Jenny Söderström [email protected] Master’s Thesis/Minor Field Study Supervisor: Desirée Nilsson

NGO Conflict Resolution Efforts – Drop in the Ocean or Drop that Ripples the Ocean? A Field Study in Georgia of the Effect of NGO-Strategy on the Occurrence of Ripple Effects

Department of Peace and Conflict Research UPPSALA UNIVERSITY 13 June 2008

Abstract NGOs have been argued to play an important role in conflict resolution through their ability to build intercommunal links, facilitate reconciliation and address the root causes for conflicts. However, although scholars as well as practitioners have been eager to stress the accomplishments of NGOs in conflict resolution, the actual impact that NGO peacebuilding activities can have on the broader conflict context, beyond the direct effects they have on the participants, has not been thoroughly evaluated. Such effects, usually referred to as ripple effects, have by some scholars been argued to be more likely if the NGO has planned for them to occur. A field study of NGO grassroots conflict resolution activities was therefore conducted in Georgia to assess whether the likelihood of ripple effects increases if an NGO has a strategy for causing such effects. Interviews were conducted with 34 respondents from seven NGOs involved in conflict resolution efforts at the grassroots-level. The conclusions drawn from withincase and comparative cross-case analysis of the interview findings support the hypothesis that NGO-strategy increases the likelihood of ripple effects – a correlation between NGO-strategy and ripple effects was found, even if the causal relationship could not be firmly established. The rather small and possibly unrepresentative sample studied, however, entails that these conclusions are highly tentative.

ii

Acknowledgments First of all, I would like to thank the Swedish International Development Agency for supporting this field study through a Minor Field Study scholarship, and the Department of Peace and Conflict Research at Uppsala University for granting me this scholarship. I also want to thank my supervisor at Uppsala University, Desirée Nilsson, for valuable comments and, not least, for making me believe that this study was possible to carry out. My sincere gratitude goes to Annelie Schlaug and Annika Karlsson at the Kvinna till Kvinna Foundation for giving me more assistance than I could ever have asked for. Without them, this study would have been substantially harder to conduct. I am, furthermore, immensely grateful for all the assistance provided by the NGOs that participated in the study. It was their hospitality and effort that made this study possible. Last, but not least, I want to express my heartfelt gratitude to all the people who took the time to meet with me during my research, although this for some entailed troublesome journeys. For their time and invaluable contribution, I am the most grateful.

iii

Table of Contents 1. Introduction .................................................................................................................1 1.1. Research Objective ..................................................................................................... 1 1.2. Delimitations of the Study ............................................................................................ 2 1.3. Disposition ................................................................................................................... 2

2. Theoretical Framework ...............................................................................................3 2.1. NGOs in Conflict Resolution and Peacebuilding.......................................................... 3 2.1.1. Definition NGO ..................................................................................................... 3 2.1.2. Definition of NGOs in Conflict Resolution and Peacebuilding............................... 3 2.1.3. An Increase in NGO Activities .............................................................................. 4 2.1.4. When the People Lead, Leaders Will Follow? ...................................................... 5 2.2. Ripple Effects .............................................................................................................. 6 2.2.1. What Are Ripple Effects? ..................................................................................... 6 2.2.2. Previous Research on Ripple Effects ................................................................... 7 2.2.3. Definition and Delimitations – Ripple Effects as Action Rather than Impact ......... 8 2.3. NGOs in Conflict Resolution - Hearts without Minds? ................................................. 10 2.3.1. NGO-Strategy Addressed in Previous Research.................................................. 11

3. Operationalisations and Assessment Criteria..........................................................13 3.1. Assessing Ripple Effects ............................................................................................. 13 3.2. Assessing NGO-Strategy ............................................................................................ 15 3.3. Control Variables ......................................................................................................... 16 3.4. Assessing the Causal Link .......................................................................................... 17

4. Research Design .........................................................................................................19 4.1. Why Georgia?.............................................................................................................. 19 4.1.1. The Conflicts over Abkhazia and South Ossetia .................................................. 20 4.2. A Comparative Approach ............................................................................................ 20 4.3. Selection Process........................................................................................................ 21 4.3.1. Selection of NGOs................................................................................................ 22 4.3.2. Selection of Respondents..................................................................................... 23 4.4. Interview Structure....................................................................................................... 24 4.5. A Note on Transcription ............................................................................................... 25 4.6. Limitations of the Field Study....................................................................................... 26

5. Research Results and Within-Case Analysis ...........................................................28 5.1. Organisation 1 ............................................................................................................. 28 5.1.1. NGO-Strategy....................................................................................................... 28 5.1.2. Ripple Effects ....................................................................................................... 29 5.1.3. The Causal Link between NGO-Strategy and Ripple Effects ............................... 29 5.2. Organisation 2 ............................................................................................................. 30 5.2.1. NGO-Strategy....................................................................................................... 30 5.2.2. Ripple Effects ....................................................................................................... 30 5.2.3. The Causal Link between NGO-Strategy and Ripple Effects ............................... 31 5.3. Organisation 3 ............................................................................................................. 31 5.3.1. NGO-Strategy....................................................................................................... 31 5.3.2. Ripple Effects ....................................................................................................... 32 5.3.3. The Causal Link between NGO-Strategy and Ripple Effects ............................... 32 5.4. Organisation 4 ............................................................................................................. 34 5.4.1. NGO-Strategy....................................................................................................... 34 5.4.2. Ripple Effects ....................................................................................................... 35 5.4.3. The Causal Link between NGO-Strategy and Ripple Effects ............................... 35

iv

5.5. Organisation 5 ............................................................................................................. 37 5.5.1. NGO-Strategy....................................................................................................... 37 5.5.2. Ripple Effects ....................................................................................................... 37 5.5.3. The Causal Link between NGO-strategy and Ripple Effects ................................ 38 5.6. Organisation 6 ............................................................................................................. 38 5.6.1. NGO-Strategy....................................................................................................... 38 5.6.2. Ripple Effects ....................................................................................................... 39 5.6.3. Causal Link between NGO-Strategy and Ripple Effects....................................... 39 5.7. Organisation 7 ............................................................................................................. 40 5.7.1. NGO-Strategy....................................................................................................... 40 5.7.2. Ripple Effects ....................................................................................................... 40 5.7.3. Causal Link between NGO-Strategy and Ripple Effects....................................... 41

6. Comparative Analysis and Conclusions...................................................................43 6.1. Conclusion about Correlation and Causality................................................................ 43 6.2. Other Explanations ...................................................................................................... 44 6.2.1. The Unexpected Result – Frequent Occurrence of NGO-Strategy and Ripple Effects ....................................................................................................... 44 6.2.2. Alternative Explanations for the Lack of Ripple Effects in Two Cases.................. 45 6.2.3. Differences between Conflict Regions and Georgia Proper? ............................... 46 6.2.4. Control Variables .................................................................................................. 47 6.3. Affecting Outside Constituencies and Decision-Makers? ............................................ 47 6.4. Concluding Remarks ................................................................................................... 49

7. References ...................................................................................................................50 Appendix 1 - Tables of Organisations and Respondents............................................56 Appendix 2 - Questionnaire A ........................................................................................57 Appendix 3 - Questionnaire B ........................................................................................59

v

1. Introduction The activities of non-governmental organisations (NGOs) in the field of conflict resolution and peacebuilding markedly increased during the 1990s and attracted substantial scholarly attention. NGOs and grassroots have been argued to play an important role in conflict resolution through their ability to build intercommunal links, facilitate reconciliation, and address the root causes for conflict. It has also been asserted that they can have an impact on conflict resolution by creating ‘constituencies for peace’ and by affecting public opinion. Although scholars, as well as practitioners, have been eager to stress the accomplishments of NGOs in conflict resolution, the actual impact of NGO activities on the broader conflict context has not been thoroughly evaluated. One obvious limitation of NGO activities is that their effects might be limited to the participants, as opposed to being transmitted to larger segments of the population or to political actors. If effects are not transferred beyond those who have partaken in trainings and seminars organised by NGOs, the impact of these activities on the conflict or peace process is doomed to be only marginal. 1.1. Research Objective What is to be studied in this thesis is the possible impact that NGO peacebuilding activities can have on a peace process beyond the direct effects they have on the participants. Such impact is often referred to as ripple effects. This is an undeveloped field of research much due to the difficulties inherent in assessing and attributing the effects of NGO grassroots conflict resolution activities, e.g. workshops and trainings, on the broader conflict context. More specifically, the focus of the study will be on one of the factors that might determine whether such ripple effects will occur. The causes for ripple effects constitute an acknowledged research gap and have only rarely been addressed in previous studies. The aim of this study is to contribute to filling this gap, by qualitatively studying one of the possible causes for ripple effects of NGO conflict resolution efforts undertaken in Georgia. Some scholars have argued that ripple effects are more likely to occur if the NGO has planned for such effects, a phenomenon referred to as NGO-strategy in this study. The hypothesis put forth in this thesis is therefore that NGO-strategy increases the likelihood of ripple effects. However, no research has thus far been conducted to study if such a causal relationship exists, wherefore the objective of this study is to answer the research question: Does NGO-strategy increase the likelihood of ripple effects? A comparative study of seven cases of NGO conflict resolution activities will be conducted, using both within-case and cross-case analysis, to assess the correlation and possible causal relationship between NGO-strategy and ripple effects.

1

1.2. Delimitations of the Study Whereas it would be hard to assess the occurrence of ripple effects at the macro-level within the scope of a master’s thesis, micro-level initiatives to cause ripple effects are possible to examine. Hence, in this study, initiatives to influence or inform other people, taken by grassroots who have participated in NGO conflict resolution activities, will be viewed as ripple effects. The study will consequently only examine NGO activities targeting grassroots. The definition of ripple effects will be further elaborated in subsection 2.2.3. Moreover, as briefly mentioned above, the study will be limited to NGO conflict resolution activities undertaken in the Republic of Georgia (proper and conflict region South Ossetia). This delimitation is necessary for practical reasons, as all material for the study will be collected through field research. Due to time constraints, it would not be possible to conduct field research in several countries. However, the study will be carried out on several locations in Georgia. 1.3. Disposition In the following section, the theoretical framework underlying the study is elaborated and key concepts are defined. In section three the operational definitions are explicated, while the research design is outlined in section four, explaining the within-case and cross-case analysis and giving a detailed description of the selection of NGOs and respondents and of the interview structure. The research results and within-case analysis are found in section five, and the concluding sixth section is devoted to the comparative cross-case analysis and to conclusions about the results and the causal relationship between the variables.

2

2. Theoretical Framework In this section, the theoretical framework underlying the study will be outlined and definitions of key concepts developed. First, the role of NGOs in conflict resolution and peacebuilding in general will be discussed, before turning to the specific phenomenon of ripple effects. Lastly, the independent variable to be studied, NGO-strategy, will be explicated. 2.1. NGOs in Conflict Resolution and Peacebuilding 2.1.1. Definition NGO

There are various definitions, and categorisations, of non-governmental organisations and these have been a matter of some debate. (Richmond 2003:3) The term NGO will in this paper be used in its broader sense, meaning ‘any non-profit organisation that is independent from government’. (Fischer 2006:3) The World Bank makes a distinction between operational NGOs, that design and implement projects, and advocacy NGOs, that seek to defend or promote a specific cause and to influence policy makers. The operational NGOs are in turn divided into three groups: national organisations, international organisations and community-based organisations (CBOs). CBOs serve the interests of a specific community in a narrow geographical area and could also be termed grassroots organisations. (Fischer 2006:3) Using the World Bank’s definitions, the organisations studied in this paper are national and community-based operational NGOs. For the purpose of this study, the NGOs studied must be involved in peacebuilding or conflict resolution. 2.1.2. Definition of NGOs in Conflict Resolution and Peacebuilding

The NGOs to be studied in this paper are only those who carry out activities explicitly aimed at peacebuilding and/or conflict resolution and not, for example, general humanitarian aid agencies who might consider peacebuilding an (un)intended side-effect of their efforts. However, the organisation need not have peacebuilding or conflict resolution as its only purpose, it is sufficient that it conducts activities with such objectives. For example, there are several IDP (internally displaced person) organisations in Georgia that are addressing various IDP issues, among which peacebuilding is one. Obviously, it is only the possible ripple effects of activities aimed at peacebuilding or conflict resolution that will be studied. Many of the concepts employed in the peacebuilding and conflict resolution literature are used interchangeably (see, for example, Aall et al. 2007:327, Davies & Kaufman 2002:3, Diamond & McDonald 1996:13, Mitchell 2006:15, van Tongeren et al. 2005), wherefore the terms peacebuilding and conflict resolution need to be further elaborated. Conflict resolution,

3

conflict transformation and conflict management are often used without distinction referring to the same approaches to conflicts. (Reimann 2004:2) Similarly, moving beyond its original definition as ‘post-conflict reconstruction’, peacebuilding as a concept has developed into encompassing a broad spectrum of activities that promote peace and support conflict prevention and management (Tschirgi 2003) and is often used broadly referring to ‘any activity undertaken with the purpose of preventing, alleviating or resolving violent, or potentially violent, conflict.’ (Goodhand & Hulme 1999:16) For the sake of simplicity and comprehension, the terminology adopted in this thesis will be confined to the two concepts conflict resolution and peacebuilding, which will be used without distinction. The reason for using two concepts instead of one is that these two concepts are used interchangeably by the NGOs themselves when referring to their activities. In this study, Goodhand’s & Hulme’s (1999) definition of peacebuilding, quoted above, will be used to define both concepts. Thus, the concepts conflict resolution and peacebuilding are defined as ‘any activity undertaken with the purpose of preventing, alleviating or resolving violent, or potentially violent, conflict.’ Moreover, only practices targeting non-official, grassroots actors will be studied, i.e. what is often referred to as track-three diplomacy.1 (Chigas 2007) Hence, track-two diplomacy, i.e. NGO peacebuilding initiatives targeting non-official but influential individuals, such as NGO-leaders (Davies & Kaufman 2002), and track-one-and-a-half diplomacy aimed at providing an unofficial conflict resolution forum for official actors (Fitzduff & Church eds. 2004), will not be included.2 2.1.3. An Increase in NGO Activities

Since the end of the Cold War, non-governmental organisations have become increasingly involved in peacebuilding. (Barnes 2006:87, Richmond 2003:1) The increase in NGO activities is partly related to changes in the nature of conflict which has demanded a revision of existing practices for conflict resolution and peacebuilding. (Fisher 1997a:332, Goodhand & Lewer 1999:14, Natsios 1997:337, Paffenholtz & Reychler 2007:31) The wars seen since the end of the Cold War have often been protracted internal conflicts based on identity issues and involving at least one non-state actor and the state is often in conflict with its own population. (Davies & Kaufman 2002:2, Rupesinghe & Anderlini 1998:22) Traditional conflict resolution methods, such as official negotiation, however, are based on the assumption that the state is the primary unit in the international system and that political leaders have the power to make agreements and gain 1 These terms are usually employed when speaking about interventions by international actors. In this study, however, only Georgian NGOs, financed by international organisations, will be included. (See subsection 4.3.1.) 2 These are the types of conflict resolution activities that are initiated by NGOs. It should, however, be noted that other unofficial actors, beside NGOs, can initiate such activities. Nevertheless, this study will focus on peacebuilding efforts made solely by NGOs.

4

the support of their constituencies for any agreement reached. However, as contemporary conflicts are multi-dimensional, formal negotiations alone cannot achieve a durable peace. (Davies & Kaufman 2002:2, Diamond & McDonald 1996:1, McKeon 2005:567pp, Rupesinghe & Anderlini 1998:22) Moreover, the conflict dynamics of protracted conflicts render communities increasingly isolated and polarised, resulting in deeply entrenched enemy images and mistrust. (Kelman 1997) There is thus a need for what McDonald (2002:56) terms ‘social peacebuilding’, i.e. addressing the fear and anger of the people and rebuilding relationships between communities. It has therefore been necessary to develop new unofficial methods for conflict resolution, involving non-state groups and civil society. One example of such unofficial peacebuilding is the activities carried out by NGOs at the grassroots-level. In the following subsection, the possible benefits of these efforts will be discussed. 2.1.4. When the People Lead, Leaders Will Follow?

