A Verdict That Fails Every Test It Sets

The European Commission asks Albania to govern by evidence, transparency, and reasoned judgment. Its own 2026 Rule of Law Report deserves to be measured by the same standards.

By Albatros Rexhaj (Tirana)

On Friday the European Commission published SWD(2026) 928 final, its annual chapter on the rule of law in Albania. On page four sits a single paragraph in bold. It opens with physical threats against prosecutors across the entire system, against the prosecutors of the Special Anti-Corruption Structure, against head prosecutors, against the former head of the High Prosecutorial Council. It continues with eight magistrates under protection, two judges and six prosecutors. Then, without a break, without a change of register, it arrives at a suspended minister, a request to lift her immunity, and the reactions of Albania’s executive and legislative branches. The paragraph is built to be read as one story, and it will be. Every reader who closes that page carries away the impression the Commission engineered: that the political class and the physical danger to Albania’s magistrates are one phenomenon, one menace, one source.

The report never says so. It does not have to. It was constructed so the reader would say it instead. And a document that convicts by construction has no business being this document, because this is the text by which the European Union sits in judgment of a candidate country, the text ministries must answer, chancelleries cite, and accession conferences treat as a verdict. An institution on that bench does not get to work this way. When it does, it invites the only judgment appropriate to a tribunal: trial by its own standards. We will read the 2026 Albania chapter against the tests the 2026 Albania chapter sets. It fails every one of them.

Start with evidence, the standard the report invokes on every page. The paragraph’s factual core, the threats and the protection details, is footnoted to the Albanian Government’s own written contribution, pages 22 and 24. The state reported its security measures to demonstrate that it protects its magistrates. The Commission took the state’s compliance data and repurposed it as the atmosphere of an accusation against the state. The claim that the judiciary’s governing institutions denounced political pressure is footnoted to “Country visit Albania, HJC, civil society organisations”: closed virtual meetings held in February, unnamed voices, no transcript, no text. Public statements by public institutions are citable by date and document; the High Judicial Council publishes its communiqués; the Commission cited a conversation instead. And the paragraph’s gravest sentence, that personal attacks on magistrates at all levels continued throughout the reporting period, carries no footnote at all. No magistrate named. No attacker named. No date, no quotation, nothing. The most inflammatory claim in the chapter is the only sentence in the paragraph with nothing under it.

Look harder at that footnote 27, because it is not merely thin sourcing. It is the report’s method of production, exposed. The Commission did not investigate whether political criticism of SPAK constitutes an attack on justice. It asked the people being criticized, and the organisations aligned with them, and wrote down their answer as its finding. The High Judicial Council attesting to pressure on the judiciary is the alleged victim certifying its own victimhood; that is testimony, and testimony from an interested party, which any tribunal on earth is obliged to weigh, test, and corroborate before adopting. The Commission adopted it verbatim. The civil society organisations, unnamed and undifferentiated, supplied the classification of speech as attack, and the Commission carried that classification into the Union’s official record without asking the one question that separates monitoring from stenography: by what test does criticism become an attack? The report contains no such test. No definition, no threshold, no criterion, applied nowhere, explained never. Somewhere between a February video call and a July publication, the vocabulary of aggrieved interlocutors became the vocabulary of the European Commission, unexamined, and was stamped with the Union’s seal. That is not due diligence. That is laundering. A grievance enters the country-visit channel, anonymity strips it of authorship, and it emerges on the other side as a finding of the European Union, unfalsifiable because no one can identify who said it, when, or about what. We know the word will be objected to in Brussels; laundering imputes intent, and intent is the one thing we cannot see. Very well. Here is the same finding rendered in the institution’s own dialect: the Commission converted an interested party’s characterization into an institutional finding without demonstrating how it verified that characterization. The reader may choose whichever register offends less. The footnote is empty in both.

Nothing forced this. The confrontation over Belinda Balluku unfolded in public, on the record, for months. The Prime Minister told the Socialist parliamentary group he had refused her resignation three times. He announced, with Taulant Balla and Ulsi Manja, the Article 242 initiative. He answered the British and German embassies publicly after Parliament declined SPAK’s immunity request in March. The government fought the suspension before the Constitutional Court, where a four to four deadlock left the measure standing, and complied. Every element is datable, quotable, attributable. And here is the deeper reason the Commission left it all uncited. Written out as chronology, dated and quoted, the record convicts no one. It shows a resignation refused, a bill proposed, an embassy answered, a case argued before the proper court, lost, and complied with. That is institutional friction conducted through lawful channels, and institutional friction is not evidence of an attack on justice. It is evidence of a constitution in use. Cited honestly, the record proves the opposite of the charge, and the drafters, who read the same newspapers we do, could not have failed to see it. So the adjectives were not a shortcut past the evidence. They were a substitute for it. A tribunal reaches for characterization at precisely the moment the facts stop cooperating with the verdict.

