Trap · reasoning move
◆ The argument that isn't meant to win
Sounds like: “Anyone who doesn't accept this is in denial.”
Some arguments convince one side completely and can never convince the other. When the other side rejects them, the rejection itself becomes the proof: 'they deny the obvious'. The argument keeps the loop going, whatever anyone intended.
Signs
the argument is framed in words the other side can only take as an insult or a denial of who they are. Its main audience is people who already agree. The other side's rejection is presented as proof ('see, they can't face it'). You have never met anyone it changed.
Questions to ask
Could this argument ever convince someone on the other side, and does its framing even try? What would a version look like that the other side could hear, and is anyone making it? Who gains when both sides stay angry?
Example
'Zionists are the modern Nazis': the criticism of the war it carries can be made without it, and only then has a chance of reaching Israelis.
The mirror
'There's no such thing as a Palestinian': the history it points to (national identities formed in the 20th century) can be told without denying people who are there. The mirror: both sides can apply this test to their own favourite lines first.
The effect is easy to see. An argument is framed in terms that only one side accepts: a comparison that wounds the other side, or a denial of the other side's existence. Its own side hears a truth. The other side hears an insult and rejects it, and that rejection is shown to the first side as evidence that the other side can't face the facts. Each round confirms both camps, and nobody on the other side has moved. This card describes the effect. It doesn't claim that anyone designs arguments this way. One explanation needs no designer. If arguments that please their own side get shared more, and those that could persuade the other side feel weaker to the people sharing them, selection alone would produce the effect.
Both sides have their versions. From the pro-Palestinian side: 'Zionists are the modern Nazis', said by a head of state in 2024 (see 'Zionists are the modern Nazis.'). It sounds like moral clarity to those who say it, and to Israelis like the inversion of their history. From the pro-Israel side: 'there is no such thing as a Palestinian people', said by an Israeli minister in 2023 (see 'There's no such thing as a Palestinian people — it was invented in 1964.'). It sounds like history to those who say it, and to Palestinians like erasure. Each is quoted by the other side as proof of what the first side is really like.
pressTimes of Israel, 'Jordan summons Israeli envoy over racist Smotrich speech, Greater Israel map' (2023) incident ✓
Speech in Paris, 19 March 2023; report of 20 March 2023 (read 2026-09-26).
Sometimes it is designed. The US Senate Intelligence Committee documented one case in detail. On 21 May 2016, Russia's Internet Research Agency used its 'Heart of Texas' Facebook page to call a 'Stop Islamization of Texas' protest at an Islamic centre in Houston. At the same time, its 'United Muslims for America' page called a counter-demonstration at 'exactly the same Islamic Da'wah Center'. 'In neither instance was the existence of a counter-protest mentioned', and the two crowds ended in 'confrontation and verbal attacks'. The operation cost $200 and was run from Saint Petersburg. More generally, its operatives were found 'simultaneously taking and attacking all sides of the arguments, all from the same internet protocol (IP) address'. That case was about American divisions. The same pattern has been found aimed at Israel: in its report for the first quarter of 2024, Meta said it had removed a network that 'originated in Iran and targeted Israel', with fake Israeli personas on the right, backing Ben-Gvir against Netanyahu, and on the left, posing as 'young liberal Israeli women' at anti-government protests. It shows that when an argument keeps both sides angry and neither persuaded, it is worth asking who benefits, without assuming that someone planned it.
primary documentUS Senate Select Committee on Intelligence, 'Russian Active Measures Campaigns and Interference in the 2016 U.S. Election', S. Rep. 116-290, Volume II: Russia's Use of Social Media (GovInfo, 2020) (2020) finding ✓
Read 2026-09-28 (scanned PDF; OCR checked against the page images). Vol. II p. 47 (report p. 117): Houston, 21 May 2016 — 'Heart of Texas' ('Stop Islamization of Texas') and 'United Muslims for America' ('Save Islamic Knowledge') events at 'exactly the same Islamic Da'wah Center'; 'In neither instance was the existence of a counter-protest mentioned'; 'escalated into confrontation and verbal attacks'; cost '$200'; run 'from the confines of the IRA's headquarters in Saint Petersburg'; John Kelly: 'kind of like arming two sides in a civil war'. Vol. II p. 54 (report p. 124): operatives 'simultaneously taking and attacking all sides of the arguments, all from the same internet protocol (IP) address'.
