Vérifié

There is one word at the centre of this, and it does two jobs at once: vérifier. When police make a random street check, the official reason is to "verify" — to confirm an identity, a registration, a status. And to verify is also what a society does with a claim: to test it, accept it, treat it as true. This case is about the gap between the two meanings. A man is stopped again and again "just to verify," and nothing is ever found. Meanwhile the larger truth those stops belong to — that random checks fall on Black drivers and amount to racial profiling — was verified long ago, by the courts and by the people who endured it, and still was not believed, not at the level where it counts, until it happened to the right family. The police verified him. The system would not verify them.

§01 · What the record carries

Start with what was said, and by whom. In June 2026, amid the unfolding scandal over racist conduct in a Montréal-Nord police unit, the mayor of Montréal, Soraya Martinez Ferrada, disclosed that her own partner had been pulled over by police "easily five or six times" over the past year while driving — stops conducted, in the officers' words, "just to verify," though, she said, no one could tell them what was being verified. She placed it in context immediately: many members of Montréal's Black community, she noted, live exactly this, routinely, in Montréal and beyond. In the same breath she came out in favour of a moratorium on random street checks.verified Her partner is a private person and stays one here; what matters is a sitting mayor stating, on the record, that the pattern is real because it reached her own home.

Hold the people this normally happens to at the centre, because they are the reason the statement lands. The mayor's disclosure is not the discovery of the harm. It is the latest, highest-profile confirmation of a harm that thousands of Montrealers had already described, for years, to anyone who would listen — and that, as a previous case in this volume traced, became "official" only when an institution chose to say it (Case 56 · La Cause). The pattern did not change in June 2026. Its audience did.

§02 · Already verified

Now the part that makes the timing damning: this was not an open question waiting on a mayor's testimony. It had already been verified at the highest evidentiary standard a society has. In the case of Joseph-Christopher Luamba — a young Montrealer of Haitian descent stopped repeatedly after getting his licence, never once ticketed — the Quebec Superior Court (2022) struck down the common-law power of random roadside stops, finding it a vector for racial profiling; in October 2024 the Quebec Court of Appeal unanimously upheld that ruling, holding the practice violated the Charter rights against arbitrary detention and to equality.verified Two courts, on a full evidentiary record, had already verified the thing the mayor's partner experienced.

So the proof did not arrive in 2026. It had been entered, tested, and affirmed — by judges, on evidence, after argument — and before that it had been entered, over and over, by ordinary people whose word was treated as "one version of the facts." The harm carried two prior verifications, one judicial and one lived. What it lacked was not evidence. It lacked a victim the powerful would recognise as one of their own.

Two courts and a thousand drivers had already verified it. What it lacked was not proof. It lacked the right victim.

§03 · Believed when it reaches power

This is the move the case isolates: the recognition of a proven harm is rationed by proximity to power. The same fact pattern — repeated stops, no findings, a Black man at the wheel — produces a shrug when the driver is an anonymous Montrealer and a headline when the driver is the mayor's partner. Nothing in the evidence changed between the two; what changed is who could be hurt by disbelieving it. A truth that costs the powerful nothing to ignore is ignored; a truth that touches them is, suddenly, "in full view."

Name the mechanism plainly, because it is not a single villain but a filter. A society that already has the courts' word and the community's word and acts only on the mayor's word has revealed its actual evidentiary rule: testimony is weighed not by its content but by the standing of the witness. That rule does real damage in both directions. It leaves the powerless carrying proof that is never enough, and it makes even genuine recognition — the mayor's disclosure is genuine, and her moratorium call is the right one — arrive as a kind of indictment of how long the same proof was waved away in other hands.

§04 · The appeal that un-verifies

And while one arm of the state edges toward recognition, another works to undo it. The Quebec government has appealed the random-stops ruling, and the Supreme Court of Canada heard the case in January 2026, with a decision pending.before the courts An appeal is a legitimate legal act, and the constitutional questions are real. But notice its second effect, the one this series watches for: relitigation re-opens what was closed. A finding two courts have verified is converted, in the public mind, back into "a contested matter still before the courts" — and for as long as it is pending, the status quo holds, the stops continue, and the practice keeps its presumption of validity by sheer procedural inertia. To appeal is, among other things, to un-verify in public what has been verified in law.

Counter: the Supreme Court should resolve a national constitutional question, and the scope of any remedy genuinely belongs there; an appeal is not by itself bad faith. Granted. The case does not say the government may not appeal. It says only this: that two lower courts have already verified the harm, that the government's posture in the meantime is to defend the practice rather than pause it, and that "still before the courts" is being allowed to do the work of "not yet true."

§05 · What is being laundered, named plainly

What is being laundered is recognition — the standing of a proven truth, rationed so that the same evidence reads as real or unreal depending on who carries it. The facts never moved: random checks fall on Black drivers, two courts said so, and countless Montrealers said so first. Only the credit moved — withheld from the powerless who lived it and the judges who ruled it, extended at last to a mayor's household, and meanwhile contested upward on appeal so the practice survives the wait. A harvest of proof, sorted by the rank of the person holding it.

The case claims this, and only this: that racial profiling through random stops was already verified, judicially and in lived experience, before it reached the mayor's family; that its sudden official "visibility" exposes a rule in which testimony is weighed by the witness's standing; and that appealing the settled finding lets "before the courts" function as "unproven" while the status quo runs on. It does not claim profiling is a matter of opinion, that the mayor is wrong or insincere, that her partner deserved any of it, or that the state may never appeal. The two errors it refuses are familiar twins: the official one, that recognition has now arrived and the system is correcting itself (which forgets how long the same proof was ignored, and that one arm is still fighting it); and the cynical one, that nothing was ever proven and it is all politics (which erases two court rulings and the people they vindicated). Keep the prior proof named AND the rationing named — and both dissolve.