Mid-century utopian visions resulted in some of the greatest tragedies in history. What went wrong?


Legibility - the state can’t control or tax you if it can’t see you, so it tends to want to force the world to conform to some simple model that is easy to measure and manipulate. Standardizing on common vocabulary across realm of management. Not inherently a bad thing - enables economies of scale and low friction exchanges. But usually also moves power and resources to the center too, and that often becomes the underlying motivation eg smallholders usually outperform plantations in raw productivity but they are harder to tax and control so various states have forcibly pushed for plantations instead.

Legible, official model is often underpinned by a complex, informal and illegible world of norms, favors, networking etc. Work-to-rule strikes are a vivid demonstration of what happens without this underpinning.



Create legibility by imposing simplified model onto a complex system. Again, not inherently a bad thing eg universal education, food and drug standards, herd immunity via compulsory vaccination, trade agreements that enable low-friction commerce etc.

Becomes dangerous when we add ideology (blind to failure) and power (can overrule resistance). Most social interventions don’t work, so intervention without feedback is very likely to be bad.

High modernism - scientism as a religion. Success of the scientific process in some domains created huge over-confidence in others (cf Uncontrolled on how generalizing from small numbers of experiments fails in complex domains). Wrapped up in ideas of Western and academic superiority - coming to teach the peasants what they are doing wrong, so any resistance to changes is interpreted as stubborn backwardness without merit. Glorifies simple generalizable laws and models, and discounts complex, locale-specific knowledge. Really into grand plans to rebuild society from scratch.

Aesthetic of rationality/legibility - tidy, orderly, straight lines, right angles, symmetry. Often takes precedence over even optimizing the target variable - strong assumption that the optimal model must be an orderly one.

Authoritarian planning tends to be frozen - it doesn’t account or allow for any change or evolution in the nature of the city or the use of its spaces.

Examples of failed interventions:


Contrasting metis and techne. Complex, implicit knowledge specific to a single locale vs simple, explicit rules that generalize across a wide variety of situations. Techne lends itself to economies of scale in creating, applying and teaching knowledge. Metis requires long experience which means that workers are not fungible; this is a roadblock to centralization and scaling eg craftsmen vs assembly lines. Led to overestimating the value of techne to the point of discounting metis entirely.

Closing section gives advice on avoiding such catastrophes, which is almost point for point the same as the advice in Uncontrolled.