Pretoria

Surrender by another name

Two parties signed a peace agreement to end a war that had three combatants. The third combatant did not sign. The signatories agreed to obligations the federal-government signatory could not actually deliver on. The agreement worked anyway, for thirty months, on the basis of an understanding both sides treated as durable.

Filed late November 2022, three weeks after the signing

The agreement that ended the open phase of the Tigray war was signed at 17:30 local time on Wednesday 2 November 2022 at the headquarters of South Africa’s Department of International Relations and Cooperation in Pretoria. The signatories were Redwan Hussien, National Security Advisor in Abiy Ahmed’s office, for the Government of the Federal Democratic Republic of Ethiopia, and Getachew Reda, the Tigray People’s Liberation Front spokesperson and lead negotiator, for the TPLF. The agreement was titled the “Agreement for Lasting Peace through a Permanent Cessation of Hostilities” — the formal name — although it has been known internationally since the day of its signing as the Pretoria Agreement or as the Cessation of Hostilities Agreement (CoHA). The witnesses were the three members of the African Union High-Level Panel for the Ethiopian Peace Process: former Nigerian President Olusegun Obasanjo as AU High Representative for the Horn of Africa, former Kenyan President Uhuru Kenyatta, and former South African Deputy President Phumzile Mlambo-Ngcuka. The talks had begun on 25 October and had run for ten days. [African Union, “Cessation of Hostilities Agreement,” 2 November 2022, https://www.peaceau.org/en/article/cessation-of-hostilities-agreement-between-the-government-of-the-federal-democratic-republic-of-ethiopia-and-the-tigray-peoples-liberation-front-tplf]

The agreement runs to fifteen articles. Its substantive provisions can be grouped under five headings.

Cessation of hostilities. The parties agreed to an “immediate and permanent” cessation of all forms of hostilities, including direct combat, propaganda, hate speech, airstrikes, landmines, support to proxy forces, and “unconstitutional correspondence and relations with foreign powers.” The reference to foreign powers was, in context, a reference to the TPLF’s relationships with Eritrean opposition groups and, implicitly, to the TPLF’s earlier diplomatic engagement with Western governments during the war.

Restoration of constitutional order. The TPLF agreed to dismantle its regional administration in Tigray and replace it with an “inclusive interim administration” formed in consultation with the federal government. The TPLF further agreed to recognise federal sovereignty over Tigray Region, to accept the Ethiopian National Defence Force as “the sole military force entrusted to safeguard the country’s security,” and to participate in a federal-recognised political process going forward.

Disarmament, demobilisation, and reintegration. The TPLF agreed to disarm its combatants under a DDR programme to begin within ten days of the signing, with heavy weapons surrendered first and small arms thereafter. The total disarmament was to be completed within thirty days — that is, by 2 December 2022. The DDR programme would include provisions for the reintegration of TPLF fighters either into the ENDF, into Tigray regional police, or into civilian life with vocational training and support.

Restoration of services and humanitarian access. The federal government agreed to lift the blockade architecture imposed since November 2020: telecommunications, banking, electricity, fuel and food supplies, commercial transport, and Ethiopian Airlines service to Mekelle would be restored. Humanitarian access would be unrestricted; aid agencies would operate without federal interference; the Tigrayan civilian population’s access to medical care, food, and basic services would be normalised.

Transitional justice and accountability. Both parties committed to a transitional-justice process to be designed by the federal government with international support. The process would address accountability for atrocities committed by all parties during the war. The specific architecture of the process — its mandate, its evidentiary standards, its prosecutorial scope, its relationship to international accountability mechanisms — was not specified in the agreement and was to be developed subsequently.

