Child Responsive Conflict Early Warning: Prioritizing Children in Peace and Security

ABSTRACT

Children affected by armed conflict are disproportionately impacted by humanitarian crises and rising global fragility. Changing conflict dynamics, emerging tactics by armed groups, and evolving radicalized violence are increasing children’s vulnerability, including the threat of being recruited and used in violence. And yet, there is a glaring omission of consideration for children in conflict early warning and no early warning indicators focused on children’s engagement in violence. Our study explores how children are considered in early warning systems and the need to focus on recruitment and use as a relevant conflict indicator, to inform more effective and child responsive prevention.

Introduction

The disproportionate impact of armed conflict on children is widely known (Machel 1996, The Fund for Peace 2020, World Bank 2020), and yet the strategic significance of children to sustainable peace continues to be undervalued and under actioned (Machel 1996, Whitman and Baillie Abidi 2020). In a time of rising fragility and emerging conflict dynamics (World Bank 2020), protecting children from the adverse effects of conflict, and particularly from the six grave violations against children – killing and maiming, recruitment and use, sexual violence, abduction, attacks against schools and hospitals, and the denial of humanitarian access – is essential to breaking cycles of violence and nurturing possibilities for peace. Machel’s (1996) ground-breaking report entitled ‘The Impact of Conflict on Children’ identified the protection of children’s rights as the cornerstone of a global peace and security agenda. From this seminal work, Machel pivoted the international community towards a more comprehensive international children and armed conflict (CAAC) agenda from which multiple UN Security Council resolutions have been adopted.And yet, there is a glaring omission of consideration for children in existing conflict early warning systems, and currently no early warning indicators focused on children’s engagement in conflict, one of the most documented violations against children and an important factor in the perpetuation of cycles of violence.

As global peace deteriorates, compounded by protracted and emerging conflicts, climate crises, enduring colonial impacts, and the devastation of the COVID-19 pandemic (OEDC 2020, Institute for Economics and Peace 2022), and as projections indicate increased global insecurity over the next decade (World Bank 2020), it is essential to improve conflict early warning systems to inform more effective violence prevention. Given children and youth (ages 0–24) make up over 41 per cent of the world’s population (UN 2019, 2022), and over 426 million children are living in conflict affected regions (Østby et al. 2020) with 337 million at risk of being recruited and used in violence (Østby et al. 2022), it is imperative that the unique challenges and perspectives of children are centred in conflict early warning. While the definition of a child varies regionally and culturally, for the purpose of this paper a child is defined as ‘every human being below the age of eighteen’ based on the United Nations Convention on the Rights of the Child (1989).

The prioritisation of children and youth require their active engagement in peace and security. The Youth, Peace and Security (YPS) agenda, established under Security Council Resolution 2250 (2015), was the first international framework to recognise the role of youth in peacebuilding, conflict prevention and countering violent extremism – seeing them as ‘part of the solution, not the problem’ (Berents 2022, p. 1). Youth play a critical role in supporting and articulating the perspectives of children and what is required to uphold their protection (Minasian et al. 2022). However, despite overlap in international definitions of what classifies a youth versus a child (Berents 2022), the YPS agenda does not ‘specifically address the unique challenges that children face nor do they recognise how children are central to achieving peace and security’ (Minasian et al. 2022, p. 2). To prevent contemporary armed violence, we must understand the diverse experiences of children living in fragile and conflict affected regions, and how these experiences can inform conflict early warning.

The Vancouver Principles on Peacekeeping and the Prevention of the Recruitment and Use of Child Soldiers is an example of a protection and prevention mechanism that prioritises children in peace and security. The Vancouver Principles are a collection of seventeen principles designed to encourage States to commit to preventing the recruitment and use of children in conflict (Government of Canada 2017). The principles range from prevention-oriented planning to the inclusion of child protection focal points throughout mission command structures, to sharing best practices in prevention work. Specifically, Principle 3 highlights the importance of early warning as a broader conflict prevention strategy and describes the need to ‘monitor, report, identify, and address early warning signs of the recruitment and use of child soldiers, recognising that such acts can amount to war crimes and can be a precursor of other war crimes, including attacks on civilians and civilian objects, crimes against humanity, and genocide’ (Government of Canada 2017).

Despite the growing traction of the Vancouver Principles globally, as well as the increased focus on conflict early warning as part of international research and policy-making agendas, early warning related to children’s vulnerability within conflict and fragile environments remains a critically understudied area (Johnson et al. 2019, Cleave and Watkins 2020). To date, more resources have been put towards reactionary measures, responding to the consequences of armed violence, and are largely invested in processes of disarmament, demobilisation, rehabilitation, and reintegration of former child soldiers (Cleave and Watkins 2020, Kupper and Young 2020, Whitman and Baillie Abidi 2020). Such measures – although critical – prove to be more costly than earlier, prevention-oriented approaches (Johnson et al. 2019). Thus, our study explores how children are considered in existing early warning systems to inform more effective and child responsive prevention. Following an overview of the theoretical perspectives on violence prevention that frame this research, and a review of the literature on children affected by armed conflict and early warning, we share our analysis based on qualitative interviews with early warning researchers and practitioners. Our findings illustrate the exclusion of children and child-centred indicators within early warning systems, important considerations for child-responsive early warning methodology, and potential indicators focused on the recruitment and use of children based on lessons learned from existing models of prevention. Our study contributes to the fields of child protection and conflict prevention by articulating the complexities of armed violence, the importance of child responsive conflict early warning, and the quintessential role children play in sustainable peace.

