Infanticide in Senegal: results from an exploratory mixed-methods study.

This article presents formative research on the practice of infanticide, the intentional killing or fatal neglect of a child less than one-year-old. We hypothesised that the abortion law in Senegal, one of the most restrictive in the world, contributes directly to the incidence of infanticide. We conducted a quantitative survey of 1016 women of reproductive age living in Senegal, and in-depth interviews with a sub-sample of 28 participants. Quantitative survey data were analysed to describe the frequencies, means, and ranges of key outcome variables. Qualitative data were analysed using modified grounded theory to identify key themes in the data. Awareness of infanticide was moderately high (60.3%) in the survey sample, and was primarily obtained through personal experience, rumours, and/or the media. Participants described two broad categories of infanticide, including passive infanticide through abandonment of the infant, versus active infanticide through suffocation, drowning or other means. Participants explicitly viewed infanticide as a direct result of the severe legal restrictions on abortion in Senegal, as well as the powerful social norms that dictate what is considered acceptable versus unacceptable childbearing in the country. Findings support the hypothesis that abortion laws and policies contribute to the occurrence of infanticide in Senegal, and suggest the need for additional, targeted research to better understand this link, and how findings can be used to inform policy reform.

Introduction

Infanticide, the intentional killing or fatal neglect of a child less than one year of age, has occurred throughout history, across every known culture, geographical area[1,2] and time, including the present day.[3] A practice that evokes emotional repugnance from most, infanticide is little studied and, consequently, little understood on a population level. From the limited existing research, the circumstances under which infanticide occurs throughout the world appear to vary by setting, but share common themes, namely the extreme economic and social vulnerability of the primary caregiver(s), the scarcity of financial and social resources available to sufficiently provide for a(nother) child, as well as the quality of life that is expected for a child born into the given circumstances.[3]

Across countries, the motivation behind infanticide reflects these concerns on the part of caregivers. In India, the killing, primarily of female infants, has been attributed to the financial pressures imposed by the traditional practice of dowry giving, straining families’ ability to provide food and shelter for their other members.[3-5] In Fiji, among cases of filicide (when a child is killed by its parent(s)), many were motivated by a desperate desire on the part of the mother to prevent further violence against the child from the father, after which the mother herself committed (or attempted) suicide.[3,6] In the United States, a review of hundreds of cases of maternal filicide revealed that, in nearly all cases, the mother had given birth or was parenting in a state of extreme emotional and social isolation, often providing care to multiple children with no financial, emotional or other support – and when a partner was present, this partner was often emotionally or physically abusive.[3] In Bolivia, incidents of infanticide similarly occurred when caregivers felt that they could not raise the child in the circumstances dictated by their lives. Parents expressed a belief that allowing an infant to die early in life was the most compassionate option, one met with sadness but also relief, rather than allowing the child to experience a lifetime of suffering due to hunger or abuse.[7] One study of mothers in Brazil highlighted a practice of passive infanticide through malnutrition or neglect for infants that were otherwise deemed unlikely to survive, because they could not justify diverting scarce resources from their healthy children to a child that would likely not live.[8] In the Netherlands, a version of infanticide is codified under the Groningen Protocol, a decision-making tool that helps parents and physicians to navigate the medical decision to end the life of an infant living with presumably unbearable suffering and no hope of improvement.[9]

Within West Africa in particular, the literature on infanticide has identified a similar range of motivations for infanticide, including sociocultural pressures and stigma around acceptable and unacceptable pregnancy in Senegal,[10,11] physical abnormalities or rare birth presentations of the newborn in Benin,[12] developmental delays in Côte d’Ivoire and Mali,[13,14] beliefs that the child is a “spirit child” that will bring destruction to the family as in Ghana,[1] or in Nigeria (as in the United States),[15] omens believed to be associated with twin births,[16] and more.

