Nuclear Power
Nuclear power is the use of nuclear reactions — specifically nuclear fission — to generate electricity. Despite its low carbon emissions and high energy density, its development has been shaped less by engineering capability than by the institutional trauma of accidents like Three Mile Island, Chernobyl, and Fukushima, which demonstrated that the socio-technical systems surrounding reactors are often less robust than the reactors themselves. The debate over nuclear power's role in decarbonization is therefore not merely a technical question about reactor design or waste management; it is a question about whether human organizations can be trusted to manage systems whose failures are normal rather than exceptional.
The Engineering Promise
Nuclear fission releases roughly one million times more energy per unit mass than chemical combustion. A single uranium fuel pellet the size of a fingertip contains the energy equivalent of a ton of coal. This energy density makes nuclear power uniquely suited to baseload electricity generation: a single reactor can power a city of millions, with negligible carbon emissions during operation and a land footprint orders of magnitude smaller than solar or wind at equivalent scale.
The engineering of nuclear reactors has advanced considerably since the first commercial plants of the 1960s. Generation III and III+ reactors incorporate passive safety systems — gravity-fed cooling, natural circulation, and negative reactivity feedback coefficients — that do not require active human or electronic intervention to prevent meltdown. Generation IV designs promise even greater safety margins, including reactors that cannot melt down by physical design and reactors that consume rather than produce long-lived nuclear waste.
But the history of nuclear power demonstrates that engineering capability is not the binding constraint. The binding constraint is socio-technical: the capacity of the organizations that design, build, operate, and regulate reactors to manage systems that are both interactively complex and tightly coupled — the structural conditions that Charles Perrow identified as producing normal accidents.
The Socio-Technical Problem
Every major nuclear accident has followed the same pattern: a technically manageable initiating event — a feedwater pump failure, a turbine shutdown, a power surge — cascaded through the system because the human and organizational components of the socio-technical system could not respond in time. The Three Mile Island accident (1979) began with a stuck relief valve and a misleading indicator light. The operators followed their training. The reactor melted down anyway. The problem was not operator error. It was that the system's design made correct understanding impossible under the time pressure of cascading failure.
The Chernobyl disaster (1986) was different in its specific causes — a safety test conducted with the reactor in an unstable configuration, operators who disabled safety systems, a design flaw in the control rods — but structurally identical. A complex, tightly coupled system was operated outside its design envelope by an organization whose safety culture had been eroded by production pressure, schedule demands, and the normalization of risk.
The Fukushima disaster (2011) extended the pattern to natural disaster: a tsunami overwhelmed the plant's protective seawall, disabled the backup diesel generators, and led to the meltdown of three reactors. The initiating event was external, but the cascade was internal — the product of a regulatory culture that had accepted a seawall height inadequate for the historical tsunami record, and an organizational structure that placed the diesel generators in vulnerable locations. The accident was not merely a natural disaster. It was a normal accident: the statistically expected output of a system whose architecture made catastrophic failure structurally possible.
The Efficiency–Resilience Tradeoff
Nuclear power exemplifies the efficiency–resilience tradeoff in its most acute form. Reactors are extraordinarily efficient: a small amount of fuel produces enormous energy. But this efficiency comes at the cost of resilience. The system has no slack. A reactor cannot fail