Silence as Self-care: Pregnant Adolescents and Adolescent Mothers Concealing Paternity in Mahama Refugee Camp, Rwanda

In Rwanda, sexual activity with and among adolescents under the age of 18 is a criminal offence. This is justified to reduce abuse and adolescent pregnancies. Despite this, the Burundian Mahama refugee camp in Rwanda is registering an escalating pregnancy rate among girls 13 to 15 years old. Drawing on ethnographic research conducted from December 2017 to April 2018, this paper shows how pregnant adolescents and adolescent mothers navigate punitive legal structures to protect their baby’s father by concealing his identity. In a challenging socioeconomic context with limited opportunities, silence provides pregnant adolescents and adolescent mothers with a strategy to protect their boyfriends from jail and to access humanitarian assistance available to single mothers. I suggest that silence can be a self-care strategy to negotiate and navigate temporalities as they seek to manage the circumstances in which they find themselves, whilst hoping for a better future for themselves and their children.


Introduction
This paper examines a trend among pregnant adolescents/adolescent mothers living in Mahama refugee camp in Rwanda who refuse to disclose their child's paternity. It draws upon findings from a wider research study that primarily explored how adolescents experience, navigate and negotiate love relationships and sexuality, wherein this particular strategy of silence, by pregnant adolescents/adolescent mothers, emerged.

3
Silence as Self-care: Pregnant Adolescents and Adolescent… In Rwanda there is growing consensus on the need to prioritise initiatives to curb adolescent sexual activities and adolescent pregnancy. The government and other stakeholders carry out awareness initiatives, including public and school talks, radio, TV, billboard and other media campaigns, and awareness raising through HIV clubs. The Burundian Mahama refugee camp in Rwanda is not exempt from these campaigns. Throughout the camp, numerous billboards aim to discourage girls from engaging in sex, with texts in Kinyarwanda/Kirundi loosely translated as: "HIV/ AIDS does not have a cure or a vaccine … Youth, avoid drugs and temptations", "She is keeping herself for marriage… what about you?", "No sex for help", "Everyone deserves to live free from violence: stop child abuse", "You have the right to say no: do not engage in forced sexual activities", and other similar messages. The country also has various policies and laws, including regarding the age of consent, to prevent and reduce sexual abuse and other unwanted negative sexual and reproductive health outcomes. The law thus forbids sex below the age of 18 and makes having sex with a minor a criminal offense; this applies to sexual activity both with and among adolescents.
Fear of prison time does not appear to be preventing sexual activity with and amongst adolescents, underscoring the need for a better understanding of these relationships from the perspectives of those involved. This paper focuses on how adolescent girls who 'find' themselves pregnant/adolescent mothers navigate the laws prohibiting sexual intercourse below the age of 18. Despite a large body of research addressing adolescent pregnancy (Ahinkorah et al., 2021;Cressey et al., 2020;French, 2019;Kassa et al., 2018;Worku et al., 2021), and evidence of some form of existing agency of girls in constrained contexts such as Mahama (Pincock, 2019;Utas, 2005aUtas, , 2005b, limited research has explored why pregnant adolescents/ adolescent mothers choose to conceal the paternity of their child, and thus appear to accept the blame and stigma of being pregnant and having a child alone. What sort of relationships are these pregnant adolescents/adolescent mothers involved in? What does keeping silent about paternity mean to pregnant adolescents/adolescent mothers?
Through presenting cases studies from ethnographic fieldwork, I aim to disrupt the current discourse of silence as problematic, arguing that it is not always indicative of an absence of agency. I specifically argue that this strategy of concealing paternity is a relatively powerful agentic move, one that denotes not only the pregnant adolescents/ adolescent mothers' understanding of the legal and humanitarian environment in which they are embedded, but also their capacity to use this to their own advantage. For these pregnant adolescents/adolescent mothers, silence is a means of self-care.