The rationale for NGO conflict resolution activities at the grassroots-level lies in the assumption that such approaches are able to transform the attitudes and behaviour of the broader communities, thereby facilitating an integrative resolution of the conflict and increasing the probability of a sustainable peace. (Barnes 2006:39, Miall et al 2002:75) It is generally assumed that the leadership takes the public ‘mood’ regarding a conflict into consideration, and that the calculations of political and military leaders therefore can be altered by the public. (McKeon 2005:568) Furthermore, it is claimed to be important to prepare the wider population for an eventual peace agreement between leaders. If public opinion is not in favour of a peaceful settlement, the leadership might be reluctant to negotiate a peace accord, out of fear of losing popular support. (Huth & Allee 2002, Ryan 1995:227) It has, consequently, been argued that a durable peace can only be achieved if the civil society includes a significant ‘peace constituency’, i.e. a critical mass of people in support of peace. (Lederach 1997:94, Miall et al. 2002:75, Riegg 2001:20). It has, for example, been claimed that NGOs played a significant role in facilitating the peace process in Northern Ireland by creating ‘peace coalitions’. (Rupesinghe & Anderlini 1998:111) NGOs are furthermore important for overcoming civilian passivity, which Rupesinghe (1995:78-79) identifies as crucial to the peace process. He finds that conflicts tend to passivate the civilian population, allowing violence to continue without civilian restraints. NGOs can therefore be important for empowering people and breaking the spell of passivity. Training and education can help people to view the situation from a new perspective and provide them with

5

new options for dealing with the conflict. This way, NGO activities can introduce new ideas and skills that might make people take constructive action toward resolving the conflict. (Babbitt 1997:366pp) A variety of methods are practiced by NGOs at the grassroots-level. Workshops, trainings, seminars and other activities are conducted with the aim of developing local capacity for conflict resolution, empowering people to take own initiatives to support the peace process, building confidence and facilitating reconciliation between polarised communities. (Babbitt 1997:371, Hoffman 2003:32) What is to be studied in this paper, broadly speaking, is if and why these efforts affect people other than those who directly participate in them. 2.2. Ripple Effects The increased activity of NGOs in conflict resolution has caused some debate about their capacities, legitimacy and possible impact. (Anderson & Olson 2003, Austin et al 2003, Paffenholtz 2007:29, Reimann & Ropers 2005:29 Richmond 2003:1) Although impact at the individual level is frequently reported from NGO activities (Anderson & Olson 2003:76-82), it has been argued that if these activities are to make a difference, knowledge and attitude changes need to be transferred to people beyond those directly participating in them. (Anderson & Olson 2003:54, Ross 2003) As Dirk Sprenger puts it: ‘[conflict resolution] training is not an end but a means’ (Sprenger 2005:2), and the challenge is therefore ‘how to work individually and impact socially’ (Sprenger 2005:5). Impact in this context refers to an activity’s effects beyond its immediate outputs or results, in this case its effects on the broader conflict context or peace process. (Bush 2003:4, Paffenholtz 2005:23) Such impact is usually referred to as ripple effects and its occurrence and causes constitute an acknowledged research gap. (Anderson & Olson 2003, Church & Shouldice 2003:3, Schmelzle & Bloomfield 2006:10, Reimann & Ropers 2005:33) The purpose of this study is to make a modest contribution to filling that research gap by studying whether NGO-strategy3 increases the likelihood of ripple effects. 2.2.1. What Are Ripple Effects?

As mentioned above, ripple effects are the effects of an activity or action beyond their direct outcome. Ilana Shapiro (2005) defines such ripple effects as ‘the impact that participants [in conflict resolution activities] can have on those within their personal and professional spheres of influence.’ She elaborates on the concept and maintains that this term is often discussed in the 3

This concept is defined in subsection 2.4 below.

6

discourse on how micro-level conflict interventions can create macro-level social change, and that individual changes occurring during NGO activities are assumed to cause ‘ever-widening circles of impact as participants take their new learning back into their respective communities and organisations.’ In this context, and in this study in general, the term micro-level refers to the individual level, whereas the term macro-level encompasses the wider society as well as sociopolitical institutions and decision-makers. (See, for example, Anderson & Olson 2003, Schmelzle & Bloomfield 2006) Scholars in this field of research use slightly different wording when speaking about ripple effects. Whereas some authors use the term ‘ripple effects’ (see, for example, Church & Shouldice 2003, Prendergast & Plumb 2002, Shapiro 2005, Spies 2006), other refer to the phenomenon as ‘multiplier effects’ (Rupesinghe 1995:89, Vukosavljevic 2007) or ‘transfer effects’ (Kelman 1995, Fisher 1997b). No consensus on terminology currently exists within the field. In this study ripple effects will be the term used. 2.2.2. Previous Research on Ripple Effects

As mentioned previously, very little research has thus far been conducted to study ripple effects of NGO activities in conflict resolution. (Mitchell 2006:13) One of the most prominent studies of such effects is Herbert Kelman’s (1995) research on the transfer effects 4 of a track two dialogue in the Israeli-Palestinian conflict, which according to Kelman had an impact on the macro-level peace process. Another study of ripple effects is a comparative analysis edited by Fisher (2005). Fisher compares a number of so called ‘successful cases’5 of transfer of effects from the individual level to the political level and the wider society, in order to draw conclusions about what factors cause ripple effects. However, no strong evidence is presented to explain the mechanism causing ripple effects, instead the tentative conclusion of the study is that ripple effects are often assumed to occur as a result of unofficial conflict resolution, but are seldom planned for. (Fischer 2005:223-224) Other studies of ripple effects include an evaluation framework designed by D’Estrée et al (2001) in which ripple effects are studied in different phases. One of the main contributions of their framework is the meso-level concept, meaning an interim step of transfer between the micro- and macro-levels, representing professional organisations, political parties, religious communities, neighbourhoods, villages, or extended family networks. (D’Estrée et al 2001:108) However, the above-mentioned scholarly contributions all focus on cases where NGOs have initiated unofficial conflict resolution processes with policy makers and people involved in 4 5

Kelman’s terminology. The cases are deemed as ‘successes’ by the authors.

7

official negotiations. Hence, they cover a different field than that which this paper attempts to explore, namely NGO conflict resolution activities targeting grassroots. Anderson’s and Olson’s (2003) exploratory research on this topic through the Reflecting on Peace Practice project (Anderson & Olson 2003) is one of few, if not the only, study of ripple effects of such activities. They find that NGOs in general do not plan for ripple effects from their activities, but seem to believe that their efforts will automatically have an impact on the broader conflict context. (Anderson & Olson 2003:13) Building on their tentative findings concerning ripple effects, this thesis will examine whether NGO-strategy increases the likelihood of such effects.

2.2.3. Definition and Delimitations – Ripple Effects as Action Rather than Impact

Assessing and attributing the impact of NGO micro-level conflict resolution activities on the wider conflict context is inherently difficult. NGOs typically seek to affect intangible factors, e.g. attitudes and behaviour, whose concrete impact on the broader conflict environment is hard to measure. Even if attitude changes could be measured, it would be a cumbersome task to isolate their effect on the conflict from the effects of other significant factors.6 (Chigas 2007:562, Gidron 2002:204, Reimann & Ropers 2005:40-41). Whereas it hence would be too big a task to assess macro-level ripple effects within the scope of a master’s thesis, micro-level initiatives to cause ripple effects are possible to examine. As mentioned above, Shapiro (2005) defines ripple effects as ‘the impact that participants can have on those within their personal and professional spheres of influence’. In accordance with this, Church and Shouldice (2003:23) define ripple effects as ‘the process through which change is transferred beyond the direct participants in an intervention’. Furthermore, Anderson and Olson argue that NGO conflict resolution activities are effective ‘if, as a result of an agency’s activities, people undertake independent initiatives, working in creative ways within their own communities to cross lines of division or to influence outside constituencies’. 7 (Anderson & Olson 2003:16) This argument will provide the basis for the definition of ripple effects to be used in this study. However, the definition will be widened to include actions taken to directly affect decision-makers, as such ripple effects should be more likely to affect the conflict context although they might be harder to achieve. In this study it will hence be claimed that ripple effects can be discerned if participants, as a result of the NGO’s activities, take independent initiatives to inform or affect outside constituencies or decision-

6

This problem is sometimes referred to as the ‘attribution gap’. (Paffenholtz 2005:17) Anderson & Olson (2003:17-18) identify three other factors that indicate that an activity is effective: creation or reform of political institutions, people’s resistance to violence and provocations to violence, and increased security for the people. An activity is deemed as more effective the more of the factors that are present.

7

8

makers.8 The term ‘outside constituencies’ refers to people beyond those who have participated in the activity. Using this definition, ripple effects will be assessed as having occurred if the NGO activity, for example, causes participants to spread information about the conflict, or try to affect the views of other people towards the conflict, invite other people to training sessions, organise their own conflict resolution activities or write articles in order to influence public opinion or decision-makers. The dependent variable ripple effects is illustrated below in Figure 1 below. Whereas Shapiro (2005), as explained above, views ripple effects as the impact the individual can have on other people, the definition used in this study implies that the mere action taken, regardless of its impact, is a ripple effect. This is not a display of disagreement with Shapiro’s definition. Rather, the definition used in this thesis is a result of resource and time constraints as well as methodological concerns pertaining to the difficulty of attributing these effects. By narrowing the larger concept of ripple effects down to the first step taken by participants to transfer knowledge, skills or information, ripple effects become researchable within the scope of a master’s thesis. Moreover, if NGO-strategy is found to increase the likelihood of ripple effects, as defined in this study, it should also, as a consequence, increase the likelihood of ripple effects as defined by Shapiro. This logically follows from the fact that ripple effects cannot occur if people do not take initiatives to cause such effects. Even if it cannot be definitely ascertained that every initiative taken has an impact, it seems reasonable to suggest that at least some of them do. To divide ripple effects into different steps in order to facilitate research is a method supported by various scholars in the field. D’Estrée et al (2001:102, 108) assert that striving to achieve structural or societal change directly by targeting small groups is unrealistic. Nevertheless, they claim that it is possible to point to ‘particular indicators’ of intermediate changes at the meso-level9. They argue that ‘[a]ssessment of conflict resolution […] has too often focused only on the two ends of the continuum – individuals and societies.’ (D’Estrée et al 2001:108) In line with this, Church and Shouldice (2003:26) suggest that the ripple effects ‘should be seen in terms of steps rather than as a direct impact from a micro project on the macro-level’. However, they underline that the most significant transfer occurs at the levels closest to that where the activity was pursued, even though activities still can have an impact at other levels. (Church & Shouldice 2003:26) Church and Shouldice also refer to Kelman’s (see subsection 2.2.2.) method of assessing the impact of problem-solving workshops by breaking down ripple effect into ‘links in a chain’ to study the causal relationship. (Church & Shouldice 2003:27) In the same vein, Paffenholtz and 8 Although implicit in the definition, the terms inform and affect are naturally related to conflict resolution. Hence, the participants need to have informed outside constituencies or decision-makers about the conflict or conflict resolution, or tried to affect these people’s attitudes toward the conflict or peace process, for example. 9 The meso-level concept refers to an interim step of transfer between the micro- and macro-levels, see subsection 2.2.2.

9

Reychler (2007:23) view ‘results chains’ as a valuable tool for measuring impact and assessing causal links. The ripple effects studied in this paper can accordingly be viewed as a ‘step’ or a ‘link’ in the transfer process. It lies outside the scope of this thesis to study whether or not these ripple effects in the end have an impact at the macro-level. (See Figure 1 below.)

Figure 1. Dependent Variable - Ripple Effects Initiative to inform/ affect

Affects

NGO

Decision-makers

?

Macro-level impact

Individual Outside constituencies Effect

?

Ripple effects

2.3. NGOs in Conflict Resolution - Hearts without Minds? Although many scholars agree that impact has to be transferred from the individual to the social level if conflict resolution activities are to affect the larger conflict context or peace process10, little research has as yet been conducted to suggest how such ripple effects are realised. (Church & Shouldice 2003:25, Çuhadar-Gürkaynak 2006:9, Schmelzle & Bloomfield 2006:10) What is to be studied in this thesis is the potential causal relationship between NGO-strategy and ripple effects. In this paper, the term NGO-strategy refers to an NGO’s strategy for causing ripple effects from their conflict resolution activities at the grassroots-level. Anderson and Olson (2003) find in their study that most NGOs do not plan for a transfer of the outcomes of their activities to the wider conflict situation. Rather, NGOs seem to believe that their efforts will automatically have a macro-level impact on the conflict, even though they do not specify how the outcome of their projects will transfer beyond the participants. (Anderson & Olson 2003:13) How transfer is to be achieved is often left unarticulated and is commonly founded on assumptions that good intentions will eventually add up to positive effects on the wider conflict context. (Anderson & Olson 2003:64, Lund 2003:21-22, Ross 2003:78) Goodhand and Lewer (1999:81) for this reason refer to NGOs as ‘hearts without minds’ and describe their activities as ‘an anarchy of good intentions’. 10 This is not to say that local conflict resolution efforts or effects on the individual are not important. However, considering the resources spent on these activities and NGOs’ aim to affect the wider conflict context, ripple effects are important and worth striving for. Nevertheless, it should be noted that this study does not intend to evaluate or make any normative judgement about the work of NGOs at the grassroots-level. The aim of this thesis is to study ripple effects, even though some might argue that such effects are not a sine qua none for NGO work at the grassroots-level.

10

Accordingly, this thesis argues that if NGOs do not consciously strive to achieve ripple effects, i.e. have a strategy for achieving ripple effects, such effects are unlikely to occur. 2.3.1. NGO-Strategy Addressed in Previous Research

The possible link between NGO-strategy and ripple effects has been discussed in previous research, although no studies of this causal relationship have been conducted.11 Barnes (2006:84), for example, argues that NGOs must ‘take a more strategic and holistic approach’ to peacebuilding and that it is essential for the impact of peacebuilding that the projects are based on an idea of how their efforts can directly contribute to the wider changes they seek. She further claims that NGO activities will miss their objectives if ‘little attention is given to how the participants can use the experience to support sustained change.’ (Barnes 2006:84) Of relevance is also the ‘Aid for Peace’ approach to planning and evaluation of peacebuilding, which was recently developed by Paffenholtz & Reychler (2007). This approach emphasises the importance of planning for achieving desired effects and the authors argue that NGOs’ planning procedures frequently fail to consider what impacts their activities will have on the peace process at large, but implicitly assume that their efforts will automatically have an effect. (Paffenholtz & Reychler 2007:59) Moreover, Babbitt (1997:372 pp) points out that the goals of any activity should be connected to its possible effect at the macro-level. She exemplifies this with activities that empower individuals and organisations at the local level to take independent action or that build networks for people within the conflict so that they can work together outside the training session. This is supported by Ward and Lekson (2007:358-359), who stress that any program should guide participants in applying skills to concrete problems and that this element should be considered when designing the program. Another interesting perspective on planning and strategy is that of Lederach (1997). He develops a strategic approach to conflict resolution training, emphasising the importance of looking beyond the direct outcome of an activity and consider what impact it will have on the broader conflict. He therefore claims that small-scale conflict resolution activities should be less about content and more about initiating a dynamic process of transformation. (Lederach 1997:108-109) According to Lederach (1997:117) it is crucial that people selected for activities are able to ‘serve as a catalyst and then create a critical mass’ with the capacity of transforming the conflict. He stresses that conflict resolution activities should be designed to generate a process of transformation. (Lederach 1997:112, 140) In line with this argument, this study claims that ripple

11

As far as the author has been able to find, no such studies have been conducted.

11

effects could be seen as such a process of transformation and that conflict resolution activities consequently should be designed to cause ripple effects. Support for the argument that NGO-strategy is important for the occurrence of ripple effects is furthermore offered by Kelman (1995) and his experiences from the workshops he carried out in the Israeli-Palestinian conflict. Kelman (1995:20) explains that the workshops were designed to emphasise the transmission of what was learned in the workshop to the wider public or decision-makers, which he claims has served to influence conflict resolution at the macro-level. Kelman (2002:89-90) describes unofficial conflict resolution as having ‘dual purposes’ of producing changes in individuals as well as a transfer of changes to the political arena, and that the latter is an ‘integral part’ of the approach. To conclude, NGO-strategy has been discussed in previous research on NGO conflict resolution, although the causal relationship between NGO-strategy and ripple effects has not been studied. In order to conduct a field study with the aim of examining this possible causal link, operational definitions of ripple effects and NGO-strategy have to be developed. Having presented and discussed the theoretical framework underlying this study, the following section will be devoted to elaborating the operational definitions of the variables.

