participate, and individuals are cast into the following prescripted roles: (1) citizen

monitor the process (field notes 2006). State agents in uniform were often armed with AK-47s or other

instruments of force, including truncheons. Plain-clothed agents of the state were omnipresent but

played no official role in gacaca proceedings. These individuals made no attempt to hide their

presence, and, in my experience, everyone in attendance knew well who the state agents were and why

they were there. Alice said it well: “Of course we all see them. And we all know why they come. To

get justice the way it is supposed to be done. They talk only to those [appointed] officials, and they

have covered shoes and sunglasses. It is very obvious they come from outside [our cell]. One who

came to witness [gacaca] proceedings didn’t actually speak enough of our language [Kinyarwanda] to

observe properly and had to get help from one of the inyangamugayo who he knows. We never saw

him again. The [current] observer always sits there, away from the other observers” (interview with

Alice, a poor Tutsi widow, 2006).

At the level of the ordinary individual, local officials who are willing to uphold the idea of the state

make it real in everyday life. As analyzed in chapter 4, both appointed and elected local officials are

chosen for their ability and willingness to effectively and efficiently promote national unity and

reconciliation or risk sanction from their bosses in Kigali.3 If a local official is unable to force

individuals resident in his bailiwick to perform the demands of the policy of national unity and

reconciliation, the official is likely to be replaced from the center with someone who is more able to

do so. Actions such as replacing local officials are an indication of the extent to which the

postgenocide government recognizes that its policy of national unity and reconciliation relies on

coercive actions; it is hardly something that many Rwandans deem legitimate or just.

The practices of the Rwandan state in pursuit of national unity and reconciliation, such as the

gacaca courts, take place in articulated “state spaces” (Scott 1998, 186). In the process of controlling

the spaces where unity and reconciliation can officially occur, the Rwandan state has neutralized or

eliminated nonstate spaces, thereby rendering these spaces suspect, which in turn constrains the ability

of many ordinary Rwandans to resist practices of national unity and reconciliation. Such acts that

occur outside the gaze of the state do not officially count, as there is no official present there to

register and legitimate the encounter. Reconciliation must be legible to all—the gacaca courts

reiterate the power of the postgenocide state to produce the image of lasting peace in the form of a

unified “Rwandan” identity. Projects such as the policy of national unity and reconciliation “reveal

both the power and the limitations of the regime’s project by announcing the gap between enforcing

participation and commanding belief ” (Wedeen 1999, 22).

members of her community look on. The individuals seated at the desk, wearing sashes, are judges.

Those seated at the desk without sashes are government observers. (photo by Anne Aghion, My

Neighbor, My Killer, © 2009)

the relation between individuals and the state and through which the idea of the state is embodied and

felt. This moves the analysis beyond “the apparatus of government to show how the magic and power

of the state are forms in everyday discursive practice” (Crais 2002, 25). An important aspect of state

power is the management and appropriation of the symbolic world (Wedeen 1999, 30). The power of

the unity-generating Rwandan state is its ability to force individuals to reconcile according to

predefined roles and in the dissemination (largely through elected local power brokers) of credible

threats of punishment for noncompliance. Individuals behave as if they are being watched because

they are—gacaca is a public and compulsory spectacle (see fig. 12). Most ordinary Rwandans also

understand that the appearance of compliance is an important aspect of performing national unity and

reconciliation. Thus, the state commands orchestrated displays of individual obedience to the policy

of unity and reconciliation while also producing tactical compliance through the actual or anticipated

use of punishment. Another important aspect of the culture of the state is the climate of fear it

produces and the self-sanctioning behavior that individuals adopt for fear of being punished for

noncompliance.

In spite of this strong state presence and the ubiquity of local officials who work to ensure that as

many individuals as possible participate to uphold the façade of national unity and reconciliation that

the gacaca courts sought to create, many ordinary Rwandans exercise their agency in strategic and

creative ways to show their opposition or indignation to what they perceive as unjust actions against

them or members of their family. As we saw in chapters 4 and 5, the risks associated with direct

—can be severe. As a result, ordinary Rwandans practice minute everyday acts of resistance. They

include some combination of persistence, prudence, and individual effort to accomplish a specific

goal that benefits the individual resister against a local official or other state agent who is unaware

that the individual is attempting to resist or subvert the demands of national unity and reconciliation.

An analytical focus on the everyday acts of resistance to gacaca trials reveals the subtle, indirect, and

nonconfrontational ways in which ordinary Rwandans do more than mitigate the demands of their

mandatory participation. They also creatively and strategically expressed dissatisfaction with their

inability to discuss RPF war crimes, land conflicts, and their continued poverty in the presence of

government officials during gacaca. Before examining in further detail the specific everyday acts of

resistance of Tutsi survivors to illustrate the extent to which gacaca is a mechanism of state power,

the next section demonstrates how individuals are expected to perform certain roles before the gacaca

courts in accordance with the script of national unity and reconciliation.