Legal Context
In Sub-Saharan Africa, premarital childbearing is very common and reportedly on the rise (Ahinkorah et al., 2021;Clark et al., 2017;Kassa et al., 2018; 1 3 Worku et al., 2021). Many adolescents have sexual intercourse before the age of 15; in Burundi and Rwanda, the median age at sexual debut is 17 years (Amo-Adjei & Tuoyire, 2018). Low contraceptive use and access among adolescents in Sub-Saharan African countries remain problematic (Kaphagawani & Kalipeni, 2017;Sedgh et al., 2015), leaving them increasingly vulnerable to unintended pregnancy (Kolling et al., 2017). The age of majority in Rwanda is 18, which is also the legal age of consent for sexual activities. This age serves primarily to discourage child abuse and exploitation, especially child marriage, although various advocates for children's wellbeing also put it forward as an imperative to encourage delayed sexual debut and reduce teenage pregnancies.
In Rwanda's penal code of May 2 nd , 2012, Chapter VI, Section II, Articles 190,191,192 and 194 state that a person who has sex with a child (under the age of 18) has committed the crime of defilement. The penal code defines a range of penetrative and non-penetrative sexual assaults and stipulates corresponding penal provisions. More specifically, Article 190 defines defilement as "any sexual intercourse or any sexual act with a child regardless of the form or means used". By legal standards, and regardless of the minor's own volition, it does not matter whether a person forces a minor to have sex, entices them with gifts or convinces them with sugary words (see Sexual Rights Initiative (SRV) 2017) 1 . In its General Comment No. 20 (2016), the UN Committee on the Rights of the Child clearly states that "states should avoid criminalizing adolescents of similar ages for factually consensual and nonexploitative sexual activity" (paragraph 40) (UN Committee on the Rights of the Child (CRC), 2016). It should also be noted that, as in other countries, the actual average age at sexual debut in Rwanda is much lower than the lawful age of consent (cf. Binagwaho et al., 2012). Moreover, research has shown that earlier sexual relationships are not always a result of sexual violence, but of adolescent exploration and experimentation (Michielsen et al., 2014). Nonetheless, in most African countries, including Rwanda, it appears that criminalising consensual sexual conduct between adolescents under 18 years is still commonplace; with the girl considered the 'victim' and the boy the 'perpetrator' (see Okwatch, 2019).
The provisions in the Rwandan penal code of 02/05/2012 cover many different forms of child sexual abuse, and efforts to bring offenders to justice are enforced through the law. However, as noted above, there is no way to define sexual relationships among adolescents other than in terms of defilement and violence. While I do not dispute that relationships between young people can be violent or non-consensual, there is nevertheless a widespread perception in Rwanda that adolescent boys risk criminalisation for engaging in consensual sexual activity; a perception amplified by the fact that the law states that a person can be held criminally responsible from the age of 14 (Binagwaho et al., 2012). In this paper, I discuss this widespread perception that consenting adolescent boys will be criminally punished for defilement, and argue that the way the law is applied provides an important piece of context to understand why pregnant adolescents/ adolescent mothers may decide to keep silent when it comes to the identity of the father of their (unborn) child.

3
Silence as Self-care: Pregnant Adolescents and Adolescent…

Cultural and Social Norms
When it comes to premarital sex, there is a double standard around gendered norms that reflects historical-cultural ideals about sexual relationships. The Rwandan and Burundian society expect girls to abstain from sex until marriage. Breaking from this expectation earns them the status of 'bad girls' and brings shame to their community (Bahimana, 2020;Michielsen et al., 2014;Nkurunziza, 2010). Boys, in contrast, often get a free pass to engage in premarital sex and are sometimes subtly expected or encouraged to do so -by their peers or parents not instilling in them a fear of sex, as they do with their daughters (Brown et al., 2001). Thus, the subject of premarital sexual activity is shrouded in denial and silence, which is broken, however, when it is materialised by a pregnancy. Indeed, Sommers, (2013), when discussing adolescent sexuality, adolescent pregnancy, and sexual violence in Burundi, cited a Burundian government official who acknowledged the topic to be a legal and social grey area, remarking that, "We only call it sexual violence if the girl becomes pregnant" (Sommers, 2013: 34).
Sexual violence is at the core of present-day discourse about young people and sexuality in Africa. A study by Stockman et al., (2013) suggests that there is a correlation between early sexual debut and sexual violence. The same study indicates the prevalence of forced sexual initiation in Africa to varies between 5 and 46%. For example, a multi-stage cluster survey conducted among 744 young women aged 15-24 in eight provinces in Burundi found that 21.8% of those aged 15-19 reported having experienced sexual violence (Elouard et al., 2018; for more on sexual violence in Burundi, see Nsengiyumva, 2018). Although there are some studies documenting the correlation of sexual violence and early sexual debut, robust data are still lacking and not necessarily representative. Furthermore, challenges exist that may hinder survivors of sexual violence from speaking out (e.g., Nguyen et al., 2018).
With growing awareness of the phenomenon, the number of initiatives to prevent sexual violence in African countries (and globally) has increased. Indeed, the general discourse around sexual violence often emphasises girls' vulnerability, as a consequence of their age, gender and economic status, presenting young men or boys as predators even in instances of consensual premarital sex. In refugee contexts, the already vulnerable condition of girls, amplified by forced displacement, heightens their exposure to sexual violence (Krause, 2017). Thus, sexual violence often comes on top of the multiple and intertwined risks that female adolescents in refugee or conflict-affected zones confront. Moreover, despite the risk for boys to experience sexual violence, many studies continue to focus only on female victimisation at the expense of men and boys (Hilhorst, 2015;Onyango & Hampanda, 2011;Sumner et al., 2016). Indeed, Turner, (2017) draws attention to a colonial legacy that constructs poor black men as inferior and inherently violent, arguing that these constructions produce certain understandings of violence.
Simultaneously, pregnant adolescents/adolescent mothers also suffer from the negative effects of stigmatisation. In Burundi and elsewhere in Africa, adolescent mothers are often not allowed to return to school, or they might be sent to a different school far away, whereas boys face no such consequences (Ruzibiza, 2020).