Theoretical & Conceptual Framework

Children affected by armed conflict, including their recruitment and use, offers a stark illustration of the interconnections of multiple forms of violence. Galtung (1969, 1996) articulated violence as a triangle of interrelated types: direct (ex. physical – often more visible), structural (exploitative systems), and cultural (norms). Dubbed the violence triangle, this framework enables an analysis beyond traditional definitions of violence that concentrate on actions and behaviours, to encompass structural and cultural elements that sustain violence. This more comprehensive analysis is essential to understanding cycles of violence and children’s place in the peace and security agenda.

Direct violence is observed with the millions of children who suffer non-fatal, and increasingly, fatal consequences of violence (WHO 2014, UN 2021). In armed conflict contexts, the global community tracks the many manifestations of direct violence, such as killing, maiming and sexual violence (UN 2021). Yet, it is important to recognise that violence is ‘strongly associated with social determinants such as weak governance; poor rule of law; cultural, social and gender norms; unemployment; income and gender inequality; rapid social change; and limited educational opportunities’ (WHO 2014, p. 33). The structural aspects which enable the conditions for violence to thrive, such as discriminatory and inequitable access to humanitarian aid, include deeply engrained sex and gender-based violence. These interconnected and exploitative systems, which largely stem from the legacy of colonialism, disproportionately impact children in conflict contexts and further exacerbate cycles of violence. Understanding violence as a social construction (Walby 2009), deeply connected with relations of power (Baillie Abidi 2018), and nested within histories of injustice (Mohanty 2003), is essential to understanding the enhanced complexities of violence against children in conflict contexts. If violence is socially constructed, and embedded within the normative frameworks of communities, then in essence, it can be dismantled and challenged to imagine possibilities for social change.

Children and Armed Conflict

Children affected by armed conflict are disproportionately impacted by complex humanitarian crises and rising global fragility. Today, one in six children live in a conflict-affected area (Østby et al. 2020), growing up in precarious situations where education is frequently interrupted, employment opportunities are limited, and violence is normalised, and yet, children remain largely invisible in the peace and security agenda. In 2021, the UN (2022) verified 23,982 grave violations against children, however it is important to note that violations are substantially underreported due to the complexities of data collection in conflict contexts, the multi-tiered UN validation process, and the politicisation of armed conflict (Cleave and Baillie Abidi 2021; Watchlist 2021). Of the six grave violations, the recruitment and use of children as soldiers has frequently been the most documented violation within the Secretary General’s annual report, affecting close to 10,000 children a year (UN 2019, 2020, 2021, etc.). One in eight children globally are vulnerable to recruitment and use based on their proximity to armed violence and armed groups (Østby et al. 2022) and their vulnerability continues to shift due to changing global contexts, emerging tactics, and increasing radicalised violence. This is a concerning trend as recruitment serves as a connecting factor to other grave violations and the associated adverse childhood experiences set a trajectory for sustained cycles of violence.

Although the recruitment and use of children has existed throughout history, the engagement of children in violence continues, and the strategic motivation for engaging children is evolving. To advance a global framework for protecting children at risk of recruitment, and in recognition of children’s essential role in the peace and security agenda, the Paris Principles on Children Associated with Armed Forces or Armed Groups were created in 2007. The Paris Principles define recruitment and use 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’ (UNICEF 2007, para 2.1). Based on this definition, children’s engagement in conflict reflects numerous roles and is not limited to direct participation in hostilities. This definition is important as it justly problematises the common portrayal of children engaged in conflict as African and boy centric. Recruitment and use are not isolated to one part of the globe, it affects both boys and girls, and the roles performed are constantly evolving (O’Neil 2018, Tynes 2019).

As global humanitarian needs continue to rise, and corresponding international funding drops, prioritising prevention becomes even more important (Wagner and Jaime 2020). Research shows that recruitment and use can have long-term implications on the ‘health and security of societies’ (Achvarina and Reich 2006, p. 130). The recruitment and use of children can also have longstanding psychological impacts on children due to broad violence exposure and potential engagement, both of which require extensive resources and care to support successful reintegration and well-being (Van Broeckhoven 2018). Conflicts where children have been engaged tend to last longer, particularly when rebel or non-state groups are recruiting children (Haer and Bohmelt 2017). Haer and Bohmelt (2016) argue that ‘child soldiers influence the probability of conflict recurrence. Or, put differently, when rebel groups recruited children as soldiers during a previous conflict, it is more likely that armed conflict recurs in a post-conflict society’ (p. 426). Breaking cycles of violence from their earliest stages is critical to create conditions for long-term security and peace where children and youth are centred.

