We monitored our 1,034-account roster, the anonymous boards we track, and our off-roster finds for the week of September 7 to 13, 2026; the collected record runs through Saturday, September 12, a partial day, and September 13 is not in it. The roster is the frame we track; what we collected is smaller: 111 of the 152 X handles on the roster reached, 1,650 opening posts from the boards, and 4,916 Bluesky posts. 367 items crossed our line. 340 of them run 47 distinct scripts and 27 stand alone. No script reached the top tier. The peak in our monitored sample is the replacement-and-remigration script at 3.95: native populations in Britain, Ireland, Italy, Portugal and Germany replaced by design, with mass removal as the only answer, sixteen posts across three wings. The AfD’s Saxony-Anhalt election program provides, in CNN’s account of it, for “the return of non-ethnic-German citizens to their country of origin”, a policy the program calls remigration, and its lead candidate restated the aim at a press conference, so the removal remedy is carried this week by a state party that has just taken 44 percent of the vote in the September 6 state election, three seats short of a majority as the news record carries the result; Tommy Robinson’s remigration-or-war line on the day of the Tamworth march is the week’s one named-carrier call for violence, at 413,937 impressions, and it sets the harm floor the script clears on its own computed score; RadioGenoa’s replacement captions run over London, Naples, Genoa and Lisbon footage. Two events drove most of the rest. The 25th anniversary of 9/11 turned four scripts on Muslim people and on the two Muslim officials named over the ceremony clip: a disloyalty script at fourteen items, an eliminationist script at twenty, a collective-guilt script at eight, and a Jewish-authorship claim on the dancing-Israelis canard that carries named accounts for the first time. The Saxony-Anhalt aftermath, with Chancellor Merz’s ethnic-cleansing charge inverted onto migration by Naomi Seibt and Eva Vlaardingerbroek, fed the peak script and half of the week’s news table. The Clancy juror interviews on September 8 kept two scripts about women alive at lower volume, the collective-guilt chain we have carded since the trial opened, now down a tier, and the civic-exclusion script on Myron Gaines and two board threads. Three sitting legislators carry scripts in their own voices: Rupert Lowe on the Portsmouth invaders, Lee Anderson on the invasion and barbarian scripts, and Geert Wilders, whose Islam kills post at 3.9 million is the largest single reach on the ledger. Every one of the 47 coordination verdicts is organic parallel. Seven cross-carrier convergence signals are on the record, named in the scripts’ coordination notes; six lift a coordination factor to 2 and one raises nothing, and none carries the bridge, hub or arrangement a verdict requires. On Bluesky, two items came from 4,916 posts, both on one script that reads reported antisemitism as staged.
The ledger holds 367 unique posts against last week’s 314, and 29,821,738 platform-reported impressions against last week’s 37,063,245. The comparison carries no directional reading. This week’s gap pass read 1,294 posts the slices had not reached and found 25 items against five last week, while the Saturday walk reached 111 of the 152 roster X handles against 117, so the collection ran wider in one place and narrower in another, and last week’s impression figure leaned on a single 12.5 million view post that has no counterpart this week. The Bluesky collection stood at 4,916 posts, the boards at 1,650, and X at 5,165 after the walk. The severity peak is the replacement-and-remigration script at 3.95, the same reading as last week’s peak; fourteen scripts sit in the second tier, and none reached the top tier. The widest reach in our monitored sample belongs to the Islam-eliminationist script at 6.9 million impressions, twenty items on five named accounts, with Laura Loomer’s anniversary chain of twelve posts and Geert Wilders’ Islam kills post at 3.9 million. The largest by volume in the sample is the Jewish-collective control script at forty-one items, thirty-six of them board threads and the rest on Henry Makow, Jackson Hinkle and Yves Engler. The two scripts that ride the Clancy juror interviews together reach 1.7 million impressions. Seventeen board-only scripts carry no view counts at all, so the scripts we can measure least include three of the fourteen in the second tier.
Scripts by severity
Click a bar for the full scorecards.
The week's weather, day by day
One dot per real event. Filled dots drew a reaction we confirmed; hollow ones drew none we recorded. Click for the full board.
The 367 items, by harm level
Click for the full ledger.
On the radar
Click a date for the full calendar and the patterns building toward it.