3
In contrast, the Rwandan government has been lauded for promoting gender equality and adolescent sexual and reproductive health and rights (SRHR) (Coast et al., 2019) due to its national education policy commitment to support school-going girls who become mothers to complete their education. As I have argued elsewhere, however, the experience of Burundian adolescent refugees in Mahama camp shows that these national policies must compete with the stronger broader social discourse that constructs school as a space incompatible with pregnancy (Ruzibiza, 2020).
Ambiguities, inconsistencies and contradictions can be observed in terms of Rwanda's age of consent law, the age of sexual debut, and the cultural and normative ideas surrounding Burundian adolescents' sexuality. The Rwandan age of consent law plays into Burundian refugees' social norms that disapprove of adolescent sexual relations before marriage; a paradox that adds more uncertainties to the experience of pregnancy, but, at the same time, sets the stage in which young women and girls negotiate their pregnancy experience.

Silence in the Context of Sexuality
Obscured within the dominant focus on cultural norms and sexual violence is girls' sexual agency. In both Burundi and Rwanda, widely adopted Christian values have imbued sexuality with silence and shame. This is a possible contributor to the contemporary stigma associated with most negative or unwanted sexual outcomes, especially HIV/AIDS and premarital pregnancy (Delius & Glaser, 2005). Thus, in the academic literature on sexuality, silence is often problematised; framed as 'cultural silence' that hinders communication about sexual matters between youth and their parents or other adults, which leads to negative or unwanted sexual outcomes. Silence is hence a major barrier to the effective implementation of adolescent SRHR programs (Denno et al., 2015), and from a health perspective is seen as a challenge to overcome (Kolling et al., 2017).
I draw on arguments made by Arnfred, (2004) to question the conceptual usefulness of the notion of a 'culture of silence' -which, in Arnfred's argument, has the effect of collapsing many different silences into one -and instead attempt to identify "different types of silences" (Arnfred, 2004:74). For example, Arnfred, in her work on sexual matters in Mozambique, conceptualises silence as discretion as opposed to something oppressive, and argues that this discretion prevents discursive rather than sexual acts. Indeed, some recent studies have explored the significance of discretion, although mostly with same-sex relationships (see Gaudio, 2009;Hoad, 2007;Pierce, 2007).
In the field of gender, development and feminist literature, silence is mostly conceptualised as a 'symbol of passivity and powerlessness' (Gal, 1991:175). Thus a woman's ability to 'break the silence' by voicing her problems and opinions, as well as her ability to make her own choices, are predominantly understood as synonymous with empowerment (Olsen, 1978;Klugman et al., 2014;hooks, 1981). Pincock's (2017) work with Tanzanian girls, however, criticises international development for imposing an already pre-defined understanding of empowerment. Finlay et al. (2017) further remind us that when empowering the Silence as Self-care: Pregnant Adolescents and Adolescent… SRHR capacities of young women, it is essential that they are not viewed as victims devoid of agency, but rather that their extant agency is acknowledged. Empowering efforts should therefore not focus on changing women, but on providing them with support and inspiration to make the right decisions throughout the course of their lives. Kabeer, (2010), and Parpart, (2010) have argued that a continued misunderstanding of silence within the empowerment and development realm fails to acknowledge that the meanings of silence are constantly socially constructed and vary across contexts and settings. It is thus important to reconsider the context and meaning of empowerment and how it works in theory and practice. Parpart, (2010) nuances the concept of silence by arguing that although resorting to silence may at times be a restricted choice, it can still have a significant positive impact, in that it provides room for agency. Others have pointed to the fact that silence can be imposed, as in being silenced, or it can be a choice, as in choosing to be silent (Fivush, 2010;Sheriff, 2000).
Earlier scholars have often pointed to various ways in which silence can be understood and exercised. Glenn, (2004) posits that disadvantaged groups use 'a rhetoric of silence' to subvert power; for instance, the use of coded/veiled languages that are ambiguous, laconic or indirect. Mazzei, (2007) speaks of 'silence as strategic': to avoid social stigma around HIV/AIDS (Parpart, 2010), or to avoid homophobia or racist remarks (Lorde, 2007). Gal points to various forms of cultural expression (including silence) adopted by women as a way of asserting power or resisting of the power of others (Gal, 1991:175), while Armstrong, (2002) contends that silence can also be embraced as a survival strategy in conflict.