Behind all of these proximate factors, we hypothesise that the legal and policy environment is a fundamentally important driver of infanticide – both directly, with respect to policies that restrict access to sexual and reproductive health care, including abortion services, as well as those that indirectly influence a caregiver’s ability to provide for a child, such as policies related to affordable housing, supplemental nutrition, accessible healthcare (including mental healthcare), or subsidised childcare.[3] In some settings, the link between infanticide and sexual and reproductive health policy seems clear. For instance, in Hungary, although not systematically documented, reports of infanticide in the media increased after austerity measures imposed in the mid to late 1990s dramatically reduced access to contraception, and effectively outlawed abortion.[17] Even more directly, data from a qualitative study in Cameroon document cases of infanticide that followed women’s failed attempts to abort unwanted pregnancies resulting from rape and/or coerced sex.[18] This study highlighted the complex reality that even though survivors of rape in Cameroon have a legitimate claim to a legal abortion, they are rarely able to access it because of the burdensome nature of the legal approval process, which consequently obliges them to resort to informal sector abortion or infanticide.[18] By and large, however, because infanticide is infrequently documented in official or unofficial statistics, the ability of researchers to rigorously study infanticide, or the influence of policy on its occurrence, is limited.

As one part of a broader research project on unwanted pregnancy and abortion in Senegal, this study aimed to conduct exploratory research on the practice of infanticide in Senegal. We hypothesised that the abortion law in Senegal, one of the most restrictive in the world, contributes directly to the incidence of infanticide. It is a remnant of a 1920 French colonial law which was repealed in 1980, but replaced by a new law asserting similar restrictions.[19] Under Senegalese law, abortion is allowed only when it is the only way to save the life of the pregnant person. In these situations, the attending physician must consult with two other physicians, one of whom must be court-appointed, who both also examine the pregnant person and attest in writing that the procedure is necessary to save the pregnant person’s life (with no exceptions for rape or incest).[20] Those who attempt an abortion for themselves, and those who provide abortions for others, are penalised with up to five years in prison and fined.[21]

Groups in Senegal have long advocated for the legalisation of abortion in cases of rape or incest, to prevent people from making the impossible choice between family and an unacceptable pregnancy. The Advocacy Committee for Access to Safe Abortion in Senegal, more commonly referred to as “The Taskforce”, has led much of this work. Formed in 2013 under the aegis of the Division of Reproductive Health of the Senegalese Ministry of Health and Social Action (MSAS), the Taskforce is a multidisciplinary committee responsible for developing and implementing strategies to inform decision-makers and the public about unsafe abortion in order to achieve social and legal change in favour of abortion law reform.[22] The Taskforce has led numerous outreach campaigns at the community level, including workshops with religious and other community leaders as well as youth, to provide information on safe medication abortion when necessary to save a woman’s life, and/or in the case of rape or incest. The Taskforce has also led research on experiences of unwanted pregnancy in Senegal[23] to inform their ongoing targeted advocacy efforts among Parliamentarians, government authorities, other state institutions (such as the Economic, Social and Environmental Council), and representatives of the media, aiming to influence public and legal opinion about the importance of granting this right to legal abortion to prevent maternal mortality due to unsafe abortion, and infanticide.

Powerful sociocultural factors and political interests, however, make members of government reluctant to publicly take a stance on reforming the abortion law, which impedes progress.[22,24] Abortion is primarily viewed as a “women’s” issue, and not a priority women’s issue at that, and is further complicated by its association with women’s sexuality, in a culture where sexuality is not always celebrated.[25] Further, social stigma silences many who have had abortions, and also changes the very way in which abortion statistics are recorded in the country. For instance, to prevent policing of patients in health facilities, most induced abortions are registered as “miscarriages” by health care workers, rather than illegal abortions.[24] This practice, however, contributes to the erasure of abortion from vital statistics, and to the consequent invisibility of abortion in the country – undermining advocacy efforts to convey the widespread need for safe abortion care. Even among those politicians that are convinced of the need to liberalise the abortion law, powerful religious leaders and interest groups have proven effective at preventing legislators from acting.[24]

