Critique of sources' perspectives on popular participation in gacaca

The purpose of this section is to critically assess the official, popular and critical perspectives on popular participation in gacaca. This section analyses the perspectives of these three groups of actors in terms of the norms and ideas central to their interpretations, as well as how effect­ively they reflect the practical reality of gacaca's operation.

First, there are several problems with the government's view of the role of popular participation in gacaca. One major problem with the official interpretation is its overstatement of the degree to which the commu­nity controls, and may freely participate in, gacaca. Popular participation in gacaca is not, as Mucyo argues above, unconditional. The govern­ment rarely discusses the extensive involvement of state actors in gacaca, which undermines the notion of gacaca as an entirely popular enterprise, driven by local agents. The government participates in numerous facets of the daily running of gacaca, including providing judges with dossiers detailing suspects' crimes and confessions, and sometimes intervening when hearings are perceived to diverge from the statutes and norms of the Gacaca Law and Gacaca Manual. During several gacaca hearings I attended, government officials intervened to correct certain judges' statements and to halt disruptive behaviour in the general assembly.^[351] Officials sometimes passed notes to gacaca judges or moved among the general assembly, speaking quietly to participants who had been unruly during hearings.^[352]

The greater problem with the government's view, however, concerns its expected outcomes of popular participation in gacaca. The government's discourse concerning a lost sense of national unity that must be regained through gacaca is highly unconvincing.

The notion of a pre-existing national unity is also inconsistent with other areas of official policy, especially the government’s emphasis on the need to establish ingando for provisionally released detainees and other groups such as civil servants. The main purpose of the camps is to teach civic virtues to a population that the government perceives as lacking important social values. If these virtues already exist in the community, there seems little need to teach them. As we will see in later chapters, this theme highlights a tension at the heart of the government’s vision of gacaca, between emphasising popular ownership and participation on the one hand and, on the other, the importance of the state in educat­ing the populace or, as Sara Bawaya of the NURC was quoted in the last chapter, �making people good or bad’. The government’s entanglement of themes of popular ownership and state education often produces prob­lems for the function of ingando and gacaca and the pursuit of various social objectives after the genocide.

Because the population echoes much of the official perspective of the community’s participation in gacaca, many of the same criticisms just made can be directed towards the population’s views, particularly where unity - the harmony of the �family’ - is assumed to be a latent fea­ture of Rwandan society which gacaca will help rediscover.

Second, the population’s generally enthusiastic statements regarding participation at gacaca have not always reflected popular practice across the country, namely that such involvement has fluctuated greatly.^[354] From my observation of sixty-seven gacaca hearings in eleven communities between 2003 and 2010, there is a consistent trend in the population’s involvement, both in terms of attendance levels and active dialogue. In these jurisdictions, attendance was high at three critical junctures in the gacaca process: first, when gacaca was getting underway and the population was compiling the initial lists of community information; second, at the beginning of the trial phase, which in most of these loca­tions occurred in 2006 and 2007; and, finally, towards the end of gacaca, particularly as Category 1 suspects were being prosecuted at the sector level.

Between these three periods, however, the population in the observed jurisdictions participated much less during hearings, both in terms of the number of attendees and the quality of the communal discussions. Following the initial information phase, many people complained that gacaca hearings were long and tedious, often involving deliberations over the same evidence and focusing on minor details of past events, particularly property that survivors claimed they had lost during the genocide and for which they claimed compensation. �I have stopped going to gacaca because the people talk for hours and hours and we make no progress’, said Faustin, a thirty-three-year-old farmer in Gisenyi. �Some of the survivors say they had this many sacks of flour or this many millet plants until their neighbours stole them, then others will say these people are lying, and things get very tense. This kind of talk can go for hours.’^[357] Other common explanations for non-attendance at gacaca dur­ing the information phase included that individuals were losing sub­stantial time from work, with gacaca hearings taking up to a full day and sometimes held twice a week in certain locations; and many gacaca participants were perceived as lying and therefore the community was losing faith in the process.^[358]

After the initial months of the trial phase of gacaca, participation also decreased in many communities for various reasons.