12

3. Operationalisations and Assessment Criteria In this section, the operational definitions of the variables will be expounded upon and the criteria used to assess the causal link between NGO-strategy and ripple effects will be explicated. First, the definitions of ripple effects and NGO-strategy will be elaborated, before presenting the control variables to be studied. Lastly, the assessment criteria for the causal relationship between these variables will be outlined. 3.1. Assessing Ripple Effects To study whether ripple effects have occurred, interviews will be conducted with people who have participated in NGO activities. In accordance with the definition of ripple effects presented in subsection 2.2.3, ripple effects will be considered as having occurred if the respondent states that s/he, as a result of the NGO’s activities, has taken independent initiatives to inform or affect outside constituencies or decision-makers. By using the wording ‘as a result of the NGO’s activities’, the definition implies that the participants have to feel that they have learnt something or gained new insights or skills a result of the NGO activity. Hence, two elements need to be present for a ripple effect to be deemed as having occurred: 1. The NGO activity must have had an effect on the participant, i.e. the participant must state that s/he has gained new insights, that her/his attitude or behaviour has changed, that they have attained new skills or that the organisation has given them new ideas on how to work with peacebuilding.12 This is related to the first part of the above-mentioned definition of ripple effects – ‘as a result of the NGO’s activities’ – and is illustrated in Figure 1 (p 10) as an arrow between NGO and individual. 2. The participant has taken independent initiatives to inform or affect outside constituencies or decisionmakers. This is the second part of the ripple effect-definition, illustrated in Figure 1 (p 10) as the arrows between individual and decision-makers/outside constituencies. Participants might for example have sought to influence people in their community (including friends and family), initiated conflict resolution activities or contacted decision-makers. (See also subsection 2.2.3) It should be noted that it is only the transfer of effects from the individual to outside constituencies or decision-makers that is termed a ripple effect. That the NGO succeeds in

This entails that the effect should be that desired by the NGO or any other ‘positive’ effect. A ‘negative effect’ would be if the respondents, for example, would report feeling increased hostility towards the other side. 12

13

affecting the participants is a presupposition that logically needs to be fulfilled for a ripple effect to be possible. Hence, although two elements are identified above as necessary for ripple effects to occur, the first element is simply the effect that the NGO has on the participant, whereas the second is the ripple effect. As follows from the definition above, a ripple effect does not demand that the participants’ attitude and behaviour change. It is possible that the participant is already peacewilling and conciliatory toward ‘the other’, but that s/he through the NGO activity attains new skills in, for example, educating other people in conflict resolution. Hence, if the participant, as a consequence of the new skills attained, takes an independent initiative to educate others, a ripple effect will have occurred. (See, for example, Vukosavjlevic 2007:15) In accordance with this line of argument, the distribution of printed information provided by the organisation will not be classified as a ripple effect. The reason for this is that the trainings do not affect people’s ability to hand out booklets and magazines, wherefore it is hard to claim that the effect of the training is closely related to the distribution of information. Such action could be taken by people who have not even participated in trainings. Printed information can be said to possibly have a direct effect on people who have not participated in trainings, which is different from a ripple effect. It should also be clarified that the participants need not have attempted to change other people’s attitudes or behaviour. The transferring of information concerning the conflict or peace process will also be regarded as ripple effects, as many conflict resolution efforts aim at spreading correct information in order to fill an information vacuum, especially in the IDP community. Considering the limited number of people interviewed from each organisation 13 , the incidence of ripple effects should be rather convincing for such effects to be deemed as prevalent. Consequently, for an NGO to be deemed as causing ripple effects, all participants14 interviewed minus one 15 must state that they have caused ripple effects. It is probable that there will be someone in the sample who differs from the other respondents, without this indicating that ripple effects have not at all occurred. Nevertheless, should the testimony from this one person indicate that ripple effects might not be very frequent, this might alter the assessment of the organisation depending on the strength of the evidence. This principle will be used also to assert that ripple effects have not occurred. Hence, for ripple effects to be deemed as non-existent, all participants minus one must state that they have not been causing ripple effects. Again, the assessment could be altered should the deviating respondent’s answers suggest that ripples are in fact widespread. Highly scattered research results, e.g. two participants have caused ripple effects See the following section. That is all respondents excluding the NGO representatives. 15 In case only two participants are interviewed, both of them have to have caused ripple effects. 13 14

14

and two have not, will be classified as uncertain. Hence, ripple effects will be assessed as occurring, not occurring or uncertain. Furthermore, the NGO representatives will also be enquired about the occurrence of ripple effects as they can contribute with a general view of the organisation’s activities and impact. Their statements can confirm the participants’ accounts or put them into question. Hence, if an NGO representative provides information that strongly and convincingly contradicts the findings from the interviews with the participants, the conclusion about ripple effects will be revised. If their statement confirms the view presented by the participants, the result will be viewed as strengthened. However, it will be taken into account that the NGO representatives may have incentives to present a positive view of the organisation and its performance, wherefore they might overstate the occurrence of ripple effects. 3.2. Assessing NGO-Strategy Building on previous research (see section 2.4.1 above), the hypothesis put forth in this thesis is that NGO-strategy increases the likelihood of ripple effects. By studying whether there seems to be such a causal relationship between NGO-strategy and ripple effects in conflict resolution activities pursued in Georgia, this field study attempts to fill a gap in contemporary research on NGOs in conflict resolution. As mentioned above, the term NGO-strategy refers to an NGO’s strategy for causing ripple effects from their conflict resolution activities at the grassroots-level. To assess whether or not the selected NGOs have such a strategy, representatives for the NGOs will be interviewed. Their information will be complemented by answers given by the participants from their activities. If the answers given by the NGO representatives should be strongly contradicted by the participants, this will affect the assessment of NGO-strategy. The independent variable NGO-strategy is dichotomous – either the NGO has a strategy or it does not. A strategy is deemed as existing if the NGO, as a part of its conflict resolution activities, makes deliberate efforts to promote ripple effects. For example, an NGO will be assessed as having a strategy if it encourages participants to spread information and knowledge to outside constituencies or decision-makers, facilitates transfer of information by enabling the participants to contact people from ‘the other side’ or suggests ways in which the participants should continue to work with conflict resolution after participating in the NGO’s activities. The deliberate training of target groups that are deemed to be able to affect decision-makers or broader segments of the population, such as journalists, is also considered as an NGO-strategy, if the purpose with the NGO activity is to give them skills or knowledge to be used in their

15

profession or in the activity they are associated with. Selection of participants has been underlined in the literature as a crucial element for causing ripple effects. (Anderson & Olson 2003:73, 79; Kelman 2002:91; Lederach 1997:114; Paffenholtz & Reychler 2007:64) Hence, a number of indicators will be studied in order to assess whether the NGOs have adopted a strategy for achieving ripple effects. However, as is explained in subsection 3.1, the mere distribution of printed information is not regarded as a ripple effect. In accordance with this, the providing of booklets or magazines for distribution is not viewed as an NGO-strategy for causing ripple effects (as defined in this study). Rather, printed information is viewed as an NGO’s strategy for causing direct effects through spreading information. 3.3. Control Variables The study also controls for the effect of factors pertaining to the individuals themselves as well as for external factors, i.e. factors that are unrelated to NGO-strategy. These factors, however, will only be discussed at length in the analysis if they should appear to be of any significance for the occurrence of ripple effects. A number of factors relating to the individual are controlled for. Age, gender and profession are factors that can be assumed to affect attitudes and behaviour. It is also possible that previous political activity or experience from other conflict resolution trainings, as well as the duration of the individual’s participation in conflict resolution activities, affect the capacity for and interest in transferring effects. Lederach (1997:52) and Orjuela (2003:200), moreover, acknowledge that participants might be too preoccupied with their day-to-day survival to be able to take independent initiatives, wherefore this factor will also be examined. Questions will furthermore be posed to find out why the participants decided to join the activity, to see if this factor effects their causing of ripple effects. Factors external to the individual and the NGO activities might also affect the participants’ ability to transfer effects. Variations caused by the town or region the NGO is working in are to some extent controlled for by interviewing two organisations on each location.16 Other factors which might affect the individual are constraints caused by the conflict environment. The so-called re-entry problem is frequently discussed in the literature. (Anderson & Olson 2003:23, Barnes 2006:10, Bush 2003:17, Fisher 2005:5, Malhotra & Liyanage 2005:911, Ryan 1995:248) This concept refers to the problems participants in NGO activities face when they return home to their respective communities. They, for example, risk being viewed as traitors (Vukosavljevic 2007:15-16, Wallach 2000:Ch 6) and might be subjected to physical attacks Except in Kutaisi, where only one NGO was visited, as the second organisation contacted was in the end unable to participate in the study. 16

16

(Ryan 1995:248) or social pressure to conform to accepted views (Malhotra & Liyanage 2005:911). As Barnes (2006:10) notes, it is hard for people to engage in conflict resolution ‘when their basic security is threatened’, wherefore the possible effect of this re-entry problem will be controlled for. Other factors, individual or external, might also be found during the interviews. These factors will be discussed if they seem to be significant for the occurrence of ripple effects. 3.4. Assessing the Causal Link If a correlation between NGO-strategy and ripple effects is found in the individual case, the causal link between these variables will be assessed in the within-case analysis. Whereas it would be next to impossible to determine whether the participants’ behaviour is a direct consequence of NGO-strategy, the likelihood of such causality will be elaborated upon. The causal link will therefore be assessed as probable, improbable or uncertain, and not, for example, as ‘confirmed’ or ‘disproved’. The participants will be asked about what they have learnt from the trainings, whether the NGO encouraged them or told them to take their own initiatives and why they decided (not) to take such initiatives.17 If the answers to these questions indicate that a respondent through the NGO activities has realised that s/he has to make an own effort to build peace or transfer knowledge and information, the causal link will be considered as probable. A direct statement from the participants, confirming that their causing of ripple effects was due to the NGOs encouragement, will obviously also entail that the causal link will be regarded as probable. The causal link will, furthermore, be assessed as probable if it is beyond doubt that the NGO-strategy has directly caused the effect, e.g. if the NGO has suggested that the participants take certain action or has provided the participants with the means, e.g. financial or other resources, for causing ripple effects. In case a causal link between no strategy and no ripple effects is to be assessed, statements indicating that the participants’ behaviour could have been altered by an NGOstrategy, will be assessed as indicating a probable causality between the lack of an NGO-strategy and the lack of ripple effects. For example, if the participants have not reflected on their own role in peacebuilding, or seem unaware of the possibility of causing ripple effects or the importance of such effects, this will be assessed as indicating a probable causality. It seems reasonable to suggest

It was deemed as being of little value to pose counterfactual questions to the participants concerning whether they would have acted the same way even if the NGO had/had not encouraged them to, as the answers to such questions are mere guesses from the participants’ side.

17

17

that the NGO could have made such participants cause ripple effects had they only been informed of the importance or possibility of doing so. If there are no statements either supporting or contradicting a causal relationship between NGO-strategy (or the lack thereof) and the occurrence of ripple effects (or the lack thereof), the causal link will be deemed as uncertain. Should the causal link be firmly contradicted, e.g. by participants stating that the NGO was not at all important for the actions they have taken, or NGOs themselves making such statements, the causal link will be deemed as improbable. The causal relationship will naturally also be discussed in the cross-case comparative analysis where other factors than NGO-strategy will be elaborated upon as well.

18

4. Research Design To assess the causal link between NGO-strategy and the occurrence of ripple effects, an interview study will be conducted in the Republic of Georgia (proper and the conflict region South Ossetia). As already mentioned, the impact of NGO conflict resolution is an undeveloped field of research and ripple effects have been acknowledged to be intrinsically difficult to assess. The purpose of the field study is to contribute to theory development by qualitatively studying the effect of NGO-strategy on the occurrence of ripple effects. Such a study is not possible to carry out without consulting the actors on the ground, who are the transmitters of ripple effects. In the choice between an interview study and a survey-questionnaire study (or a combination of the two), the former was deemed as the most appropriate as it should increase the validity of the study (see further below) as well as ensure a high response frequency. (Keeter 2005:159) Moreover a survey-questionnaire study would have been all but impossible to carry out. Firstly, such a survey would have demanded far more cooperation from the NGOs, as they would have had to assist in distributing and collecting the questionnaires. Secondly, due to insufficient participants’ directories it would have been practically impossible to reach out to a significant number of representative respondents, and to ensure a satisfactory response frequency. 4.1. Why Georgia? Due to time and resource constraints, the study had to be limited to only one country. Practicalities such as working up a network of contacts and arranging travels to and within the country demands substantial time and resources, wherefore it would not have been possible to conduct field research in several different countries within the scope of this master’s thesis. There are several reasons for selecting Georgia as the country to study. Firstly, there are currently two unsettled territorial conflicts on the territory internationally recognised as the Republic of Georgia. These conflicts are of a protracted nature and involve identity issues. As was elaborated in section 2, NGOs are believed to be of particular importance for peacebuilding in such conflicts, as much work needs to be done at the grassroots-level. Secondly, international interest in Georgia has been substantial since the break-up of the Soviet Union (Matveeva 2002:425 pp.), and in the aftermath of the Rose Revolution in 2003, which ended Eduard Shevardnadze’s unpopular rule, the interest has further increased. Several international organisations are today involved in conflict resolution activities in Georgia, and local peacebuilding NGOs are plentiful. (Matveeva 2002:433 pp., OUNRC 2006)

However, the

impact of these organisations on the conflicts over Abkhazia and South Ossetia appears to be

19

limited. Nevertheless, there is a substantial IDP population in Georgia proper which possibly could be able to affect the peace process if mobilised. (King 2005:286, Matveeva 2005:416 p., 433 pp.) Given these preconditions, it is interesting to study the possible ripple effects of grassroots conflict resolution efforts in these conflicts. Finally, there are practical reasons for selecting Georgia as the location for the field research, as the researcher is acquainted with the country, speaks Russian and has personal contacts in the capital Tbilisi. Hence, doing the field study in Georgia facilitates the preparations and the conducting of the interviews, which in turn should enhance the quality of the research. 4.1.1. The Conflicts over Abkhazia and South Ossetia

In connection with Georgia’s independence from the Soviet Union, territorial conflicts erupted between the Georgian government and the two regions Abkhazia and South Ossetia. Although ceasefires were reached in 1993 and 1992 respectively, and a partial peace agreement was signed between Georgia and Abkhazia in 1994, the conflicts are still simmering and no political solution to the conflicts has yet been reached. The two regions have become increasingly independent and have functioning de facto authorities, although they have not been internationally recognised. (UCDP 2008) The conflicts involve an ethnic element and the Abkhazian and Ossetian communities on one hand, and the Georgian community on the other hand, have lived isolated from each other since the wars. These communities have become increasingly polarised, entailing deeply entrenched enemy images and stereotypes, wherefore confidence-building, reconciliation and impartial information about the conflict are needed. The conflicts have been referred to as ‘frozen’ and have been more or less stalemated during the last decade, although incidents during the 2000s have destabilised the conflicts temporarily. (See, for example, UCDP 2008, ICG 2004; 2007) However, tension in Abkhazia grew during the time of the field research. The escalation was sparked in April 2008, when Russia declared its intent to formalise diplomatic and economic relations with the de facto authorities in the unrecognised states. This was followed by incidents where at least two unmanned Georgian reconnaissance drones were downed, allegedly by a Russian fighter jet. Adding to the tension, Russia markedly increased its number of peacekeepers in Abkhazia in early May, an act viewed by Georgia as a mobilization of troops. (Akhmeteli 200805-09, ICG 2008) 4.2. A Comparative Approach Seven NGOs were selected for the field research (see below), entailing that seven cases of NGO conflict resolution activities will be studied. To assess whether NGO-strategy increases the

20

likelihood of ripple effects, within-case analysis will be used to examine the correlation between these variables in each of the seven cases. In the event that a correlation is found, the possible causal link between NGO-strategy and the occurrence of ripple effects will be discussed. These seven cases will then be compared with each other in order to further analyse the possible causal relationship. (See section 6.) The study hence has a comparative approach. A correlation between NGO-strategy (or the lack thereof) and ripple effects (or the lack thereof) in the cases examined will be regarded as support for the hypothesis put forth in this thesis, although the assessment of the causal relationship might strengthen or weaken this support. As no data were available before the study, the selection of NGOs could not be made on the basis of variation in the occurrence of ripple effects, which might otherwise have been desirable. Rather, the selection of cases was guided by a will to study ‘typical’ NGOs involved in peacebuilding in order to find out whether or not these NGOs caused ripple effects and if NGOstrategy seemed to be the cause for the possible variation in ripple effects. As will be further discussed in the subsection below, there is a variation between the NGOs studied, regarding target groups and location of activity. This difference among the cases was considered an advantage, rather than a problem, as it strengthens the claim that a correlation found is a consequence of NGO-strategy rather than of factors pertaining to the NGOs themselves. For instance, if NGOs targeting different segments of the population show the same correlation between NGO-strategy and ripple effects, the occurrence of such effects cannot be claimed to be related to a particular target group. However, the possible consequences of this variation on the research results will naturally be discussed in the analysis. 4.3. Selection Process For the study, 3418 respondents from seven NGOs were interviewed. The selection process of the organisations and respondents is further described below. As the research has a comparative approach, various NGOs needed to be studied. A smaller number than seven cases might not have yielded a convincing correlation between the variables studied, considering that the possible variation in ripple effects was not known beforehand. The upper limit was determined by the number of NGOs that were able to participate in the study. (See below.) In order to assess the strategies and ripple effects of each NGO, it was considered necessary to interview approximately six respondents from each organisation.