Methodology
This paper draws on findings from a wider ethnographic study conducted among Burundian adolescent refugees (boys and girls aged 13-19) in Mahama refugee camp, which was part of a project that included Nakivale settlement in Uganda. The overall research study aimed to understand how Burundian adolescent refugees experience, navigate and negotiate love relationships and sexuality in their everyday lives. This paper discusses findings obtained in Mahama, where I spent four months in the field carrying out ethnographic research from December 2017 to April 2018.
As the fieldwork in Mahama progressed, one of the major emerging findings was that pregnancies among girls aged 13-15 were on the rise (see also my discussion elsewhere about teenage pregnancy and schooling: Ruzibiza, 2020). This paper is about a core theme that arose in this context, namely the silence of pregnant adolescents/ adolescent mothers concerning paternity. It offers an opportunity to explore how pregnant adolescents/adolescent mothers navigate the different challenges they face around adolescent pregnancies, and it is an opportunity to understand their experiences from their own perspectives.
The data was collected through focus group discussions (FGDs), individual indepth interviews (IDIs) and sexual diaries (maintained by young people aged 18, for a period of two weeks). Participants were from different provinces of Burundi, and different ethnic and social backgrounds. Some of my interlocutors were in Mahama with their whole families, others with just one or a few family members, and yet others were living with a foster family or on their own 2 . Over the course of the fieldwork, approximately 60 adolescents actively participated in FGDs on varying themes that were iteratively developed over time. In total, 13 FGDs took place, with most adolescents participating in more than one. I respected participants' requests to have separate groups for girls and boys; I also hoped it would encourage open discussion. Of the 13 FGDs, four were with younger adolescents aged 13-15 (two with boys and two with girls), while the other nine involved older adolescents aged 16-19. Moreover, 45 IDIs were held with 32 girls and 13 boys, lasting between 30 and 45 min. Over the course of the fieldwork, I also met informally with most of the research participants.
The main criteria for participant selection in the larger study was unmarried adolescents. Within this group, a sub-category of unmarried pregnant adolescents/adolescent mothers (who had become pregnant while in Mahama) emerged. In addition to the data from the pregnant adolescents/adolescent mothers, I use data from interviews with other respondents (adolescent boys and other adult participants) when it pertains to the topics of adolescent pregnancy and the concealing of paternity.
This research took place two years after most of the refugees had arrived in Mahama. Most respondents were either at an advanced stage of pregnancy or had a young baby up to one year old. Aware of the stigma attached to teenage pregnancy, I held IDIs (n = 10) with pregnant adolescents/adolescent mothers instead of involving them in FGDs. The stigma around adolescent pregnancy made it difficult to recruit respondents, let alone encourage them to express their opinions freely. I did nevertheless manage to negotiate and create a safe space. Due to the fact that most of them had no help in caring for their newborn (due to stigma and family rejection), I would often meet them when they went alone for a check-up at one of the camp's two health care centres; there we would find a quiet place to talk.
For a deeper understanding of the context and guided by themes that emerged in my discussions with adolescents, I interviewed key informants including community leaders and some NGO representatives in Mahama. It was very challenging to hold a formal and recorded interview with NGO staff because of their self-censorship. Therefore, interactions were informal and superficial. I also interviewed parents, although not necessarily the parents of the adolescents included in the study. This facilitated the crosschecking and triangulation of data. All IDIs and FGDs were conducted in Kirundi.
Before, during and after fieldwork, ethical considerations were dealt with according to research ethics committee guidelines. Informed consent was obtained from all participants, with the guarantee of anonymity. I therefore used pseudonyms throughout data recordings and writings, and personal details that could potentially reveal the identity of informants have been altered or omitted. Due to the qualitative character of the research, priority was given to an inductive approach, which involved paying attention not to force topics onto respondents.

3
Silence as Self-care: Pregnant Adolescents and Adolescent…

Findings
Fifteen-year-old Ella considered the age of consent law, which criminalises sexual activities with and among adolescents, to be particularly unfair; she insisted that she had been in a consensual sexual relationship. She therefore developed a strategy to uphold what she believed was right: she decided to keep silent about the identity of her baby's father. Reviewing this strategy employed by Ella, and other pregnant adolescents/adolescent mothers, this section highlights how individuals resist the boundaries of the existing legal framework while simultaneously existing within and working with it.