Conflict Early Warning

Early warning has evolved over the past three decades from its origins in Cold War era military intelligence, to an important tool for detecting impending humanitarian emergencies, particularly following the failure of the international community to respond to the 1994 Rwanda genocide (von Keyserlingk and Kopfmüller 2006, Pham and Vinck 2012, Nyheim 2015, Leaning 2016). Today, there are a multitude of early warning systems focused on various elements of conflict and armed violence – from threats of terrorism, political violence, and coups, to displacement, famine, and mass atrocity. While there is no universal definition of conflict early warning (Centre for Civilians in Conflict 2021), the OECD define it as ‘a process that (a) alerts decision makers to the potential outbreak, escalation, and resurgence of violent conflict; and (b) promotes an understanding among decision makers of the nature and impacts of violent conflict’ (Nyheim 2008, pp. 11–12). Nyheim (2015) outlines three generations of early warning systems, where the first two generations (1995 to 1999; 1999 to 2003) typically focus on international and regional systems respectively, and the third generation (2003 to present) is focused on more localised applications. Approaches to early warning and response continue to evolve through new methodologies and technologies – notably the widespread reach of the internet and cellular service (Bock 2015). Regardless of scope and scale, early warning involves regular data collection and analysis that link warning with response (Matveeva 2006), and it is during this juncture that early warning systems struggle to translate warning into time-sensitive action (Harff 2009, Bock 2015, Wagner and Jaime 2020).

To advance predictive capabilities, early warning systems utilise indicators, which act as observable activities that measure changes, shifts or emerging trends within a situation of focus. Such indicators are used as the basis for warnings – often paired with a qualitative level assessment – to alert decision-makers and key actors that action may be required. However, early warning systems have largely overlooked the significance of social and identity factors. Early research on gender and early warning, for example, presented foundational critiques on the use of generic indicators that did not disaggregate or recognise differing impacts and vulnerabilities across demographics, notably gender identities (Schmeidl and Piza-Lopez 2002). Developing a warning system that assumes vulnerability is the same across demographics will prove ineffectual in a society that is already unequal, thus ‘increasing the marginalisation and vulnerability of groups who have less power and influence’ (Brown et al. 2019, p. 1). Breaking down indicators by gender provides a greater opportunity to identify signs of instability that affect diverse groups differently (Schmeidl and Piza-Lopez 2002). Considerations around gender are inherent to developing a system where everyone is included (Brown et al. 2019), and thus must consider those who may be hardest to reach but may require the warnings of impending crises the most (Wulf and Debiel 2009). Considering other social and identify factors, such as age, throughout all stages of the early warning processes are equally important.

Early Warning & Children

Since Graça Machel’s 1996 report, both research and policymaking have sought to better understand and prevent the impact of conflict on children. Several treaties, political tools, and protection mechanisms have been created to protect children in conflict and fragile contexts, such as the Optional Protocol to the Convention on the Rights of Child, the Safe Schools Declaration, and the UN’s Monitoring and Reporting Mechanism (MRM). Established in 2005 through Security Council Resolution 1612, the MRM provides the most comprehensive global system for documenting violations against children, including cases of recruitment and use (UN Security Council 2005). Although not designed to be an early warning system, the purpose of the MRM is to capture the data required to support ‘well informed, concerted and effective advocacy and responses to protect and care for children and foster compliance with international child protection standards and norms’ (UNICEF 2014, p. 2). The MRM provides important data on trends related to children and armed conflict to deter and hold perpetrators to account. However, given that the MRM is not triggered until violations against children have already occurred, and the release of data is completed annually, early interventions to break such cycles are often missed or limited. Moreover, an Eminent Persons Group convened by Watchlist (2021) found that since 2015 the process for listing perpetrators of grave violations against children has become increasingly politicised, as influential Member States within the UN system have successfully lobbied being delisted or not listed at all. Such discrepancies within the listing process demonstrates the layers and complexities of conflict prevention, and the lack of recognition of how children are both impacted by and influence structural and cultural violence.

Conflict prevention is extremely complicated and thus, there is a need to move beyond singularly focused warning systems and indicators to ‘take into account the needs of diverse groups at every stage of the conflict cycle’ (Schmeidl and Piza-Lopez 2002, p. 19). Mary Anderson’s principles of Do No Harm argue that approaches to development and peacebuilding that are universal in scope and do not consider the diverse social groupings and vulnerabilities of a country will be more likely to prove ineffectual (Anderson 1999, 2000). Despite this, an explicit focus on children in conflict early warning systems remains limited (Cleave and Watkins 2020). Although children affected by armed conflict, including those recruited and used, represent a heterogenous group – diverse in race, ethnicity, gender, class, religion, geolocation, and many other positionalities – there is a tendency within early warning systems to lump children together as one broad demographic. O’Neil (2018) argues ‘even within the same conflict, there is likely contextual variation across time, space and demographics’ (p. 42). For instance, considering how gender influences the methods of recruitment and use and the distinct impacts on boys, girls and gender diverse children is critical to better understand indicators of vulnerability and dynamics of conflict (Brett and Specht 2004). Protection and prevention mechanisms, including early warning systems, that fail to recognise the diversity of children’s identities and lived experiences, will fail to respond to the needs of children living in fragile and conflict contexts (Baillie Abidi 2021, Kapur and Thompson 2021).

