The Security Dilemma

This is the idea promised at the end of Module Three, and it is the one most likely to change how you read the news. Here is the whole of it.

Two neighbors live at the end of a long road. There is no sheriff. There has never been any trouble between them.

One of them starts hearing about break-ins in the area, so he buys a dog and puts up a fence.

The second neighbor sees the fence go up. He does not know about the break-ins. What he knows is that the man next door has decided he needs protection, and the only thing to protect against out here is him. So he buys a gun.

The first neighbor sees the gun. His fence, which was about strangers, has apparently been answered with a weapon, which can only be about him. He buys two more dogs and a second gun.

Neither of these men wants trouble. Neither of them started anything. Every single step either one took was defensive, and they are now closer to shooting each other than they were before anybody did anything about safety.

Why Sincerity Does Not Help

The name for this is the security dilemma, and the important thing about it is that it does not require anybody to be aggressive, deceitful, or unreasonable.

It requires one condition only. You can see what the other side is able to do. You can never see what they mean by it.

A defensive move and a threatening move look exactly the same from the outside.

A wall keeps people out and it also protects the men gathering behind it. A radar station warns you of an attack and it also guides yours. An army near a border is either a shield or a starting line, and the difference is entirely in the intention, which is the one thing that never crosses the fence.

Now add the part that makes it worse rather than better. Each side has perfect knowledge of its own intentions. You know you are not planning anything. Your own defensiveness is transparent to you and therefore feels obviously innocent, so the other side's alarm at it looks like paranoia or pretext. Meanwhile they are having the identical experience about you, and they are equally sincere.

That is why this is so hard to talk your way out of. Both sides are telling the truth. Both find the other's account unbelievable. Sincerity has never once been visible from the outside.

The Ratchet

These situations also climb easily and come down with great difficulty.

Going up costs money and nothing else. Coming down costs safety, or looks like it does. Any step that reduces the tension can be read at home as weakness and across the fence as a trick. So the natural resting position is one more dog.

Before the First World War, Britain and Germany spent an extraordinary share of their national wealth building battleships against each other. Neither government was trying to start the war it got. Each was answering the fleet the other was building. When the fighting came, it came through alliances and mobilization schedules that had been built to make war less likely.

What Actually Lowers It

The useful part of this lesson is that the dilemma has known solvents, and none of them are trust.

The first is making the difference visible. Some capabilities can only defend, and everybody can tell. A fixed shelter is not an invasion. Where the two things can be told apart, the spiral runs slower. A good deal of arms control has consisted of arguing about which category a particular thing belongs in.

The second is being checkable. If the other side can verify what you have, your restraint becomes evidence instead of a claim. This is why inspections, monitoring, published numbers, and overhead observation matter far more than declarations do. Two governments that despise each other can still both prefer a world where each can count the other's launchers.

The third is giving something up in a way that would be expensive to fake. Words are free, so words prove nothing. Pulling forces back from a border, or dismantling something you would need in order to attack, costs you if you were lying. That is what makes it worth anything.

The fourth is somebody above both of you. Two neighbors with a working sheriff have a much smaller version of this problem, because neither has to guarantee his own safety. Countries have no sheriff. That absence is why Lesson 2.3 described the fifth rung as running on fear and force.

When It Is Not A Security Dilemma

Here the lesson has to be careful, because this idea can be turned into something foolish.

Not every conflict is a misunderstanding between frightened people. Some parties want what you have and intend to take it. Treating a genuine predator as an anxious neighbor is not sophistication. It is a way of arriving late.

So here is the test, and it is the only reliable one available. Watch what happens to the other side's behavior when the threat you pose to them goes down.

If you pull something back and their buildup slows, you were in a security dilemma, and you have just found the lever. If you pull something back and their buildup continues at the same rate or speeds up, then their arming was never a response to you, and you are dealing with something else that will need something else.

That test is slow and it costs something to run. It is still better than assuming, and assuming is what almost everybody does.

The Small Version

None of this is confined to countries.

Two colleagues start copying their manager on messages. Neither began it as an attack. The first wanted a record. The second saw a record being kept and wanted his side in it. Six weeks later every ordinary question between them is documented, the relationship is unrecoverable, and both of them will tell you truthfully that they were only protecting themselves.

A married couple does the same thing with money, or with what they tell their families.

Same structure, same sincerity, same result. The fence went up first and nobody remembers why.