Talking to people you disagree with

The instinct on encountering a view you consider wrong is to correct it, which is the approach least likely to work.

Why argument rarely works

The mechanisms.

Being told you are wrong produces defence rather than reconsideration.

Positions held publicly are harder to abandon, since doing so has a social cost.

Beliefs are frequently connected to identity and group membership rather than to evidence, which means evidence does not address what is holding them.

And a certain amount of research suggests that counter-evidence can strengthen a position in some circumstances, though the extent of this effect is debated.

Which does not mean minds never change — they do, generally slowly and rarely during the conversation itself.

What works better

Approaches with more support.

Genuine curiosity: asking how someone came to a view, what experiences informed it, what would change it.

Which produces information and lowers defences simultaneously.

Asking for explanation of mechanism: research on the illusion of explanatory depth has found that asking people to explain in detail how something works reduces confidence in their position, where asking for reasons does not.

Which is a specific and useful technique.

Finding the actual disagreement, which is frequently narrower than it appears and is sometimes about values rather than facts.

Stating their position back in a form they accept before responding, which demonstrates understanding and reveals whether you have it.

And conceding what is valid, which almost always exists and which changes the character of the exchange.

Separating the disagreements

Which clarifies a great deal.

Disagreement about facts: resolvable in principle by evidence, and frequently not the actual issue.

Disagreement about the interpretation of facts.

Disagreement about values and priorities, which evidence cannot settle.

Disagreement about who to trust, which underlies many apparent factual disputes.

And disagreement about identity and belonging, which is not really about the topic.

Establishing which you are having prevents the frustration of arguing facts with someone whose disagreement is about values.

The questions worth asking

Which move a conversation.

How did you come to think that?

What would change your mind?

What is the strongest argument on the other side?

Can you explain the mechanism — how does that actually work?

What do you think I believe, and why do you think I believe it?

And, asked of yourself: what would change my mind, and can I answer that as readily as I expect them to?

Your own position

Which deserves the same scrutiny.

Can you state the opposing view in a form its holders would accept?

Do you know why intelligent people hold it?

What is the strongest evidence against your position?

How confident should you actually be, given what you know?

And are you defending a conclusion or examining a question?

An honest answer to the last generally identifies whether the conversation is worth having.

When to stop

Which is a skill.

When the exchange has become about winning rather than understanding.

When either party is repeating rather than developing.

When it is damaging a relationship that matters more than the point.

When the other person is not engaging in good faith.

And when you have said what you have to say, which is generally sooner than the conversation ends.

"I do not think we are going to agree on this, and I understand your view better than I did" is a legitimate ending.

Family and close relationships

Where the stakes are different.

The relationship generally matters more than the disagreement.

Which means the objective is frequently to maintain the relationship rather than to resolve the issue.

Practical measures: agreeing which topics are off the table; addressing behaviour rather than belief where behaviour is the actual problem; and separating disagreement from disapproval, which is difficult and is what determines whether the relationship survives it.

And accepting that some people will not change, which is a fact rather than a failure of your argument.

Online

Where all of this is harder.

Absence of tone, absence of relationship, presence of an audience, and incentives towards performance rather than exchange.

Which means the probability of a productive disagreement online is low, and the probability of an unproductive one is high.

Anyone genuinely wanting to change a mind is better served by a private conversation, and anyone wanting to demonstrate a position to onlookers should be honest that this is what they are doing.