The Killing of Karine Buisset in Goma: Rwanda's Occupation, a Drone Strike, and the Long Pattern of Targeted Violence
In the early hours of Wednesday, 11 March 2026, a drone struck a two-storey residential building in the Himbi neighbourhood of Goma, a city held by Rwanda-backed RDF/M23 rebels since January 2025. Karine Buisset, a 54-year-old French national from Belz in Morbihan and a UNICEF child protection officer, was sleeping in the apartment of Christine Guinot, UNICEF's head of security in the DRC, who was not present that night. Buisset died at the scene. Two other people were also killed. By 4:12 a.m., a second wave of strikes had hit the city.
RDF/M23 spokesperson Lawrence Kanyuka attributed the drone attack to the Armed Forces of the Democratic Republic of the Congo (FARDC), describing it as a "combat drone" strike and a "terrorist attack" on civilian areas. France's President Emmanuel Macron confirmed Buisset's death on social media and called for respect for humanitarian law. UNICEF Executive Director Catherine Russell stated unequivocally: "Civilians, including aid workers, must never be targeted." On 13 March 2026, France's Parquet National Antiterroriste opened a formal investigation for murder constituting a war crime.
This article does not contest the question of which side fired the specific drone that killed Karine Buisset. That is a matter for the PNAT investigation and for independent forensic inquiry. What this article examines is the broader and equally urgent question: who created the conditions in which a UNICEF worker could be killed in a pre-dawn drone strike in a residential neighbourhood of a Congolese city? The answer to that question runs through Kigali, through three decades of documented targeted violence, and through an international community that has consistently refused to impose the accountability that might have interrupted the pattern.
Part One: Goma Under RDF/M23 — A City Surrendered to Armed Impunity
Who Was Karine Buisset?
Karine Buisset had extended her mission in Goma at the last moment. She was working on child protection programmes in eastern DRC, one of the most dangerous environments in the world for children: a region where RDF/M23 and other armed groups have recruited, abducted, and deployed child soldiers, where sexual violence against children remains endemic, and where the civilian protection architecture has been progressively dismantled by years of armed conflict. The work of a UNICEF child protection officer in North Kivu means navigating bureaucracies, negotiating access with armed factions, and documenting violations for accountability processes that rarely materialise. Buisset chose to stay when staying was dangerous. She was not near any frontline. She was in a residential building, asleep, in a neighbourhood that draws humanitarian workers for its relative quiet.
UNICEF described her as "a dedicated humanitarian who worked tirelessly to support children and families affected by conflict and crisis." Her colleagues described someone who was passionate about child protection and committed to the communities she served. She was 54 years old. She came from Brittany. She was killed before dawn in a city that an internationally-backed armed group has occupied for over a year.
The Drone Strike and the Question of Attribution
The attribution of the drone strike requires careful handling. RDF/M23, operating through its AFC (Alliance Fleuve Congo) political structure, publicly blamed FARDC for the attack. AFC/M23 Vice President Bertrand Bisimwa stated that the drones struck residential areas far from active frontlines. The incident occurred a day after the DRC military announced it had shot down two drones belonging to Rwandan forces in South Kivu, near the Minembwe area, describing the incursion as a violation of Congolese airspace and of the Washington Accord. That context matters. The airspace over Goma and the surrounding region is militarised by multiple actors. Drone capability in this conflict is not the exclusive possession of any single party.
However, attribution of the specific strike does not exhaust the question of responsibility. Under international humanitarian law, the party that controls a civilian population bears primary responsibility for that population's protection. RDF/M23 has controlled Goma since January 2025. It has governed the city by force for more than a year. If FARDC drones are striking civilian areas of an RDF/M23-held city, the failure of civilian protection is a consequence of the military situation RDF/M23 created with Rwandan support. The war that placed Karine Buisset in danger is Rwanda's war, conducted through a proxy force whose supply chain, personnel, and command structure run directly to Kigali.
Rwanda's Role: Beyond Denial
Rwanda has denied direct involvement in RDF/M23 at every diplomatic forum. Those denials do not survive contact with the evidence. The UN Group of Experts on the DRC, in reports spanning more than a decade, has documented Rwandan Defence Force troop deployments inside DRC territory, the supply of weapons and ammunition to RDF/M23, and the command relationship between Kigali and the group's military leadership. Human Rights Watch and Amnesty International have corroborated these findings independently. The United Nations and Western nations publicly accuse Rwanda of backing RDF/M23. Kigali continues to deny it.
Rwanda's standing justification is the FDLR — the Democratic Forces for the Liberation of Rwanda — whose presence in eastern DRC it presents as an existential security threat. That framing functions as a blanket authorisation for military operations in DRC regardless of their actual objectives. The FDLR itself has a history that demands nuance: it was formed in part by Hutu refugees who fled into the DRC facing violence in the aftermath of the RPF's seizure of power and who feared — with documented justification — that return to Rwanda meant death. Rwanda's security narrative, however, does not distinguish between armed combatants and civilians. It never has.
Part Two: The Historical Signature — RPF Targeted Killings and What They Reveal
A Doctrine of Targeted Violence: The RPF Before 1994
Understanding the operational culture of the force that commands RDF/M23 requires going back to the RPF's campaign of targeted violence in Rwanda from 1990 onwards. When the RPF invaded from Uganda in October 1990, it did not only seek military control of territory. It pursued a systematic strategy of eliminating individuals perceived as obstacles to its political objectives. Filip Reyntjens, one of the foremost academic authorities on Rwanda, has documented this pattern in detail. The RPF used infiltration networks and political assassinations to remove moderate Hutu politicians who might have provided a credible political alternative — individuals who represented the possibility of a negotiated peace that would have constrained RPF dominance.
