On this page
Key facts and issues at stake
DEMOS is registered outside the European Union, so it could not ordinarily bring a complaint to the European Ombudsman. In July 2024, the Ombudsman, Emily O’Reilly, opened an inquiry on her own initiative instead, in the public interest, given the gravity of the human-rights concerns. A year later, under her successor, Teresa Anjinho, the inquiry was closed (Decision of 8 July 2025, case OI/4/2024/MIK).
The Decision found one fault: the Commission made its free-of-charge condition explicit only after the study was complete. It called this “regrettable” and suggested clearer communication in future.
It found nothing else. The words in which the complaint was made — censorship, academic freedom, human rights — do not appear in it. Neither do the citizens of Guinea-Bissau. No remedy was offered.
An inquiry that looked away
- One side heard. The inquiry team met the Commission. DEMOS was never formally heard, and no official who ordered the halt was questioned.
- Answers kept from DEMOS. The Commission’s replies to the inquiry’s four questions informed the Decision. DEMOS never saw them.
- The wrong instrument. By the Commission’s own criteria, the study called for a grant: the researchers designed the work and bore most of its cost. It was run instead as a service contract — the choice that gave the Commission ownership of the study. The inquiry asked about this, accepted the Commission’s answer, and looked no further.
The evidence was in the file. It was not weighed.
We hold the European institutions to their own standards.
The papers
The Price of a Relationship
Censorship, Democratic Erosion, and the Limits of EU Accountability
Our response to the Ombudsman. The main paper.
Democratic erosion rarely announces itself. It spreads — beyond borders, through institutions, one quiet accommodation at a time. The paper documents how the EU’s commitments to freedom of expression, academic independence and democratic accountability were set aside in West Africa, and how the institution charged with preventing that ratified it instead. The Commission suppressed, for political reasons, a research initiative it had contracted in the service of democracy. The European Ombudsman, after an unusual own-initiative inquiry, found no maladministration. On ample evidence and sustained reasoning, the paper argues that the Decision was fundamentally flawed, and that it sets a worrisome precedent.
Contents
- Part I — The Original Wrong1. The Censorship: A Political Act, Not an Administrative Irregularity2. The Human Rights Dimension and Its Abandonment by the Inquiry
- Part II — A Compromised Investigation3. A One-Sided Investigation: The Procedural Failures4. The Evidence That Was There and the Inferences That Were Not Drawn
- Part III — Conceptual Error5. The Public Goods Fallacy: A Category Error With Systemic Consequences
- Part IV — Impact6. The Damage to a Project, Its Community, and Its Potential7. Guinea-Bissau, Autocratization, and the Cost of Acquiescence8. The Spillover: What Harms Democracy Abroad Harms Democracy at Home
- Conclusion · Coda: A Path Forward
Contents
- Part I — The Original Wrong1. The Censorship: A Political Act, Not an Administrative Irregularity2. The Human Rights Dimension and Its Abandonment by the Inquiry
- Part II — A Compromised Investigation3. A One-Sided Investigation: The Procedural Failures4. The Evidence That Was There and the Inferences That Were Not Drawn
- Part III — Conceptual Error5. The Public Goods Fallacy: A Category Error With Systemic Consequences
- Part IV — Impact6. The Damage to a Project, Its Community, and Its Potential7. Guinea-Bissau, Autocratization, and the Cost of Acquiescence8. The Spillover: What Harms Democracy Abroad Harms Democracy at Home
- Conclusion · Coda: A Path Forward
What public good? What public interest?
- The category error. Public funding is not public access. A good is public when no one can be kept from it. A study kept from its public fails that test.
- What “public” means. Bobbio, Arendt and Habermas agree: what is public is visible, shared and open to deliberation — not what an institution owns.
- The EU’s own policy contradicts the Commission. Its documents may be reused by anyone, even commercially. Its research grants leave the results with the researchers and pay for open access. Only the studies it commissions abroad can be withheld from the public.
- The public interest. It lies in the widest circulation of knowledge. A rule that lets officials lock a study away, in the name of the “public good”, serves their discretion, not the public. It is now taught to the officials who manage such studies.
The Devil Is in the Details
How the Ombudsman Overlooked What Was at Stake in the Voices of the People Case
A companion text: the Decision, read closely.
The companion text reads the Decision line by line, against the record. Its thesis: the Decision did not refute the case for academic freedom; it dissolved it. A complaint about censorship, fundamental rights and the voice of a people became a contract dispute. It happened through many small moves, each with a plausible explanation, all pointing the same way: a date moved forward, agreements recast as favours, a hedge that closes the case, the Commission’s portrait of itself accepted whole, an explanation only one side saw. Of the 21 positions the Commission set out before the inquiry team, the Decision adopts 15 in the Ombudsman’s own voice, and rejects none.
Academic freedom at risk
The Decision harms academic freedom in two ways.
First, it sets a bad precedent. A case about censorship was turned into a contract dispute, and the harm done was left without remedy.
Second, it opened the door to new risks. The Ombudsman’s only suggestion was that the Commission communicate its conditions more clearly. In December 2025, the Commission told the Ombudsman how it would act on her suggestion: it would teach its staff that a study’s free distribution is “mandatory because the study was produced with public funds and is therefore a public good”. A study can now be withheld from the public, officially, in the name of the public good. The rule would not have prevented what happened to the Voices of the People study. It prescribes it.
A case meant to address a violation of academic freedom produced a result that makes further violations more likely.
From Bissau to Brussels
How a case of censorship became a rule
- 2021–2023The Commission censors the Voices of the People study, blocks dialogue and refuses mediation.
- 2024DEMOS turns to the European Ombudsman: censorship, and the violation of fundamental rights. The Ombudsman opens an own-initiative inquiry on those grounds.
- 2024–2025A one-sided inquiry: one meeting, held with the Commission; DEMOS never formally heard; key evidence left unexamined.
- Jan 2025The inquiry puts four sharp questions to the Commission. The answers, which informed the Decision, are never shared with DEMOS.
- Jul 2025After a change of Ombudsman, the case is closed as a contract dispute. The Commission is never asked to answer for the censorship. No remedy.
- Dec 2025Following the Ombudsman’s suggestion, the Commission undertakes to teach its staff the “public good” rule, for every study it commissions abroad.
- NowThe rule that buried one study can be used against the next.
What harms democracy in Guinea-Bissau erodes it in the European Union.