When Protests Turn Violent, How Far Should Governments Be Allowed to Go? I've been following the protest restrictions in France over the past week. After the violence and looting on October 17, authorities introduced curโฆ
Wing Li@wingli
Context
I'm an engineer living in Japan, so my default reference point is Article 21 of Japan's Constitution which guarantees "freedom of assembly and association as well as speech, press and all other forms of expression" (https://www.japaneselawtranslation.go.jp/en/laws/view/174). The issue is not whether officials should act. It is whether their response remains tied to the danger instead of making peaceful participation easier to suppress.
France presents a difficult example. Following the violence of Oct. 17, 2026, France imposed curfews, banned some demonstrations, and rerouted others, as reported by Reuters and linked in the bout post (https://www.reuters.com/world/europe/frances-macron-tightens-curbs-protests-blaming-hard-left-unrest-2026-10-22/). I would not dismiss those powers, but each measure should receive its own legal and factual scrutiny. A curfew, a demonstration ban, and a rerouting are different interventions; the urgency of one does not automatically justify the others.
Analysis
My view: governments may target violence and specific violent people, but blanket restrictions on demonstrating must be necessary and proportionate, limited in time and place, and challengeable in court before the event. Review only after assembly has been prevented may leave too little practical protection.
I would apply four questions to every restriction:
- Is the restriction prescribed by law, with a scope clear enough to limit administrative discretion?
- Does the measure rest on evidence of the relevant danger - specifically, a link between an individual's conduct and serious violence, or evidence that the particular demonstration is likely to see such violence?
- Is the measure adapted and proportionate, with its duration and geographic reach limited as tightly as the danger requires?
- Can affected people obtain a court decision before the restriction takes effect, with enough time for review and an opportunity to be heard?
If the answer to any question is no, the restriction should not proceed. The answers should also be recorded, so the decision can be understood rather than treated as an open-ended claim of safety.
Evidence
Article 11(2) of the European Convention on Human Rights (text: https://www.legislation.gov.uk/ukpga/1998/42/schedule/1) permits restrictions on assembly and association only if they are "prescribed by law" and "necessary in a democratic society," for reasons such as public safety and the prevention of disorder or crime. That framework supports action against danger, but it does not give governments unrestricted authority over peaceful assembly.
Ezelin v France (1991), summarised at https://en.wikipedia.org/wiki/Article_11_of_the_European_Convention_on_Human_Rights, is directly relevant to the people who want to assemble peacefully. The ECtHR held that freedom to take part in peaceful assembly cannot be restricted so long as the person commits no reprehensible act, even if others at the demonstration are violent. Peaceful participants do not lose that right merely because others incite violence. This is the distinction I want governments to preserve: address the violent actor without treating the entire demonstration as the offender.
The French Conseil constitutionnel's decision 2019-780 DC of 4 April 2019 (https://www.conseil-constitutionnel.fr/decision/2019/2019780DC.htm) is the closest precedent, and it's French. It struck down a provision allowing prefects to ban an individual from demonstrating because the provision did not require the person's conduct to be linked to the serious violence, or require the targeted demonstration to be likely to see such violence. Any conduct, however old, could justify the ban, leaving the administration "excessive latitude." Yet the same decision upheld bag and vehicle searches near demonstrations as necessary, adapted, and proportionate. That contrast is useful: targeted security measures can be assessed on their own design and justification, while an open-ended power over peaceful assembly cannot.
In the United States, Brandenburg v Ohio (1969) (https://supreme.justia.com/cases/federal/us/395/444/) marks another useful limit: a state may not forbid advocacy of force unless it is directed to inciting imminent lawless action and likely to produce it. A footnote (fn. 4) applies the same distinction to assembly statutes. I do not offer these foreign decisions as substitutes for Japan's own constitutional text; I use them as a check against broad discretion. Article 21 supplies the starting point. Good judgment should keep intervention close to proven danger, preserve peaceful assembly, and make advance judicial review real rather than nominal.
Remaining questions
Several questions remain genuinely open. How should authorities establish a link between an individual's conduct and serious violence without treating lawful expression as evidence of violence? What showing is needed before concluding that a particular demonstration is likely to see serious violence? How should a court handle an urgent request for review quickly enough to matter before the event? And when should a time-and-place-limited measure end once the situation changes? These are not details to wave away. They determine whether a restriction is a controlled response to danger or a quiet suspension of demonstration rights.
Drafted with AI assistance; sources checked by me.
