RW-M-01: Ndi Umunyarwanda and the Post-Ethnic Citizenship Project (1994–2026)

Status: DRAFTWords: 9,960

1. Key Takeaways

Post-genocide Rwanda is conducting the modern world's most radical experiment in state-led identity engineering: not the management of ethnicity (the consociational route), not its celebration (the multicultural route), but its attempted legal abolition. Since 1994 the Rwandan state has removed ethnic designations from identity cards and the census, criminalised public ethnic categorisation under the divisionism and genocide-ideology laws, rebuilt the school curriculum around a unitary national history, and constructed a citizenship-formation machinery — ingando solidarity camps, the itorero civic academy, the Ndi Umunyarwanda ("I am Rwandan") programme, the Kwibuka commemoration cycle — whose explicit objective is that the categories Hutu, Tutsi, and Twa cease to function as social identities. The official formula is categorical: there are no Hutu and no Tutsi, only Banyarwanda. No other state has attempted ethnicity's erasure as formal policy; Rwanda's project is therefore the limiting case against which the entire comparative literature on post-conflict identity management must be read.
The project rests on a specific historical diagnosis — that ethnicity in Rwanda was a colonial fabrication, and that what colonialism made, the state can unmake. The official historiography holds that pre-colonial Rwanda was a unified nation sharing one language, one territory, and one set of institutions, in which "Hutu" and "Tutsi" were fluid socio-economic positions rather than ethnic identities; that German and especially Belgian colonialism racialised these categories through Hamitic-hypothesis pseudo-science and the 1933–34 census that fixed ethnicity on identity cards; and that the 1959–94 sequence of revolution, quota-state, and genocide was the terminal harvest of that fabrication (the colonial/structural account is carried in full at RW-J-01 §4). Professional historiography complicates the narrative — the categories pre-existed colonialism and hardened under the pre-colonial expansionist court before the Belgians racialised them — but the diagnosis's political function is independent of its historiographical precision: it locates the genocide's cause outside the nation, making unity recoverable rather than aspirational, and it licenses the prescription that follows.
The prescription was enacted with unusual completeness: ethnicity was deleted from the state's entire information architecture, and its public invocation was criminalised. The new identity cards issued from the mid-1990s carry no ethnic field — a direct repudiation of the cards that had functioned as death warrants at the 1994 roadblocks; no census since 1991 has collected ethnic data; the 2003 Constitution commits the state to "eradication of ethnic, regional and other divisions" and entrenches the fight against genocide ideology; and a layered penal architecture — the 2001 sectarianism ("divisionism") law, the 2008 genocide-ideology law, its 2013 revision narrowing intent requirements after international criticism, and the 2018 penal-code restatement — makes public ethnic categorisation legally hazardous. The legal enforcement layer is the project's signature and its most contested element: the same statutes that ban the genocide's vocabulary have been deployed against opposition figures, journalists, and researchers (RW-N-01 §4).
The project's formation machinery is the most elaborate civic-pedagogy apparatus in Africa. Ingando solidarity camps (from 1996–99) processed returning refugees, demobilised combatants, released gacaca convicts, and university-bound students through weeks of residential re-education in the official history and the unity doctrine (RW-G-01 §6). Itorero ry'Igihugu, launched in November 2007 and given a national commission in 2013, revived a pre-colonial court institution as a civic-values academy that has put hundreds of thousands of Rwandans through training as intore — "the chosen," tasked with carrying national values into their communities (RW-G-01 §5; the RDF's parallel domestic-formation role at RW-I-01 §6). Ndi Umunyarwanda, launched in 2013, added a testimonial politics of public truth-telling and apology. The history curriculum, suspended for over a decade after 1994 and rebuilt around the unity narrative, and the annual Kwibuka commemoration's hundred days of mourning pedagogy (RW-N-01 §6) complete the apparatus. Together these instruments mean a Rwandan born after 1994 — now the large majority of the population — has been formed inside the framework at every institutional encounter from primary school to university entry to public employment.
Ndi Umunyarwanda itself, the programme that gives this document its title, exposed the project's central unresolved tension: between unity as shared citizenship and unity as scripted moral asymmetry. Launched in 2013 out of a youth-forum process and embraced by President Kagame, the programme staged public testimonials in which Rwandans affirmed Rwandan-ness above ethnicity — and in which Hutu participants, including senior officials, apologised for the genocide committed "in their name." Defenders read the apology dimension as voluntary moral leadership clearing the ground for trust; critics — including opposition figures prosecuted afterwards and the gospel singer Kizito Mihigo, whose 2014 song questioning the commemoration's selectivity preceded his arrest and his 2020 death in custody — read it as the imposition of collective ethnic guilt by a state that simultaneously declares ethnicity not to exist [TBD-VERIFY: the programme's contested episodes and the official responses are documented unevenly; see §3.3]. The controversy is not peripheral: it is the project's deepest paradox made visible, the moment the post-ethnic state required participants to speak as Hutu in order to transcend being Hutu.
The achievements ledger is real and must be stated at full strength: thirty-two years without ethnic mass violence inside Rwanda, in a country that experienced genocide within living memory, is the project's irreducible exhibit. No serious observer predicted in 1994 that the post-genocide settlement would hold without recurrence; comparable post-atrocity settings (Burundi next door, eastern DRC across the border) have cycled through repeated violence in the same period. The state's Rwanda Reconciliation Barometer reports reconciliation indicators above ninety per cent [TBD-VERIFY: the Barometer's editions, headline findings, and the methodology debates around survey research under dominant-party conditions; §4.1]; ethnographic work documents genuine cross-category cohabitation, intermarriage, and a generational cohort for whom the official identity is the default public self. Whatever discount is applied for preference falsification, the behavioural floor — no pogroms, no ethnic riots, no communal massacres on Rwandan soil since 1994 — is not a survey artefact.
The critique literature is equally serious and must be stated at the same strength: the dominant scholarly reading is suppression-not-transcendence — that the categories persist beneath an enforced public silence, and that the silence itself is asymmetric. Ethnographic scholarship (Susanne Buckley-Zistel's "chosen amnesia," Susan Thomson's everyday-resistance findings, Andrea Purdeková's ingando studies, Lyndsay McLean Hilker's work on youth and intermarriage [TBD-VERIFY: the precise findings and their dates; §5.1]) reports that Rwandans navigate ethnicity constantly in private while performing unity in public. The asymmetry critique cuts deeper: the official naming — the genocide against the Tutsi — is historically accurate as to the genocide's character, yet the surrounding architecture leaves Hutu victims of 1990–98 (war massacres, RPF/RPA killings, the Congo refugee deaths) without a public mourning category, making "Hutu" simultaneously unspeakable and operative: deniable as identity, inescapable as ascribed perpetrator lineage (RW-J-01 §5). The political-instrument reading completes the critique: the divisionism laws police the opposition as much as the genocide's ideology, and the unity project doubles as the RPF's legitimation architecture — the claim that ethnic politics means genocide, and that the RPF alone stands between Rwanda and recurrence, converts identity policy into regime insurance (RW-N-01 §4, §6).
The project's sharpest contradiction is regional: the ethnic vocabulary abolished at home is instrumentalised abroad. Rwanda's quarter-century of intervention in eastern DRC has been continuously justified in ethnic terms — the FDLR as a génocidaire threat, the Banyamulenge and Congolese Tutsi as kin requiring protection, M23's grievances as an extension of the anti-Tutsi persecution the genocide exemplified (RW-D-08 §12; RW-F-01). The state that prosecutes its citizens for saying "Hutu" and "Tutsi" deploys both categories fluently in its foreign-policy communications. Defenders resolve the tension by distinguishing domestic nation-building from external threat-description; critics read it as proof that the categories remain fully operative in the state's own strategic cognition, abolished only where their abolition serves control.
The project is approaching its first true stress test untested: it has never operated without its enforcer, and its subjects are now mostly people who never held an ethnic identity card. Kagame's terms run through 2034 under the 2015 constitutional arrangements (RW-K-01); the succession question (RW-O-01 §2) is therefore also an identity question — whether Ndi Umunyarwanda is an internalised settlement that survives political opening or a compliance regime that decompresses when enforcement weakens, on the precedent of Yugoslavia's "brotherhood and unity" or the Soviet druzhba narodov. The post-memory generation cuts both ways: it is the project's greatest asset (a majority formed entirely inside the framework, with no lived memory of ethnic classification) and its greatest unknown (an identity settlement never chosen, never contested, and never tested against open politics). The diaspora opposition's explicitly ethnic framings and the double-genocide discourse wars (RW-J-01 §5) preview what contestation would sound like.
Comparatively, Rwanda is the limiting case in two literatures at once. In the nation-building-after-atrocity genre it stands at the unity pole of the unity-versus-truth trade-off — against Germany's perpetrator-centred memory regime, Spain's pact of forgetting and its revisitation, and Yugoslavia's cautionary suppression — having chosen a single authorised narrative enforced by law. In the engineered-identity literature it radicalises the high-modernist nation-building tradition: Tanzania's ujamaa-and-Swahili nationhood and Singapore's managed multiracialism are its closest echoes, but Nyerere built a supra-ethnic identity over acknowledged ethnicities and Singapore institutionalised categories rather than abolishing them; only Rwanda attempts deletion. What the Rwandan case ultimately tests is the proposition embedded in its own founding diagnosis: whether states can unmake what states made. The verdict, on the evidence assembled below, is genuinely open — which is itself the most important finding.

