Trap Β· reasoning move
β Reading an incident story
Sounds like: βLook what they did β it's all over the news.β
A single violent incident, told in one frame, stands in for the whole conflict.
Signs
one incident presented as the whole picture; a headline with one direction and no before; numbers without a denominator or definition; footage that starts mid-scene.
Questions to ask
What happened just before? Who was injured, by whom β settlers, soldiers, residents, or both sides of a clash? What exactly is being counted, and how was it verified? What's the denominator, and what's the comparison β per capita, per year, against which other place? Do the other side's authorities confirm or dispute it?
What to do
hold the story until those questions are answered β both ways.
Example
'settlers attacked a village' for a clash where both sides threw stones and soldiers caused the injuries.
The mirror
'terrorists attacked a community' where settlers had entered the village first.
What it is.
Almost every violent incident in this conflict is instantly folded into one of two stories, 'settler terror' or 'Palestinian terror'. Headlines compress a sequence β provocation, clash, response β into one direction; datasets count different things (incidents including trespass and intimidation, or only attacks with casualties) and validate them differently (OCHA uses two sources for incidents against Palestinians, mostly one for incidents against settlers); and per-capita scale is rarely given (for a worked example with a denominator, see 'Reading the numbers' on 'Settler violence is a handful of extremists β the 'hilltop youth' β and Israel deals with them.'). Whether this conflict gets more coverage per death than larger wars elsewhere is a claim not yet measured here β . None of this makes incidents unreal; it makes single stories poor evidence of scale.
dataUN OCHA oPt, Settler-related violence dashboard β definitions and methodology β
Read (2026-09-28). The dashboard covers incidents 'since January 2017' ('Data going back to January 2006 will be added soon'); scope: 'violence, intimidation or trespass carried out by or against Israeli settlers and other Israeli civilians'; settler identity is sometimes 'presumed based on the circumstances of the case'; for incidents against Palestinians the perpetrator 'can be an Israeli settler, a member of the security forces who intervened in a settler-related incident, or both'; 'incident against Palestinians are validated by at least two independent and reliable sources, most incidents against settlers are entered based on one source'; intimidation-only figures 'are not comprehensive'.
Working it out: what one incident can carry, with two examples.
An incident can establish that something happened, who did it and how. It cannot establish how often such things happen; that needs a count with a definition and a denominator. Most incident stories fail by carrying the second claim on the first kind of evidence, and by cutting the sequence so that it runs one way. Two examples. Huwara, 2023: a Palestinian gunman killed two Israeli brothers, Hillel and Yagel Yaniv; settlers then rampaged through the town, at least one Palestinian man was killed and hundreds injured, and Israel's own West Bank commander called it a 'pogrom'; the army's inquiry found it had 'failed to deploy enough soldiers' (CNN). Told as 'settler terror', the story drops the killing that came first; told as 'Palestinian terror', it drops the rampage that followed. Both halves are documented; neither, alone, says how common settler or Palestinian attacks are (for counts, see 'Settler violence is a handful of extremists β the 'hilltop youth' β and Israel deals with them.'). Al-Ahli hospital, 17 October 2023: the UN humanitarian office relayed the Gaza Health Ministry's figure of '471 people β¦ killed', and the New York Times later said its own coverage 'relied too heavily on claims by Hamas, and did not make clear that those claims could not immediately be verified'. Human Rights Watch's later investigation found it 'resulted from an apparent rocket-propelled munition, such as those commonly used by Palestinian armed groups', and an Israeli air-dropped bomb 'highly unlikely', while saying 'further investigation is needed'. The first story carried a general claim (Israel bombs hospitals) on an incident whose facts were not yet known (see 'The correction nobody sees'). What both examples teach: hold the general claim until a count exists, and hold the incident until its sequence and its attribution are known, in both directions.
What the labels (Holds, Open, Arguable β¦) mean
pressCNN, 'Huwara: The settler attack on this West Bank town was called a βpogromβ by Israelβs military. Videos show soldiers did little to stop it' (15 June 2023) (2023) incident β
Read 2026-10-01 (the ToI link the card had is walled). Triggered by the killing of Hillel and Yagel Yaniv of Har Bracha by a Palestinian gunman; 'at least one Palestinian man dead and hundreds of others injured'; the West Bank commander's 'pogrom' and the IDF 'had not been sufficiently prepared'; IDF inquiry: 'failed to deploy enough soldiers'; Halevi's statement; CNN's video analysis: soldiers fired tear gas at Palestinians throwing rocks, fire trucks stopped; the IDF's answers (forces 'required to act to stop the violation'; additional forces arrived 'after the riot'); several arrested, 'All were released without charges'; no one charged over Sameh Aqtash's death. Re-read 2026-10-12: photo caption, 'vehicles torched by Israeli settlers in an attack on Huwara on February 26, triggered by the killing of two Israelis'. Re-read 2026-10-13 (CNN): Halevi, 'This is a severe incident that took place under our responsibility and should not have happened'.
advocacy / viewHuman Rights Watch, 'Gaza: Findings on October 17 al-Ahli Hospital Explosion' (26 November 2023) (2023) finding β
Checked: 'resulted from an apparent rocket-propelled munition, such as those commonly used by Palestinian armed groups'; 'a large air-dropped bomb, such as those Israel has used extensively in Gaza, highly unlikely'; 'further investigation is needed'.
primary documentOCHA oPt, 'Escalation / Hostilities in the Gaza Strip and Israel' flash updates nos. 1β15 (7β21 October 2023) (2023) aggregate β
Read 2026-09-29 and 2026-09-29 (UN humanitarian office; relays Gaza MoH figures). No. 1 (7 Oct): MoH '198 Palestinians had been killed'; no. 7 (13 Oct): 'In total, 1,799 have been killed'; nos. 1β4, 6, 7, 9, 10 give no cumulative Gaza children figure in their text (5 and 8 not loaded). No. 11 (16 Oct): 'cumulative fatality toll in the Gaza Strip to 3,000, including at least 853 children'. No. 12 (18 Oct): 3,478, 'including at least 853 children (as of 17 October)'; al-Ahli '471 people β¦ killed', per MoH. No. 13 (19 Oct): 3,785, 'including at least 1,524 children'. Nos. 14β15: '60 per cent' then '62 per cent of the fatalities being children and women'. Early updates use the slug 'escalation-gaza-strip-and-israel-flash-update-N'.
pressNew York Times, Editors' Note on al-Ahli hospital coverage (23 October 2023), as reported by The Times of Israel (2023) incident β
Checked (via The Times of Israel's reproduction, 23 Oct 2023): coverage 'relied too heavily on claims by Hamas, and did not make clear that those claims could not immediately be verified'.
Why it works on us
- The incidents are often real and terrible.
- Video feels like proof.
- Numbers from official bodies look neutral.
Where this trap shows up
Your side does it too β look at both columns.