2.1 Conceptualization

1In order to give my study a strong basis on which to build an analysis, a clear definition and conceptualization of the core concepts of the research is required. Thus, I identify the concepts here as ideal-types that I then confront with reality. I have created an “analysis sheet” (appendix 1) that enables me to carry out my research while referring accurately to the theoretical aspects as defined.

2.1.1. Accountability

2I use the definition of accountability as given by Bovens (2007a):

A relationship between an actor and a forum, in which the actor has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgement, and the actor may face consequences. (2007, p. 447a)

3There are five basic questions that stem from Bovens’s work to assess the functioning concept of accountability: Who? To whom? For what? How? And with what outcome? (Bovens, 2007a, p. 451).

4Thus, accountability represents a relation between power-wielders and those holding them accountable (Grant & Keohane, 2005, pp. 30–31). Therefore, it can be considered as a proactive process by which “public officials inform about and justify their plans of action, their behavior and results and are sanctioned accordingly’” (Ackerman, 2005, p. 12). Moreover, accountability will not be seen as a retrospective process but rather as an ongoing one where participation of all actors leads to more effectivity and representativity (Burral & Neligan, 2005, p. 7).

5I conceptualize accountability in both its horizontal and vertical forms. Bovens explains that vertical accountability reflects a situation where “the forum formally wields power over the actor, […] due to the hierarchical relationship between actor and forum” (2004, p. 460). Regarding the horizontal form of accountability, he describes it as “giving account to various stakeholders in society […] on a voluntary basis with no intervention on the part of a principal” (Bovens, 2004, p. 460). Concerning social accountability, I focus on civic engagement through several elements: the use by NGOs and Ios of media, public demonstrations, protests, advocacy campaigns, investigative journalism and participatory data gathering; and, on a governmental level, the inclusion of participatory public policy-making, participatory budgeting and citizen evaluation reports (Malena et al., 2004, p. 3; Kohli, 2012, p. 14; Brinkerhoff & Wetterberg, 1999, p. 283). As evidence, authors mention evidence-based advocacy, capacity-building training and monitoring (Kubalasa & Nsapato, 2010, p. 106).

6Furthermore, Grant & Keohane provide an interesting framework to classify different types of accountability in world politics. Four of them are based on delegation: hierarchical, supervisory, fiscal and legal. Three other types pertain to diverse forms of participation: market, peer and reputational accountability (Grant & Keohane, 2005, p. 30).

Fig. 1: Grant & Keohane, 2005, p. 30

7I use this framework and its definitions to analyze each actor in the synthesis. Yet, this classification does not allow for a flexible and dynamic analysis since it provides only unidirectional paths for accountability holders and power-wielders. To better represent reality, I assume that power-wielders’ and accountability holders’ roles as conceptualized by Grant and Keohane (2005) can be swapped depending on the situation. This is why I highlight the importance of multidirectional accountability that can be defined as a relation not going in only one direction and refers to power-wielders being subject to several accountability-holders. Moreover, I intend to develop further this framework as it focuses uniquely on accountability in world politics. Indeed, my goal is to work on policies of organizations surrounding treaties, on a micro-level, assessing the multidirectionality of the concept. Furthermore, this classification does not make it possible to grasp all of the mechanisms I want to analyze such as TRAN, direct and indirect accountability mechanisms. Thus, the need to create a new classification for the purpose of my research.

2.1.2. Treaty accountability network

8The idea of a treaty accountability network (TRAN) emerges from an understanding of accountability in a multidirectional dimension, which is essential to a comprehensive overview. According to Harlow-Rawlings (2007), the creation of accountability networks is significant since it gathers various actors forming a collective identity. This enables them to formulate demands and be able to defend themselves (Harlow-Rawling, 2007, p. 560). My conceptualization of a TRAN differs from the definition of the network by Harlow-Rawlings, since it focuses on a looser tie between stakeholders. It maps relevant actors rather than the ones uniting formally to tackle an issue together. Thus, in my view, a TRAN embodies all the actors and mechanisms that pertain to holding states and other stakeholders accountable.

9Peer accountability is another form of horizontal accountability mentioned in the literature that is “the result of mutual evaluation of organizations by their counterparts” (Grant & Keohane, 2005, p. 30). It is similar to the concept of “mutual accountability between bodies standing on equal footing” (Bovens, 2004, p. 460). Thus, this concept involves all the actors of a treaty accountability network.

