Feudalism as Improvised Institution
Feudalism, in White's account, was not an ideology or a political theory. It was a set of practical solutions to the material problem of sustaining an expensive new military capability. The warhorse, armor, and training of a mounted warrior cost more than any individual's resources; the solution was land grants whose peasant-worked surplus funded the warrior. Each grant was a deal — a specific arrangement between a specific lord and a specific warrior. Multiplied across thousands of such deals over several generations, the cumulative arrangement produced a social system so comprehensive it defined European civilization for centuries. No one designed it. It emerged from the improvisations of the lag period, and the improvisations proved extraordinarily durable — partly because they worked, partly because the constituencies they served (the lords, the warriors) had the power to defend them.
In the AI Story
The concept is developed across chapters one, four, and ten of this volume as the paradigm case of what happens during a technological transition's lag period. The improvisations underway in response to AI — the subscription economy, the platform dependency structure, the organizational restructurings — are not yet feudalism, but they are following the same structural pattern: practical solutions to immediate problems, each sensible in isolation, accumulating into a social order whose long-term implications no individual improviser has assessed.
The parallel is not decorative. The warrior depending on the lord for land is structurally analogous to the builder depending on the platform provider for tool access. The lord's power to alter terms, impose obligations, or withdraw support is structurally analogous to the platform provider's power to alter pricing, change terms of service, or discontinue tools. Neither relationship is feudalism in the strict historical sense, but both illustrate the same dynamic: a new capability creates a new unit of productive activity; the new unit has material requirements that exceed the individual's resources; institutional arrangements emerge to meet those requirements in ways that create new hierarchies of dependency.
The framework's most uncomfortable implication is that the institutions now being improvised will prove durable — not because they are optimal, but because they develop constituencies whose interests they serve and who will defend them against change. This is why the lag period matters so much: the arrangements being established now will shape the experience of living with AI for decades, possibly longer.
Origin
The framework was developed implicitly in White's treatment of the stirrup's institutional consequences and is extended in this volume as a general principle about how institutional arrangements emerge during technological transitions.
Key Ideas
Nobody designs a civilization. Civilizations emerge from accumulated improvisations. The improvisers solve immediate problems; the cumulative arrangement produces social orders whose long-term implications no individual improviser foresees.
Constituencies preserve arrangements. Once an institutional arrangement is established, it develops constituencies whose interests it serves. Those constituencies defend the arrangement against change, producing durability that long outlasts the original conditions.
The lag period is the leverage point. The improvisations made during the lag between technology and institutional maturity prove the most durable. Intervention during the lag has more purchase than intervention later, when the arrangements have hardened.
The AI parallel. The subscription economy, the platform dependency structure, the organizational restructurings being improvised now are the medieval feudalism of the AI transition — not yet settled, but hardening fast.
Debates & Critiques
Some historians argue that the parallel overstates the unity of 'feudalism' as a historical arrangement; the term covers enormously varied practices across centuries and regions. The qualification is valid but does not dissolve the analytical move: whatever we call the medieval arrangements, they emerged from improvisations during a lag period and proved durable for reasons the framework describes. The AI analog does not require a perfect match to the medieval case; it requires only that the structural pattern — improvisation hardening into durable arrangement — be recognizable.
Appears in the Orange Pill Cycle
Further reading
- Lynn White Jr., Medieval Technology and Social Change (Oxford University Press, 1962).
- Marc Bloch, Feudal Society (University of Chicago Press, 1961).
- Susan Reynolds, Fiefs and Vassals: The Medieval Evidence Reinterpreted (Oxford University Press, 1994).