Why You Stay With Your Side Even When the Other Side Is Right

Imagine a band of a hundred and fifty people, which is roughly the size of the human group for most of our history. A hard question arises: go north or go south. The band splits. Two leaders emerge, one for each direction, and the argument is bitter, because a wrong choice costs lives.

Now suppose you are in the northern faction, and you listen carefully, and you come to the private conclusion that the southern arguments are better. Not slightly better. Clearly better.

What do you do? More to the point, what would most people do?

The Answer Nobody Is Proud Of

Most people stay with their side. They do not switch. And if you think that is a failure of character, you have not thought about what switching would cost.

The sides in a band are not instruments for deciding one question. A side is a political and social alliance that happens to be voting on this question today. You belong to it because you agree with most of its positions, because your friends are in it, because you have eaten with these people and fought beside them and expect to depend on them next month. The north-or-south question is one item on a very long ledger.

Now count what a switch does to that ledger.

The other side may not take you. If they refuse, you have cut every tie to your own group and gained nothing. In a band of a hundred and fifty, with no other band in reach, that is not a social setback. It is a threat to your life.

If they take you, they will not take you fully. You have a long history of siding against them. Every member of the southern faction remembers it. You will be the last to be trusted and the first to be suspected, for years. Your position in the group has got worse, not better.

And they will expect you to vote with them from now on. You joined over one question, but you disagree with them on a dozen others, some of them important to you. Having switched once, you cannot switch back. So you have traded being right about one question for being outvoted on many.

Put the three together and the rational move is clear. You keep your mouth shut about the south, you vote north, and you hope the north is not as wrong as you fear.

What This Does to the Group

The consequence is that the band decides badly. Not because its members are stupid, not because they cannot tell a good argument from a bad one, but because the cost of acting on the good argument is borne entirely by the individual and the benefit is spread across everyone.

That is the structure of a collective-action failure, and it is worth naming as one. Every individual in the northern faction may privately know the south is right. As long as each expects the others to stay, each stays. The group goes north with a majority that does not believe in it.

The alliance structure converts a question of fact into a question of loyalty, and loyalty is decided by different machinery. The machinery is older, faster and much more strongly felt, because for most of our history being expelled was fatal and being wrong about a route was survivable.

Scale It Up

Everything above was about a band. Now look at a modern electorate.

The sides are parties. The ledger is the party platform, forty issues bundled together, and you joined for a handful of them. The social ties are real: your family, your colleagues, your friends mostly vote the way you do, and they know how you vote. And the costs of switching are exactly the band's costs, softened but not removed. The other side will treat you as a convert with a suspicious past. Your own side will treat you as a defector. And once you have crossed you will be expected to carry the whole of the new bundle, including the parts you find wrong.

So people do not switch when the other side has the better argument on a particular question. They stay, and they find reasons, and the reasons feel like their own. I have written about how the bundling itself turns every dispute into a defence of the whole beachhead. This is the same mechanism seen from inside the individual: the bundle makes the personal cost of conceding one point enormous, so no point is conceded.

The result at the national scale is the same as in the band, only with more people. Groups make bad decisions because sub-group alliances make individual honesty expensive. That sentence is most of what is wrong with politics as practised.

Why Leaders Get the Loyalty

There is a second thing the band explains, which is the intensity of attachment to leaders.

You cannot love a platform. A list of forty positions is not a thing a human mind can be loyal to. But a person is. The leader becomes the label for the group: the face you picture when you think of your side, the name that stands in for the whole ledger. We are a social species; we remember people effortlessly and abstractions with difficulty. So the loyalty that is really owed to the alliance gets attached to the individual who represents it, and then defended as if it were a judgement about him.

That is why supporters forgive a leader almost anything. They are not defending the man. They are defending the thing the man stands for in their heads, which is their own membership, and the cost of withdrawing it.

What Would Have to Change

The band could not fix this. It had no way to vote on north-or-south without the vote also being a test of allegiance, because the same people, in the same room, would be deciding everything else next week.

We do not have that excuse. A system that put questions one at a time, to everyone, without a party standing behind each answer, would let a person say the south is right about this without saying anything about who his friends are. The question would be detached from the ledger. Nobody would be a defector for answering it honestly, because there would be nothing to defect from.

That is not a small change, and I do not pretend it is. But the diagnosis says what the cure has to look like. As long as questions are decided by sides, they will be decided by loyalty, and loyalty does not care which way is north.