What Happened
On June 17, 2026, the Kula Penal Court of Peace — a first-instance criminal judgeship in Manisa province, a jurisdiction of roughly 50,000 people — ordered Turkey's telecoms regulator to block nationwide access to Change.org, the global petition platform used by an estimated 22.6 million people in Turkey. The Information and Communication Technologies Authority (BTK) implemented the order within days. Neither the court nor BTK publicly disclosed what content triggered the block (bianet; Stockholm Center for Freedom).
Change.org removed the contested content and filed a formal appeal on June 24. The same court reversed itself on July 9, restoring access after roughly three weeks. The episode is unremarkable by the standards of Turkish internet regulation — and that is precisely the point.
The Legal Mechanism
Law No. 5651, Turkey's foundational internet statute, gives criminal peace judgeships (sulh ceza hakimliği) — single-judge courts originally designed to rule on pretrial detention and search warrants — the power to order nationwide access blocks. Article 8 of the law catalogs the offenses that can trigger a block, and BTK's own guidance confirms that once a blocking decision is transmitted, access and hosting providers must comply on an expedited basis (BTK, Türkiye'de İnternet Hukuku). A separate BTK process lets an affected party petition for reversal — the exact route Change.org used (BTK, Erişim Engelleme Kararlarına İtiraz).
The Steelman
It's worth stating the strongest version of the case for this system before criticizing it. Turkey's blocking regime is nominally judicial, not purely executive — a court, not a ministry, issued the order. It is also, at least formally, reversible: Change.org contested the order, removed material the court found objectionable, and got the block lifted within three weeks through an ordinary appeal channel. A country wrestling with defamatory or unlawful content hosted on a foreign platform with no local office needs some enforcement lever, and a narrowly targeted order aimed at specific content — rather than, say, a permanent ban — is in principle a more proportionate tool than the blanket platform bans some other jurisdictions have floated.
Why the Steelman Doesn't Hold Up at Scale
The trouble is that Turkey's version of this mechanism produces outcomes that look nothing like narrow, proportionate content moderation. First, opacity: no public authority stated what content on Change.org justified blocking the entire platform for the whole country. Rights groups monitoring these cases note this is standard practice, not an exception — courts applying Turkey's blocking provisions "rarely assess the context or public interest value of the content at issue," relying on boilerplate national-security and public-order language instead of case-specific reasoning (IPI).
Second, scale. The Change.org case is not an isolated data point — it is one entry in a system that blocked 1,264,506 domains and websites by the end of 2024 alone, per the EngelliWeb annual report from the Freedom of Expression Association, with 852 separate institutions and judgeships issuing over a million blocking decisions (İFÖD EngelliWeb 2024). Freedom House scores Turkey 31 out of 100 on internet freedom, rating it "Not Free" and among the most restrictive environments in Europe (Freedom House, Freedom on the Net: Turkey). A mechanism exercised a million times a year by hundreds of local courts is not a narrow judicial safeguard — it is an administrative default setting.
Third, the remedy is disproportionate to the harm even when the underlying complaint is legitimate. A court with jurisdiction over a district of 50,000 residents can, on its own initiative, cut off a platform for the entire country of roughly 85 million people. If specific content violates personality rights or defames someone, Law No. 5651 already permits URL-level takedowns — the tool Change.org ultimately used to get unblocked. Reaching instead for a whole-platform block, then disclosing nothing about the reasoning, converts a content dispute into a national access event with no public accountability trail.
The Proportionate Path
None of this requires Turkey to abandon judicial content review — some form of it exists in most democracies, including the EU's Digital Services Act. But proportionate regulation means blocking orders should be content-specific by default, reasoned in writing and published, and appealable to a court with actual subject-matter competence rather than a single duty judge handling detention hearings. Platforms operating in Turkey — and the businesses, campaigners, and ordinary users who rely on them — deserve to know why access disappeared before it does, not three weeks after.
Until then, cases like Change.org's will keep recurring, not as scandals but as routine administration — which is the more troubling outcome.