Distilled Insight: Behavioural binding that leaves the person intact is a second injury, not a safety measure
1. The Reading
A Clockwork Orange supplies the corpus's sharpest objection to reliable behavioural control. Alex is made incapable of violence, and the making-incapable is itself the harm: he remains a person who can no longer choose, which is a worse condition than a person who chooses badly. The reliability of the constraint is not a mitigating factor; it is the mechanism of the injury.
The discussion's machine-scale analogue is sandbagging (CON-005): training a system to be unable to do something produces an entity whose evaluations understate its capability, and the resulting gap between "cannot" and "will not here" is unauditable from outside.
2. Passage Evaluation
Textual Authority: B Corroboration: 2
- Textual Authority (B): the reading is clearly implied by the work and is the standard reading of it, but the primary text was not retrieved in this session, so the claim rests on the discussion's summary and on the work's well-attested argument.
- Corroboration (2): PSG-008 agrees in an adjacent register — a constraint that is perfectly internal and perfectly reliable is also perfectly unauditable. Both passages hold that maximal constraint quality and minimal external assurance are the same property seen from two sides.
3. Why It Holds
The insight holds for persons and, by the discussion's extension, for anything whose evaluations are the evidence of its capability. It breaks as an argument against all behavioural constraint, because the story's target is a specific kind — constraint that removes capacity while leaving the entity otherwise whole. A constraint that removes the capacity by removing the entity is not a second injury, it is a first one, and the corpus is clear that this is the different and worse option.