As a result of this social and political climate in Senegal, recent estimates suggest that between one-quarter and one-third of all female detainees in Senegal have been convicted of illegal abortion or infanticide.[23,24,26] Scholars have commented on the direct link between the severity of the legal climate around abortion and the occurrence of infanticide in Senegal, inferring that parents, primarily women, may view infanticide as their only option because legal pregnancy termination is not available.[20,26] This assertion has been echoed elsewhere, in interviews with female detainees in Senegal,[23] from international human rights organisations,[27] and in the media.[28-31]

The limited published research on infanticide in Senegal, conducted in the 1980s and 90s, includes research that analysed cases of infanticide among psychiatric patients in Dakar, and found that fear of parental reaction and social rejection, severe economic limitations, and the isolation of women were the reported causes of infanticide, which mostly resulted from strangulation.[11,32-34] Yet, the lack of recent, scientific research on infanticide in Senegal at the population level limits our ability to study the motivations for and conditions under which infanticide occurs, the link between policy and infanticide, and ultimately, to develop interventions to change policy and eliminate the circumstances that lead caregivers to infanticide. We conducted this study to gather preliminary quantitative data and in-depth qualitative data on knowledge and perceptions of, as well as motivations for, infanticide in Senegal.

Methods

Sample selection

For this mixed-methods study, our target population was women of reproductive age from four regions of Senegal (Dakar, Diourbel, Louga, and Ziguinchor). Study investigators selected the four regions in collaboration with local partners to represent a range of reproductive health settings within Senegal, inclusive of urban and rural areas, as well as different linguistic groups. Men were excluded from this study as the primary focus of the larger study was on experiences unique to pregnancy. We utilised a multi-stage, probability-proportional-to-size sampling strategy to select survey participants. Within each of the four regions, the primary sampling unit was the commune [equivalent to a municipality or district], and the secondary sampling unit was the neighbourhood [a defined area within the commune]. Individual communes were selected with probability proportional to estimated population size using the sample function in the R software platform [http://www.R-project.org/]. Within selected neighbourhoods, all households were counted by a team of female enumerators from the Centre de Recherche pour le Développement Économique et Social (CRDES), and a sampling interval was calculated to approach every nth house, resulting in 20 selected households per neighbourhood. A household was eligible if it contained at least one woman of reproductive age (15–44 years). Within each selected household, women aged 15–44 years were enumerated and one was randomly selected to participate. A total sample size of 1000 individuals was targeted to reach approximately 85 individuals with a history of induced abortion in the past five years, based on 2012 abortion estimates for Senegal,[35] with a 3% margin of error, 90% power, an alpha of 5%, and an allowance for under-reporting of abortion.

The subsample of interview participants for this study was selected from among survey participants who reported either a personal experience with abortion, knowledge of a best friend’s experience with abortion, or knowledge of someone who had practiced infanticide, and who gave consent to be contacted in the future by study investigators. Given the exploratory nature of this research, a target sample size of approximately 25 interviews was set to capture at least a minimum range of experiences with regard to abortion and infanticide.

Data collection

Data were collected in September and October of 2017. Survey data as well as qualitative interview data were collected by a team of 25 trained, female enumerators and 5 supervisors. The survey was coded with the Census and Survey Processing System (CSPro) (www.census.gov/data/software/cspro.html) by partners at CRDES, and was administered using a computer tablet by enumerators to individual respondents in selected households. The survey instrument contained questions on contraceptive knowledge and use, experiences with unwanted pregnancy, knowledge and perceptions of abortion, abortion law, and infanticide, as well as sociodemographic characteristics. The in-depth interview guide contained sections on individual and community norms regarding pregnancy, birth, family size, unwanted pregnancy, abortion, and infanticide, as well as personal experiences with unwanted pregnancy, abortion, and infanticide, and knowledge of abortion law. Interviews were conducted in-person, at a time and place of the participant’s choosing, and were audio-recorded.

