Introduction

2During the time of my fieldwork, the Itorero programme was managed by a special taskforce situated within the public institution called the National Unity and Reconciliation Commission (NURC). The programme is officially modelled on a pre-colonial educational institution carrying the same name, which sought to train the future (male) leaders of the civic and military administration under the Rwandan king. Those trainees were called ‘Intore’, which is also the name given to a person participating in the ‘modern’ Itorero programme.

3The contemporary programme teaches a set of ‘cultural values and taboos’ which are meant to help the citizenry contribute to national development efforts as articulated in the Rwandan state’s overarching policy framework for development, called the Vision 2020. This Vision, adopted in the year 2000, is meant to turn Rwanda into a middle-income country by the year 2020. Beyond teachings on Rwanda’s cultural values and taboos, the theoretical curriculum includes lessons on national history, as well as contemporary government policies, programmes and laws.

4Itorero targets the entire Rwandan population, even citizens in the diaspora. Participation is mandatory for all Rwandans living in the country who are above the age of seven. The programme was launched in 2007. During its first phase of implementation, it targeted formal and informal leaders in society who could serve as future Itorero educators in their local communities. So far, about three hundred thousand Rwandans have been trained in thirty or so ‘categories’ of such leaders, including heads of central and local government, primary and secondary school teachers, high school graduates, prison staff, health workers, and members of Rwandan diaspora organisations abroad. The content of these trainings has been quite comprehensive. In addition to the theoretical teachings mentioned above, the camps often comprise military exercises, physical training, and cultural art forms traced back to the pre-colonial kingdom, entailing dancing, singing and poetry.

5In 2011, Itorero ry’Igihugu was officially launched in all of Rwanda’s villages and schools as part of the second, ’decentralised’ phase of the programme. The objective is for people to be educated where they live, work and go to school. Usually, such Itorero trainings are undertaken a few hours once a week across a period of approximately six months, and they are led by the local administration and school teachers who were trained during the first programme phase. Plans are also underway to initiate Itorero trainings in Rwanda’s workplaces and institutions, such as private business companies and churches.

6My field study in Rwanda was largely divided into three 'sub-fields'. It included participant observation in a variety of different Itorero trainings across the country, coupled with interviews with persons connected to the programme (managers, trainers and participants). Secondly, it involved talking to (other) Rwandans living in Kigali about topics addressed in Itorero, and inquiring about their knowledge and perspectives of the programme as such. Lastly, the study entailed participant observation in the daily life and governance of one particular neighbourhood in Kigali where I lived during most of my fieldwork.

7As an ethnographer, I sought to understand imaginaries of the state and its relationship to the citizens both from within and from without. On the one hand, this implied listening to Itorero teachings and the testimonies given by programme managers and government officials. Those shed light on the government’s views of the Rwandan nation-state, its history and its people. On the other hand, it entailed relating such political narratives to those of Rwanda’s citizens who did not associate themselves with the state administration. Some of these persons had participated in the Itorero programme, while others had not.

8As with any ethnographic study, there were a number of challenges that I as a researcher encountered in the field. The rest of this paper reflects on three such challenges. In my view, they are characteristic both of the socio-political and cultural context of Rwanda as a larger geographical field, and of my own particular interest in public civic education and state- and citizen-making.

Talking about the ‘un-talkable’: Coming to grips with censorship by the state and the self

9One of the major challenges I came to face as an ethnographer pertains to the kind of social reservation, wariness and distancing that characterised most of my initial encounters with Rwandans. This has many, overlapping explanation. For example, secrecy, silence and social reservation have been described as inherent features of Rwandan culture. Such traits are believed to have been cultivated by the royal kingdom and to date back to the formation of the pre-colonial state (De Lame, 2005). More importantly, Rwanda’s colonial and post-colonial history of violence and conflict has played a central part in raising fears and suspicion among citizens. This violence culminated in the 1994 genocide, in which thousands of Rwandans either facilitated or actively participated in the killing of up to 800 000 fellow citizens, most of whom belonged to a particular ethnic group. Yet another critical factor can be attributed to the extensive apparatus of control and surveillance maintained by the post-genocide government in Rwanda. This political leadership, dominated by the Rwandan Patriotic Front, was born out of an armed movement of refugees residing in Uganda, which launched a guerrilla war against the sitting Rwandan President and his regime in 1990. Four years later, in July 1994, the rebels toppled that regime. Their coup put an end to the genocide initiated by the government against people identifying themselves with the same ethnic group as the refugee rebels. Ever since, the leadership of this guerrilla movement has maintained a strong hold of power in Rwanda, not the least by means of a complex security apparatus that penetrates the most local levels of Rwandan society (Thomson, 2009).

