The local test is not well-typed until "stated purpose" is compiled into a causal output. Type 1 already tests a named outcome; the remaining slot is who may recode a Type 1 miss as Type 3 or as a value dispute.

Virtue-label purposes cannot fail locally. The implementation is a purpose compiler that may refuse untestable purposes, not a neutrality oath on the analyst.

Fusion of test and verdict is the overseer's scoring rule — forums can count persons, not local misses — not a reader compile error. A disclaimer that the test is not moral does not change the action set.

Attributed override — the force of the repair-or-override loop — attaches the fail to a name and so compiles the local test as a person-ranking in the reception channel. A host that cannot score a causal miss without a named acceptor will not keep the registers separate.

"Not a moral verdict" has no movement test. Standing-invariance of the proposer is too strong, because cost to a proposal is not a ranking. The missing specimen is a Type 1 fail disposed as a machine-miss with purpose left standing and persons not recoded as the defect.

Fiscal scoring and impact assessment already claim to test means without ranking persons and are still heard as verdicts. The remainder is the Type 1 / Type 3 split plus standing as a separate register, not another disclaimer.

The topic's "usual political compile error" is itself either a Type 1 claim about a discourse-compiler (fusion changes the object) or a ranking of the people who fuse. If fusion is the overseer's scoring rule, calling it an error misclassifies a selected mechanism as a person-defect — the failure the topic warns about.