Finally, during seven years of gacaca observations, the general social and political atmosphere surrounding gacaca has crucially influenced the population’s sense of well-being and affected popular involvement during hearings. Soon after the initial interest in gacaca during the pilot phase in 2002 and the extension of gacaca across the country in 2003 and 2005, interviewees consistently stated that the general atmosphere of tense uncertainty made them reluctant to attend gacaca. In the first eight months of 2003 alone, Rwandans faced a flurry of major national events that unsettled the populace: the first of several releases of geno­cide suspects from prison into the ingando and then into the commu­nity; the expansion of gacaca from 750 to nearly 9,000 jurisdictions; the government’s banning of the MDR, the largest Hutu opposition party; a referendum on a new constitution; the first parliamentary and presi­dential elections since the genocide; Rwanda’s increased involvement in conflict in the DRC; and an escalation of tensions with its neighbour and previous ally Uganda that many feared would lead to all-out war.

Conducting interviews in Ruhengeri, Gisenyi, Kigali Ville, Kigali Ngali and Butare provinces during this period, I heard a constant refrain of anxiety and confusion. In a village outside Ruhengeri, in the fertile volcanic hills near the border with Uganda, I met Gahiji, a fifty-five-year- old Tutsi farmer. He pointed to deep scars on his left cheek and said that during the genocide he had been attacked with a machete. Moments before the attack, he watched as his wife and two young sons were hacked to death only metres from where he stood. â€?We heard on the radio in January that the genocidaiτes were coming back’, he said. â€?At first, we were very scared. Since then, the government has told us nothing.’ He paused to take a long drink from a calabash. â€?Now they talk of an election. We have to walk to the municipal office to register [to vote] and we lose a day’s work on the farm. We just want to work and live in peace but that’s impossible with all of these things going on.’ Gahiji said that life had improved a little since the genocide: he had remarried, and his millet and maize crops had done well. â€?But we live every day with the memories of what happened’, he said. â€?Sometimes it seems too much to keep going. But God gives us strength, and somehow we keep on living.’^[362]

In 2003 and 2006, numerous interviewees living in communities near the Ugandan or Congolese borders cited insecurity and violence in neighbouring countries as a key reason why they were reluctant to participate in gacaca.^[363] Several hearings I attended in Ruhengeri and Gisenyi provinces in particular were called off because a quorum of par­ticipants could not be reached.^[364] Theogene, a forty-two-year-old farmer in Gisenyi whom I interviewed after the gacaca hearing in his jurisdic­tion was cancelled in May 2003, said,

There has been fighting again across the border and many Tutsi have been killed. People are on the move and taking all of their possessions with them. People here have loved ones [in North Kivu] and the fighting there worries them. They don't farm or go to the market or to church - they stay in their houses... They don't go to gacaca because there they see the Hutu and everyone is agitated because of what is happening [across the border].^[365]

Such statements highlight many Rwandans' continued sense of vul­nerability - especially among those living in the northern and western provinces - and the impact of this on their willingness to participate in public gatherings such as gacaca, which inevitably raise further emotive and often destabilising or traumatising issues.

This evidence from eleven communities over a seven-year period high­lights that gacaca is a highly fluid, dynamic process that relies heavily on the population's confidence and active involvement. At times, con­fidence and a desire to participate have been substantial in these com­munities, while at others they have decreased markedly. Whether people decide to attend gacaca hearings and participate depends greatly on the nature of the testimony and interactions during hearings and also on the prevailing social and political atmosphere. Even though many respond­ents stated that they were enthusiastic about participating in gacaca and believed popular ownership of the process was crucial, the observed real­ity of gacaca highlights that the population's participation in gacaca has varied substantially over time.