10In this context, asking people to speak their mind to a person they have not known for a very long time is to demand a lot. Not long into my research, I realised that reaching any kind of sincere testimony would turn out to take time and effort. Moreover, expectations placed on somebody else to ‘speak out’, I learned, need to be accompanied by an active effort on behalf of the listener to recognise and interpret silences and omissions, rumours and proverbs.

11Even more challenging is it to undertake research on subjects widely seen as politically and morally charged. Due to the perceived omnipresence of the state in Rwanda, such subjects are numerous. The Itorero programme is one of them, since its syllabus addresses highly contested subjects, such as ethnic identity, the causes of the genocide, and policies and laws on justice and reconciliation. One of my neighbours told me for instance that since I was working with Itorero I could be sure that my mobile phone and email were tapped.

12As with most politically sensitive topics in Rwanda, there are ready-made, official narratives that one can adopt when asked to voice an opinion. And many of the persons I spoke with about Itorero – especially central government officials and people I did not know very well – did apply these. Others replied that they did not know anything or had no interest in talking to me about Itorero, because ‘politics did not interest’ them. Yet others returned the question to me, asking me why I wanted to know. Most of those persons who eventually agreed to share their views and thoughts, did so on the condition that I could ensure absolute anonymity, provide a setting where no one could hear us, and promise not to record anything. In fact, it turned out almost impossible to tape record any interviews or conversations. Those I did record often turned out compromised by the interviewee’s self-censorship as well as his/her censorship of what I was allowed to save and use from my recordings. At several times, I was asked to delete from the tape recorder utterances made following my interlocutor’s regret of having formulated him/herself ‘badly’. At other times, interviewees shared with me strikingly different perspectives once the tape recorder had been turned off.

13My participation and presence at the Itorero trainings is another example of the challenges of auto-censorship and mistrust in Rwanda. At the camp sites, it turned out impossible for me to ask participants what they thought of the programme. Most often, I received a standard response of “it’s good” or “I like it”, even in cases where I had seen the persons sleeping through most classes or trying to avoid group activities and individual responsibilities. Weeks or months later, however, a few of them confessed to having a whole other take on it. Thus it turned out critical for my study to follow up with some of the participants I had gotten to know during the training, and allow them a chance to express their views outside the direct realms of state control and surveillance.

14The tendency of only receiving ’state-sanctioned’ responses to politically sensitive questions in Rwanda is widely, if not openly, acknowledged. In fact, during my research, several of my informants brought up the entrenched political correctness of Rwandans when in the company of authorities or persons not considered close friends or kin. It was even publically addressed in the Itorero programme, as a source of concern among both trainers and participants.

15The problematic around auto-censorship has also been addressed in numerous publications on field research in Rwanda, such as those by Larissa Begley (2009, 2013), Jennie E. Burnet (2012), Bert Ingelaere (2010), Andrea Purdekovà (2011) and Thomson, Ansoms & Murison (2013). These scholars bring to the fore how the government’s extensive network of surveillance, coupled with its elaborate set of official narratives, governs what can be said and not said, and under what circumstances. While my study indeed affirms these observations, I also believe we should be wary of equating the official narrative with one that is less true for the narrator. In the context of my fieldwork, certain personal recitations of the politically dominant discourse, such as those which wholeheartedly endorsed the government’s version of the genocide, reflected what I believe my interlocutors held as true; they were not necessarily a veil for a hidden, alternative script.

16Nevertheless, these scholars put their finger on a pertinent challenge to ethnographic research in Rwanda, notably how to create meaning from censored testimonies. This is a central topic in the recent anthology Emotional and Ethical Challenges for Field Research in Africa (Thomson, Ansoms & Murison, 2013). Yolande Bouka (2013), one of the contributors to that volume, highlights how the accuracy of accounts may not be as important as the way people present them within a given context. Drawing on her research in Rwanda among released prisoners and their views and experiences of the Rwandan justice system, Bouka writes that the manner in which various ‘facts’ were presented had much to say about how the narrator conceived of and positioned him/herself vis-à-vis other actors, events, practices or ideas in the context narrated. An insightful analysis of a similar problematic can be found in Jennie E. Burnet’s (2012) work on Rwandan women’s narratives of the genocide. During her research, Burnet explains, silences turned out to be a critical part of her interlocutors’ testimonies. Their hesitance to share memories and the omissions created in their testimonies were just as charged with meaning as the words they actually spoke.