2. The Project's Premise: The Post-Genocide Diagnosis

2.1 Ethnicity as the colonial fabrication that killed

Every identity project rests on a story about how the identities it addresses came to exist, and post-genocide Rwanda's story is unusually specific. The official historiography — articulated in the National Unity and Reconciliation Commission's foundational documents, in the ingando and itorero curricula, in the rebuilt school history programme, and in three decades of presidential commemoration speeches — runs as follows. Pre-colonial Rwanda was a unified nation: one language (Kinyarwanda), one territory progressively consolidated under the Nyiginya monarchy, one religion-and-ritual complex, shared clans (ubwoko in its older sense) that crossed the Hutu–Tutsi–Twa distinction, and institutions — cattle clientage, the court's itorero schools, collective cultivation — that bound the categories together. "Hutu" and "Tutsi" in this account were socio-economic and political positions, not ethnic groups: a Hutu who acquired cattle and standing could become Tutsi (kwihutura); the categories carried no myth of separate origin and no barrier to intermarriage. The Twa, forest-dwelling potters and hunters, were a small distinct community but within the same national fabric.

Onto this unity, the account continues, colonialism imposed race. German indirect rule (1897–1916) ruled through the court and absorbed the Hamitic hypothesis — the European pseudo-scientific doctrine that anything resembling statecraft in central Africa must have been imported by a superior, quasi-Caucasoid pastoralist race from the north-east. Belgian rule (1916–62) operationalised the doctrine: the 1933–34 census measured, counted, and registered every Rwandan as Hutu, Tutsi, or Twa, fixed the classification on compulsory identity booklets, reserved education, church advancement, and administrative office for the Tutsi minority it had racialised as natural rulers, and then — in the terminal colonial reversal of the 1950s — switched patronage to the Hutu counter-elite it had itself schooled in the vocabulary of race, midwifing the 1959 revolution and the quota-republics that followed. The First and Second Republics (1962–94) inherited and inverted the colonial race-state: the identity card remained, ethnic quotas governed schooling and employment, periodic pogroms (1959–64, 1973) drove the Tutsi exile waves from which the RPF eventually came, and the genocide of 1994 was the logic's terminus — the moment the card became a death warrant at the roadblock. The full colonial/structural account, including its use in the international-blame debate, is carried at RW-J-01 §4; what matters here is its domestic function as the project's premise.

The diagnosis does specific political work, and the work explains the project's confidence. First, it locates the genocide's deep cause outside the nation: colonialism fabricated the categories, the post-colonial republics administered them, and the genocide harvested them — therefore the nation itself is not congenitally divided, and unity is recovery, not invention. Second, it renders ethnicity artificial and therefore reversible: what an administrative act created (the 1933 census), an administrative act can dissolve (the post-1994 deletion). Third, it indicts the political use of ethnicity as such: if ethnic categories are the genocide's instrument, then ethnic politics — counting, mobilising, or claiming by category — is not a legitimate competing interest but a recurrence risk, which is the premise on which the criminal-law layer (§2.3) stands.

Professional historiography accepts much of the indictment and complicates the idealisation. The scholarship of Catharine Newbury (The Cohesion of Oppression, 1988), David Newbury, and Jan Vansina (Antecedents to Modern Rwanda, 2004) finds that the Hutu–Tutsi distinction pre-existed colonialism and had already hardened into a hierarchical, increasingly hereditary cleavage under the expansionist court of Rwabugiri in the late nineteenth century — colonialism racialised and bureaucratised a stratification it found, rather than inventing one from nothing. The pre-colonial-harmony narrative, on this reading, is itself a political construction: accurate about the Belgian racialisation, selective about the indigenous hierarchy. The corpus does not need to adjudicate the historiography here (RW-J-01 carries the contest); the analytical point is that the project's premise is stronger than it needs to be. Even if ethnicity were partly indigenous, its lethal racialised form was demonstrably administrative — and the project's real wager is not historiographical but practical: that whatever the categories' origins, a state that controls schooling, law, commemoration, and public language for two generations can change what its citizens are.

2.2 The prescription: deletion from the state's information architecture

The prescription followed the diagnosis with the directness that has characterised RPF governance throughout (RW-A-01 §7). The Government of National Unity's 1994 founding programme, built on the Arusha framework's protocols minus the defeated parties, declared the end of ethnic and regional discrimination; the operative deletions came in sequence. The identity card was reissued without the ethnic field in the mid-1990s — the single most symbolically charged act available to the new state, since the old card's ubwoko line had sorted the living from the dead at the 1994 roadblocks (RW-B-01 §6). The census stopped counting: the 1991 census was the last to record ethnicity, and the 2002, 2012, and 2022 censuses collect none, with the consequence — remarkable and insufficiently remarked — that the Hutu–Tutsi demographic ratios cited in every account of Rwanda are now thirty-five-year-old extrapolations. No Rwandan official statistic, school register, employment file, or administrative record has carried an ethnic field for three decades. The doctrine received its formula in the RPF's standing rhetorical repertoire: there are no Hutu, no Tutsi, no Twa — only Banyarwanda, the people of Rwanda. Kagame's variants over the years — that he is the president of Rwandans, not of ethnicities; that ethnicity is the politics of those who killed — made the formula the regime's first-person creed.

