A Peace With an Autocracy Is a Peace With One Man

Every few years someone proposes that Israel make peace with one or another of the brutal regimes around it, and the proposal is debated as a question of territory, of guarantees, of whether this particular ruler can be trusted. I want to argue that the debate is held at the wrong level, and that the free world, when it asks Israel or anyone else to make peace, is asking the wrong thing.

The right thing to ask for is democracy on the other side. Not as a moral preference but as a structural one. A treaty with a democracy is a treaty with a country. A treaty with an autocracy is a treaty with a man.

What a Treaty With a Man Is Worth

Consider what a peace agreement actually is: a promise about the future behaviour of a state. Its value depends entirely on who is bound by it and for how long.

When the counterparty is an autocrat, the promise is bound to him. Not to his people, who were not asked. Not to his institutions, which exist to execute his will and have no independent commitment to anything. Not to his successor, who inherits the throne and none of the obligations, and who may well have taken the throne by presenting the predecessor's peace as a betrayal. The agreement lives exactly as long as the man's interest in it does, and the man's interest is subject to his health, his rivals, his need for an external enemy, and his mood.

I have written before about the mirror-image problem: a democracy cannot easily bind its future governments either, because they turn over. That is a real weakness. But the autocracy's version is worse in the way that matters for peace, because the democracy's turnover is public, constrained, and arguable, while the autocrat's change of heart is none of those.

What a Treaty With a Democracy Is Worth

A democracy that signs a peace has had to sell it to its own public first. The public argued about it, its opponents lost, and the losing side now has to win an election and repudiate a settled arrangement in front of everyone before the treaty can be broken. That is a high bar. It is not infinitely high, but it is orders of magnitude higher than one man's reconsideration.

More than that: a democratic public has an interest in peace that a dictator does not have. The public pays for wars with its sons and its taxes. A dictator pays with other people's sons and other people's taxes, and frequently gains from the war, because an external enemy is the oldest tool for keeping a population in line. The interests of the two parties to a peace are aligned only when both parties are the people who bear the cost.

This is the old observation that democracies rarely fight each other, and I do not want to overstate it or pretend that it is a law. But the mechanism behind it is not mysterious. Peace holds where the people who would pay for war are the people who decide on it.

Why the Demand Is Not Made

If this is right, the natural demand for the free world to make, of Israel and of everyone, is not "make peace with the regime next door" but "help the people next door become a regime that can make peace." Yet that demand is almost never made, and it is worth asking why.

Partly because it is slow, and diplomats are rewarded for signatures within their tenure. Partly because it sounds like interference, and the free world has a bad record of interfering. But mostly, I think, because dealing with one man is convenient. He can deliver. He does not need to consult anyone. A signature from him is a signature, whereas a democratic neighbour would have to be persuaded, and its parliament would ask questions, and the whole thing would take years.

That convenience is exactly the property that makes the resulting peace worthless. The man who can sign without consulting anyone is the man who can tear it up without consulting anyone. We prefer him as a counterparty for the same reason we should not trust him as one.

The Case Nobody Wants to Discuss

This argument applies with full force to the state that may one day exist next to Israel. Discussions of a Palestinian state are conducted almost entirely in terms of borders, security arrangements, and who recognises whom. The question of what kind of state it would be, whether its government would answer to its people or to whoever holds the guns, is treated as an internal matter, or as something that can be sorted out later.

It cannot be sorted out later, because it is the whole question. A state that is another autocracy is another treaty with one man, subject to everything above, and a peace signed with it is worth what its ruler's survival is worth. A state whose government has to face its own public is a counterparty whose promises mean something. If one wants the peace to last longer than a ruler, one has to want the second, and one has to say so, out loud, as a condition rather than a hope.

I am aware of how this sounds to people who want any agreement at all rather than none. But an agreement that collapses in a decade is not better than none; it is worse, because it spends the credibility that the next agreement would need.

The Principle

Put generally, because it is general: the insurance policy for any peace is the political system on the other side of it, not the text of the agreement. Guarantees, monitors, and signatures are all attempts to compensate for a counterparty that cannot bind itself. A counterparty that can bind itself, because it is bound to its own people, needs far less of them.

So when the free world asks anyone to make peace with a tyranny, it should be asked back: peace with whom, exactly, and for how long after he is gone?