primary documentMeta, 'Quarterly Adversarial Threat Report, Q1 2024' section 'Israel-based network' (2024) finding ✓
Read (2026-09-28). 'found links to STOIC, a political marketing and business intelligence firm based in Tel Aviv, Israel. It is now banned from our platform'; 510 Facebook accounts, 11 Pages, one Group, 32 Instagram accounts; about 500 Page followers, ~2,000 Instagram followers; targeted the US and Canada; commented on pages of media and 'political and public figures, including US lawmakers'; 'supportive commentary about Israel' (incl. its military actions, per the report's summary table); criticism of Muslims and UNRWA; comments 'often met with critical responses from authentic users calling them propaganda'; some comments 'likely generated using artificial intelligence'. The report does not name who commissioned the operation. Iran-based network: 22 Facebook accounts, eight Pages, eight Groups, 23 Instagram accounts; 'originated in Iran and targeted Israel'; 'fictitious personas posing as Israelis'; 'supportive commentary about Israel'; support for Itamar Ben-Gvir while criticising Netanyahu; accounts that 'posed as young liberal Israeli women and posted about anti-government protests'. Re-read 2026-10-12: the comments 'included links to the operation's websites and were often met with critical responses from authentic users calling them propaganda'.
Working it out: designed, or by itself?
The question: when an argument keeps both camps angry, is that designed or does it happen by itself? The page offers both explanations; the evidence on this site speaks to each. Designed, documented: Russia's Internet Research Agency called a protest and its counter-protest at the same Islamic centre in Houston in 2016, at a cost of $200, and its operatives took 'all sides of the arguments, all from the same internet protocol (IP) address' (US Senate Intelligence Committee). In the first quarter of 2024 Meta removed two networks aimed at this conflict, one from each direction: an Iran-based network targeting Israel with fake Israeli personas on the right and the left (22 Facebook accounts, 8 Pages, 8 Groups, 23 Instagram accounts), and an Israel-based one run by a Tel Aviv marketing firm, STOIC, targeting the US and Canada with pro-Israel comments (510 Facebook accounts, 11 Pages, 32 Instagram accounts). In Europe, on the same conflict: in November 2023 France's foreign ministry said its agency VIGINUM had detected a network of '1095 bots' on X that published '2589 posts' amplifying photos of Stars of David stencilled on walls in Paris, affiliated 'with a high degree of confidence' to Russia's RRN/Doppelgänger network; the bots posted the photos from 28 October, 'nearly 48 hours' before the first authentic post. The ministry read it as a strategy 'to exploit international crises to sow confusion and create tensions'; who commissioned the stencils it left to the judicial inquiry (French; this site's translation). That operation fed one fear, Jews' fear of antisemitism. In the same month the same campaign worked the other direction: a spoofed copy of Le Parisien's website ran a fake article claiming that President Macron has 'the blood of Palestinians on his hands' (Euronews, November 2023, which attributes the spoofs to Doppelgänger; their recurring theme was Western aid 'diverted from Ukraine to Israel'). Fear for one community, anger for the other, from one operation: the pattern of the Houston case, on this conflict. Reading the scale: account counts measure effort, not effect. The one reach figure given is for the Israel-based network: about 500 Page followers and about 2,000 Instagram followers, and its comments were 'often met with critical responses from authentic users calling them propaganda'. No reach figure is given here for the Iran-based network. Weight of evidence: each removal is an incident-level finding by a platform or a committee; none measures how much of the conflict's online argument such networks make up. Why the loop feeds itself, by