Daniel's been watching the Venice coverage closely, and he's written in with something that's been nagging at him. The NAZA documentary, the one that won the Special Jury Prize at Venice, produced by The Guardian, directed by Yuval Abraham and Rachel Szor. It alleges the Israeli army used AI-assisted targeting in Gaza and tolerated civilian casualty figures that, if true, would be extraordinary. And yesterday, October first, the IDF published a ten-page English rebuttal calling the film distorted, misleading, and disingenuous.
Ten pages. That's not a press release, that's a legal brief.
It reads like one. And Daniel's not asking us to relitigate the strikes. He's asking about something narrower. The IDF says it was never offered the chance to see the material or respond on camera before publication, that basic journalistic best practice wasn't followed, and that it took weeks before a representative could even watch the full film. Daniel says he agrees instinctively that this seems improper. But he wants to know what the generally held standards actually are for the right of response when filmmakers allege wrongdoing. Who must be contacted, when, and in what form. And how this case measures against them.
So the question is whether the IDF was treated unfairly, procedurally.
Right. And before we can judge that, we need to know what the rulebook actually says.
Let's set the scene first, because the timeline matters. NAZA is eighty minutes. Shot in secret on Tel Aviv rooftops over three years. Built on twenty-four anonymous Israeli insiders, mostly described as Unit 8200 intelligence personnel. It premiered at Venice on September tenth and eleventh. It's set for release in fifty-six European territories, and it'll stream free worldwide after the theatrical run.
And the rebuttal landed October first. Nearly a month after Venice.
Which is itself part of the story. The IDF says the delay happened because the filmmakers never gave advance notice of the claims and never provided a copy of the film, despite an explicit request. So the army says it had to rely on a detailed account by a representative who attended a public screening, plus other public accounts and the filmmakers' own public statements.
That's the procedural charge in one sentence. And here's what I want to do with this episode. This is not a debate about whether the strikes were lawful. That's contested, the research is thin in places, and neither of us is in a position to adjudicate it. The narrower question is what the right of response actually requires, and whether the IDF's complaint measures up against it.
And the asymmetry we're going to exploit is this. The IDF's rebuttal bundles a valid procedural complaint with an invalid one. Separating them is the whole game.
So let's start with the baseline obligation. What does every major code actually say?
The BBC Editorial Guidelines are the clearest on this. Offering a right of reply to those subject to significant criticism or allegations of wrongdoing is a fairness obligation under the Ofcom Broadcasting Code. It's not a courtesy. It's not a nicety. It's a duty.
And the ABC Editorial Policies phrase it as an obligation to make reasonable efforts in the circumstances to provide a fair opportunity to respond. Where allegations are made about a person or organisation.
The Society of Professional Journalists Code says journalists should diligently seek out subjects of news stories to give them the opportunity to respond to allegations of wrongdoing. Diligently. That's the word.
Channel 4's compliance guidance puts it flatly. Providing a right to reply is a cornerstone of responsible journalism.
So the baseline is established. If you're going to accuse someone of something serious, you put it to them first. That's not controversial. Every code agrees.
Now here's the nuance the IDF's complaint glosses over. Format is flexible.
The BBC guidance is explicit. There is no prescribed format that a right of reply should take. An interview, a written statement, or simply a phone call whose response is reflected fairly. Any of those satisfies the standard.
So we weren't offered the chance to react on camera is a format complaint. And format complaints don't establish a breach. If the filmmakers had called the IDF press office, read them the allegations, and printed their denial, that would satisfy the code. They didn't have to put anyone on camera.
Right. The obligation is to give the subject an informed opportunity to respond. How that response is captured is a production decision, not an ethical requirement.
Second nuance. There is no requirement to hand over evidence.
BBC again. Respondents should be given enough information and detail about the arguments and allegations to understand them and give an informed response. So long as that is achieved, there is no requirement to make copies of evidence available or to show the subject any secret filming.
