Theories of instability, tested against the record
- Investigate
- Theories
- Data and accuracy
Investigate now brings eleven theories from the comparative literature to every case, and a new Compare cases mode tests them across the whole atlas.
- A library of eleven theories: Goldstone’s demographic-structural breakdown, Turchin’s fathers-and-sons cycle, Ibn Khaldun’s dynastic cycle, Huntington’s modernisation gap, Skocpol and Tilly on war and revolution, Tilly’s multiple sovereignty, Hilton and Cohn on plague and revolt, climate shock (Manning, Büntgen, Parker, Campbell), Fearon and Laitin on state weakness, the conflict trap (Collier, Hoeffler, Walter) and Toynbee’s disintegration. Each carries its sources, its strongest objection, and conditions the atlas can test for any state and year.
- “What the theories point to” sits directly under “What led to it” and gives every theory one of four answers: its conditions were present, they were tested and absent, they cannot be tested here, or the theory does not ask about this kind of subject.
- Where a theory’s conditions are met, the things it names become causes “by a theory”, cited to its authors and ranked below anything a source says about the event. The Peasants’ Revolt of 1381 now shows the Black Death in England (Hilton, Cohn). You can filter these links by theory.
- Compare cases: each theory’s rate across the whole record, with and without its conditions, by region and by how well documented the period is, with the near-misses where the conditions held and nothing followed.
- Compare up to six cases side by side. An empty cell says why it is empty: checked and nothing found, not known yet, not applicable, or outside the years a dataset covers.
- “What if a cause is removed” counts what goes with a cause, and never claims to know whether the event would still have happened.
- Instability on record, by region and half-century, drawn only where enough states and events are recorded to say anything; the rest is hatched and says why.
- A test of chance: do changes of ruler fall near volcanic winters or severe droughts more often than chance would put them? On this record, no.
- The area each state held, year by year, measured from Cliopatria’s borders: the Roman Empire at 5.26 million km² in 117.