itself: in ten studies (N = 3,440), Nour Kteily, Gordon Hodson and Emile Bruneau found that people who feel an outgroup sees them as less than human dehumanise it in return, and that this 'plays a central role in outgroup aggression'; one study was of Israelis' perceptions of Palestinians. It also ran the other way: Americans who learned 'that Muslims humanize Americans' humanised Muslims 'in turn' (2016; abstract read). On this conflict both sides were measured during the 2014 Gaza war: on a 0–100 'Ascent of Man' scale, Israelis rated Palestinians 39.81 points lower than their own group, and Palestinians in the West Bank rated Israelis 37.03 points lower; the authors call the levels 'extreme (and comparable)' (Bruneau and Kteily, 2017; online wartime samples, not representative of either population, as the authors say). And each side misjudges the other: during the May 2021 riots inside Israel, 'the vast majority of Jews and Arabs opposed violence', yet 'each group perceived their respective out-group as significantly less opposed'; telling people that the other group 'vastly opposes violence' increased their own opposition to it, and a replication a year later also decreased support for violence (Nir, Nassir, Hasson and Halperin, 2023; Israeli citizens only, abstract read). The misjudgement is not always negative: in three preregistered studies (N = 1081), Muslim Palestinians in the West Bank and religious Jewish Israelis each predicted that 'belief in God motivates outgroup members to give more money in intergroup exchanges' and to 'place a greater value on outgroup members' lives' (Shackleford, Pasek, Vishkin and Ginges, 2024), which cuts against the view, which the authors trace to writers such as Richard Dawkins, that 'diversity of religious belief causes intergroup hostility'. So the loop needs no operator, and one thing that eases it, in these studies, is evidence that the other side sees you as human. The worst case for each side. For 'it's all bots and foreign agents': take the networks at their largest; the documented ones were small where reach was measured (VIGINUM's count, 1,095 accounts and 2,589 posts, measures effort, not reach), and the two most-quoted divisive lines on this page, 'Zionists are the modern Nazis' and 'there is no such thing as a Palestinian people', were said in public by a head of state and a minister, not by fake accounts. For 'nobody manufactures this': the Houston case shows both sides of a confrontation being staged for $200, networks from both directions were found working on this conflict in a single quarter, and one Russian network fed fear on one side and anger on the other in the same month of 2023. Holds on the evidence here: staged division is documented, including on this conflict from both directions, and the most divisive lines are also said openly by real leaders. Open, needs platform-wide reach data: that influence operations drive a meaningful share of the argument. Not supported: that the loop is mainly manufactured, or that it never is.
What the labels (Holds, Open, Arguable …) mean
primary documentUS Senate Select Committee on Intelligence, 'Russian Active Measures Campaigns and Interference in the 2016 U.S. Election', S. Rep. 116-290, Volume II: Russia's Use of Social Media (GovInfo, 2020) (2020) finding ✓
Read 2026-09-28 (scanned PDF; OCR checked against the page images). Vol. II p. 47 (report p. 117): Houston, 21 May 2016 — 'Heart of Texas' ('Stop Islamization of Texas') and 'United Muslims for America' ('Save Islamic Knowledge') events at 'exactly the same Islamic Da'wah Center'; 'In neither instance was the existence of a counter-protest mentioned'; 'escalated into confrontation and verbal attacks'; cost '$200'; run 'from the confines of the IRA's headquarters in Saint Petersburg'; John Kelly: 'kind of like arming two sides in a civil war'. Vol. II p. 54 (report p. 124): operatives 'simultaneously taking and attacking all sides of the arguments, all from the same internet protocol (IP) address'.
