Standfirst: Terrorism charges in Manchester, a Belgian intelligence attribution and new government records in New York and France show that antisemitism is no longer documented only through incident counts. It is also appearing in counterterrorism cases, threat assessments and institutional security files. Those categories must still be kept separate.

Publication date: October 4, 2026

Dateline: PARIS

Abstract: Two men arrested in Manchester on September 20 have now been charged with preparing acts of terrorism allegedly targeting Jewish sites. Belgian security officials have separately attributed attacks on Jewish locations in Liège and Antwerp to an Iran-inspired campaign. In London, however, prosecutors removed the terrorism connection from a different stabbing case while retaining an alleged antisemitic hate-crime element. New York released its first municipal antisemitism strategy, France published broader midyear bias-crime data, and Canadian police opened a new investigation into antisemitic flyers near the synagogue attacked on Yom Kippur. Together, the records show a persistent and increasingly security-intensive threat environment, not a broad decline in antisemitism.

Manchester moved from arrest to charge

Eight days changed the legal status of the Manchester case. When British police announced the arrests of two men on September 22, the public record supported one description: a suspected terrorist plot against the Jewish community was under investigation. On October 2, the Crown Prosecution Service authorized charges against Rahman Salehi and Salam Ahmadyan for allegedly engaging in conduct in preparation for terrorist acts.

The Metropolitan Police said both men are Iranian nationals and were arrested in central Manchester on September 20, hours before Yom Kippur began. The charges allege that, on or before that date, they communicated through encrypted applications to obtain components for a homemade explosive device, acquired written and video bomb-making instructions, obtained liquids and equipment, researched potential targets, collected maps and images, and conducted reconnaissance.

At Westminster Magistrates’ Court on October 3, the prosecution said the alleged targets included the Manchester Jewish Museum and Heaton Park Hebrew Congregation, where two worshippers were killed in a terrorist attack on October 2, 2025. Prosecutors alleged that a third person, assessed to be an Iranian national operating from Iran, coordinated the planning and supplied instructions for a device using chlorine and brake fluid.

Neither defendant entered a plea. Both were remanded in custody for an October 23 hearing at the Old Bailey. These are detailed criminal allegations, not findings of fact. The CPS expressly stated that its decision means only that prosecutors consider there to be sufficient evidence to present the case to a court. Guilt can be determined only after the judicial process.

The procedural change still matters. A police suspicion has become a formal terrorism prosecution with identified defendants and particulars of charge. It would be inaccurate to continue describing the matter merely as two arrests. It would be equally inaccurate to describe the alleged plot as proved.

One London case moved out of the terrorism category

Another British case moved in the opposite direction. On April 29, two visibly Jewish men were stabbed in Golders Green, north London. Police initially declared the incident terrorist and the national threat level was subsequently raised to “severe,” although the government said the decision reflected the wider threat environment and not that attack alone.

At a September 25 pretrial hearing, defendant Essa Suleiman pleaded not guilty to attempted murder charges. Prosecutors told the court that they would no longer pursue a terrorism connection. The Metropolitan Police later said the CPS remained satisfied that there was evidence supporting an alleged hate-crime element, which could be presented as an aggravating factor if there were a conviction.

The reclassification does not mean that the stabbings did not occur, that the victims were not Jewish, or that antisemitic motive has been disproved. It means the prosecution does not intend to establish the statutory terrorist connection in the criminal proceedings. The attempted-murder allegations and the alleged hate aggravation remain to be tried.

This is a useful warning against treating legal labels as permanent facts. An incident can enter a dataset as suspected terrorism, later proceed as an alleged hate crime and ultimately produce a conviction, acquittal or different finding. Each stage answers a different question. None should be silently substituted for another.

Britain’s broader record remains elevated

The category change in Golders Green does not establish that the wider threat has receded. The Community Security Trust recorded 1,926 antisemitic incidents in Britain during the first half of 2026, 21 percent more than in the same period of 2025 and almost twice the first-half total in 2023.

Severity moved upward. CST recorded 135 assaults and one case of extreme violence, for 136 physical attacks in all. That was an 82 percent increase from the first half of 2025 and the highest six-month assault total in the organization’s series. Threats rose 27 percent, while damage and desecration fell 18 percent. The fall in one category therefore cannot support a conclusion that antisemitism as a whole declined.

