The 12 scenario questions of the 2026 civic exam: the map of expected themes and the four meta-traps that keep coming back

Of the forty questions on the new civic exam, twelve are now scenario questions1. That share — less than a third — has become candidates' main source of anxiety, because it is the only one where you can't get by on having read the Citizen's Booklet. And because, on these twelve, the ministry made a singular choice: publish nothing.

This article does three things. It explains why the ministerial silence is deliberate and what it means. It draws the map of patterns that French preparation converges on in 2026, staying honest about what is published knowledge and what is calibrated hypothesis. And it highlights the four meta-traps that recur in almost every scenario question, regardless of theme — those, you can learn.

The Order of 10 October 2025 fixes the mechanics of the civic exam in force since 1 January 20261. Forty questions, forty-five minutes, a pass threshold of thirty-two. The split between knowledge and scenario isn't a detail: it appears in the order's programme.

The twelve scenario questions are distributed within this grid — without the order specifying how many fall under each theme. French preparation observes in practice a concentration on the "Principles and values" and "Life in society" themes, that is, the areas where the rule applies to a concrete case rather than to a fact to be memorised.

On its official site, the Ministry of the Interior publishes the complete list of knowledge questions for the three tracks (multi-year residence permit, resident card, naturalisation)4. For the scenario questions, it states explicitly that they "are not made public". This choice is deliberate: if the twelve prompts circulated openly, they would stop testing the capacity to apply a principle and become a simple memory exercise — the opposite of what the reform seeks to measure.

On top of this, a more prosaic reality: the exam only entered into force on 1 January 2026. As we write, four months later, the corpus of public testimony remains very thin. Specialist forums report a handful of accounts — often anonymous, rarely detailed on the exact wording.11A systematic search across Reddit, prefecture forums, Facebook groups, and French preparation blogs in May 2026 surfaces a single named account of a specific case. The rest is indirect observation or calibrated hypothesis. What circulates online is therefore, in the great majority, the product of the preparation ecosystem itself: competing sites that calibrate their "examples" against the official programme and quote each other.

The consequence is healthy for anyone preparing the exam: no list, even one sold by an "expert", contains the real questions. Everything rests on the programme — and on the capacity to apply it to a concrete case.

Reading several French preparation sites in parallel — letestcivique, eduf, prepacivique, parlez-vous-french, FrenchCitizen — a pattern emerges. The twelve motifs below recur everywhere. They don't reproduce the exam questions; they describe the families of situations that the official programme makes likely.

Principles and values

• Laïcité — public-service employee wearing a visible religious sign. Town-hall civil servant, teacher, hospital reception agent. The expected rule: neutrality applies to the agent in the exercise of their duties.
• Laïcité — public-service user wearing a religious sign. The mirror image of the previous one. The rule changes: the user may, except in state schools for pupils (Law of 15 March 2004).
• Hiring discrimination on grounds of gender, origin, or disability. "Freedom to do business" is offered as a distractor; the right answer recalls that article 225-1 of the Penal Code prohibits this refusal7.

Rights and duties

• Witness to discrimination in access to housing, a service, or a nightclub. The trap: "stay out of it". The expected response: report it (Defender of Rights, 3928, file a complaint).
• Hateful speech online, under a pseudonym. Anonymity removes neither civil liability nor criminal sanction. Candidates systematically underestimate the legal exposure.
• Authorised demonstration that drifts into property damage. Prefectural authorisation does not cover violence; responsibility remains individual.

Life in society

• Neighbours and repeated nighttime noise. The most cited motif in the published corpus of testimony. Expected hierarchy: dialogue, mediation (justice conciliator, building manager, town hall), procedure if necessary.
• Child refusing to go to school. Schooling is compulsory from 3 to 168; responsibility lies with the parents and the sanction exists.
• Request for a school exemption on religious grounds (swimming, mixed-gender outing, natural sciences). The "compromise" is offered; the rule is clear: attendance and mixing are not negotiable.