Ten days after the Pretoria signing, on 12 November 2022, in Nairobi, the federal-Tigrayan implementation agreement was signed by senior military commanders. The signatories were Field Marshal Birhanu Jula, Chief of the General Staff of the ENDF, for the federal government, and Lieutenant General Tadesse Werede, Commander of the Tigray Armed Combatants, for the TPLF. The Nairobi document specified the operational modalities for the disarmament process and confirmed the 2 December 2022 deadline for completion. [PA-X Gender, “Declaration of the Senior Commanders on the Modalities for the Implementation of the Agreement,” 12 November 2022, https://www.peaceagreements.org/agreements/wgg/2464/]

The Pretoria Agreement was, by every contemporary standard of conflict resolution, a substantive achievement. It ended what had become — by late October 2022 — the world’s deadliest active armed conflict. The Wilson Center’s February 2023 retrospective characterised the agreement as having stopped a war that had killed between 383,000 and 600,000 people; the Ethiopia-Tigray peace agreement Wikipedia entry cites a comparable range. The agreement’s signing produced, within weeks, the operational restoration of basic services to Tigray — telecommunications service resumed on 7 December 2022, federal-supplied electricity began returning to parts of the region in mid-December, banking access was partially restored in January 2023, Ethiopian Airlines resumed limited Mekelle service. Humanitarian aid flows into Tigray, which had been operating at a fraction of need throughout the open-conflict phase, began increasing through November and December 2022. The TDF began the disarmament process on schedule; by 4 December 2022, Tadesse Werede stated publicly that approximately 65% of TDF combatants had begun the disarmament protocol. The fighting, in the operational sense, stopped within days of the signing.

These outcomes were real. They were also the more straightforward dimensions of the agreement. The structural defects of what had been signed at Pretoria became visible within weeks, became operationally consequential within months, and have produced — by mid-2026 — the conditions of the post-Pretoria standoff that the Tigray-Eritrea-federal triangle now operates within. Three structural defects deserve particular emphasis.

The first structural defect is the absence of the Eritrean Defence Force from the agreement.

The war that Pretoria ended had been fought, on the federal coalition side, by three armed forces: the ENDF, the Eritrean Defence Force, and the Amhara regional special forces (alongside Fano militia and the Gaachana Sirna federally-backed militia in some operational zones). The agreement was signed by the federal government and the TPLF. The EDF was not a party. The Eritrean government was not represented at Pretoria. The Eritrean government had not been invited to the negotiations. The Amhara regional government and the Fano militias were similarly not represented.

The federal government’s mechanism for managing the non-parties to the agreement was to commit, in Article 6, to use “all possible means” to ensure that “non-ENDF” forces would either be withdrawn from Tigray or would cease hostile operations. The mechanism’s operational reality, from the day of the signing onwards, was that the federal government had limited and contested leverage over the EDF, the Amhara regional forces, and the Fano militias. The EDF was a foreign army under the command of an Eritrean president who had his own strategic objectives in Tigray; the Amhara regional forces were part of the post-2018 federalist architecture that the federal government did not formally control; the Fano militias were irregular formations that no federal authority could compel to withdraw. The agreement’s commitment to ensure non-ENDF withdrawal was, in operational terms, a commitment to do something the signatory was not in a position to actually do.

The consequences began to surface immediately. Through November, December, and January, Eritrean Defence Force units remained in northern and western Tigray. They did not withdraw. They continued, in fact, to conduct documented atrocities against Tigrayan civilians in zones the EDF still occupied. Amnesty International’s September 2023 investigation, “Ethiopia: Eritrean soldiers committed war crimes and possible crimes against humanity after signing of agreement to end hostilities” — based on field interviews with 49 survivors and witnesses — documented EDF conduct at Kokob Tsibah, Mariam Shewito, and other locations in central Tigray through January 2023. The conduct included intentional killings of Tigrayan men in house-to-house searches and conditions amounting to sexual slavery for Tigrayan women and girls. The pattern documented was continuous with EDF conduct at Axum and Dengelat in November 2020 — the same forces, often the same units, the same impunity, more than two years later and after the federal government’s principal signatory in the war had purportedly ended that war. [Amnesty International, “Ethiopia: Eritrean soldiers committed war crimes and possible crimes against humanity after signing of agreement to end hostilities – new report,” 5 September 2023, https://www.amnesty.org/en/latest/news/2023/09/eritrean-soldiers-committed-war-crimes-and-possible-crimes-against-humanity-in-the-tigray-region-after-signing-of-agreement-to-end-hostilities/]

