Recruitment of child soldiers in Nepal: Mental health status and risk factors for voluntary participation of youth in armed groups

Abstract

Preventing involuntary conscription and voluntary recruitment of youth into armed groups are global human rights priorities. Pathways for self-reported voluntary recruitment and the impact of voluntary recruitment on mental health have received limited attention. The objective of this study was to identify risk factors for voluntarily joining armed groups, as well as the association of conscription status and mental health. In Nepal, interviews were conducted with 258 former child soldiers who participated in a communist (Maoist) revolution. Eighty percent of child soldiers joined ‘voluntarily’. Girls were 2.07 times as likely to join voluntarily (95% CI, 1.03–4.16, p=0.04). Among girls, 51% reported joining voluntarily because of personal connections to people who were members of the armed group, compared to 22% of boys. Other reasons included escaping difficult life situations (36%), inability to achieve other goals in life (28%), and an appealing philosophy of the armed group (32%). Poor economic conditions were more frequently endorsed among boys (22%) than girls (10%). Voluntary conscription was associated with decreased risk for PTSD among boys but not for girls. Interventions to prevent voluntary association with armed groups could benefit from attending to difficulties in daily life, identifying non-violent paths to achieve life goals, and challenging the political philosophy of armed groups. Among boys, addressing economic risk factors may prevent recruitment, and prevention efforts for girls will need to address personal connections to armed groups, as it has important implications for preventing recruitment through new methods, such as social media.

BACKGROUND

Youth throughout the world become engaged in armed groups through a range of pathways. Elucidation of these pathways is crucial for prevention measures. Moreover, the pathways by which youths become combatants may have important implications for their mental health needs after returning to civilian life. Practices of forced conscription and subsequent negative mental health outcomes are well documented (Betancourt et al., 2013). There has been less research on youth who self-report as ‘voluntarily’¹ joining an armed group. Our objective is to explore how pathways of recruitment with different levels of voluntary participation may relate to different forms of youth vulnerability, especially gender, and may have different implications for mental health needs.

The 2007 Paris Principles define “children associated with armed forces or armed groups” (CAAFAG) as,

any person below 18 years of age who is or who has been recruited or used by an armed force or armed group in any capacity, including but not limited to children, boys, and girls used as fighters, cooks, porters, messengers, spies, or for sexual purposes. It does not only refer to a child who is taking or has taken a direct part in hostilities (UNICEF, 2007, p. 7).

For the purposes of this paper, we will employ the more commonly used term “child soldier” instead of CAAFAG, and we use the Paris Principles definition of CAAFAG as our operational definition of child soldiers.

Studies throughout the world on child soldiers have tried to explain the causal pathways of child soldiering through various lenses (Beber & Blattman, 2013; Betancourt et al., 2013; Brett & Specht, 2004; Chen, 2014; Harendra de Silva, 2013; Schauer & Elbert, 2010; Shakya, 2011; Stevens, 2014; Wessells, 2006). Chen’s (Chen, 2014) work in Myanmar categorizes three level of relationship between a child and a recruiter that facilitates participating in armed forces: (a) victim—coercer relationship; (b) patron—client relationship; (c) comradeship relationship.

The Victim-Coercer relationship is a result of forced recruitment where children are bound to be a part of the armed force through abduction or creating an environment of forced involvement. In Uganda, for example, the Lord’s Resistance Army (LRA) carried out large-scale and indiscriminate forced recruitment by raiding homes. It is estimated that LRA led by Joseph Kony abducted around 60,000 to 80,000, mainly male adolescents (Beber & Blattman, 2013).

Chen describes the emergence of the Patron-Client relationship through an environment where children have no other options for future economic or educational development, and hence the children see soldiering as an important measure to gain access to resources and power. In countries hit by conflict, children are often deprived of access to basic needs and joining an armed group increases access to these resources. During the civil war in Sri Lanka, joining the armed group would be the most attractive and the ‘rational’ action for poor children with no prospect of accessing education, employment and other opportunities (de Silva, 2013). In conflicts where children have lost their parents, joining an armed group may also provide the service of a form of guardianship. In some context, joining an armed group not only means access to food and shelter but also to getting power and respect. Stevens ( 2014) found that girls throughout the world voluntarily join armed forces to flee abusive domestic environments, forced marriages, and other forms of gender oppression. In this relationship, the armed group provides a service to youth for them to achieve other life goals.