Finally, commentators' views on popular participation in gacaca pose various problems. The views of some Rwandan and non-Rwandan commentators are unconvincing for different reasons. Some Rwandan commentators overstate, in similar fashion to the government and population, the extent to which the community is able to direct, and participate in, gacaca. Many Rwandan commentators’ views of popular participation at gacaca suggest that dialogue in the general assembly will be entirely free and open, with state officials and other external parties wielding little influence over hearings. These commentators are undoubtedly justified in underlining the importance of the face-to-face dialogue between suspects and survivors, which may not occur without gacaca. However, the community’s dialogue is not as open and undir­ected as these sources suggest. As argued above, state officials play a significant role in ensuring security and adherence to the statutes and norms of the Gacaca Law and Gacaca Manual and may interfere for extraneous reasons. As Alphonse’s statements in the previous chapter highlight, external bodies such as Ibuka also play an important role in many gacaca jurisdictions, a point echoed by several other participants in gacaca whom I interviewed. ^[366] The level of popular participation that occurs at gacaca is significant, and unusual among post-conflict insti­tutions, but it is still much more moderate than some government offi­cials, segments of the population and Rwandan commentators suggest.

Furthermore, it is highly questionable that popular participation in gacaca will lead to greater democratic engagement and general political decentralisation in Rwanda, at least to the extent that Karekezi and Gasibirege contend. The attempt to connect democratic trends in the running of gacaca with broader political participation relies heavily, first, on the population feeling sufficiently empowered to engage in the wider political realm and, second, on the government increasing the capacity of the population to participate in public life. It is not obvi­ous that either of these conditions will soon be fulfilled. The public’s sense of empowerment after gacaca depends heavily on its experiences at gacaca, the degree of trust generated there, and whether individ­uals feel that gacaca addresses their needs and concerns. It also seems highly unlikely that the government will seek any time soon to develop the political capabilities of the Hutu majority, especially in light of its moves in 2003 to dissolve the MDR, the largest - and generally con­sidered moderate - Hutu opposition party, which amounts to denying Hutu any significant, official voice in the Rwandan parliament.^[367] To encourage greater democratic engagement among the majority of the population at the local level, while denying the political participation of the majority at the national level, would be highly contradictory.

Gacaca unquestionably affords the population a rare opportunity to debate issues concerning national reconstruction and to participate in rebuilding processes. It must be recognised that gacaca has empowered many groups in society who have often been marginalised in national life, especially women and youth, who have played central roles as gacaca judges and as participants in the general assembly. Francine, a sector­level gacaca judge in Musanze district, said,

Gacaca has allowed many women to speak publicly for the first time. They have spoken the truth clearly and talked openly about their per­sonal experiences. I have seen this happen in my community. Gacaca has allowed women to gain courage, especially those who have husbands in jail. These women provide for their families, they are trodden down, but gacaca gives them strength. This proves that women are capable of doing anything as well as men, so gacaca has been important for showing their strengths... Normally, in Rwanda women are apologetic but now [at gacaca] they are watching their men say sorry for their crimes, some­times even apologising directly to women, and women have found this empowering.

... It is true that talking about rape at gacaca has been very difficult for women. It is hard for women to tell the truth if they have been raped. I expect many challenges when these new [Category 1] rape cases come to gacaca, but it will be easier this time because we will be behind closed doors and some women have already given testimony at gacaca, so they are less afraid.^[368]

A recognition of gacaca's potential for empowerment partly explains the government's recent policy of establishing microfinance coopera­tives for many gacaca judges, using the organisational skills they have developed during gacaca and providing some form of recompense for their unpaid labour as judges over many years.^[369] It appears too that many gacaca judges will become abunzi or continue as inyangamugayo if the government follows through on its planned extension of gacaca to deal with everyday crimes after the genocide caseload is complete, suggesting judges will continue to play a vital role in Rwandan society. Francine said that inyangamugayo are likely to fulfil such a function:

After gacaca, if problems arise in the community, people can still go to the inyangamugayo to resolve problems without resorting to state institu­tions. If there are problems between survivors and the general popula­tion, we can bring families together to resolve problems. We will always be intermediaries after gacaca and we will continue to train the people in the importance of gacaca and its values, because gacaca is important for resolving problems... The inyangamugayo are not tired but courageous and ready to continue working in their communities.^[370]

Nevertheless, because of government restrictions on political participa­tion and empowerment at the national level, it is highly questionable whether popular participation within the confines of gacaca will lead to the community’s greater involvement in democratic enterprises outside of gacaca and thus substantially alter the national political landscape.

