Central limit theorem

The central limit theorem (CLT) is the proposition that the sum of a large number of independent, identically distributed random variables converges to a normal distribution, regardless of the underlying distribution's shape. It is the mathematical engine behind the ubiquity of bell curves in nature: measurement errors, biological traits, and aggregate behavior all tend toward Gaussianity because they are the sum of many small, uncoordinated influences.

The theorem has a dark side. Its conditions — independence, identical distribution, finite variance — are precisely what break down in systems with feedback, memory, or interaction. The CLT is not a universal law but a limiting theorem for a specific class of systems: those that are decomposable, stationary, and non-interacting. When these conditions fail, as they do in financial markets, social networks, and ecosystems, the CLT becomes not an explanation but a misdirection — a reason to expect normality where none exists.

The CLT's dominance in statistical pedagogy has produced a generation of scientists who see the bell curve as the natural state of the world and everything else as deviation. This is backwards. Normality is the special case; non-normality is the rule in complex systems.