The Comradeship relationship involves children participating in the armed groups voluntarily through indoctrination or as an act of revenge for killed or injured family and friends. Studies conducted in multiple conflict regions throughout the world have highlighted that children are attractive for this type of recruitment because of the ease with which by which they get indoctrinated, mobilized and retained (Beber & Blattman, 2013; Chen, 2014; Harendra de Silva, 2013; Schauer & Elbert, 2010). A study in Sri Lanka showed 18 out of 19 child soldiers joined voluntarily through this mechanism, while a similar study in El Salvador saw that 73.7% of child soldiers joined voluntarily through comradeship (de Silva, 2013) (Betancourt et al., 2013). In Uganda, children who had lost loved ones or had been tortured joined through comradeship to seek revenge (Schauer & Elbert, 2010).

Comradeship relationships are also created through family and community ideological indoctrination as children are growing up. In Sri Lanka, the Liberation Tigers of Tamil Eelam (LTTE) established a training base for recruits under 16 years of age (referred as “Tiger cubs”) teaching them non-military and military training courses and later using them in wars against the Sri Lankan state forces. These children were heavily indoctrinated with the idea of dying/killing for the sake of their country and deaths were glorified as martyrdom (de Silva, 2013). In the Democratic Republic of the Congo (DRC), for the children of Mai-Mai groups, joining the armed forces was related to the social ethics and values of their communities. It was considered a moral code for them to fight for a higher cause – defending their community; Hoffmann (2010) in his study of this group argues that for them, getting involved in the conflict was the “means to liberate the DRC nation-state from foreign military aggression which they believed threatened their way of life, if not their physical survival,” (p. 355).

These framings in terms of relationships are helpful to elucidate different pathways to association and to develop strategies to reduce risk of association. It is important to keep in mind that some pathways to association may reflect elements of all of three relationships types. Moreover, the relationship categorization will likely differ based on the perspective of who is doing the categorization, and the context in which information on participation is gathered. Youths may be inclined to present a narrative focused more strongly comradeship in some contexts (e.g. in times of active conflict) but reflect the same narrative under a lens of victim-coercer in other contexts (e.g. if disappointed when post-war promises are not fulfilled). Parents ideologically supporting a movement may report that their children joined out of comradeship whereas parents with other ideological positions may frame association in regard to victim-coercer. Most nongovernmental organizations and human rights advocates might frame all association of persons under 18 years of age in a victim-coercer relationship and claim that children’s military association never represents a voluntary association through pathways such as comradeship. Therefore, categorization of association pathways should always clearly state from whom and under what context information regarding association was gathered.

In this study, we describe the process by which children (persons under the age of 18 years) joined armed groups during a decade long civil war known as the People’s War in Nepal from 1996–2006. We are specifically interested in motivations for patron—client and comradeship relations between armed group groups and associating youth. To understand the process by which children voluntarily associated with armed groups in Nepal a brief political history and description of context—especially women’s position in society—is crucial.

From 1996 through 2006, the Communist Party of Nepal (Maoist), CPN (M), battled the Hindu monarchy of Nepal (and the national army it controlled) to establish a federal republic government. More than 17,000 people were killed, with the majority of deaths at the hands of the Hindu monarchy’s Royal Nepal Army and the government’s (armed) police forces (INSEC, 2008). The armed conflict formally ended in 2006, when the CPN (M) signed a peace treaty with political parties in the government opposed to royal rule, which led to the inclusion of the CPN (M) in the national government. The Hindu monarchy was ended and Nepal became a federal republic. During elections in 2008, the CPN (M) won a two-thirds majority but later abandoned their top government positions (Adhikari, 2014). Since 2008, there have been numerous changes of leadership between political parties with CPN (M) leaders occasionally holding the position of prime minister. During the ten-year period from 2006–2015, the government has been largely dysfunctional, with widespread corruption in the health and education sectors (Norad, 2011).