The deletion's designers understood that an unrecorded category is not an abolished one, and the prescription therefore had a second, positive face: the construction of Ndi Umunyarwanda — Rwandan-ness — as a thick replacement identity, with its own history (the recovered pre-colonial unity), its own values vocabulary (ubumwe, unity; ubupfura, integrity/nobility; kwigira, self-reliance; agaciro, dignity — the last elevated into a national keyword and the name of the sovereign Agaciro Development Fund), its own revived institutions (umuganda communal work, RW-G-03; itorero; ubudehe mutual assistance; gacaca, RW-A-02), and its own formation machinery (§3). The home-grown-solutions framing did double duty: it supplied authentically Rwandan content for the new identity, and it positioned the unity project as decolonisation — the recovery of what colonialism broke — rather than as social engineering.

2.3 The constitutional and penal architecture: the project's legal enforcement

The third element of the prescription made the new identity legally compulsory in public space. The 2003 Constitution (RW-A-01 §8) wove the project into the state's foundations: the preamble's resolve to "fight the ideology of genocide and all its manifestations" and to "eradicate ethnic, regional and any other form of divisions"; the entrenchment of power-sharing provisions designed to prevent ethnic majoritarianism (the constitutional bar on the winning party holding more than half the cabinet, the prohibition of parties based on "ethnic group, tribe, clan, region, sex, religion"); and the establishment of the National Unity and Reconciliation Commission (created by statute in 1999, constitutionalised in 2003, folded into the Ministry of National Unity and Civic Engagement in 2020/2021 [TBD-VERIFY: the exact date and statutory form of the NURC's absorption into MINUBUMWE]) as the project's permanent bureaucracy. The 2008 constitutional amendment renamed the event itself — "the genocide against the Tutsi" replacing the more generic genocide references — a naming whose accuracy and whose asymmetric implications are treated at §5.2.

The penal layer came in three waves. Law 47/2001 criminalised "discrimination and sectarianism" (amacakubiri — divisionism): the use of speech, writing, or action based on ethnicity, region, or religion to divide. Law 18/2008 criminalised "genocide ideology" as such — defined broadly enough (thoughts, speech, "dehumanising" conduct, even by minors) to draw sustained criticism from Amnesty International, Human Rights Watch, and donor governments for vagueness and for its use against political speech; the 2013 revision (Law 84/2013) responded by adding an intent requirement and more specific definitions, and the 2018 penal code restated the offences with adjusted penalties [TBD-VERIFY: the precise penalty ranges across the 2008/2013/2018 instruments and the annual prosecution statistics, which are reported unevenly]. The architecture's dual character is the document's recurring theme stated in legal form: the laws criminalise the genocide's actual vocabulary — the radio-borne categories that organised the killing (RW-B-01 §5) — and they criminalise much of what political opposition in a post-genocide society would naturally say, since any appeal to Hutu majority grievance, any claim that RPF crimes deserve equal mourning, any contestation of the official history is chargeable as divisionism or genocide ideology. Victoire Ingabire's 2012 conviction (on charges including genocide-ideology offences, following her 2010 statement at the Kigali Genocide Memorial that Hutu victims also deserved remembrance) is the canonical case, and the opposition-policing function is documented at RW-N-01 §4. Both readings of the architecture — necessary firewall and political weapon — are developed at full strength in §4 and §5.

3. The Project's Instruments: The Citizenship-Formation Machinery

3.1 Ingando: the solidarity-camp architecture

The project's first mass instrument predated its formal doctrine. Ingando — from kuganda, to halt and encamp, a term with pre-colonial military resonances — began in 1996–97 as the receiving mechanism for the enormous returnee flows from Zaire and Tanzania (RW-A-01 §5): residential camps, typically several weeks to three months, combining civic re-education in the official history with practical reintegration. The format was rapidly generalised into the new state's standard processing institution for populations crossing a status boundary: demobilised soldiers and ex-FAR/militia combatants returning from the DRC; provisionally released génocidaire prisoners ahead of gacaca (some 40,000+ in the 2003 and subsequent release waves passed through ingando before returning to their hills); students between secondary school and university, for whom ingando became an effective entry requirement for government scholarships; and cohorts of officials, teachers, and community leaders. The curriculum's core was the diagnosis of §2 taught as catechism: the pre-colonial unity, the colonial fabrication, the genocide as the fabrication's harvest, the RPF as the nation's restoration, and the participant's own obligations within the unity project. [TBD-VERIFY: the programme's evolution, throughput numbers, and current form — sources diverge on whether ingando continues as a distinct programme or has been substantially folded into itorero since the late 2000s; NURC reporting and the academic literature (Mgbako 2005, Purdeková 2011–15) describe different phases.]

Ingando is also where the scholarly critique first attached to a specific instrument. Andrea Purdeková's fieldwork-based studies and Chi Mgbako's earlier account read the camps as ideological induction wrapped in reconciliation language — the single narrative taught without contest, the RPF's history merged with the nation's, dissenting questions structurally impossible in a camp setting, and the graduates enrolled as the project's local relays. Defenders, including many participants on the record, describe genuine reintegration effects, particularly for released prisoners re-entering communities of survivors. The two readings are not mutually exclusive, and their coexistence — an instrument that demonstrably processes reintegration and demonstrably indoctrinates — is the pattern every subsequent instrument repeats.

3.2 Itorero: the civic-values academy

Itorero ry'Igihugu, launched in November 2007 and given a statutory National Itorero Commission in 2013, is the project's permanent formation institution and its most explicit act of invented tradition. The pre-colonial itorero was the court school in which young men of the elite were trained in dance, rhetoric, war, and the values of ubupfura; the revived national itorero re-purposes the institution as a civic academy for the whole population — teachers, local officials, health workers, executives, diaspora youth, and school cohorts pass through training cycles in national values (ubumwe, agaciro, kwigira, gukunda igihugu — patriotism), the official history, and the government's programme priorities, graduating as intore, "the chosen," with an explicit mandate to model and transmit the values in their institutions. Throughput claims run to the hundreds of thousands per multi-year cycle [TBD-VERIFY: cumulative intore numbers — official statements have cited figures from several hundred thousand to over a million trained since 2007; independent verification is thin]. The fuller institutional treatment sits at RW-G-01 §5, and the RDF's adjacent role in domestic civic formation — the army as the regime's most trusted formation institution, running components of ingando/itorero and the urugerero national-service scheme attached to itorero from 2013 — at RW-I-01 §6.

Analytically, itorero marks the project's shift from processing (the ingando model: re-educate populations at status boundaries) to production (form the entire citizenry continuously, with a trained vanguard as multiplier). The vanguard design is the tell: intore is a hierarchical, selective identity within the egalitarian Banyarwanda identity, and the critique literature reads the structure as Leninist in form — a cadre system transmitting the party-state's line — beneath its traditionalist costume. The defence reads it as the only realistic transmission mechanism a poor state possesses: values do not self-propagate, and a society being asked to change its deepest categories needs carriers.