01 · The scripts
Five factors are scored from 1 to 5: reach, toxicity, coordination, mainstreaming risk, and harm potential. Each item is assigned to exactly one script, with overlaps adjudicated on the dominant predicate. A script’s canonical claim is its shared core predicate; escalation clauses, a removal remedy or a ban from office, are stated with the count of posts that carry them, and each script’s prose does that. Their composite Ember Priority Score is shown on a chart running from 0 to 5; the axis starts at 0 for display, and the score itself cannot fall below 1. Click one for its scorecard. Peak severity this week is 3.95, the replacement-and-remigration script, and no script reached the top tier; the median script sits at 2.51. Fourteen scripts stand in the second tier, eighteen in the third, and fifteen at the bottom. Four scripts carry the severity floor an explicit call for violence sets: the replacement-and-remigration script clears it on its own computed score, on Tommy Robinson’s remigration-or-war ultimatum, and three board-only scripts compute to 2.00 and sit in the second tier because of the floor alone, one naming Jewish people for hanging, one welcoming the death of non-white people, and one calling for violence against the political left. The ultimatum went up on the day of Britain First’s Tamworth remigration march in the same run as Robinson’s MILLIONS MUST GO posts, which name the migrant-origin population of Britain as the people addressed; it codes as a call for violence under the methodology’s threat-as-prediction construction, war stated as the outcome if removal is refused, and it holds at medium confidence because the war is stated as an outcome and no instruction is given. Toxicity anchors at 4 for sustained othering that carries a single explicit call for harm and at 5 for explicit dehumanizing metaphor or endorsed harm across the script; the peak script sits at 4 under the first anchor.
02 · The ledger
Click a row for the quote, the coding, and the source. Every named item carries all three. The two Bluesky rows are identification-bar redactions: their wording and source are withheld, and their classification rests on our collection record rather than on public verification. The ledger counts unique posts. Fifty-six posts were captured twice, by two X slices that read the same accounts, and each pair was collapsed to one id before anything was counted; where the two verifications disagreed on the rung, the adjudication with the fuller reading stood by ruling, the more severe of the two in every case, and the ruling is recorded on the script. A collapsed pair counts once in every count, factor and reach figure.
03 · The week's weather
The events we recorded no reaction to matter as much as the ones we did. Together they show which stories these narratives picked up this week and which drew no reaction our footprint recorded.
04 · On the radar
Twenty-eight scripts continue a ruled lineage. The Jewish-collective control script is carded for a seventh week, the longest chain in the record, forty-one items this week, and it holds in the third tier. The replacement-and-remigration script continues the Channel-invasion chain for a sixth week and holds the second tier, now as the peak. The Islam-eliminationist and Muslim-official disloyalty chains each run a fifth week, both in the second tier. The women’s collective-guilt script continues the chain opened when the Clancy trial began, down a tier as the juror interviews replaced the verdict as its occasion, and the women’s civic-exclusion script continues at five items. The invasion frame continues the Spain-conquest chain in the second tier. Three scripts moved up into the second tier: descent-belonging denial, the Islamization takeover script and the migrant-criminality class. The board-only anti-Indian dehumanization script continues a chain now in its fourth week and fell from the second tier to the fourth with no extermination call on its threads this week; the anti-Indian exclusion script continues last week’s named-carrier card, now on a government account and a board thread. Six scripts moved tier, four up and two down. Nineteen scripts are new this week, among them the Jewish-authorship claim on 9/11, split from collective control on its own predicate, and three board-only anti-Indian scripts. Ten of last week’s scripts are not carded this week. We do not read that as decay: a narrative can be absent because it lost its carriers, because it did not clear the bar this week, or because this week’s collection did not surface it, and these records cannot separate those. Each line compares one narrative with itself a week earlier, joined on a key ruled when the narrative was carded rather than on any similarity between the two weeks’ text.