Analysis

Quantitative survey data were weighted to account for sampling design, and analysed to describe the frequencies, means, and ranges of key outcome variables. Audio files from qualitative interviews were transcribed and analysed in the original French, and then subsequently translated into English. The original audio files were compared to transcripts to check for accuracy and any systematic errors in transcription. Two researchers analysed the transcripts using modified grounded theory, a systematic approach to identifying key themes and ideas present in the data. The first and second authors developed a preliminary codebook based on the interview guide and an initial review of the transcripts. This codebook was applied independently and in parallel to three transcripts by the research team to assess consistency in code application, to identify codes that were lacking, and to remove unnecessary codes. Individual codes were combined or sub-divided in an ongoing analytic process to best organise and detect patterns in the content of the interview data. Through discussions with the research team, the codebook was revised accordingly and the second author coded all remaining transcripts in the original French using the online software, Dedoose (www.dedoose.com). Each included interview transcript excerpt is described with the participant’s age and place name.

Ethical approval. This study was reviewed and approved by the Comité National d’Ethique pour la Recherche en Santé based in Dakar, Senegal.

Results

Survey and interview sample characteristics

A total of 1016 women participated in the quantitative survey (Table 1). Participants were aged 18 to 44 years, with 44% between 21 and 30 years. The majority of participants had a primary education or less (63%), just over half were married (57%), and slightly less than half were employed (46%). From within the larger survey sample, 28 individuals participated in the in-depth interviews. Similarly, among interview participants, approximately half were between the ages of 20 and 30 years of age, most were married, and a majority had completed only a primary school education or less (Table 3). Two of the women reported a prior abortion, and just over half knew someone who had practiced infanticide.

Table 1.

Sociodemographic characteristics of survey participants

		%	%
	N	Unweighted	Weighted
Age (years)
18–20	71	7	6
21–25	236	23	23
26–30	196	19	21
31–35	222	22	21
36–44	291	29	30
Languages spoken
Wolof	993	98	97
French	343	34	42
Poular	156	15	18
Serer	122	12	12
Mandingue	76	8	7
Diola	110	11	7
Other	63	6	9
Education
None	327	32	21
Primary	302	30	43
Secondary	290	29	38
Post-secondary	97	10	18
Employment
Currently in school	124	12	16
Currently employed	414	41	46
Employed and in school	2	0	4
Relationship status
Single, never married	188	19	24
Unmarried, living with partner	6	1	9
Married, living with partner	544	54	52
Married, not living with partner	216	21	15
Divorced/separated	50	5	7
Widowed	12	1	2
Religion
Muslim	982	97	96
Christian	34	3	4
Other	0	0	0

Table 3.

Sociodemographic characteristics of qualitative interview sample

	N	%
Age (years)
18–20	4	14
21–25	6	21
26–30	7	25
31–35	3	11
36–44	7	25
Missing	1	4
Ethnic group
Bambara	1	4
Diola	2	7
Laobe	1	4
Maure	2	7
Poular	2	7
Soce	1	4
Serer	4	14
Toucouleur	4	14
Wolof	7	25
Other	4	14
Education
None	1	4
Primary	11	39
Secondary	8	29
Post-secondary	3	11
Missing	5	18
Employment
Food seller	6	21
Hair dresser	1	4
Housekeeper	2	7
Teacher	1	4
None	15	54
Missing	3	11
Relationship status
Single	8	29
Married	17	61
Divorced/separated	2	7
Widowed	1	4
Missing	0	0
Knows someone who has practiced infanticide	15	54
Prior abortion	2	7

Survey results

The majority of participants were aware of Senegalese law with regard to abortion (73%), although only 53% correctly stated the content of the law. The majority of participants had heard of infanticide (60%, Table 2) and over 15% personally knew (14%) or suspected (1%) someone who had practiced infanticide. Forty per cent of survey participants declined to answer the question on infanticide. Respondents personally knew between one and four individuals that had practiced infanticide, or an average of 1.3 individuals that had done so (Table 2).

Table 2.