17As in Burnet’s study, silences and secrecy have been found to constitute psycho-somatic symptoms of trauma (see also Ross, 2003), but they have also been interpreted as coping-strategies applied in settings of post-conflict tension (Mannergren Selimovic, 2010), the threat of arbitrary terror (Taussig, 1992), or racism and discrimination (Green, 2002). Yet in other cases, secrecy has been understood as signs of resistance, encrypted in for example cultural forms like story-telling and riddles (Ferme, 2009; Martin, 1992). In her study of everyday forms of violence among persons living on the margins in eastern Brazil, Nancy Scheper-Hughes (1992 : 229) writes that half-truths became part of the ‘mundane surrealism of ethnographic description’. This entailed a constant struggle to try to distinguish fact from fiction and rumours from actual past and present occurrences.

18In my own interactions with neighbours, authorities and Itorero participants, it was not uncommon for some of my questions to be answered with another question or with a seemingly esoteric proverb. Although my research assistants were of invaluable help, I, as well as they, still struggled with trying to understand the meaning of cryptic responses and subtle indications. Moreover, the amplitude of the government’s secret intelligence apparatus was a frequent topic of conversation among both Rwandans and foreigners in Kigali. Within the international aid community, more or less serious rumours circulated among Americans and Europeans about spies on the street, video cameras in restaurants, and planted microphones in certain hotels. The same rumours were ample among my Rwandan friends and informants, with the difference that these latter often referred to actors and systems of hidden surveillance in their local communities.

19This context of wariness and auto-censorship, I soon found, also required me to know when to stop probing for information. I was often reminded of the fact that asking questions was simply not something you did in Rwanda. In Kinyarwanda, the word for ‘question’, ikibazo, also means ‘problem’. A pertinent problem with questions, I found, was that they raised suspicion. When my close friends and research assistants censored my questions or advised me to act or formulate myself in certain ways before persons I knew less well, it was almost invariably to save me from raising suspicion, not to help me show respect or sensitivity. For the same reason – trying not to attract suspicion – I rarely came to take any photographs of my surroundings.

20Despite the ‘precautions’ I took to respect people’s privacy and not overstep my boundaries, different rumours circulated about me, even in the small neighbourhood where I lived. One said I was a spy for a Rwandan diaspora movement that wanted to overthrow the President, another that I was a Jehovah’s Witness. In my absence, it happened that my research assistants were addressed by local authorities about my research. One was for example asked what my research was ‘really’ about and whether there was any ulterior motive behind my questions. Unintentionally, I placed several persons in situations where they feared state repercussions because they had happened to accompany me as I had approached state authorities with questions that they deemed in hindsight to be too political. These experiences bring to mind Allen Feldman’s (1991) emphasis on the importance of knowing where the line goes for making enquiries in politically polarised and socially tense environments. Drawing on his ethnographic research on terror and violence in Northern Ireland, Feldman writes that ethnographic achievements do not so much depend on the ability to access all pieces of information as it hinges on the capacity to be selective in one’s explorations.

Studying and working with politicised state institutions: The researcher as both ‘ambassador’ and ‘whistle-blower’

21Not only the questions I posed but my identity writ large became in some respects the target of political projections. For example, not far into my fieldwork, I found myself expected to serve both as an advocate or a spokesperson of the Itorero programme, and as a channel of voice and source of empathy for persons who feared and opposed that very same programme.

22Accessing the inner workings of the Itorero institution depended on the maintenance of good social relations with its staff. Several Itorero stakeholders, especially members of the programme Taskforce and employees of the NURC, saw me as a potential ambassador for the Itorero programme. As a researcher, I could tell the academic world about the good intentions and benefits of the programme through the publishing of my research results. Although I was never allowed to attend any management meetings held either by the Itorero Taskforce or the NURC, I was encouraged to work for the programme as an intern. While I did come to help out with a few minor tasks, I declined to take on the formal responsibility that an internship would imply. Nevertheless, during some Itorero trainings, not only was I publically used as an example of a role model trainee, I was even introduced as an employee of the Itorero Taskforce managing the programme. Hardly surprisingly, then, I came to discover that Itorero participants at various sites had initially mistaken me for a representative or staff member of the Itorero programme.