primary documentMeta, 'Quarterly Adversarial Threat Report, Q1 2024' section 'Israel-based network' (2024) finding ✓
Read (2026-09-28). 'found links to STOIC, a political marketing and business intelligence firm based in Tel Aviv, Israel. It is now banned from our platform'; 510 Facebook accounts, 11 Pages, one Group, 32 Instagram accounts; about 500 Page followers, ~2,000 Instagram followers; targeted the US and Canada; commented on pages of media and 'political and public figures, including US lawmakers'; 'supportive commentary about Israel' (incl. its military actions, per the report's summary table); criticism of Muslims and UNRWA; comments 'often met with critical responses from authentic users calling them propaganda'; some comments 'likely generated using artificial intelligence'. The report does not name who commissioned the operation. Iran-based network: 22 Facebook accounts, eight Pages, eight Groups, 23 Instagram accounts; 'originated in Iran and targeted Israel'; 'fictitious personas posing as Israelis'; 'supportive commentary about Israel'; support for Itamar Ben-Gvir while criticising Netanyahu; accounts that 'posed as young liberal Israeli women and posted about anti-government protests'. Re-read 2026-10-12: the comments 'included links to the operation's websites and were often met with critical responses from authentic users calling them propaganda'.
primary documentFrench Ministry for Europe and Foreign Affairs, 'Russie – Nouvelle ingérence numérique russe contre la France' (statement, 9 November 2023) (2023) finding ✓
Read 2026-10-13 (French; translations here are this site's). 'La France condamne avec fermeté l'implication du réseau russe Recent Reliable News (RRN/Doppelgänger) dans l'amplification artificielle et la primo-diffusion sur les réseaux sociaux des photos des tags représentant des étoiles de David dans le 10e arrondissement de Paris.' 'S'agissant des faits eux-mêmes, l'enquête judiciaire en cours devra établir la possible responsabilité d'un commanditaire étranger.' VIGINUM 'a détecté le 6 novembre 2023 l'implication d'un réseau de 1095 bots sur la plateforme X (anciennement Twitter), ayant publié 2589 posts'; affiliated to RRN 'avec un haut degré de confiance'; 'la première publication authentique des photos des tags semble être intervenue le 30 octobre à 19h37 sur X, VIGINUM a identifié les premières publications du réseau de bots RRN dès le 28 octobre, à partir de 19h24, soit près de 48h avant.' The ministry's reading: 'une stratégie opportuniste et irresponsable visant à exploiter les crises internationales pour semer la confusion et à créer des tensions dans le débat public en France et en Europe.' A government's attribution, from its own agency's detection. This site's translations of the phrases quoted on its pages: '1095 bots', '2589 posts', 'with a high degree of confidence', 'nearly 48 hours', 'to exploit international crises to sow confusion and create tensions'.
pressEuronews (The Cube), Sophia Khatsenkova, 'Doppelganger: How a Russian disinformation campaign is exploiting the Israel-Hamas war' (22 November 2023) (2023) finding ✓
Read 2026-10-13. 'From the font to the layout, everything looks exactly like the website of the French daily Le Parisien. … the article that claims French President Emmanuel Macron has "the blood of Palestinians on his hands." In reality, it's a fake article, published on an almost identical platform with just a slightly different URL'; spoofed Spiegel pages in German; 'there is a common theme: the suggestion that the financial support coming from Western powers has been diverted from Ukraine to Israel'; the X accounts sharing the spoofs were mostly 'created in September and October 2023' with no followers (per the Antibot4Navalny collective). On the Paris Stars of David: the Moldovan couple 'claimed to have acted on the orders of a Russian individual in exchange for money', but 'a direct link between this incident and the Doppelganger disinformation campaign has not y[et been established]' (text cut off in the copy read).
dataNour Kteily, Gordon Hodson, Emile Bruneau, 'They See Us as Less Than Human: Metadehumanization Predicts Intergroup Conflict via Reciprocal Dehumanization', Journal of Personality and Social Psychology 110(3) (2016) (2016) finding ✓
Abstract read 2026-10-13 (University of Pennsylvania repository; doi 10.1037/pspa0000044). 'Across 10 studies (N = 3,440) involving several real-world conflicts spanning 3 continents, we provide the first empirical evidence that metadehumanization (a) plays a central role in outgroup aggression that is (b) mediated by outgroup dehumanization, and (c) distinct from metaprejudice.' 'Study 4 examines Israelis' metaperceptions with respect to Palestinians, showing that: (a) feeling dehumanized (i.e., metadehumanization) is distinct from feeling disliked (i.e., metaprejudice), and (b) metadehumanization uniquely influences aggression through outgroup dehumanization'. 'In Study 6, American participants learning that Muslims humanize Americans (i.e., metahumanization) humanize Muslims in turn.' The full paper is not read; the Israeli study is of Israelis only, not Palestinians' perceptions of Israelis.