Dealing with an Unplanned Pregnancy
Ella was a 15-year-old Burundian girl from a middle-class 3 family, originally from Kirundo, who had been living in Mahama refugee camp for two years when I met her and her 11-month-old baby girl in 2018. She had fled the 2015 Burundian conflict with both parents, three sisters and one brother. Ella explained that when she discovered she was pregnant and had to decide how to manage her situation, she tried to imagine and construct the best possible outcome.
For Ella, getting pregnant was entirely unplanned. She and her boyfriend, who was about two years older than her, had what she called "a consensual" sexual encounter. Several weeks later, she discovered she was pregnant. She knew it was X's child (she never disclosed the boy's name to me): he was the first and only boy she had ever had sex with, she told me. She shyly confided that it was not their first time: […] It is not like it was our first-time having sex. We have been in a girlfriendboyfriend relationship for at least seven months or so […]. I knew people will say he forced me, but this is wrong. He did not, we were in love. Here they say that boys always lure girls to sleep with them. But for me, it was not the case. We both wanted it. […] we had sex a few times, maybe four or five if I try to remember. All those times, we never used a condom, but I had not imagined that it would reach to getting pregnant. […] well, after all, I was not so lucky, because I got pregnant. It was a shock for me when I found out.
Ella's parents were unhappy with the news, and they rejected her immediately, refusing to help her and demanding she leave their house: … you know, it is already hard and shameful to be pregnant out of a marriage and not to add that the father is unknown, where the assumptions become then that you are a slut and slept with different people and can't tell who the father is […] my parents didn't want the judgments of other Burundian families who would say they were tolerating my prostitution […] I suffer from their rejection and judgments… Ella further explained that in addition to the shame and judgment that her pregnancy brought to her family, it was clear that by rejecting her, they were also rejecting the responsibility of having an extra mouth to feed. Life in the refugee camp was already so challenging without the added responsibility of a newborn. Ella confessed that the thought of aborting had crossed her mind when she was contemplating the coming responsibilities of raising a child in the harsh economic and health conditions of Mahama; especially given that she could not rely on her boyfriend, who was also not old enough or financially able to take on the responsibilities of providing for her and the baby. Ella nevertheless continued the pregnancy, because the only way to get an abortion was through illegal means, with potentially high health and legal risks (including imprisonment). When I met her, she was taking care of her child alone. The only help she received was from Save the Children, 4 who had stepped in after she had been rejected by her family. They helped her to find a shelter and facilitated her getting her own UNHCR food provision card.
Ella believed that keeping the identity of her 17-year-old boyfriend secret was her only option, since this allowed her to circumvent a legal framework that she believed was unfair: "You see, I was not going to report him because it [the law] is wrong […] he did not force me. And I had to think of the future". Ella spoke about her baby's father fondly; she emphasised that she would not be responsible for him being prosecuted for something they had both consented to. She explained that before the pregnancy, he had always been good to her, they were in love. When he could get a little money from his siblings or family friends, he would help her with small things like body lotion and sanitary pads.
Ella explained that pregnancy among adolescents in Mahama was common, and a common response was to blame -and try to prosecute -the father: One never knows what will happen if the boy who is involved with the pregnancy is known […] some parents of the girls use that to take to court the boy. Well, I am not saying it is wrong for cases when the pregnancy resulted from sexual violence. […] it is simply wrong when the girl consented.
She added that one of the realities in Mahama is that some adolescent girls are willingly involved with older men ('sugar daddies'), whom they expose when they get pregnant. Ella considered this wrong, though she found it especially wrong when the boy involved was himself a teenager. She brought up these other cases to explain that her own situation was not unique: "It is not like these girls are innocents […] they engage in love relationships and sometimes problems like pregnancy arise". When it came to her case, however, Ella explained her decision to keep silent: […] My only consolation is that I keep hoping that maybe in the future my boyfriend will come back and be grateful that I prevented him from jail time and recognise his daughter and support us.
[…] You see, I had to think of what is better for me and my child. It may not be better for now […] but who knows what the future will bring. Maybe in the future, he will become rich and support us financially […] or we could marry.

Silence as Self-care: Pregnant Adolescents and Adolescent…
Ella knew she could create trouble for her boyfriend, but for her it was important that he did not go to jail. She added that by being a single mother, she also had more chances of getting support from Save the Children.
To my question of whether she ever regretted her decision to remain silent, especially after she told me that her boyfriend had run away, she responded: I don't regret anything. Even now that I don't know his whereabouts.
[…] My boyfriend is not here anymore. He run away when I told him that I was maybe pregnant. I don't know why he decided to run away […] I'm not going to judge him, it is possible that maybe because he knew cases of other young boys who went to prison, he too must have been afraid and run.
[…] you see, he didn't know if I or even my family was not going to tell lies about him forcing me into having sex and getting me pregnant […] another thing, it wouldn't have mattered that he didn't force me, he would have been sentenced. Nobody cares about him […] but me, I care, and I would know that it is because of me.
Deliberately keeping silent about the identity of her child's father offered Ella a way to prevent a bad situation from becoming worse, and she also used it to negotiate some potential gains. She took a pragmatic approach to her challenging situation of unplanned pregnancy and tried to turn it into a self-care opportunity. She explained that her situation was not as bad as it could be, in particular because Save the Children, with its focus on helping the most vulnerable, prioritises the needs and wellbeing of single adolescent mothers. Moreover, deliberately keeping silent offered her some relief in that it allowed her to craft imaginaries of potential futures with her boyfriend: living together, him recognising the child, and the possibility of receiving financial help from him.