These were not battlefield deaths. They were deliberate eliminations of political competition. The Arusha peace process of 1993, which was supposed to produce a power-sharing agreement, was repeatedly destabilised by violence that served RPF strategic interests by foreclosing compromise. International observers at the time noted the systematic quality of the killings. The pattern was not incidental to the RPF's political project. It was constitutive of it.
The Assassination of President Habyarimana: The Shot That Started the Genocide
On 6 April 1994, the aircraft carrying President Juvénal Habyarimana and President Cyprien Ntaryamira of Burundi was shot down over Kigali. All on board were killed. Within hours, the genocide against the Tutsi began. For decades, the dominant narrative attributed the attack to Hutu extremists seeking a pretext for mass killing. That narrative has been subjected to serious forensic and judicial challenge.
French investigating magistrate Jean-Louis Bruguière, whose investigation spanned nearly a decade and drew on witness testimony, ballistic evidence, and intelligence analysis, concluded in 2006 that responsibility for the attack lay with the RPF, acting under orders from Paul Kagame. Michael Hourigan, an investigator working for the International Criminal Tribunal for Rwanda in the late 1990s, submitted an affidavit stating that he had obtained credible evidence pointing to RPF responsibility and that his investigation had been shut down by ICTR Chief Prosecutor Louise Arbour under political pressure. Prosecutor Carla Del Ponte, who succeeded Arbour, confirmed in her memoir that attempts to investigate RPF crimes were blocked at the political level.
The strategic logic of the Habyarimana assassination is not difficult to reconstruct. The Arusha Accords required the RPF to share power with existing Rwandan institutions, substantially reducing RPF dominance. Habyarimana's death triggered the genocide. The genocide provided the moral and political justification for the RPF's military seizure of Rwanda. The RPF assumed full power not as a party to a negotiated settlement but as a conquering force that had ended a catastrophe — a catastrophe whose origins remain, at minimum, a subject of serious scholarly and judicial dispute.
The UN Mapping Report: Crimes Without Accountability
The 2010 UN Mapping Report on the DRC documented in forensic detail the mass killings carried out by RPF-linked forces across the DRC between 1993 and 2003. It documented the systematic targeting of Rwandan Hutu refugees — men, women, and children — who had fled into the DRC following the RPF's seizure of power. The report noted that the scale, method, and targeting of the killings bore the hallmarks of acts committed with intent to destroy a group in whole or in part. Rwanda threatened to withdraw its peacekeeping forces from Darfur if the report was published in full. The report was published in redacted form. No prosecutions followed. Many of the commanders implicated went on to occupy senior positions in the Rwandan Defence Force — the same force that commands RDF/M23 in eastern DRC today.
Kill, Then Blame: The RPF's Documented Pattern of False Attribution
RDF/M23's attribution of the Goma drone strike to FARDC is not simply a denial. It is a manoeuvre with a precise historical precedent. During the RPF's wars of the 1990s — both the 1990–1994 campaign inside Rwanda and the subsequent operations across the DRC — a consistent operational pattern emerged: the RPF would carry out, or facilitate, an act of violence against civilians or officials, and then attribute that act to the Rwandan government, to Hutu extremists, or to another convenient party. The attribution served a dual function. It deflected accountability from the RPF while simultaneously providing political justification for further RPF military action.
This pattern is not conjecture. It is documented across multiple credible sources. Filip Reyntjens has catalogued incidents in which killings of Hutu officials and moderate political figures — individuals whose deaths served clear RPF political objectives — were publicly attributed to Hutu extremist militias or to the Habyarimana government, when available evidence pointed toward RPF infiltration networks. The logic was consistent: the RPF had both the motive and the means; the attribution to others served the RPF's narrative of itself as a liberation force confronting a genocidal enemy rather than a political actor pursuing power through strategic violence.
The assassination of President Habyarimana on 6 April 1994 is the most consequential application of this pattern. The killing of a head of state was attributed immediately and comprehensively to Hutu extremists by RPF-aligned sources, a narrative that became politically dominant within days and has been reinforced by the post-genocide Rwandan government's control over historical memory ever since. Yet the forensic and judicial evidence assembled by Bruguière, the Hourigan affidavit, and the Spanish judicial investigation points in a different direction. The assassination bore the hallmarks of the same operational logic: kill an obstacle, attribute the killing to the enemy, exploit the consequences.
The same template operated in the DRC. During the First and Second Congo Wars, RPF-linked forces and their Congolese allies carried out mass killings of Hutu refugees that were systematically attributed in official Rwandan communications to inter-ethnic violence, crossfire, or FDLR operations. The UN Mapping Report dismantled that narrative with forensic specificity, documenting organised, targeted operations against civilian populations that bore no resemblance to accidental casualties of combat. The perpetrators and the propagandists were the same institutional entity.
The Goma drone strike and RDF/M23's immediate attribution to FARDC must be read within this documented pattern. RDF/M23 controls Goma. It has controlled Goma for over a year. It has an established institutional interest in presenting itself as the protector of a city under attack from a reckless Congolese government, rather than as an occupying force whose presence has made civilians the collateral of a war of Rwanda's choosing. Whether or not FARDC operated the specific drone that killed Karine Buisset — a question the PNAT investigation must resolve — the attribution by RDF/M23 is not a neutral factual claim. It is the latest iteration of a false-flag attribution strategy that the RPF and its successor institutions have deployed consistently for thirty years. Attribution by the perpetrator of a pattern should never be accepted at face value. It should be treated as evidence of that pattern.
There is one detail that strips RDF/M23's attribution of any remaining credibility as a genuine factual claim: RDF/M23 did not call for an independent investigation. A party that has just suffered the killing of a UNICEF worker in a city it controls, and that sincerely believes another military is responsible, has every incentive to demand independent forensic inquiry — to invite investigators, to preserve evidence, to engage the United Nations and the international humanitarian community in establishing accountability. RDF/M23 did none of these things. It issued a statement blaming FARDC and moved on. This is not the behaviour of a party seeking truth. It is the behaviour of a party executing a communications strategy.
