Field Research and Literary Sources

The current study employed, among other tools, semi-structured interviews with a sample of informants in each research component. The choice of this method was considered to be most suitable, as it allows both parties to explore the meaning of the questions and answers involved. The method naturally allows the interviewer to depart significantly from any schedule or guide that is being used and can ask new questions following up interviewees’ replies. In this way, there is a clear sharing through interactions, an element which is usually not so central or lacking all together in other research procedures. As Bryman¹³⁰ claims, the interview is the most widely employed method for the collection of qualitative data. It is generally assumed that the interview acts as a medium through which the researcher can gain access to information and share knowledge with those who have it and thus be able to organise and remember the presentation of their knowledge.

Interviews were conducted at different times, with respondents identified in three different data collection areas. In Tanzania, the selected data collection areas for this study were such centers as Mwalimu Nyerere Foundation; Center for the Study of Forced Migration; three refugee camps in North Western Tanzania, namely Kibondo, Kasulu and Nyalugusu; Ministry of Justice and Constitutional Affairs and the Ministry of Home Affairs, the latter being the government unit responsible for all migrations and refugees issues in the country. In these areas, respondents consisted of refugees from the three refugee camps mentioned above. The selection of respondents was such that from each camp, three refuges were interviewed. Four were from Congo while the rest were from Burundi. The aim was to maintain gender balance in ideas and also to gain insight on different responses from each country. The group further consisted of three diplomats who were very instrumental and played key roles in the Burundi Peace Process which took place in Arusha. The last in this group of respondents was a famous professor and researcher who had researched and written at length on the Africa’s Great Lakes Conflicts.

130 Ibid, p.319.

From Burundi, respondents consisted of legal practitioners (advocates and two judges); long-experienced members of parliament of Burundi; leaders of human rights NGOs; leaders of the principal political parties in Burundi and members of the local population from the four areas which have been mostly affected by civil wars in Burundi. From the DRC respondents interviewed consisted of church leaders;

MONUC personnel, including those from the mission’s human rights department;

leaders of civil society organisations; members of the local community; government officials; academicians and legal practitioners (advocates).

Before embarking on the actual interviews, a pilot study of few of the respondents was conducted in order to test the interview guide and develop the subsequent revision of the instrument used for data collection. Interviews were then conducted with each subject. Each interview lasted between 45 to 75 minutes, the longest being 90 minutes. Subsequent interviews were conducted either in the respondents’

respective offices or in their places of residence. Preparations for the interviews were such that respondents were given prior information, either through telephone or through research assistants who arranged the interview sessions. Before the start of each interview, each respondent was notified on the aim of the interview, that it only serves academic purposes and that the information passed on will be kept strictly confidential. The option for anonymity was also explained to each respondent.

Most interviews were audio-recorded using a digital recorder and transcribed in order to provide accuracy in data collection. Interviews with some of the respondents were however jotted down some time after the informal interview time elapsed. These practices were considered important in that they allowed the interviewer to check for errors in interpretation of data and to use extensive quotations in the text of this thesis.

Field notes (scratches or jotted notes) were taken during the interview to record important and meaningful interactions between subjects and the interviewer.

Transcriptions were coded and categorized inductively using content analysis to organise data.

It must be pointed out however that interviewing, though it is one crucial method of qualitative research, is not free from problems. The main challenges perceived in applying interviews as a data collection research method are numerous. In the first place, given the sensitivity of the research topic with its main focus on armed and

ethnic conflicts, respondents’ scepticism was prevalent as they became alarmed at the prospect on their words being preserved using a recorder, despite prior notice of the purpose of the interview and accession. As a result, some of the interviews were not as interesting as they were expected to be. Another big stumbling block was a linguistic one. With the exception of informants from Tanzania, interviewees from both Burundi and the DRC had difficulties in expressing themselves in a language in which the interviewer was fluent. Both countries are the so-called “francophone”, using French as a second or official language. At times, those respondents who could speak Swahili,¹³¹ poor as it might be, were preferred in the selection process, to those who spoke only French and their local native languages.