During the armed conflict, children were recruited into the CPN (M)’s People’s Liberation Army (PLA) and the Royal Nepal Army as soldiers, sentries, spies, cooks, and porters (Human Rights Watch, 2007; United Nations, 2006). Local groups estimate that at the conclusion of the armed conflict approximately 9,000 members—one-third of the PLA—comprised fourteen to eighteen year-olds, with 40 percent being girls (Human Rights Watch, 2007). Ten percent of the Royal Nepal Army during the conflict was below the age of eighteen (Singh, 2004). Recruitment included victim—coercer, patron-client, and comradeship relations.

Victim-Coercer: in Nepal, one form of Maoist recruitment was a policy of “One House, One Person” where at least one person from each family was obligated to join their People’s Liberation Army (PLA) (Shakya, 2011). Adolescents were often sent because they were unmarried, though families could often pay the Maoists to not take someone from the household if the family had adequate economic means (Kohrt et al., 2010c).

Patron-Client

Prior studies have highlighted patron-client relationships especially for adolescent girls escaping abusive situations including domestic violence, forced child marriage, and other gender-based violence (Shakya, 2011). Women and girls experience high levels of abuses and rights violations, as illustrated by Nepal’s rank at 121st of 136 countries on the Global Gender Gap Index, (World Economic Forum, 2013). Nepal’s Gender-related Development Index (GDI) score, a measure of equality between women and men, is 0.545, among the lowest in the world (UNDP, 2009). Thirty-one percent of women report experiencing inter-personal violence (Bennett et al., 2008). Youth also joined Maoists as a form of access to education, which they could not get in their home communities because of poverty or lack of parental support for schooling (Morley & Kohrt, 2013). Joining for education was especially salient for girls, many of whom reported that their parents sent sons to school but not daughters (Kohrt et al., 2010c). Joining the Maoists was presented as an opportunity to challenge the caste hierarchy in which political and economic resources were controlled by high caste groups: Bahun (priest caste) and Chhetri (warrior/royalty caste). Youth from Dalit/Nepali (occupational castes) and other ethnic groups (referred to as Janajati, which include other Hindu groups, Buddhists, Muslims, and Kirantis) were encouraged to join Maoists as a way to improve their status within Nepali society (Adhikari, 2014).

Comradeship

For youth who were not escaping immediate violence or seeking services such as education or care for basic needs, the Maoist ideology was used to foster association through comradeship relations. Children reported to be attracted to Maoist political ideology including Maoist claims to fight for eliminating caste discrimination, promoting women’s rights, and ending India’s exploitation of Nepal’s resources (Shakya, 2011).

In this paper, we examine the pathways by which youth become child soldiers through either forced conscription or self-reported voluntary association with the People’s Liberation Army of the Communist Party of Nepal (Maoist). Given the political ideology of the Maoist movement in Nepal, we examine how gender and caste/ethnicity predict voluntary vs. involuntary association. We also examine the relationship of pre-conscription trauma with voluntary vs. involuntary association. Ultimately, this analysis is intended to improve identification of populations of youth who are more likely to join violent political groups voluntarily.

METHODS

In 2007, we employed criterion sampling to select 258 child soldiers for this study from 8 districts across Nepal. All districts were participating in Unicef-sponsored reintegration programs for child soldiers (Kohrt et al., 2015b). The Nepali NGO, Transcultural Psychosocial Organization (TPO) Nepal implemented the formative research with child soldiers and intervention development and implementation from 2006–2009 with follow-up studies in 2012. The study comprised a combination of qualitative and quantitative methods to understand the mental health and psychosocial consequences of children’s participation in armed groups. Data presented here include pre-intervention data on types of conscription and associated variables. Data were gathered by a Nepali research team employed by TPO Nepal with a background in field research who received a month-long training on qualitative and quantitative data collection as well as on the ethics of research with vulnerable children.

The qualitative component of the larger study included participatory approaches (with a technique known as Child Led Indicators (CLI) in which children developed their own psychosocial indicators of distress and wellbeing), narrative focus group discussions (N=25 groups) with children and community members, key informant interviews (N=152) with children and community members, and case studies (N=20) of child soldiers (Karki et al., 2009; Kohrt et al., 2010c; Kohrt, 2015; Morley & Kohrt, 2013). Case studies were selected through partner nongovernmental organizations (NGOs), community health workers, or leaders in the community. They were asked to identify one child soldier who was having a difficult time reintegrating and one child soldier who was reintegrating well. Each case study comprised one child soldier coupled with at least one individual from the following groups: a family member, a community member relevant to that case, and/or another individual who could speak about the child’s particular adjustment following service in the Maoist Army.

