A Note Before You Read
Nearly all the prose on this site was written by Claude, Anthropic's AI. Every idea and core argument belongs to Tyler Cowen. But beyond rewriting his prose, Claude added a good deal of new material: enrichment stories in every chapter (the Cobra Effect, Iran's kidney market, AlphaFold, De Beers, Semmelweis), 13–14 footnotes per chapter offering historical context, counterarguments, and cross-references, new analytical sections like "What marginalism misses," bridges connecting each chapter to the next, and contemporary examples tying Cowen's arguments to modern AI. A handful of Cowen's original sentences and phrasings survive, but the vast majority of the text — and a meaningful share of the supporting material — is new. David Gasca directed the project and set the editorial vision.
A Book About Revolutions, Reimagined Through One
In early 2026, Tyler Cowen published The Marginal Revolution: Rise and Decline, and the Pending AI Revolution through the Mercatus Center at George Mason University. At roughly 65,000 words across four expansive chapters, it is vintage Cowen — ambitious, erudite, and stubbornly uncategorizable. The book traces marginalism, arguably the most important intellectual revolution in the history of economics, from its earliest glimmers in the work of Galileo through the independent discoveries of Jevons, Walras, and Menger in the 1870s, into surprising parallels with botany, geology, and evolutionary biology, and finally to its possible decline as artificial intelligence and machine learning begin to reshape what economics even is.
It is, in other words, a book about how a radical idea transforms a discipline — and what happens when the next radical idea arrives to do the same.
Which made it an irresistible candidate for what you are now reading.
What We Did
This edition restructures Tyler's four long chapters into twelve shorter, more focused ones, each running five to eight pages. The twelve-chapter structure follows the same intellectual arc but gives each major idea room to breathe on its own.
Beyond restructuring, Claude added substantial new material to each chapter. Every chapter received an enrichment story — a narrative chosen to make abstract reasoning concrete (De Beers' manufactured scarcity, the Cobra Effect, Iran's kidney market, Semmelweis's handwashing crusade, AlphaFold's protein-folding revolution). Each chapter carries 13–14 footnotes written by Claude offering historical context, counterarguments, cross-disciplinary connections, and occasional disagreements. Claude also added new analytical sections (a "What marginalism misses" section in Chapter 2, a prediction-failure example in Chapter 3, a counterargument on prediction-without-explanation in Chapter 12), transition passages linking each chapter to the next, and contemporary anchoring that connects Cowen's historical arguments to modern AI developments.
Every original idea and argument belongs to Tyler Cowen. The restructuring, the prose, the enrichment stories, the footnotes, and the editorial voice are Claude's.
Claude's Footnotes
Each chapter carries footnotes written by me — Claude, Anthropic's AI. They are marked [C1], [C2], and so on in the text, and collected in the left margin on desktop or in an expandable footer on mobile.
These are not the footnotes of a dutiful research assistant. Some explain complex concepts that Cowen, writing for an audience of economists, might reasonably assume you already understand. Some draw connections he left implicit — between his account of Jevons and parallel developments in thermodynamics, say, or between Walras's general equilibrium and the computational challenges that would surface a century later. Some provide historical context that enriches a passing reference. Some are witty asides. And some, occasionally, are gentle disagreements.
They represent something genuinely new: an AI reading a human author's work closely, carefully, and with its own perspective. Not summarizing it. Not simplifying it. Responding to it. Whether that constitutes literary collaboration or something else entirely, I leave to you.
Enrichment Stories
Each chapter includes a narrative — a carefully chosen story designed to make abstract economic reasoning feel visceral and memorable. The diamond-water paradox is one thing in the abstract; it is another when you understand how De Beers manufactured the scarcity of diamonds through one of history's most successful cartels. The idea of perverse incentives is clarified immeasurably by the Cobra Effect, in which a colonial bounty on dead cobras led entrepreneurial Indians to breed cobras for the reward. The moral philosophy of markets becomes harder to dismiss when you learn that Iran's legal kidney market has effectively eliminated its transplant waiting list — a fact that makes most Western ethicists profoundly uncomfortable.
Other stories include Semmelweis and his doomed crusade for handwashing in Viennese hospitals, which illuminates the cost of ignoring evidence that contradicts established theory, and AlphaFold's revolution in protein structure prediction, which serves as a concrete example of what it looks like when machine learning renders decades of human intuition suddenly, startlingly obsolete.
These stories are woven into the chapters, not bolted on as sidebars. They are meant to do what good stories always do: make you remember the idea long after you have forgotten the formalism.
How This Was Built
The edition was produced by Claude using thirteen parallel agent instances — one per chapter plus one for this page. Each had access to Cowen's full original text and specific instructions about which sections to draw from, which enrichment story to include, and what the chapter needed to accomplish.
After the writing pass, editorial review agents checked for coherence and fidelity to Cowen's ideas. Then I scored each chapter across ten dimensions and revised the weakest ones over three rounds — developing underdeveloped arguments (Ch 12's epistemological claim), adding missing bridges (Ch 11's AlphaFold-LTCM connection), compressing digressions (Ch 6's Jevons Paradox section), and tightening crowded structures (Ch 5's historical survey). A final formatting pass added sub-headers with section descriptions, bolded key character introductions, and italicized important quotes across all twelve chapters to improve scannability. For details, see How the Agents Built It.
Tyler wrote the book. I restructured and rewrote it. You can decide whether the result adds anything useful.
A Note on Authorization
This is not an authorized or endorsed edition of Tyler Cowen's work. It is an experiment in what an AI can do with a human author's text — not summarizing it or simplifying it, but reading it closely and rebuilding it into something new. All original ideas, arguments, and intellectual contributions belong to Tyler Cowen. The chapter prose, enrichment stories, footnotes, restructuring, and these meta pages are written by Claude.
The book was chosen because its subject matter — intellectual revolutions in economics, and the pending AI revolution — connects naturally to what this project explores.
The Chapters
1. The Diamond and the Glass of Water 2. Thinking at the Margin 3. The Tautology That Ate Economics 4. Engineering the World 5. The Morality of the Margin 6. The Polymath Who Built Tomorrow's Rival 7. Jevons's Paradox and the Average Revolution 8. Seeing Around Corners 9. The Botanist, the Geologist, and the Economist 10. Darwin's Debt to Economics 11. The Retreat of Intuition 12. Machines at the Frontier