The way out

Violence is system-shaped

Which is the whole argument. A thing that emerges from situations and relationships is a thing that can be built differently. War is not a fact about human beings. It is a fact about the arrangements human beings are currently living inside.

The research this page draws on studies violence at the smallest scale — one person, one room, one night. Its finding is that violence is rarely a decision and almost always a trajectory: an escalating situation that had points along it where a different outcome was still cheap.

Those points are off-ramps. The work is about building more of them, and earlier.

1. People escalate when they cannot get out

Aggressors go furthest when they feel trapped with no exit that lets them keep their standing. Charging forward becomes the only move that is not humiliation.

The intervention that works is not force. It is the presence of someone who makes stepping back survivable — a witness, a caring responder, a face-saving route down.

Scale that up and it is the same shape as the trap between states. A leader who cannot stop without it being called a defeat will not stop. So build the ladder down before it is needed, and make using it something other than surrender.

2. Below a threshold nobody acts. Above it, everybody does

Community response has threshold dynamics. If too few people take a watchful role, willingness decays — everyone assumes someone else will act, and nobody does.

Past a certain level of participation, it flips. Intervention becomes the ordinary thing to do, and prevention starts sustaining itself.

This is the most hopeful finding in the literature, because it means the goal is not persuading everyone. It is reaching the threshold. Most people are already willing; what is missing is the belief that others are too.

3. Who intervenes is predictable

Willingness to step in follows a gradient — it drops off with social distance, and it collapses when responsibility is diffuse.

That is a design specification, not a lament. If intervention depends on closeness, build systems that shorten the distance: alerts that reach the people who already care, responsibility assigned to someone specific rather than to everyone in general.

What this looks like above the scale of a room

Nothing on this page is a treaty proposal. These are the properties any arrangement needs if it is going to bend the curve in the memorial.

Make the decision wider than the few who take it

The people who bear war are not the people who declare it. Any mechanism that closes that gap — that makes the bearer a participant in the decision rather than its recipient — attacks the thing at its root.

Build the off-ramp before the crisis

A route down invented during a confrontation is a concession. The same route, agreed in advance and used by everyone, is a procedure. Identical act, completely different political cost.

Make the count visible while it is happening

Every figure in this memorial was compiled afterwards. Costs that arrive after the decision cannot inform it. The single most useful thing is a count that is legible in week one, not in the history written thirty years later.

Reach the threshold, do not wait for consensus

Prevention does not need everyone. It needs enough people, visible to each other, that acting stops feeling like standing alone.

Why this is not naive

The claim here is narrow and it is worth stating precisely. It is not that people are good. It is not that conflict ends.

It is that war is a particular technology for resolving conflict, that it is the most expensive one ever used, that its costs fall overwhelmingly on people with no part in choosing it, and that it persists largely because the arrangements around it make stopping harder than continuing.

Every one of those is a property of a system. Systems get rebuilt. This one is about four hundred years old, which is young.

A hundred and three million people are not a natural disaster. They are the running total of a design we have not yet replaced. The fourteen wars in this memorial, 1853–2024

What you can do with this page

Read the numbers and refuse to round them. Notice, when a case for a war is being made, whether it is built on the tests that already exist or on the urgency of the moment.

And be the presence that makes stepping back survivable — in a room, in an argument, at whatever scale you actually operate. The research says that is not a small thing. It says it is the mechanism.