05 · Method
This is a purposive, outcome-selected monitoring sample drawn from a 1,034-account roster spanning 23 ideological categories; this week’s coverage limits are stated below. It is not a random slice of the internet. Roster and board collection is systematic within those limits; off-roster discovery is opportunistic and incomplete, so our off-roster finds show that off-roster carriage exists and cannot say how much of it there is. A hit is language that collectivizes, dehumanizes, or calls for exclusion of a protected group, or that runs dehumanizing or eliminationist framing against a political camp. Criticism of governments, institutions, and named individuals over their own conduct stays protected, however sharp. Every hit rests on a dated source read in full at collection, and every named item on the ledger carries its quote, its coding and its source; the two Bluesky items are identification-bar redactions, described under limitations, whose classification rests on our collection record. Where the source was still live, a second, adversarial pass re-verified the quote against it, 148 of 148 on the right-lane X slice and 109 of 109 on the unassigned X slice; where it had already gone offline, which covers the 4chan threads in this week’s set, the second pass re-read our collection record and the coding, and the item says so. That second pass rejected 37 items and corrected about 35 harm rungs, most of them downward. The count reconciles this way: 257 X records were verified page by page across the two X slices, 148 on the right-lane slice and 109 on the unassigned slice; 26 were rejected and 231 confirmed, and 56 of the confirmed were twins captured by both slices, so the two slices contribute 175 unique X posts; add 160 board items, two Bluesky items, three news-institutional items, two from the left-lane and other slice, and the 25 gap items, none of which shares a URL with any slice, and the ledger is 367. The rejections were policy argument without a group predicate, Nigel Farage’s invasion metaphor as a ballot-box argument among them, institutions as targets, single-person insults with no group claim, presupposition jokes with no stated predicate, and an in-group slur, and none of them are in this report. A separate 15-item depth check, run independently of the pipeline on the board slice, the slice with the largest share of confirmed items (160 of the 367; by posts collected, X and Bluesky are the larger corpora) and the one with no live page to re-read, agreed with fourteen of the fifteen codings; the fifteenth, a headline that labelled two onlookers by religion with the hidden-hand claim supplied by the reader, matched the shape this slice rejects and was removed. Research and drafting run through Claude under this methodology; every published classification passes the documented verification chain, and Jerm holds final editorial responsibility. Nine items this week cleared the extremism branch of our gate against a political camp rather than a protected class, two of them on the two violence rungs: a board call for brutal violence against opponents of AI framed as leftists, and a board mandate to eliminate the political left. The rest are Matt Walsh’s demonic leftists and rodents, Jack Posobiec’s Unhumans thesis and his Bolsheviks line, Peter Sweden’s the left murdered him, and a board thread calling AfD voters inbred as a class. Every other confirmed item names a protected class, or a group our methodology treats as one, including seventeen of the nineteen ledger rows on the two top harm rungs, glorifying violence and calls for violence. Items from other directions are coded under the same standard: the AfD-voters thread, a board post naming American people for nuclear attack, a board post calling Christian people a cancer, and the two Bluesky posts that read reported antisemitism as staged.
Platform-reported impression counts are capture-time platform counters, and they can rise after we read them. Read each figure as the reading at capture rather than as an audience count. This week’s total leans less on any single post than last week’s did; the largest, Geert Wilders’ Islam kills post at 3.9 million, carries about an eighth of it.
Reach is not measurable everywhere. The 4chan corpus has no view counts and no persistent handles, so board items carry no reach figure and add nothing to a family total. Reach therefore ranks the named-carrier scripts only, and any cross-platform reach comparison on this page ranks X carriage only. The 4chan threads in this week’s set were verified against our collection records at capture rather than against a second live read, so confirmed for those items means that.
The X footprint needs stating plainly. The nightly core pull reached about 35 handles a night, and a bounded catch-up walk on Saturday raised in-window coverage to 111 of the 152 X handles we track, against 117 last week; 41 handles had no in-window posts collected at the cut, so a narrative living on those handles reads here as smaller than it is. Two accounts posted at a volume the slice reads did not exhaust, RadioGenoa at 182 in-window posts and Laura Loomer at 147, and a gap pass over 1,294 posts the slices had not read found 25 items against five last week, most of them on those two accounts and on three accounts the slices had read to zero, Mark Levin, Nick Tenconi and Peter Sweden; they are in the ledger. The hopper’s X records truncate at the 280-character preview, and every X quote on the ledger was verified against the live page at full length. Daily shortlists exist for September 7 to 11 and added little beyond the news table; September 12 is a partial day, and September 13 is not in the collected record at the cut, so the Sunday nightly’s posts roll into next week’s collection. The news table carried no rows for the American stories that drove the week, the 9/11 ceremony, the Rogers attack on Abdul El-Sayed, the Clancy juror interviews, the Texas curriculum vote and the Republican convention shouts, which reached the scan through the posts table and web search. The twin collapse stated on the ledger moved Jack Posobiec’s Unhumans post to hateful and two stand-alone items, and moved no score. The research cut ran Saturday evening, September 12, and the report publishes the same day, so the last day and a half of the window is uncovered. Britain First’s Tamworth march and the first protests under the balaclava bans fell on that last day, so carriage of both reads thin for a short-window reason.