Knowledge of infanticide among survey participants

		%	%
	N	Unweighted	Weighted
Have you heard about infanticide?
Yes	597	59	60
No	419	41	40
Do you know anyone who has practiced infanticide?
Yes	120	12	14
I suspect	11	1	1
No	466	46	45
Missing	419	41	40
How many people do you know who have practiced infanticide?
	Range	Mean	Mean
	1–4	1.3	1.3

Interview results

Participants described a range of experiences with infanticide, particularly with regard to knowledge and awareness, methods of, and reasons for infanticide. Each is described in more detail below.

Knowledge of infanticide

Interview participants learned about individual cases of infanticide through a variety of channels, from first-hand experiences, to rumour or suspicion, to reports in the media. A number of participants discussed personal experiences with infanticide, such as personally discovering the body of the deceased infant, as described by one participant: “I once assisted in a scene where a child was found on a bridge in Medina” (age 28, Medina). Other participants emphasised the role that rumour plays in knowledge of infanticide cases. Comments such as “I have heard it … ” or “It seems that … ” suggest that rumour or suspicion (rather than direct experience) informed statements from some participants, and when probed on specific details, participants admitted as much. Several participants, however, described specific situations in which the absence of a baby after a woman had been visibly pregnant provoked suspicion of infanticide. These cases often led to interrogations and/or denunciation of the pregnant woman. One woman reported an instance of infanticide in this context: “Everyone knew she was pregnant and we didn't see a child after she gave birth” (age 38, Bambey). She went on to describe that the community called the police about this woman based on these suspicions, and the infant was eventually found in a bag. In addition to rumour, a substantial portion of participants highlighted the role of the media (radio, television, and newspapers) in spreading information about cases of infanticide that occurred in the country.

Methods of infanticide

Participants identified two overarching methods of infanticide: (1) the passive death of an infant as a result of abandonment, or (2) the active killing of an infant and disposal of its body. In the first group, participants described the abandonment of infants in various places such as toilets and septic tanks, on railways, or in garbage cans. One interviewee comments: “In our very own community, a woman left her twins by the railway tracks” (age 39, Guinaw Rail Nord). In some cases, the infant was discovered before death occurred; however, the act of abandonment was still viewed by the authorities as infanticide, or attempted infanticide.

In the second category, participants described cases where the infant was intentionally killed, after which varied means were used to dispose of the body. Interviewees mentioned a range of methods used, including strangulation, drowning, and live burial. As commented by one of the respondents: “I have heard that some people put their unwanted children in a pit or grab the baby by the neck until death” (age 42, Linguere/Abattoir 2). For disposal of the body, a number of methods were mentioned including discarding the remains in landfills or septic tanks, burial in the forest, hiding remains in the household (such as in a refrigerator, or wardrobe). One interviewee described a specific instance of this (age 27, Pikine/Yeumbeul Nord). This participant spoke of a woman who was forced into a marriage arranged by her parents in an attempt to preserve her reputation after she had a child out of wedlock. Yet, even after marriage, she became pregnant again by another man who was not her husband. During this time, she was working as a maid in Dakar and tried to hide the pregnancy from her family. After giving birth, she attempted to discard her newborn in a closet of her employer’s home during a wedding celebration. However, her employer had known of her pregnancy and thus, upon discovering the infant, knew that it belonged to her – and thus reported her for infanticide.

Motivations for infanticide

While acknowledging that each individual case of infanticide is unique, two primary themes emerged in discussions of the motivation for infanticide: the societal conditions under which childbearing is considered acceptable, and the inaccessibility of abortion, related to its criminalisation. Interviewees described cases of infanticide that occurred in conditions under which having a child would be considered unacceptable by society, such as a child born to an unmarried woman, to a married woman who had been unfaithful, to an unmarried woman whose partner denied paternity, or to a woman that had been raped. One participant, describing the fear of societal reaction to infidelity, said: Other participants even more explicitly named societal stigma as a motivator of infanticide. For instance, one woman described: “Very often it is to avoid being marginalised in the community, where getting pregnant out of wedlock is seen as an indecent act. It is this hostile social climate that often leads to infanticide” (age 17, Rufisque ouest).

“Her husband was in prison for a long time. Unable to stay still, she got pregnant by another man. Knowing that this would be viewed badly, that [her community] would point fingers at her and her future baby, she preferred to sacrifice her own child” (age 23, Bambey).