The Gacaca Courts in the Promotion of National Unity and Reconciliation


The gacaca courts are at the heart of government efforts to achieve national unity and reconciliation

since the 1994 genocide “ruptured the social fabric” (interview with NURC official 2006). Little is

known about the actual functioning of the gacaca courts in pursuit of the goal of promoting national

unity and reconciliation despite a voluminous academic literature (for analysis see Ingelaere 2012).

This is largely because researchers have focused predominantly on the protection of human rights and

the ability of gacacacourts to uphold international standards of criminal justice and legal

accountability, rather than on outcomes at the level of the individual. Political psychologists and

trauma specialists have focused on the healing potential of the courts and the importance of justice as

a precondition for reconciliation but are silent on how unity and reconciliation play out in the lives of

ordinary Rwandans.

A common thread in these analyses of the gacaca process is the assumption that ordinary Rwandans

are willing participants. Schabas (2005, 897) claims, without any reference to his methodology or

sources, that “Rwandans have consistently rejected any compromise with full accountability, insisting

upon criminal prosecution for all alleged perpetrators.” Pham et al. (2004, 603) provide more

information about their methods but do not state the specifics of their sample or the circumstances in

which their survey was completed: “More respondents supported the local judicial responses (90.8

percent supported gacaca trials and 67.8 percent the Rwanda national trials) than the ICTR (42.1

percent in support).” Official government documents contend that “the autocratically divisive political

structures that once denied minorities a political voice have been replaced, for instance with the

implementation of democratic cellular councils that involve local communities in important decisions

at grassroots level” (interview with NURC official 2006; NURC 2010, 30–32). What these sources

obscure is the extent to which gacaca represents yet another “state space” where individual Rwandans

are observed and monitored for their compliance with the script of national unity and reconciliation.

Gacaca started as a pilot project in June 2002 and was later refined in 2004. The most notable

refinement was the merging of categories 2 and 3 (intentional murder, attempted murder,

manslaughter, and assault cases) into a single category. Gacaca was launched nationwide in January

2005, with the government estimating that it would take ten years for the courts to complete the

process. In fact, the government declared the process completed in 2012, with an estimated few

my fieldwork, the gacaca trials at the sector and cell levels were taking place throughout the

communities where the ordinary Rwandans who participated in my research lived. Trials in these

communities had already passed the first stage of gacaca, the information-gathering stage, and the

courts were involved in the second and final stage of judgment and sentencing.

The gacaca process is a top-down and highly centralized process that is overseen by the National

Service of Gacaca Jurisdictions (NSGJ), which is a chamber of the Supreme Court of Rwanda. Its

primary task is to coordinate the transmission of prisoner dossiers between the cell- and the sector-

level courts and to connect these courts to the local offices of the NSGJ, which are housed in the

capital of each province (NSGJ 2005/6, 2006). Each district-level office is linked with other

government offices, including the prosecutor’s office, the prison, and the relevant line ministries, and

with the central coordinating office in Kigali (NSGJ/ASF 2002, 3–5). The NSGJ also has the legal

authority to monitor and advise the gacaca courts, including overseeing the election process for

judges, and to intervene at the local level when judges “are not in control of the proceedings”

(interview with local NSGJ official 2006). NSGJ staff at the lower levels of the bureaucracy must

remain abreast of activities at the level of the cell and the sector and report these to authorities at the

provincial and the national levels. They must also complete weekly and monthly reports for

transmission to authorities at MINIJUST (NSGJ 2005, 7).

Local NSGJ staff also work with senior government officials from the Ministry of Justice and

NSGJ in Kigali to “sensitize the population” and to ensure that nyumbakumi (“responsible for ten

houses”)—the lowest administrative official—assist gacaca judges to investigate any involvement of

individuals in committing acts of genocide in 1994 (interview with local NSGJ official 2006).

Sensitization campaigns target rural populations to encourage people to participate out of self-interest

and to further national unity and reconciliation. This quotation from a local official reveals the extent

to which the full participation of the population is required: “As for you saying you will not forgive

him or that you will not do this or that, that is very bad. Whether you like it or not, that is the law”

(quoted in PRI 2002, 23). Mass participation is crucial to the Rwandan state’s ability to generate an

image of national unity and reconciliation: perpetrators (génocidaires) must provide truthful

information about what they and their accomplices did during the genocide or go to prison for at least

fifteen years. Before any truth telling takes place, a panel of nine judges (inyangamugayo) has already

amassed a dossier of evidence against which the truth is adjudicated. Witnesses corroborate, revise, or

reject the evidence presented by the inyangamugayo or the testimony of survivors and/or

génocidaires. Citizen spectators observe the proceedings, and their en masse presence adds an air of

credibility and legitimacy to the proceedings, as it shows external observers (i.e., Westerners) that the

government at least tries to treat its citizens fairly in the pursuit of justice for crimes committed

during the genocide. The performance of survivors is critical to the spectacle; survivors act both as

accusers and, once the truth has been established, as magnanimous individuals who are able to forgive.

