I Do Not Trust the Doomsday Clock Is Three Different Objections, and All of Them Fail

I was talking to an intelligent friend about the Doomsday Clock, the figure the Bulletin of the Atomic Scientists publishes each year as a number of seconds to midnight. He said he had absolutely no trust in that thing.

Fair enough. But the sentence does not say what it appears to say, because "no trust" can mean three quite different objections, and they need separate answers. When I separated them, none of the three survived.

Objection One: The Scientists Are Biased

Of course they are. Everyone who produces an estimate about anything is biased. The people who set the clock have views about nuclear weapons, they have institutional interests, and they know that a frightening number gets more attention than a reassuring one.

But notice what the objection asks for. It asks for an estimate produced by people with no biases, and there is no such estimate on offer, on this or any other subject. The choice is not between a biased estimate and an unbiased one. The choice is between a biased estimate made by people who publish their reasoning and can be argued with, and no estimate at all, which in practice means the estimate you already carry in your head, which is also biased and which nobody can inspect.

Science is not a procedure for finding unbiased people. It is the only procedure we have found for extracting something usable from biased ones: state the reasoning, expose it to people with different biases, and correct. The clock is a crude instance of that. Refusing it because its makers are biased is refusing the only known remedy for the disease you are complaining of.

Objection Two: Seconds to Midnight Is a Silly Unit

This one has more force, because the unit is theatrical. A probability of a catastrophic exchange, whatever it is, is not naturally a time of day, and the conversion has no scale behind it. Ninety seconds does not mean anything that a hundred seconds does not.

So what would my friend prefer? A probability. Per year? Per decade? With what confidence interval? And how many people would read it?

The honest answer is that the theatrical unit exists because most of the public cannot absorb a probability. Not because they are stupid, but because our schools teach Euclidean geometry and not statistics, and a population that has never been taught what a two per cent annual risk compounds to over a lifetime will not be moved by being told it. A clock face is a translation into the only vocabulary the audience was given.

That is a real indictment. It is an indictment of the education system, and fixing it would take a generation or two under the best conditions imaginable. It is not an indictment of the people who, faced with that public, chose a unit the public could feel.

Objection Three: You Cannot Quantify This at All

The deepest version of the complaint: the probability of a nuclear event is not the sort of thing that can be given a number. The event has almost never happened, the causes are political and chaotic, and any figure is a guess dressed up.

Grant every word of that. Now ask the only question that matters: what is the alternative?

The alternative is intuition. It is the feeling, when you read the news, that things are fine or that things are bad. And human intuition about rare catastrophic events is not merely imperfect; it is about as bad as intuition gets. We are frightened by vivid risks and calm about diffuse ones. We treat a risk we have not personally seen as zero. We update on the last headline and forget it by the next. If you decline to quantify, you have not avoided estimating. You have delegated the estimate to the least reliable instrument you own, and hidden the fact from yourself.

A number produced by a committee reading published evidence, however crude, however theatrical its packaging, is a better instrument than that. Not because it is right. Because it is checkable, and it can be argued with, and it does not evaporate when the news cycle moves on.

What "I Do Not Trust It" Actually Means

My friend is not unusual. Most objections to any attempt to put a number on something people feel strongly about have this shape: one of the three complaints above, usually unseparated, delivered with a confidence the complaint does not earn.

The pattern is worth naming. Refusing to quantify feels like humility and functions as evasion. It sounds like an admission that the matter is too grave for numbers. What it does is leave the matter to instinct, which is to say, to whoever tells the most frightening story most recently.

If the clock is wrong, argue that it is wrong. Say it should read ten minutes, or two seconds, and say why. That is a conversation the makers of the clock invited, and it is the conversation quantification exists to make possible. "I do not trust that thing" is a refusal to have it.