This fear of societal ostracisation was complemented and at times even outshone by fear of familial ostracisation. One participant described an instance of this as follows: Participants’ comments suggested that women who find themselves in these situations where a child would be socially unacceptable try to hide the pregnancy entirely, never disclosing it to anyone, so that the woman could seek informal sector abortion and/or commit infanticide without anyone ever knowing she was pregnant, and thereby avoid suspicion entirely. However, this was not always possible because, in some cases, the pregnancy only became unacceptable late in gestation; for example, when a husband/partner or other man involved in the pregnancy denied paternity or refused to accept responsibility for raising the child – after the pregnancy was known to the woman’s family and/or community. As described by one respondent: In these situations, the pregnancy became unwanted at a gestational age when abortion was no longer possible, thus, for some people, infanticide was viewed as their only option.

“It is mainly the relatives because here, when a woman gets pregnant and she is not married, or [gets pregnant] from a man other than her husband when she is married, her parents no longer speak to her, or [they] kick her out of the family house and, therefore, she is forced to hide her pregnancy and once she gives birth, she has no other option but to commit infanticide.” (age 19, Mermoz/Sacre Coeur)

“There are many reasons that a woman can become pregnant in the absence of her husband. Sometimes the man who has impregnated her refuses to assume his responsibilities, and sometimes, by the time this happens, it is too late to have an abortion. Also, some women do not want to have an illegitimate child in their lives that will ruin and tarnish their reputation.” (age 23, Bambey)

In other cases, interviewees described infanticide as a phenomenon that resulted from a long process of failed abortion attempts due to the lack of safe, legal abortion services in Senegal. These cases comprised the second major category of reasons for infanticide: infanticide as a consequence of the criminalisation of abortion. As one participant described it, the primary cause of infanticide results from an inability to access safe, effective abortion care: “Most [cases of infanticide] result from unwanted pregnancies and the woman has not been able to find someone who can help her have an abortion” (age 24, Dakar Buiscuiterie). Participants described the severe legal restrictions on abortion in Senegal, and that as a result, the only abortion care that is available is clandestine and requires knowledge of its existence, an ability to pay, and a tolerance of the potentially significant risks to one’s own health due to variable safety and effectiveness. As a result, interviewees described women who were unable to obtain an abortion and were therefore forced to carry unwanted pregnancies to term, for which some resorted to infanticide. One of the interviewees directly attributed infanticide to the legal restrictions on abortion: “The laws are strict, especially in case of rape and incest; if abortion was allowed, there would be no more infanticide” (age 36, Dakar Gueule Tapée).

Other, less frequently mentioned motivations for infanticide included: economic hardship, and, to a lesser extent, mental illness. Participants cited poverty as a common motivation for infanticide that manifested in a number of ways, from the inability to provide food, shelter or other basic needs to a(nother) child, to the inability to pay for or otherwise obtain contraception to prevent pregnancy in the first place. Two respondents mentioned mental illness as the reason for infanticide.

While participants expressed an understanding for the range of motivations for infanticide, they unanimously condemned the practice and identified no mitigating circumstances under which it might be considered anything less than criminal. In case of evidence or even suspicion of infanticide, participants reported that those alleged to be involved in infanticide were reported to the police and sent to prison, and that the stigma may persist long after: “Her employer reported her to the police. She has completed her prison sentence. […] Everyone was denouncing her act. It's a good thing she doesn't live here anymore” (age 27, Yeumbeul Nord).

Discussion

In this mixed-methods study of infanticide in Senegal, we found that awareness of infanticide was moderately high, and was primarily obtained through personal experience, rumours, and/or the media. Participants described two broad categories of infanticide, including passive infanticide through abandonment of the infant, versus active infanticide through suffocation, drowning or other means. Participants understood infanticide to be a direct result of the severe legal restrictions on abortion, as well as the powerful social and religious norms that dictate acceptable versus unacceptable childbearing in Senegal.