Compare this with the RPF's conduct after the assassination of President Habyarimana in 1994. The RPF did not call for an independent investigation into the plane attack. It attributed responsibility to Hutu extremists within hours, consolidated that narrative through its control of post-genocide Rwandan institutions, and obstructed every subsequent attempt at forensic inquiry — including the Bruguière investigation and the Hourigan ICTR inquiry. The institutional reflex is identical across thirty years: attribute immediately, deny loudly, obstruct investigation. RDF/M23's response to the Buisset killing follows this template with precision. The absence of any call for independent inquiry is not a neutral omission. It is evidence of the pattern.
This point carries direct analytical weight for the PNAT investigation and for international diplomatic responses. If the investigation focuses exclusively on the technical question of which military operated the drone, without situating that question in the broader pattern of RPF/RDF false attribution, it risks being manipulated into providing legitimacy for RDF/M23's narrative. Forensic attribution must be accompanied by contextual analysis. The killing of Karine Buisset did not occur in a vacuum. It occurred in a city under armed occupation, in a war initiated and sustained by Rwanda, conducted by a force with a documented thirty-year history of killing and then blaming others.
Scene Contamination: RDF/M23 Inside the House Before Any Independent Body
What happened in the hours immediately following the strike on 11 March 2026 compounds the evidentiary crisis at the heart of this case. Reports confirmed that RDF/M23 officials, alongside firefighters and UN staff, were present at the scene on Wednesday morning — inside the building where Karine Buisset was killed. RDF/M23, the armed group that controls Goma by force, had access to the physical site of the killing before any independent investigative body had secured it.
The Forensic Question: Drone Cover or Cause of Death?
There is a forensic question at the centre of this case that has not received the attention it demands: the physical evidence at the scene where Karine Buisset was killed is not consistent exclusively with a drone strike. The building shows signs of shooting. This is not a minor evidentiary detail. It raises a question that the PNAT investigation must answer before any other: was Karine Buisset killed by the drone, or was she killed by gunfire, with the drone deployed subsequently as cover?
These are not equivalent scenarios with the same implications. If Buisset was killed by gunfire and a drone strike was then used to mask that method of killing, the entire framing of this case as a drone attribution dispute is a misdirection — one that serves RDF/M23's interests precisely because it focuses international attention and legal inquiry on the question of which military operated a drone, rather than on who shot her and why. A drone strike, even one attributed to FARDC, carries the narrative of collateral damage in an active conflict: a tragic but explicable consequence of warfare between state armies. A targeted shooting, masked by a subsequent drone strike, is something categorically different. It is an execution.
The RPF's documented operational history is directly relevant here. The RPF did not only kill through military operations. It killed through infiltration, through targeted assassination, through the placement of operatives in close proximity to individuals it sought to eliminate. Filip Reyntjens and Gérard Prunier have both documented the use of small arms and close-range killings against Rwandan political figures in the early 1990s — killings that were then attributed to Hutu extremists, to random violence, or to the chaos of a deteriorating security environment. The method was to kill by a means that could be explained away, then construct an alternative narrative around the killing. A drone strike provides exactly such a narrative. It is visible, explosive, attributable to an external actor, and creates sufficient physical disruption to complicate forensic analysis of what happened inside the building before the strike.
This possibility — that shooting preceded the drone, and that the drone was deployed to cover the shooting — must be treated by the PNAT as a primary investigative hypothesis, not a secondary one. The physical evidence requires it. Bullet trajectories, entry and exit points, ballistic patterns, and the relationship between the structural damage caused by the drone and the evidence of shooting must all be forensically mapped and sequenced. The critical question is not only what caused the structural damage to the building but what caused the death of Karine Buisset — and whether those two things were the same event or two separate events staged to appear as one.
Why Scene Integrity Was Critical and Why It Has Been Destroyed
This forensic complexity makes RDF/M23's immediate presence inside the building not merely procedurally improper but potentially determinative of whether justice is possible in this case at all. If Buisset was killed by gunfire before the drone strike, the physical evidence of that shooting — bullet casings, blood distribution patterns, ballistic traces on interior surfaces — is precisely the evidence that RDF/M23 officials, moving through the building before independent investigators arrived, had the opportunity to remove. A scene that showed signs of shooting before it showed signs of a drone strike is a scene that RDF/M23 had every institutional interest in reordering.
In any criminal investigation, the integrity of the crime scene is foundational. The preservation of physical evidence determines whether forensic attribution is possible at all. A crime scene to which interested parties have had unrestricted access before independent investigators arrive is a scene that has been compromised. Whether or not evidence was deliberately removed or destroyed, the conditions for its removal or destruction existed from the moment RDF/M23 officials entered that building.
The correct procedure, under international humanitarian law and standard investigation protocols, was clear. RDF/M23, as the occupying authority in Goma, had an obligation to secure the perimeter and ensure that no party with a direct interest in the outcome of any attribution inquiry was permitted access to the physical evidence before an independent body had taken control of the scene. The United Nations Organisation Stabilisation Mission in the Democratic Republic of the Congo — MONUSCO — or another independent international body, should have been called immediately and granted exclusive access to document the site. RDF/M23 did not do this. Its own officials were among the first persons inside the building.
The PNAT must formally request that RDF/M23's presence at the scene, the timeline of access, and the identity of all individuals who entered the building before independent investigators arrived be treated as a core element of the investigation. It must also formally establish, through whatever physical evidence remains, whether the signs of shooting predate or postdate the structural damage from the drone — and whether Buisset's death is consistent with drone blast injuries, with gunshot wounds, or with both. That sequencing is the forensic key to this case.