CST classified the Golders Green stabbings as the period’s sole case of extreme violence. That classification describes the severity and apparent targeting under CST’s methodology. It is separate from the terrorism threshold prosecutors later declined to pursue. A monitoring organization, police investigators and a criminal court can properly attach different labels because they apply different tests.

Belgium makes the foreign-state dimension explicit

The British charges are not the only development placing attacks on Jewish sites inside a transnational security file. On September 25, The Brussels Times reported that Gert Vercauteren, head of Belgium’s Coordination Unit for Threat Analysis, attributed two March attacks to calls from Iran to strike Jewish and Israeli interests abroad.

The incidents were an explosive device detonated outside the Liège synagogue on March 9 and the burning of a car in Antwerp’s Jewish quarter during the same week. No one was physically injured. Belgian federal prosecutors opened a terrorism investigation into the Liège explosion. Vercauteren described the incidents as part of a coordinated campaign rather than isolated vandalism and referred to the disruption of related operations, including an alleged plan to bomb a synagogue in Heemstede in the Netherlands.

Belgium’s own public threat assessment had already stated in March that Iran had demonstrated the capacity to conduct operations in Europe, directly or through proxies, against Israeli and Jewish interests, American interests and Iranian opposition figures. The national threat level remained at three on a four-point scale, with heightened vigilance around Jewish and Israeli sites.

These are intelligence and prosecutorial assessments. They carry more evidentiary weight than anonymous speculation, but they are not convictions of the people or state institutions allegedly responsible. The Manchester prosecution likewise alleges contact with a person who may have been in Iran; the court has not established Iranian state direction. Similar subject matter does not make the two investigations interchangeable.

France: a broader indicator rose, but it is not an antisemitism total

France’s Interior Ministry published a midyear crime review on September 30 covering the twelve months through June 2026. For the first time, the review included supplementary indicators for offenses of a racist, xenophobic or anti-religious character. That combined category rose 2 percent.

The figure is relevant to the general bias-crime environment, but it cannot be presented as a 2 percent increase in antisemitism. The published headline does not separate anti-Jewish offenses from anti-Muslim, anti-Christian, racist or xenophobic cases. It also measures police- and gendarmerie-recorded offenses, not every incident experienced or reported to community organizations.

The latest complete antisemitism-specific Interior Ministry total remains 1,320 recorded acts in 2025. That was 16 percent fewer than the exceptional 1,570 recorded in 2024. The ministry nevertheless described the level as historically high and noted that the last three years were the highest of the previous quarter-century. The 2025 figure remained more than three times the 436 acts recorded in 2022. Physical or verbal attacks and online hatred directed at people accounted for 67 percent of the total.

The statistical framing is therefore straightforward. France recorded fewer antisemitic acts than at the 2024 peak; the available evidence does not establish less antisemitism in the broader social sense. The longer baseline, composition of the total, continued protection of Jewish institutions and underreporting all prevent that conclusion.

New York places antisemitism inside city government

On September 29, New York City released what it described as its first municipal Strategy to Combat Antisemitism. The plan commits $6.4 million to community hate-prevention and survivor-support grants, $1 million to security at synagogues and other houses of worship, and support for a $26 million investment in the Office for the Prevention of Hate Crimes. It also relaunches an interagency Security Infrastructure Working Group and expands Holocaust and Jewish-history education.

The strategy is an official policy document, not a measure of prevalence and not evidence that any intervention has already produced an effect. Its evidentiary value is different: it records the city government’s assessment that antisemitism requires coordinated work across security, civil rights, education, cultural institutions and survivor services.

The most recent NYPD figures help explain that assessment. Confirmed anti-Jewish hate crimes fell from 22 in August 2025 to 19 in August 2026. Across the first eight months of the year, however, they rose from 212 to 229 and represented 53.9 percent of all confirmed hate crimes. The monthly decrease and the cumulative increase are both accurate. Selecting only the first creates a misleading impression of improvement.

The NYPD also distinguishes incidents initially reported as hate crimes from those later confirmed after investigation. That distinction improves clarity, but it means the two series cannot be mixed. Nor do police records measure every instance of harassment, discrimination, concealment of Jewish identity or unreported abuse.

Canada: propaganda appears at the site of an armed attack

On October 1, Belleville Police opened an investigation into antisemitic flyers distributed in the city’s east end, including the area around the Sons of Jacob Synagogue. The same synagogue had been targeted on September 20 by an armed attacker whom police later said intended to carry out a mass killing during Yom Kippur services. A police officer was critically wounded while preventing entry; no congregants were shot.