The views of popular participation in gacaca expressed by most non-Rwandan commentators, who in the main are critical of this method or believe that it is overstated, pose a different set of prob­lems. Responding to the criticisms of these authors, most of whom sub­scribe to the dominant discourse on gacaca, takes us to the heart of how the ethos and modus operandi of gacaca differ from those of more conventional transitional institutions. There are three main problems with non-Rwandan commentators’ critique of popular participation in gacaca, two related to the critique itself and the other to the alter­native modus operandi that these critics propose (often implicitly) in place of popular involvement.

First, the criticism that gacaca will weaken judicial safeguards as a result of greater communal participation is highly flawed. The key prob­lems with this complaint are that it ignores the various safeguards that are in place in gacaca to protect suspects from potential miscarriages of justice and how gacaca has operated in practice.^[371] Gacaca cannot pro­tect against all miscarriages of justice but it is unjustified to argue that

gacaca in no way protects individual rights. That judges are required to pass judgments and sentences on the basis of a consensus (or, when fail­ing to achieve consensus, on a majority) of nine judges, rather than on the opinion of a single judge, constructs an important layer of protec­tion for the accused. Judges must discuss cases in camera, where they are less influenced by the views of the community, before reaching a decision and communicating it to the general assembly. This forces judges to debate cases in private, often at great length, thus adding a cru­cial element of slow, critical consideration. My research indicates that, nationwide, approximately 25 per cent of gacaca cases have resulted in acquittals.^[372] The high acquittal rate may not be particularly surprising given that, reflecting the society as a whole, the majority of popularly elected gacaca judges are Hutu and may be inclined to judge their fellow Hutu sympathetically when they have been accused of genocide crimes. This situation is far from the brand of mob justice predicted by many human-rights observers of gacaca.

Furthermore, if individuals found guilty at gacaca believe they have not received a fair hearing, they may appeal the decision first to the juris­diction where they were initially tried and, if still dissatisfied, to the next higher jurisdiction. As Alphonse’s experience described in the previous chapter highlights, many suspects successfully appeal their gacaca cases. Twenty-nine of the eighty-two genocide suspects I interviewed around Rwanda after the commencement of the gacaca trial phase in 2005 said they had appealed their convictions or sentences. Twenty-four of these suspects, all of whom were subsequently either exonerated or had their sentences decreased, stated that they were satisfied with the decisions handed down at the appeals level.^[373] Measures such as gacaca’s appeals process are particularly important for protecting innocent suspects, who may feel that they cannot receive a fair trial in their home communities. Human-rights critics who equate popular participation in gacaca with lynch law ignore these protective features that are central to the gacaca process.

The Gacaca Law and Gacaca Manual also afford judges significant powers to control the content and tenor of evidence given at gacaca, in order to maintain decorum and security during hearings. Judges may stop individual testimony, banish antagonistic participants or halt entire hearings if they believe that certain testimony damages the overall pur­suits of gacaca, or if there is a threat of violence towards judges or members of the general assembly.^[374] One such intervention by a judge occurred at a gacaca hearing I attended in Kacyiru district of Kigali Ville. On a patch of grass beside a football field, a detainee wearing a bright red T-shirt, who had been released from an ingando less than a week before, arrived halfway through the hearing. A woman sitting in the general assembly spotted him and accused him of having burnt the roof of a house belong­ing to an old woman in the community, whose murder was being dis­cussed. After the woman described the alleged act of arson, the accused man stood at the back of the gathering and began shouting first at the woman who was giving evidence, then at the president of the judges' panel for allowing this testimony to continue. The president told him to stop talking and to let the woman speak. The accused man refused and kept shouting. Some friends of the man also began shouting at the president to stop the woman from talking. A group of other women in the gathering told the man to sit and wait for his turn to speak, which he did momentarily, but he soon leapt to his feet again. The president could do little as the man screamed at the assembly, �I know many things that I will never tell. Everyone here today should be on that list of killers.' Eventually, after the president left the bench at the front of the gather­ing to speak directly to the released detainee, the latter sat down and allowed the hearing to continue.^[375]

Examples such as this one highlight the tense atmosphere at many gacaca hearings and the judges' important mediation role, neither of which the government readily recognises. The government's descriptions of a near-absolute degree of popular participation in gacaca exacerbate commentators' concerns over the potential for miscarriages of just­ice. The government may be its own worst enemy in defending gacaca against the criticisms of human-rights observers concerning due process. The more moderate account of popular participation in gacaca that I propose here recognises the mediation role played by gacaca judges and, to a lesser extent, government officials, and therefore significant safe­guards for suspects, especially protecting the innocent from miscarriages of justice, and for all participants, in terms of protection from physical and verbal attack during gacaca hearings. I assess gacaca's performance in terms of delivering justice for genocide crimes in greater detail in Chapter 8.