Three further details compound the evidentiary picture and each demands specific investigative attention. First, two other people were killed in the same building on the same night. Their names have not been released. They are believed to be Congolese nationals. No public explanation has been offered for their presence in a building primarily occupied by international humanitarian personnel. Second, no debris from any drone has been recovered or publicly identified at the scene. In any genuine drone strike investigation, physical remnants of the munition — fragments of the airframe, propulsion components, guidance systems — constitute primary forensic evidence for establishing the origin, type, and operator of the weapon. Their absence is not a neutral fact. It is a gap that either requires explanation or constitutes evidence in itself. Third, the identity and role of the two Congolese victims has not been established. In a context where RDF/M23 officials were inside the building within hours of the killing, the question of why those individuals were present — and whether they were known to RDF/M23 — must be treated as a live investigative issue.
Taken together, these details raise a possibility that the PNAT investigation cannot ignore: that the two Congolese victims may have been killed precisely because they were witnesses. If Karine Buisset was killed by gunfire before the drone was deployed as cover, anyone else present in the building at the time of the shooting would have been a witness to the method of killing. The elimination of witnesses is not a theoretical risk in cases involving RDF/M23 and the RPF's institutional predecessors. It is a documented operational practice. The pattern of killing individuals who could provide testimony inconvenient to the RPF narrative — from political moderates in the early 1990s to refugee community leaders in the DRC — runs through the same three-decade institutional history that shapes every other element of this case.
The cumulative effect of these facts — RDF/M23 officials inside the building before investigators arrived, signs of shooting alongside drone damage, no drone debris recovered, two unidentified Congolese victims whose presence has not been explained, and no call by RDF/M23 for independent investigation — constitutes a serious and documented risk of spoliation of evidence. Spoliation, in legal terms, is the destruction, alteration, concealment, or failure to preserve evidence relevant to litigation or investigation. It does not require proof of deliberate intent to meet the threshold of legal significance. The conditions for spoliation existed from the moment RDF/M23 took control of the scene. The PNAT must treat the totality of these circumstances — not merely the drone attribution question — as the evidentiary foundation of its inquiry.
This behaviour is again consistent with the RPF's historic operational pattern. After the assassination of President Habyarimana, the crash site and physical evidence were not preserved for independent forensic analysis. Rwandan authorities controlled access to critical evidence. The Bruguière investigation was conducted at a permanent evidentiary disadvantage as a result. The contamination or loss of primary evidence has been a recurring feature of incidents in which Rwanda or its proxies have had an interest in shaping attribution. It is not a coincidence. It is a method.
The French PNAT investigation must also grapple with the broader context of MONUSCO's position in Goma. An occupying armed group does not facilitate the independent operation of a UN peacekeeping mission whose mandate includes investigating violations of international humanitarian law. In RDF/M23-held Goma, the independent investigative capacity that should have secured the Buisset crime scene either had no authority to act or was not positioned to do so before RDF/M23 had already been inside the building. This is a structural consequence of the occupation itself — further evidence that Rwanda's military presence in eastern DRC destroys the very conditions necessary for accountability.
From Kigali to Goma: The Continuity of Impunity
There is a direct institutional line from the RPF of the early 1990s to the RDF of 2026. The commanders who directed operations in the DRC during the First and Second Congo Wars are the same generation that now commands the force backing RDF/M23. The doctrine of targeted killing, the willingness to eliminate individuals and communities perceived as obstacles, and the use of strategic violence as a tool of political control are not aberrations. They are institutional inheritances. The killing of moderate Rwandan politicians before 1994, the assassination of President Habyarimana, the mass killing of refugees in the DRC, and the ongoing RDF/M23 offensive are chapters in a single account of what happens when strategic violence is deployed without accountability.
Karine Buisset was killed in a city governed by the product of that doctrine. Whether the specific drone that killed her was Congolese or Rwandan, she died because Rwanda created the conditions — the occupation of Goma, the militarisation of civilian space, the destruction of civilian protection architecture — in which a UNICEF worker sleeping in a residential building could be reached by a combat drone before dawn.
France's Neutrality, Kagame in Paris, and the Alibi of Distance
The Paris Conference, the Goma Airport, and France's Misjudgment of Kagame
The killing of Karine Buisset did not occur against a blank diplomatic canvas. It occurred five months after France had invested enormous political capital, financial resources, and institutional credibility in a major international effort to address the very humanitarian crisis in which she was working. On 30 October 2025, France and Togo co-hosted the International Conference for Peace and Prosperity in the Great Lakes Region in Paris, bringing together seventy states and international organisations. At its close, President Macron announced that more than €1.5 billion had been pledged in humanitarian aid for displaced populations in eastern DRC and the wider Great Lakes region. The conference was a significant diplomatic undertaking. France had positioned itself as the driving force behind an international humanitarian response to a crisis it publicly attributed to the RDF/M23 offensive backed by Rwanda.
At that same conference, Macron announced that Goma airport would reopen to humanitarian flights within weeks, along with secure corridors for aid delivery. This was not a vague aspiration. It was a public commitment made on behalf of the parties to the conference — a commitment that required RDF/M23, and by extension Rwanda, to implement it. It was not implemented. By January 2026, three months after Macron made that announcement, Goma airport remained closed. The EU's own humanitarian documentation confirmed that the airport shutdown was continuing to severely restrict the humanitarian response, including access for the very organisations working to reach displaced populations in North and South Kivu. Humanitarian air bridges were being operated from Nairobi, routing supplies overland into a region whose primary airport RDF/M23 continued to control and deny.