10It is important to emphasize that I am focusing on the role of states, NGOs, Ios and populations but, for the sake of the study’s length and coherence, I leave aside the roles of other actor such as donors.

11As part of a TRAN, a non-governmental organization is defined as an “organization that tries to achieve social or political aims, but is not controlled by a government” (Cambridge Dictionary, 2020). Characteristically, NGOs embody four basic features: they are independent from states; they work for the common good; they are non-profit directed; and their mandates, functions, size and relationships with other actors can vary substantially (Baylis et al., 2017, p. 354). As the other main actor of a TRAN, an international organization is defined as an “autonomous site of authority, independent from the state principals” (Barnett & Finnemore, 1999, p. 707). The relative autonomy Ios acquire from states comes from a bureaucratization process (Weber, 2013, pp. 196–197) and from “specialized technical knowledge, training, and experience that is not immediately available to other actors” (Barnett & Finnemore, 1999, p. 707). These organizations’ main attributes originate from “their charters, voting procedures, committee structures, and the like, [which] account for what they do (Kratochwil & Ruggie, 1986, p. 755).

12The graph below (figure 2) depicts the relations of accountability between actors of a TRAN. The main actor held accountable in the TRAN is the state. The relations among non-state actors are mainly peer accountability (double-arrows) and are otherwise either vertical (between population and states, for example) or horizontal (between population and NGOs or Ios and states). Moreover, the relations involve both direct and indirect forms of accountability. Thus, the graph offers an overview of relevant actors implied in a TRAN and the directionality of accountability relations.

Fig. 2 : Treaty Accountability Network

2.1.3. Direct accountability and indirect accountability through empowerment

13I intend to portray the TRAN by distinguishing direct and indirect accountability mechanisms. Since there are several mechanisms existing for its participants to hold states responsible, I need to go beyond Grant and Keohane’s categories to better represent the empirical situation.

14Direct mechanisms are based on a one-to-one relationship where an actor directly holds another actor accountable. There are different mechanisms for this, either horizontal or vertical. Direct procedures are set up to hold governments accountable in a direct and “linear” way: for instance, C is held accountable by A through the denunciation of certain of its practices with the help of reports or campaigns. The source of authority for Ios can be based in law: for instance, a convention itself could state the role of the organization as assuming monitoring or reporting functions. Regarding NGOs, their source of authority often originates from their moral authority but it could also be formally delegated by states or other non-state actors for specific purposes.

15The graph below depicts the direct accountability mechanism:

Fig. 3 : Direct Accountability Mechanism

16Indirect mechanisms imply a third-party in the relationship: for instance, A uses a third-party called B to hold C accountable. In my research I mainly focus on the empowerment of a population provided by NGOs and Ios to hold their governments responsible. Yet, I do not exclude the possibility of identifying other paths of indirect accountability, which would then simply mean the relation is not unique but is developed through other actors.

17These mechanisms rely mainly on horizontal accountability and NGOs and Ios have an important role to increase power-wielders’ accountability. For instance, they are able to impact the reputation of a state (through direct or indirect mechanisms) and, they can empower a population, which will then be more informed and exercise political pressure on governments.

18Empowerment can be defined as a process through which people, groups or communities gain greater authority and power over their lives and situations (Ucar Martinez et al., 2017, p. 408). It improves relational, positional and societal power in general. Indeed, “empowered individuals can significantly advance a collective agenda, in some cases even spurring emboldened publics to action” (Brinkerhoff & Wetterberg, 2016, p. 276). It is especially relevant for minorities or categories of populations that do not have access to ways of expressing their demands: empowerment “creat[es] social structural change to benefit oppressed people” (Yeich & Levine, 1992, p. 1895). However, I do not rely uniquely on the notion of “oppressed people” which seems too paternalistic and exclusive since it is not only this category of population that can be empowered. Indicators of empowerment can be participation and education of populations (Danhoudo et al., 2018, p. 6). Moreover, of interest for this research, an insightful definition of the case of the empowerment of youth and children is provided by Lundy: they should have voice, space, people that hear their demands and genuine influence on policies (Lundy, 2007, p. 940).