23Likewise, when studying the everyday governance of my living area in Kigali, access to the local administration relied on maintaining good relations with the authorities. As a foreigner in Rwanda, an easy way to do this is simply to show up for and take part in the many community-level civic duties that are compulsory for all Rwandan citizens. By participating in sessions of collective, manual labour, attending village meetings and paying local fees and taxes on time, I soon found myself being used as a show case of a model citizen by the local authorities. Seeking to pressure my neighbours to improve their performance as good citizens, the administration used my engagement to put the rest of the neighbourhood’s population to shame, arguing that even the “white lady” (that is, one of the foreigners or wealthy inhabitants who were unofficially exempted from petty issues such as community labour) contributed more than the ‘normal’ people. I was also pointed out as a potential or indirect member of the government party. Many times, the neighbourhood authorities, both publically and in private, asked me to join the Rwandan Patriotic Front. If I joined, they argued, many others would follow.

24While I never did join, I strongly doubt that doing so would have resulted in any dramatic rise in recruitments. It may however have implied the distancing of some of my neighbours. Possibilities to build relationships of trust with persons who were not affiliated with the Itorero programme depended on my ability to empathise with them also. And many of them were rather critical of the programme, as well as of the government at large. Some lived in fear of what they saw as a coercive, politicised and punishing state. In me, they found a potential whistle blower who could lift the lid on the forms of state oppression, repression and surveillance that they were exposed to, but too afraid to speak out about. As a gateway to the ‘outside’, my findings could influence foreign policy and, in the long run, change domestic political conditions for their children and grandchildren.

25The balancing act required in navigating through this politically polarised milieu was in other words quite challenging. Much is at stake for any Rwandan who voices critique of certain political truths, or who shows support for political taboos. For example, a few of my informants admitted that they were known by the local authorities as ‘critical’ persons inclined to address foreigners like me to ‘kiss and tell’ and give the state a bad reputation. They believed that making my acquaintance implied that the authorities were keeping an extra eye on them.

26Naturally therefore, some of my informants hoped that taking the risk of facing state repercussions by sharing their views with me should give something in return. Similarly, Itorero stakeholders and other government representatives believed that they took risks in allowing me access to the workings of a government agency they knew was questioned and criticised, not the least by certain foreigners. Hence they, too, hoped for some sign of reciprocity. Larissa Begley writes that one’s success as a field researcher in Rwanda hinges on one’s capacity to “appear to be on everyone’s sides” (Begley, 2013 : 77). While there is certainly much truth in this, my usual response was rather to downplay my role as any kind of potential or actual intervener. I insisted that my task as a researcher was primarily to listen and try to understand. Most people accepted this, at least formally and at least in the beginning. But, of course, as time passed and our relationships deepened, expectations were raised that I should take sides with them, defend them, and most importantly, show sympathy for their cause.

27This predicament resounds with what Olivier de Sardan (1995) calls ‘enclicage’. It entails the fact that the progressive assimilation of the researcher into his/her field of study does not imply integration into society at large, but always into certain ‘cliques’ or factions of that society. Within these latter, people have their own logics, perspectives and interests, in which the researcher often comes to find him/herself entangled. The risks affiliated with this enclicage, Olivier de Sardan writes, concern the tendency of the researcher to end up defending or advocating the interests and ideas of certain informants. This, in turn, may close the door to other informants who do not share those same perspectives and interests.

The researcher with a professional background in ‘development’: A key that opens some doors but closes others

28To some extent, the enclicage (Olivier de Sardan, 1995) I experienced as an anthropologist in Rwanda dated back to a time before my fieldwork. A year prior to my ethnographic research, I was working in Kigali for the Swedish International Development Cooperation Agency (Sida). A part of that work involved managing financial support for the Rwandan government’s on-going demobilisation programme. That programme included a component of civic education that is widely seen as the forerunner of the Itorero programme, and that is generally managed by the NURC, the same institution hosting the Itorero Taskforce. Moreover, in the past, Sida has supported other projects managed by the NURC. Hence, a few of my colleagues at the Swedish embassy were familiar with certain people among the staff managing the Itorero programme. Personal experiences at the NURC of good professional relations with Sida, and possibly also anticipations of additional Swedish funding, may very well have facilitated my access to the workings of the Itorero Taskforce. Accommodating a researcher who had been a former employee of a foreign donor agency which had collaborated with them earlier was probably seen as an opportunity to promote the Itorero programme within strategic areas of the foreign aid community. Having maintained my friendships at the embassy, I held potential in helping the programme managers to raise funds, political support or, at the very least, ‘goodwill’.