This is not a minor footnote about logistics. Kagame rejected a commitment that France had made publicly, at a conference France had organised, with the stated purpose of delivering aid to the populations whose suffering France had placed at the centre of its diplomacy. The rejection was not diplomatic. It was operational: RDF/M23 continued to control Goma airport and continued to deny humanitarian access. France had mobilised €1.5 billion in pledges and the full weight of its diplomatic convening power, and the entity it was implicitly relying upon to implement the result chose not to comply. The appropriate response to that rejection — in any honest accounting of diplomatic realities — was a fundamental reassessment of France's relationship with Kigali.
That reassessment did not happen. Four months after Kagame rejected the airport commitment, France hosted him at the Élysée for the Nuclear Energy Summit. The France-Rwanda bilateral relationship continued undisturbed, as if the conference of October 2025 had produced nothing of consequence. France chose to interpret Kagame's rejection not as evidence that Rwanda's objectives in the DRC were incompatible with humanitarian access, but as a matter for ongoing negotiation. The diplomatic relationship was preserved at the expense of the diplomatic outcome.
France's Misjudgment: Believing Humanitarian Workers Were Safe in RDF/M23 Territory
Embedded in France's management of its relationship with Rwanda was a specific and consequential assumption: that France's good relations with Kigali conferred some degree of protection on humanitarian workers operating in RDF/M23-controlled territory. This assumption was never stated explicitly. It did not need to be. It was implicit in the decision to allow UNICEF personnel — including French nationals — to continue working in Goma under RDF/M23 occupation without triggering the kind of security reassessment that a clear-eyed reading of RDF/M23's conduct demanded. France believed that the Paris conference, the diplomatic relationship with Kagame, and its role as a principal interlocutor in the Great Lakes process gave it a measure of influence over conditions on the ground in RDF/M23-held areas. That belief was wrong.
The evidence that it was wrong was available before 11 March 2026. Kagame had rejected the airport commitment. RDF/M23 had continued its military offensive in clear violation of the Washington Accords, capturing Uvira days after the accord was signed. The United States had imposed sanctions on the entire Rwandan Defence Force. The EU had sanctioned Rwandan officials. These were not ambiguous signals. They were a documented pattern demonstrating that RDF/M23 and Rwanda did not regard diplomatic relationships with Western governments as operational constraints. The protection that France believed its relationship with Kigali extended to humanitarian workers in RDF/M23-held Goma was not real. It was an assumption that had no basis in RDF/M23's actual conduct.
France's misjudgment of Kagame's objectives in the DRC has direct implications for the safety assessment that governed Karine Buisset's continued presence in Goma. A government that accurately understood what Rwanda and RDF/M23 were doing — that they were prosecuting a war of occupation with no intention of implementing humanitarian commitments, that the airport rejection was a signal not a negotiating position, that RDF/M23-controlled territory was becoming progressively more dangerous for international personnel — would have issued different security guidance. It would have recognised that the diplomatic relationship with Kigali did not neutralise the operational threat to humanitarian workers on the ground. France did not make that recognition in time.
This does not shift primary responsibility away from Rwanda, from RDF/M23, and from the occupation that created the conditions for Buisset's death. That responsibility is absolute and remains with Kigali. But it does establish that France bears a secondary accountability: for the gap between its diplomatic ambitions and its security judgments; for continuing a relationship with Kagame that encouraged a false sense of protection; and for failing to reassess the safety of French nationals working in RDF/M23-controlled territory after Kagame demonstrated, repeatedly and publicly, that his forces were not subject to the diplomatic constraints France believed it had helped to construct. The Paris conference generated €1.5 billion in pledges and produced a commitment on the Goma airport that was never honoured. It also produced a diplomatic comfort zone around the France-Rwanda relationship that outlasted the evidence on which it was based. Karine Buisset worked inside that comfort zone. She should not have been left there.
While the United States, the United Kingdom, Germany, and the European Union have all taken concrete measures against Rwanda in response to its documented support for RDF/M23, France has not followed suit. The contrast is not marginal. It defines France's position in this conflict as one of structural complicity through inaction.
The record of allied action against Rwanda is substantial. On 2 March 2026, the United States sanctioned the entire Rwandan Defence Force and four of its senior commanders — including the army chief of staff and the chief of defence staff — for direct operational support of RDF/M23. On 6 March 2026, the US State Department announced visa restrictions on senior Rwandan officials for fuelling instability in eastern DRC in violation of the Washington Accords. The European Union had sanctioned RDF/M23 leaders and Rwandan officials in March 2025 and withheld funds from the Rwandan army's participation in the EU-backed Mozambique intervention. Germany suspended aid programmes. The United Kingdom suspended most financial aid to Rwanda. Rwanda responded to the EU sanctions by cutting diplomatic ties with Belgium.
France has taken none of these steps. France authored UN Security Council Resolution 2773 in 2025, which called on RDF/M23 to withdraw and on Rwanda to cease support for the group. France's UN representative stated that the Council had risen to its responsibility and that Rwanda's support for RDF/M23 must end. These are words. They have not been followed by the suspension of bilateral aid, the imposition of sanctions, or any other mechanism that might alter Rwanda's calculus. France's relationship with Rwanda — rebuilt since 2021 when Macron acknowledged France's heavy responsibility in the 1994 genocide and the two countries restored full diplomatic ties — has been maintained at the direct expense of accountability for what Rwanda is doing in the DRC.
The consequence of this neutrality is not merely diplomatic inconsistency. It is a structural enabler of violence. When one major Western power maintains business-as-usual relations with a state whose military is under US Treasury sanctions for supporting a group committing summary executions and mass displacement, the message received in Kigali is that the cost of the war in the DRC remains manageable. France's failure to act, against the explicit position of its closest allies, has contributed to that calculation.
Kagame in Paris: The Significance of the Timing
On 10 March 2026, President Paul Kagame arrived in Paris for the Nuclear Energy Summit hosted by Emmanuel Macron. He met with Macron at the Élysée, addressed an audience of sixty heads of state and international officials, and presented Rwanda as a responsible, forward-looking state committed to clean energy and economic transformation. The following morning, 11 March 2026, Karine Buisset was killed in a drone strike in Rwanda-backed RDF/M23-held Goma.