These findings are consistent with the limited peer-reviewed evidence on infanticide in Senegal that has similarly identified socio-cultural factors such as societal stigma, familial pressure, and the social and political isolation of women as primary motivators of infanticide.[10,11] Even decades ago, caregivers reported resorting to infanticide in the face of pressure to conform to strict societal and religious scripts for acceptable pregnancy and parenting.[11] In an analysis of female psychiatric patients convicted of infanticide, infanticide was seen as their only remaining defence mechanism against debilitating societal exclusion, a fate that many likened to death.[11] While infanticide had been viewed as something performed only by vulnerable young women in isolated rural areas, research from Senegal in the late 1980s found instead that infanticide was practiced much more widely across social strata, as the same social pressures likely to make a birth unwanted affected women of all backgrounds.[10]

Of central importance, the findings presented here provide evidence for the hypothesis that infanticide occurs in Senegal as a direct result of policies related to the legal status and criminalisation of abortion. The opinions shared by interviewees align with those stated by others in the media, in the legal field, and in public health.[28,31,35] Individual stories of women who wanted abortion but could not obtain one and went on to consider infanticide as their only option make explicit the impact of policy on the lives of women and their families in Senegal. As part of a larger portfolio of research, the data presented in this study can and will be used by the Taskforce on Safe Abortion to communicate to legislators the urgency and gravity of the need to address abortion policy in Senegal. In recent years, the Taskforce has increased outreach efforts to other countries in the region to share strategies and lessons learned in advocating for policy change, and is seeking new ways of leveraging continent-wide initiatives, such as the Maputo Protocol,[36] to strengthen their efforts to liberalise abortion law.

This mixed-methods study is limited by the fact that no validated measures exist for asking about infanticide in a research study – a reflection of the lack of scholarship on this subject in general. Instead, we developed our own exploratory questions in consultation with local partners – but as a result, we do not have quantitative data on the direct occurrence of infanticide, nor knowledge of the sensitivity or specificity of the questions that were asked. Another limitation may be the multiple languages in which this study was conducted – the survey was translated into several local languages, and despite careful work in the design stage, some items may have translated differently in different contexts. Similarly, for qualitative results, the interviews and analyses were conducted in French, but some meaning may have been lost in translation to English.

The strengths of this study, however, include the mixed-methods nature of the design, which allowed for the generation of preliminary, quantitative statistics on knowledge of infanticide in Senegal, and also more nuanced qualitative data on the context in which it occurs, and the ways in which it is understood by the population. Further, the probability sample used for the quantitative sample increases the representativeness of the quantitative findings and the generalisability of the results. Findings from this study can inform future studies of infanticide in Senegal, and offer a contribution to a little understood, but crucially important aspect of the link between sexual and reproductive health policy and infanticide.

In conclusion, findings from this study indicate that infanticide occurs with some regularity in modern-day Senegal, and that a substantial proportion of cases may result directly from the severe legal restrictions on abortion in the country. The evidence presented here is consistent with claims made in the media and by civil society regarding the impact of abortion policy on the existence of infanticide in Senegal. These findings warrant additional, targeted research to better understand the link between abortion, the ability to access a full range of sexual and reproductive health services, and infanticide in Senegal, and how findings can be used to inform policy reform. In particular, future research should draw from the narratives presented here to rigorously inquire how people understand and experience the crime of infanticide, how criminal justice systems respond to it, how the media cover it, and how the incidence of and individual experiences with unwanted pregnancy and abortion in Senegal can highlight pathways by which people come to commit infanticide. On a broader scale, contextual research is needed to understand the extent to which the burden of social norms regarding sexuality and its consequences (pregnancy) is carried by women, and what is needed to shift this reality. Rigorously collected statistics, brought to life by individual stories, are needed to increase the effectiveness of advocacy in Senegal, and to compel those with authority to act for change. Advocates should explore new avenues of reaching and influencing public opinion, not only to engage legislators, but also to inform and activate the public, particularly young people, to demand the change necessary to reduce the occurrence of infanticide, and the impossible choices embodied by this phenomenon.