19The graph below depicts the relation between Ios/NGOs and the population. The arrows signify accountability of Ios/NGOs towards the population and from the population to states. Through the empowerment of the population thanks to Ios/NGOs’ programs (green arrow), people will then be able to hold states responsible.

Fig. 4 : Indirect Accountability Mechanism

20Finally, accountability pathways pertain to the analysis of states in this research. I intend to demonstrate that states can develop mechanisms to allow different actors such as NGOs, Ios and individuals to verify whether governments comply with their obligations. This term is the equivalent of “mechanism” as it is used in the research, so, to avoid confusion, the word pathway is employed to refer to mechanisms created by states.

2.2 Independent variables, dependent variables and argument

21On one hand, the outcome to be reached is for states to be held accountable for a specific treaty they ratified. On the other hand, the main explanatory factors considered are the role of the TRAN and, the direct or indirect accountability mechanisms that are set up by NGOs and Ios and pathways created by the state itself. The relation expected is that states that have ratified a convention are consequently held accountable by several actors through these different procedures.

22My argument is therefore that NGOs and Ios are core actors of a treaty accountability network surrounding each international agreement. The treaty accountability network enables them to hold states responsible for their compliance with treaties, in the case I consider here, with the United Nations Child Rights Convention (UNCRC). Certain countries also develop pathways for the population to hold them accountable. Several mechanisms, conceptualized as direct or indirect accountability, make it possible to further develop an understanding of the concept. I argue that NGOs and Ios intend to increase empowerment of the population and improve their capacity to hold states accountable for their international commitments.

2.3 Hypotheses

23The following three hypotheses formulated in the case of the UNCRC concern Terre des Hommes Suisse, UNICEF in the last 5 years and Switzerland in the last 15 years. These hypotheses draw on the conceptualization of accountability, TRAN and direct and indirect accountability mechanisms regarding treaties as elaborated above. The hypotheses are linked to each other, yet can be independently verified.

241. If a state ratifies an international treaty, it is constrained to be accountable to its population. Consequently, a TRAN develops, allowing actors such as NGOs and Ios to hold states accountable.

252. If states, Ios and NGOs are subject to accountability, this can be either direct or indirect. In the latter case, it involves accountability mechanisms that can be developed by Ios and NGOs to empower aid recipients to hold their governments accountable.

263. If peer accountability also exists among Ios and NGOs, it emphasizes power relations among them for their various actions undertaken regarding the UNCRC.

2.4 Alternative explanations

27Numerous other explanations could explain the mechanisms I intend to identify in this research. First of all, one could argue that NGOs and Ios do not rely on the ratification of an international treaty to hold a state accountable. Actually, they could advocate for states to ratify a treaty. Thus, the hypothesis that a TRAN enables actors to hold states accountable might not be the only explanation for their influence.

28Secondly, in many cases, even though empowerment mechanisms exist, this does not mean that they will be used by aid recipients or that they will be effective. Indeed, such use and effectiveness need a particular context such as functional infrastructures and tools that the population understands and values (Grimes, 2013, p. 380; McNeil & Malena, 2010, p. 186; Kihongo & Lubuva, 2010, p. 68).

29Thirdly, actors do not necessarily cooperate or have a clear vision of what other actors’ actions are, thereby limiting their capacity to hold peers accountable. Indeed, often actors support cooperation but cannot implement it (Hensell, 2015, p. 106). The reasons for this include, for instance, disagreements, threats to actors’ autonomy and lack of resources. Another explanation could be what Carpenter describes as the role of gatekeepers, where certain organizations prevent the participation of certain actors in talks and cooperation schemes based, for instance, on the credibility or reputation of the actors (Carpenter, 2010, pp. 213–215).

30Finally, the contexts in which non-state actors operate may have a significant influence on their ability to engage in direct and indirect accountability. Indeed, in some countries, the possibility to voice demands and concerns might be limited and could explain the presence of these mechanisms and the way they operate.

31However, it seems that my explanation is the most relevant so far. Studies including these alternative explanations would be interesting in order to evaluate whether they are empirically relevant or whether my hypotheses would be disproved.