This coincidence is not incidental. It is consistent with a documented operational pattern in Kagame's conduct across three decades of conflict. When the RDF or RDF/M23 have planned and executed major military operations — the seizure of towns and strategic sites, significant offensive advances, killings of civilians or high-profile targets — Kagame has frequently been abroad. The function of this pattern is not difficult to identify. Physical distance from Rwanda at the moment of a major operation serves as a built-in denial mechanism. Kagame is in Paris for an energy summit; Rwanda could not possibly be directing events in Goma; the drone strike must be the work of FARDC. The logic is circular, and it is designed to be.
This is not a new strategy. It mirrors the operational logic behind Kagame's reported movements during critical phases of the 1994 genocide and the subsequent Congo wars. The RPF's pattern of killing and then attributing — documented in the Bruguière investigation, the Hourigan affidavit, and the Mapping Report — was invariably accompanied by the performance of distance: senior RPF figures were elsewhere, at peace talks, at summits, or otherwise visibly engaged in diplomacy while operations on the ground continued. The alibi of distance has always been central to Rwanda's deniability architecture.
That Macron hosted Kagame on 10 March 2026, welcomed him publicly, and engaged him as a partner in a major international summit, while Kagame's proxy force held a Congolese city under armed occupation, while US sanctions were already in effect against the Rwandan military, and while a French UNICEF worker was killed in that city the very next morning, is not merely a diplomatic embarrassment. It is a demonstration of the structural problem at the heart of the international response to this conflict. France provided Kagame with exactly the stage he required: international visibility, diplomatic legitimacy, and the performance of normality at the precise moment his forces' operations were claiming lives.
The Cost of Looking Away: The Broader International Failure
The Parquet National Antiterroriste's war crimes investigation, opened on 13 March 2026, is a more substantive response than a statement. It is the kind of investigative mechanism that should have been activated years ago for RPF crimes documented in the Congo. Its opening now, following the killing of a French national, reflects the reality that international political will is mobilised more readily when the victim is a citizen of a powerful state. That is not an argument against the investigation. It is an argument for extending the same investigative seriousness to the thousands of Congolese who have died in this conflict without triggering comparable international attention.
The PNAT's investigation must proceed without deference to the France-Rwanda diplomatic relationship. It must examine the full context of the drone strike — including who created the military conditions in Goma that made the strike possible. If France's judiciary pursues this investigation with genuine independence, it will inevitably confront the question of Rwanda's responsibility for the war in which Karine Buisset was killed. That will be an uncomfortable finding for the Élysée. It is nevertheless the finding that justice requires.
Conclusion: The Root Cause Is the Occupation
The question of which military operated the drone that killed Karine Buisset is important, and the PNAT must establish the answer. But there is a prior question, simpler and more fundamental, that requires no forensic investigation to resolve: if Rwanda were not occupying eastern DRC, Karine Buisset would not have died.
She was not in Goma because a conflict between Congolese parties had made the city dangerous. She was in Goma because Rwanda's proxy force had seized the city, transformed it into a theatre of active military occupation, militarised its airspace, and created the conditions in which combat drones operate in residential neighbourhoods before dawn. The drone — whoever fired it — flew over a city that Rwanda made into a war zone. The building that was destroyed was struck in a city that Rwanda's forces hold by force. The woman who died worked in a city that would be at peace if Rwanda's military and its proxy had not invaded and occupied it.
This is the simplest and most consequential fact in this case. Every layer of complexity that surrounds it — the attribution debate, RDF/M23's communications strategy, the diplomatic manoeuvrings in Paris, the question of which party fired which weapon — is downstream of this single root cause. Rwanda's military occupation of Congolese territory is the condition that made Karine Buisset's death possible. Remove the occupation, and she is alive. That is not an analytical abstraction. It is a statement about cause and responsibility that no amount of diplomatic hedging can dissolve.
The killing of Karine Buisset in Goma on 11 March 2026 is the latest consequence of a political culture of impunity that has been constructed and maintained around Rwanda for more than three decades. The RPF's targeted killings of Rwandan officials and political moderates in the early 1990s, the assassination of President Habyarimana, the mass killing of Hutu refugees in the DRC, RDF/M23's immediate blame of FARDC without any call for independent investigation, and the ongoing military occupation are not separate events. They are chapters in a single account of how strategic violence, deployed without accountability, becomes a permanent instrument of governance.
The PNAT investigation must be pursued wherever the forensic evidence leads. France must end its diplomatic neutrality and align with its allies in imposing concrete costs on Rwanda. The international community must reckon with the thirty-year accountability deficit that has made this conflict, and this killing, structurally possible. And the demand that anchors all of these responses must be the one demand that addresses the root cause: Rwanda must end its military occupation of the Democratic Republic of Congo.
Security concerns, however genuine, cannot justify the creation of insecurity for others. That principle — applied here with full force — means that Rwanda's stated security interests do not entitle it to occupy Congolese cities, control Congolese airspace, or operate a proxy military force that kills UNICEF workers in their sleep. Karine Buisset, child protection officer, 54 years old, from Belz in Brittany, deserves no less than the truth: she died because Rwanda occupied Goma. Neither she, nor the millions of Congolese who have lived and died under this conflict, should be denied that clarity.
Frequently Asked Questions
Who was Karine Buisset?
Karine Buisset was a 54-year-old French national from Belz, Morbihan, in Brittany. She was a UNICEF child protection officer working in Goma, eastern DRC. She was killed in a drone strike on 11 March 2026 in a residential building in Goma's Himbi neighbourhood. Her mission had been extended at the last moment. UNICEF described her as a dedicated humanitarian who worked tirelessly to support children and families affected by conflict and crisis.