3.3 Ndi Umunyarwanda (2013): the testimonial politics and the apology controversy

The programme that names the project's mature phase emerged from the YouthConnekt dialogues of mid-2013, in which young Rwandans — many born in exile or after 1994 — discussed the genocide's inheritance with unusual frankness, including the question of what the children of perpetrators owe and carry. Out of these sessions, endorsed by Kagame at the national leaders' retreat and rolled out from late 2013, came Ndi Umunyarwanda ("I am Rwandan"): a national programme of facilitated public conversations and testimonials in which participants affirmed Rwandan-ness above ethnicity, survivors testified, and — the element that made the programme famous and contested — Hutu participants, including the then–prime minister and senior clergy, publicly apologised for the genocide committed "in the name of Hutu," asking forgiveness for crimes they had not personally committed [TBD-VERIFY: the programme's launch chronology, the specific senior-official apologies and their wording, and the official position on whether apology was expected of all Hutu — contemporary accounts (2013–14) attribute to the programme's defenders the position that apology was voluntary moral leadership, while opposition figures and some clergy reported social and professional pressure; the contested episodes are documented unevenly and largely in Kinyarwanda-language sources].

The apology dimension split even sympathetic observers, and the controversy deserves precise statement because it is the project's core paradox made operational. The programme's logic: reconciliation requires truth spoken in the categories in which the crime was committed; the genocide was committed by Hutu in the name of Hutu; therefore healing requires those who carry the name to disavow the crime publicly, voluntarily, generationally — after which the categories can finally rest. The critique's logic: a state that declares ethnicity abolished cannot coherently require (or "encourage," under conditions where declining carries risk) citizens to step forward as Hutu to assume a collective moral debt; doing so re-inscribes the categories it bans, converts them into a permanent perpetrator/victim moral hierarchy, and extends guilt to a generation born after the crime. PS Imberakuri and FDU-Inkingi figures who voiced versions of this critique faced divisionism exposure; the gospel singer Kizito Mihigo — a Tutsi survivor and commemoration laureate — released Igisobanuro cy'Urupfu ("The Meaning of Death") in March 2014, a song mourning all victims including those of RPF violence and gently questioning the official mourning's selectivity; he was arrested within weeks, convicted in 2015 of conspiracy charges, pardoned in 2018, re-arrested attempting to leave the country in February 2020, and found dead in his police cell days later — officially suicide, widely disbelieved (RW-N-01 §4 carries the case in the political-space inventory). Whatever the intent behind Ndi Umunyarwanda, the Mihigo arc fixed its meaning for the critique literature: the programme polices the boundary of permissible mourning, and the boundary runs along the ethnic line the state says does not exist.

3.4 The history curriculum and the single-narrative schooling

Schooling is the project's longest-horizon instrument, and its history is itself diagnostic. History teaching was effectively suspended after 1994 — the old curriculum was the ethnic republics' curriculum, and no agreed replacement existed — and for over a decade Rwandan schools taught no national history at all, a vacuum the literature treats as the project's most honest moment: the state knew the past was a weapon and preferred silence to contest. The rebuilt curriculum, developed through the 2000s with NURC, the National Curriculum Development Centre, and partners including the University of California–Berkeley Human Rights Center and the Aegis Trust's educational programmes, restored history teaching around the unity narrative and a "teaching the genocide" framework; the 2015/2016 competence-based curriculum embedded genocide studies, citizenship, and Ndi Umunyarwanda values across subjects. The full treatment — including the teacher-level evidence that classroom practice avoids the dangerous questions, the single-narrative critique, and the comparative pedagogy literature — is carried at RW-G-01 (especially §10–§11) and is not duplicated here. The instrument-level point for this document: the curriculum means the post-1994 cohorts encounter the official diagnosis (§2.1) as schooled fact before they encounter any alternative, and the alternatives remain criminally chargeable (§2.3). No other instrument so directly tests whether formed identity becomes felt identity.

3.5 The commemoration regime: Kwibuka as annual pedagogy

The final instrument is the calendar. Kwibuka ("to remember") — the commemoration cycle opening every 7 April with the national ceremony at the Kigali Genocide Memorial, the flame of remembrance, the week of national mourning, and the hundred days of commemoration to 4 July's Liberation Day — is the project's annual whole-population pedagogy, restating the diagnosis, mourning the victims, honouring the rescue, and renewing the never again compact, with the milestone years (Kwibuka20 in 2014, Kwibuka25 in 2019, Kwibuka30 in 2024) staged as major international events. The commemoration regime's external function — genocide memory as diplomatic asset, and the counter-reading that memory is instrumentalised to immunise the government against criticism — is carried at RW-N-01 §6; the memorial-and-mourning architecture's contested inclusions and exclusions at RW-J-01 §7 and §5. What belongs to this document is the domestic identity function: Kwibuka is where the project's two faces meet most visibly, the season in which the nation is simultaneously unified in mourning and re-divided by the mourning's categories — survivors mourn publicly, the descendants of perpetrators attend in scripted contrition, and the unmournable dead (§5.2) are remembered, if at all, in private. The commemoration is the project's solemn heart and the critique's annual exhibit, and both descriptions are accurate.

4. The Project's Achievements Ledger

4.1 The measured outcomes and their measurement problem

The project's flagship metric is the Rwanda Reconciliation Barometer, the NURC's periodic national survey of reconciliation attitudes, with editions in 2010, 2015, and 2020 [TBD-VERIFY: the Barometer's full edition list and whether a post-2020 edition has been conducted under MINUBUMWE]. The headline trajectory is of steadily high and rising indicators: the 2020 edition reported the overall reconciliation status at approximately 94.7 per cent, up from approximately 92.5 per cent in 2015 and 82.3 per cent in 2010, with component indicators — citizens' identification as Rwandan above ethnicity, trust between communities, perceived fairness of institutions — in the high eighties and nineties [TBD-VERIFY: the precise component figures and the instrument's composition across editions; the Barometer's published methodology and any independent replication]. The figures are taken seriously enough by donors to anchor programme evaluations, and dismissed quickly enough by critics to anchor the opposite literature.

The measurement problem is structural and must be stated plainly: survey research on regime-defined values, conducted by a state commission, in a country where the wrong answer to "do ethnic divisions persist?" is adjacent to a criminal offence, faces a preference-falsification problem that no questionnaire design fully escapes. The scholarship on political attitudes under dominant-party conditions — and the Rwanda-specific ethnography of Susan Thomson and others on rehearsed public speech (§5.1) — counsels treating the Barometer's levels as unreliable and its trends as weakly informative. Yet the symmetrical error should also be avoided: independent and donor-commissioned research (Afrobarometer rounds in Rwanda, academic survey work, the World Values Survey's Rwanda waves) broadly confirms high expressed national identification and high institutional trust, with the same caveats; and qualitative work that is sharply critical of the project nonetheless documents genuine attitudinal movement, particularly among the young. The honest summary: the measured outcomes are consistent with both deep internalisation and disciplined performance, and the survey instruments cannot distinguish the two. The behavioural record can — which is why §4.2 is the ledger's real content.