2.5 Operationalization

2.5.1. Case selection

32In this paper, I focus on one NGO and one IO and select one state in order to explore empirically the development of accountability mechanisms and the treaty accountability network with respect to state compliance with specific international norms related to the rights of the child. My units of analysis are these three different actors and their policies in their context for the past 5 years for the NGO and IO and, in the last 15 years for the state. I chose this time frame because it is mostly during these periods of time that social forms of accountability and participatory methods have developed at the non-state and state levels (Marcel, 2015). It also makes it possible to do an up-to-date study.

33Firstly, my case focuses on the application of the United Nations Child Rights Convention (UNCRC). The UNCRC, as one of the most ratified conventions in the world, is a salient case to demonstrate my hypotheses. Moreover, it fosters rights of children and youth who lack a voice to express their demands (Allard, 1996, p. 167; Johnson, 2017, p. 3). The literature raises the importance of providing tools for excluded or marginalized populations, such as children and youth (Nieuwenhuys, 1997; Feinstein et al., 2009; Day, 2009). For example, Muchabaiwa and Johnson (2010) provide an interesting case study on child rights. Their research emphasizes that if children are included in decision-making and are aware of their rights, they will not allow policies that ignore them (Muchabaiwa & Johnson, 2010, p. 109). Moreover, the UN Committee on the Rights of the Child (henceforth referred to as the Committee) is a significant mechanism developed by the UNCRC. It has the duty to monitor state parties’ behaviors regarding the treaty (Garbarino & Briggs, 2014, p. 1). It would then be interesting to identify how, via the Committee, the NGO and IO studied in my research assess the role and the actions of the concerned states.

34The UNCRC thus provides a salient case study of an international agreement for human rights that implicates many different actors and, most prominently NGOs and Ios. For my analysis, I therefore focus on the UNCRC and its three related optional protocols.

35Secondly, I consider the NGO Terre des Hommes Suisse (TDHS). TDHS is a prominent NGO working on child rights. It is considered to be of medium size with a budget of 8 million CHF per year and a staff consisting of approximately 25 people (TDHS, 2020a). Because TDHS delegates much of its advocacy to the International Terre des Hommes Federation (FITDH), which groups together 10 Terre des Hommes NGOs (Denmark, Italy, Germany, France, Netherlands, Spain, Terre des Hommes Foundation Lausanne, Terre des Hommes Schweiz and Terre des Hommes Suisse) (FITDH, 2020), it is necessary to analyze TDHS and the FITDH together. Actually, the FITDH constitutes one of the most influential and active NGO federations for children in the world (Child Rights Connect, 2010, p. 17). The fact that both Terre des Hommes Suisse and the FITDH have advocacy and participatory programs fosters my choice to focus on their case in my research since the literature acknowledges these tools as central for holding states accountable. Hickey and Mohan (2004, pp. 165–166) identify that advocacy of NGOs is a powerful resource for the causes they defend and ensures their effective participation in decision-making. The role NGOs play in holding governments accountable also has much to do with the growing presence they have among multilateral I. In these contexts, they have the possibility, for example, to name and shame states to draw backlash from other governments, civil society and public opinion and push for policy changes (Winston, 2002, p. 87). Moreover, NGOs have a significant role in relaying information and reaching out to particular populations that might often be excluded from discussions such as poorly represented segments of the population (Danhoudo et al., 2018, p. 6).

36Thus, Terre des Hommes Suisse, including the FITDH, constitutes a relevant case because its profile is coherent with the goal of my research and it is a good example to test my hypotheses. I also focus on this NGO because I personally have contacts with its staff and prior knowledge of its work.

37Thirdly, regarding the international organization, I focus on the United Nations Children’s Emergency Fund (UNICEF). UNICEF is the core UN organization working for child rights (UN, 2020a) and its structure is unique, notably in its functioning and relation to the UNCRC (Black, 1987, p. 16). It operates in a decentralized manner and since it works exclusively for children, it allows more autonomy than other UN agencies (Andonova, 2017, p. 147). It was also one of the first IOs to address the issue of accountability and social accountability, making it a relevant case study.

