A historian of science figured out who reaches for numbers and why. His answer is a map of power in your meetings.
Picture two people in the same product review. The head of product says a feature won't work; she has seen a hundred launches like it. The room nods. A junior researcher says the same thing, with a deck. The room asks about her sample size.
Same conclusion. Two receptions. We tend to read this as a story about seniority, or rigour, or confidence. Theodore Porter would say it is a story about trust — and that the numbers are being demanded from exactly the person who has the least of it.
Porter, a historian of science at UCLA, spent years in the archives of nineteenth- and twentieth-century engineers, actuaries and accountants asking a question that sounds naive: why does anyone agree to subordinate their own judgment to a calculation?
The answer he found in Trust in Numbers runs against instinct. The pull toward quantification comes not from the strongest but from the weakest. An expert whom society trusts can say "believe my judgment" and be believed. An expert who is not trusted has to show the arithmetic. French state engineers, drawn from an elite that was trusted, ran on expert opinion for decades. American army engineers, whom Congress did not trust, invented cost-benefit analysis — not because it was more accurate than judgment, but because it could travel across a wall of suspicion where personal credibility could not.
Porter called it a technology of distance. Numbers let a decision move to where trust doesn't reach: across hierarchies, across departments, across the gap between someone who did the work and someone who didn't watch them do it.
The lens does two things at once when you point it at a modern organisation.
It tells you that the demand for evidence is not a sign of maturity. It is a symptom of a trust deficit, and it lands on the people lowest in the hierarchy. The junior researcher is asked for the sample size because nobody in the room has a reason to believe her yet. The head of product is not, because they do. Neither reaction has anything to do with whether the feature will work.
And it tells you where power sits. Ask a simple question of any team: who here can say "I just see it this way" and be followed without numbers? The answer — founder only, any executive, recognised experts at any level, almost no one — is a cleaner map of authority than the org chart. The person exempt from evidence is the person with power. Everyone else is quantifying their way toward being believed.
This is why "we're a data-driven company" can be true and hollow at the same time. Data-driven for whom? If the requirement to show numbers scales inversely with rank, the company has not adopted evidence. It has adopted a way of managing people it does not trust, and given it a flattering name.
For researchers and analysts the lens is uncomfortable and useful. The question you are really being asked, when someone says "show me the data", is often not "is this true" but "why should I believe you". Answering the first with more rigour when the second was the question is how careful people burn out. Sometimes the right move is a better chart. Sometimes it's a sponsor with authority carrying the same finding into the room.
For leaders the lens is sharper. Every time you decide without numbers, you are spending trust you have. That is legitimate; it is what expertise is for, when the environment has actually taught you something. But notice who in your company is never allowed to do the same, and ask whether that's because their judgment is worse — or because nobody has checked.
Our first study asks people to reconstruct one real decision and answer, among other things, who in their company can decide without evidence and be followed. We are testing a prediction that follows from Porter: that the demand "back it with data" correlates with the rank of the person being asked, not with the quality of what they said.
If you've been on either side of that sentence, four minutes of your time is one data point toward finding out.
Four minutes. One real decision. Take the diagnostic →
Institute for Human Reasoning is a research and education organisation strengthening human reasoning for consequential decisions in a technology-shaped world.
Seventy years of research on evidence and decisions, in six findings that still embarrass every company calling itself data-driven.
ReadEvery company will say yes. The only measurement that means anything is where AI enters a decision — and whether anyone checks it when it disagrees with the boss.
Read