The Amhara regional administration of Western Tigray (the Welkait area), which had been operating in the contested zone since November 2020, was similarly unaffected by the Pretoria signing in its operational conduct. The ethnic-expulsion campaign against Tigrayan residents documented by Amnesty/HRW in April 2022 continued through November 2022, December 2022, January 2023, and into the months that followed. Human Rights Watch’s 2023–2024 documentation extended the pattern through August 2023. The federal commitment to ensure non-ENDF withdrawal, as applied to the Amhara regional forces in Western Tigray, was an empty commitment. The Amhara forces did not withdraw. The Tigrayan civilians who had been expelled from Western Tigray did not return. The administrative reality on the ground in late 2022 and through 2023 was that approximately 1.2 million displaced Tigrayans had been forced from a region that the agreement, on its surface, had committed to return to TPLF administration.

The second structural defect is the deferred resolution of the contested territories.

Article 7 of the Pretoria Agreement committed the parties to resolve the status of contested territories — Welkait/Tegede (in Western Tigray) and Raya/Alamata (in southern Tigray) — in line with Article 39 of the 1995 federal constitution. Article 39 provides for the resolution of internal-boundary disputes through referendum processes administered by the House of Federation. The constitutional process, fully implemented, would require a defined electorate (including who counts as a “resident” of a contested area given the displacement that had occurred), a defined ballot question, an independent administering authority, and a mechanism for implementing the result.

None of these elements were specified at Pretoria. None have been specified since. The federal government’s posture in the post-Pretoria period has been that the contested territories will be resolved “in due course” through the constitutional process; the operational reality has been that the territories remain under Amhara administration, that the Amhara federal Prosperity Party constituent is politically aligned against any referendum that would risk the territories’ return to Tigray, and that the federal government — dependent on Amhara political support, fighting a Fano insurgency that took up arms partly over the territorial question, and unwilling to expend political capital on a referendum whose outcome it could not control — has no realistic prospect of conducting the constitutional process in any near term.

The contested territories were, for the TPLF, the heart of the political question Pretoria was supposed to address. The TPLF had been demanding, throughout the war, the return of Welkait and Raya to Tigray administration. The Pretoria Agreement’s commitment to resolve the contested territories was the TPLF’s principal political gain from the agreement. The non-implementation of that commitment, through 2023, 2024, 2025, and into 2026, has produced — by mid-2026 — the operational basis of the TPLF’s claim that the federal government has breached the agreement. The TPLF’s restoration of its pre-war regional legislative council in May 2026 has been justified, by the Debretsion faction’s public communications, on the basis that the federal government’s failure to implement the contested-territories commitment has rendered the broader agreement unenforceable in its remaining provisions.

The third structural defect is the transitional-justice failure.

The Pretoria Agreement committed both parties to a transitional-justice process that would address accountability for war-time atrocities. The Ethiopian federal government’s transitional-justice policy framework was published in April 2024, eighteen months after the signing. The framework specifies a national-level process that includes truth-telling, criminal accountability, institutional reform, and reparations components. The framework’s operational implementation, through 2024, 2025, and into 2026, has been minimal. No senior federal official has been prosecuted for war-related conduct. No senior ENDF officer has been prosecuted. No senior TPLF official has been prosecuted. No senior EDF official has been investigated. The ICHREE, the only international accountability mechanism that operated during the war, had its mandate not renewed in October 2023 — a development the federal government had lobbied for and that the federal government’s African Union and European partners had ultimately not blocked.