Two of these 34 respondents had to be omitted from the study after the interviews, as they were in fact representatives for the NGOs rather than ordinary participants. (See subsection 5.3.) 18

21

4.3.1. Selection of NGOs

Two months in advance, emails were sent out to around thirty NGOs, presumably involved in conflict resolution in Georgia, requesting for their participation in the study. Information about these organisations was found in an NGO directory provided by van Tongeren et al. (2002:601607), on a UN list of humanitarian and development contacts (OUNRC 2006) and on NGO websites. Additional NGOs were found through the organisations initially contacted, hence snowball sampling was used to some extent. Out of the organisations contacted, ten were willing or able to participate in the study, whereas the remaining twenty something NGOs were currently not involved in conflict resolution, did not target grassroots, or did not reply to emails or telephone calls. This selection procedure might have entailed a sample of NGOs that is unrepresentative for the entire NGO population in Georgia. It might be the case that the NGOs that agreed to participate were those that knew that their activities were successful or those that have the most resources available, as this may make them more willing to be studied or more able to set aside resources for the study. However, most of the organisations that did not answer the request for participation were those that were not explicitly involved in peacebuilding, e.g. a local Red Cross Society, or organisations whose current activity is uncertain as they were found on outdated websites. Hence, whereas this self-selection bias is inevitable, its possible effect on the study is deemed as marginal. Out of the ten organisations that agreed to participate in the study, only seven were in the end visited. One organisation turned out to be unable to participate, and the two NGOs in Abkhazia could not be visited due to the increased tensions in the region during the time for the research. The seven NGOs visited are located in four different towns in Georgia (proper and South Ossetia) 19 and target different segments of the population, e.g. women, youth and excombatants. These conditions render the study more interesting as they enhance the universality of any conclusions drawn and thereby the generalisability of the study. However, most grassroots NGOs20, in Georgia proper, involved in conflict resolution seem to target IDPs21. Hence the respondents from four of the five organisations in Georgia proper were IDPs. Moreover, five out of the total seven organisations only target women or youth. This is often the case with grassroots activities, as they aim at involving marginalised groups in the peace process, or target people who have special concerns or attributes that are common over the conflict lines. (Barnes 2006:9, Ropers 2004:5) Kutaisi, Tbilisi and Zugdidi in Georgia proper, and Tskhinvali in South Ossetia. All NGOs that were found during the preparatory research for this study, that is. 21 Internally displaced persons. 19 20

22

Furthermore, the organisations all carry out similar conflict resolution trainings, aimed at educating the participants in conflict theory, conflict analysis, and conflict resolution and at informing them about the conflicts they are living in/with. Hence, the activities, whose possible ripple effects are to be studied, are what the NGOs themselves refer to as conflict resolution trainings. Finally it should be mentioned that all NGOs studied are national or community-based organisations financed by international governmental organisations or international NGOs, such as UNDP and International Alert. 4.3.2. Selection of Respondents

Of the 34 people interviewed, 26 were participants and 8 NGO representatives. The aim was to interview six respondents from each organisation, one NGO representative and five participants from conflict resolution activities. 22 Due to time constraints, a larger number of respondents could not be interviewed. However, due to unforeseen incidents, e.g. illness and delayed transports, it was not always possible to interview five participants. The number of participants interviewed from each NGO therefore varies between two and five.23 Interviewing merely two participants from an organisation could be viewed as insufficient for assessing the possible occurrence, or lack, of ripple effects. However, as explained in the previous section, the NGO representatives’ answers will also be taken into account when drawing conclusions about ripple effects, which to some extent might compensate for the small number of participants. Nonetheless, the consequences of this small sample will be further discussed in the analysis. This does not, however, affect the assessment criteria discussed in the previous section, which will be used for all cases in the within-case analysis. All NGO representatives, except those from organisation 1, were NGO leaders, as these were deemed to be able to give the most information. The leader of organisation 1, however, was not involved in the actual trainings, wherefore two trainers were interviewed instead. It should also be noted that merely seven out of 34 respondents were men, although only three out of the seven NGOs explicitly target women. The possible effects of this gender biased sample will be discussed in the analysis. The selection of respondents was not uncomplicated. The intent was to participate during training sessions or seminars and randomly select people24 for the interviews. However, only two of the organisations25 were conducting conflict resolution trainings during the time of 22 There are hence two categories of respondents: participants and NGO representatives. This should be kept in mind when reading this and the following sections. 23 See Appendix 1. 24 Obviously, first-time participants would not have been selected. 25 Organisations 4 and 7. (See Appendix 1 for information about the organisations)

23

the field research, wherefore such a selection of respondents was only used twice. In the other cases, the NGOs had to assist in arranging the interviews, and were hence able to affect the selection of respondents to a certain extent. An effort was therefore made to make sure that the organisations did not solely pick out the most active participants from their previous trainings, which often was their first suggestion. Instead, a variation in age, level of activity, and duration of participation in activities was requested. The selection was further complicated by the fact that some of the organisations did not keep detailed records of (former) participants, wherefore only people with whom the organisation were in contact could be interviewed.26 It is possible that the respondents in these cases were those who are participating the most actively in the organisation’s activities, implying that they are more likely to be causing ripple effects. Nevertheless, this possible selection bias was unavoidable. However, the NGOs were not informed about the exact purpose of the interviews, but were given the general explanation that the aim of the field research was to study NGO activities aimed at peacebuilding. This should to some extent mitigate the risk that the organisations arranged interviews with people they knew had produced ripple effects. 4.4. Interview Structure Interviews were conducted between 1 May and 25 May 2008. For the most part they were carried out in facilities provided by the organisations, although they once had to be conducted at the respondents’ work place and once in the researcher’s home. Before interviewing the participants, general questions about the trainings were posed to the NGO representative in order to get necessary information about the trainings and the NGO’s activities. The interviews with the NGO representatives were conducted after the participants had been interviewed. This way, information valuable for the interviews was obtained from NGO representatives before interviewing participants and vice versa. The respondents were interviewed in person, and the duration of the interviews varied between 30 and 60 minutes. Interviews were conducted in English or Russian depending on what was more convenient for the respondent, and when needed an English-Georgian or Russian-Georgian interpreter was used, however these occasions were rare.27 The inherent risk that any fundamental information was ‘lost in translation’ during interviews is deemed as low, considering that the interview questions were rather straightforward and generally rendered straightforward answers. Moreover, the interpreters were familiar with the topic studied. 26 Organisations 1 and 3. (See Appendix 1) Organisation 2 conducts trainings quite rarely and has generally carried out trainings with very specific purposes, e.g. to enable participants to put up a puppet theatre on conflict resolution. The organisation arranged interviews with three participants from the two most recent trainings. 27 Two respondents from each of organisations 1 and 7.

24

A semi-structured interview process was employed, utilising standardised questions but allowing for control questions and a flexible order of the questions. The semi-structured process increases the validity of the study as it enables the interviewer and respondent to elaborate on the subject at hand. In order to increase the reliability of the study and the comparability between the objects of analysis, standardised questions 28 were used and each question was posed to every respondent.29 Such a process furthermore ensured that all the variables of interest for the study were examined. One way of increasing the validity and reliability of interview studies is to use control and follow-up questions. (Kvale 2007:88-89) Hence, to be sure that a question was correctly interpreted by the respondent, and to extract as much information as possible from her/him, follow-up questions were frequently used. Control questions were also used to make sure that the respondent’s answers were correctly interpreted. Hence, the reply, as interpreted by the interviewer, was often repeated to the respondent. Moreover, the respondents often answered many of the interview questions before the actual question was posed. In these cases, control questions were used to make sure that the respondent was correctly understood. Before the interviews, the respondents were informed about the general purpose of the study, i.e. NGO peacebuilding activities. To reduce the risk of respondents giving the answers they thought were sought, they were not given complete information about the variables studied. Participants were told that the interviews aimed to find out more about their experiences from participating in the NGOs activities, and NGO representatives were simply informed that they would be interviewed about their peacebuilding activities. Different questions were, naturally, posed to the participants and the NGO representatives. The participants were primarily asked about their own actions and incentives for taking/not taking independent initiatives after partaking in conflict resolution activities, while the NGO representatives were mainly enquired about their strategies. However, both categories of respondents were probed for information on all variables studied: ripple effects, NGO-strategy and the control variables. 4.5. A Note on Transcription The interviews were recorded with the permission of the respondents, and notes were taken as a complement. The interviews were then transcribed and translated into English as soon as See Appendices 2 and 3 for questionnaires. Questions 18, 20 and 21 on questionnaire A and questions 3 and 5 on questionnaire B were only posed if the previous question/s did not yield elaborate replies. Moreover, as is apparent from the questionnaires (see Appendices 2 and 3), the wording of the questions was slightly altered depending on the answers to previous questions. (See, for example, questions 12 and 22 on questionnaire A) 28 29

25

possible after they had been completed with all respondents from the same organisation. Lengthy personal stories recounted during interviews were not transcribed, as such transcription would have been time-consuming, and sometimes ethically questionable, whereas these stories did not contribute with any valuable information to the study. It should be underlined that only information which indisputably was unrelated to the NGO activities and the variables studied was excluded in the transcription process. For example, the participants, having lived with the conflict for almost two decades, often referred to tragic incidents and losses of family members during interviews. Whereas information provided as a direct answer to the interview questions was naturally never excluded, the details of tragic incidents were excluded as they were not relevant for the research.30 Had this study had an exploratory purpose this way of transcribing could not have been employed. However, as the purpose of this thesis is to see if any support can be found for the hypothesis that NGO-strategy increases the likelihood of ripple effects, this transcription technique is believed to be useful and sufficient. 4.6. Limitations of the Field Study A qualitative study based on interviews conducted in merely one country, albeit in two conflict contexts, implies limits to the generalisability of the results. Studying NGOs from different regions and with different target groups is a way of increasing generalisability, although it is possible that there are certain characteristics pertaining to the country or region itself that could affect the occurrence of ripple effects. Nevertheless, seeing as very little research has addressed this issue, this study should be able to contribute to the research on ripple effects even though it is limited in scope, as it can offer insights and tentative conclusions about the causal link between NGO-strategy and ripple effects. The conclusions should, however, preferably be limited to NGOs working with conflict resolution in protracted territorial conflicts, as this is the type of conflict addressed by the Georgian organisations. It is, for example, plausible that the type of conflict affects the individual’s possibility to cause ripple effects. Moreover, there is a reliability problem inherent in doing interview research. The answers to the questions posed might to some extent be affected by the interviewer entailing that different interviewers will get different answers and hence different research results. However, using standardised questions should mitigate this problem. (Kvale 2007:86) It is, furthermore, plausible that a respondent will give untruthful answers, either to please the interviewer or to appear more effective or active than s/he or the organisation in reality is. (Kvale 2007:14) This For example, respondent 10 (see Appendix 1) explained that she started taking part in conflict resolution activities after her son was killed. This piece of information was naturally included in the transcription, while the horrifying details of how her son was killed, and how she found out that he had been killed, were omitted. 30

26

validity problem should be somewhat reduced by the anonymous nature of the interviews. (Keeter 2005:137 p) The respondent’s activities or thoughts will then not be ‘judged’ by anyone else than the interviewer, a fact that should decrease the respondent’s incentives to ‘boost’ their abilities and activities. Although the respondents might anyway choose to enhance reality, it should at least give them less incitement to do so than if, for example, the participants knew that the organisation they are associated with might read their replies. Furthermore, it may be argued that the possible selection bias discussed in section 4.3.2. could affect the validity of the study, as there is a risk that the study is not assessing the ripple effects of NGO conflict resolution activities in general, but rather the ripple effects caused by the NGOs’ most active participants. Although this risk could not be completely eliminated, measures were taken to reduce it. Nonetheless, the consequences of this possible bias will be discussed in the analysis. Moreover, considered that the number of respondents is rather small, especially in the cases were only two participants could be interviewed, only tentative conclusions can be drawn from this study.

27

5. Research Results and Within-Case Analysis In this section, the research results for each of the seven cases will be presented and analysed separately, using within-case analysis. The aim is to present as many quotes as possible, without making the text incomprehensive.31 The results will thus be exemplified by illustrative quotations from the respondents, and contradictory findings and answers that put the hypothesis into question will always be quoted without exception. The results for the independent and the dependent variables are presented and analysed under separate headings for each organisation, and the analysis of the possible causal link between these variables is found in the end of each subsection.32 The research results are summarised in Table 1 on page 42. As explicated in subsection 3.1., two elements need to be present for a ripple effect to be assessed as having occurred: the NGO activity must have had an effect on the participant, and the participant must have taken independent initiatives to inform or affect outside constituencies or decision-makers. The former is a precondition for the latter and such effects were reported from all participants interviewed. The occurrence of these effects will therefore not be elaborated in the analysis. Hence, whenever a ripple effect is stated to be present, effects have been reported from the respondents, which is often obvious from the quotes included in the analysis. The people interviewed, referred to as the respondents, are either NGO representatives (NGO leaders in all cases but one) or participants, i.e. people who have taken part in conflict resolution trainings. The respondents were given anonymity, but their gender, age and the organisation whose trainings they participated in can be found in Appendix 1. 5.1. Organisation 1 5.1.1. NGO-Strategy

The NGO has a strategy for causing ripple effects. Respondent 1, an NGO representative, explains that they begin ‘each training […] with asking every woman […] what she herself suggests she could do’ and the other NGO representative states that they ask the women ‘what they themselves are doing to facilitate a peaceful resolution of the conflict, so that they understand their own role in the peace process.’ (Respondent 2) Hence, an emphasis is put on each participant’s own role in the conflict resolution process. The participants also state that the organisation has told them to speak to other people about what they have learnt. Respondent 3, Including Russian original quotes was deemed as superfluous and page-consuming, wherefore all quotes are presented in English, even when originally expressed in Russian. 32 The results for NGO-strategy will be presented before ripple effects in each subsection, as the strategy in several cases provide details that explain the ripple effects, wherefore the analysis is more comprehensive this way. 31

28

for example, maintains that the NGO has encouraged her to spread her knowledge ‘because it’s their goal that people become more peaceful, so they want us to talk to other people.’ The NGO, moreover, encourages the participants to phone Abkhazia and to travel there to meet with Abkhazians, and assist them with the practicalities, i.e. telephone connection and mini-grants to finance the trip. This is a clear example of a strategy for spreading information across community lines, and thereby causing ripple effects.33 5.1.2. Ripple Effects

