Confidence Is Not Evidence

Why are we so easily persuaded by people who sound certain?

There is something reassuring about certainty.

Someone speaks clearly. No hesitation. No qualification. No visible doubt.

They know what happened. They know why it happened. They know what should happen next.

We listen.

And somewhere in the process, an interesting substitution can occur.

We stop evaluating what the person knows.

We begin responding to how certain they appear to be.

Those are not the same thing.

Certainty has a sound

You can hear it.

Compare these two statements:

I think this is probably what happened, but I could be missing something.

This is what happened.

The second feels stronger. It may even feel more credible.

But nothing about the sentence tells us whether the speaker possesses better information. It tells us only that the speaker has expressed less doubt.

We encounter this constantly: in meetings, in families, on television, online, and in almost every place where people are asked to explain complicated things before all the facts are available.

The person willing to say I don’t know can sound less informed than the person willing to pretend that he does.

That creates a strange incentive.

Certainty attracts attention.

Qualification slows things down.

And attention often rewards the person who arrives with the cleanest explanation rather than the most accurate one.

Expertise often sounds different

Years ago, I began noticing something about people who know a great deal about complicated subjects.

They frequently qualify their answers.

Usually.

Under these circumstances.

Based on what we know.

There are exceptions.

I would need more information.

At first, that can sound evasive.

We asked a question. We want an answer.

But perhaps qualification isn’t always weakness.

Sometimes it is evidence that someone understands how much there is to know.

A person who has spent twenty years studying something may see variables invisible to the person who learned about it yesterday.

The novice sees an answer.

The expert sees conditions.

That doesn’t mean experts are always right.

They aren’t.

It means uncertainty can sometimes be the product of knowledge rather than its absence.

The more clearly we see complexity, the harder absolute certainty becomes.

We prefer clean stories

Unfortunately, complexity is difficult to carry around.

Simple explanations travel better.

One person caused the problem. One decision changed everything. One side was right. One side was wrong. One event explains what came afterward.

Maybe.

Sometimes reality really is that simple.

But often a clean explanation is appealing precisely because it removes the uncomfortable parts: ambiguity, contradiction, incomplete information, chance, our own uncertainty.

A confident speaker resolves those things for us.

They hand us a finished story.

We no longer have to sit with the unfinished one.

That may be part of certainty’s appeal.

It doesn’t merely give us an answer.

It relieves us of the burden of not having one.

Confidence moves through groups

Imagine six people sitting around a table discussing a problem.

Five are uncertain.

One isn’t.

The confident person speaks first. He explains what happened and what should be done.

The others may have reservations. One may possess information that complicates his explanation. Another may suspect the proposed solution won’t work.

But now something has changed.

There is a position in the room.

Disagreeing requires effort.

Someone has to interrupt the momentum and say:

I’m not sure that’s right.

That sentence is harder to say than it appears, especially when the confident person has status, experience, authority, or simply a personality that occupies more space.

Sometimes the group follows the most knowledgeable person.

Sometimes it follows the person most willing to sound knowledgeable.

Those outcomes can look remarkably similar from the outside.

Until they don’t.

We do this to ourselves, too

Confidence doesn’t always come from another person.

Sometimes the persuasive voice is our own.

We remember why someone did something. We know what another person meant. We know how a conversation would have gone. We know what would have happened if we had made the other choice.

But do we?

Memory is particularly good at producing certainty.

Over time, the distinction between what happened and what we concluded about what happened can become difficult to see.

An interpretation becomes a recollection.

A recollection becomes a story.

A story becomes something we no longer think to question.

Eventually we can become very confident about things we never actually knew.

That possibility should probably make us a little uncomfortable.

It does me.

The problem isn’t confidence

Confidence is useful.

Decisions have to be made. Leaders have to lead. Parents have to parent. Doctors have to recommend treatments. Architects have to sign drawings.

People cannot move through life prefacing every sentence with maybe.

There are moments when experience, preparation, and judgment justify confidence.

The problem begins when we use confidence as evidence of those things.

Because confidence tells us something about the speaker.

Evidence tells us something about the claim.

One can exist without the other.

A hesitant person can be right. A confident person can be wrong. A room full of people can agree. An institution can issue a statement. Millions of people can repeat the same explanation.

None of those things, by themselves, determine whether the underlying claim is true.

Perhaps we are listening for the wrong thing

When someone speaks with certainty, our instinct is often to ask whether we believe them.

There may be a better question.

What would make this true?

What evidence supports it? What assumptions does it require? What information would change the conclusion?

And perhaps most importantly:

What does the speaker actually know?

Those questions don’t require cynicism.

They require separation.

Separating the person from the proposition.

Separating authority from accuracy.

Separating conviction from proof.

That is harder than it sounds because human beings don’t encounter information in laboratory conditions.

Information comes through people.

And people come with voices, reputations, titles, relationships, histories, and varying degrees of confidence.

All of those things affect us.

They probably always will.

The goal isn’t to become immune to influence.

I’m not sure anyone is.

The goal may simply be to notice it.

Because certainty can be persuasive. Confidence can be useful. Authority can be deserved.

But none of them changes the standard.

Confidence is not evidence.

— Thomas