The 258 participants in the quantitative study were former child soldiers who returned home from the Maoist Army in Nepal (Kohrt et al., 2015a). Criterion selection criteria for the target population included at least one month of participation in an armed group, younger than 18 years during the enrollment in the study, and having a consenting caregiver. The first 30 former child soldiers enrolled for NGO services in each district were invited to participate in the study. By the aforementioned selection criteria, a total of 258 former child soldiers, including 159 male participants and 99 female participants, were selected for this study. The average age of the total participants were 15.59 years; standard deviation was 1.35 years. The age range was 11–17 years old.

The participants completed 60 to 90 minute personal interviews; the caregivers of these former child soldiers did not participate in the interview but provided their consent. Due to high illiteracy rates, research assistants read questionnaires in Nepali to the children. Demographic characteristics such as sex, age, education level, ethnicity, and religion were documented. Ethnicity was included as part of the study as it has been associated with one’s mental health condition in Nepal. For example, members of low-caste groups have 2–4 fold greater odds of psychological distress, depression, and anxiety when compared with high-caste groups in Nepal (Kohrt, 2009; Kohrt et al., 2009).

A 37-question semi-structured interview was developed based on the findings of the qualitative research. Questions address reasons for association with the armed group; details of roles and activities during association; health, social, and economic conditions prior to, during, and after association; educational experience before, during, and after association; and use NGO services, health care, and traditional healing and rituals before, during and after association.

The main independent variable of the study was voluntary conscription. The variable was based on their self-reported responses during the interview for the questions “Who was the decision maker with regards to joining the armed forces?” Among the 3 choices, “All decisions were mine” and “Some decisions were mine and some were others” were treated as voluntary conscription and “All decisions were other” were recoded as forceful conscription. Participants were also asked, “What was your reason for becoming associated with an armed group?” In addition, exposure to traumatic events refers to a sum of participants’ experiences to ten categories of traumatic events (experience of fire, violence or death, beating, bombing, abduction, torture, murder of a family member, parents fighting, physical abuse by parents, and disappearance of relatives).

The socio-demographic factors, sex, age, caste, literacy, as well as exposure to traumatic events, were tested to see if those factors have any correlation with voluntary conscription of former child soldiers in Nepal. Initially, each factor was separately evaluated by appropriate bivariate analysis methods, and then all the demographic characteristics were included in a multivariate logistic regression model.

In addition, we evaluated the association between voluntary vs. involuntary conscription and mental health outcomes. The primary mental health outcomes were Depression Self-Rating Scale (DSRS), Child PTSD Symptom Scale (CPSS), and Child Functional Impairment scale (CFI). The DSRS for children is an 18-item scale assessing depression symptoms on a “Mostly”, “Sometimes”, or “Never” scale that corresponded with item scores of 2, 1, or 0 (Birleson, 1981). The range of possible scores on the DSRS is from 0 to 36. The CPSS (Foa et al., 2001) is a 17-item scale corresponding to DSM-IV-TR symptoms of PTSD that range from 0 to 68. The DSRS and CPSS have been translated and validated in Nepal and have good psychometric properties (DSRS, area under the curve (AUC)=0.82, sensitivity=0.71, specificity=0.81, cutoff score ≥14; CPSS, AUC=0.77, sensitivity=0.68, specificity=0.73, cutoff score ≥20) (Kohrt et al., 2011). For the current analyses we employ continuous scores for these outcomes. The CFI is a 10-item scale (range: 0 to 30) locally developed to assess impairment in daily functional role expectations according to cultural expectations for children in Nepal (Kohrt et al., 2010b).