Which tells us what this chapter is, because it is not a finding of fact. It is a drama, and the roles were cast before the evidence arrived. Read the verbs. Wherever the chapter records progress, agency vanishes into the passive voice: “legislative amendments have been adopted,” “the vetting process was fully completed,” Albania “continued implementing.” Wherever the chapter assigns blame, the author suddenly appears: “attempts continued by public officials and politicians, including at the highest levels.” The government of Albania is permitted to be the subject of a sentence chiefly when the sentence condemns it. And yet the villain wrote most of the good news. Every reform this chapter praises, the defamation amendments, the media ownership law, the Asset Recovery Office, the conflict of interest and lobbying and whistleblower legislation, the consensually adopted Rules of Procedure, the security budget for the courts, was drafted by this executive and voted by this majority. The government’s written contribution is the most cited source in the entire chapter; footnote after footnote reads “Albanian Government (2026).” The Commission harvests the accused for its facts, assigns its achievements to no one, and reserves authorship for the menace. Once the casting is understood, everything else follows. A villain needs no defined offense; the role is the offense. Interested testimony needs no corroboration; it confirms the script. A document that sorts every fact by pre-assigned role is not evaluating a country. It is staging one.

Now hold the method against the chapter’s own scorecard, item by item.

The report faults Albania for inconsistent evidence-based policymaking. Its own central political finding has no evidence attached: a claim of continuing personal attacks resting on air. The report faults Albania’s public consultations for opacity, late drafts, stakeholders unheard. Its own gravest claims are sourced to meetings no one can attend, held with interlocutors no one can name, bundled into a footnote that fuses a constitutional council with an anonymous crowd so that no reader can assign a single claim to a single mouth. The report records that Albanian courts fail to provide timely reasoning for their decisions. Its own decisive paragraph delivers verdicts with no reasoning at all: attack and undue influence appear as labels, never as conclusions argued from identified facts. The report faults Albanian authorities for refusing information requests behind generic references to confidentiality. “Country visit Albania, civil society organisations” is a generic reference, and it is not an isolated one. We counted. The country-visit citation appears in some twenty-five of the chapter’s 155 footnotes; the civil-society variant in eighteen; and in about fourteen of those, the unnamed organisations are the only source given, with no document beside them. Now look at which claims those fourteen footnotes carry: political interference in procurement still widespread, parliamentary oversight of the executive further shrunk, journalists intimidated including by government officials, a perception of impunity, discreditation and smear campaigns against civil society. Every statistic in this chapter has a document behind it. Nearly every judgment about the government’s conduct has a closed February meeting behind it. The division of labor is exact: the documents carry the facts, the anonymous channel carries the verdicts. And the report grades Albania on transparency and verifiability while its own annex cites, for the media risk scores deployed across an entire pillar, the 2026 Media Pluralism Monitor country report, marked in the Commission’s own bibliography with the words “not yet published.” The examiner demands that Albania show its work while resting its findings on a source the public cannot read.

There is a name for a proceeding in which the charge is never specified, the evidence never produced, the sources never identified, the accuser’s classification adopted without scrutiny, and the verdict delivered through the vocabulary of the accusation. Every standard in this report exists to abolish that kind of proceeding from Albanian life. The Commission has now conducted one against Albania’s constitutional organs, in writing, under its own seal.

Convicted them of what? Read the charge sheet. Parliament voted on an immunity request. Deciding immunity is Parliament’s enumerated function; a vote that can only go one way is not a vote, it is a ceremony. The government petitioned the Constitutional Court over the suspension of a minister. That court exists to hear precisely that dispute; the government argued, lost the round, complied. Majority deputies proposed amendments to the Criminal Procedure Code. Proposing legislation is what legislatures do, and the proposal sits in the open where it can be gutted, defeated, or struck down. The Prime Minister criticized SPAK’s decisions, harshly, repeatedly, on camera. Speech is speech, and the head of an elected government does not forfeit it because the subject is a prosecutor. That is the complete inventory. Not one unlawful act appears in it. No order defied, no ruling ignored, no prosecutor removed, no budget touched, no file seized. Every condemned act is the exercise of a constitutional power through its designated channel. What the report calls an assault on the justice system is the separation of powers doing the only thing it has ever done: branches contesting each other with the instruments the constitution puts in their hands. The friction is not a malfunction of the design. The friction is the design.

The chapter concedes this in a sentence its drafters did not reread. In the checks and balances section, the same report certifies that the Constitutional Court “continued to be effective in upholding institutional checks and balances,” deliberating on milestone cases about the division of power. Weigh that. The government objected to a judicial measure, carried the objection to the proper court, the court functioned, the system held. The document certifies the referee on one page and, pages earlier, calls the act of appealing to the referee an attack on the game. Both sentences carry the Commission’s signature. Both cannot be true.

