27MAY

Amnesty: 39 executions since war began

4 min read

15:19UTC

Amnesty International documented at least 39 political executions and more than 6,000 arbitrary arrests in Iran since the war began on 28 February, in a report dated 28 May. The repression kept wartime tempo throughout the Doha talks.

← Back to Brussels slips sovereignty law a third time Jump to analysis ↓

TechnologyDeveloping

Key takeaway

Iran's prisons kept wartime execution tempo through the Doha talks; the deal timetable never reached them.

Amnesty International, the London-based human rights watchdog founded in 1961, published the first comprehensive count of the war window in a report dated Thursday 28 May: at least 39 political executions and more than 6,000 arbitrary arrests in Iran since the war began on 28 February ¹. The 39 break down as 16 protesters, 9 dissidents, 10 people convicted of spying for the US or Israel, and 4 for armed rebellion ². Amnesty also logged an 88-day internet blackout, forced confessions aired on state media, and asset seizures targeting more than 750 individuals ³.

The tally sits on top of a register that had already passed 200 executions for 2026 by mid-May, when Ramin Zaleh and Karim Maroufpour were hanged at Naqadeh Central Prison without notice to their families . It also encompasses the 25-27 May cluster, including the execution of Turkish national Gholamreza Khani Shakarab on spying-for-Israel charges, over which NATO member Ankara issued no public protest .

The two clocks never synchronised. While negotiators in Qatar debated whether the Hormuz memorandum was "largely negotiated", Iran's judiciary and the Islamic Revolutionary Guard Corps (IRGC, the ideological military branch whose security operations underpin the arrests) carried executions at the same pace they ran before the Doha track opened. The diplomacy moved on a 60-day horizon; the prisons moved on a daily one. For the 39 already executed and the 6,000 detained, a briefing room discussing a tentative deal changed nothing.

Deep Analysis

In plain English

On 28 May, Amnesty International published its first full count of what Iran has done to its own people since the war started on 28 February 2026. Amnesty is a London-based charity that monitors human rights abuses by governments worldwide; it accepts no government funding so that it can report on any country without fear of losing support. The numbers: 39 people executed for political reasons, more than 6,000 arrested without proper legal process, and 88 days of near-total internet shutdown. The 39 executions break into four groups: 16 were protesters, nine were political dissidents, ten were people convicted of spying for the US or Israel, and four were convicted of armed rebellion. These executions happened while Iran was simultaneously negotiating with the United States over ending the conflict. That is not a coincidence Amnesty glosses over. The report documents that Iran's security apparatus kept running at wartime pace even as diplomats talked peace in Doha and Islamabad. A separate monitoring group called Hengaw, based in Norway and focused on Kurdish communities in Iran, has been tracking individual executions prison by prison. Hengaw's running register records a higher total than Amnesty's 39, because Hengaw counts all security-linked cases, while Amnesty counts only the 39 cases it has individually confirmed as politically motivated.

Deep Analysis

Root Causes

Iran's wartime judicial system has three structural features that produce accelerated execution tempos independently of any political instruction.

First, the 88-day internet blackout (documented in and partially lifted on 25 May) severed the connection between defence lawyers and their clients in custody. Under normal operating conditions, Iranian defence lawyers can monitor case status, file appeals, and alert families and international monitors.

The blackout eliminated this function, allowing cases to move from arrest to execution in under 50 days, as in Mojtaba Kian's case (arrested March 2026, executed 24 May), without any observable escalation visible to external monitors.

Second, wartime espionage charges carry mandatory death sentences under Article 183 of Iran's Islamic Penal Code, and the Supreme National Security Council's wartime classification of 'supporting the enemy' is broad enough to encompass any contact with foreign media, international organisations, or diaspora networks.

The 10 espionage-category executions in Amnesty's count were processed through the Revolutionary Court, which holds closed hearings, bars independent legal representation, and accepts intelligence service testimony without cross-examination.

