The thread installs Chernobyl Unit 4 as a case in which a positive void coefficient, a shutdown rod that could insert reactivity, and a test procedure that isolated emergency cooling were already properties of the plant. The unresolved question is which repair, required before the rundown test, would have closed that channel. Forbidding jumpering of ECCS and the turbine trip is one rule: the Chief Engineer could authorize the isolation, and it sat in the test programme. Redesigning the graphite displacers and reducing the void coefficient is a different first rule: Ignalina had named the positive scram in 1983, and design documents that called the void coefficient negative were already false for the burned core. Putting the operating reactivity margin on the console as a safety limit, not a spatial-control number ten minutes late and fifty metres away, is a third. Those are not substitutes. A jumper ban with the rods still adding reactivity on AZ-5 still lets a fully withdrawn bank destroy the core. A rod redesign with ECCS still isolatable for eleven hours still lets a test run without emergency cooling. An ORM display without a negative void coefficient still leaves the physics. INSAG-1, INSAG-7, and the two Soviet annexes already record all three. They do not say which one, required when the test was written, would have turned a signed rundown programme back into a non-decision for the core.