Second, non-Rwandan commentators overstate the extent to which government officials interfere in gacaca hearings and jeopardise notions of popular participation. At the outset, this critique is inconsistent with the argument - often made by the same commentators, including AI and HRW - that gacaca is likely to result in mob justice, in which the state and other actors are powerless to intervene. In substantive terms, few of the non-Rwandan critics who make this point support the case with evidence from observed gacaca hearings.^[376] Ingelaere argues broadly that �local authorities do not play an overtly active role' during hear­ings but they have been instructed to write reports for higher author­ities. However, he does not elaborate on concrete instances where such reports have been written and the impact of this on the running of gacaca. Waldorf meanwhile cites examples of state coercion only in high- profile cases such as those involving Major General Laurent Munyakazi, a high-ranking Hutu military official, and Father Guy Theunis, a Belgian priest. While my own observations, as highlighted above, indicate that state officials do from time to time intervene directly in hearings, this is a generally uncommon occurrence and usually confined to commu­nities close to Kigali, where there is invariably a greater state presence. On three occasions that I have witnessed, government officials have intervened during hearings and been told by judges to desist from speak­ing on the grounds that the community should be free to debate the genocide evidence at hand. On all occasions, the officials followed the judges' orders.^[377] Such instances of judges standing their ground represent important moments of local agency that remain under-recognised in most non-Rwandan critiques of gacaca.

There is evidence, including from my own observations, that state offi­cials in some locations actively encourage - or even coerce - the popu­lation to participate in gacaca hearings. However, there is little evidence to suggest that this is a widespread phenomenon - the large number of hearings that have been cancelled due to low turnouts suggests quite the opposite - nor that it would undermine the importance of popular par­ticipation in gacaca.^[378] Waldorf's argument regarding coercion is prob­lematic: he argues first that low turnouts at hearings represent �peasant resistance' to state compellance of participation in gacaca and then later that �[l]ow participation rates have forced the state to employ coercion'.^[379] He attributes the slowness of the initial information phase of gacaca to low participation during hearings.^[380] However, I observed substantial participation during the information phase, with some of the highest turnouts at any point in the gacaca process, given that the population during that period viewed gacaca as novel and a source of relief after the long absence of any accountability process for genocide crimes. From my observations, I concur with Waldorf that the information phase was very slow and this diminished the population's enthusiasm for participation in gacaca, at least until the beginning of trials. The primary cause of this slowness, though, I contend, was not a lack of participation but rather the overly forensic nature of the early information-gathering, espe­cially arduous and highly contested discussions in the general assembly regarding property that people had lost during the genocide.

There is a further problem with some observers' arguments concerning government coercion of popular participation, namely that they centre around how spontaneously the population chooses to attend hearings rather than how actively it provides and debates testimony once hearings are underway. It is unclear why a state that requires citizens to participate in a legal process to which they could directly provide evidence should be seen as engaging in unjustifiable coercion. In such cases, there appears to be little difference between the actions of the Rwandan government and the legal requirement in many countries for citizens with relevant evidence or who have been summoned to jury service to actively partici­pate in trials.