Much of the existent research on recruitment and use has focused on incentives for children to join armed forces or armed groups, or the rationale for recruiting children from the perspective of these actors, drawing on indicators such as: high rates of poverty (Singer 2005); proliferation of light arms (Singer 2005) and early, widespread distribution of arms (Booth et al. 2020); unaccompanied minors and internally displaced children (Singer 2005, Achvarina and Reich 2006); limited or non-existent child protection mechanisms in refugee and internally displaced camps (Achvarina and Reich 2006); youth bulges (Peters et al. 2003); history of state violations of human rights, particularly related to children (Government of Canada 2019); normalisations of violence against children (Baillie Abidi 2021); and limited capacity and human capital of armed forces and armed groups (Booth et al. 2020). Although this list is not exhaustive, it is important to note that such indicators cannot be viewed in isolation from one another. Poverty on its own does not make children vulnerable to recruitment and use as there are countless contexts where such socio-economic conditions exist without instances of recruitment taking place (Haer 2019). Combining multiple indicators, including indicators disaggregated by social and identify factors, present a fuller risk analysis of children’s vulnerability.

Equally important to the consideration of children’s diversity and the range of relevant indicators, is the recognition of children’s agency. Through analysis of interviews conducted as part of the Government of Colombia’s former combatant database, Rosenau et al. (2014) note that ‘children were not simply passive victims, but quasi-independent agents faced with a limited menu of unpalatable options’ (p. 282). Factors such as social exclusion and marginalisation, familial connections to armed groups, and the desire for a greater quality of life were identified as reasons for children’s engagement in the conflict. When prevention and protection mechanisms are created from the perspective of adults, key considerations are often missed or misjudged leading to ineffective implementation. Children’s voices, perspectives, and participation in peace and security innovation are critical to violence prevention (Baillie Abidi 2018, 2021, O’Neil 2018, Pruitt 2020). Participation is a right within the Convention on the Rights of the Child (UN 1989), and yet, the engagement of children in the design, implementation and evaluation of violence prevention, peacebuilding programming, and early warning systems, is universally lacking.

Methodology

The central objective of this study was to better understand how children are considered in existing early warning systems, to enhance child responsiveness in early warning and to strengthen child protection mechanisms in conflict and fragile environments. The secondary objective of the study was to explore early warning and the recruitment and use of children into armed forces and armed groups, given the significance of this grave violation in perpetuating cycles of armed violence. Two primary research questions guided the research process: Question 1 – How are children considered in existing early warning systems of conflict prevention?; and Question 2 - How can early warning enhance early and collaborative action to better protect children and communities from recruitment?

The study employed semi-structured interviews to investigate the research questions. Semi-structured interviews with key informants were selected to enable professionals to share insight and perspective from their experience (Blee and Taylor 2002). Semi-structured interviews are also valuable for conveying and unpacking experiences and issues that may be personal and hard to voice (Johnson 2001, Brounéus 2011). Working in conflict prevention, and particularly in relation to the impacts of conflict on children, can be a sensitive discussion and semi-structured interviews can allow participants to guide the discussion in a safer way. We conducted 20 interviews with early warning researchers and practitioners from a variety of sectors, including the security sector, international non-governmental organisations, national non-governmental organisations, and research institutes. The interviews took place between February 2021 and May 2021, were conducted on-line, and the participants were selected from the researchers’ professional networks based on a mapping of existing early warning systems. The following Table 1 features an overview of the participants’ gender, professional background, and the geographic areas of focus of their respective early warning systems.

Table 1. Participant demographics.

Interview Participants	Sector	Region of Focus	Gender
001	Security Sector	Global	M
002	International NGO	Global	M
003	Research Institute	Global	F
004	National NGO	Africa	M
005	Research Institute	Africa	F
006	African Union	Africa	M
007	United Nations	Africa & South America	F
008	National NGO	Africa	M
009	Research Institute	Global	M
010	International NGO	Africa	M
011	International NGO	Africa	M
012	International NGO	Global	M
013	United Nations	Global	F
014	Research Institute	Global	M
015	Research Institute	Africa	F
016	International NGO	Global	F
017	Security Sector	Global	M
018	International NGO	Africa	F
019	Research Institute	Africa	M
020	Research Institute	South America	F

Findings

Our study convened researchers and practitioners to better understand how children are considered in early warning systems, particularly in relation to children’s recruitment into armed forces and armed groups. The research findings illustrate the lack of consideration for children in conflict early warning, important elements for child-responsive warning methodologies, and relevant indicators related to the recruitment and use of children as soldiers – a key feature in conflict prevention. The following section explores the key findings.

Exclusion of Children from Conflict Early Warning

Children are one of the most vulnerable and usually the largest population within fragile and conflict environments and yet they remain absent from conflict early warning. As one participant described, ‘children are not something we really talk about’ in conflict prevention (Participant 003). ‘Children are often the hidden figures of humanitarian emergencies and response’ (Participant 002), hence ‘tangible steps to implement child-related indicators, particularly for recruitment and use and broader violations that impact children directly, have not been championed’ (Participant 006).

The participants engaged in this study were not aware of any early warning system with indicators focused on children, despite a recognition that children are increasingly becoming politicised in contemporary conflicts. A participant (005) highlighted how:

children are extremely political in this conflict. When the conflict began, around January 2017 many parents pulled their children out of school out of protest and have not been back since then. Children have been tortured and kidnapped for going to school, teachers have been actively targeted.

