# Choose a mode for assumptions ## Intention I want to evaluate an assumption in an explicitly chosen mode. Each mode accepts only the fact origins it names. ## Incorrect form and why it stays silent A typo does not select the nearest mode. ```law title="Incorrect form" context { mode simulaton; } ``` The correct word simulation also matters: outside it, assumed_for_simulation is not accepted. ## Correct form Mode is set as a context axis. ```law language "law.core" version "0.2"; package recipes.n.r08 version "0.1.0"; namespace "urn:recipe:n-package:08"; relation assumed(); case Experiment { context { mode simulation; } assert assumed() { origin assumed_for_simulation; } } ``` ## Frozen execution scene | Facts | Question | Answer | |---|---|---| | assumed_for_simulation, simulation | assumed() | TRUE_ONLY | | Same, compliance | assumed() | NEITHER, ASSERTION_NOT_ACCEPTED | ```law test "simulation accepts assumption" { given { context { legal_time @2026-09-13; decision_time @2026-09-13T09:00:00+05:00; knowledge_time @2026-09-13T09:00:00+05:00; timezone "Asia/Almaty"; mode simulation; } assert assumed() { origin assumed_for_simulation; } } evaluate truth(assumed()); expect truth_status == TRUE_ONLY; } ``` ```law test "compliance rejects assumption" { given { context { legal_time @2026-09-13; decision_time @2026-09-13T09:00:00+05:00; knowledge_time @2026-09-13T09:00:00+05:00; timezone "Asia/Almaty"; mode compliance; } assert assumed() { origin assumed_for_simulation; } } evaluate truth(assumed()); expect truth_status == NEITHER; expect issue(ASSERTION_NOT_ACCEPTED); } ``` ## Counterfactual `teaches` corrupts the mode name and requires LDC-E1353. The second scene keeps the fact, changes only the lawful mode, and pins the acceptance rejection. ## Boundary Mode does not rewrite norms. Mode may affect the result only through an explicit policy; a court fact and a judgment belong to the evidence and interpretations books. ## Pitfall An assumption outside simulation stays Not established, not refuted, with ASSERTION_NOT_ACCEPTED. The fix is to choose simulation explicitly, keeping the origin, not to recolour the fact as case_input.