38The role of IOs in holding states accountable is particularly interesting because it differs from that of NGOs since their relationship to states has different stakes and interests (Willetts, 2001, pp. 14–15). IOs have a different position regarding governments than NGOs and often have less agency (Willetts, 2001, pp. 14–15; Hawkins et al., 2006, p. 7; Carraro, 2019, p. 1079). IOs are the repository of some of the major international covenants, which gives them a particular importance in the eyes of states (Blokker, 2004, p. 144): for instance, UNICEF’s role is clearly stated in the UNCRC. Furthermore, it is important to remember that IOs are the result of a principal–agent model where states delegate power to IOs for specific tasks (Andonova, 2017, p. 37). Thus, IOs may to some extent reflect the opinion of the governments that fund them. However, this does not imply that they do not possess their own agency and are mere agents of states. Indeed, as Barnett and Finnemore put it, IOs are not solely empty shells, filled with states’ interests as argued by Waltz (2010, p. 81), but rather are much more complex (Barnett & Finnemore, 1999, p. 704). Moreover, IOs are subject to diverse forms of accountability. Suzuki and Nanwani argue that “it is now clear that the legal personality of international organizations entails a responsibility for their conduct” (2005, p. 179). However, IOs’ accountability is complex to grasp, which bolsters my ambition to analyze it.

39Thus, the fact that UNICEF is a leading IO working for child rights and that it is explicitly referred to in the UNCRC confirms the choice to use it as a case for my research.

40In addition, as part of the discourse on the democratic deficit IOs face, they have gradually opened up to NGOs and other actors (Kriesi et al., 2013, p. 121). Consequently, this has led NGOs to develop advocacy programs that confront IOs (Bendell & United Nations, 2006, p. 31). This is where the concept of peer accountability can be useful to assess relations among the different actors participating in a TRAN. Thus, it is essential to focus on them in this research. Furthermore, the literature does not provide a real comparative study, mapping and identifying the related mechanisms of NGOs and IOs, or the relations between them.

41Finally, I focus on the case of Switzerland as an example of a state. Switzerland is one of the most democratic countries in the world (The Economist, 2020) and it has been proactive in the implementation of the UNCRC. Since it ratified the Convention in 1997, Switzerland has received four positive evaluations by the Committee, twice in the last decade. Nevertheless, it still has significant areas to improve its implementation of the UNCRC (Leybold-Johnson, 2017); therefore, I am able to assess the work that has been done and that is still in progress. Switzerland, due to its federal organization, provides useful evidence of accountability pathways developed at different administrative levels such as the municipal, cantonal and federal levels. The Office of the United Nations High Commissioner for Human Rights (OHCHR) stated that “federalism, while entirely compatible with Switzerland’s international obligations, sometimes made it difficult to have an overview of efforts to uphold the rights of the child” (OHCHR, 2015). Therefore, it identifies the need to provide an analysis of this kind of overview of the country.

42More generally, the relevance of focusing on the responsibility of a state is also demonstrated by the literature because it shows that there are multiple relations netween non-state actors and governments. As a matter of fact, non-state actors’ ability to work with and formulate demands to governments relies heavily on the form of their relationships with ruling authorities (Scholte, 2004, p. 226). There is a large scale of depth of cooperation which ultimately entails different policies. While in some cases it is not necessary or possible to work with states, in other cases, it is crucial in order to obtain information or help in managing or designing policies, as demonstrated in the case of Médecins sans Frontières in Biafra (Simeant, 2001, p. 4). Furthermore, pathways are also used by states to open up space for civil society to participate in different political decisions.

43Thus, Switzerland’s position regarding the UNCRC and the related pathways it has developed make it possible for its population and non-state actors to identify the impact the Convention has on the country. Finally, my access to documentation and relevant regulations in Switzerland is facilitated by the fact that it is where I live and I speak the languages.

2.5.2. Methodology

44My methodology relies on descriptive qualitative research based on content analysis. I detail a specific situation and sum up the data related to it. Quantitative analysis would not be relevant in this study because I aim to identify actors and the mechanisms they have created. Because this study offers a first framework of a treaty accountability network, it is difficult to imagine a large-N analysis. It is instead beneficial to focus on a small-N analysis to have accurate and precise identification of mechanisms related to my hypotheses.

45Hereafter, I explain my method which can be divided into 6 main steps. At each step I confront the ideal-types of the core concepts previously defined with the empirical world. I use the analysis sheet in appendix 1 to correctly distinguish the concepts related to my three hypotheses.

