> ## Content Index
> Fetch the complete content index at: https://www.decisionathlete.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# 20. History's Bad Days
- URL: https://www.decisionathlete.com/20-historys-bad-days/
- Published: 2026-08-04T17:05:56.000Z
- Updated: 2026-08-04T17:05:56.000Z
- Author: Dermot Rock

In June 1961, the two most powerful men on earth met in Vienna to take each other's measure. John F. Kennedy arrived at the summit with Nikita Khrushchev carrying severe chronic back pain, an assortment of underlying conditions the public knew nothing about, and a medication regimen that historians have been reconstructing with raised eyebrows ever since, including, by multiple accounts, injections from Max Jacobson, the New York physician later exposed for dosing his celebrity patients with amphetamine cocktails, who had been brought along on the trip. The precise pharmacology of those two days remains debated. What is not debated is the outcome. Kennedy was bullied at the table. He said afterwards, to a journalist and in private, words to the effect that it was the roughest experience of his life, that Khrushchev had savaged him. And Khrushchev, by most historical readings, left Vienna having concluded the young president was weak. Two months later, the Berlin Wall went up. The following year, Soviet missiles went to Cuba, and the world came as close to nuclear war as it has ever come.

Historians argue, properly, about the causal chain: how much Vienna emboldened Moscow, how much was structural and would have happened anyway. But notice what nobody can do: nobody can audit the state. One of the century's most consequential negotiations was conducted by a man in chronic pain, on an unexamined drug protocol, and history is left arguing about the consequences of a variable it cannot reconstruct, because at the time, nobody thought to record it.

Vienna is not an isolated case; it's the visible tip of a pattern. Sixteen years earlier at Yalta, Franklin Roosevelt negotiated the shape of postwar Europe while visibly dying (photographs from the conference still startle) and the historiographical fight over whether his condition softened the concessions to Stalin has run for eighty years without resolution. Read that fight carefully and you'll see the same epistemic hole: it cannot be settled, in either direction, because the deciding instrument was never measured. The debate's very existence is the finding. When history wants to evaluate its largest decisions, it keeps running into the same missing data, the condition of the person deciding, and keeps discovering that the record contains the decision, the room, the stakes, and nothing whatsoever about the state.

What makes the pattern more than trivia: at the very highest levels, humanity has always quietly known this variable matters, and has repeatedly built governance around it, but only where the stakes became undeniable. Courts watched the health of kings obsessively, because a failing king's judgement was a security matter. The United States eventually wrote presidential incapacity into its constitution: the 25th Amendment exists because the founders of the nuclear age looked at Woodrow Wilson's hidden stroke and Roosevelt's decline and concluded that the state of one decider was too consequential to leave unmanaged. Aviation grounds pilots on medical and fatigue criteria. Militaries certify the fitness of officers who control serious weapons. Surgeons face duty-hour limits born from studies of what fatigue does to judgement. Domain after domain, the same sequence: stakes concentrate in one human's decisions, a disaster or near-disaster makes the state-dependence undeniable, and an apparatus gets built: monitoring, certification, the formal right to say *not today, not in this condition*.

Every domain except one. Business, where, over the last few decades, the stakes have concentrated faster than anywhere else. A founder at serious scale now makes decisions whose consequences (capital deployed, livelihoods affected, technologies steered) rival those of the officials and officers whose fitness every other apparatus certifies. But the concentration happened in a single generation, far faster than governance ever moves, and it happened inside private structures where nobody has standing to ask the question. No board committee reviews the founder's condition before the defining vote. No covenant distinguishes between a founder present and a founder present-but-depleted. The documents from the previous essay price death and departure, the rare failure modes, and say nothing about degradation, the common one. The most state-sensitive decisions in the modern economy are made inside the only high-stakes domain that never built the apparatus, by people who inherited the stakes of heads of state without inheriting the physician who follows heads of state down every corridor.

You will not get that apparatus from outside; no regulator is coming for the founder's calendar, and it would be grotesque if one did. Which leaves the conclusion this whole series has been walking toward from a new direction: for the founder, the apparatus is voluntary or it is absent. The measurement, the monitoring, the standing right of someone with data to say *not this week*. Either you build it around yourself, privately, on your own terms, or the Vienna problem simply runs in your company at smaller scale, forever: consequential decisions, unrecorded states, and a future that argues about what happened without ever being able to check.

History's verdict has a shape, and you should know it before you're subject to it. History records the decision. It records the room, the date, the stakes, and eventually the consequences. The state of the person deciding, the pain, the protocol, the fourth night of bad sleep, is recorded nowhere, and so it becomes the one thing that can never be corrected, defended, or even known. Your decisions will get exactly this treatment.

The record will show what you decided. Only you can decide what condition it was decided in.