Institutional system · History and culture

• Voter card received, voting procedures. Tests the secrecy of the ballot, the age (18), and the dual nature of voting as a right and a civic duty.
• Republican symbol recognised in context (flag, the Marseillaise, the motto). A simple scenario, testing cultural familiarity more than memorisation.
• Visible religious sign at work in a private company. Meta-trap no. 2 in its purest form: the answer is not the same as in a state school or a public service.

Reading the twelve motifs above, the same kinds of wrong choices keep coming back. Effective preparation isn't about memorising thirty particular situations: it's about recognising the four traps that repeat.

1. The polite answer is almost always wrong

On nearly every scenario question testing a discrimination, a refusal, an exemption, a disruptive behaviour, or a report, the most diplomatic distractor is wrong. "Find a compromise", "respect everyone's culture", "don't make things worse" sound civilised; they misread the rule. The exam evaluates the application of law, not personal diplomacy.

2. Public and private don't follow the same rule on laïcité

Laïcité applies to the State and its agents (Law of 9 December 19056). It does not apply to ordinary users, nor to private-sector employees — except where a justified neutrality clause exists. This public/private confusion is, according to several preparation sites, the leading cause of error on scenario questions. It appears across several motifs in different guises (municipal agent, mini-market employee, schoolchild).

3. A "freedom to" is never absolute

Freedom of expression, freedom to do business, freedom of religion, freedom of conscience: each is regularly offered as the right answer when in fact a legal limit applies to the case. The most tested limits: hate speech and defamation for expression, discrimination for business, breach of public order for religion, legal majority for conscience.

4. The hierarchy of recourse must be respected

On motifs of neighbours, discrimination, harassment, the right order is always the same: dialogue first, mediation next, authorities if necessary. Jumping straight to the police is wrong; doing nothing is wrong. The right answer knows how to name the intermediate level (justice conciliator, town hall, building manager, Defender of Rights, consumer mediator, 3928).

Forty-five minutes for forty questions is just over a minute per question. On scenario questions the pace is tighter because the prompt is longer. Three quick passes are enough:

1. First pass — the principle at stake. Read the prompt looking not for "what would you do?" but which principle is being tested. Laïcité? Equality? Freedom to demonstrate? Compulsory schooling? Discrimination? Once the principle is identified, you almost have the answer.
2. Second pass — eliminate the extremes. Of four choices, two are almost always easy to discard: the one that "does nothing" and the one that "calls the police right away" are rarely correct. Two plausible choices remain.
3. Third pass — apply the meta-traps. Between the two remaining choices, ask: which is more diplomatic? If one is clearly more polite or more "accommodating", it's probably the other one that's correct. Also check whether a "freedom to" is being invoked out of place.

The ministry publishes no scenario question and has authorised no third party to do so. Any list claiming to reproduce the exact prompts is, by construction, either reconstructed from partial accounts or invented. Three reliable warning signs:

• The argument from authority. "List obtained from a prefecture trainer", "real exam questions", "internal ministry document". None of these can exist publicly. Real prefecture trainers themselves point out that they don't know the prompts.
• The single, unverifiable source. An anonymous account on a forum, a screenshot without context, a PDF with no author. If the same list appears identically on three different sites, it's because they're quoting each other — not because they have a source.
• The polite "right answer". If the list offers the most diplomatic response (compromise, waiting, withdrawal) as correct, it was written by someone who has never prepared a civic exam. The construction rule for official multiple-choice questions is the opposite.

We write our own scenario questions. None is copied from a list circulating online. Each prompt is calibrated against the five themes of the Order of 10 October 2025, validated against the four meta-traps described above, and accompanied by an explanation that cites the law or the republican principle that grounds the right answer.

200+ scenario questions today form our reference library, and we publish new ones each week as circulars and case law evolve. Difficulty is calibrated by track: accessible for the multi-year residence permit, intermediate for the resident card, demanding for naturalisation.