Ella's Story: not an Isolated Case
Ella's strategy was far from unique. I met other adolescent mothers in Mahama who had also chosen to conceal their child's paternity. For example, Pascaline (16 years old, pregnant adolescent), explained that she had nothing to gain by naming the person who made her pregnant: 'Boys and men who make girls pregnant are prosecuted. I gain nothing by sending him to jail. But maybe by not making his name public I may get something … it may not be now but maybe in the future. If I make it public, things will get complicated for him, especially because the authorities get involved. They make it a case of sexual violence and yet I know I consented. He will be jailed. Then, of course he will blame me, be angry and hate me. Then, of course he will break up with me, which means that there will not be any possibility of being back with him. Even if he gets out of jail, he will not be willing to have anything to do with me and the child at all! […] I think that concealing his name may pay off in the long run.' In Mahama, teenage pregnancy is both common and widely discussed. Mahama residents (of all ages) said they were mortified by the immorality of the girls getting pregnant. For instance, in discussions with camp community members, my research assistant and I would start with general questions about the challenges camp residents faced, and participants often spontaneously talked about the high number of adolescent pregnancies. At the same time, there was a widespread perception among the community that adolescent boys were being unfairly criminalised for engaging in consensual sexual activity. Because of Rwandan law, they said, in most cases the pregnant girls' parents usually take legal action, because they care more about family honour than doing the right thing. The narratives of pregnant adolescents endorsed the view that parents often take legal action when they find out who made their daughter pregnant. For instance, Aurelie (16 years old, adolescent mother) explained that her parents had no other way of knowing the person who impregnated her unless she told them. She opted to lie that she had sex with more than one boy, and thus had no way of knowing who the father was. She added: 'So, if you are trying to do the right thing, which is protecting yourself by protecting your lover, it is better to say nothing. I had to hide our relationship and his identity from my parents and other people, because my parents would have forced him to marry me, which would have been impossible for him as he is also still a boy not a man. And if he failed to do so, they may tell lies about him that he raped me.' These quotes show how in response to the possible legal actions that the parents and the authorities can enforce, pregnant girls had started to refuse to reveal the identity of their baby's father to circumvent what they believed to be an unfair legal framework.
Most of the community respondents also alluded to the practice of adolescent girls willingly getting involved with an older 'sugar daddy'. Like Ella, they critiqued girls who did this and then exposed the man when they got pregnant, taking him to court using the sexual violence script. Moreover, most of the young people I encountered during my time in Mahama voiced their concerns regarding the reality of adolescent boys being criminalised for engaging in consensual sexual activity with their adolescent girlfriends. They wished the law would not be applied blindly, that teenage boys would also be considered children. Others stated that there should be some consideration of the fact that girls could lie and frame a boy. As 17-year-old Louis stated: Impregnating a girl when young is a problem, because one is not a man ready to assume parental responsibility, and that is something that adds to the mental pressure of the young boy […] so already, being a father at an earlier age is wrong and sending them to jail for something they both did consensually is adding wrong to wrong […] running away or relocating is the solution many boys have taken […] you see, going to jail takes away any potential aspiration that the boy might have for a future […] a future that could be successful and benefit his child.
Louis also mentioned that while running away might be the best option available to most boys -besides counting on the mercy or love of their girlfriend to protect them -its success as a strategy was not guaranteed. Indeed, stories of boys who had tried to run away but had been caught were common in Mahama.