Third, the IRGC's internal security role expanded dramatically after the 28 February strikes decapitated civilian state structures. The IRGC now runs both the physical detention estate and the intelligence services that generate the prosecution cases. This vertical integration between arrest authority and prosecution means no independent state body checks the evidence quality before cases reach the Revolutionary Court.

Escalation

The execution tempo has not slowed during the diplomatic window and may be accelerating. Amnesty's 39-figure for 90 days of war works out to roughly one political execution every 2.3 days. Hengaw's 25-27 May cluster, with five executions over three days , suggests the pace is not declining.

Iran's government is running a negotiating posture externally while running a consolidation posture internally; both can be sustained simultaneously as long as the internet blackout limits the information flow that would make the contradiction politically costly.

What could happen next?

Consequence
Amnesty's 28 May report provides the evidentiary foundation for any post-conflict accountability process: the 39 named execution categories and 6,000-plus arrest records constitute a documented dataset that UN Special Procedures and the International Criminal Court can draw on for referral or monitoring purposes.
Medium term · Assessed
Risk
If any peace agreement is signed before the internal execution campaign concludes, Iran's negotiating partners, particularly the European states involved in the 40-nation Hormuz coalition, will face domestic pressure to condition sanctions-relief implementation on cessation of political executions, creating a conditionality layer the current MOU structure does not address.
Short term · Assessed
Meaning
NATO member Turkey's silence over the 26 May execution of Turkish national Gholamreza Khani Shakarab (ID:3659), while Turkey serves as an active mediator in Iran-related diplomacy, signals that Ankara has calculated that mediation access is worth more than consular protest over a single citizen, a calculation that may be tested if further foreign nationals are executed.
Short term · Assessed
Precedent
The 88-day internet blackout, confirmed by Amnesty as the longest sustained wartime internet shutdown in modern history, establishes a template for information control during armed conflict that other authoritarian states have already begun studying, per Freedom House reporting.
Long term · Assessed

Source Landscape

This story draws on centre-leaning sources from United Kingdom

United Kingdom

LeftRight

Primary parallel: In the summer of 1988, Iran executed between 3,000 and 5,000 political prisoners, primarily Mojahedin-e Khalq and Marxist Fedayeen, in a three-month purge that coincided with the final phase of ceasefire negotiations ending the Iran-Iraq War. The Khomeini-era fatwa authorised the killings as a precondition for post-war political consolidation: the regime eliminated an organised opposition before the external threat ended, reasoning that a ceasefire would otherwise allow domestic dissidents to reconstitute.

The 2026 pattern is structurally parallel. Iran is executing protesters, dissidents, and wartime 'spies' at accelerating pace as ceasefire negotiations advance, in categories (16 protesters, 9 dissidents, 10 espionage-convicted) that mirror the 1988 categories of political opposition and foreign-linked agents. The lesson the Islamic Republic drew from 1988 was that wartime exception powers are most safely exercised before, not after, a political settlement.

Counter-parallel: During the Algerian civil war of the 1990s, the military government also conducted mass executions of Islamist prisoners through the war years, but the tempo increased after ceasefires rather than before them, as the state used post-conflict instability to clear prison populations. The 2026 Iran pattern is the inverse: the execution pace is highest during the active diplomatic window, suggesting internal consolidation is front-loaded rather than deferred.

Consensus view: UN Special Rapporteur on Human Rights in Iran and Iran Human Rights Monitor (IHR) director Mahmood Amiry-Moghaddam both assess that the 39 political executions documented by Amnesty International from 28 February to 28 May 2026 represent the fastest sustained wartime political-execution tempo Iran has recorded since the 1988 prison massacres.

IHR's methodology, which cross-references Hengaw's prison-by-prison documentation with Amnesty's category register, places the credibility interval at 35-42 executions, with Amnesty's 39 figure within the agreed range. Amiry-Moghaddam notes the 88-day internet blackout is structurally functional for execution concealment: families cannot verify, lawyers cannot communicate, and international monitors receive delayed and partial information.