Who carried out the drone strike that killed Karine Buisset?
RDF/M23 immediately attributed the drone strike to FARDC without calling for any independent investigation. France's Parquet National Antiterroriste opened a formal war crimes investigation on 13 March 2026. The failure of RDF/M23 to demand independent forensic inquiry is itself analytically significant: a party genuinely seeking truth calls for investigation. RDF/M23 issued a statement and moved on, consistent with the RPF's historic pattern of immediate attribution followed by obstruction of inquiry. Regardless of which military fired the specific drone, Rwanda's occupation of Goma created the conditions in which the strike was possible.
Why does RDF/M23's presence at the crime scene matter for the investigation?
Multiple features of the Buisset crime scene raise serious concerns about spoliation of evidence. RDF/M23 officials entered the building before any independent investigative body secured it. The building shows signs of shooting as well as drone damage, raising the possibility that she was killed by gunfire before the drone was deployed as cover. No debris from any drone has been recovered or identified at the scene. Two other people — believed to be Congolese nationals — were also killed, but their names have not been released and no explanation has been given for their presence. They may have been witnesses to the method of killing. RDF/M23 has not called for any independent investigation. Taken together, these facts constitute a documented risk of spoliation — the destruction, alteration, or concealment of evidence relevant to establishing how and by whom Karine Buisset was killed. The PNAT must treat all of these circumstances as core elements of its inquiry.
Was Karine Buisset definitely killed by the drone strike?
This has not been established. The building where she was killed shows signs of shooting as well as drone damage, raising the forensic possibility that she was killed by gunfire and that the drone was deployed subsequently as cover to frame FARDC. These are categorically different scenarios: drone strike damage suggests tragic but explicable military action by another party; a targeted shooting masked by a drone strike is an execution. The PNAT must forensically sequence the drone damage and the shooting evidence and establish the cause of death before any attribution claim can be evaluated. RDF/M23's presence inside the building before independent investigators arrived means the physical evidence that would answer this question may have been compromised.
If Rwanda had not occupied Goma, would Karine Buisset have died?
No. Buisset died in a city that Rwanda's proxy force seized in January 2025, in a building struck in a militarised urban environment created by Rwanda's occupation. Without the occupation, there is no militarised airspace, no combat drones over residential neighbourhoods, no pre-dawn strike on a building housing aid workers. The root cause of her death is Rwanda's military occupation of eastern DRC. Every other question — attribution, cause of death, diplomatic responses, legal investigations — is downstream of that single fact.
Did the RPF carry out killings and blame others during the 1990s conflicts?
Yes. Scholars including Filip Reyntjens have documented a consistent RPF pattern of killing Hutu officials, moderate politicians, and civilians, then attributing those killings to Hutu extremists or the Habyarimana government. The assassination of President Habyarimana — attributed immediately to Hutu extremists but the subject of serious judicial challenge pointing toward RPF responsibility — is the most consequential instance of this pattern. The UN Mapping Report similarly documented how mass killings of Hutu refugees in the DRC were attributed to crossfire or inter-ethnic violence rather than to the organised RPF-linked operations the Report identified. This false-flag attribution strategy is an institutional inheritance that RDF/M23, under Rwandan command, continues to deploy.
What is RDF/M23 and what is Rwanda's role in it?
RDF/M23 (Mouvement du 23 Mars) is an armed group that has controlled large swathes of eastern DRC, including Goma, since January 2025. Multiple UN Group of Experts reports have documented Rwandan Defence Force troop deployments alongside RDF/M23, Rwandan weapons supply, and command links to Kigali. Rwanda denies direct involvement. The UN and Western nations publicly accuse Rwanda of backing the group.
Why was France's Paris conference on the DRC relevant to Karine Buisset's death?
On 30 October 2025, France and Togo co-hosted an international conference in Paris that raised more than €1.5 billion in humanitarian pledges for displaced people in eastern DRC. At the conference, Macron announced that Goma airport would reopen to humanitarian flights within weeks. Kagame and RDF/M23 did not implement this commitment. By January 2026, the airport remained closed. France continued its diplomatic relationship with Rwanda regardless. The conference created a false sense of French influence over conditions in RDF/M23-held Goma — an assumption that humanitarian workers, including French nationals, were operating under a degree of diplomatic protection that did not exist in practice.
Does France bear any responsibility for Karine Buisset's safety?
Yes, partially. Primary responsibility for Buisset's death rests with Rwanda and its occupation of Goma. However, France bears a secondary accountability for misjudging Kagame's objectives. France assumed that its good diplomatic relations with Kigali conferred a degree of protection on humanitarian workers in RDF/M23-held territory. That assumption was contradicted by clear evidence before 11 March 2026: Kagame had rejected the Goma airport commitment, RDF/M23 had violated the Washington Accords, and the US had sanctioned the entire Rwandan military. France did not revise its security assumptions or its relationship with Kigali in response to any of these signals. Karine Buisset was left working in territory that was more dangerous than France's diplomatic framework acknowledged.
Kagame arrived in Paris on 10 March 2026 for the Nuclear Energy Summit hosted by President Macron. Buisset was killed in Goma the following morning. The timing is consistent with a documented operational pattern in which Kagame travels abroad during major RDF/M23 operations, providing built-in deniability. His physical distance from Rwanda serves as a denial mechanism: Rwanda could not be directing operations in Goma if Kagame is at a summit in Paris. This pattern mirrors the RPF's historic use of high-level diplomatic visibility to create distance from operations on the ground.
Why has France not sanctioned Rwanda like the US, UK, and EU have?
