estimates provided by officials, we were confident of being able to interview at least 20 women incarcerated after convictions for filicide.Information from prison headquarters was that there were nine prisons and two hospitals in Malaysia where we would find 37 women in the prisons and two women in the hospitals who potentially met inclusion criteria.However, when fieldwork was conducted in 2013, only ten convicted women who were capable of giving informed consent were available.Most of the 1096 cases associated with the 1147 arrests were unsolved, many accused women were still engaged in court proceedings, and others had been released from prison.
The procedure of conducting research in prisons and forensic psychiatric hospitals presented further challenges.Once the environment, rules and regulations of each institution were understood and good relations had been established with a person at each level of the hierarchy who could facilitate access, practical matters had to be managed.Visits to prisons and hospitals all over peninsular Malaysia did not always result in appointments being kept.Tight security meant pat-down searches before entry to each prison and repeated scrupulous checking of all documentation: of appointments, information about the research, official approval letters and personal identification.This was repeated at the entrance to the cell complex.It was only at the first visit that we discovered that recording devices are not allowed in prisons.The interviewer had thus to make a written record of each woman's account, which meant interrupting her from time to time to keep pace with what she said.Prison staff were required to observe the women at all times; the planned confidential interview had therefore to be conducted in whispers as a warder stood just outside the open door.In the locked forensic psychiatric wards, patients lived behind barred windows and were continuously monitored.Nevertheless, private and comfortable meetings with women were possible and interviews could be audio-recorded.It is remarkable that women told their stories under these conditions.

Other challenges
Finally, in common with other low-and middle-income countries, lack of resources is a major barrier in Malaysia.There is a serious shortage of financial and human resources in the various health professions responsible for the prevention and management of filicide, in particular perinatal and child health (Gupta et al. 2011), women's, maternal and reproductive health (Fauveau, Sherratt and de Bernis 2008), and mental health (Kakuma et al. 2011) (2000), only about 10 per cent of global spending on research has been used to understand the needs of the poorest 90 per cent of the world's population.Lack of human resources in this area in Malaysia is evident.When this research started, it was estimated that there were only 0.6 psychiatrists per 100,000 population, while the neighbouring country in which the project was hosted, Australia, had 14.0 psychiatrists per 100,000 population (more than twenty times that of Malaysia) (Jenkins et al. 2010).This shortage of human resources means that most professionals in Malaysia must take on many tasks and can devote little time to research.Experts in qualitative research on sensitive topics such as filicide are extremely rare.

Conclusion
Although researching filicide is difficult, research conditions are not ideal, and it may not be possible to achieve best-practice samples and data, these are not reasons to avoid it.Vulnerable people from marginalised groups who have no other means of being heard, especially when there are institutional barriers to access, need us to conduct the best research possible, given the circumstances (Daly and Lumley 2005).
Our investigation of women and filicide in Malaysia has been a challenging yet gratifying task.It required perseverance; adaptability; openness to acquiring new knowledge and skills; careful attention to the details of rigorous research; willingness to engage in reflection about culture, religion, nationality and personal sensitivities; and committed collaboration.It has produced an original, valuable contribution to knowledge.We have analysed Malaysia's early initiatives in collecting data and illuminated areas requiring further attention (Razali et al. 2014).We have been able to demonstrate that women convicted of filicide are themselves victims (Razali 2015).
The difficulties confronted and overcome are outlined in this paper which we hope will encourage other researchers who may be contemplating research with vulnerable populations on sensitive topics.The rewards are not limited to the research results but extend to inform policy, programs and clinical practice in Malaysia. Correspondence Conducting structured clinical interviews, performing laboratory investigations and psychological tests, giving diagnoses, planning biopsychosocial interventions, and prescribing medications are routine clinical activities for psychiatrists.Such professional activities are the antithesis of the task of the qualitative researcher.It was challenging to allow the participants to tell their stories freely without interrupting or attempting to diagnose symptoms of depression, psychosis or anxiety.It was difficult not to explore further whether their psychiatric symptoms met the diagnostic criteria of the Diagnostic and Statistical Manual of Mental Disorders (DSM) or to ask specific questions about treatments.The researcher had to keep in mind that participants are not patients.The researcher's task was to understand the meaning of filicide to women convicted of killing their children and how they interpreted and explained what had led them to this predicament.
. Resources for IJCJ&SD 42 Online version via www.crimejusticejournal.com © 2017 6(2)research are even more seriously affected.According to the Global Forum of Health Research

:
Dr Maggie Kirkman, Senior Research Fellow-Jean Hailes Research Unit, School of Public Health and Preventive Medicine, 553 St Kilda Road, Melbourne VIC 3004, Australia.Email: Maggie.Kirkman@monash.edu