Counter-view: Professor Ali Ansari (St Andrews University, Iran expert) argues that the divergence between Amnesty's 39-figure and the higher counts in Hengaw's running register reflects a methodological difference rather than a coverage gap. Amnesty counts only cases where it has independently verified the political motive for the execution, while Hengaw includes all wartime security-related executions including narcotics cases with a political-prisoner profile.

Ansari notes the same methodological divergence inflated apparent disagreement between human rights bodies during the 2019 crackdown, when Amnesty's confirmed figure (304) and the government's denials coexisted with Hengaw's higher estimate. The underlying consensus on direction and magnitude holds regardless of counting method.

Key tension: Whether the execution tempo reflects a deliberate IRGC-designed deterrence campaign timed to run alongside the diplomatic track. The 24 May execution of Mojtaba Kian for wartime espionage was announced on the same day as Trump's 'largely negotiated' deal statement, suggesting at minimum that Tehran's security apparatus treats diplomatic windows as cover rather than constraint.

First Reported In

Update #111 · US sanctions the strait its deal reopens

Amnesty International· 29 May 2026

Read original →

Causes and effects

Caused by

Naqadeh executes two Kurdish PDKI prisoners

Amnesty's 28 May 39-execution tally builds on the execution register that had already passed 200 for 2026 by mid-May, covering the same wartime political category.

Occurred 21 May 2026

Read story →

Iran hangs Turkish national, toll rises

The Amnesty report's war-window count encompasses the 25-27 May cluster of executions including Turkish national Shakarab, contextualising the pace of the repression campaign.

Occurred 26 May 2026

Read story →

This Event

Amnesty: 39 executions since war began

Iran's security state held its execution and arrest tempo through the diplomatic track, indifferent to whether the deal was close.

Led to

Iran's killing moves to the field

Amnesty's 28 May tally of 39 political executions since February established the judicial baseline; the 28-30 May extrajudicial field killings represent an escalation beyond the auditable court-based process.

Occurred 28 May 2026

Read story →

Different Perspectives

ASML / European tech industry

ASML's Q2 2026 guidance came in €300m below consensus as China DUV revenue collapsed 17 percentage points; the company's CEO wrote US export-control outcomes directly into 2026 guidance. European tech firms named on the USTR retaliation list alongside SAP, Siemens and Spotify face the same calculus: US trade exposure constrains what Brussels can legislate on their behalf.

France / Anne Le Henanff

Le Henanff chaired the G7 Digital Ministerial at Bercy on 29 May with CAIDA off the agenda, pivoting France's presidency to AI safety principles it had not designed the week around. France backs CAIDA but cannot override Berlin's tariff calculus, so the ministerial produced no new French-led commitment.

Germany / Federal government

Berlin's automotive sector faces up to $200bn in threatened US tariffs, a commercial exposure that dwarfs any benefit CAIDA's public-sector cloud rules would deliver to German digital firms. Federal silence inside the College of Commissioners functions as a block under consensus adoption rules without requiring a formal veto.

USTR / Ambassador Andrew Puzder

Puzder's public warning on 25 May that CAIDA is inconsistent with the EU-US trade framework was the first time Washington made its bilateral pressure visible before a Commission adoption vote rather than after. The USTR Section 301 determination on 24 July provides the enforcement backstop.

European Commission / Henna Virkkunen

Virkkunen framed the third slip as a procedural delay in finalising a 400-page text without addressing Puzder's trade-framework red line publicly. The Commission enforces existing law against Google while losing the legislative timeline on CAIDA, exposing an asymmetric position: enforcement holds; new sovereignty legislation does not.

OpenForum Europe / open-source community

The EUR 350m Sovereign Tech Fund has no Commission host, no budget line, and no commissioner's name attached six weeks after the April conference, while Germany is already paying maintainers to staff international standards bodies. The CRA open-source guidance resolves contributor liability but leaves the financial-donations grey area open with the 11 September reporting clock running.