There are examples of ripple effects occurring from this organisation’s trainings. The two participants interviewed have been spreading information to people in their communities and have invited new people to join the activities. Respondent 4 says: ‘[w]hen people have other opinions than me about the conflict, I tell them what I’ve learnt from the organisation. I also talk about the new information I have about the conflict.’ The NGO representatives also state that the participants have told them that they use their new skills in their daily lives. Respondent 2, for example, states that the participants ‘also work as “propagandists” in their own homes and communities where they can also reach the men.’ The participants furthermore describe how the organisation has helped them and other participants to phone and travel to Abkhazia, which supposedly has led to that ‘stories are being spread in both communities that “the other” also wants peace’, according to one of the NGO representatives (respondent 1). Hence, even though only two participants from this organisation were interviewed, it seems as though other participants have also caused ripple effects. If the organisation through its activities encourages and enables participants to contact and/or visit their former home region, and thereby spread the message to ‘the other side’ that Georgians do not want war, this is an obvious example of ripple effects. 5.1.3. The Causal Link between NGO-Strategy and Ripple Effects

That the NGO encourages the participants, and provides them with the possibility, to make phone calls and trips to Abkhazia is a clear example of ripple effects caused directly by the NGO’s strategy. As for the other ripple effects identified, there is no clear indication that the NGO has in fact caused these through their strategy. It could, however, be argued that the NGO’s emphasis on the participants’ own role in the peace process may have increased the likelihood of the participants taking own initiatives to spread their knowledge, although this causal link cannot be confirmed. Had this been a part of the NGOs trainings it could have been regarded as a form of reconciliation rather than as a ripple effect. However, as this is a separate and ‘voluntary’ activity, it will be regarded as a part of the NGO’s strategy. 33

29

To conclude, the NGO’s strategy has undoubtedly caused some of the ripple effects, but it is uncertain whether the NGO is the reason that participants have been spreading information to other people in their community. Hence, the causal link in this case will be assessed as uncertain/probable, indicating that there are two separate paths of ripples from the trainings, one closely related to the NGO-strategy, and one whose causal relation to strategy remains unclear. 5.2. Organisation 2 5.2.1. NGO-Strategy

The NGO has a clear strategy for causing ripple effects. The NGO leader invites teachers or other people that will be able to carry out peacebuilding projects in schools, for example, although she has also arranged summer schools for children. The trainings of adults often have the specific purpose of training the participants so that they can partake in subsequent projects or conduct conflict resolution trainings on their own. The NGO leader states that the participants often ‘are trained to do certain things, for example teachers are trained to teach conflict resolution in their schools.’ (Respondent 5) Hence the aim with the trainings seems to be to reach beyond the participants. Moreover, the NGO leader states that they ‘tell the participants that they themselves must take every opportunity to spread the message at all levels in society, and that they have to try to work independently with conflict resolution.’ 5.2.2. Ripple Effects

Ripple effects are occurring in this case. The participants interviewed were selected for trainings with the purpose that they would carry out specific peacebuilding activities with school children, which they also have done. Respondents 6 and 7 have put up a puppet show with a peacebuilding theme. Respondent 8 has conducted conflict resolution trainings in the school where she is a teacher. She also reports having changed her pupils’ attitudes through these activities: ‘I didn’t think that I would be able to change them, but they have become more and more positive and peace-willing. […] The children say that they are grateful [for this]’. However, she states that it is hard for the children to in turn try to affect their parents, who are still hostile. The NGO leader also claims that what the participants learn is ‘transferred through word of mouth’ and that children, who have participated in summer schools organised by the NGO, talk to their parents about what they have learnt and that the parents in turn thank her for this. (Respondent 5)

30

5.2.3. The Causal Link between NGO-Strategy and Ripple Effects

The causal relationship is obvious in this case. The NGO selects participants with the explicit aim of training them so that they can carry out specific peacebuilding activities. Respondents state that it was the NGO leader’s idea that they carry out these activities. Respondent 7 states that ‘it was all thanks to [the NGO leader]. It was her idea to do the show. That was the reason we participated in the training and put up the show.’ Respondent 8 states that the leader of the NGO invited her to the training, so that she ‘could then teach the children something about the conflict.’ She maintains that she ‘didn’t expect the training to have this effect on’ her and that she ‘didn’t think that [she] would become a peacebuilder’, but that she has ‘understood that we can make peace ourselves, in our everyday life’. She also says that she has been given new instruments for talking with the children and continues: ‘I can teach them that we, ourselves, can build peace.’ Hence, although sceptic at first the respondent was affected by the trainings, and used her new knowledge and skills for their predefined purpose, i.e. to educate her pupils. It seems probable that the NGO’s strategy of training people with the aim of making them conduct specific activities have made many participants, even among those not interviewed, cause ripple effects. It is, however, hard to evaluate the possible ripple effects caused by children participating in summer schools. Nevertheless, considering that the organisation’s trainings target only adults, the causal link is in this case deemed as probable. 5.3. Organisation 3 5.3.1. NGO-Strategy

The NGO does not have an explicit strategy for causing ripple effects. The NGO leader herself states that she does not ‘carry out any special activities or take any special action to encourage’ participants to spread what they have learnt. (Respondent 9) Furthermore, the NGO leader appears to be unaware of the desirability of spreading information as she does not mention the spreading knowledge as a way of affecting the peace process. Moreover, the NGO leader states that when selecting participants what is most important is ‘that they are good people and that they are tolerant’, hence she does not explicitly select people who are likely to be able to make an impact outside of the trainings. (Respondent 9) However, some of the respondents state that they were selected because of their previous experience of conflict resolution or other society work. That the NGO leader did not mention this as a criterion for her trainings might indicate that such selection is made only on rare occasions, and that it probably is not a part of the NGO’s strategy.

31

5.3.2. Ripple Effects

The research results concerning ripple effects are unfortunately distorted by the fact that respondents 11 and 12 turned out to be members of the organisation since ten years back, and are paid to work for the organisation. They are hence not ordinary participants of training sessions and cannot be said to be representative for the overall group of participants. These respondents were therefore omitted from the study, since they were in fact NGO representatives rather than participants. The analysis of the participants’ view is therefore based on the interviews with respondents 10, 13 and 14 who are possibly more representative. None of them have taken any initiatives to transfer their knowledge or use their skills. Respondent 10 states that she has been talking to people in her village, not about the things she has learnt through the organisation, however, but about her own experiences from losing her son in the war. Moreover, she says that she ‘was talking to people before the organisation contacted’ her. She can hence not be said to have caused ripple effects, as defined in this study, since her activity is unrelated to the NGO trainings. The answers given by respondents 13 and 14 are very similar to each other. Both have told their friends that they have participated in trainings, but have not spread any knowledge or information, hence they have not caused ripple effects. Respondent 13, for example, has told her friends that she has ‘been to these trainings’, but has not ‘talked to them exactly about what’ the organisation was teaching them. They have not spoken to anyone else about the trainings and have not taken any other initiatives to cause ripple effects. Concerning ripple effects caused by other former participants, the NGO leader asserts that ‘[t]here are examples of women who have started their own organisations’. However, when asked if her activities have had an impact on people outside the training, she answered: ‘To be quite honest, I don’t know.’ The same answer was given when she was asked whether she thinks that the participants have been trying to spread their new knowledge to people in their communities. The NGO leader’s statement suggests that ripple effects might not be very common, even if there might of course be the random instance of such effects. To conclude, based on these results, ripple effects are assessed as not occurring. 5.3.3. The Causal Link between NGO-Strategy and Ripple Effects

The respondents claim that the trainers did not say anything about spreading what they learnt during the training sessions. However, when asked whether they have tried to spread their knowledge to, or affect the views of, their friends respondents 13 and 14 both say that they have not done so, as their friends ‘know a lot about the conflict already’ (respondent 14) or are ‘very peaceful’ (respondent 13), and hence do not need to learn more about the conflict. Hence, there 32

are conditions unrelated to the NGO that explain why these respondents have not been causing ripple effects. Nonetheless, these circumstances do not offer an explanation to why the respondents have not taken any other initiatives to spread what they have learnt. Concerning affecting other people than their friends, respondent 13, for example, states that she has not been trying to affect her parents, as ‘it’s harder for them and their generation to see the conflict from a new perspective. And I think they should be allowed to have their own view.’ She has not been talking to anyone else about what she has learnt and says that she does not know who she should talk to. Respondent 14 says that she has not tried to change the views of anyone else than her friends either. When asked why, she answered: ‘I haven’t really thought about it I guess.’ The respondents hence seem unaware of their own role as information transmitters, in stark contrast to the respondents of most of the other organisations who underline the importance of spreading information and doing something themselves. Not knowing who to talk to or not having thought about transferring knowledge, are examples of information failures that could have been remedied by the NGO. The behaviour of these respondents could probably easily have been altered, if the NGO had told them that they should talk to other people. However, as will be shown, some of the respondents from the other organisations have primarily caused ripple effects by talking to their friends. Hence, it is plausible that the fact that these respondents’ friends were already well-informed and peaceful is the reason behind the lack of ripple effects. Respondent 10 has personal reasons for wanting to work with conflict resolution, but does not seem to have been affected much by the NGO trainings. Moreover, considering that she states that she talks to people about things that she thinks ‘are more important’ than what she has learnt from the NGO, an NGO-strategy for causing ripple effects might not have altered her behaviour. The lack of ripple effects of her part would hence not be due to the lack of an NGOstrategy. However, had the NGO stressed the importance of spreading their message to more people, it is possible that the respondent would have forwarded also this information and knowledge, besides talking to people about the ‘more important’ things. To conclude, the findings do not decisively point to a causal link between NGOstrategy and ripple effects, although such a link is possible. The causal relationship is therefore deemed uncertain.

33

5.4. Organisation 4 5.4.1. NGO-Strategy

The NGO undoubtedly has a strategy for causing ripple effects. The NGO leader describes that she believes that the organisation can affect the peace process by creating a peace constituency: ‘[w]hat we do is that we widen the circle of people who want a peaceful resolution to the conflict. If the participants spread what they learn so that this circle gets bigger and bigger, then this will promote a peaceful resolution.’ She asserts that they ‘educate […] people so that they themselves will be “propagandists” in their own communities. So that they will become peacebuilders’ and she further states that the trainers, ‘give [the participants] the tools and tell them that they themselves have to spread their knowledge.’ They ‘also tell [the participants] to tell other people about [their] activities’. This is confirmed by the participants. Respondent 19, for example, states that ‘[t]hey tell us to be active and to do things and not just talk about our problems.’ Furthermore, during a training session visited in connection with the interviews, the trainer was explaining for the participants that three levels are important in conflict resolution: the grassroots-level, the middle-level (e.g. NGO leaders, academics) and the top-level (decisionmakers). (See, for example, Chigas 2007:555, Lederach 1997:39) She emphasised the importance of transferring information between these different levels and explained that the participants themselves are a part of the grassroots-level and that it is important that they try to affect the toplevel and the middle-level in order to have an impact on the peace process. The trainer also underlined that it is important to put up your own concrete goals in peacebuilding, as opposed to just think about what you would like to do to build peace. She suggested that the participants create their own women’s groups where they can discuss and come up with ideas that should be transferred to decision-makers. Moreover, the organisation provides the participants with the possibility to use the internet to contact women in Abkhazia, in order to create links with the Abkhazian side. The NGO leader calls this ‘facilitation of contact between Abkhaz and Georgian women’ and through the internet the participants are expected to make ‘contact with Abkhazian women and [tell] them that Georgians don’t want war’. (Respondent 15) This is clearly an elaborate strategy to cause ripple effects.34 Finally, beside IDP women, the NGO deliberately targets journalists and teachers, as these are believed to be able to affect the conflict context and children respectively. This is also an indication of a strategy.

As explained in subsection 5.1.1., as these activities lie outside of the actual trainings, they are viewed as ripple effects rather than as, for example, reconciliation activities. 34

34

5.4.2. Ripple Effects

Ripple effects are deemed as occurring. All participants have invited other people to trainings, and four out of five state that they have spread their knowledge. Respondent 20, for example, is a teacher and has been talking to her ‘pupils to show them that there is a way out of this conflict.’ Respondent 16 says that during a conflict resolution training in school she could use her new skills and tell the others ‘that it is possible to solve the conflict peacefully, and that it’s possible to have a dialogue with the Abkhazians’. Moreover respondent 16 maintains that she has ‘succeeded in changing’ the views of other people. The exception is respondent 17, a journalist, who says that she has not had the possibility to use her new knowledge in her work so far, since she has not been able to visit Abkhazia where she usually works as a correspondent. She has, however, talked to her friends about the conflict, although she has not tried to spread any knowledge or information to them. Her ‘friends are very peaceful’ and ‘know a lot about the conflict’ she thinks, hence she has merely told them about what they ‘had been doing during the training’. Moreover, she states that she has not had the time to take any own initiatives to spread what she has learnt to other people than her friends, although she would like to. The NGO leader, moreover, states that there are examples of people spreading knowledge. When asked if the participants have spread what they have learnt, she said: ‘I know they have. Each month there are more and more people coming to our meetings so it’s obvious that the participants have been talking about our activities and that they have invited new women. The circle gets wider and wider.’ Furthermore, she mentions that the participants through the organisation have been able to contact Abkhazian women to make them understand that Georgians do not want war. ‘IDPs have visited Abkhazia and we give IDPs the possibility to contact Abkhaz women via the internet. This way our beneficiaries35 spread the message to them that we don’t want war and that Georgians only want a peaceful resolution to the conflict.’ Moreover, the NGO leader asserts that teachers have been teaching their pupils conflict resolution. However, it is uncertain if journalists have used their supposedly new knowledge in their work. 5.4.3. The Causal Link between NGO-Strategy and Ripple Effects

The four respondents who have spread their knowledge state that the NGO has told them to speak to other people about what they have learnt during the training sessions and that they should invite other people to the trainings. Respondent 16 states that already on inviting her to

35

Beneficiaries are referred to as participants in this study. (Author’s comment)

35

the trainings, ‘[the organisation] said […] that we would learn a lot and that we then could spread the information’. That the spreading of knowledge was from the beginning presented as the aim of the training might have made the participant more prone to do so. Respondent 20, however, underlines that she herself naturally understood that this information is important to spread to other people, implying that even if the NGO had not said anything about spreading information, she would have acted the same way. ‘I understand myself that it’s important to talk to other people about it’. Hence, the organisation’s strategy was in this case not crucial for the spreading of knowledge, which to some extent contradicts the hypothesis studied. However, that some individuals will cause ripple effects regardless of whether the NGO tells them to or not, is rather expected. The question is whether the general picture seems to point to that NGO-strategy makes ripple effects more likely. Respondent 17, who is a journalist, has not spread any knowledge, but has invited people to the trainings. She states that she has not been able to use her new knowledge in her work as a correspondent because of the current restraints on working in Abkhazia. Nevertheless, relating to NGO-strategy, she says that to use the knowledge in her work ‘was the whole point of the training. That was why I was invited.’ Hence, she is well aware of what is expected of her, but external factors have prevented her from causing ripple effects through her work. Furthermore, she has not caused any ripple effects among her friends, since she does not seem to think that they need to be educated. Neither has she spread her knowledge to anyone else, as she has not had the time. However, she states that she would like to do so. It hence seems as though external factors have prevented the respondent from spreading information, although she seems aware of the importance of doing so. The NGO-strategy can therefore not be said to have failed, as the respondent is aware of that she should be using her knowledge in her work, which was the reason she was selected for the training. The causal link is therefore not deemed as weakened by this respondent’s relative failure to cause ripple effects. However, the respondent puts focus on the impact of external factors on the occurrence of ripple effects. It is obvious that there is a causal link when the NGO-strategy has implied the provision of internet contacts with Abkhazian women. However, there are no other strong indications that the NGO-strategy is the primary reason for the participants’ spreading of knowledge. Respondent 20 furthermore puts the causal link into question, stating that she would have spread information even if the NGO had not told her to. Whether this would really have been the case is impossible to say. However, this remark brings to mind that not everyone needs to be encouraged to spread knowledge. Similar to organisation 1, this NGO causes ripple effects

36

through two different paths, whose causal link to the NGO-strategy differ. Consequently, the causal relationship will also in this case be deemed as uncertain/probable. 5.5. Organisation 5 5.5.1. NGO-Strategy

The NGO has an explicit strategy for causing ripple effects. The NGO leader maintains that she believes she can affect the peace process by training young people who in turn talk to friends and family and travel around the country and spread their knowledge. That way the trainings affect more people. The NGO leader states that they ‘begin each training session with writing down the principles for the training’ and that ‘item number three on that list is “the spreading of knowledge”’. (Respondent 21) Hence the spreading of knowledge is one of the fundamental principles of the conflict resolution trainings, and participants are told at the very first training session that they should spread what they learn. These principles are also referred to by respondents 23 and 25 when talking about how the NGO encourages them to spread information. Moreover, respondent 22 states that ‘[the NGO leader] told us that we have to speak to everyone: kids, young people, our parents, and she told us to invite people to trainings […] She told us to try to affect people that have a different opinion than us’. 5.5.2. Ripple Effects

The participants have been causing ripple effects. The NGO leader argues that ‘if I train one person, I have affected 15 people’ (respondent 21), as the participants spread the knowledge they obtain. She says that the students talk to other students at the university and inform them about the trainings. Furthermore, the NGO leader maintains that the participants also ‘conduct their own conflict resolution trainings at home with their families’. (Respondent 21) As for the participants, they all state that they have spread what they have learnt during the training sessions. They have spoken to parents and friends and invited people to trainings. Respondent 23 furthermore states that she used her skills when she visited Abkhazia and managed to change the attitudes of her Abkhazian neighbours. ‘I used the skills I had learnt in the trainings and listened to their opinions and I managed to change their attitudes by using my conflict resolution skills’, she says. Respondent 25 and 26 also report having changed the views of friends and family. Respondent 24, on the other hand, has talked to other people but states: ‘I’ve tried to change their attitudes but it’s hard. They don’t believe what I say’. Although this respondent is still viewed as causing ripple effects, as defined in this thesis, it is interesting to note that the respondents have different experiences from trying to affect people.

