This case attracts confident claims in both directions, and a surprising number of them are simply wrong. What follows is a plain check of the ones that come up most, with the accurate version attached. Several of them will be unwelcome to whichever side you started on. That is rather the point.

Claims about the airstrike itself

ClaimVerdictWhat the record shows
The children in the van were killedFalseBoth were badly wounded and both survived. Their page.
Nobody in the street had a weaponFalseAt least one man appears to be carrying a rocket-propelled grenade launcher.
Everyone in the street was a fighterUnsupportedTwo were Reuters staff with camera equipment. Most of the others were never identified either way.
Someone in the van was armedUnsupportedNo weapon appears in the footage of that phase. The van.
It was one attackFalseThree separate engagements over about forty minutes, with different facts and different law.
The crew knew they were shooting at journalistsNo evidenceNothing in the record supports it. The sight could not have shown press equipment.

Claims about the recording and the release

ClaimVerdictWhat the record shows
WikiLeaks hid the full versionFalseBoth the full recording and the short edit went online the same day. The two versions.
The edit is the only version most people sawTruePublishing both is not the same as both being watched. Almost all viewing was the short one.
The footage was doctoredNo evidenceIts authenticity has never been seriously disputed, including by the US government.
Nobody knew about the incident before 2010FalseA book published in 2009 described it and quoted the cockpit radio. Here.
Reuters was given the video and sat on itFalseReuters asked for nearly three years and never received it. The FOIA fight.

Claims about the legal outcome

Two mirror-image errors turn up constantly, and neither is right.

  • "A court found the airstrike lawful." No court has ever examined it. The only inquiry was an internal Army administrative investigation, which cannot make a legal ruling and cannot charge anyone.
  • "A court found it a war crime." Also untrue. No prosecutor anywhere has brought a case. Serious criticism from lawyers and human rights organisations is not a finding.

The accurate position is uncomfortable and easy to state: the question was never tested. That is the story.

The accurate position, stated plainly

No court has ever examined this airstrike. Not to clear it, not to condemn it. An internal Army inquiry looked at it in 2007 and closed it, and nothing since has reopened the question.

Claims about the people who published it

  • "Manning leaked only this video." No. The video went to WikiLeaks as part of a disclosure of hundreds of thousands of documents, including the Iraq War Logs.
  • "Manning was convicted of treason." No. The conviction was under the Espionage Act and other charges. She was acquitted of aiding the enemy, the gravest count. The verdict.
  • "Manning was pardoned." No. The sentence was commuted in January 2017, which shortens a sentence but does not undo a conviction. The difference matters.
  • "Assange was extradited to the United States." No. He was never extradited. He pleaded guilty in a US court in the Northern Mariana Islands in June 2024 and flew home. The plea deal.

Why these keep spreading

Almost all of them come from the same place: someone watched the short edit once, years ago, and has been arguing from memory ever since. Memory smooths a complicated forty minutes into one image, usually the most upsetting one, and then fills the gaps with whatever the person already believed.

Summaries of summaries

The rest come from summaries of summaries. An article cites an article that cites a forum post. Following that chain back to a record, as the sources page suggests, usually ends the dispute in about five minutes.

The check that ends most of these

Follow the claim back to a record. Most chains end at another website, and the ones that reach a document usually contradict the claim. Five minutes is normally enough.

Why the myths make the record easier to dismiss

Every myth on this page, in either direction, makes the story worse as a piece of evidence. Say the children died and someone will check, find they lived, and stop believing anything else you told them. Say nobody had a weapon and the first person who watches the footage will see one.

What is actually established

What is actually established is damaging enough on its own: two journalists killed, a man shot for stopping to help, two children hurt in the front seat, a building hit with three missiles and never accounted for, an internal inquiry that cleared everyone, and prison only for the people who made it public. None of that needs a single exaggeration to land, and every exaggeration makes it easier to wave away.

Questions people ask about this

Was the video edited to mislead?

The seventeen-minute version is edited and captioned, and it emphasises some phases over others. The full recording was published at the same time, so the material to check it against was always available.

Has any court ruled on the airstrike?

No. There has never been a criminal case. The only examination was an internal Army administrative inquiry, which cannot make legal rulings.

Were the men in the street insurgents?

Some may have been; at least one appears to be carrying a launcher. Two were Reuters staff. Most were never identified either way, and that uncertainty is part of the record.

Where this comes from

  1. July 12, 2007, Baghdad airstrike — Wikipedia
  2. Federal court dockets and filings — CourtListener
  3. Department of Defense FOIA reading rooms — US Department of Defense

Spotted something wrong? Tell us and we will fix it.