The participant (005) further added that ‘children are the silent victims of this crisis. The impacts will be felt for a very long time’ and suggested that tracking indicators related to children, including their recruitment and use, would fill gaps in current knowledge of conflict trends. Without a concentrated focus on child-centred indicators, the explicit collection of data on children is absent from conflict early warning. One participant (002) argued that even though their research is focused on children, their early warning analyzes broader conditions related to conflict and forced migration ‘because of lack of data [on children]’. Several participants discussed challenges of limited data disaggregation, and that although gender related data is scarce it is typically more readily available than data related to age. Alarmingly, one participant (019) suggested that the lack of data on children is further challenged because ‘even fighters know pictures of children is not good’, and armed groups are actively concealing impacts related to children. Another participant (011) who was outside the UN system, argued that there is also a lack of trust in the reliability of data on children and armed, as they recounted how they ‘personally contributed data [to the MRM] that reflect higher numbers’ than the UN reported for a whole country.

Child-centred indicators and reliable data are not the only elements missing from conflict early warning, children’s perspectives and engagement are completely overlooked. ‘Children are an area that tend to be avoided, especially in terms of engagement … instead they are typically considered in post-conflict settings rather than early interventions’ (Participant 017) and they are seen ‘as beneficiaries instead of stakeholders’ (Participant 004). One participant (020) described how children and violations of children’s rights are largely featured in ‘special reports’ as opposed to being included in conflict assessments, and these reports are only produced if there is ‘a dedicated person to focus on children’. Other participants discussed how children are often invited to attend and speak at high-level humanitarian events focused on child protection but are not meaningfully engaged in early warning or conflict transformation policy and programming. One participant (013) described this approach as ‘essentially asking someone to talk about their trauma, but not giving them the opportunity to link their advocacy with change’. Another participant (004) highlighted how ‘young people are only at the level of providing information’, however, ‘young people are responding and need to be part of the process from start to finish’. In this regard, a participant (020) reflected on the importance of creatively engaging and communicating with children and described a process whereby warnings were ‘put into public spaces where adolescents were at risk of being recruited… they were able to reduce the risk just by saying they were watching’ and by recognising the importance of communicating directly with children.

Children are key to sustaining peace and, in some cases, sustaining conflict. For instance, ‘youth groups have been instrumental in reducing violence in communities … ’ yet have also ‘carried out atrocities on behalf of adults as seen in Rwanda’ (Participant 010). One research participant (019) reflected on a conversation with a colleague who ‘met a girl who was a child soldier … she had an AK strapped to her and he drove her somewhere … they had an interesting conversation about why she became a child soldier’ and how domestic violence was a push factor. In this regard, most of the participants discussed the importance of listening and learning from children, with one participant (008) noting ‘they [children] have experienced conflict in different ways from their parents and have valuable insights to share’. Thus, there is a need to move away from ‘showcasing children strictly as victims, and instead actually listening to them and designing programming with them’ and engaging with children as ‘equal partners’ in peacebuilding (Participant 004).

Another participant (007) noted that organisations ‘tend to have gender experts and children and armed conflict experts, but very rarely do they overlap’, noting a lack of intersectional consideration within conflict early warning and a lack of capacity to engage children from a gender responsive perspective. Participants cited the importance of engaging girls in early warning, with one participant (006) sharing: ‘girls should not just remain in the kitchen – as happens in many African settings – they should be partners in the work’. Discussions related to the engagement of girls included recognition of sex and gender-based violence and the stark realities of gender inequities which are exacerbated in fragile and conflict contexts.

Child-Responsive Warning Methodology

Engaging children’s perspectives in early warning is complicated given ethical standards around consent, safety, and security. Yet, the need to prioritise children in early warning was emphasised as important for breaking cycles and normalisations of violence. As one participant (010) noted, children are rarely seen ‘as agents of change’ however, ‘long-term social protection strategies – that include strong child protection framing’ and meaningful engagement of children ‘has the capacity to enable a greater prioritisation of children’s rights and the prevention of violations against children’. Understanding methodologies from child development, violence prevention, and protection lenses, is essential for the continued evolution of conflict early warning. While early warning models are diversely constructed, implemented, and evaluated, all interview participants emphasised the importance of a few essential elements for successful early warning: well-defined policies, clear indicators, and engaging diverse perspectives throughout the life cycle of the early warning process.

Effective early warning is rooted in confidentiality policies that prioritise the safety and security of data sources and those participating in data collection. Policies must guide and enable the ability ‘to classify warnings related to space, time and mandate’ (Participant 001), and in the context of a child-responsive early warning system, extra attention must be placed on ensuring a child centred analysis of the policies and practices that guide the warning methodology. Employing the principle of ‘Do No Harm’ (Anderson 1999, 2000) must be front and centre in the development of warning methodologies which centre children and are essential for systems to develop credibility and trust, particularly from the perspective of affected parties. Participants described many safety and security tactics, including the use of code words, database security measures, anonymous reporting, reporting by third parties such as diaspora communities, and sometimes the need to withhold the release of data for a particular period, but acknowledged child responsive policies are lacking in the field. A participant (011) noted that ‘when we ask witnesses to share, we are asking them to take risks which does not often result in justice for the survivors’, and these risks may look different for children. Another participant argued a ‘system needs to be secure … if contributors were located that would be an abject failure of the system’ (Participant 019). To improve safety and security for participants, and particularly for children, strategies must be developed from the earliest stages to identify risks and adverse effects with clearly defined mitigation plans that recognise existing social inequities, including gender-based violence and children’s enhanced vulnerabilities, their rights and existing protection mechanisms. The development of child-responsive policies requires collaboration between diverse fields and communities, including humanitarian, security, health sectors, early warning specialists, and community-based and child-centred regional organisations.