4.2 The behavioural floor: the absence, and what lived integration shows

The project's irreducible exhibit is an absence: no episode of ethnic mass violence on Rwandan soil since July 1994. The claim's force requires its baseline: Rwanda 1959–1994 experienced ethnic pogroms roughly once a decade (1959–64, 1973, the 1990–93 massacres, then the genocide); Burundi, the structural twin, has cycled through mass violence in 1972, 1988, 1993–2005, and the 2015 crisis; eastern DRC, hosting the genocide's exported demography, has burned continuously (RW-D-08). Against that regional and historical baseline, thirty-two years of internal communal peace — through the stress events of the prisoner releases into survivor communities (2003 onward), the gacaca decade's reopening of every hill's wounds (RW-A-02), the 2010 and 2024 election cycles, and the DRC wars' spillover pressures — is not a trivial accomplishment to be footnoted past. Critics correctly attribute part of the absence to the state's saturating security presence rather than to transformed hearts; but the consociational and power-sharing literatures would predict that pure suppression without any attitudinal movement produces periodic eruption, and Rwanda has had none.

Beneath the absence, lived integration is real and documentable. Survivors and perpetrators' families share hills, churches, cooperatives, and umuganda work-parties (RW-G-03); the reconciliation-village programmes, however curated for visitors, rest on actual cohabitation; intermarriage across the old categories — always present in Rwanda, one reason the 1994 killing was so intimate — continues, and the offspring of mixed unions are the categories' quiet dissolvers [TBD-VERIFY: no official intermarriage statistics exist precisely because the categories are unrecorded; the evidence is ethnographic — McLean Hilker's youth studies and the cohabitation literature — and cannot be quantified]. The unrecordability cuts both ways analytically: the state cannot prove integration with numbers, and critics cannot prove its absence.

4.3 The generational dimension: the post-genocide majority

Demography is the project's compounding asset. With a median age around twenty, roughly three-quarters of Rwandans alive in 2026 were born after 1994 [TBD-VERIFY: the 2022 census-based share of the population born post-1994 — figures cited range from 65 to 78 per cent depending on date and source]. This majority has never held an ethnic identity card, never been schooled in ethnic quotas, never experienced ethnicity as an official category — and has been formed, at every institutional encounter from primary school through itorero to Kwibuka, inside the Ndi Umunyarwanda framework. For this cohort the official identity is not a replacement identity but the default one; ethnicity arrives, when it arrives, as family transmission, commemoration-season ascription, or diaspora internet discourse, against the grain of everything official. The project's wager is that each cohort raised inside the framework holds it more cheaply — less as discipline, more as fact — until the categories' social reproduction fails. Whether the wager is being won is precisely what cannot yet be observed (§6.3); but the wager's structure is sound, and it distinguishes Rwanda's project from suppression regimes (the Yugoslav comparison, §7.1) that froze adult identities without building a formation machinery beneath them.

4.4 The comparative singularity

The achievements ledger closes with the classificatory fact: Rwanda is the only state attempting ethnicity's legal abolition, and the contrast inventory defines the experiment's stakes. Burundi is the consociational opposite: same categories, same colonial history, same demographic structure — and the 2000 Arusha Accord answered them by constitutionalising ethnicity, with 60:40 Hutu–Tutsi quotas in government and the assembly, 50:50 in the army and senate. Burundi recognises, counts, and balances; Rwanda deletes and criminalises. The two states are the controlled experiment the comparative literature could not have designed, and the verdict at 2026 is genuinely mixed: Burundi's quotas absorbed the 2015 crisis without ethnic civil war (the crisis ran intra-Hutu, which consociationalists count as success), while Rwanda's deletion has delivered superior stability and development at the price documented in §5. Mauritius (MU-M-01) is the recognition-and-management model: communal categories constitutionally acknowledged (the best-loser system), politically balanced, and culturally celebrated, within a competitive democracy — the multicultural route Rwanda's diagnosis forecloses, since for Kigali the recognition of categories is the recurrence mechanism. South Africa's rainbow-nation project kept racial categories fully visible — recorded, counted, and used for redress — betting that acknowledged difference plus truth-telling heals better than mandated unity; the TRC's truth-for-amnesty architecture is the standing alternative to Rwanda's unity-over-truth sequencing (§7.1). Rwanda's singularity is thus not merely descriptive but theoretical: every other post-conflict identity regime manages categories it admits exist. Rwanda's results — the absence, the formed generation, the unresolved silence — are the only empirical evidence the world possesses on what deletion does.

5. The Critiques

The corpus discipline requires the critique literature at the same strength as the achievements ledger, and the critique is not a single argument but three, of escalating depth: that the project suppresses rather than transcends; that its memory architecture is asymmetric; and that it functions as the RPF's legitimation machinery.

5.1 Suppression, not transcendence: the everyday-ethnicity ethnographies

The dominant scholarly reading rests on a body of fieldwork conducted mostly in the 2000s, under conditions of access that have since narrowed. Susanne Buckley-Zistel's southern-Rwanda fieldwork produced the literature's most cited concept, "chosen amnesia": communities deliberately not discussing the past's causes — pretending, in her informants' rendering, "as if nothing happened" — because coexistence on the same hills requires silence; the finding's force is that the silence is locally rational and officially convenient at once, and that it postpones rather than processes. Susan Thomson's work (Whispering Truth to Power, 2013) documented everyday resistance to the reconciliation apparatus — peasants performing the required unity scripts at ingando and gacaca while privately maintaining ethnic readings of their dispossession — and supplied the methodological warning that governs all Rwandan survey data: public speech in Rwanda is a performance with known required content. Andrea Purdeková's institutional ethnographies of ingando and the unity apparatus traced how "unity" operates as governmentality — a technology of surveillance and conduct-shaping reaching to the hill level. Lyndsay McLean Hilker's studies of Kigali youth found the categories fully legible to the post-genocide generation — transmitted in families, deployed in friendship and marriage decisions, readable through commemoration behaviour — beneath fluent public non-ethnicity. [TBD-VERIFY: the precise publication details and findings attributed here — Buckley-Zistel's 2006 Africa article "Remembering to Forget"; Thomson 2013; Purdeková's 2015 monograph Making Ubumwe; McLean Hilker's 2009–2014 articles — are cited from the secondary literature and should be checked against the original texts.]

The composite finding: the categories persist beneath the silence, transmitted privately, and the project has changed what may be said more demonstrably than what is felt. The finding's limits should be stated with equal care: the core fieldwork is now fifteen to twenty-five years old, conducted on cohorts formed before or during the genocide; the formed generation of §4.3 has not been ethnographically studied at comparable depth, partly because research access has narrowed; and "persistence beneath silence" is what one would expect mid-project even if the project were succeeding — identity change is generational by nature, and the ethnographies cannot distinguish a failing suppression from a slow transcendence in progress. The suppression reading is the best-evidenced reading; it is not yet a verdict.

