Where Good Ideas Come From

Johnson argues that innovation is an environmental and network property rather than an individual one, and identifies seven patterns that produce it: the adjacent possible, meaning that innovation moves into the space of what current components make reachable rather than jumping arbitrarily; liquid networks, meaning environments where ideas connect and recombine; the slow hunch, meaning that most breakthroughs develop over years as incomplete intuitions before they resolve; serendipity, meaning accidental connection between unrelated ideas; error, which he argues produces innovation by forcing exploration outside the productive path; exaptation, meaning the repurposing of something developed for another use; and platforms, meaning that innovations create the conditions for further innovations built on top.

His evidence includes historical cases such as Darwin's notebooks showing the theory developing years before he reported the insight, the coffee house as a physical liquid network, the double-entry ledger, Gutenberg's use of the wine press, the invention of the printing press from existing components, and the development of GPS.

He also presents an analysis of major innovations across four hundred years classified by whether they came from individuals or networks and from market or non-market settings, and reports that networked non-market innovation dominates, particularly in the recent period, which he uses to argue against strong intellectual property protection.

The historical cases are chosen after the outcome and the seven patterns are broad enough to describe almost any innovation retrospectively. The classification exercise depends heavily on how each innovation is coded. The adjacent possible and exaptation are the two concepts with the most analytical use.