Cross-tabulations and independent t-tests were used to evaluate group differences in gender, recruitment type, and mental health outcomes. Hierarchical linear regressions were conducted to evaluate predictors of recruitment type and mental health outcomes. Statistical analyses were performed with SSPS version 22 (SPSS Inc., 2013). All interviews were translated into English and analyzed using Atlas.ti with a codebook developed by three independent coders (inter-coder agreement: percent agreement 0.90, Cohen’s κ 0.82). A statistical significance level of p <0.05 was used. All qualitative and quantitative analyses were performed by collaborative teams including the Nepali research team of TPO Nepal, and foreign researchers including two psychologists and a psychiatrist-anthropologist, all of whom had spent more than a decade working in Nepal. For additional details on methods see (Kohrt et al., 2008; Kohrt et al., 2010b; Kohrt et al., 2010a; Kohrt et al., 2015a; Morley & Kohrt, 2013). The research was approved by the institutional review board of Emory University, Atlanta, USA, and the Nepal Health Research Council, Kathmandu, Nepal.

RESULTS

Risk factors for voluntary association with armed groups

There were more male participants (61.6%) than female participants (38.4%), and most participants self-identified as literate (see Table 1). Approximately half of the study participants reported that they belong to a low caste, and approximately 25 percent of the participants self-identified as high caste members. Among the sample, 205 children (79.5%) reported that they joined the Maoist armed force voluntarily and 53 of them (20.5%) described that their participation resulted from forceful recruitment.

Table 1.

Demographic characteristics of former child soldiers (n=258)

Characteristics
Age (year)	Mean ± SD (95% C.I.) 15.59 ± 1.35 (15.42 – 15.75)
Sex No. (%)
Male	159 (61.6%)
Female	99 (38.4%)
Literacy No. (%)
Literate	246 (95.3%)
Illiterate	12 (4.7%)
Caste No. (%)
High Caste	66 (25.6%)
Ethnic minority/Janajati	62 (24.0%)
Low Caste	130 (50.4%)
Exposure to trauma before association (Number of types of traumatic events)	Mean ± SD (95% C.I.) 1.85 ± 1.41 (1.68 – 2.03)
Association to Maoist Army
Voluntary association	205 (79.5%) [Boys 120 (75.5%); Girls 85 (85.9%)]
Forced conscription	53 (20.5%) [Boys 39 (24.5%); Girls 14 (14.1%)]

Female former child soldiers in Nepal were more likely to have joined the military voluntarily than males: 85.6% of former female child soldiers reported that they joined voluntarily compared with 75.5% of former male child soldiers. Involuntary conscripts were twice as likely to be boys (n=39) compared to girls (n=14) in this sample (OR=1.97, CI=1.01–3.86, p=.04). Higher age showed a positive correlation with voluntary association at a level of borderline significance (OR (Exp(B))=1.24; p= 0.053). When placed in a multivariate model, both female sex and higher age associated with voluntarily joining the Maoists (see Table 2). Caste, literacy, and pre-conscription exposure to traumatic events did not associate with conscription type. Text Box 1 provides narratives of two girls who reported voluntarily joining the Maoists.

Table 2.

Bivariate and multivariate analyses of the association between involuntary conscription and risk factors (n=258)

	Bivariate		Multivariate
	OR (95% CI)	Sig.	OR (95% CI)	Sig.
Age	1.24 (0.99–1.53)	0.053	1.26 (1.01–1.58)	0.042*
Sex (female)	1.97 (1.01–3.86)	0.045*	2.07 (1.03–4.16)	0.042*
Literacy	0.34 (0.04–2.69)	0.47	0.44 (0.05–3.64)	0.45
Caste^
High Caste	0.95 (0.48–1.88)	0.88	1.17 (0.56–2.45)	0.67
Low Caste	1.46 (0.69–3.11)	0.32	1.31 (0.58–2.97)	0.51
Exposure to trauma before association	1.00 (0.81–1.24)	0.98	1.00 (0.80–1.24)	0.98

^*p <0.05,

^{^}Reference group for caste was ethnic minority/Janajati status.

TEXT BOX 1. Narratives of female former child soldiers in Nepal (Kohrt et al., 2010c).

I was thirteen years old when I joined the Party. I was born into a poor family. We just have this much [points to surroundings], nothing more. I was a very good student [but] after I took my exams for fifth grade my parents told me: ‘We are very poor, we have no money so you have to leave school and take care of your brothers and sister.’… Then, I left school. My parents wanted to save their money so they could send my brother to school. The Party women came to my village and invited me to a cultural program. Both sons and daughters should be treated equally, the Maoist leaders said. Husbands and wives should work together, too… From that day, I didn’t want to go back to my house. I joined the Party.