37

5.5.3. The Causal Link between NGO-strategy and Ripple Effects

The participants all state that the trainer has told them that they must talk to other people about what they have learnt. Two of the respondents also refer to the principles of the training, explained above, and state that one of the first things they were told was that they have to spread information. These principles and the respondents’ referring to them indicate that the spreading of information is a fundamental part of the trainings, which might make it more likely that the participants are affected by the NGO-strategy. As respondent 23 states: ‘the first thing we did on the first training session was to go through the principles of the training and one of the principles was that we should spread the information.’ Moreover the respondents also maintain that they have learnt through the trainings that they themselves have to be active in conflict resolution, as opposed to leaving the peace process to the governments. Respondent 26, for example, states that she has learnt that they themselves ‘have to participate in the conflict resolution process’, and respondent 25 says: ‘we must be active and participate. It is important that we speak our mind and say that we want a peaceful resolution.’ This might suggest that the respondents would not have taken action themselves, had it not been for the NGO-strategy of stressing the importance of doing so. Overall the respondents seem very aware of that they have learnt something and that they should spread this to other people. This awareness of a ‘duty’ to spread information and knowledge possibly indicates that it is indeed the organisation’s strategy that has made them take initiatives that they otherwise would not have taken. Respondent 26, for instance, states that it is the NGO leader who has ‘taught [them] what to say and do to spread the information to other people’ and continues to say that she ‘[a]t first […] didn’t think that it would make any difference, [she] didn’t think that [she] would affect anyone’, but now she has seen results. Hence, it is unlikely that she would have taken any initiative to affect people if it had not been for the NGO’s encouragement. Considering that the respondents clearly refer to the NGO’s principles and encouragement concerning spreading knowledge and taking action, and their explicit awareness of the importance of transmitting information, a causal link between NGO-strategy and ripple effects will in this case be assessed as probable. 5.6. Organisation 6 5.6.1. NGO-Strategy

The NGO has a strategy for causing ripple effects. The NGO leader states that they tell the participants to talk to other people about what they have learnt during the training sessions and that they expect such spreading of information from the participants. They also encourage

38

participants to write articles. The respondents confirm that the NGO has told them to spread their knowledge to other people. According to respondent 29, the trainers said that the participants ‘need to make more people understand what the consequences of war would be’ and that a peaceful resolution of the conflict is better. 5.6.2. Ripple Effects

The NGO representative asserts that people are telling him that they have talked to their friends and neighbours, and maintains that those living in IDP collective centres can gather up to 50 people at once who they can address. Moreover, he claims that there are several examples of people writing articles for newspapers. Accordingly, the two participants state that they have told their friends about what they have learnt and that they have tried to change the opinion of those who are supportive of a military solution to the conflict. They furthermore claim that some of the people they have talked to have changed their views. Respondent 28 says that he has ‘tried to change those who were propagandising war’ and that he thinks that ‘they have changed after hearing what [he has] told them.’ Respondent 28 has also invited other people to the trainings and has made phone calls to Abkhazia. Respondent 29 says that ‘[t]here were those who were saying that we need to start a new war and that that is the only way to resolve the conflict’ and that he has ‘tried to explain to them that war will do no good.’ According to him ‘many of the people [he has] talked to have changed their views.’ Hence, ripple effects are deemed as occurring, as the two participants have caused ripple effects. However, it should be kept in mind that only two participants could be interviewed from this organisation, entailing that the assessment of this organisation is based on rather limited material, even though the NGO representative, who might be overstating, also maintains that ripple effects have occurred. 5.6.3. Causal Link between NGO-Strategy and Ripple Effects

Respondent 28 asserts that the trainings ‘woke [him] up’ and made him realise that he himself could do something and that he ‘can’t only sit and hope that something will happen or that someone else will do something.’ According to him, the trainers said that the participants ‘must do something’ themselves and he furthermore states they were discussing what every person can do, for example speak to their friends, and that he ‘understood that this was [his] responsibility’. These statements suggest that there is a causal link between NGO-strategy and ripple effects, seeing as the respondent clearly states that the trainings have made him understand his own role in working with conflict resolution. Respondent 29, however, does not as clearly state that the NGO’s encouragement has been important for his spreading of knowledge. Since no other 39

participants could be interviewed, and since the NGO representative does not give any strong evidence strengthening the causal relationship, the causal link between NGO-strategy and ripple effects is uncertain. 5.7. Organisation 7 5.7.1. NGO-Strategy

The NGO does not have a strategy for causing ripple effects through their trainings. Anyone can apply for the trainings because the NGO leader thinks it is important that everyone has ‘at least a minimum of understanding for the situation.’ Furthermore, she does not have ‘very high expectations’ on the participants to take action after the trainings. When asked about whether they encourage the participants to spread what they have learnt, she stated: ‘No, that they do by themselves. We don’t have to tell them to. Each society is based on communication, so they automatically talk to their friends about what they’ve learnt.’ This statement is especially interesting as it confirms the findings of previous studies of ripple effects, namely that NGOs simply assume that the impact of their activities will automatically transfer to other people than the participants. (See, for example, subsection 2.4.) 5.7.2. Ripple Effects

Three out of four participants state that they have not taken any own initiatives to spread what they have learnt, hence ripple effects are assessed as non-occurring. Respondent 31, on the other hand, states that he has spoken to his friends about what he has learnt, as he ‘usually talks to [his] friends about what [he has] been doing.’ He also helped the NGO arrange an information meeting at his university ‘when the trainers asked [him] to’. However, he has not tried to change people’s views about the conflict, as he thinks that ‘it’s hard to change the views’ of Georgian people. Respondent 33 says he has not taken any initiatives yet because ‘[t]he trainings are not finished yet, so [he has not] learnt everything’. He claims that he might talk to people when he has finished his training. The NGO leader maintains that she ‘thinks’ that the trainings have had an impact on people other than those who have participated in the trainings. She argues that ‘the IDP community isn’t that big in Georgia’ and that ‘young people share their views with their friends and discuss these issues’. ‘So indirectly [she thinks] that the trainings affect other people’. She cannot give any concrete examples of such spreading of knowledge, but states that she knows that the participants talk to their friends at school. Contrary to other NGO representatives,

40

however, she does not state that the participants have told her that they have been talking to other people. Hence, her statements about ripple effects are mere guesses. Moreover, she asserts that young people ‘can’t affect their parents’, since their parents have very strong opinions. This is supported by respondent 31 who states that he knows his friends ‘can’t really talk to their parents about [a peaceful resolution to the conflict]’. 5.7.3. Causal Link between NGO-Strategy and Ripple Effects

It seems as if there might be a causal link between the lack of NGO-strategy and the relative lack of ripple effects. Respondent 32 states that ‘the trainers are better at teaching people about the conflict, they know much more than I do.’ Such an opinion indicates that the importance of everyone contributing to conflict resolution has not been made clear during trainings. It seems reasonable to suggest that the trainings could have changed his understanding of his own role and made him realise that he himself can indeed transfer knowledge about the conflict. Respondent 33 has not yet caused any ripple effects. When asked if he has talked to his friends or other people about what he has learnt he claims that he thinks that what he has learnt is ‘personal’ and that he does not think that he knows ‘what to tell other people about this’. He also says that he has not learnt everything yet, as the training programme is not finished, and that he might talk to people when he has left the programme. However, this is a one-year programme and the respondent has already participated for six months and states that he has learnt things and that he has changed. Other respondents in this study have participated in fewer training sessions, but have still told people about what they have learnt. Hence, it seems strange that he would not be able to tell people about what he has learnt, particularly as he himself feels that he has changed. It is likely that if the trainers had been talking about the importance of spreading knowledge and information, the respondent would have understood that what he has learnt is something that is interesting and important also for other people to know, and that he should transfer this knowledge. Respondent 34, who has not caused any ripple effects, states: ‘[the trainers] only teach us about the conflict. What we want to do with that information is up to us.’ It seems probable that he would have had a different view on what to do with the information if the trainers had in fact told him that he should spread it to other people in his community. The ‘outlier’ of this case is respondent 31, who has indeed been talking to his friends, even if he has not tried to change their opinion as he finds this hard to do. He states that the reason for talking to his friends about what he has learnt is that they, like him, are IDPs and that the information is interesting for them. Hence, the lack of an NGO-strategy does not entail that

41

no ripple effects at all occur. However, it should be noted that this respondent has participated in conflict resolution activities before, organised by another NGO.36 It is possible that his previous experiences from those activities explain his deviating behaviour. The causal link is to some extent weakened by the fact that respondent 31 has been causing ripple effects in spite of the lack of NGO-strategy. Nevertheless, based on the discussion on the other respondents’ statements, the causal link is assessed as probable.

Table 1. Research results

36

Organisation 1

Ripple Effects Occurring

NGO-Strategy Yes

2 3 4

Occurring Not occurring Occurring

Yes No Yes

5 6 7

Occurring Occurring Not occurring

Yes Yes No

Causal Link Uncertain/ probable Probable Uncertain Uncertain/ probable Probable Uncertain Probable

This respondent is in fact the only participant who has previous experiences from conflict resolution activities.

42

6. Comparative Analysis and Conclusions In Table 1 on the previous page, the results from the within-case analysis are presented. In this concluding chapter, the research results will be further analysed and conclusions about the possible causal link between NGO-strategy and ripple effects will be elaborated. Other factors that might have affected the research results will also be expounded upon. The last subsection is devoted to a brief summary of the study and the conclusions.

6.1. Conclusion about Correlation and Causality The research results presented above support the hypothesis underlying this study, as they show the expected correlation between NGO-strategy and ripple effects. In the five cases where ripple effects were assessed to have occurred, the NGO had a strategy for causing such effects. In the two cases were ripple effects were absent, there was no such NGO-strategy. This correlation suggests that there might be a causal relationship between the variables studied. It is, however, hard to assess this causal link. The assessment in this study has been made with great caution and the causal relationship has been categorised as probable, uncertain or, for organisations causing ripple effects through two different channels, a combination of both.37 That NGO-strategy (or the lack thereof) has caused ripple effects (or the lack thereof) was deemed as probable in three out of seven cases. If the uncertain/probable cases are included, five out of seven cases show a causality deemed as probable. Hence, only two cases (organisations 3 and 6) showed no or little indication of a causal relationship between the variables studied, although such a relationship could not be disproved in these cases. Furthermore, the causal link was not contradicted or rejected in any of the cases studied. Only one respondent, number 20, explicitly stated that her behaviour was not caused by the NGO. To conclude, this study suggests that NGO-strategy might increase the likelihood of ripple effects, as there is a correlation between the two variables. However, the findings do not firmly support the causal relationship between NGO-strategy and ripple effects, even if causality is probable in at least three cases and was not discarded in any of the cases. However, as stated previously, hefty generalisations should not be made from the research results as they are based on a rather small, and possibly unrepresentative, sample of participants from conflict resolution trainings.

37

The classification ‘improbable’ was not used.

43

6.2. Other Explanations Although the research results support the hypothesis put forth in this thesis, there is a possibility that other factors than NGO-strategy affected the research results. The factors identified will be discussed below. 6.2.1. The Unexpected Result – Frequent Occurrence of NGO-Strategy and Ripple Effects

The overall picture projected in this study is that NGOs in general have a strategy for causing ripple effects and that such effects are occurring relatively frequently.38 This result is somewhat surprising, considering the academic scepticism concerning ripple effects. The result expected was that most NGOs would not have a strategy and would not be causing ripples. It should therefore be asked whether this study puts the general view into question or if the results are explained by other factors. Firstly, the finding that the majority of the NGOs had a strategy for causing ripple effects demands some attention. The NGOs, in spite of the fact that the interviews were anonymous, might have had an interest in presenting themselves in a positive light and may therefore have wanted to exaggerate the weight they put on the spreading of information and knowledge. However, it seems unlikely that the NGOs during interviews would have fabricated details about their trainings. Moreover, the existence of an NGO-strategy was confirmed by the participants. Secondly, the ripple effects seen in five out of seven cases should be commented. As previously discussed, the participants may have wanted to make themselves appear more active than they in fact are. However, many of these respondents provided trustworthy and elaborate recounts on how, why and when they had been causing ripple effects, wherefore their answers are believed to be reliable. Moreover, not all participants stated that they had been causing ripple effects, which indicates that the respondents might not have felt a need to ‘boost’ their own performance. Furthermore, those who had not taken any such initiatives had all been trained by NGOs without a strategy. It is unlikely that the participants from those NGOs would simply, coincidentally, answer the questions more truthfully than the participants who said that they had caused ripple effects. Hence, the correlation found between NGO-strategy and ripple effects is not deemed to be significantly affected by respondents giving untruthful answers. However, as previously discussed, there is a risk that the interviews arranged by the NGOs themselves, although with a rather detailed request on who they should select, entailed a certain bias in favour of ripple effects. The relatively frequent occurrence of ripple effects might 38

It should also be mentioned that no ‘negative ripple effects’, as discussed in footnote 12 on page 13, were reported.

44

therefore be a consequence of this selection bias, which unfortunately could not be completely avoided. Nevertheless, the respondents from organisation 4, who were selected by the researcher without the help of the NGO, had all caused ripple effects, whereas the respondents from organisation 3, who were selected by the NGO, had not caused any such effects. This indicates that such a selection bias might in fact not be present in the sample studied. 6.2.2. Alternative Explanations for the Lack of Ripple Effects in Two Cases

The lack of ripple effects from organisations 3 and 7 need further elaboration. At least two other possible explanations to the results from organisation 7 can be discussed. Firstly, the participants in this training session, and consequently in the interviews, were all male. It could therefore be claimed that the lack of ripple effects is due to the respondents’ gender rather than to NGOstrategy. However, respondent 31 from organisation 7 has indeed caused ripple effects, as has the other two men interviewed in the study (participants from organisation 6). Furthermore, the other organisation that has not caused ripple effects, organisation 3, target only women. Hence, gender is not a viable explanation for the variance found in this study. Still, as the proportion of men in this study was small, it would be interesting to examine the gender perspective more closely with a different sample of participants. Secondly, the respondents from organisation 7 were all selected by the researcher, without the help of the NGO, which was the case with five of the other organisations. It is therefore possible that these respondents were more ‘representative’ than the respondents from the majority of the NGOs39, and that the same result would have been obtained for most of the organisations if the respondents had all been selected the same way. However, the respondents from organisation 4 were also selected this way, but had in fact caused ripple effects. Moreover, although the selection of participants from organisation 3 was largely made by the NGO, their participants had not caused ripple effects. Hence, the way of selecting respondents does not seem to be correlated to the occurrence of ripple effects. Furthermore, it should be underlined that external factors can, obviously, affect the individual respondent. Due to tensions in Abkhazia, for example, respondent 17, a journalist usually working as a correspondent in the conflict region, has not been able to cause ripple effects to the extent that she might have otherwise. Moreover, this respondent points to lack of time as the reason why she is not talking to more people about what she has learnt. Finally, as is apparent from the interviews with respondents 13, 14 and 17, ripple effects might be less common when people in the participants’ personal or professional sphere of As discussed in subsections 4.3.2. and 4.6., there is a risk that the NGOs arranged interviews with only very active respondents.