Which is a sentence the IDF's complaint runs straight into. We weren't shown the film is not, by itself, a breach of the standard.
The ABC says the same thing in different words. In most cases there is no need to provide access to the raw material on which allegations are based.
So now we can apply the benchmark point by point. What's the IDF's strongest argument?
The strongest point is the no-contact-before-publication one. The filmmakers apparently did not contact the IDF at all before publication to put the allegations to it. That maps directly onto the baseline obligation every code imposes. If that's accurate, it's a real breach.
And the weakest?
The demand to see the material or a copy of the film. That's explicitly not required by BBC or ABC guidance. The not offered the chance to react on camera framing is a format question, and format is flexible. So the IDF's rebuttal is doing something clever. It's bundling a valid complaint with an invalid one and letting the invalid one borrow the valid one's authority.
Hold on. Say that again, because I think that's the move.
The complaint reads as one grievance. We weren't contacted and we weren't shown the film. But those are two different things. The first is a breach. The second is not. And by stating them together, the rebuttal makes the second sound as serious as the first.
That's the kind of thing that works in a press release and falls apart when you read the actual guidance.
Which is why Daniel's instinct is right but incomplete. He says it seems highly improper. Part of it is. Part of it isn't. And the part that isn't is doing a lot of rhetorical work.
Let's talk about timing, because that's the part that cuts both ways.
The IDF says it only watched the full film in recent days, weeks after publication, and that this is why its response was delayed. That's consistent with the filmmakers not having provided a screener. If you don't send the subject a copy, they can't watch it before release.
But.
But the IDF also says a representative attended a public screening earlier, on which it partly relied. So if the IDF could attend a screening, the weeks to watch it framing is partly a choice, not purely the filmmakers' fault.
That's a real tension. You can't say we couldn't see it and also say we sent someone to see it.
You can, actually, if you're precise about it. Attending a public screening and watching the full film are different things. But the rhetorical effect is that the delay sounds like the filmmakers' fault when it's at least partly the IDF's own scheduling.
And conversely, if the filmmakers had sent a screener, the IDF's response could have come before or alongside release. So the timing isn't a clean indictment of either side.
There's one more standard worth quoting directly, because it reframes the whole dispute. The ABC says: what is a reasonable time will be determined by looking at the situation from that person's perspective, not the perspective of the media.
From the subject's perspective.
Time allowed depends on complexity, prior familiarity, the subject's resources, and public-interest urgency. So the question isn't whether the filmmakers thought they'd given enough time. It's whether the IDF, from its own vantage point, had a fair opportunity.
And the IDF is not a resource-poor subject. That's the thing. This isn't a small charity getting ambushed by a local paper. It's a military with a press office, legal advisers, and a communications apparatus.
Which cuts against the delay argument. A well-resourced subject has less standing to say it needed weeks, especially when it also had a representative at a screening.
So where does that leave the IDF's complaint overall?
Partly well-grounded, partly not. No contact before publication is a real breach of every major code. No access to the film is not required by any code. The on-camera framing is a format question, and format is flexible. And the timing complaint is weakened by the IDF's own account of attending a screening.
So that's the framework. Now let's apply it to the harder part, which is what happens when the sources themselves can't be checked.
The anonymity tension. The film rests on twenty-four anonymous Israeli insiders, mostly Unit 8200 personnel. The IDF argues this is a structural problem. The film does not reveal who is making the claims, the officers' identities are obscured, or even which specific attacks they are referring to.
And that's a real objection. If you can't see who's talking, you can't assess their vantage point. You can't check whether they'd know what they claim to know.
But anonymity is what makes whistleblower testimony possible at all. If you name your sources in a military intelligence unit, you end their careers, you may end their freedom. The standards guidance acknowledges this tension. The BBC says on occasion there may be a balance to be struck between providing information to the subject of an investigation and protecting sources.
And the ABC says it's preferable if the person is told the source of the allegations, subject to the need to protect source confidentiality. Preferable. Not required.