Tutsi survivors bear the brunt of the burden of providing the truthful record of how the genocide

happened, and they can be punished if they fail to “tell all of their truth” (interview with local NSGJ

official 2006).

In practice, however, gacaca courts are a contested and conflicted state space and are characterized

by discord and tension between the various actors. Citizen spectators I consulted spoke of the ways in

which local authorities, usually one or more of the nyumbakumi, the cell or sector coordinators,

demand individual attendance, with the threat of official sanction. Witness this from Béatha, a

destitute Hutu widow with children:





He [the cell coordinator] came, and he asked me why I didn’t attend the gacaca.

He came on a motorcycle, so everyone [in the community] knew. I told him

because my son is sick and I stayed home with him. He wrote something in his

book and said, “Next time you have to come. I don’t care the reason.” So I got

scared because my son wasn’t sick at all! It was that I had no money to spare for

transport, but I didn’t tell him because I know that is no excuse. The radio says

we have a responsibility to participate. Last time we had a meeting [at the office

of the cell coordinator], the official said that there was no excuse for not

participating. . . .

But I didn’t see anything, so what can I say? I heard people getting killed, but

I was in hiding. I mean, I hid when they [the killers] came. Who didn’t? We all

hid in those days [during the genocide]. I don’t know who saw what, but I say

that I saw nothing. I heard the official tell someone whose name I know to kill,

but I didn’t see it. If he killed, I don’t know. So why go if I have nothing to say

and if I have no money? How do I get money to go? I am not a survivor so am

not a member of an association; money for school uniforms is now due. Now I

have a big headache because he knows I did not go and gacaca is again next

week. Next week! And now my neighbors know he [the cell coordinator] has

come, so I have to go.

I also fear because if my neighbors find out that I said my son was sick they

might use that information to denounce me [to the cell coordinator]. What will I

do then? (Interview 2006)

those who, in the eyes of the government, are peripheral to the success of gacaca as a tool to generate

national unity and reconciliation. As a Hutu woman, she has a limited role in the gacaca, since she is

not expected to testify as a Tutsi survivor or to tell her truth as a perpetrator. Instead, her role is to act

as a citizen spectator to give the appearance of popular support for the courts. Her words are also

symbolic of the self-monitoring behavior that characterizes postgenocide processes of justice and

reconciliation. In addition to the strain of meeting the official requirements of gacaca, Béatha refers

to her neighbors’ awareness of her reason for not attending the gacaca session. Official state-based

sanctions (fines, imprisonment) for nonparticipation matter, but so do the watchful eyes of friends,

neighbors, and, in some instances, family in ensuring the full participation of the population at

gacaca. There may indeed be other reasons why Béatha prefers not to attend the weekly gacaca

session. She may want to avoid being falsely accused by neighbors, or she may know more than she is

willing to share and prefer not to participate to avoid forced contact with her neighbors. The point is

that the gacaca process is for many ordinary Rwandans an oppressive form of state power that forces

them to participate in ways that are not necessarily in line with their own lived realities.

Key actors in ensuring the smooth operation of the gacaca courts are the judges, the

inyangamugayo. Judges are officially “elected” by members of the communities they are to serve

under local authorities. Judges are constantly balancing their privileged role as elected (although

unpaid) officials against the requirements of living as a member of the community they are to serve.

Judges have little autonomy when it comes to rendering judgment. NSGJ official in Kigali resolve any

confusion about the scope of the gacaca law or the relevant procedures. Judges simply conduct the

gacaca trial in a fair and impartial way. In theory, “fair and impartial” means that, in the course of

procedural integrity of the trial, judges can draw on their own personal experiences and insights. In

practice, it means that judges are under constant surveillance by both local government authorities and

community members, both of whom can report any wrongdoing, real or perceived, to state authorities.

The gacaca law itself is also a constraint, as judges are duty bound to respect its rules and regulations

at all times, even though they receive minimal training and support in how to do so. Judges, because

of the important role they play in facilitating the smooth operation of the gacaca courts, are expected

to spur génocidaires to tell the truth and also to ensure that survivors are able to offer an act of

forgiveness. Because of this power, judges are often reminded by the state of their duty to ensure the

active participation of the population at gacaca.

Ensuring the active participation of the population is not without its risks. Didier’s lament is

emblematic of the thoughts of the judges with whom I spoke:

I can share my difficulties with. I am a survivor like the other survivors, but

even they look at me differently. I feel isolated since I became a judge. And

because I have a family, I have responsibilities to them. My wife even asked

how I could let my family starve when it was time to harvest. But I couldn’t do

that work because it [being a judge] takes at least two days a week, sometimes

more. . . . If I don’t undertake my duties as a judge, I can get into serious