39

45

influence are already peaceful and well-informed and hence do not need to be informed or influenced. To affect people outside your immediate social network demands a greater effort than speaking to your friends and family does. Most of the respondents, who had caused ripple effects, had done so by speaking to their friends or family. However, this explanation is not applicable to those respondents from organisation 7 who had not been causing ripple effects. 6.2.3. Differences between Conflict Regions and Georgia Proper?

The possible differences between conflict resolution efforts in Georgia proper and the conflict regions need to be elaborated upon, wherefore it is unfortunate that Abkhazia could not be visited. During the interviews with the NGOs in Georgia proper, some respondents expressed a belief that peacebuilding activities are harder in Abkhazia, since the political climate does not permit discussions on peace and reconciliation. Respondent 1, for example, states that ‘[i]t’s dangerous to talk about reconciliation in Abkhazia and to claim that Georgians want peace’, a view supported by Respondent 29 who claims that ‘[i]t’s hard to talk about peace in Abkhazia. They’re not allowed to talk about living together with us.’ Other respondents also indicated that Abkhazians face a risk when discussing politics or if they would say that they have been talking to Georgians. Such unfavourable conditions are logically plausible, as the democracy level in an unrecognised state might be lower than in the state it de jure belongs to. (Sambanis 2000:463) It would therefore have been interesting to study ripple effects in Abkhazia, as such effects might be less likely under such unfavourable circumstances. However, the research results from the conflict region visited, South Ossetia, does not indicate any such difficulties in carrying out peace work. Even though one of the NGOs had not caused any ripple effects, none of the respondents expressed any concern over political restraints on discussing conflict resolution. Respondent 10 had even been talking about reconciliation on the local radio and television. Moreover, respondent 8, a teacher, has been discussing conflict resolution with her pupils, which indicates that these questions are not ‘forbidden’ in the public discourse. Although the situation might be different in Abkhazia than in South Ossetia, it might simply be the case that the respondents in Georgia proper, affected by nearly two decades of stereotyping, believe the government of ‘the other side’ to be adverse to conflict resolution and reconciliation. Moreover, even though information about the democracy level in these regions is hard to obtain, recent reports on this topic give no support to the claims that civil society would fare substantially worse in Abkhazia than in Georgia. (ICG 2006:15, Popescu 2006:16)

46

6.2.4. Control Variables

Beside the gender question discussed above, none of the control variables yielded any results. One of the most interesting findings, regarding the control variables, is perhaps that the neither the frequency nor the duration of participation in trainings seems to affect the occurrence of ripple effects. The respondents who have not spread ripple effects have all taken part in a series of conflict resolution trainings, stretching over three to six months. The participants who have spread ripples have various experiences from conflict resolution trainings – some have taken part in only one training, whereas others have taken part at several occasions. Nor can it be argued that it is important that the trainings are ‘finished’ as respondent 31 puts it. Although this respondent claims that he might take own initiatives after finishing the series of trainings, this has not been important for the respondents from organisations 4 and 5, most of whom are also currently participating in trainings. The study does not systematically control for the effect of the scope of the change in the respondent, or the level of skills attained through trainings, on the likelihood of ripple effects. However, these factors do not seem to have been significant. Both those who assert that they have changed completely after the trainings, and those who simply state that they have learnt new information about the conflict, have caused ripple effects. 6.3. Affecting Outside Constituencies and Decision-Makers? It is interesting to note that of the twelve participants who have caused ripple effects by trying to change the attitudes of other people towards the conflict40, ten report having been able to change the attitudes of the people they have been talking to. For example, respondent 8, a teacher, states that she ‘didn’t think that [she] would be able to change [the pupils], but they have become more and more positive and peace-willing’. Furthermore, respondent 22 explains that ‘[s]ome of [her] friends were hostile towards Abkhazians’ and that she has ‘tried to change them and some of them have different opinions now.’ This means that these respondents have in fact had an impact on outside constituencies and not just taken action as suffices for the definition of ripple effects in this study. However, three participants state that they think it is hard to change people’s attitudes. Respondent 24, for example, claims that she has ‘tried to change [other people’s] attitudes but it’s hard’ because ‘[t]hey don’t believe what [she says]’. Hence, the actions assessed as ripple effects in this study might not always have had an impact on people beyond the participants. Moreover, even though most respondents claim to have affected other people, this does not necessarily mean 40

As opposed to solely transmitting information or putting up a puppet theatre.

47

that their initiatives have in fact made an impact. Consequently, even though ripple effects, as defined in this study, have occurred and might have been caused by NGO-strategy, it cannot be ascertained that they have indeed affected outside constituencies. 41 Nevertheless, as argued in section 2.2.3, that participants take action to inform or affect other people increases the likelihood of affecting these people, even if the actual impact is not assessed in this study. It should also be noted that none of the participants have tried to inform or affect decision-makers. Neither have they started up their own organisations or taken any initiatives to carry out their own conflict resolution activities. The ripple effects found in this study were in the form of spreading of knowledge and information, initiatives to affect other people’s42 views on the conflict or ‘the other’, and inviting/encouraging other people to participate in conflict resolution activities. This is perhaps not surprising as the most undemanding and obvious way to cause ripple effects is to speak to people around you about what you have learnt. However, ripple effects would probably have a greater impact if immediately transferred to the general public or the top-level. Interesting to note is that the NGO-strategies in general do not envision the participants’ spreading knowledge to people beyond their own community.43 The organisations primarily encourage participants to speak to people within their immediate sphere of influence, rather than to contact decision-makers or take more far-reaching initiatives aimed at conflict resolution. This emphasis can be viewed as a logical consequence of the fact that their main purpose is to work with grassroots. Had the organisations targeted other groups, for example NGO leaders, emphasis would probably have been on influencing also decision-makers and public opinion. It could still be considered whether a change in NGO-strategy, entailing an emphasis on influencing public opinion in general or decision-makers, would make people take such initiatives, or if ordinary grassroots would still be unlikely to cause ripple effects through those channels. It does, however, seem likely that people in general would not try to affect public opinion or decision-makers, or that those who wish to affect the conflict more broadly would prefer to simply join the organisation and work with conflict resolution through them, rather than take independent action.

As will be discussed below, the respondents have not tried to affect decision-makers. Only people within their personal or professional sphere of influence and not public opinion in general. 43 The exception is perhaps organisation 4 which discusses the importance of influencing the middle- and top-levels during trainings. 41 42

48

6.4. Concluding Remarks The aim of this study was to contribute to theory development in the field of NGO conflict resolution by studying whether NGO-strategy increases the likelihood of ripple effects. Whether or not NGO peacebuilding efforts at the grassroots-level are transferred to people beyond those directly involved in these activities has been a matter of scholarly debate. Previous research on NGOs in conflict resolution suggested that ripple effects from NGO peacebuilding activities might be more likely if the NGO has planned for such effects to occur. As very little research has been conducted in this field, this study aimed to fill a research gap by assessing whether NGOstrategy increases the likelihood of ripple effects. For this purpose, field research was conducted in Georgia (proper and the conflict region South Ossetia) during May 2008. 34 respondents from seven different NGOs involved in conflict resolution were interviewed, entailing that seven cases of NGO conflict resolution activities were studied. The interview results were analysed using both within-case and comparative cross-case analysis. The study supports the hypothesis that NGO-strategy increases the likelihood of ripple effects, even if the causal relationship could not be definitively determined. It should be kept in mind, however, that the study was limited in scope and possibly involved an unrepresentative sample of respondents, wherefore far-reaching conclusions should not be drawn. Nevertheless, there is a correlation between the two variables, which motivates further research on the causal relationship. The findings consequently suggest that NGOs working with grassroots conflict resolution should adopt a strategy for causing ripple effects, if they wish for such effects to occur. This study cannot, however, offer any conclusions regarding the possible macro-level impact of these ripple effects. Still, the study indicates that NGO conflict resolution activities might indeed be the drop that ripples the ocean.

49

7. References Aall, Pamela R.; Helsing, Jeffrey W. & Tidwell, Alan C. (2007) ’Addressing Conflict through Education’ in Zartman, William I., editor. Peacemaking in International Conflict: Methods & Techniques, Washington D.C.: United States Institute of Peace Press Akhmeteli, Nina (2008-05-09). ‘Georgia: US and EU Support for Tbilisi Grows Amid Escalating Tension with Russia’, Eurasianet.org, URL: http://www.eurasianet.org/ (Accessed 2008-05-14) Anderson, Mary B. & Olson, Lara (2003). Confronting War: Critical Lessons for Peace Practitioners. Reflecting on Peace Practice Project. The Collaborative for Development Action, Inc. Babbitt, Eileen F. (1997). ‘Contributions of Training to International Conflict Resolution’, in Zartman, William I. & Rasmussen J. Lewis, eds., Peacemaking in International Conflict: Methods & Techniques, Washington D.C.: United States Institute of Peace Press Barnes, Catherine (2006). Agents for Change: Civil Society Roles in Preventing War & Building Peace. (Issue Paper 2), Amsterdam: European Centre for Conflict Prevention Bush, Kenneth (2003). Hands-on PCIA. Part I A Handbook for Peace And Conflict Impact Assessment (PCIA), URL: http://action.web.ca/home/cpcc/attach/Hands-On%20PCIA%20-%20Handbook%20X%20--%20BUSH%20Final%20Author%5C's%20Version1.pdf (Downloaded 2008-03-27) Chigas, Diana (2007). ‘Capacities and Limits of NGOs as Conflict Managers’ in Crocker et al, Leashing the Dogs of War: Conflict Management in a Divided World, Washington D.C.: United States Institute of Peace Press Church, Cheyanne & Shouldice, Julie (2003). The Evaluation of Conflict Resolution Interventions, Part II: Emerging Practice and Theory, Derry/Londonderry: INCORE International Conflict Research (University of Ulster & The United Nations University) Çuhadar-Gürkaynak, Esra, (2006). "Towards improving the tools of assessment: expanding the d'Estrée et al. framework for problem solving initiatives" (Paper presented at the annual meeting of the International Studies Association, Town & Country Resort and Convention Center, San Diego, California, USA, Mar 22, 2006) URL: http://www.allacademic.com/meta/p98212_index.html (Downloaded 2008-03-31) Davies, John & Kaufman, Edy (2002). ‘Second Track/Citizens’ Diplomacy: An Overview.’, in Davies, John & Kaufman Edward (Edy), eds, Second Track/Citizens’ Diplomacy: An Overview. Concepts and Techniques for Conflict Transformation. Lanham: Rowman & Littlefield Publishers, Inc. D’Estrée, Tamra Pearson; Fast, Larissa A.; Weiss, Joshua N. & Jakobsen, Monica S. (2001). ‘In Theory: Changing the Debate about “Success” in Conflict Resolution Efforts’, Negotiation Journal, April 2001. Diamond, Louise & McDonald, John (1996). Multi-Track Diplomacy: A Systems Approach to Peace. (Third Edition), West Hartford: Kumarian Press

50

Fischer, Martina (2006). ‘Civil Society in Conflict Transformation: Ambivalence, Potentials and Challenges’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31) Fisher, Ronald J. (1997a). ‘Training as Interactive Conflict Resolution: Characteristics and Challenges’, International Negotiation, Issue 2, pp 331-351 Fisher, Ronald J. (1997b). ‘The Potential Contribution of Training to Resolving International Conflict’, International Negotiation, Issue 2, pp 471–486 Fisher, Ronald J. (2005). ‘Introduction: Analyzing Successful Transfer Effects in Interactive Conflict Resolution’, in Fisher, Ronald J. ed., Paving the Way: Contributions of Interactive Conflict Resolution to Peacemaking, Oxford: Lexington Books. Fitzduff, Mari & Church, Cheyanne eds. (2004). NGOs at the Table: Strategies for Influencing Policies in Areas of Conflict. New York: Rowman and Littlefield Gidron, Benjamin; Katz, Stanley N. & Hasenfeld, Yeheskel, editors (2002). Mobilizing for Peace: Conflict Resolution in Northern Ireland, Israel/Palestine and South Africa, New York: Oxford University Press, Inc. Goodhand, Jonathan & Hulme, David (1999). ‘From Wars to Complex Political Emergencies: Understanding Conflict and Peace-Building in the New World Disorder’, Third World Quarterly, Vol 20. No.1, pp 13-26 Goodhand, Jonathan & Lewer, Nick (1999). 'Sri Lanka: NGOs and peace-building in complex political emergencies', Third World Quarterly, Vol 20. No.1 Hall, Michael (2004) The Search for Conflict Resolution: Lessons Drawn from a Community Development Strategy, Island Pamphlets 61, Belfast: Regency Press Hoffman, Mark (2003). ‘PCIA Methodology: Evolving Art Form or Practical Dead End?’, in Austin, A.; Fischer, M. & Wils, O, eds., Peace and Conflict Impact Assessment: Critical Views on Theory and Practice. Berghof Handbook Dialogue Series, Issue No. 1, Berlin: Berghof Research Center for Constructive Conflict Management. URL: http://www.berghof-handbook.net (Downloaded 2008-03-27) Huth, Paul K. & Allee, Todd L., 2002, ‘Domestic Political Accountability and the Escalation and Settlement of International Disputes’ The Journal of Conflict Resolution, Vol. 46, No. 6, pp 754-790 International Crisis Group, ICG (2007). Abkhazia: Ways Forward, Europe Report No 179, Tbilisi/Brussels, URL: http://www.crisisgroup.org/ (Downloaded 2008-05-12) International Crisis Group, ICG (2006). Abkhazia Today, Europe Report No 176, Tbilisi/Brussels, URL: http://www.crisisgroup.org/ (Downloaded 2008-06-08) International Crisis Group, ICG, (2004). Georgia – Avoiding War in South Ossetia. Europe Report No 159, Tbilisi/Brussels, URL: http://www.crisisgroup.org/ (Downloaded 2008-05-12)

51

Keeter, Scott (2005). ‘Survey Research’, in Druckman, Daniel, ed., Doing Research: Methods of Inquiry for Conflict Analysis, Thousand Oaks: SAGE Publications Kelman, Herbert C. (1995). ‘Contributions of an Unofficial Conflict Resolution Effort to the Israeli-Palestinian Breakthrough’, Negotiation Journal, Vol. 11, pp 19-27 Kelman, Herbert C. (1997). ‘Social-Psychological Dimensions of International Conflict’, in Zartman, William I. & Rasmussen J. Lewis, eds., Peacemaking in International Conflict: Methods & Techniques, Washington D.C.: United States Institute of Peace Press Kelman, Herbert C. (2002). ‘Interactive Problem Solving as a Tool for Second Track Diplomacy’ in Davies, John & Kaufman Edward (Edy), eds, Second Track/Citizens’ Diplomacy: An Overview. Concepts and Techniques for Conflict Transformation. Lanham: Rowman & Littlefield Publishers, Inc. King, Charles (2005). 'The Uses of Deadlocks: Intractability in Euroasia' in Chester A. Crocker, Fen Osler Hampson & Pamela R. Aall, eds, Grasping The Nettle: Analyzing Cases Of Intractable Conflict Washington, D.C.: United States Institute of Peace Press Kvale, Steinar (2007). ‘Doing Interviews’, in Uwe Flick ed., The SAGE Qualitative Research Kit, Los Angeles, London & New Dehli: SAGE Publications. Lederach, John Paul (1997). Building Peace: Sustainable Reconciliation in Divided Societies. Washington, D.C.: United States Institute of Peace Press Lund, Michael (2003). What Kind of Peace is Being Built? Taking Stock of Post-Conflict Peacebuilding and Charting Future Directions. A Discussion Paper, (Prepared on the occasion of the tenth anniversary of Agenda for Peace for the International Development Research Centre (IDRC), Ottawa, Canada), URL: http://www.idrc.ca/uploads/user-S/10527469720lund_final_mar_20.pdf (Downloaded 2008-03-31) Malhotra, Deepak & Liyanage, Sumanasiri (2005). ‘Long-Term Effects of Peace Workshops in Protracted Conflicts’, Journal of Conflict Resolution, Vol 49, No 6, pp 908-924 Matveeva, Anna (2002). ‘Georgia: Peace Remains Elusive in Ethnic Patchwork’ in van Tongeren et al, Searching for Peace in Europe and Eurasia: An Overview of Conflict Prevention and Peacebuilding Activities. London: Lynne Rienner Publishers, Inc. McDonald, John W. (2002). ‘The Need for Multi-Track Diplomacy’, in Davies, John & Kaufman Edward (Edy), eds, Second Track/Citizens’ Diplomacy: An Overview. Concepts and Techniques for Conflict Transformation. Lanham: Rowman & Littlefield Publishers, Inc. McKeon, Celia (2005). ‘Civil Society: Participating in Peace Processes’, in van Tongeren, Paul; Brenk, Malin; Hellema, Marte & Verhoeven, Juliette, People Building Peace II: Successful Stories of Civil Society. London: Lynne Rienner Publishers, Inc. Miall, Hugh; Ramsbotham, Oliver &Woodhouse, Tom (2002). ‘Calling for a Broad Approach to Conflict Resolution’ in van Tongeren, Paul; van de Veen, Hans & Verhoeven, Juliette, editors, Searching for Peace in Europe and Eurasia: An Overview of Conflict Prevention and Peacebuilding Activities, Boulder: Lynne Rienner Publishers, Inc.