The flyer distribution is a new reported incident. Police have described the material as antisemitic and asked for video evidence, but they have not publicly identified a suspect or established a connection to the September shooting. Proximity in place and time does not prove common authorship, organization or motive beyond what is visible in the material itself.

Canada’s record now contains two sharply different events at one location: a completed armed hate-motivated attack and a subsequent propaganda distribution under investigation. Counting both as one unit each may be appropriate for an incident ledger. It does not communicate the difference in severity, evidentiary status or institutional consequences.

Israel-related context requires two distinctions

International events involving Israel remain a major trigger in several datasets. CST reported that 787 incidents, 41 percent of its first-half 2026 total, referenced Israel, Palestine, the October 7 attacks or the subsequent conflicts. CST counted those cases because they also contained anti-Jewish language, motivation or targeting. It did not classify criticism of Israeli government decisions as antisemitic merely because Israel was mentioned.

That boundary is essential. Political opposition to an Israeli policy, including severe or angry criticism, is not in itself evidence of hostility toward Jews. Conduct becomes relevant to an antisemitism record when there is anti-Jewish targeting, collective blame, a recognized antisemitic trope, a threat or another documented bias indicator. Conversely, describing all Israel-related hostility as political criticism can erase cases in which “Zionist” is used as a substitute for “Jew” or Jewish people are targeted for actions of the Israeli state.

Why a lower total can still mislead

The latest evidence reinforces five recurring measurement limits:

• Peak selection: comparing only with an exceptional 2024 total can conceal how far current levels remain above 2022 or a longer historical series.
• Category composition: fewer reports of property damage or campus encampment-related incidents do not describe what happened to assaults, threats or terrorism cases.
• Severity flattening: a flyer, an online message, an assault and an alleged bomb plot may each count as one case while creating radically different harm and security demands.
• Legal attrition: police suspicion, charge, trial and conviction apply progressively different evidentiary thresholds. Reclassification can change a dataset without erasing the underlying conduct.
• Underreporting: the EU Agency for Fundamental Rights says national methods remain inconsistent and very few Jewish respondents report incidents to authorities or monitoring bodies.

A number can be accurate while the conclusion attached to it is unsupported. Nothing in the public records reviewed for this edition establishes deliberate statistical manipulation by authorities or monitoring organizations. Proving intent would require internal instructions, knowingly false descriptions or a documented pattern of strategic omission. The demonstrable issue is the misleading effect produced when scope, baseline and categories are left unexplained.

Conclusion

This week’s record does not demonstrate a broad decline in antisemitism. It shows terrorism charges over an alleged plot against Manchester Jewish sites, a separate British case reclassified from terrorism to an alleged hate crime, Belgian intelligence attribution of attacks to an Iran-inspired campaign, and continuing institutional responses in New York, France and Canada. The evidence increasingly reaches beyond incident monitoring into counterterrorism and national-security systems, while remaining subject to distinct legal standards.

The next significant records include the October 23 Old Bailey hearing in the Manchester case, the Golders Green trial, full-year 2026 data from CST and French authorities, the next ADL national audit, and the Australian Royal Commission on Antisemitism and Social Cohesion’s final report, due December 18. No comparable new national release from New Zealand was identified this week. Until those records arrive, short-term falls in selected categories do not establish that antisemitism itself is receding.

Notes and references

1. Crown Prosecution Service, “CPS to prosecute two men for preparing act of terrorism in Manchester”, October 2, 2026.
2. Metropolitan Police, Manchester terrorism charges and particulars, October 2, 2026.
3. Reuters, Westminster Magistrates’ Court hearing in the Manchester case, October 3, 2026.
4. Metropolitan Police, update on the Golders Green stabbing investigation, September 2026.
5. Crown Prosecution Service, original Golders Green attempted-murder charges, April 30, 2026.
6. Community Security Trust, Antisemitic Incidents January–June 2026.
7. The Brussels Times, Belgian threat-analysis attribution concerning Liège and Antwerp, September 25, 2026.
8. Belgian Coordination Unit for Threat Analysis, Middle East conflict threat assessment, March 13, 2026.
9. French Interior Ministry, midyear 2026 police- and gendarmerie-recorded crime review, September 30, 2026.
10. French Interior Ministry, Actes antireligieux — Tendances 2025, February 12, 2026.
11. New York City Mayor’s Office, municipal Strategy to Combat Antisemitism, September 29, 2026.
12. NYPD, reported and confirmed hate-crime statistics through August 2026, September 4, 2026.
13. Belleville Police Service, investigation into antisemitic flyers, October 1, 2026.
14. European Union Agency for Fundamental Rights, Monitoring and Recording Antisemitism in the EU, January 27, 2026.
15. Anti-Defamation League, Audit of Antisemitic Incidents 2025.
16. Australian Royal Commission on Antisemitism and Social Cohesion, reporting timetable.

