Dossier mode
The Hannibal Directive on October 7
The same investigation, restaged one beat at a time. Drive it with the arrow keys, space, or autoplay. Nothing is cut from the piece — long runs are split across frames. Read the full investigation or open the Israeli Influence on the United States hub.
The Hannibal Directive on October 7.
Israel's own largest newspaper reported the military invoked a protocol to stop soldiers being captured — even at the risk of killing them. An unknown number of Israelis died under Israeli fire that day.
Hamas killed the large majority of the ~1,200 and committed documented atrocities — not in dispute. Separately and also documented: in the chaos, the IDF used 'Hannibal' orders, and at least some Israelis were killed by their own forces. The official narrative omits it.
We grade what's established, refuse to invent a toll no one has (Haaretz says it's unknown), carry the IDF's position, and ask why a story Israel's press told loudly barely reached the West.
What the Hannibal Directive is.
A real Israeli military protocol (1986) authorizing heavy force to stop a soldier being captured — accepting that the soldier may be killed rather than taken hostage. Long secret, hugely controversial inside Israel, officially curtailed/abolished in its original form ~2016. Undisputed background.
It was used on October 7: 'not a single vehicle' back to Gaza.
July 2024, Haaretz (Israel's paper of record) reported Hannibal-style orders during the Oct 7 attack: invoked at three infiltrated army facilities, and a Southern Command division-wide order over military radio that 'not a single vehicle' cross back to Gaza — many carrying abducted Israelis. Existence of the orders = FACT (attributed to Haaretz); the toll the orders caused = uncertain, which the paper stressed.
Kibbutz Be'eri: tank shells fired on a house full of hostages.
The clearest single case: Brig. Gen. Barak Hiram told the NYT (Dec 2023) he ordered tank shells fired at a Be'eri house where Hamas held ~15 Israeli hostages, to end the standoff; only 2 of those hostages survived. Freed captive Yasmin Porat described heavy Israeli fire at the scene. Order and outcome = FACT; it illustrates commanders choosing overwhelming force where hostages were present.
The toll is unknown — and we will not invent one.
That Israeli fire killed some Israeli civilians/soldiers on Oct 7 is well-supported (the orders; Be'eri; Haaretz: 'many of the kidnapped people were at risk, exposed to Israeli gunfire'). But the number is genuinely unknown — Haaretz wrote it 'does not know whether or how many' were hit. So: self-inflicted casualties = PROBABLY TRUE, no figure assigned. The viral 'Israel killed most of its own' claim is NOT supported and is rejected — as firmly as the erasure of the deaths that did occur.
How this page is graded.
- FACT: the Hannibal protocol; that Hannibal-style orders were issued on Oct 7 (Haaretz); the Be'eri tank shelling (NYT / Barak Hiram); the near-silence of US coverage; the IDF's disputing/under-review position.
- PROBABLY TRUE: that Israeli fire killed some Israelis on Oct 7. UNKNOWN (no number asserted): how many — Haaretz says the toll is unknown.
- REJECTED: 'Israel killed most of its own on Oct 7' (unsupported) and the erasure of self-inflicted deaths (contradicted). Hamas killed the majority and committed documented atrocities — not exonerated; intent to kill hostages not asserted.
Why it matters.
The official account of Oct 7 is total: 1,200 innocents, murdered by Hamas, full stop. Most is true and terrible — but 'full stop' does work the evidence doesn't support, because some deaths came from Israeli fire under a doctrine Israel's own journalists have tried to investigate, and acknowledging it complicates a story deployed within hours to justify everything after. We're not the arbiters of how many; we insist the question is legitimate and that its disappearance from Western coverage is itself a finding. It sits with the rest of the october-7 cluster — the warnings ignored, the contested Al-Ahli attribution — under one through-line: the first casualty is the part of the record that doesn't fit.
Help us fill these lines.
- OpenAn official, published accounting of how many Israelis were killed by Israeli fire on October 7 — still not released.
- OpenThe scope and authorization of the 'no vehicle returns to Gaza' order: how far up the chain, and how widely applied.
- OpenVerification of specific individual cases (e.g., particular civilians reported killed by IDF fire) beyond the documented Be'eri shelling.