The participants argued that for warnings to be viewed as credible, they must be relevant, contextualised, transparent, gender-responsive, communicated clearly, connected to potential interventions, and most importantly, be based on well-developed indicators. Identifying child-responsive indicators related to conflict prevention is highly complex and requires an analysis from multiple perspectives. It was noted that indicator development ‘cannot look at violations in silos’ (Participant 012) and must create space for comparative and contextual analysis. One participant (002) stated it is ‘important to understand the drivers that lead to one phenomenon versus another, and how equally they reinforce one another’. Another participant (012) highlighted the ‘need to situate where children are in the broader picture of harm and risks to the civilian population’, emphasising children’s essential place within communities as well as the diversity among children. The participant (012) further argued that it is important to observe global trends but equally important to look at the whole life of a child, ‘to put kids in that picture; put kids in society, put kids in school … how is the daily life of a Sri Lankan Tamil kid, for example … this is what they do on the weekend, this is the frequency’. Similarly, another participant (017) emphasised the importance of developing indicators in relation to the ‘rhythm of life in the village’ and of being informed by local context. The localisation must take into consideration different concepts of childhood. Understanding when and how violations against children occur in each context, is critical to prevention. And developing accessible indicators to make ‘it easy to understand what the indicators are and the reason behind them’ is important for broad application (Participant 003).

The participants also emphasised the importance of creating gender responsive indicators to recognise the different ways boys, girls and gender diverse children are impacted by conflict. Participants argued for enhanced gender analysis to be centred within early warnings and acknowledged a glaring weakness in ‘doing gender’, with one participant (016) stating: ‘if you are not thinking about this through a gender lens you are not thinking about it’ because indicators, solutions and interventions are unique based on gender – among other intersectional considerations. Another participant (007) argued a ‘gender perspective depends on conflict trends, duration of conflict, strategies of armed groups… how girls, and non-binary children, are recruited and the impact on girls who are often invisible in the process of recruitment, release and reintegration’ require a gender responsive approach. The participant (007) further argued there is a ‘need to go beyond just prioritising the disaggregation of data after the fact. We need to create indicators on how certain issues affect women and girls differently’.

Early warning is ‘as much art as science’ (Participant 014). On one hand, if there is too much focus on the quantitative side, there is the risk that early warning may lose the ‘qualitative-human element’ (Participant 005), running the risk that local contexts, including children’s perspectives, could be overlooked. One participant (013) acknowledged how quantitative tools, such as surveys, may miss important nuances, for example, ‘depending on how you phrase the question, you could totally miss recruitment … but if you spend time with someone you might uncover it’. Engaging a wide range of actors, including actors in health, education, social welfare, and children and youth focused organisations, are critical to adequately reflect diverse perspectives and to enable effective information sharing and the identification of potential challenges and interventions. Several participants recommended that engaging older youth in the design of early warning is a strategy to ensure children’s and youth’s perspectives and engagement are centred within the methodology. As one participant (010) stated, ‘capacity building of the people so they are part of the work’ and ‘agents in their own protection’ and ‘developing protection in a better way because they are already doing it’ is key. However, enabling child responsive warning methodology is challenged by the limited engagement of donors in supporting long-term and prevention focused interventions, the failure of the peace and security agenda to identify children as key stakeholders, and the complexities of conducting early warning in a manner that honours the diverse development and experiences of children.

Child-Responsive Early Warning: Recruitment & Use

Defining child responsive indicators within early warning systems, and particularly indicators which address the multiple layers of violence enabling the recruitment of children to occur, were identified as critical to effective conflict prevention. The participants identified multiple instances where the recruitment and use of children led to long-term impacts and sustained cycles of violence, for example, in Sierra Leone, Liberia, Democratic Republic of Congo, South Sudan, Syria, and Yemen. Reflecting on years of professional experience working in peace and conflict, the participants identified three main themes related to recruitment and use and broader conflict prevention: socio-economic factors, state security and violence trends, and the actions of non-state armed groups.

Socio-Economic Factors

All participants identified socio-economic factors as foundational to the conditions for armed groups to recruit children. Several participants spoke of the correlation between a high level of instability and the recruitment of children, with one participant (007) indicating that ‘weakened social systems’ are intentionally ‘capitalised by armed groups’. Another participant (012) argued: ‘in cases where the economic situation worsens and there is no longer the same ability to pay for food, risk factors for children to end up in armed groups become higher as seen in Myanmar and Afghanistan’. Participants discussed multiple connections related to changing economic security, including: the vulnerability of families who are ‘in positions where parents may need their children to work to survive’ (013); unaccompanied children due to parental death or family members travelling to earn an income; security actors feeling compelled to engage children, for example, ‘there are cases where armed forces will bring in children and adolescents to take on smaller tasks to support families…this is seen as providing the support required to get basic needs met – and may not be perceived by some groups as formal recruitment and use’ (013); and children who are incentivised by economic opportunity in contexts where options are depleting.

