MacCulloch says it flatly: "We underestimate the role of lunacy in history. Some people in the past are just mad." The sentence comes near the end of the interview, after the carrot game pedagogy passage, while he's discussing the bizarre-to-humility chain and what entertaining history actually permits.
The provocation is sharper than it sounds. He gives the Sinclair Earl who built Rosslyn Chapel as one case — "had lots of money, but I think he was also mad. He had enough money to indulge his lunacy to build this really quite silly building, which is beautiful, but isn't very sensible." He notes that historians keep theorizing about Rosslyn — Templars, Holy Grail, mystical geometry — when the simpler explanation is that the patron was insane in the terms of his own period and had enough money to act on his insanity without correction.
Then he gives Henry VIII's last decade as the ill-but-not-mad case: in constant physical pain, irrational because of the pain, mood governed by how he felt that morning. A doctor had to point this out to MacCulloch; historians had been treating Henry's last-year decisions as political maneuvering when much of the variance was actually pain-driven mood.
What MacCulloch is doing here is methodologically aggressive. He's saying that the standard historian's apparatus systematically over-rationalizes the historical record, and that the corrective is to make explicit room for lunacy, illness, and irrationality-via-physical-state in the explanatory toolkit.
First wire (obvious): Histories of "great men" should pay more attention to the actors' mental and physical health. Reductive but reasonable.
Second wire (deeper): The over-rationalization isn't accidental; it's constitutive of the discipline's training. Historians are trained to find motives, calculations, strategic logic. The training is what produces the systematic blind spot. The fix isn't more training in the same direction; it's the explicit addition of non-rational explanatory categories to the historian's toolkit, with the discipline of using them when the evidence demands.
Third wire (uncomfortable): The same correction probably applies to contemporary readings of contemporary people. The colleague whose behavior makes no sense, the family member whose decisions seem self-defeating, the public figure whose choices appear inexplicable. Some of them are mad. Some are ill. The rational-actor model is the standard mode in which we read each other, and the same systematic over-rationalization that produces bad history produces bad personal-relationship readings. If MacCulloch is right at the historiographic level, most of us are mis-reading most of the people around us most of the time, because the model we're using doesn't have slots for the conditions actually shaping their behavior.
Adjacent: Lunacy as Historical Explanatory Variable (the concept page this spark feeds into), Cross-Discipline Nerving (the doctor told MacCulloch about Henry VIII), Narcissism Spectrum (the neighboring slot).
Cross-domain: connects to depth-psychology literature on personality structure and to the philosophy-of-madness literature (Foucault). Also connects to behavioral-mechanics narcissism literature where Greene and Lieberman map patterns that look mad from outside but are intelligible from inside the narcissism. The distinction mad / narcissistic / ill / mismatched is currently underdeveloped in the vault.
The gap: vault has plenty on individual psychology and plenty on historical figures, but the intersection — "how do you read historical figures whose internal states are non-standard?" — is undertheorized. MacCulloch's blunt provocation is generative because it forces the intersection.
Essay seed: The Rational-Actor Failure Mode in History and Life. The angle: catalog historical figures over-rationalized by standard historiography, name what the lunacy or illness or mismatch interpretation would look like in each case, then extend to contemporary parallels. Risk: easy to slide into amateur diagnosis.
Open question filed: For the historical figures most over-rationalized by current scholarship, who are the cross-discipline practitioners (doctors, psychiatrists, neurologists) whose perspective would reveal what historians have habitually missed? Are those conversations happening?
Concept page potential: Rational-Actor Failure Modes in Interpretive Work — a more general page covering history, biography, criminology, journalism, and personal relationships. The over-rationalization may be a pan-disciplinary problem with the same structural fix.