dataEmile Bruneau, Nour Kteily, 'The enemy as animal: Symmetric dehumanization during asymmetric warfare', PLoS ONE 12(7): e0181422 (26 July 2017) (2017) finding ✓
Read 2026-10-13 (open access). Israelis (Study 1; N = 521) and Palestinians (Study 2; N = 354) during the 2014 Gaza war; 'community samples of Israelis and Palestinians expressed extreme (and comparable) levels of blatant dehumanization', associated with outgroup hostility in both. Measure: the 'Ascent of Man' scale, 0–100. 'On average, Israelis rated Palestinians 39.81 points lower than their own group'; Palestinians rated Israelis 'an average of 37.03 points lower than Palestinians'. Absolute ratings: Israelis rated Palestinians M = 41.21; Palestinians rated Israelis M = 45.03, both 'closer in ‘evolvedness’ to the quadrupedal primate than to the ‘fully evolved’ human'. Palestinian sample: 'community residents of the West Bank, recruited through a text-based employment agency', online, late August 2014. Limits stated by the authors: 'the samples we collected were not national probability samples, and thus they cannot be considered fully representative'; the Israeli sample was about 10 years older; collected in wartime, which 'might be expected to accentuate outgroup dehumanization'.
dataNimrod Nir, Yara Nassir, Yossi Hasson, Eran Halperin, 'Kill or be killed: Can correcting misperceptions of out-group hostility de-escalate a violent inter-group out-break?', European Journal of Social Psychology 53(5): 1004–1018 (August 2023) (2023) finding ✓
Abstract read 2026-10-13 (Hebrew University research portal; doi 10.1002/ejsp.2949). During the May 2021 riots between Israel's Jewish and Arab citizens: 'although the vast majority of Jews and Arabs opposed violence, each group perceived their respective out-group as significantly less opposed to such violence, and that meta-perception was strongly associated with participants' own opposition to inter-group violence'; 'experimentally exposing Jewish and Arab citizens to information showing that their out-group vastly opposes violence increased participants' own opposition to inter-group violence in the midst of nationwide violent inter-group attacks'; a third study during the 'June 22 wave of terror' replicated this, 'both increasing opposition to violence and decreasing support for violence'. Citizens of Israel only (Jews and Arabs), not Palestinians in the territories; the full paper is not read.
dataCrystal M. Shackleford, Michael H. Pasek, Allon Vishkin, Jeremy Ginges, 'Palestinians and Israelis believe the other's God encourages intergroup benevolence: a case of positive intergroup meta-perceptions', Journal of Experimental Social Psychology 110 (2024) (2024) finding ✓
Read 2026-10-13 (LSE Research Online, the authors' accepted version; doi 10.1016/j.jesp.2023.104551). 'Contrary to this hypothesis, three preregistered studies (N = 1081) provide consistent evidence that Palestinians and Israelis predict that belief in God motivates outgroup members to give more money in intergroup exchanges (Studies 1 and 2) and to place a greater value on outgroup members' lives (Study 3).' Samples: Muslim Palestinians in the West Bank (in person) and Jewish Israelis identifying as religious or ultra-Orthodox (Study 2 Israeli sample: 'religious (57%) or ultra-orthodox (43%)'). 'We note that 19% of Palestinian participants interviewed as part of the larger protocol … declined to participate in this prediction experiment'. Accepted version; may differ from the published one. Introduction: 'Scholars often argue that diversity of religious belief causes intergroup hostility (e.g., Dawkins, 2006; for a review see Armstrong, 2014).'
Where this trap shows up
Your side does it too — look at both columns.