So this is the tightrope Daniel flagged. And it's tight, not a gotcha. The filmmakers have a legitimate interest in protecting their sources. The IDF has a legitimate interest in knowing who's accusing it and on what basis. Both of those are real. And the codes don't resolve the tension, they just say balance it.
Now here's the IDF's substantive counter to the insiders' vantage point, and I think this is the strongest thing in the whole rebuttal.
The IDF says the interviewees handled targeting data midstream. They saw only limited, initial stages dealing with intelligence, not the later planning and approval stages run by operational bodies with legal advisers.
So even granting their sincerity, the IDF says they lacked the vantage point to establish final strike decisions.
Which is a real epistemic problem with anonymous mid-process testimony. And it is not answered by the right-of-response framework at all. This is a separate issue. Even a procedurally perfect process wouldn't fix it.
Because the problem isn't that the IDF wasn't contacted. The problem is that the sources, by their position in the process, may not have seen the thing they're testifying about.
And that's the kind of objection that a right of response can't cure, because it's not about fairness to the subject. It's about the reliability of the evidence.
Let's do the specific factual rebuttals, because they matter for context even if they're not the procedural question.
The IDF says it has never planned, approved or carried out any strike in Gaza in which five hundred civilians, or anything close to that number, were expected to be killed. That's a flat denial of the film's central claim.
And on the AI point?
Artificial intelligence has never independently selected targets. That's the IDF's line. And on Lavender specifically: Lavender is not a list of targets approved to strike, but a database that assists intelligence analysts.
So the framing is that the system is assistive, not autonomous. That's a meaningful distinction if it holds.
And the twenty and two hundred figures. The IDF says those are approval-authority levels, not permitted casualty numbers. No approval-authority threshold of two hundred casualties has ever applied or existed.
There's a nice framing in the Jerusalem Post's coverage. The IDF says documents citing hundred-to-one or twenty-to-one ratios are maximums that cannot even be considered, analogous to statutory maximum sentences that courts rarely impose.
A twenty-to-one cutoff might yield an actual five-to-one or seven-to-one strike. The maximum is a ceiling, not a target.
Which is a plausible reading of how approval thresholds work in any bureaucracy. You set a ceiling so nobody goes above it, and in practice you come in well below.
But here's the thing the Jerusalem Post itself notes, and it's worth sitting with. The IDF has still failed to give an on-record discussion of its actual combatant-civilian casualty ratios.
Right.
It has not truly addressed the vast cumulative number of civilian casualties. So even a procedurally perfect right-of-response process would leave the substantive question open.
The IDF's rebuttal answers process and specific claims but not the aggregate. It says we never targeted five hundred civilians, we never let AI pick targets, Lavender is a database. But it doesn't say here is our actual ratio, here is the cumulative toll, here is how we account for it.
Which is the gap. And it's important because it shows that the procedural complaint is not a substitute for the substantive answer. Even if the filmmakers had done everything by the book, the IDF would still owe an answer on the aggregate.
The procedural fight is not the substantive fight. They're related but distinct, and conflating them is how one becomes a substitute for the other.
There's one more piece of context worth weighing. Abraham and Szor previously made the Oscar-winning No Other Land, which also drew intense Israeli official pushback.
There's a pattern in how Israeli institutions respond to these filmmakers.
A pattern is not an argument. But it is context for judging whether the IDF's complaint is a genuine procedural grievance or a rhetorical move. If the same institution has pushed back hard on the same filmmakers before, you'd want to look closely at whether this complaint is about the process or about the conclusion.
The honest answer is it might be both. The procedural complaint can be partly valid and also partly motivated. Those aren't mutually exclusive.
The no-contact point stands on its own merits. The pattern doesn't change that. But it does inform how much weight you give the framing around it.
There's one more thing. The research found no evidence of a direct on-record response from Abraham or Szor to the October first rebuttal.