In the context of ‘broken social fabrics’ the participants emphasised the importance of institutional protection mechanisms to ensure children’s rights are prioritised, especially during situations of duress. One participant (010) argued that as a starting point for exploring indicators of recruitment, it is essential to know ‘what there is institutionally … does the country have policies and structures that are effective for the protection of the rights… is there a policy against child recruitment? Is it effective? … Is there a strong rule of law?’ Several participants raised concerns over the lack of systems for unaccompanied children and the lack of options for children to be engaged in the community. A participant (016) identified the importance of recognising that the ‘needs of girls and boys… are always different’ and ‘we have to always strive to put gender dynamics in the program to ensure no one is left behind’. Connected to declining economic security and limited protection mechanisms, are the challenges associated with access to education. One participant (004) stated: ‘when there is a lack of education and basic needs are not being met, there is a higher risk of children joining armed groups’. In complex contexts, recruitment patterns were observed in areas lacking educational facilities, those with low rates of education, and those with recorded attacks on schools. Furthermore, several participants discussed a lack of options as a significant contributor to recruitment, with one participant (008) stating that ‘being out of school creates an idle population of children. In certain areas, children can tell the difference between Separatist and military bullets’ but struggle with basic numeracy and literacy. Alarmingly, another participant shared that armed groups are recognising that school-based recruitment is being noticed by communities, and thus, ‘not a good indicator because the actors realised that if they [children] leave the schools, they [authorities] will release the alert and that is not what they [armed groups recruiting] want. Now they want children to keep going to schools, so they do not draw attention’ (Participant 020).

State Security & Violence Trends

Shifts in violence are monitored within conflict early warning, which focus on diverse sets of indicators such as interpersonal violence, abduction, and political violence, depending on their mandate. No participant was aware of any early warning system of conflict prevention that included indicators on the recruitment and use of children as soldiers. This omission is particularly interesting given many participants identified the recruitment of children as an indicator of escalating conflict, diminishing human rights, and three participants indicated that recruitment is an indicator of genocide.

All participants discussed the importance of including indicators on broader violence trends to reflect the interconnections between various levels and forms of violence and the need ‘to not just look at things in isolation’ (012). Government ‘willingness to breach other norms in conflict’, including ‘government statements that show a disregard for children’ were identified as important indicators of the potential recruitment of children and demonstrate structural and cultural violence (Participant 007). Other forms of violence that were raised as areas of concern included: forced disappearances, forced and sporadic displacement, use of torture, high crime and trafficking rates, control of media, presence of arms, lack of humanitarian access, and polarisation in society. Participants highlighted the need to monitor changes in normal societal behaviour and routines, such as people no longer congregating at the markets (Participant 017), to consider if these changes are related to escalations of violence. Finally, the presence of international security was raised as a possible protective factor against manifestations of violence against children.

Norms of violence across a range of social contexts were also identified as principal elements in relation to the recruitment and use of children as soldiers. ‘If you have a feeling that you will not wake up in a safe environment … and see patterns of violence/abuse that are designed to rip apart the fabric of society’ (005), and recognise high levels of impunity, recruitment may seem like an appropriate pathway for safety and survival. Similarly, if children have been exposed to other forms of violence, such as seen through the mobilisation of children in cattle raiding in South Sudan, the concept of being recruited and used may already be normalised. One participant (005) stated:

Having so many people witness atrocities, radicalizes large portions of the population. There are videos where attacks are going on and communities are surrounding and watching or witnessing, where children are at the forefront looking on. Even in children’s games there are cases where children have divided themselves as Separatists and military agents. Radicalization on the playground without fully recognizing what is happening. Children calling themselves ‘commander’ or other security-based terms.

Furthermore, if local violence is increasing such as ‘when the cycle of violence goes for longer than two weeks, the local area is at a greater threat of being destabilised. As such, children will be more vulnerable to abductions. An extension in violence also usually means that groups are being supported by someone with a lot of power’ (016). Given the complexities of conflict and community norms, most participants discussed the need to ensure that a cultural lens is employed in the development of indicators, and that affected communities and end-users feel part of the process. In this regard, one participant (004) stated: ‘it is always important to bear in mind that the vulnerabilities [to recruitment and use] will not always be the same’.

Actions of Non-State Armed Groups

The presence and actions of armed groups is an important and often ‘taken for granted’ (004) factor in the recruitment and use of children as soldiers. Considering if ‘there are armed actors who are interested in children …, why they are interested in children’ (004) and understanding the typologies of the armed groups and the dynamics they operate within, are key to developing relevant indicators related to recruitment. One participant (010) stated: ‘from the angle of the recruiters, [you] need to consider why they target young children and young girls especially. Children are seen as very docile. Children have a soft spot in the community’. At the same time, the use of children can be seen as a strategic advantage by some armed groups. One participant (007) reflected observing the strategic use of children to garner attention within peace processes, stating: ‘when you see certain organisations beginning to make a lot of noise or incite chaos, they may be attempting to get enough attention that they are invited to the negotiating table as a way to access greater power. Recruitment can thus be used as a tactic in this regard’. Another participant (015) argued that girls are strategically used for many reasons, one of which is to gain international attention. And yet another participant (011) shared how a recruiter acknowledged the strategic recruitment of young people and the tactics employed to intentionally seek out children who have little social connections but who also demonstrate leadership potential.