52

Mitchell, Christopher R. (2006). ‘Conflict, Social Change and Conflict Resolution. An Enquiry.’, in Bloomfield, David; Fischer, Martina & Schmelzle, Beatrix, eds., Social Change and Conflict Transformation Berghof Handbook Dialogue Series, Issue No. 5, Berlin: Berghof Research Center for Constructive Conflict Management. URL: http://www.berghof-handbook.net (Downloaded 2008-03-27) Natsios, Andrew S. (1997). ‘An NGO Perspective’, in Zartman, William I. & Rasmussen J. Lewis, eds., Peacemaking in International Conflict: Methods & Techniques, Washington D.C.: United States Institute of Peace Press Office of the United Nations Resident Coordinator (OUNRC) (2006). Humanitarian and Development Contacts Tbilisi, URL: www.ungeorgia.ge/files/upload/Contact/dir_dec.pdf (Downloaded 2008-03-05) Orjuela, Camilla (2003). ‘Building Peace in Sri Lanka: A Role for Civil Society?’ Journal of Peace Research, Vol. 40, No. 2, pp 195-212 Paffenholtz, Thania (2004). ‘Designing Transformation and Intervention Processes’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31) Paffenholtz, Thania (2005). ‘Third-generation PCIA: Introducing the Aid for Peace Approach’ Schmelzle, Beatrix ed., New Trends in Peace and Conflict Impact Assessment (PCIA), Berghof Handbook Dialogue Series, Issue No. 4, Berlin: Berghof Research Center for Constructive Conflict Management. URL: http://www.berghof-handbook.net (Downloaded 2008-03-27) Paffenholtz, Thania & Reychler, Luc (2007). Aid for Peace: A Guide to Planning and Evaluation for Conflict Zones. Baden-Baden: Nomos. Popescu, Nicu (2006). Democracy in Secessionism: Transnistria and Abkhazia’s Domestic Policies, CPS International Policy Fellowship Program 2005/2006, Budapest, Hungary: Open Society Institute, URL: http://www.policy.hu/news/NPopescu-PS/22 (Downloaded 2008-06-05) Prendergast, John & Plumb, Emily (2002). ‘Building Local Capacity: From Implementation to Peacebulding’ in Stedman et al, Ending Civil Wars: The Implementation of Peace Agreements, London: Lynne Rienner Publishers, Inc. Reimann, Cordula (2004). ‘Assessing the State-of-the-Art in Conflict Transformation’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31) Reimann, Cordula & Ropers, Norbert (2005). ‘Discourses on Peace Practices: Learning to Change by Learning from Change?’ in van Tongeren et al, 2005, People Building Peace II: Successful Stories of Civil Society. London: Lynne Rienner Publishers, Inc. Richmond, Oliver P. (2003). ‘Introduction: NGOs, Peace and Human Security’, in Carey, Henry F. & Richmond, Oliver P., eds., Mitigating Conflict: The Role of NGOs, London, Portland: Frank Cass Publishers

53

Riegg, Natalya Tovmasyan (2001). Conflicts in the Second World: A View on Track 2 Diplomacy, Working Paper No. 17, George Mason University, Institute for Conflict Analysis and Resolution Ropers, Norbert, (2004). ‘From Resolution to Transformation: The Role of Dialogue Projects’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31) Ross, Marc H. (2003). ‘PCIA as a Peacebuilding Tool’, in Austin, A.; Fischer, M. & Wils, O, eds., Peace and Conflict Impact Assessment: Critical Views on Theory and Practice. Berghof Handbook Dialogue Series, Issue No. 1, Berlin: Berghof Research Center for Constructive Conflict Management. URL: http://www.berghof-handbook.net (Downloaded 2008-03-27) Rupesinghe, Kumar (1995). ‘Conflict Transformation’, in Rupesinghe, Kumar ed. Conflict Transformation, London: St. Martin’s Press Rupesinghe, Kumar & Anderlini, Sanam B. (1998). Civil Wars, Civil Peace: An Introduction to Conflict Resolution. International Alert, London: Pluto Press Ryan, Stephen (1995). ‘Transforming Violent Intercommunal Conflict’, in Rupesinghe, Kumar ed., Conflict Transformation, London: St. Martin’s Press Sambanis, Nicholas (2000). 'Partition as a Solution to Ethnic War: An Empirical Critique of the Theoretical Literature', World Politics No. 52, pp 437-483. Schmelzle, Beatrix (2006). ‘Training for Conflict Transformation – An Overview of Approaches and Resources’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31) Schmelzle, Beatrix & Bloomfield David (2006). ‘Approaching Social Change’, in Bloomfield, David; Fischer, Martina & Schmelzle, Beatrix, eds., Social Change and Conflict Transformation, Berghof Handbook Dialogue Series, Issue No. 5, Berlin: Berghof Research Center for Constructive Conflict Management. URL: http://www.berghof-handbook.net (Downloaded 2008-03-27) Shapiro, Ilana (2005). ‘Theories of Change’ (Knowledge Base Essay), Beyond Intractabiliy.org, URL: http://www.beyondintractability.org/essay/theories_of_change/?nid=1256 (Accessed 2008-04-14) Spies, Chris F. J. (2006). ‘Resolutionary Change: The Art of Awakening Dormant Faculties in Others’, in Bloomfield, David; Fischer, Martina & Schmelzle, Beatrix, eds., Social Change and Conflict Transformation Berghof Handbook Dialogue Series, Issue No. 5, Berlin: Berghof Research Center for Constructive Conflict Management. URL: http://www.berghof-handbook.net (Downloaded 2008-03-27) Sprenger, Dirk (2005). ‘The Training Process: Achieving social impact by training individuals? How to make sure that training for conflict transformation has an impact on conflict transformation.’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31)

54

Tschirgi, Necla (2003). ‘Peacebuilding as the Link between Security and Development: Is the Window of Opportunity Closing?’ Strengthening the Security-Development Nexus: Conflict, Peace and Development in the 21st Century, Security-Development Nexus Program International Peace Academy Studies in Security and Development, New York URL: http://www.un.org/esa/peacebuilding/Library/Peacebuilding_as_link_IPA.pdf (Downloaded 2008-04-24) Uppsala Conflict Data Program, UCDP (2008), Georgia: Abkhazia/South Ossetia, URL: www.ucdp.uu.se/database (Accessed: 2008-05-12) Van Tongeren, Paul; van de Veen, Hans & Verhoeven, Juliette, editors (2002) Searching for Peace in Europe and Eurasia: An Overview of Conflict Prevention and Peacebuilding Activities, Boulder: Lynne Rienner Publishers, Inc. Van Tongeren, Paul; Verhoeven, Juliette & Wake, Jim (2005). ’People Building Peace: Key Messages and Essential Findings’, in Van Tongeren, Paul; Brenk, Malin; Hellema, Marte & Verhoeven, Juliette, eds., People Building Peace II: Successful Stories of Civil Society. London: Lynne Rienner Publishers, Inc. Vukosavljevic, Nenad (2007). ‘Training for Peacebuilding and Conflict Transformation. Experiences of the “Centre for Nonviolent Action” in the Western Balkans’, in Fischer, Martina & Schmelzle, Beatrix eds. Berghof Handbook for Conflict Transformation, Berlin: Berghof Research Center for Constructive Conflict Management, URL: http://www.berghof-handbook.net/ (Downloaded 2008-03-31) Wallach, John (2000). The Enemy Has a Face: The Seeds of Peace Expercience, Washington D.C.: The Endowment of the United States Institute of Peace Ward, George F. & Lekson, Michael J. (2007). ‘Dealing with Conflict: The Contributions of Training’, in Zartman, William I., ed., Peacemaking in International Conflict: Methods & Techniques, Washington D.C.: United States Institute of Peace Press

55

Appendix 1 – Tables of Organisations and Respondents Organisations Org. No. Location

Target Group

Date of Interviews

1 Kutaisi IDP women 2 Tskhinvali* Teachers, journalists, children 3 Tskhinvali* Georgian and Ossetian women 4 Zugdidi IDP women 5 Zugdidi IDP youth 6 Tbilisi Ex-combatants 7 Tbilisi IDP youth *Located in the conflict region South Ossetia.

1 May 2008 6 May 2008 7 May 2008 15 May 2008 16 May 2008 23 May 2008 25 May 2008

Respondents Resp. Location No.

Position

Org. Gender and Age No.

1 Kutaisi NGO Representative 1 Woman 35-40 years 2 Kutaisi NGO Representative 1 Woman 35-40 years 3 Kutaisi Participant 1 Woman 55-60 years 4 Kutaisi Participant 1 Woman 35-40 years 5 Tskhinvali (SO) NGO Representative 2 Woman 40-45 years 6 Tskhinvali (SO) Participant 2 Woman 45-50 years 7 Tskhinvali (SO) Participant 2 Woman 45-50 years 8 Tskhinvali (SO) Participant 2 Woman 30-35 years 9 Tskhinvali (SO) NGO Representative 3 Woman 50-55 years 10 Tskhinvali (SO) Participant 3 Woman 65-70 years 11* Tskhinvali (SO) Participant 3 Woman 65-70 years 12* Tskhinvali (SO) Participant 3 Woman 60-65 years 13 Tskhinvali (SO) Participant 3 Woman 30-35 years 14 Tskhinvali (SO) Participant 3 Woman 30-35 years 15 Zugdidi NGO Representative 4 Woman 50-55 years 16 Zugdidi Participant 4 Woman 30-35 years 17 Zugdidi Participant 4 Woman 30-35 years 18 Zugdidi Participant 4 Woman 25-30 years 19 Zugdidi Participant 4 Woman 15-20 years 20 Zugdidi Participant 4 Woman 25-30 years 21 Zugdidi NGO Representative 5 Woman 60-65 years 22 Zugdidi Participant 5 Woman 15-20 years 23 Zugdidi Participant 5 Woman 15-20 years 24 Zugdidi Participant 5 Woman 20-25 years 25 Zugdidi Participant 5 Woman 20-25 years 26 Zugdidi Participant 5 Woman 15-20 years 27 Tbilisi NGO Representative 6 Man 65-70 years 28 Tbilisi Participant 6 Man 40-45 years 29 Tbilisi Participant 6 Man 40-45 years 30 Tbilisi NGO Representative 7 Woman 50-55 years 31 Tbilisi Participant 7 Man 15-20 years 32 Tbilisi Participant 7 Man 15-20 years 33 Tbilisi Participant 7 Man 15-20 years 34 Tbilisi Participant 7 Man 15-20 years * Respondent deemed as unrepresentative and was omitted from the study. 56

Ripple Effects

n/a n/a Yes Yes n/a Yes Yes Yes n/a No No No n/a Yes Yes Yes Yes Yes n/a Yes Yes Yes Yes Yes n/a Yes Yes n/a Yes No No No

Appendix 2 – Questionnaire A Questionnaire A (participants) 1) What is your profession? 2) What level of education have you attained? 3) Do you have any other kind of responsibilities at your work place or within an organisation? (For example, board member, leader of interest group.) 4) Were you participating in conflict resolution activities before you joined (the organisation’s) trainings? 5) Had you been politically active before participating in the trainings? 6) How were you selected to participate in the trainings? 7) Why did you decide to participate in the trainings? 8) How many times have you participated in the trainings?/How long have you been taking part in trainings? 9) What would you say that you have learnt from participating?/Have you gained any new skills through your participation? 10) Have your participation changed your views on the conflict? 11) Have your participation changed your attitude towards Abkhazians/South Ossetians/Georgians? 12) Have you taken any own initiatives to spread what you have learnt to your community/family? OR Have you used the skills you have gained? In what way? 13) Have you initiated any activities aimed at a peaceful resolution of the conflict? For example, meetings or seminars. 14) Have you spread your new knowledge to your family and friends or to other people in your community?/Have you been talking to other people about what you have learnt in the trainings? 15) Have you tried to change your family’s or community’s attitude to the conflict? (Do you think that you have succeeded in changing their views?) 16) Have you encouraged/invited other people to participate in the trainings? 17) Have you contacted decision-makers or other officials to influence them? 18) Have you taken any other initiatives to affect the peace process?* IF NO to questions 12-18: 19) What would you say is the primary reason for not taking action to spread what you have learnt? using the skills you have attained?

57

Appendix 2 – Questionnaire A 20) Does any of the following apply to your situation: - preoccupied with day-to-day survival - do not have the time - afraid of negative reaction from community/family, e.g. afraid of being viewed as a traitor, fear of threats - social pressure to conform to accepted views* 21) Would you be willing to talk to other people about what you have learnt use the skills you have gained …or to initiate conflict resolution activities?* If YES to any of the questions 12-18: 22) What made you decide to spread what you have learnt to your community/family? use your new skills? take own initiatives to affect the peace process? contact decision-makers? For both answers: 23) Did the organisation (trainers) encourage you to/did the trainers tell you that you that you should spread what you have learnt to your community/family? talk to other people about what you have learnt? use your new skills? take own initiatives to affect the peace process? contact decision-makers?

* Question only posed if the preceding question did not yield an elaborate reply.

58

Appendix 3 – Questionnaire B Questionnaire B (NGO representatives) 1) (Tell me about your peacebuilding activities) 2) What is your goal? 3) What is the long-term objective of your organisation?* 4) How/through what mechanisms do you assume that your activities will contribute to the peace process on the macro-level? 5) What kind of impact do you think that your activities can have on the conflict/peace process on the macro-level?* 6) Would you say that your activities have had an impact on other people than those participating in your activities? 7) How do you select the participants for your activities? 8) What do you expect of the participants after they have participated in your activities? Are they expected to make efforts outside the activities they participate in? 9) Do the participants usually take part in several training sessions? Why? Do you expected them to as a part of your programme or do they simply want to participate more than once? 10) Do you encourage the participants to/do you tell the participants that they should -

spread what you have learnt to their community/family? use their new skills outside your activities? take own initiatives to affect the peace process? contact decision-makers?

11) Have there been any examples of people spreading their knowledge? using their skills? taking their own initiatives to affect the peace process? contacting decision-makers? 12) Are the participants facing any kind of risk because of their involvement with your organisation? For example, harassment, physical attacks?

* Question only posed if the preceding question did not yield an elaborate reply.

59