The Jerusalem Post explicitly states it has requested a response from the NAZA filmmakers, implying none had been given at publication time.
As of the research cutoff, the filmmakers have not answered the procedural charge at all. They haven't said we did contact the IDF, here's the log. They haven't said we didn't, and here's why.
Which is a genuine gap. If you're accused of skipping a basic step, the obvious response is to show you didn't skip it. And that hasn't happened yet.
Which is where I want to bring in something that's been nagging at me this whole episode.
Hilbert: I agree with you.
...About which part?
Hilbert: The no-contact point being the strong one. The we weren't shown the film part is the kind of thing a subject says when it wants the process complaint to do the work the substance can't. But the thing that actually bothers me is the timeline. A representative attended a public screening, and the army still says it took weeks to watch the full film. That was always the tell.
The tell of what?
Hilbert: Of a delay that's a choice dressed up as a grievance. I spent a stretch of my life as a fact-checker for a regional news wire. Not glamorous. But it taught me one thing I've never forgotten. The difference between we were not contacted and we were not shown the material is the difference between a real complaint and a complaint that sounds real.
That's a clean distinction.
Hilbert: In that job, the standard internal rule was that if you couldn't reach a subject, you documented every attempt. Date, time, method. Because the documentation was the defense. If someone later said you never called, you had the log.
The question is whether the filmmakers have published any such log.
Hilbert: That's what I'd want to see. The IDF says they weren't contacted at all. The filmmakers haven't said anything. If there's a log, it settles it. If there isn't, that tells you something too.
You'd want to see it before you decided.
Hilbert: I'd want to see it. I suspect I never will. Anyway, the levels on the second mic are drifting. I'll fix it in the edit.
That's the piece we didn't have. Not a verdict, but a working journalist's instinct for which procedural complaints are real and which are performance.
The asymmetry Hilbert points to is the thing I keep coming back to. No log from the filmmakers. A detailed procedural complaint from the IDF. One side has shown its work and the other hasn't.
Which doesn't settle the substantive question either way. It just tells you where the burden currently sits.
Before we close, there's one detail from the research that didn't fit the main discussion. The IDF's rebuttal includes a line that's worth reading on its own. The moral burden is real, and the IDF does not dismiss it. The moral burden reflects the seriousness with which the decisions were made.
Which is an unusual thing to put in a procedural rebuttal. It's not a legal argument. It's an acknowledgment that the thing being described, whatever the specifics, carries weight.
It sits oddly next to the distorted, misleading, and disingenuous framing. One line is dismissive and the other is somber, and they're in the same document.
It reads like two different hands wrote it. The press office wrote the denial and someone else wrote the moral burden line.
Which is the kind of thing you notice when you read the whole thing instead of the pull quotes.
Here's the open question, then. The IDF's procedural complaint is partly well-grounded. No contact before publication is a real breach of every major code. And partly not. No code requires showing the subject the film or offering an on-camera response. The filmmakers have not, as of the research, answered the procedural charge at all.
The deeper unresolved tension is that even a procedurally perfect right-of-response process would leave the substantive question open. The cumulative civilian toll. The combatant-civilian ratios. The IDF's rebuttal answers process and specific claims but not the aggregate.
The right of response is a fairness mechanism, not a truth mechanism. It ensures the subject gets a fair hearing. It does not ensure the subject's answer is adequate.
Those are different things. And conflating them is how procedural complaints become substitutes for substantive ones.
Which is the thing to hold onto. Both sides in this dispute have an interest in blurring that line. The IDF benefits from treating a process complaint as if it answered the substance. The filmmakers benefit from treating the absence of a substantive answer as if it proved the procedural complaint was bad faith.
Neither of those moves is valid. The process question and the substance question are separate, and you have to answer them separately.
That's the landing. Thanks to Hilbert Flumingtop for producing. This has been My Weird Prompts. If you want to support the show, leave us a review wherever you're listening.
We'll be back soon.
See you tomorrow.