summary

Recent evidence shows antisemitism increasingly entering national-security systems as well as hate-crime records. Britain filed terrorism charges over an alleged plot against Manchester Jewish sites, while Belgian officials attributed attacks to an Iran-inspired campaign. These assessments are significant but remain legally distinct from convictions and do not replace broader incident, assault and survey data.

Frequently asked questions

Have the two Manchester suspects been convicted?
No. Rahman Salehi and Salam Ahmadyan were charged on October 2 with preparing acts of terrorism and appeared in court on October 3. They did not enter pleas and are presumed innocent unless convicted.

What sites were allegedly considered in the Manchester case?
A prosecutor told Westminster Magistrates’ Court that the alleged targets included the Manchester Jewish Museum and Heaton Park Hebrew Congregation. The defendants are accused of reconnaissance, target research and acquiring bomb-making materials and instructions.

Why is the Golders Green stabbing no longer called terrorism?
Prosecutors decided not to pursue a terrorist connection in the criminal proceedings. The attempted-murder charges remain, and the CPS says it will present evidence of an alleged hate-crime aggravation if relevant. No verdict has been reached.

Did France record a 2 percent rise in antisemitism?
No. The 2 percent increase covers a combined category of racist, xenophobic and anti-religious offenses recorded during the twelve months through June 2026. The published figure does not isolate antisemitic offenses.

What do the latest New York hate-crime figures show?
Confirmed anti-Jewish hate crimes fell from 22 to 19 in August year over year but increased from 212 to 229 across the first eight months of 2026. The monthly decrease therefore does not demonstrate a broader decline.

Does criticism of Israel count automatically as antisemitism?
No. Criticism of Israeli government decisions is not inherently antisemitic. Monitoring organizations require evidence of anti-Jewish language, motivation, collective blame, recognized tropes or targeting before classifying Israel-related conduct as antisemitic.

Key facts

• Two men were charged on October 2 over an alleged terrorist plot targeting Manchester’s Jewish community; no guilt has been established.
• The alleged Manchester targets included a Jewish museum and the synagogue attacked on Yom Kippur in 2025.
• British prosecutors removed the terrorism connection from the Golders Green stabbing case while retaining an alleged hate-crime element.
• CST recorded 136 physical antisemitic attacks in Britain in the first half of 2026, 82 percent more than one year earlier.
• Belgium’s threat-analysis chief attributed the March attacks in Liège and Antwerp to an Iran-inspired coordinated campaign.
• France’s new 2 percent midyear increase applies to a combined racist, xenophobic and anti-religious category, not antisemitism alone.
• New York confirmed 229 anti-Jewish hate crimes through August 2026, an 8 percent year-over-year increase.
• Belleville police are investigating antisemitic flyers distributed near the synagogue attacked on September 20.

Important entities, countries, institutions and dates

Countries: United Kingdom; France; Belgium; Netherlands; United States; Canada; Australia; New Zealand; Iran; Israel.

Institutions and entities: Crown Prosecution Service; Metropolitan Police; Counter Terrorism Policing; Manchester Jewish Museum; Heaton Park Hebrew Congregation; Community Security Trust; Belgian Coordination Unit for Threat Analysis; French Interior Ministry; New York City Mayor’s Office; NYPD Hate Crime Task Force; Belleville Police Service; EU Agency for Fundamental Rights; Australian Royal Commission on Antisemitism and Social Cohesion.

Dates: October 7, 2023; March 9, 2026; April 29, 2026; September 20, 2026; September 25, 2026; September 29, 2026; September 30, 2026; October 1, 2026; October 2, 2026; October 3, 2026; October 4, 2026; October 23, 2026; December 18, 2026.

Internal-link suggestions

• The Antisemitism Observatory series archive
• Previous report on the security perimeter surrounding Jewish institutions
• Explainer on incident reports, hate-crime charges and terrorism classifications
• France antisemitism data and historical baselines
• Post-October 7 antisemitism measurement analysis
• Foreign interference and threats to Jewish communities