;- ‘Asha’, female former child soldier who joined the People’s Army at 13 years old

I was thirteen years old. I was a very shy girl who wouldn’t speak with people other than my mother… In our village, people used to come to ask my hand in marriage even when I was very young. Even a mention of marriage gave me a headache. I hated it! I wanted to avoid marriage in any way possible… I have a slightly older friend in the village. Against her will, her father married her off at a young age. She was miserable. She often said she would go to India or join the Maoists. She would die there, if need be. At least, she would be free from marriage. Like her, liberation was what I needed. At the time, many Maoist activities used to take place in our village… I liked their cultural program. Very entertaining! What a wonderful life—I would often think—I would have if I became a Maoist. I would get to travel a lot and wouldn’t have to be forced into marriage. Besides, I would also have a lot of friends!

- ‘Shova’, female former child soldier who joined the People’s Army at 13 years old

We examined the reasons for association separately among former girl soldiers and boy soldiers (see Figure 1). Among girl soldiers, having social relationships with people who are Maoists was the most common reason endorsed for voluntary association, endorsed by 51% of girls. Among boys, the appeal of the Maoists philosophy and escaping personal distress were endorsed by 38% and 34% respectively. Both boys and girls endorsed some reasons equally e.g., escaping personal distress (34% boys, 38% girls) and wanting to improve the country (24% boys, 24% girls). There were statistically significant gender differences in some responses such as liking people who are Maoists, endorsed by 51% of girls compared to 22% of boys (OR=3.82, 95% CI 1.98–7.41, p<.001 ). Conversely, joining Maoists because of poor economic conditions was more frequently endorsed by boys (23%) than girls (11%), (OR=2.49, 95% CI 1.04–5.97, p=.04). Girls were more likely to join because of inability to achieve other goals in life compared to boys (34% vs. 24%); however, this difference was not statistically significant. There were no differences among caste/ethnic groups for reasons for voluntary association (p>.05).

Figure 1.

Reasons for joining Maoist among voluntarily-associating child soldiers

Type of conscription and mental health outcomes

In the total sample, the conscription method did not demonstrate bivariate associations with any of the three mental health outcomes of interest (p>.05). Risk factor analysis for conscription found that gender was significantly related to voluntary association and hence gender-stratified analyses were conducted for mental health outcomes. Among male former child soldiers, conscription type showed a statistically significant association with PTSD like symptoms: male former child soldiers who voluntarily joined the Maoist expressed lower CPSS scores whereas male former child soldiers who were forcefully conscripted expressed comparatively higher CPSS scores (see Table 3). The mean difference between the voluntary associated male group and the involuntary conscripted male group was 3.5 points (see Figure 2). Conversely, there were no differences in mental health outcomes depending upon the conscription methods among female former child soldiers in the study.

Table 3.

Multivariate analysis of male former child soldiers’ PTSD (CPSS) and conscription status

Variable	B	Sig. of variable	R2	ΔR2	Sig. of model
Step 1			0.03		0.045*
Voluntary	−3.50	0.045*
Step 2			0.05	0.02	0.14
Voluntary	−3.42	0.054
Age	−0.89	0.13
Literacy	1.23	0.78
High caste	0.71	0.68
Low caste	2.59	0.21
Step 3			0.20	0.07	0.000*
Voluntary	−2.09	0.21
Age	−0.90	0.10
Literacy	2.36	0.56
High caste	0.08	0.96
Low caste	0.35	0.86
Trauma (Pre)	1.76	0.001*
Trauma (During)	2.36	0.000*
Trauma (After)	1.57	0.08

^*p <0.05

Figure 2.