5.2 The asymmetry critique: the unmournable dead and the unspeakable-but-operative category

The second critique concedes the unity project's sincerity and attacks its architecture. The official naming — the genocide against the Tutsi, constitutionalised in 2008 and adopted by the UN in 2018 — is historically accurate as a description of the genocide: the systematic extermination campaign targeted Tutsi as such, and the generic "Rwandan genocide" label had been exploited by deniers to dissolve the crime into mutual violence. But the naming sits inside an architecture in which it does asymmetric work. The Hutu dead of 1990–98 — the civil-war massacres, the killings documented in RPF-controlled zones in 1994 (the Gersony reporting episode), Kibeho 1995, and the Congo refugee deaths catalogued by the 2010 UN Mapping Report — have no public mourning category, no memorial inclusion, no commemoration season; the full contested accounting is carried at RW-J-01 §5 and is not re-litigated here. The identity consequence is what belongs to this document: a mourning regime in which one category's dead are nationally sacralised and the other's are privately grieved (or chargeable as "double genocide" revisionism if publicly equated) makes "Hutu" an unspeakable-but-operative category — officially non-existent, unusable for claims or mourning, yet fully operative as ascribed perpetrator lineage in Ndi Umunyarwanda's apology politics (§3.3), in gacaca's defendant demography, and in commemoration-season social ascription. The critique's sharpest formulation: the project has not abolished ethnicity; it has abolished one ethnicity's political and mnemonic existence while building the national narrative around the other's martyrdom. The defence's sharpest reply: equivalence is precisely what genocide forbids — the crimes are real but categorically different, a genocide is not a civil war, and the "double genocide" framing the asymmetry critique edges toward is the deniers' framework (the taboo is a firewall, not a silencing). Both arguments are serious; the corpus holds them open, noting only that the asymmetry critique is made from within the anti-denial consensus by scholars (Reyntjens, Longman, the Remaking Rwanda contributors) who fully affirm the genocide against the Tutsi.

5.3 The political-instrument reading: unity as the RPF's legitimation architecture

The third critique is the coldest: that whatever the project's reconciliation content, its load-bearing function is regime security. The argument's elements. First, the opposition-policing function: the divisionism and genocide-ideology statutes have been deployed against essentially every consequential challenger — Ingabire (2010–12), Bizimungu before her (RW-A-01 §9), PS Imberakuri's leadership, Mihigo, journalists and researchers — such that the laws' observable jurisprudence is more political than communal (RW-N-01 §4 carries the inventory). Second, the competition-foreclosing function: since any appeal to majority-community grievance is chargeable, and since the constitutional party rules ban organisation along the only cleavage on which a mass opposition could plausibly mobilise, the unity framework structurally guarantees that no legal opposition can ever assemble a majority coalition against the RPF — the project defines the terms of politics such that the incumbent cannot lose. Third, the legitimation function: the official syllogism — ethnic politics produced genocide; the RPF ended the genocide and guards against its return; therefore the RPF's rule is the unity project and its critics are the recurrence risk — converts identity policy into a permanent incumbency argument, renewed annually at Kwibuka (RW-N-01 §6) and underwriting the 99-per-cent election results (RW-D-05's territory). On this reading, the sincerity question is moot: a minority-rooted regime governing a society where the majority community was implicated in the genocide rationally requires the abolition of ethnic counting, and the project is that requirement made virtuous.

The reply, at full strength: the political-instrument reading explains the project's enforcement but not its content. A purely cynical regime would not need the formation machinery's depth — the curricula, itorero, the village-level reconciliation programmes — to stay in power; coercion alone would suffice, as it has for cruder regimes elsewhere. The project's defenders, including non-Rwandan scholars of the developmental-state school, argue the regime-security and nation-building readings are simply both true and mutually reinforcing — that in post-genocide conditions there was no available identity policy that did not serve the incumbent, and that judging the project by its convenience to the RPF evades the question of what alternative policy a responsible post-1994 government should have adopted. The Burundian alternative (recognition and quotas) was on the table at Arusha and was rejected by the RPF precisely because counting would have made it a permanent minority government; the critique and the defence agree on that fact and diverge on its moral reading. The corpus's position is the discipline stated in §1: both ledgers are real, the same instruments appear in both, and the project cannot be honestly described with either ledger alone.

6. The Project Under Stress

6.1 The regional contradiction: abolished at home, instrumentalised abroad

The project's most visible stress point is geographic: the categories deleted from Rwanda's domestic vocabulary are the working language of its foreign policy. The pattern runs the full length of the post-1994 period. The 1996–97 and 1998–2003 Congo wars were justified by the génocidaire threat in the camps — the ex-FAR/Interahamwe formations that became the FDLR, whose continued existence in eastern DRC remains Kigali's standing casus belli (RW-F-01). The 2012–13 and 2021–26 M23 episodes were framed, in Rwandan official communication, substantially in ethnic-protection terms: the persecution of Congolese Tutsi and Banyamulenge, the hate speech and pogrom dynamics in the Kivus read explicitly as 1994's prologue replayed, the FDLR's integration into the FARDC's ground coalition presented as a genocidal alliance on the border (RW-D-08 §5, §12 carries the strategic-logic assessment, including the contending mineral-stake and security-buffer readings). The official discourse thus performs, abroad, exactly the operations it criminalises at home: counting by category, mobilising by category, claiming protection duties by category.

The tension admits two resolutions, and they map onto the document's two ledgers. The coherence defence: there is no contradiction in abolishing ethnicity as a domestic political category while describing external threats accurately — the FDLR is a genocidal organisation by inheritance and ideology, the Kivu anti-Tutsi violence is real and documented by UN reporting, and a state built on "never again" would be derelict to unsee ethnicity across a border because it has transcended it within one. The instrumentalisation critique: the selective fluency is the tell — the state's strategic cognition runs fully on the categories (who is kin, who is threat, who must be protected), which means the abolition is not a cognitive transformation but a domestic speech regime; and the regional deployments have repeatedly re-ethnicised the neighbourhood (the Kivus' Rwandophone question, the anti-Tutsi backlash each intervention feeds), exporting the dynamics the project claims to be extinguishing. The 2025 Washington and Doha frameworks' neutralisation-of-the-FDLR provisions (RW-D-08 §11) will test whether the external ethnic-threat framing can be retired; the project's domestic credibility is partly hostage to that question, since a permanent external ethnic enemy keeps the categories permanently warm.

6.2 The diaspora dimension: the discourse war the project cannot police

The project's writ ends at the border, and beyond it the abolished categories flourish. The exile opposition is substantially organised along, and speaks fluently in, the old categories: the FDU-Inkingi lineage and the Hutu-diaspora media ecosystem carry majority-grievance framings (the 1994–98 RPF killings, the Congo dead, the "victor's justice" reading of the post-genocide order); the RNC of the 2010 RPF defectors (Kayumba Nyamwasa, the late Patrick Karegeya) brought insider Tutsi voices into the anti-Kagame coalition and complicated the regime's equation of opposition with génocidaire revanchism; and the diaspora internet — YouTube channels, francophone forums, the commentary ecosystem around works like the 2014 BBC documentary Rwanda's Untold Story (banned in Rwanda, prosecuted rhetorically as denial) — conducts the double-genocide discourse war in exactly the vocabulary domestic law forbids (the full historiographical contest at RW-J-01 §5). The state's response has been extraterritorial: transnational repression documented by Freedom House and others, the Karegeya assassination in Johannesburg (2014), pressure on host governments, and the digital-era contest for the diaspora generation's narrative — Ndi Umunyarwanda programming for diaspora youth, Itorero abroad, commemoration diplomacy. The stress is structural: the post-memory generation inside Rwanda (§4.3) is one search query away from the counter-narrative at all times, and the project's information monopoly, complete in the schoolroom, is zero online. Whether formed identity survives contact with uncensored contestation is being tested continuously in the diaspora — and the diaspora's persistent ethnic polarisation is, on the critique reading, the project's shadow control group: Rwandans outside the enforcement perimeter have not converged on Banyarwanda.