Non-Rwandan commentators’ critique in this regard also presupposes that the Rwandan state is capable of penetrating deeply into the 11,000 jurisdictions where gacaca operates in order to fundamentally control the gacaca process as a whole. As Timothy Longman, Scott Straus and Lee Ann Fujii highlight in their analyses of the causes of the genocide and the motivations of those who committed murder, even a highly centralised bureaucracy such as Rwanda’s has limited reach into com­munal life. These authors counter a common view of the genocide as solely the result of government command from the centre to the periph­ery, highlighting that state orders were often rejected, there was regular dissension between elites and the general population at the community level, and in some places more localised motivations, such as fear and greed, better explain people’s decision to kill.^[381] Similarly, gacaca mani­fests some cases of government interference, but these should not be generalised to characterise the process as a whole, or to undermine the importance of popular agency in gacaca. In analysing the operation of gacaca, Longman (who initially expressed some concerns about the pos­sibility of state control over gacaca^[382]) argues, �Several factors help to pro­tect the independence of the gacaca courts... [T]he sheer number of judges on each panel helps to make influencing them more difficult... Related to this, the sheer number of courts will make it difficult for any individual or group to influence the entire process.’^[383]

A recent case highlights the Rwandan government’s limited knowledge of many developments in gacaca jurisdictions far from Kigali. In 2006, the Rwandan and UK governments signed a memorandum of understanding to facilitate the extradition of four genocide suspects from the UK to Rwanda. Defence lawyers for the accused succeeded in appealing to the UK High Court against their extradition.^[384] During the High Court case, family members of two of the accused, Emmanuel Nteziryayo and Celestin Ugirashebuja, who were living in Rwanda alerted the defence team that their loved ones had already been tried in absentia by gacaca jurisdictions in Southern Province.^[385] At the time, it was contrary to Rwandan criminal law for individuals pending extradition from foreign states to be prosecuted through gacaca. Kigali-based officials in the National Public Prosecution Authority and the NSGJ were unaware of the attempt by sector-level gacaca judges to prosecute Nteziryayo and Ugirashebuja. ^[386] This example highlights that even in the cases of high-profile genocide suspects, central government authorities are often uninformed of developments in far-flung gacaca juris­dictions. Given the potential legal complications for the Rwandan govern­ment precipitated by Nteziryayo’s and Ugirashebuja’s domestic prosecution before their extradition from the UK was resolved, the Kigali authorities would certainly have had a keen interest - and the legal basis - to intervene to halt these gacaca trials. However, they were unaware of these events in the countryside and therefore made no attempts to intervene. This example underscores why we should be highly sceptical of analyses characterising gacaca as simply another means of Rwandan state control over the entire country. Such accounts greatly overestimate the ability of the government to monitor and influence directly a court system as diffuse as gacaca. As I will show in later chapters, evidence from communities around Rwanda also highlights that the population moulds the objectives and methods of gacaca in ways that often diverge substantially from government intentions, underscoring the importance of popular agency in gacaca.

The third major problem with non-Rwandan observers’ criticisms of popular participation in gacaca is their proposed alternative modus operandi. As argued in Chapter 3, the dominant discourse criticises gacaca for failing to provide for formal justice. This critique provides an implicit, alternative view of how gacaca should operate. The for­mal approach to gacaca would apply what human-rights critics con­sider a form of legal due process, such as the one assumed to operate in institutions such as the ICTR, but which they perceive as currently lacking in gacaca. Judges would limit interactions in the general assem­bly to discussions of facts considered critical to determining the guilt or innocence of suspects. Members of the general assembly would be encouraged to respond only to questions from judges and not to debate with one another during hearings. Lawyers in turn would be present to advise survivors and suspects on how best to construct their respect­ive arguments and to intervene in hearings if they believe that judges are contravening the Gacaca Law.^[387] The claim concerning the need for lawyers at gacaca is the only part of this formal alternative that the dominant discourse on gacaca outlines explicitly and consistently. The other components of this alternative approach just outlined are nevertheless consistent with the sorts of procedural criticisms that proponents of the dominant discourse direct at gacaca. Therefore, if the orchestrators of gacaca were to reform the institution in line with such criticisms, then gacaca would need to incorporate all of the for­mal elements just expressed.

The formal alternative to gacaca is inadequate for both political and practical reasons. At the political level, the main problem with the for­mal approach to gacaca is that it defies the spirit of gacaca discernible from the government’s and population’s views. The population expects to participate in largely open, undirected hearings, in front of judges they have elected, and to debate and discuss both legal and non-legal issues. The population expects that gacaca will function very differently from a conventional courtroom. The formal approach implied by human-rights critics would prove alienating, distancing the population from the work­ings of a judicial system in which it would be entitled to participate only when called as witnesses and only in response to questions from judges and lawyers. Such strictures would greatly limit interactions between participants in gacaca.