France authored UN Security Council Resolution 2773 (2025) calling for RDF/M23 to withdraw and Rwanda to cease support for the group, but has not followed other Western allies in imposing sanctions or suspending bilateral aid. The US sanctioned the entire Rwandan Defence Force in March 2026; Germany suspended aid; the UK suspended most financial aid; the EU sanctioned Rwandan officials. France has maintained its diplomatic and economic relationship with Rwanda, rebuilt after Macron acknowledged France's heavy responsibility in the 1994 genocide in 2021. France's failure to impose concrete measures has contributed to Rwanda's calculation that the cost of war in the DRC remains politically manageable.
President Emmanuel Macron confirmed Karine Buisset's death on social media and called for respect for humanitarian law. On 13 March 2026, France's Parquet National Antiterroriste opened a formal investigation for murder constituting a war crime. UNICEF Executive Director Catherine Russell stated that civilians and aid workers must never be targeted.
Who shot down President Habyarimana's plane?
The question remains formally unresolved but is the subject of serious judicial and scholarly inquiry. French magistrate Jean-Louis Bruguière concluded in 2006 that the RPF, under orders from Paul Kagame, was responsible. A Spanish judicial investigation reached similar conclusions. ICTR investigator Michael Hourigan's affidavit stated that credible evidence pointing to RPF responsibility was suppressed. Scholars including Filip Reyntjens and Gérard Prunier have engaged critically with this evidence.
What is the UN Mapping Report?
The UN Mapping Report (2010) is a 550-page United Nations document that documented serious violations of international humanitarian and human rights law committed in the DRC between 1993 and 2003. It documented RPF-linked forces killing Rwandan Hutu refugees at scale and noted that some incidents may constitute acts of genocide. No prosecutions have followed.
What should the international community do?
Credible responses require: full implementation of UN sanctions; suspension of bilateral aid to Rwanda contingent on RDF/M23 disengagement; referral of the situation to the International Criminal Court; transparent conduct of the PNAT war crimes investigation; and a diplomatic reckoning with the impunity that has enabled Rwanda to operate without accountability in the DRC for three decades.
References
Bruguière, J.-L. (2006) Ordonnance de soit communiqué. Tribunal de grande instance de Paris. Paris: Ministère de la Justice.
Del Ponte, C. and Sudetic, C. (2009) Madame Prosecutor: Confrontations with Humanity's Worst Criminals and the Culture of Impunity. New York: Other Press.
Défis Humanitaires (2025) Humanitarian situation in the DRC. December. Available at: https://defishumanitaires.com [Accessed 13 March 2026].
European Commission (2026) Democratic Republic of the Congo — Humanitarian Aid. Available at: https://civil-protection-humanitarian-aid.ec.europa.eu [Accessed 13 March 2026].
French Ministry for Europe and Foreign Affairs (2026) Multipartite Meeting in Support of the Peace Process in the Democratic Republic of Congo and the Great Lakes Region, 19 January. Available at: https://www.diplomatie.gouv.fr [Accessed 13 March 2026].
News24 (2025) DRC's Goma airport to reopen for humanitarian flights — Macron. 30 October. Available at: https://www.news24.com [Accessed 13 March 2026].
RFI/AllAfrica (2025) Over €1.5 Billion Pledged for Africa's Great Lakes Region at Paris Conference. 31 October. Available at: https://allafrica.com [Accessed 13 March 2026].
Hourigan, M. (2006) Affidavit to the International Criminal Tribunal for Rwanda. On file with ICTR. Referenced in Reyntjens, F. (2011).
Human Rights Watch (2026) US Sanctions Rwandan Army, Commanders. 3 March. Available at: https://www.hrw.org [Accessed 13 March 2026].
International Crisis Group (2025) The RDF/M23 Offensive: Elusive Peace in the Great Lakes. Brussels: ICG. Available at: https://www.crisisgroup.org [Accessed 13 March 2026].
KT Press (2026) President Kagame in Paris to Advance Rwanda's Nuclear Energy Ambitions. 10 March. Available at: https://www.ktpress.rw [Accessed 13 March 2026].
Top Africa News (2026) Kagame to Join Macron in Paris for Landmark Nuclear Summit Amid Rising Regional Tensions. 4 March. Available at: https://www.topafricanews.com [Accessed 13 March 2026].
United Nations Security Council (2025) Resolution 2773 (2025): Strongly Condemning Ongoing Offensives by RDF/M23 in Eastern Democratic Republic of the Congo. New York: United Nations. Available at: https://press.un.org [Accessed 13 March 2026].
United States Department of State (2026) Sanctioning Rwandan Violators of the Washington Accords for Peace and Prosperity. 2 March. Available at: https://www.state.gov [Accessed 13 March 2026].
Prunier, G. (2009) Africa's World War: Congo, the Rwandan Genocide, and the Making of a Continental Catastrophe. Oxford: Oxford University Press.
Reyntjens, F. (2011) 'Constructing the truth, dealing with dissent, domesticating the world: Governance in post-genocide Rwanda', African Affairs, 110(438), pp. 1–34.
Stearns, J. (2012) Dancing in the Glory of Monsters: The Collapse of the Congo and the Great War of Africa. New York: PublicAffairs.
UNICEF (2026) Statement by UNICEF on the killing of a staff member in the Democratic Republic of Congo, 11 March. Available at: https://www.unicef.org/press-releases [Accessed 13 March 2026].
United Nations (2010) Report of the Mapping Exercise documenting the most serious violations of human rights and international humanitarian law committed within the territory of the Democratic Republic of the Congo between March 1993 and June 2003. Geneva: OHCHR.
United Nations Group of Experts on the Democratic Republic of the Congo (2024) Final Report to the Security Council. New York: United Nations Security Council.
Al Jazeera (2026) UN aid worker killed in drone strike in DR Congo's rebel-held Goma. 11 March. Available at: https://www.aljazeera.com [Accessed 13 March 2026].
Author: The African Rights Campaign, London, UK
Comments
Post a Comment