3
Silence as Self-care: Pregnant Adolescents and Adolescent…

Discussion and Conclusion
In the wider scholarly literature, silence around sexuality is often described as a social practice informed by cultural taboos. Oftentimes, silence dominates communication on sexuality-related issues between adolescents and parents (Denno et al., 2015;Van Reeuwijk & Nahar, 2013). Parents may be uncomfortable with or even morally opposed to the idea of discussing sexuality with adolescents, and in some cases they do not possess the skills to do so (Bushaija et al., 2013;De Haas & Hutter, 2019;Kamangu et al., 2017). The reticence of parents and other adults to discuss or provide knowledge about sexuality-related subjects to adolescents is also related to how parents and adults understand, construct and approach adolescent sexuality from the lens of abstinence until marriage, where premarital sex is always wrong and harmful. In Mahama, this perspective is dominant.
For adolescents, not speaking about sexuality-related subjects is similarly shaped by the socio-cultural structures in which they live. In contexts where abstinence until marriage is valued, silence about sexuality primarily serves the function of upholding the image that they are not having sex, thus protecting their reputation as good, innocent children, and consequently their family's reputation and honour. The function of silence is thus to maintain the appearance that adolescents are observing societal values and beliefs as expected. Yet once a girl becomes pregnant, the language or illusion of sexual innocence loses its meaning, as does the related silence.
While pregnancy breaks the illusion of abstinence among adolescents, it also brings up new opportunities to retain the illusion of sexual innocence. Where a generalised silence around sexuality is the norm, girls tend to be portrayed as (potential) victims (Kolawole, 2004). Given such a gendered logic, adolescent girls in Mahama can easily claim or be portrayed as victims of sexual violence. The social construction of womanhood and manhood in both the Burundian and Rwandan contexts sits well with this narrative. Girls are often construed as passive and submissive, while boys are seen as active and therefore more prone to violence. Moreover, these gendered logics fit the approaches of development interventions meant to improve adolescents' sexual wellbeing, a discourse that reproduces risky and violent men, while concurrently inviting measures to save the (often young, female) victims (Turner, 2017).
The Rwandan penal code of 02/05/2012 explicitly aims to address a number of important issues regarding child sexual abuse and sexual assault. However, it is not free of the abovementioned gendered and normative biases that are usually invoked regarding adolescent sexuality. It has thus been criticised in terms of how it has been practically implemented to criminalise consenting sexual activities among adolescents. The age of consent provides a systematised assumption of non-consent for all sexual relationships involving persons below the age of 18 and leaves no rooms to define sexual relationships among adolescents in Rwanda other than in terms of defilement and violence. As the perspectives of Ella and the refugee community in Mahama show, adolescents may have sex with peers for all kind of reasons, including love and experimentation (cf. Michielsen et al., 2014).
Legally speaking, there is a complete denial of adolescents' sexual agency. Yet Ella's case shows that adolescent girls who become pregnant are far from passive concerning pregnancy management and negotiating their environment and future (Ngabaza, 2011). For Ella, silence was an active strategy expressing her agency, even in the face of stigma. My findings reveal that Ella and other pregnant adolescents/ adolescent mothers such as Pascaline and Aurelie, in Mahama resist, through silence, the dominant discourse that considers sexual encounters among adolescents as sexual violence (cf. Parpart, 2010). Their practice of silence is a powerful manifestation of agency, intended to open up possibilities for greater personal control over their situation in an otherwise uncertain context, both for the present and a hoped for future (cf. Pincock, 2017;Van Reeuwijk & Nahar, 2013).
More specifically, I suggest that among Ella's motivations to stay silent is an attempt at self-care and care for her baby. This echoes Cense's, (2019) findings of how looking after oneself enters into the equation as young women appraise their decision-making around teenage pregnancy. While for Cense's interlocutors, looking after themselves relates to avoiding getting pregnant or their decision to seek an abortion, for my interlocutors, it was in relation to the decision not to disclose the identity of their child's father. While previous academic literature on sexual violence points to the dynamic of financial dependency (e.g. Gerver, 2013, who describes how perpetrators bribe their victims to stay silent after rape), Ella's case as well as Pascaline and Aurelie is different; they were not explicitly pressured by their boyfriend to remain silent (though this of course does not dismiss other dependencies that may exist). In weighing their choices, however, Ella, Pascale and Aurelie did not see any benefits of naming their boyfriends; rather, they saw clear risks. Additionally, Ella explained that by remaining silent, she would fall in the category of single adolescent mother, which would enable her to seek support from Save the Children. Such support is especially important in Mahama, as her own family and that of boyfriend, like most families in the refugee camp, were struggling to make ends meet.
Another reason why Ella wanted to save her boyfriend from jail related to an important normative kinship obligation prevalent in the community in general. In the long-term, keeping silent allowed Ella, and other pregnant adolescents/ adolescent mothers like her, to craft imaginaries of potential normative futures with their boyfriends; she nurtured hopes that he would potentially marry her, would recognise the child and take on his paternal responsibilities. A future in marriage with the father of her child (and the one who took her virginity) would allow for a return to and reaffirmation of the norm that had been broken by Ella's adolescent pregnancy. This echoes Van Eerdewijk's, (2009) research in Senegal, which found that girls are active agents, and in as much as they can reproduce and reinforce dominant discourse around their sexuality, they can also find opportunities to manoeuvre and deviate from it. Such an understanding of girls' agency sheds light on the coexistence of silence as self-care while maintaining the status quo.