The transitional-justice failure is, in operational terms, the most consequential structural defect of the Pretoria framework. It has meant that the war’s perpetrators — on all sides — have faced no significant legal accountability for what was done during the war. It has produced what the Crisis Group and the Heinrich-Böll-Stiftung have characterised as “impunity” — a condition in which actors who committed serious war crimes know that they will face no consequences and that future similar conduct will face no consequences. The impunity condition has, by 2026, produced a Horn of Africa regional environment in which the threat of renewed armed conflict between the same parties carries the same operational expectation of impunity. The political effect of the transitional-justice failure has been to maintain the structural pre-conditions for renewed conflict rather than to alter them. [International Crisis Group, “Turning the Pretoria Deal into Lasting Peace in Ethiopia,” 23 November 2022, https://www.crisisgroup.org/africa/ethiopia/turning-pretoria-deal-lasting-peace-ethiopia; Heinrich-Böll-Stiftung, “The Pretoria Agreement for Tigray: One Year After,” 1 November 2023, https://hoa.boell.org/en/2023/11/01/pretoria-agreement-tigray-one-year-after]

The retrospective question, from June 2026, is whether the Pretoria Agreement could have been structured differently — whether a more substantive agreement, addressing the EDF, the contested territories, and the transitional-justice architecture with more specificity, was politically achievable at the moment of signing.

The honest answer is probably not. The federal government’s negotiating position at Pretoria was substantially stronger than the TPLF’s — the TDF was being progressively defeated in the renewed federal offensive of August–October 2022, the famine was producing political pressure on the TPLF leadership to sign, the international community had limited leverage. The federal government had reasons to want the agreement as it was: bilateral, focused on TPLF concessions, deferring the harder questions to later processes the federal government would control. The TPLF had reasons to accept what was on offer: any agreement was better than continued war, and the TPLF leadership calculated that the federal government’s commitments would, even if partially implemented, produce material improvement in Tigrayan civilian conditions and a survival pathway for the TPLF as a political organisation. Both calculations were rational on the day of signing. Both calculations have been, in significant respects, refuted by the events of the following thirty-eight months.

What the Pretoria Agreement accomplished is what its signatories knew it would accomplish: it stopped the open phase of the war, it restored basic services to Tigray, and it gave both parties a survival framework. What it did not accomplish is what its signatories knew it would not accomplish: it did not produce withdrawal of non-signatory forces from Tigrayan territory, it did not resolve the contested-territories question, and it did not deliver accountability for war-time atrocities. The first set of accomplishments was substantial. The second set of non-accomplishments was structural. The combination was not a comprehensive peace settlement. It was a bilateral cessation of hostilities between two of the war’s three principal combatants, with the third combatant (Eritrea) treated as an absence the federal signatory would manage on the bilateral signatories’ behalf — a management the federal signatory was not actually positioned to deliver.

The Pretoria Agreement bought, on this assessment, thirty months of relative peace between Addis and Mekelle. It bought no peace between Addis and Asmara, between Mekelle and Asmara, or between the federal coalition and the constituencies (Amhara, Fano, OLA) whose grievances the agreement did not address. By April 2023 — five months after the signing — the Amhara federal Prosperity Party constituent had taken up arms against the federal government, and the agreement that had ended one war had become the structural cause of another. By May 2025 — thirty months after the signing — the TPLF had been deregistered as a political party, and the agreement’s commitment to TPLF political reintegration had become a fiction. By May 2026 — forty-two months after the signing — the TPLF had reinstated its pre-war regional legislative council in operational defiance of the agreement, and the agreement’s commitment to the inclusive interim administration had become a dead letter. By June 2026 — forty-three months after the signing — the federal government and the Eritrean government were publicly accusing each other of military aggression, and the agreement’s commitment to the cessation of hostilities was operative only on the bilateral signatories’ formal level, not in the broader regional reality.

Olusegun Obasanjo, speaking to the AU Peace and Security Council on 9 November 2022 — seven days after the signing — characterised the Pretoria Agreement as the realisation of “African solutions to African problems” and as a foundation for “the restoration of peace, security and stability in the Tigray Region.” The characterisation was, on the bilateral terms in which the signatories had constructed the agreement, defensible. It has also proven, in the broader political reality the agreement was supposed to transform, substantially inadequate.

