Eighty percent.

The question

A number this large should generate coverage in every language that touches Iran. It did not. Amnesty International published its annual death penalty report on May 18, 2026. The headline figure — 2,707 executions worldwide in 2025, the highest recorded since 1981 — was followed immediately by a second: Iran alone carried out 2,159 of them. One country. Eighty percent of the global total.

Our ingest data for the week of May 18–24 shows 4 outlets carrying the story. All four operate in English. Zero in Persian. Zero in Kurdish. Zero in Arabic. Zero in Turkish.

This is not a coverage gap in the ordinary sense. A story about Iran, in which Iran is the overwhelming protagonist, generated no coverage in the language most Iranians read. The question is not what the report said — Amnesty's documentation is thorough. The question is what the distribution of coverage reveals about who is permitted to name what Iran is doing, and from where.

The source topology

This brief draws on 4 primary sources and 7 contextual sources.

Amnesty International (English) — transnational / human rights organization / London. The primary source document, published May 18. CBS News — English / US mainstream / corporate / New York. Covered the report as international news. Democracy Now! — English / independent / left-progressive / New York. Featured Raha Bahreini, Amnesty's Iran researcher, in a live interview May 20. Iran International (English edition) — English / exile-Iranian / anti-regime / London / reportedly Saudi-funded. Covered it in its English edition only — not its Persian edition.

Contextual: Radio Farda (Persian / exile / anti-regime / Prague — RFE/RL's Persian service. Published an interview with Raha Bahreini in Persian — the only Persian-language coverage our engine found. Diaspora press, not domestic). BBC Persian (Persian / state-funded UK / editorially independent / London. A social post on the report. Diaspora-facing). Iran HRM (English / exile / human rights monitor). IranHR.net (English + Persian / independent / Oslo-based. Covers individual execution cases; no consolidated Amnesty report piece this week).

State Persian press monitored — zero coverage found: IRNA (state-owned, Foreign Ministry wire). Tasnim (IRGC-aligned). Fars (semi-official). Mehr (state-adjacent). The apparatus that reports daily on judicial proceedings, drug enforcement, and national security published nothing on the Amnesty report.

No Kurdish-language source. No Arabic-language source. No Turkish-language source. No domestic Persian-language source.

The discourse map

The structure of the silence

The most analytically significant fact about this week's coverage is not what was published but what wasn't. Iranian state press covers drug enforcement as a public safety success. It covers executions — individually and in aggregate — when framing them as legitimate state action against criminals and enemies of the state. Previous Amnesty death penalty reports have been addressed by state media, typically to dispute the methodology or deny the figures.

This year: nothing. The scale of 2,159 — a number so large that denial becomes incoherent — appears to have produced a different response. Not refutation. Absence.

Who covered it and from where

Every outlet that covered the Amnesty report operates outside Iran. Iran International, Radio Farda, BBC Persian — exile or foreign-state-funded outlets whose audiences are primarily diaspora. They are read in Tehran on VPN, if at all. The people most directly affected — Iranian citizens, families of those executed, communities in Sistan-Baluchestan and Kurdistan where execution rates are highest — have no domestic press carrying this number.

Democracy Now!'s interview with Raha Bahreini is the most analytically rich piece in our corpus. Bahreini named the mechanism: the Iranian authorities use the death penalty as a tool of political repression to create an atmosphere of terror and fear. She also named the composition: roughly half of the 2,159 are drug-related offenses, disproportionately falling on Baluch, Kurdish, Arab, and Turkmen minorities. Baluch people — 2–6 percent of Iran's population — accounted for approximately 32 percent of drug-related executions in some months. Kurdish people are similarly overrepresented.

The Kurdish press absence

Kurds are disproportionately executed in Iran. Kurdish political organizations document this systematically. ANF News, Rudaw, Kurdistan24 cover Iranian state repression of Kurdish communities regularly. None of them, in our corpus, carried the Amnesty report this week.

This absence is distinct from the state-Persian absence. Kurdish press is not suppressing the story. It appears to be treating the Amnesty annual report as a non-event because the underlying facts — Kurdish executions in Iran — are already being covered case-by-case throughout the year. The Amnesty report aggregates what Kurdish press has been reporting in fragments for years. The aggregate doesn't register as new information.

The Arabic press gap

Al Jazeera covered Iran extensively this week. Its silence on the Amnesty report is consistent with a pattern: its Iran coverage prioritizes geopolitical events — US-Iran tensions, the ceasefire, regional alignment — over human rights documentation. The Amnesty report is a human rights document. It does not fit the geopolitical frame Al Jazeera applies to Iran.

Asharq Al-Awsat (Saudi-aligned, London) covered the report briefly in its English edition but not its Arabic edition — a platform asymmetry that mirrors Iran International's English-only coverage. The story that is politically sensitive in the primary language runs in the secondary language for international audiences, and not the other way around.

The English frame

English-language coverage treated the Amnesty report as a human rights story with a numerical hook. CBS News ran the record-high figure. Democracy Now! ran the political repression argument. The Algemeiner focused on the Israel angle — Amnesty's documentation of Israeli executions in the same report, vastly fewer but politically charged for its readership. The frame across all four outlets is accurate. What it doesn't carry: the structural argument about why this number went unreported in the language of the country responsible for 80 percent of it.

The cross-dimensional synthesis

The coverage gap on the Amnesty Iran execution report is not a coincidence of news cycles. It is a structural outcome produced by two reinforcing mechanisms.

The first is censorship pressure × state alignment. Iran's domestic press operates under documented threat of closure, editor detention, and judicial action. A story that would require a domestic outlet to print "2,159 executions — the highest since 1981" under its masthead is a story no domestic outlet will run. The silence is rational survival behavior, not editorial oversight.

The second is audience targeting × geographic origin. The outlets that did cover it reach diaspora audiences, not domestic ones. The coverage exists, but it exists in a location where the people most affected by the story cannot easily access it. The gap between where coverage appears and where the affected population lives is the accountability deficit this brief is naming.

Censorship pressure is the most direct dimension doing analytical work. Iranian domestic press ran nothing — not a refutation, not a counter-statistic, not a dismissal as anti-Iran propaganda. That specific form of silence — after years of addressing previous Amnesty reports — signals that the 2,159 figure is too large to engage with defensively.

Language corpus × geographic origin is the second. Persian-language coverage exists, but only from outside Iran. The language is present. The geography is absent. The people inside Iran who would need to read "2,159" in Persian, published domestically, to have a public reckoning with that number are not being reached by outlets in Prague and London.