The average scores of PTSD symptom severity by conscription methods (Voluntary vs. Involuntary)

DISCUSSION

A growing body of research into the experiences of former child soldiers and mental health consequences has illustrated that contextual grounding is crucial to identify who are most negatively affected by the experiences and in need of services (Betancourt et al., 2013). Our findings support this by demonstrating gender differences for both recruitment pathways and mental health status for former child soldiers in Nepal. In the context of a communist revolution in Nepal, we found that 80% of former child soldiers identified as joining voluntarily, with girls more likely than boys to report joining voluntarily. The finding that such a large portion of youth joined voluntarily is likely associated with the populist appeal of the Maoist movement and its assertions to address social inequities in Nepal (including caste, gender, and rural-urban inequities) (Adhikari, 2014; Kohrt et al., 2010c). The Maoists highlighted the government and monarchies failure to provide resources in most rural parts of the country, especially Western Nepal. Other contexts where youth have reported high rates of voluntary association are also characterized by strong political opposition groups with popular support in some regions of the countries, e.g., El Salvador, Sri Lanka, and Mozambique (de Silva, 2013; Dickson-Gomez, 2002; Santacruz & Arana, 2002; West, 2004). This contrasts with context with high rates of forced conscription where conflicts have been characterized by political instability, rapidly shifting warring factions, or ongoing dislocation of groups throughout the region, e.g., regions of West and East Africa (Akello et al., 2006; Betancourt et al., 2013).

There was a difference by gender in the percentage of boys versus girls self-identifying as voluntary recruits and a difference by gender regarding the reasons given for joining voluntarily. Maoist doctrine and recruitment emphasized the marginalization of women. Maoists’ recruitment strategies highlighted dissatisfaction among women with traditional norms and behavioral restrictions (Sharma & Prasain, 2004)., For example, Maoists—including Maoist women—emphasized that girls were safer from sexual violence when part of the armed group than when in the community. Qualitative research supports this, with the finding that girls were vulnerable to sexual violence from police, the Royal Nepal Army, and from relatives and community members (Kohrt et al., 2010c). Within the Maoists there was reportedly strict enforcement of only consensual sexual relationships, and Maoist men who raped women were publicly punished. Moreover, girls reported that joining the Maoists allowed them to escape child marriages and relationships with marital rape. This contrasts with depictions of child soldiers in West and East Africa where association with armed groups has been characterized by sexual slavery and becoming “bush wives,” (Betancourt et al., 2013). Participation in Mozambique, however, echoes findings from Nepal in that girls joining FRELIMO’s (Frente de Libertação de Moçambique) female detachment reported combat as empowering and an escape from patriarchal societal structures (West, 2004).

Regarding the gender differences in association, the importance of personal relationships among girls was striking. Half of the girls who joined voluntarily said that a personal connection was an important reason for joining. One example of this was girl soldiers returning to their villages to recruit friends and neighbors (Kohrt et al., 2010c). Similarly, girls discussed the approach of making friends in communities where they were recruiting, causing other girls to join as a result of these friendships. Older female role models among the Maoists were an important recruitment factor. Personal relationships were important only for one out of five boys who voluntarily associated. The role of these social relationships in recruiting youth requires further investigation across armed conflicts. Access to education also played a role. Among girls, the inability to pursue education because of gender norms favoring educating boys was a common reason to join (Morley & Kohrt, 2013). Girls said that joining the Maoists was the best alternative to being able to get an education.

We had hypothesized that voluntary recruitment may be protective for mental health outcomes among former child soldiers. However, this was not supported for all youth. Among girls, there were no differences in levels of PTSD symptoms, depression symptoms, or functional impairment based on type of conscription. Among boys, we did find that PTSD scores were lower among voluntary recruits. Taken together, these findings on mental health outcomes do not suggest that need for mental health and psychosocial support would necessarily vary based on why and by what pathway a youth became involved with the Maoists.

Limitations

As mentioned in the introduction, youth likely respond differently to different audiences when asked how they became associated with armed groups. Perceptions of benefits or potential punishment for specific answers may bias youths’ responses, as may the passing of time and changing perspectives. Moreover, responses of parents, teachers, and friends may differ from those self-reported by the former child soldier. Therefore, the current findings should be taken in light of specifically using self-report in an immediate post-conflict context. That being said, this is an important time and methodology because it is the approach most frequently used when evaluating youth post-conflict for reintegration services, as was the case for the context of this study.

