What Success Leaves Behind

Successful containment is commonly treated as evidence that a system remained capable or learned from experience. That inference is underidentified. The same acceptable outcome can leave greater internally owned capacity, unchanged capacity, residual burden, persistent vulnerability, or dependence on external support. This article develops a capacity inheritance framework for repeated-load systems. The inherited state contains adaptive capacity, residual burden, vulnerability, and external scaffolding. Realized load also depends on the operational state selected before impact, allowing learning to work by reducing the next demand rather than enlarging response capacity. The framework distinguishes regenerative, preservative, consumptive, protected, and maladaptive success; separates immediate reaction from durable inheritance; and specifies observation, verification, implementation, testing, persistence, and autonomy as sequential conversion gates. A process trace of weather-related loss-of-offsite-power events at Pilgrim Nuclear Power Station from 2013 through 2018 illustrates the framework. The sequence shows partial procedural learning, recurrent electrical exposure, externally scaffolded recovery, and terminal censoring when permanent shutdown ended the opportunity to test autonomous persistence. A stylized numerical demonstration shows that identical successful outputs can rest on different combinations of internal capacity, external support, and anticipatory load shaping. The analysis shifts resilience assessment from whether a system survived an episode to what state the episode transmitted forward.