The Kosovo–Serbia dialogue was built to normalise relations. In practice, it normalised a deeper asymmetry: accountability on one side, revision on the other.
Mark Gojcaj | 17 June 2026
In the summer of 2020, wartime accountability finally interrupted the Kosovo-Serbia dialogue, and it removed a head of state. The defendant was not a Serbian official. It was Hashim Thaçi, the president of Kosovo, indicted by a court in The Hague that Kosovo had itself been required to create, who resigned to face charges over the conduct of a war his side, by every international account, had not begun. A White House summit was called off within days of the indictment. It was the most consequential meeting of war crimes and the normalisation process in a decade, and its subject was the side that had been bombed.
Any honest account of normalisation has to begin there, because that moment fixes the ledger. The machinery of wartime accountability is real and it is active, and in the years that have mattered most to the dialogue it has fallen hardest on Pristina. The Kosovo Specialist Chambers, sitting in the Netherlands, were built by a Kosovo constitutional amendment to try Kosovo Albanians, and they are trying them now. This is not a grievance. Crimes are crimes, and a court that reaches the side that won the moral argument as readily as the side that lost it is doing something rare and right. The grievance is about the other pan of the scale, the one that has stayed empty.
Set beside Thaçi in the dock, consider what the dialogue has asked of Belgrade about its own war. The process opened in 2011 as a technical exercise, civil registries and freedom of movement and licence plates, and across more than two hundred meetings and thirty-eight agreements it has stayed resolutely turned toward the future. Its governing instruction, written into the text the parties endorsed at Ohrid, is to proceed "without prejudice to the different view of the Parties on fundamental questions." The past enters the entire architecture once, as a declaration on missing persons folded into an implementation annex, and that is the commitment honoured least. Serbia has opened accession chapters on the strength of promises about what comes next. Suva Reka was tried in a Belgrade courtroom; the graves at Batajnica, where bodies were trucked north from Kosovo and buried on Serbian soil, were exhumed and investigated; Reçak anchored a judgment in The Hague. Each of these crimes was confronted somewhere. Not one of them has ever reached the dialogue's agenda. The dialogue keeps a column for relations and no column for responsibility.
The frame that permits this is the oldest one the international community carries about these wars: that all sides committed crimes. In a courtroom the sentence is not only true, it is indispensable; it is why the Specialist Chambers exist and why Kosovo Albanians answer before them. But a sentence that opens an inquiry in The Hague does very different work in Brussels. In the negotiating room "all sides" is not a finding to be tested against evidence. It is a settlement assumed in advance, a way of spreading responsibility evenly enough that no one at the table has to be the guilty party. And once responsibility is distributed by assumption rather than weighed by proof, the heavier ledger receives the larger discount. The state with the most to answer for benefits most from the agreement never to ask.
This is the quiet transaction at the centre of the process, and it required no one to intend it. To keep Belgrade at the table, the facilitators adopted an operating standard under which the past is out of scope, blame is shared by default, and nothing about the war need be conceded in order to keep moving. That standard did not arrive from nowhere. It is, almost exactly, Belgrade's own: settle nothing, distribute everything, defer the rest. The dialogue did not overcome Serbia's posture toward its war. It absorbed it, and renamed the absorption neutrality.
A hostile reader will demand the document where this was decided, the memo in which Europe resolved to underwrite a forgery. There is none, and none is needed. The proof is not in anyone's intentions; it is in the ledger, and the ledger is open to inspection. Over fifteen years the accountability that materialised fell on the side that was attacked. The process built to address relations declared the past off-limits in its own text. And the active revisionism this paper has catalogued this week, the judge and the documentarian and the party line, met no diplomatic cost at any point along the road. Three observable facts, no motive required. A pattern does not need a confession to be a pattern.
Normalisation built on that foundation is not the end of a conflict. It is the conflict settled on Belgrade's terms and ratified by the parties who were meant to hold the balance. It asks Kosovo to accept as a neighbour a state the process has required to pay nothing for its war and to concede nothing about it, and to call the arrangement peace.
The dialogue's defenders have a real answer to this, and it deserves to be stated at its strongest. The past, they will say, cannot be a precondition; make it one and everything freezes; the living need bridges and number plates and a list of their missing more than they need a reckoning that may never come. That is the best case the process has, and there is truth in it. But there is a difference between declining to make the past a precondition and building a process that guarantees the past will never carry a price. The first is pragmatism. The second is a subsidy, and a subsidy is what the architecture delivers, whatever the intention behind it.
Until the war costs Belgrade something at the table, the dialogue is not normalising relations between two states. It is normalising one state's account of what it did. A state still revising that account in real time, on its broadcasts and in its courtrooms and through the men it sends to defend a clean Reçak, is not a neighbour bargaining in good faith. It is an author who has not agreed to stop revising the book, handed a treaty to sign on a page he fully intends to rewrite.
The other pan of the scale is still empty. The process contains no mechanism that would ever fill it, and a peace that leaves it empty is not peace. It is the forgery, notarised.