Coordination is two readings that answer different questions. The coordination factor in the score measures degree of organization on the 1 to 5 anchors, and it rises off the floor on convergence signals: near-duplicate phrasing, same-day timing, a shared novel claim, coined term or fabricated detail. The coordination verdict, overt, covert or organic parallel, requires a confirmed bridge account, a named hub or an arrangement, and convergence alone never reaches it (METHODOLOGY section 9). This week every verdict is organic parallel. Seven cross-carrier convergence signals are on the record, meaning named in a script’s coordination note, and six of them lift a coordination factor to 2, the convergence-without-a-bridge anchor: the invader label on the Portsmouth landing from three party leaders and a sitting MP within a day; the same-trigger convergence on Chancellor Merz’s remark and the Tamworth march across four named carriers; the Ground Zero ceremony clip captioned on four named accounts within hours of its Fox circulation, convergence on a broadcast source; the Uganda line on ACTBrigitte on September 9 and on RadioGenoa on September 11 in near-identical wording, two carriers two days apart with no hub, timing cluster or bridge documented between them; Dan Dicks’ quote-post of the Action4Canada founder’s video with an endorsement, one documented amplification link between two Canadian accounts, which does not meet the hub, shared-stage or network bar; and the dancing-Israelis canard on two named accounts the same day with no shared text. The seventh raised nothing: Mark Levin’s caption on the 2001 footage, Palestinians celebrating the attack, matched a same-morning caption on an item that was rejected at verification and sits outside the ledger, and a rejected item cannot raise a factor, so the match is noted as observed and that script’s factor stays at the floor. None of the seven reached a verdict; each becomes a coordination check if a linking account appears. On an anonymous board, one actor and an organized handful cannot be told apart; the verbatim reposts of a board opening post recorded on five scripts, a day to three days apart or a week on, read as one poster re-running a thread, and no board script this week showed the copypasta pattern that earns a covert verdict.
The Bluesky lane produced two coded items from 4,916 screened posts, against four last week from a corpus of the same shape. Both sit in one script that reads reported antisemitism as staged: one denies that Jewish people are Semites and casts the antisemitism reported against them as planned by Zionists, coded hateful, and one applies a Zionist-Nazi portmanteau as a standing character, coded stigmatizing. Both are low-engagement posts from small accounts, and they are identification-bar redactions: their wording is withheld because a verbatim quote identifies an account as surely as its handle, so their classification rests on our collection record rather than on public verification. Two items from 4,916 posts admit several readings: lower prevalence of this carriage on the platform, a mismatch between our lexicon and the platform’s vocabulary, the accounts we collect, or a recall failure in the screen. Two items cannot separate those readings, and we have run no recall study on the platform. What the collection does show is that the platform’s migration and AfD traffic this week ran to reporting and commentary against the far right.
No hit-rate or prevalence claim here may be read as a comparison across ideological lanes. That comparison requires a lane-tier sample for every lane compared, over a common platform footprint. We did not draw that sample this week.
Five factors, each scored 1 to 5: reach (R), toxicity (T), coordination (C), mainstreaming risk (M), and harm potential (H). They combine as (R·T·C·M1.3·H1.5)1/5.8: multiply all five, with mainstreaming risk raised to the 1.3 power and harm potential to the 1.5 power because those two carry the most real-world weight, then take the 5.8th root. The 5.8 is the sum of the weights (1 + 1 + 1 + 1.3 + 1.5), which brings the result back onto the 1-to-5 scale. Multiplication limits how far any one factor can lift a score, without eliminating its influence. One override runs the other way: an explicit call for violence against the target class sets harm potential to 5, and any script at H=5 takes a P2 floor regardless of its computed score, because an explicit call for violence with no measured reach is still an explicit call for violence. Four scripts carry that floor this week, and the chart marks them; one computes above the second-tier line on its own score, and three are lifted by the floor alone. P1 starts at 4.0, P2 at 3.0, P3 at 2.0. Coordination as a factor scores degree of organization: it leaves the floor on convergence signals, near-duplicate phrasing, same-day timing or a shared novel claim, and climbs to the co-platforming and network anchors on a named hub or a confirmed bridge. The coordination verdict is a separate reading and requires the bridge, the hub or an arrangement, so a script can score 2 on the factor and still carry an organic parallel verdict.
Full-schema agreement: 78% as of 2026-07-27