Furthermore, in a context like Mahama where everyone believes girls to be innocent and men to be perpetrators, pregnant girls realise that their story of love will not be heard. Their silence is thus born from a restricted power, yet it allows them to reaffirm their position. For instance, Ella made it clear that it was important to her 1 3 Silence as Self-care: Pregnant Adolescents and Adolescent… to do the right thing. At the same time, this sense of justice and moral ethics shows that, from her standpoint, caring for others leads to self-care. Ella's case illuminates Surrey's claim (1985) that for women, the primary experience of self is relational. Since Ella's relationship with her family, and especially her parents, had broken down following her pregnancy, the only remaining relationship of value that she nurtured was the one with her boyfriend. Her decision for silence was informed by her sense of responsibility for others (her boyfriend, her child), however this intersected and coexisted with considerations of her own self-interest.
While the legal age of consent in Rwanda is meant to protect adolescents, it has unintended negative consequences. The adolescent boy who experiments consensually with love and sex may be labelled a sexual predator. Furthermore, the law forces pregnant girls to carry the burden by themselves and forgo asking for help from the baby's father and his family. Moreover, the pregnant adolescents' desire to protect her lover who impregnated her from the law through silence, also allows the boy/ young man to not have to deal with the heavy responsibilities of parenthood. One may wonder how this phenomenon affects the increasing number of females headed households in African societies, including Burundi (Courtois, 2016). In addition, though, for now, pregnant adolescents can seek support from NGOs like Save the Children, there is no guarantee that institutional help will be available in the long run. Moreover, as I have shown elsewhere (Ruzibiza, 2020), this institutional support does not fully cover all of the pregnant adolescents/adolescent mothers' needs; for example, in the absence of childcare (which would usually come from extended family support structures), many cannot return to school. Consequently, the chances are small that the pregnant adolescents/adolescent mothers will finish their education, and their children will likely grow up fatherless.
While the law concerns everyone in Rwanda, including Burundian refugees in Mahama, the unintended negative consequences appear exaggerated in the camp, where there is very little room for the families of pregnant girls and their boyfriends to find alternative ways of dealing with the situation. Mahama is very densely organised, leading to little privacy among camp residents. Everyone knows and sees what others are doing, and it is hard for pregnant girls to pass unnoticed. Moreover, the presence of the camp administrators, as well as local and international NGOs and UN organisations providing assistance, is felt by everyone in the community. In this context, certain pressing problems are quickly acted upon, as the camp administrator might fear exposure and external scrutiny. It is unsurprising then that in Mahama, where adolescent pregnancies are increasingly common, the camp administrator was making efforts to 'fix' the problem, which in turn made such pregnancies even more visible.
To summarise, pregnant adolescents/adolescent mothers' silence about their child's paternity is, I argue, best understood as a strategy for relational self-care. Such silence shares similarities with other forms of silence found in the literature, such as silence as a coping and survival strategy (Lorde, 2007;Parpart, 2010), silence as resistance (Gal, 1991), and silence as withholding information (Bakuri et al., 2020). Pregnant adolescents/adolescent mothers keep silent to improve their own and their child's current material situation, and nurture hopes of future marriage and support. And, despite having transgressed societal norms, they feel proud to be doing 'the right thing'. Given that the future outcome of this strategy of concealing the name of their child, as envisioned by the pregnant adolescents may be a form of wishful thinking, long-term research can help uncover what happens to these mothers and their children in the long run, especially when their hopes do not materialise.

Limitations of the Study
Although the research study yielded contextualised findings relating to the strategic negotiations of adolescent girls confronted with an unintended pregnancy, some limitations are worth mentioning. One notable limitation was my inability to document the perspectives of adolescent fathers due to the challenges of identifying and locating them. Another is that at the time of fieldwork, there was much gossip among adolescents and other camps residents pertaining to scandalous suspicions that some NGO staff members were romantically involved with young girls, and that some had pressured the girls for abortions. This situation resulted in NGO staff being very hostile towards me, which made it impossible to gain a deeper understanding of their perspectives on the issues at hand. Notes 1. For this paper, I refer to the old law, as it was the one in application during the fieldwork. Currently applicable is Law No. 68/2018 of the penal code of 30/08/2018; Article 133 relates to defilement. Important changes were made in the 2018 law to introduce reforms and improve age sensitivity in rape trials. 2. Such household variations were due to the political situation in Burundi, where not everyone faced similar levels of persecution. It was thus quite common to find that only a few members of a family had fled while others remained. 3. Middle-class here refers to their social class before arriving at Mahama. 4. Save the Children, an International NGO with a focus on child protection services, was one of the NGOs operating in the camp. Silence as Self-care: Pregnant Adolescents and Adolescent… are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.