The peace that ended the Tigray war was, in the most precise sense, a settlement of a bilateral conflict that had been only one component of a multilateral war. The broader war’s other components — the Eritrean intervention, the Amhara territorial dispute, the federalist-centralist political struggle, the regional alignment dynamics — were not settled at Pretoria. They have been waiting for the next form their resolution takes.

In June 2026, the next form is visible on the horizon. The Pretoria Agreement has, in its operational reality, ended. The agreement that was signed three and a half years ago purchased a strategic respite that has now expired. What comes after the respite — renewed war, deferred political settlement, or the structural collapse of the post-2018 Ethiopian federation into something else — is the central forward question of Abiy Ahmed’s premiership.

The agreement signed in Pretoria on 2 November 2022 was a real accomplishment in its day. Its limitations were also real. The limitations have, in the intervening thirty-eight months, become the operational reality of post-war Ethiopia.

---

Sources for Article 13

1. Wikipedia, “Ethiopia–Tigray peace agreement,” accessed June 2026 — https://en.wikipedia.org/wiki/Ethiopia%E2%80%93Tigray_peace_agreement
2. African Union, “Cessation of Hostilities Agreement between the Government of the Federal Democratic Republic of Ethiopia and the Tigray Peoples’ Liberation Front (TPLF),” 2 November 2022 — https://www.peaceau.org/en/article/cessation-of-hostilities-agreement-between-the-government-of-the-federal-democratic-republic-of-ethiopia-and-the-tigray-peoples-liberation-front-tplf
3. PA-X Gender, “Declaration of the Senior Commanders on the Modalities for the Implementation of the Agreement for a Lasting Peace Through a Permanent Cessation of Hostilities,” 12 November 2022 — https://www.peaceagreements.org/agreements/wgg/2464/
4. International Crisis Group, “Turning the Pretoria Deal into Lasting Peace in Ethiopia,” 23 November 2022 — https://www.crisisgroup.org/africa/ethiopia/turning-pretoria-deal-lasting-peace-ethiopia
5. Heinrich-Böll-Stiftung, “The Pretoria Agreement for Tigray: One Year After,” 1 November 2023 — https://hoa.boell.org/en/2023/11/01/pretoria-agreement-tigray-one-year-after
6. Wilson Center, “Ethiopia’s Tigray Conflict Peace Deal,” 15 February 2023 — https://www.wilsoncenter.org/blog-post/ethiopias-tigray-conflict-peace-deal-african-union
7. ISS Africa, “A year after the Pretoria agreement, hard work remains for Ethiopia,” 2 November 2023 — https://issafrica.org/iss-today/a-year-after-the-pretoria-agreement-hard-work-remains-for-ethiopia
8. Amnesty International, “Ethiopia: Eritrean soldiers committed war crimes and possible crimes against humanity after signing of agreement to end hostilities,” 5 September 2023 — https://www.amnesty.org/en/latest/news/2023/09/eritrean-soldiers-committed-war-crimes-and-possible-crimes-against-humanity-in-the-tigray-region-after-signing-of-agreement-to-end-hostilities/
9. Martin Plaut, “Text of today’s peace agreement between the Ethiopian government and Tigray administration signed in South Africa,” 2 November 2022 — https://martinplaut.com/2022/11/02/text-of-todays-peace-agreement-between-the-ethiopian-government-and-tigray-administration-signed-in-south-africa/
10. SSRC Kujenga Amani, “Ethiopia’s Pretoria Peace Agreement and the Fate of the ‘Contested’ Areas,” 20 March 2024 — https://kujenga-amani.ssrc.org/2024/03/20/ethiopias-pretoria-peace-agreement-and-the-fate-of-the-contested-areas/
11. Amani Africa, “Briefing on the agreement for lasting peace,” 9 November 2022 — https://amaniafrica-et.org/briefing-on-the-agreement-for-lasting-peace-through-permanent-cessation-of-hostilities-in-ethiopia/

---

End of expanded article for Dispatch 13. One article, approximately 3,000 words.