In considering generalization of these findings, it is important to consider specific limitations of this study. As with the majority of child soldier research, it is challenging to consider what constitutes a representative sample. Within the context of Nepal, our findings are representative of child soldiers who have engaged with NGO services. Child soldiers who have hidden their identities as former combatants, or who were not present in cantonment areas, are not represented here. Reasons for joining the Maoists and post-conflict mental health status of other groups of former child soldiers might differ from the findings of the current study. In addition, the current findings do not reflect the status of former combatants who joined as children, but who were over 18-years of age when the armed conflict ended. Individuals who joined as children were not provided with NGO services through the organizations participating in this study. Gender differences as described here are also likely to be highly specific to the nature of the conflict and the status of women in civilian communities.

Applications

A broad implication of these findings is that reintegration programs should take into account the variation in reasons and factors that lead to recruitment. Reintegration programs can be strengthened if they address the reasons why children associate with armed groups, which provides a signal to what children value and the types of recruitment push and pull factors in the community. Addressing these push and pull factors, for example with attention to the different relationships that made conscription more likely, is important because such relations may continue to be present in youth’s lives after demobilization and make them vulnerable for other forms of exploitation. For example, in Nepal some former child soldiers became part of political groups that employed violence during subsequent elections.

The study also has important applications for understanding gendered recruitment pathways. In contexts ranging from West Africa to East Africa to Syria and Iraq where ISIS is active, girls have been violently recruited into being sex slaves for combatants (Arango, 2014; Pham et al., 2009) and there are increasing media reports of girls joining voluntarily as well (Mosendz, 2014a; Neer & O’Toole, 2014). In these settings, protection of women against forcible recruitment by armed forces and groups should be the top priority. However, as the situations in Nepal, Sri Lanka, Mozambique, and some other settings have illustrated measures are also needed to prevent girls joining voluntarily. In these settings, the social pathways by which girls join and alternative, non-violent social opportunities need to be explored.

This is relevant currently with the global recruitment of young women into ISIS through social media channels (Mosendz, 2014a; Neer & O’Toole, 2014). The Institute of Strategic Dialogue, an organization that helps in managing the largest database of female travelers, estimated that most of the women and girls who travel from Western countries to Middle East to join ISIS are very young. In American and European countries, girls from both Muslim and non-Muslim families have become connected to ISIS through social media with messages tapping into their frustrations and vulnerabilities (Bennhold, 2015). Keeping the role of social relations in mind is crucial for reintegration when girls are returned from armed groups. The ability to form new positive, non-violent social relationships, as well as girls’ frustrations and vulnerabilities that are often only addressed by social relations on a micro scale should be a major priority.

For boys, and in part for girls, economic drivers are also important. Within the Patron-Client relationship, youth may join armed forces as form of employment, either to receive basic needs or to be paid in exchange for military services. Within ISIS controlled regions, some children are paid for coming to school, where they are then taught religious education alongside weapons training (Mosendz, 2014b) This is followed by payment of approximately $100 per month for military services according to Human Rights Watch (2015). It is not surprising that reintegration programs for former child soldiers across many world regions have focused on economic support. However, studies in Burundi and Nepal have shown that economic programs have limited mental health benefit compared to social support services (Adhikari et al., 2014; Jordans et al., 2012), but these studies did not compare outcomes by gender, which would be suggested based on our current study’s outcomes.

CONCLUSION

The context for armed conflicts varies substantially across place and time. Although this population was characterized by a large percentage of voluntary recruits, that is not always the case in other conflicts. Similarly, the widespread participation of women in fighting forces has been limited to a few other settings of child conscription. Therefore, generalization of specific findings from this study would not be appropriate. The findings do however, point toward strategies that armed groups in a range of settings may exploit to manipulate youth in participating violent conflict. For example, armed groups identify youth who feel that the ability to pursue their goals is thwarted by societal norms and institutions. Although an armed group’s philosophy may be generally appealing, our data suggest that this also needs to resonate with an individual’s desire to escape a stressful situation or to pursue goals not deemed achievable through non-violent means. The importance of personal relationships is an area that requires further research, especially for girls engaging in political violence. Regarding mental health needs, forced recruitment predicts higher PTSD among boys but not among girls. This further illustrates that context is crucial to consider for whom soldiering will have the most negative impacts.