6.3 The succession exposure: whether the project survives its enforcer

The deepest stress is temporal. The project has had one political guarantor for its entire existence: every instrument in §3 was built under Kagame's authority, the enforcement architecture answers to the presidency, and the official syllogism of §5.3 binds the unity project to the RPF's — in practice, to Kagame's — continuation. The 2015 constitutional settlement runs his eligibility to 2034 (RW-K-01); the succession question is the corpus's designated master variable (RW-O-01 §2), and its identity dimension is this document's closing problem. The scenarios bracket cleanly. If the project has succeeded — if the §4.3 generation holds Ndi Umunyarwanda as fact rather than discipline — then succession, even disorderly succession, finds no ethnic tinder: the categories' social reproduction has failed, and the post-Kagame contest runs on other cleavages (regional, generational, economic). If the suppression reading is right, succession is the decompression event: the silence ends when the enforcer goes, the suppressed grievances (the unmourned dead, the apology politics' resentments, the majority's political exclusion) surface in the first open contest, and the precedents are grim — Yugoslavia's "brotherhood and unity" survived Tito by barely a decade; the Soviet druzhba narodov dissolved into ethnic successor politics within years of enforcement weakening. The memory-regime question at forty (RW-O-01 §5) compounds the exposure: the commemoration architecture, the criminal-law firewall, and the asymmetric mourning settlement were all calibrated for a survivor generation that will be passing from the scene through the 2030s, and recalibration — any relaxation of the speech regime, any widening of the mourning categories — risks being read as the settlement's unravelling by survivors and as too little by the critique. The honest statement: the project's terminal evaluation is unavailable before the succession, because its central question — internalised or imposed? — is only answerable when imposition lapses. Rwanda's post-ethnic citizenship is, at 2026, an experiment whose decisive observation has not yet occurred.

7. Comparative Perspective

7.1 The nation-building-after-atrocity genre: the unity-versus-truth trade-off

In the transitional-justice literature Rwanda anchors one pole of the field's central trade-off. The canonical post-atrocity memory regimes array as follows. Germany is the truth pole: perpetrator-centred memory, decades of contested but ultimately institutionalised acknowledgment (Vergangenheitsbewältigung), the crime taught as the nation's own act — a regime built on the perpetrator state's continuity and defeat, conditions Rwanda does not share, since in Rwanda perpetrators and survivors must share hills, not merely textbooks. Spain chose the pacto del olvido — amnesty and silence as the transition's price — and the model's apparent success unravelled generationally: the historical-memory mobilisations from 2000 onward, the 2007 and 2022 memory laws, and the exhumation politics demonstrated that suppressed mourning returns on a two-generation delay, the precedent the suppression reading of Rwanda cites most. Yugoslavia is the cautionary tale stated in §6.3: an enforced supra-ethnic identity ("brotherhood and unity"), a criminalised nationalist vocabulary, suppressed wartime grievances (Bleiburg, Jasenovac's contested tolls) — and, when the enforcing structure decayed, the suppressed categories returned as mobilisation-ready identities with their grievances compounded by the silence. South Africa's TRC inverted Rwanda's sequencing — truth first, even at amnesty's price, on the theory that acknowledged truth is reconciliation's precondition — and its mixed legacy (racial cleavage politically alive, but no recurrence and a durable constitutional settlement) keeps the alternative credible. Rwanda's distinctiveness within the genre: it chose unity over open truth-telling (the gacaca process produced massive fact-finding about the genocide while foreclosing the symmetric accounting — RW-A-02; RW-J-01 §5), enforced the choice with criminal law, and built a formation machinery to make the choice generational. The genre's accumulated finding — that suppressed memory returns — is the strongest comparative argument against the project; the project's reply is that none of the genre's cases combined suppression with formation at Rwanda's depth, and that the analogy to Yugoslavia ignores the machinery of §3.

7.2 The engineered-identity literature: high-modernist nation-building and its echoes

The second literature reads Rwanda not through atrocity but through statecraft: the high-modernist tradition of states that set out to manufacture their citizens. The genre's classics — revolutionary France's manufacture of Frenchmen from provincial diversity, Atatürk's republican Turk, the Soviet new man — established the pattern Rwanda repeats: a founding rupture, a discredited old identity, a vanguard with a diagnosis, control of schooling and language, invented tradition, and a formation calendar. The African post-colonial cohort supplies the nearer echoes. Tanzania under Nyerere is the genre's African success: Swahili as unifying language, ujamaa's civic nationalism, the deliberate demotion of tribe from politics — and Tanzania's durable immunity to ethnic politics is the existence proof that African states can build supra-ethnic identity. But the Tanzanian contrast defines Rwanda's radicalism: Nyerere built over acknowledged ethnicities (some 120, none demographically dominant, none recently weaponised in mass killing), demoting rather than deleting them; Rwanda attempts deletion of exactly two categories, recently genocidal, in a society where everyone knows which one they carry. Singapore — Rwanda's chosen developmental referent (RW-C-02; RW-O-01 §7) — is the instructive opposite on identity despite the governance kinship: Singapore institutionalised its categories (the CMIO framework, recorded race, managed proportions in housing and politics) and built the national identity as a roof over them, betting on management where Rwanda bets on erasure; the two states share the formation instinct (national service, civic curricula, commemoration pedagogy) while standing at opposite poles on recognition. The high-modernist literature's standard finding (Scott's Seeing Like a State lineage) is that engineered identity succeeds where the state's scheme aligns with lived social process and fails where it overrides it — which restates, in the genre's vocabulary, exactly the open question of §5.1: whether the post-1994 generation's lived process is converging with the scheme or merely performing it.

7.3 What the Rwandan case tests

The comparative synthesis can be stated as a single proposition under test. Rwanda's founding diagnosis holds that the lethal form of Hutu and Tutsi was made by states — the colonial census, the quota republics, the génocidaire radio — and the project wagers that what states made, a state can unmake, given monopoly over formation and two generations of time. The comparative literatures supply the priors on both sides: the constructivist scholarship on ethnicity (the consensus that the categories are historically constructed and politically activated, not primordial) says the wager is theoretically sound; the memory-regime record (Spain, Yugoslavia) says suppression-led versions of the wager fail on a generational delay; the formation-state record (Tanzania, Singapore, the high-modernist canon) says deep formation can durably reshape identity when it does not require the denial of lived experience. Rwanda combines the heaviest suppression and the deepest formation in the modern record, applied to the hardest case — categories sanctified by genocide within living memory. No existing case predicts the outcome, which is why the project, whatever its verdict, will be load-bearing for social theory: it is the cleanest test ever run of whether ethnicity is, as the constructivists hold, unmade the way it was made — by the patient operation of state power on the young — or whether the categories, once written in blood, are beyond administrative recall.