The vocabulary gives the framework away. Everywhere criticism of SPAK appears in this chapter, it arrives pre-convicted: media scrutiny is “attempts to discredit,” political disagreement is “pressure,” public argument is “inappropriate rhetoric,” reaction to a court order is “direct attacks.” Now search the document for the sentence that does not exist: an acknowledgment, anywhere, that the conduct of a prosecutorial body wielding special courts, sweeping investigative powers, and deliberate insulation from domestic accountability could ever be a legitimate object of public debate. There is none. The category is empty by design, which means the classification of any statement as an attack is not a finding this report reached. It is the axiom it started from, imported wholesale from the interlocutors of footnote 27. Within this framework no event can ever register as the prosecution erring and being rightly criticized. An instrument incapable of recording that outcome is not monitoring an institution. It is consecrating one. Europe ran this experiment already. For a decade Romania’s anticorruption directorate was the Commission’s model, its critics dismissed as the self-defense of the corrupt, until the secret protocols with the intelligence service surfaced and it emerged that some of what Brussels had waved away as attacks on justice had been pointing at real abuses the whole time. The lesson was never that prosecutors are villains. It was that an institution spared all criticism compounds its excesses in the dark until they detonate. And to the answer already forming in Brussels, that lawfulness was Hungary’s alibi too: the abuse-of-lawful-powers doctrine carries a burden of proof. It requires showing the systematic closure of contestation, the capture of the referee, the engineered impossibility of losing. This reporting period shows one case, fought in open court, before a referee the Commission itself certifies, while SPAK goes on investigating the government’s inner circle and in June went on seizing property tied to the very former minister at the center of the storm. Governments executing that playbook do not keep losing rounds and complying with the losses. Invoke the doctrine. Then discharge its burden. The report does not try.

We do not claim the government’s motives were pure. We do not know them, we do not need them, and neither did the Commission, which is the point. We do not claim the immunity vote was wise; the Constitutional Court’s own doctrine treats immunity as a functional guarantee rather than a personal shield, and a rigorous report could have driven that argument home. We do not claim the danger to SPAK is imagined; eight protection details are eight protection details, and the men who threaten Albanian prosecutors are not parliamentarians but the defendants those prosecutors pursue, which is exactly why welding the two into one paragraph was rhetoric, not analysis. Our claim is this. The institution that examines Albania has failed its own examination on every criterion it set: evidence, reasoning, transparency, consultation, verifiability, due diligence toward its own sources, the specification of a charge before the delivery of a verdict. It took the testimony of interested parties at face value, adopted their vocabulary as its findings, cast the government of a candidate country as the villain of a story written before the evidence was gathered, and convicted that country’s constitutional organs of an offense it never defined, by a method its own report exists to condemn.

And there is a question this report obliges Albania to sit with, quietly, without slogans. For three decades the Union has been this country’s destination, and the destination was never the single market or the structural funds. It was a manner of exercising power: bound by rules, answerable to evidence, reviewable by someone. Albania did not vet its entire judiciary, dismiss more than half of it, and rebuild its courts from the studs in order to join a club whose examiner is exempt from examination. If the Union waiting at the end of accession is one whose institutions judge without evidence, source without names, and receive no judgment in return, an authority that speaks and is never spoken back to, a deity that issues verdicts and answers no appeal, then Albania is entitled to ask whether that is the Union it has been suffering toward. We believe it is not. We believe the Union worth joining is the one described in this report’s standards, not the one revealed in this report’s method. But the labor of keeping those two Unions the same body does not fall on Tirana. It falls on the institution that signed this document.

The first casualty is the standards themselves. For a decade Albania was taught that the rule of law means power exercised within rules that bind everyone, proven by evidence anyone can check. That teaching survives only as long as the teacher submits to it. A report that redefines the concept as power exercised toward outcomes Brussels prefers, issued in the Union’s name, does more durable damage to the rule of law in Albania than anything that happened in Parliament in March, because it teaches every actor in this region the oldest cynical lesson available: standards are for the examined, never for the examiner. The second casualty is SPAK, which needs criticism the way every lasting institution needs it, to answer, to correct, to mature, and which is instead being taught that its critics are illegitimate by definition. The Balluku file will close one day, in conviction or acquittal, and either outcome is survivable. What will remain is the norm this report has planted: that to contest the special prosecution is itself the offense. Someday, because it is staffed by human beings, SPAK will get a case wrong. On that day Albania will reach for the vocabulary of legitimate criticism and find that the guardians of its rule of law deleted it, year by year, footnote by absent footnote, in reports exactly like this one.

Albatros Rexhaj is an author, playwright, and analyst with a background in national-security studies and nearly three decades of experience with international organisations dealing with political and security affairs.