29Researchers on Rwanda such as Bert Ingelaere (2010) have stressed how Rwanda’s elite often seek to encapsulate field researchers and development workers to make sure they only reach state-sanctioned versions of reality. Therefore, it is important to ask oneself as a researcher which voices one listens to and how one accesses these voices. For me, this became evident in instances such as when persons I did not know contacted me after having heard about my background as a former Sida employee who was now collaborating with the Itorero Taskforce.

30The very same connections to the foreign aid sector may in other cases have impeded my access to views and experiences alternative to those of the Itorero proponents. Persons critical of the programme and of the many foreign states supporting the Rwandan government may have been dissuaded from confiding in me due to my affiliation with a donor agency known to pursue state-to-state cooperation with Rwanda and its many public institutions. Some persons told me how Western foreigners like me were usually naïve and uncritical when it came to the Rwandan political machinery. Once we settled in Kigali and started to work with the government, we wholeheartedly bought into the dominant political rhetoric on economic development and good governance.

31In any case, here again, my strategy usually came to be one of downplay and discretion. I tried to lessen the weight of my affiliation with Sida and the international donor community, and instead emphasise my professional role as a researcher. In doing so, I hoped to convince my informants that my academic position offered independence from any political constraints or financial opportunities that may otherwise have influenced my analysis. To what extent I came about as convincing probably depended on whom I was addressing.

32Moreover, as a researcher, I too, had to learn the practice of auto-censorship. In densely populated and intensely surveyed societies such as Rwanda, commitments to anonymity and confidentiality can be very challenging. Yet, due to the political polarisation of Rwandan society, anonymity and the ability to keep secrets are critical for any field researcher. This is especially true for those who, like me, have prior ties to certain ‘cliques’ in Rwanda, and for those who seek to understand and engage with a wider spectrum of persons – from those said to represent the state to those who fear and oppose it. For me, it came to require a daily routine of choosing my words carefully, where the truth was but one of several selection criteria. Only such tactics could safeguard the confidence I managed to build over time with certain persons, and without which I would end up neither here nor there in Rwanda’s political spectrum.

33Pierre-Joseph Laurent (2012) writes that anthropologists have an obligation to be engaged, because the validity of anthropology as a sciencelies in the researcher’s ability to critically reflect on his/her own subjectivity as well as that of his/her informants. In light of my experience of having to auto-censor my words and restrain my involvement as a researcher, I believe that Rwanda is a good example of how anthropological engagement also hinges on simultaneous forms of disengagement. Such disengagement, meanwhile, certainly poses its own risks. For example, building trust in Rwanda, I found, much depended on my willingness to share with my interlocutors information about my own life, my background and my work. Yet, due to my obligation to protect the confidentiality of what had been shared by other informants, I sometimes had to refuse answering seemingly innocent questions, such as what my plans were for the afternoon or where I had been to attend a public meeting. These forced silences may certainly have raised suspicions among some persons. On the other hand, they helped me to understand why people I came to know sometimes felt compelled to avoid answering questions which at first glance seemed perfectly innocent to me. They reminded me of the fact that most of the balancing acts I as an ethnographer had to perform to navigate in Rwanda’s politically and socially charged landscape, my informants had had to learn much earlier in life.

Conclusion

34In all, the basic lessons of my fieldwork in Rwanda are in no way unique. As Hagberg and Ouattara (2012) write, anthropologists everywhere influence in various ways the social settings they study, and this is ultimately a challenge facing all empirically grounded studies. Moreover, just like the ethnographer undertakes research on his/her informants, so do the informants in most contexts undertake research on the ethnographer (Ansoms, 2013). However, different settings and different research interests impact on the ways in which these processes play out in practice. Rwanda exemplifies how population density, state surveillance, political polarisation and social distrust may generate particular challenges for ethnography on behalf of both researchers and informants. Such challenges, I have tried to show, include creating meaning from censored testimonies, maneuvring in and across various ‘cliques’ of society, and knowing when to disengage in order to stay engaged.