8. Conclusion

Ndi Umunyarwanda is the idea on which everything else in the Rwandan corpus rests. The developmental state (RW-C-02), the electoral architecture (RW-D-05's territory), the commemoration diplomacy (RW-N-01), the DRC posture (RW-D-08), and the succession question (RW-O-01) all presuppose the identity settlement this document has anatomised: a state that diagnosed ethnicity as the administratively fabricated cause of its founding catastrophe, deleted it from every record, criminalised its public invocation, and built the most elaborate citizenship-formation machinery in Africa to fill the cleared ground with Banyarwanda.

Three findings organise the assessment. First, the project is real — not a slogan but a four-decade institutional programme with a premise, instruments, a budget, a bureaucracy, and a formed generation; analyses that treat it as mere regime propaganda miss the machinery's depth, just as analyses that take its survey metrics at face value miss the speech regime those metrics are collected under. Second, the project's two ledgers are produced by the same instruments and cannot be netted against each other: the laws that bank the thirty-two-year absence of ethnic violence are the laws that imprisoned Ingabire and Mihigo; the commemoration that binds the nation in mourning is the commemoration that sorts the mournable from the unmournable; the programme that asks Rwandans to transcend their categories is the programme that required some to apologise within them. Every honest sentence about the project must hold both clauses. Third, the project's verdict is structurally unavailable before its first transfer of power: the question that matters — whether the post-1994 generation carries Ndi Umunyarwanda as identity or as discipline — can only be observed when the enforcement that makes the two indistinguishable lapses, which places the experiment's decisive observation somewhere on the far side of the 2034 horizon (RW-K-01; RW-O-01 §2).

What can be said at 2026 is the comparative finding: among all the ways states have answered catastrophic ethnic division — Burundi's quotas, Mauritius's recognition (MU-M-01), South Africa's truth-telling, Spain's forgetting, Yugoslavia's enforced brotherhood — Rwanda alone chose deletion, and Rwanda alone is therefore generating evidence on whether the constructivist premise of modern social science survives its hardest application. If the project holds through succession, it will stand as the demonstration that even genocide-sanctified categories are unmade the way they were made. If it decompresses, it will stand as the demonstration that suppression-plus-formation, at any depth, cannot substitute for the symmetric truth-telling the project's architecture forecloses. Either outcome will be among the most consequential findings in the modern politics of identity — which is the measure of what Rwanda, for better and for worse, has undertaken.

Sources

National Unity and Reconciliation Commission (NURC), Rwanda Reconciliation Barometer (editions 2010, 2015, 2020), Kigali.
Republic of Rwanda, Constitution of the Republic of Rwanda (2003, as amended 2008 and 2015), including the preamble's unity and anti-genocide-ideology commitments.
Law No. 47/2001 on Prevention, Suppression and Punishment of the Crime of Discrimination and Sectarianism; Law No. 18/2008 Relating to the Punishment of the Crime of Genocide Ideology; Law No. 84/2013 on the Crime of Genocide Ideology and Related Crimes.
Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton University Press, 2001).
Catharine Newbury, The Cohesion of Oppression: Clientship and Ethnicity in Rwanda, 1860–1960 (Columbia University Press, 1988).
Jan Vansina, Antecedents to Modern Rwanda: The Nyiginya Kingdom (University of Wisconsin Press, 2004).
Johan Pottier, Re-Imagining Rwanda: Conflict, Survival and Disinformation in the Late Twentieth Century (Cambridge University Press, 2002).
Susanne Buckley-Zistel, "Remembering to Forget: Chosen Amnesia as a Strategy for Local Coexistence in Post-Genocide Rwanda," Africa 76:2 (2006).
Susan Thomson, Whispering Truth to Power: Everyday Resistance to Reconciliation in Postgenocide Rwanda (University of Wisconsin Press, 2013).
Andrea Purdeková, Making Ubumwe: Power, State and Camps in Rwanda's Unity-Building Project (Berghahn, 2015); and "'Even If I Am Not Here, There Are So Many Eyes': Surveillance and State Reach in Rwanda," Journal of Modern African Studies 49:3 (2011).
Lyndsay McLean Hilker, "Everyday Ethnicities: Identity and Reconciliation Among Rwandan Youth," Journal of Genocide Research 11:1 (2009).
Scott Straus and Lars Waldorf (eds.), Remaking Rwanda: State Building and Human Rights after Mass Violence (University of Wisconsin Press, 2011).
Filip Reyntjens, Political Governance in Post-Genocide Rwanda (Cambridge University Press, 2013).
Timothy Longman, Memory and Justice in Post-Genocide Rwanda (Cambridge University Press, 2017).
Nigel Eltringham, Accounting for Horror: Post-Genocide Debates in Rwanda (Pluto Press, 2004).
Chi Mgbako, "Ingando Solidarity Camps: Reconciliation and Political Indoctrination in Post-Genocide Rwanda," Harvard Human Rights Journal 18 (2005).
Molly Sundberg, Training for Model Citizenship: An Ethnography of Civic Education and State-Making in Rwanda (Palgrave Macmillan, 2016) — on itorero and the intore.
Phil Clark, The Gacaca Courts, Post-Genocide Justice and Reconciliation in Rwanda (Cambridge University Press, 2010).
Amnesty International, Safer to Stay Silent: The Chilling Effect of Rwanda's Laws on "Genocide Ideology" and "Sectarianism" (2010); Human Rights Watch reporting on the genocide-ideology prosecutions (2008–2024).
Republic of Rwanda / Ndi Umunyarwanda programme documentation and contemporary press coverage of the 2013 launch and the apology debate (The New Times, Kigali; BBC Great Lakes; Jeune Afrique, 2013–2014).
UN General Assembly Decision 72/550 (26 January 2018, based on draft resolution A/72/L.31) [Tier-1 audit 2026-08-29: corrected from "Resolution A/RES/72/550"] designating the International Day of Reflection on the 1994 Genocide against the Tutsi in Rwanda.
James C. Scott, Seeing Like a State (Yale University Press, 1998) — comparative high-modernist frame; and the Tanzania/Singapore nation-building literatures for §7.2.

RW-A-01: Genocide Aftermath (1994–2003) — the Government of National Unity, the 2003 Constitution, and the consolidation within which the project was founded
RW-A-02: Gacaca Courts (2002–2012) — the justice architecture's reconciliation and truth-finding functions
RW-B-01: The 1994 Genocide Against the Tutsi — the catastrophe the project answers; the identity card at the roadblocks
RW-D-08: Rwanda–DRC M23 Conflict and the Doha/Washington Tracks (2021–2026) — the regional deployment of the ethnic vocabulary (§6.1)
RW-G-01: Rwandan Education and Genocide Memory — the curriculum, ingando, itorero, and memory-pedagogy machinery in full
RW-I-01: Rwanda Defence Force — the army's domestic civic-formation functions (§6 there)
RW-J-01: Three Accounts — 1994 Genocide, Prevention Failure, and International Response — the colonial/structural account (§4 there) and the RPF-accountability contest (§5 there)
RW-K-01: 2015 Constitutional Referendum and the Third-Term Decision — the enforcement horizon to 2034
RW-N-01: Rwanda in International Perceptions — the authoritarian-critique frame (§4 there) and the instrumentalised-memory frame (§6 there)
RW-O-01: Rwanda Megatrends — the succession question (§2 there) and the memory-regime question (§5 there)
MU-M-01: The Mauritian Multicultural Model — the recognition-model contrast (§4.4)