461. Firstly, I define the treaty accountability network surrounding the UNCRC and its related protocols. In doing so, I look at what roles and procedural mechanisms exist according to these texts. I do not operate a network analysis per se since I carry out a content analysis of the legal document. This makes it possible to grasp the formal aspects of accountability developed by the Convention. I map actors participating in the TRAN and quote in annex key words and sentences as part of the content analysis.
Secondly, I identify accountability mechanisms developed by Terre des Hommes Suisse and UNICEF using a content analysis. I present two sections according to the data gathered from each of the two actors studied. Tables summarizing the sources I use are available in the appendices 2, 3 and 4.

472. For each organization, I thoroughly identify the core concepts and their related mechanisms (TRAN, direct accountability mechanisms, indirect accountability mechanisms and peer accountability) and list them to provide a clear insight into what exists. I present information from interviews conducted with professionals from UNICEF and Terre des Hommes Suisse, which allowed me to grasp whether there are informal elements adding to the formal accountability mechanisms and to compare my content analysis with their point of view. I focus on relevant keywords, sentences and names of programs pertaining to the related mechanisms, explaining the most relevant ones in the text and listing the others in appendices 2, 3 and 4.

483. I focus on Switzerland for the analysis of a state and look for TRAN, accountability pathways and peer accountability. My aim is to analyze the avenues developed by the country to be held accountable by diverse actors. In order to achieve this goal, I identify the various pathways developed by using content analysis.
At the end of the analysis of the IO, NGO and state, I provide a table synthesizing all mechanisms and pathways, classify these and count the number of mechanisms existing per category in order to make a comparison between the IO and the NGO.

494. I conclude with a general synthesis of the data presented, a map of the actors surrounding the UNCRC, and a proposal for a comparative analysis, focusing on each actor involved and based on the number of mechanisms developed. My aim is to test my hypotheses and assess whether IOs and NGOs are more inclined to developing one mechanism rather than the other. I also furnish an overview with graphics showing which procedures are the most relevant for each actor. Based on Grant and Keohane’s classification of accountability, I present a summary of the types of accountability the different actors develop and whether it matches their expectations. I compare these results with my concepts and hypotheses.

505. Finally, I provide a conclusion to the study and offer a reflection on its impact for the literature as well as ways to conduct further research in the field. The final section of the thesis is an epilogue where I propose an insight into the work of the Kofi Annan Foundation, an NGO working on peacebuilding with youth, yet not specifically working on the UNCRC. This makes it possible to assess the relevance of my research to related fields.

2.6. Sources

51Firstly, thanks to a network of contacts and respondents that I established with Terre des Hommes Suisse, I was able to study the organization’s internal documents and reports pertaining to my study. I also contacted staff members working on matters relevant to my topic to get further insight into the policies and strategies of this NGO and conducted interviews with people working in the organization. Secondly, regarding UNICEF, I rely on publicly available documents and on an interview I conducted with a senior employee of the organization to get a better understanding of their programs pertaining to my study. Finally, to analyze the case of the state’s perspective on accountability and child rights, I rely on open-access documentation such as regulations, laws and programs.

52The data I analyzed consist of reports, annual reports, strategic frameworks, statements, guidelines, strategic plans, toolkits, charters, program documents, procedures and interviews, as well as regulations and laws. Overall, I analyzed 44 sources: 17 regarding UNICEF, 19 for TDHS and 8 regarding Switzerland.

53While I strived to conduct research that respects the highest scientific standards, it is important to underline that the sources used in this research might be partially unreliable, notably the annual reports on accountability that might be biased to ensure the organization a good image rather than portraying the empirical reality. Such open-access sources might be partial and not reflect reality. Furthermore, the three interviews engaged only the people speaking and their perceptions of the work certainly does not objectively reflect reality. This supports the necessity of crosschecking data in order to ensure a complete and truthful picture.

54This empirical research contributes to the literature on accountability and non-state actors’ role towards states and their commitments. It provides insight into the various direct and indirect mechanisms existing to achieve such goals. It also assesses the importance of a TRAN as an unavoidable theoretical tool. This study is also able to acknowledge peer accountability among various actors composing the TRAN. Moreover, it paves the way for further research on a larger scale and including non-democratic countries as well as other non-state actors such as businesses and religious groups.