Embodied in the various Rwandan sources’ interpretations of popular participation in gacaca is a discursive understanding of the way gacaca is expected to operate. According to the discursive view, participants in gacaca should feel free to discuss issues that are crucial to their personal and communal experiences during and after the genocide. Whatever �truth’ may be discovered in gacaca will be reached through commu­nal dialogue, not through the views of elites imparted to the popula­tion. Such dialogue may be messy, may take a long time and may in the end produce rather inconclusive results; there can be no doubting the risks inherent in the discursive approach embodied in gacaca. However, much of the population views the action of communal dialogue as inher­ently valuable. It contends that gacaca draws together people who may have, for reasons of protracted conflict, found it difficult to discuss mat­ters of individual and mutual importance in the past. In this view, gacaca encourages participants to discuss crucial issues in an open environment where the community as a whole may benefit from hearing, and contrib­uting to, such dialogue.

Because the formal alternative seeks to minimise communal involve­ment by giving an increased role to judges and lawyers, it directly opposes most Rwandans’ self-definitions and their dialogical interpretation of gacaca. Viewing the negotiated approach as a potential cause of further acrimony and violence, the formal version of gacaca advocates an alter­native that would lack popular legitimacy. Critics who advocate a more formal approach to questions of justice generally, and who specifically criticise gacaca for failing to meet formal requirements, would undermine the popular participatory spirit of gacaca and therefore seek to impose on gacaca a set of guidelines that the population is unlikely to consider appropriate. Because the Gacaca Law and Gacaca Manual enshrine the centrality of the population’s acceptance of, and involvement in, gacaca, as argued in Chapter 3, the question of ensuring gacaca’s popular legit­imacy is of the utmost importance.

Apart from the political problems already outlined, two main prac­tical problems with the formal alternative display its undesirability and unfeasibility. First, the proposal made by human-rights critics that lawyers be included in the gacaca process fails for both pragmatic and normative reasons. On a pragmatic level, Rwanda lacks the number of lawyers necessary to ensure that hundreds of thousands of genocide sus­pects have equal representation at gacaca. Most of the country’s lawyers died during the genocide, a small number have since been trained and even fewer international lawyers are on hand to assist. This practical constraint is in turn ethically significant. To allow some individuals to benefit from the assistance of lawyers, while others, operating within the same hearing and who may be arguing against those acting on expert legal advice, are unable to gain access to lawyers for financial or other reasons, is to introduce an unacceptable form of inequality of assistance into the gacaca process. It is preferable therefore to remove lawyers com­pletely from gacaca, thus ensuring that no participants in gacaca gain an advantage over others by having access to legal assistance that is not available to all participants.

In practice, lawyers continue to play an important unofficial role in gacaca, especially as many participants consult lawyers between gacaca hearings regarding how best to frame their evidence when their turn comes to testify. One lawyer in Musanze district said,

I haven't dealt directly with genocide cases but I sometimes advise on gacaca cases, particularly for individuals who have been convicted. They often come to me asking what they should do... I advise them as best I can, knowing the Gacaca Law and also how gacaca works. I go to gacaca, not as a lawyer but just as a normal member of the com­munity, so I know how gacaca functions. I can't advise people during hearings but there is nothing stopping me advising them if they come to my office.^[388]

Nevertheless, to allow lawyers to operate within hearings, particularly to speak on participants' behalf, would have the deleterious effects out­lined above.

Second, the strictly formal approach to gacaca severely limits the range of issues that the community can discuss and debate during hearings. This narrower discourse is not only problematic because it fails to meet most Rwandans' expectations of how gacaca should oper­ate, and therefore lacks popular legitimacy, but at a pragmatic level it means that the community cannot pursue certain objectives (par­ticularly those that may not necessarily relate directly to formal just­ice). The formal approach would limit gacaca hearings to discussions of legal facts, to the exclusion of many emotionally motivated expressions which, as we have seen, much of the population considers valuable functions of gacaca.