Also, at times, some of the interviewees distorted the required information through selective perceptions, with a desire to please the interviewer. In such cases, there was no option but to replace such an interviewee with a more straightforward one.

Interviewing, the transcription of interviews and the analysis of the transcripts are all very time-consuming,¹³² but the advantages derived therefrom are immense.¹³³ The views of respondents certainly gave the interviewer and researcher of this study valuable insight in approaching the topic on the application of human rights in conflict prevention and management.

1.3.2.2 Questionnaires

A survey questionnaire method was found appropriate to extract information from informants who could not be directly interviewed due to various reasons. Survey questionnaires usually inform research by providing information from a known sample of informants selected on the basis of specified criteria. Typically, data from survey questionnaires is collected without the researcher intervention.

Specific questions were prepared in the light of the current study research objectives.

Different sets of questionnaires were designed and administered to different target respondents. In the first set, I administered both structured and semi-structured questionnaires to different respondents in the research areas. Given the multilingual

131 The local native language of the interviewer.

132 See Bryman, supra note 116, p.319.

133 On the advantages of interviewing vis-à-vis other methods of data collection in qualitative research, see Ibid p.338.

nature of the informants selected from the DRC and Burundi, questionnaires were administered in three different languages, namely Swahili, French and English. The contents, structure, format and sequence of the questionnaires used are as shown in appendix “C.”

Pre-testing of the survey questionnaires was carried out among a few refugees in the three refugee camps in Tanzania (men and women), and they were collected immediately for verification. This allowed minor changes to be made to improve the contents and form of the questionnaires. The questionnaires were sent to contact persons identified through research assistants in both Burundi and the DRC, others identified through the embassies of both countries and the Mwalimu Nyerere Foundation Centre. In the course of field research in Congo and Burundi, more persons were identified, making a total number of persons who received the questionnaires to be 45. The survey questionnaire was administered via electronic mails, personal visits and telephoning. Two research assistants in the DRC and one in Burundi also facilitated the work. The number of people who responded and returned the survey questionnaires was 31. As some questionnaires were responded to in Swahili and French language, proper translation was sought before the findings were analysed.

1.3.2.3 Case Study

The need to study a complex topic within its organisational context led me to adopt a case study methodology. A case study involves a detailed examination of a single or a particular occurrence.¹³⁴ Its main concern is the detail and complexity of the case, which it treats as a bound system. In general, a case study methodology is a preferred method for investigating questions that ask how or why of both historical and contemporary events.¹³⁵ It can produce an in-depth understanding of a particular situation and an analysis of its meaning in a discovery process.¹³⁶ Case study evidence can be drawn from a variety of sources, including interviews, direct observations, participant observation, and even documents and archival records. Three main types of case studies are distinguished: namely, explanatory, causal and descriptive.¹³⁷

134 See Burrel and Morgan, supra note 117, p.49.

135 Yin, R (2003), Case Study Research: Design and Methods, 3^rd Ed, London/New Delhi: SAGE Publications, p. 112.

136 See Merriam, supra note 106, p. 46.

137 See Yin, supra note 135, p. 114.

Explanatory case studies collect data before theories or specific research questions are formulated. Causal case study focuses in looking at the cause-and-effect relationships, and search for explanatory theories of the phenomena.

Descriptive case study requires a theory to guide the collection of data and this theory should then be openly stated at the outset. The more thoughtful the theory, the better the descriptive case study will be. Insights from this type of research can exert a strong influence on policy, practice and research, with the role of the researcher being that of a versatile, functioning simultaneously as a scientist, a fieldworker, and a toolmaker and as a technical developer.¹³⁸ While the conclusions from such case study research may not be generalizable, they focus strongly on qualitative issues;

they may nevertheless suggest important implications for other similar contexts.¹³⁹ In the current study, the Africa’s Great Lakes Region is an overall research area.

However, given the vastness of the region,¹⁴⁰ it will be unrealistic, if not impossible to try and conduct the current research for the whole region. Although the Africa’s Great Lakes is considered as a conflict system, the DRC conflict, which is among the many other conflicts in the region, was selected for this study. Again, only some aspects of the conflict in the DRC are thoroughly studied.