The Pretoria Peace Is Effectively Dead. A 24-Hour Battle in Tigray Just Confirmed It.

Ethiopian federal forces and Tigrayan fighters fought a roughly 24-hour battle in western Tigray on 1-2 August 2026, the most serious clash since the 2022 Pretoria agreement ended a war that the African Union has estimated killed at least 600,000 people and displaced some two million. Fighting spread from Humera, on the border with Sudan, to Shererina further south, with both sides deploying artillery, mechanised units and drones before a lull returned to the border area several days later. The two sides blame each other for starting it: Tigray's regional government in Mekelle says federal troops, joined by an allied Amhara militia, advanced first from Shererina at dawn on Saturday; a federal official countered that the Tigray People's Liberation Front had been organising a "tripartite" push with the Amhara Fano militia and the Oromo Liberation Army near the tri-border area where Ethiopia, Eritrea and Sudan meet. Civilians paid the immediate price regardless of which account is accurate, with refugees crossing into Sudan as the fighting intensified.

The clash matters less for its scale, which by the standards of the 2020-2022 war was limited and short-lived, than for what it confirms: the institutional architecture that ended that war has effectively collapsed, and the actor best positioned to benefit from its collapse, Eritrea, was never a signatory to it in the first place.

How the peace unravelled from the inside

The Pretoria deal was never a settlement between Addis Ababa and a unified Tigray; it was a settlement between Addis Ababa and the wing of the TPLF, led by Getachew Reda as head of the interim Tigray Interim Administration, willing to trade autonomy for demobilisation. A hardline faction under TPLF chairman Debretsion Gebremichael rejected that bargain from the outset and spent the following three years dismantling it. It expelled Getachew and his allies from the party in 2024, forced the Tigray Interim Administration out in a de facto takeover in March 2025 that installed a Debretsion loyalist in its place, and in May 2026 went further still, reconstituting the pre-war Tigray Regional Council and naming Debretsion regional president, formally erasing the interim governance structure the Pretoria agreement had built. Getachew fled to Addis Ababa, took a position advising Prime Minister Abiy Ahmed, and founded a rival regional party, Simret, that so far commands no forces on the ground. The institutional bridge Pretoria built between Mekelle and the federal government no longer exists; what remains in Mekelle is a leadership with every incentive to test the federal government militarily rather than negotiate with it.

Eritrea's shadow presence

Eritrea was never a party to Pretoria and has never fully withdrawn the forces it deployed into Tigray during the war. Eritrean commanders have continued to meet TPLF officials, and Ethiopia's federal government has accused Asmara of backing both the hardline TPLF faction and, separately, the Fano insurgency active in Amhara; Eritrea denies both charges. That relationship gives the current flare-up a second, more dangerous dimension. Analysts at the International Crisis Group and the Critical Threats Project have each warned that further clashes between Ethiopian forces and the TPLF carry a real risk of drawing in Eritrea and Sudan directly, given both states' interest in a weakened, fragmented federal government in Addis Ababa. A local battle over a border town becomes materially more dangerous once a neighbouring state has both forces already in place nearby and an incentive to see the fighting continue.

The Assab question sits underneath everything

The Tigray flashpoint cannot be separated from the larger dispute simmering between Addis Ababa and Asmara over the Eritrean port of Assab. Since late 2025, Prime Minister Abiy has described Ethiopia's access to the sea, lost when Eritrea won independence in 1993, as a "matter of survival" and has called reclaiming it a matter of correcting a historical "mistake." Ethiopian military officials have used comparably absolute language: ENDF official Teshome Gemechu has said controlling the port "has become our survival interest worth paying any price for." Abiy has called publicly for international mediation, naming the United States, China, Russia and the European Union as candidates, a list that notably omits Saudi Arabia, which brokered the 2018 Jeddah Peace Agreement between the two countries but has since built a closer relationship with Asmara and appears reluctant to be seen taking Addis Ababa's side. The International Crisis Group's assessment, published earlier this year, that the risk of an Ethiopia-Eritrea war "remains high three years after the last major war in the area" reads, after the first week of August, less like a warning about a distant possibility and more like a description of where things already stand.

A regionalising conflict

The Tigray-Eritrea axis is not unfolding in isolation. Ethiopia has separately been accused, in reporting Ethiopian officials dispute, of hosting a training camp for Sudan's Rapid Support Forces in its Benishangul-Gumuz region near the Sudanese border, built with Emirati financing after Addis Ababa's own defence relationship with the UAE fractured. Whether or not that specific allegation holds up to scrutiny, it illustrates a wider pattern FIKIR Institute has tracked across the Horn this year: as external powers, the UAE, Saudi Arabia, Turkey, Egypt and now Israel, compete for footholds along the Red Sea and in Somalia, Sudan and Ethiopia, local disputes that might once have stayed contained increasingly acquire outside sponsors on multiple sides. A TPLF leadership that can draw on Eritrean backing, and a federal government watching Gulf capital and security partnerships flow toward its rivals, both have less reason to treat their own restraint as cost-free.

What to watch

Three signals will indicate whether the 1-2 August clash was a contained flare-up or an opening exchange. First, whether Eritrean forces remain garrisoned rather than move, and whether Asmara issues any statement distancing itself from the fighting rather than staying silent. Second, whether Abiy's government pursues the international mediation on Assab it has publicly requested, or lets the offer lapse while continuing to reinforce its northern frontier, a sign that mediation rhetoric is not being matched by preparation for talks. Third, whether the African Union, whose Pretoria framework has now been dismantled from within Tigray by the same party that signed it, offers any updated mechanism, or treats the collapse of its own flagship peace process as someone else's problem.

Policy implications

For the African Union and international partners, the priority is to stop treating Pretoria as a settled fact. The interim governance structure it created has been dismantled by the Tigrayan side that negotiated it, and continuing to reference the 2022 agreement as the operative framework, without acknowledging that collapse, risks missing the point at which renewed mediation is still possible. For Ethiopia, the choice remains genuinely open: pursuing internationally mediated talks over Red Sea access, rather than treating it as an issue to be resolved through military pressure on Eritrea or Tigray, is the one course that does not risk reopening a war whose last iteration killed hundreds of thousands of people. For Gulf and regional actors expanding their footprint in the Horn, the lesson of the past week is that competitive external investment in fragile states does not stay confined to the sector it targets; a security vacuum in Tigray is now difficult to separate from a wider contest over ports, patronage and alignment being fought from Assab to Berbera to Port Sudan.