The tactics, culture and composition of armed groups were presented as essential elements to further consider. The participants pointed to several behaviours, or changes in behaviour, that were notable in relation to risks for child recruitment, including: the creation of new/independent armed groups; increased defections; sporadic attacks on communities; changes in command; the use of fear tactics; positioning near refugee camps or schools; and demonstrating a disregard for children, including within social media. Several participants highlighted links between rank achievement and the recruitment and use of children, with one participant (006) illustrating: ‘if you want to say you are a captain, you want to say that you have so many people under your control’. Another participant (007) highlighted that in many contexts, ‘having children connected to you as a warlord is seen as a symbol of power … armed groups are incentivised in advance of unification processes to quickly build-up their numbers … Where numbers increase, Generals are able to position themselves to take on higher ranks by having more personnel who report to them’ - in many cases children are used to bolster numbers quickly.

Cultural and historical context are additional factors related to the actions of armed groups and the potential for recruiting children. One participant (011) acknowledged that ‘recruitment may be reflective of the society where these [armed] groups operate and may not be seen as unusual at all’. The mere presence of armed groups within communities may foster stronger relational connections to armed actors and a sensitisation to arms and violence. For example, it is often easier to recruit children from one’s own community: ‘when young boys see their elder brothers with arms and already terrorising their village, paired with a lack of education, this presents an opportunity for the elder brothers to be seen as role models – and from there the younger children are more easily influenced to join the conflict’ (Participant 013).

Discussion & Conclusion

Children continue to be disproportionately affected by conflict and violence based on a diverse set of realities related to their age, gender, political affiliations, economic status, and other positionalities (Baillie Abidi 2021; UN Security Council 2021). Given children’s significance to long term peace and security, the central objective of this study was to explore how children are considered within conflict early warning to inform more effective and comprehensive prevention mechanisms. The findings demonstrate that consideration for children remains glaringly absent within the conflict early warning field, including within indicator development, data collection, methodological approaches, and broader conflict trend analyses. Without understanding the contexts underlining the rights and vulnerabilities of children, early warning systems fail to recognise the role of children in mitigating or exacerbating conflict, leaving little to no space for their prioritisation in prevention efforts.

The lack of focus on children in conflict early warning is problematic for many reasons. It is well-documented that grave violations against children in conflict have the potential to exasperate fragile safety nets and protection mechanisms, to normalise violence during formative years, and to contribute to the creation and maintenance of cycles of insecurity and violence. Notably, situations where children have been recruited and used in violence can be linked to both increased conflict duration and likelihood of conflict reoccurrence (Haer and Bohmelt 2017). Across most modern conflicts, the recruitment and use of children has become a defining feature (Achvarina and Reich 2006), with estimates that the number of children recruited globally has nearly doubled since 2012 (Child Soldiers International 2019). This is concerning given the links identified under the Vancouver Principles between child protection and conflict prevention (Government of Canada 2017), and the links identified by the study participants between child recruitment and increasing risk of mass atrocity and genocide. Hence, a greater prioritisation of children’s experiences, including a focus on recruitment prevention, is required to break cycles and normalisations of violence from the earliest stages.

Yet, to date, limited research within the early warning field has focused on children as essential to broader conflict prevention. Understanding the risks that children face, and how their vulnerability in conflict is influenced by their age and gender, is essential to inform effective conflict prevention. Conflict early warning could be improved by further incorporating an intersectional analysis of conflict contexts, including an analysis of children’s every day, going beyond elements of direct violence, to include structural (i.e., inequitable protections for children from certain communities) and cultural violence (i.e., norms of violence towards children). It is important to recognise that children experience conflict differently than adults, thus, there is much to be learned from childhood experiences and the indicators that increase their vulnerability or their resiliency to better inform prevention and response. Developing relevant child centred indicators and child responsive early warning methodologies, are critical to create opportunities for timelier and more comprehensive interventions.

If violence is socially constructed, and systematically upheld, part of early warning is understanding how violence becomes normalised to shift towards norms of peace more effectively. One interview participant claimed, ‘it is important to recognise that there is no gold standard for early warning’ (Participant 009). However, by building child responsive early warning, there is an opportunity to create more effective mechanisms to protect children from violations of their rights and to protect communities from intergenerational cycles of violence. For too long, children have been overlooked as active agents in peace and security. Through the development of child responsive early warning, there is an opportunity to position children as central to the international peace and security agenda. Without adequate consideration of children, and the vulnerabilities they face in their formative years, cycles of conflict and violence will continue to perpetuate. Engaging children and incorporating child-centred analyses within early warning is fundamental to sustainable peace.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Catherine Baillie Abidi

Catherine Baillie Abidi is a professor in Child and Youth Study at Mount Saint Vincent University. Her research interests include: children, peace and security; violence prevention; youth engagement; and participatory social action. Her book, Pedagogies for Building Cultures of Peace explores how normalizations of violence are constructed from the perspective of young people and how pedagogies can be created toward building cultures of peace.

Laura Cleave

Laura Cleave is a research advisor at the Dallaire Institute's African Centre of Excellence in Kigali, Rwanda. Her research is focused on conflict early warning and children, peace and security. She facilitates a growing community of practice designed to provide a knowledge-sharing platform for subject matter experts in the fields of recruitment prevention, child protection and early warning.