A bubble is a price that has come loose from its anchor. Assets are ordinarily worth the discounted stream of what they will pay; in a bubble they are worth what the next buyer will pay, which is a different and more fragile proposition. The logic is internally coherent — buying an overpriced asset is rational if someone more optimistic is behind you — right up to the moment the queue of greater fools ends.
The canonical year is 1720, which produced two bubbles at once: the South Sea Company in London and John Law’s Mississippi scheme in Paris. Both wrapped a thin commercial story around a machine for converting government debt into shares, both drew in a public that had never before owned securities, and both collapsed within months of their peaks. The pattern they established — a plausible new-era story, easy credit, broadening participation, then sudden reversal — has needed no updating since.
What changes between episodes is the story, not the shape. In 1929 it was the permanent prosperity of the American industrial machine; in 2000 it was the internet’s abolition of the old rules of valuation. The stories matter because they are load-bearing: a bubble is a coordination device, and it holds exactly as long as the narrative that coordinates it.
Bubbles are easy to name in retrospect and disputed in real time — every one had contemporaries who called it and were early enough to look wrong. The atlas treats the diagnostic question as less useful than the structural one: how much leverage is under the price, and who is holding it when the narrative breaks.