Part II: The Persistence Engine

Why Wrong Ideas Stay


Part I explained how wrong ideas get in. That is the less interesting half of the problem. The truly consequential question -- the one that determines whether a wrong consensus costs a field five years or fifty -- is why wrong ideas stay. How does a demonstrably incorrect belief survive the accumulation of counter-evidence, the objections of informed critics, and the visible failure of its predictions? The answer is not that people are irrational. The answer is that the systems surrounding the belief are doing exactly what they were designed to do: protect the consensus.

These eight chapters map the machinery of persistence. You will see how institutional sunk costs -- careers built, textbooks written, Nobel Prizes awarded -- create switching costs so enormous that abandoning a wrong answer feels more threatening than defending it. You will examine the replication problem, where findings that cannot be reproduced continue to be cited because nobody checks. You will learn how incentive structures manufacture error at industrial scale, rewarding the production of publishable results over the production of true ones. You will encounter precision without accuracy, the dangerous state in which a field's numbers look impressively exact while being systematically wrong. You will study the Einstellung effect, where expertise itself becomes a trap, blinding experienced practitioners to solutions that a novice would see. You will watch consensus enforcement machines punish dissent, observe complexity hiding inside apparent simplicity, and meet the zombie idea -- the belief that has been killed by evidence multiple times and keeps coming back.

The persistence engine is not a conspiracy. It is an emergent property of institutions that value stability, reward conformity, and penalize the kind of disruptive correction that progress requires. Understanding how it works is the prerequisite for understanding how to break it -- which is the subject of Part III. But before we can talk about correction, we need to understand exactly what correction is up against. The forces documented in these chapters are formidable, interlocking, and largely invisible to the people operating inside them. Making them visible is the first step toward overcoming them.

Chapters in This Part