ExpertsFMWhat Robert Caro learned about power after studying it for 50 years
Dwarkesh Podcast · October 30, 2026
5 Selected Lessons:
heuristic · 3 case · 2
Refuse to take no for an answer when chasing a necessary source
Source
Dwarkesh Podcast — What Robert Caro learned about power after studying it for 50 years (October 30, 2026) · at 0:00 ▸
When this applies
when you need information or buy-in from someone who has no obligation to give it
The takeaway
Treat 'no' from a necessary stakeholder as a starting position, not a final answer, when the information is essential to the work.
Why it works
Persistence signals seriousness and often outlasts the other party's reluctance, especially when the person eventually feels their story deserves telling.
One level up
The people who 'do everything' to get what they need usually get further than those who stop at the first refusal.
Where it stops holding
This stops working when persistence tips into harassment or when the subject has nothing to gain and everything to lose from talking, making continued pressure counterproductive or unethical.
Detail
rule: If someone says no to an interview you feel you must have, don't accept it as final — keep pursuing it.
threshold: Applies when the source is irreplaceable to understanding the subject (e.g., the only person present for key events).
rationale: Caro says 'you feel there are certain people you have to interview... if they don't want to give you an interview, you can't, I feel, you can't say no.'
In their words
Well, you feel there are certain people that you have to interview. SPEAKER_00: And if they don't want to give you an interview, you can't, I feel, you can't say no.
Leverage, not persuasion, is the real engine behind a powerful persuader
Source
Dwarkesh Podcast — What Robert Caro learned about power after studying it for 50 years (October 30, 2026) · at 32:51 ▸
When this applies
when evaluating why someone's persuasive style actually works on other people
The takeaway
Don't mistake someone's skill at influence for the true source of their power — look for what they can give or withhold that the other party actually needs.
Why it works
A persuasive interaction only moves someone when the persuader also controls outcomes the target cares about (committee assignments, legislation, favors); the charisma is the delivery mechanism, but the leverage is the actual cause of compliance.
One level up
Soft power (charm, persuasion) is usually parasitic on hard power (control over resources and outcomes), not a substitute for it.
Where it stops holding
This stops working when the persuader genuinely has no leverage over the other party — in that narrower case, pure persuasion by argument or charisma may be the whole story.
Detail
rule: Behind persuasion, look for the leverage — what the persuader can give you or deny you.
threshold: Applies whenever a powerful figure's charm or personal force is the visible mechanism of influence over someone with something to lose or gain.
rationale: Caro: 'Behind his persuasion was the power... it's not just the force of his personality that he's bringing to bear on you. Behind that personality, you know he has the power to do things to you or for you.'
In their words
Behind his persuasion was the power. SPEAKER_00: When you were talking to him, SPEAKER_00: if he asked something of you, SPEAKER_00: you knew he had the power behind him to give you or deny you things that you wanted.
Digital archives speed finding documents, not understanding them
Source
Dwarkesh Podcast — What Robert Caro learned about power after studying it for 50 years (October 30, 2026) · at 43:11 ▸
When this applies
when evaluating whether a new information-access tool will meaningfully accelerate deep analytical or research work
The takeaway
Before assuming a tool that speeds up access to information will speed up the whole task, identify whether access or comprehension is the true bottleneck.
Why it works
The bottleneck in deep research is reading, synthesizing, and understanding material, not locating it; a tool that only speeds retrieval leaves the actual rate-limiting step untouched.
One level up
Tools that remove friction from the easy part of a task don't necessarily shrink the time the hard part takes.
Where it stops holding
This stops working when the research task genuinely is retrieval-bound (e.g., simple fact-checking) rather than comprehension-bound, in which case faster access does meaningfully speed the whole task.
Detail
rule: Finding documents is a small thing compared to reading them.
threshold: Digital search helps locate where oral histories or documents are archived, but Caro says it 'wouldn't have made things' go meaningfully faster for his actual process.
rationale: Caro: 'I can't quite tell you why, but certainly you could find the documents faster. But finding them is a small thing compared to reading them.'
In their words
But finding them is a small thing compared to reading them.
Robert Moses builds Jones Beach's road despite hundreds of hostile property owners
Source
Dwarkesh Podcast — What Robert Caro learned about power after studying it for 50 years (October 30, 2026) · at 5:04 ▸ · at 8:45 ▸
When this applies
when a scarce, dispersed set of veto-holders can each block progress on a large undertaking
The takeaway
Large-scale obstruction by many small, uncoordinated opponents is often a solvable sequence of individual problems, not a single blocking wall — if someone has the will to work through them one by one.
Why it works
A determined actor who simply decides 'I'm going to build it' and attacks each obstacle individually can overcome what looks like an impossible, systemically-blocked situation because the obstacles, though numerous, are not coordinated.
One level up
What looks structurally impossible is sometimes just unattempted by anyone with sufficient will.
Where it stops holding
This stops working when opponents are coordinated rather than dispersed, or when the legal/political system genuinely gives any single veto-holder the power to stop the whole project (not just their piece).
Detail
situation: Moses wants to build a road to a beach 27 miles from NYC, but dozens of property owners refuse to let a road cross their land, and it seems impossible to build.
decision: Moses declares 'I'm going to build it' and works around each individual obstacle rather than accepting the aggregate block as final.
outcome: Jones Beach opens and becomes accessible to New York City; Caro calls it 'an act of political genius and also an act of will.'
lesson: Seemingly systemic blockage from many individual opponents can be overcome piece by piece by someone with enough will and ingenuity.
In their words
But he basically says, I'm going to build it. SPEAKER_00: And to watch him do it, to overcome each obstacle, there's so many.
Moses hardens from idealist reformer into ruthless operator through repeated failure
Source
Dwarkesh Podcast — What Robert Caro learned about power after studying it for 50 years (October 30, 2026) · at 11:55 ▸ · at 24:05 ▸
When this applies
when someone with a genuine mission keeps hitting walls using 'clean' methods
The takeaway
Sustained failure using principled methods can train capable, mission-driven people to adopt ruthless methods — and success with those methods entrenches the shift permanently.
Why it works
Early failure to accomplish goals through merit and persuasion teaches an ambitious actor that exploiting others' weaknesses (affairs, money, leverage) gets results, and success at this reinforces the behavior until it becomes the default mode.
One level up
The means by which someone first tastes real success often become the permanent operating method, regardless of how they started.
Where it stops holding
This stops working as an explanatory pattern when the person's early failures are due to lack of skill rather than lack of ruthlessness — in that case, the person may just fail again, not harden into something new.
Detail
situation: Young idealistic Moses tries to reform civil service and build beaches for ordinary New Yorkers using merit and vision.
decision: As he faces resistance over decades, Moses begins to use people's personal weaknesses (affairs, bribes) and steamroll opposition regardless of human cost.
outcome: Moses builds the Cross Bronx Expressway over fierce objection, evicting tens of thousands without relocation, and crushes critics like Henry Epstein by exposing an affair.
lesson: An idealist's method can be overtaken by the ruthless tactics that first proved effective, and the change can go unrecognized by the person themselves.
In their words
He learns that, in my opinion, by failing to get things done when he's very young. SPEAKER_00: He has things that he wants to get done, and they don't get done. SPEAKER_00: And the awareness comes on him, … that he has to find ways Weaknesses in people. SPEAKER_00: He has to use it.