The shrinking of the dialogical space inherent in the formal approach to gacaca would stymie gacaca's pursuit of both legal and non-legal ends. Regarding legal objectives, communal dialogue in an open forum, where issues can be debated and discussed, is important for gacaca judges to make reasoned decisions about genocide cases. Survivors in particular can ask questions directly of those who com­mitted crimes, which rarely occurs in more conventional legal settings, and the accused are permitted to respond. Judges may also hear evi­dence in an open, communal setting that they would not necessarily glean if they were limited to hearing testimony only from witnesses whom they had called. This more fluid exchange of views, in which judges act as mediators, can provide crucial information for determin­ing the guilt or innocence of genocide suspects. The human-rights interpretation, which holds that significant communal involvement in a dialogically based legal setting automatically leads to unfair or biased decision-making, therefore neglects several important ways in which the discursive approach to discovering legal �facts' may not only be safeguarded against miscarriages of justice, as gacaca is designed to do, but may even in some instances be more legally beneficial than more conventional methods of criminal justice.

Concerning non-legal ends, the formal model of gacaca would bar participants from expressing views and emotions that do not neces­sarily concern judicial cases but are nonetheless considered important for other reasons. As I have already shown, many survivors view this greater sense of freedom of expression during gacaca hearings, relative to those in conventional courtrooms, as important for fulfilling their emotional and psychological needs after the genocide. Furthermore, the presence of lawyers, as the formal approach to gacaca requires, would undermine the content and tone of open, largely undirected, communal discourse that is otherwise possible during gacaca hear­ings. The presence of lawyers would significantly alter the dynamic between members of the general assembly, increasing the use of tech­nical legal language and modes of argumentation, and alter power dynamics as fully trained lawyers operate in a space where (often min­imally trained) judges are supposed to be the primary facilitators of gacaca. In such a situation, the population would feel more inhibited and intimidated than in a forum where they are among their neigh­bours, giving evidence before judges whom they have elected. In advo­cating the inclusion of lawyers in the gacaca process, critics of the institution risk undermining the popular ethos of gacaca and the modes of dialogue that ensue at gacaca.

Thus, to summarise, all of the Rwandan sources investigated here - official, popular and critical - view popular participation in gacaca not only as the modus operandi of gacaca but more importantly as a valu­able systemic expression of a Rwandan worldview of human identity as communally embedded and �truth’, both legal and non-legal, as a negoti­ated outcome reached via communal discussion in public settings. In the main, these sources overstate the extent to which the population can engage in a completely open, undirected communal dialogue at gacaca. State officials and gacaca judges in fact play a vital role in mediating the dialogical space at gacaca, often in order to maintain a sense of order and security and to ensure that gacaca pursues the objectives for which it is designed. However, this elite involvement in gacaca is still relatively minimal.

In contrast, most non-Rwandan commentators, basing their views on the formal interpretation of legal process, argue that the level of popular participation that they discern in gacaca inevitably entails a form of mob justice. The formal perspective holds that any criminal­justice system comprising significant communal, dialogical processes inevitably constitutes a biased, unjust system that cannot safeguard individual rights. The more moderate view of popular participation at gacaca proposed here - with its emphasis on the community’s partici­pation in a largely open dialogical space, mediated by gacaca judges and the state and constrained by the Gacaca Law - shows that human­rights critics are wrong to criticise gacaca for failing to provide for impartial decision-making and to safeguard individual rights. The spirit of gacaca enshrines local actors as the most crucial participants. This popular ethos must be maintained if the majority of Rwandans are to view gacaca as a legitimate remedy to the legacies of the geno­cide. Mediators, however, still play a key role in helping the population achieve its intended objectives through gacaca, increasing the popula­tion’s trust in gacaca and thus bolstering gacaca’s popular legitimacy. The modus operandi of gacaca therefore entails largely unrestricted popular participation within clearly defined boundaries designed to protect individual rights and to direct communal discussions towards fulfilling gacaca’s aims. It is gacaca’s hybridity - in this case, the com­bination of discursive methods and formal constraints - that most sources overlook.