Foreign and Security Policy: More Guts Than Brains?

This is an old article I published at the time on the Imago website...

An ancient dispute among political philosophers is rooted in a dilemma: which matters more in a ruler, or in a system of government — the courage and capacity to make decisions, or the intelligence to understand hard problems in depth and know which decision ought to be made? Since we cannot have everything in life, we would like an answer to this question so that we can choose the system of government that best serves us. In this essay I will try to grapple with it.

If there is one thing that has become clear to humanity over time, it is that the world we live in is immeasurably more complex than the banal abstractions of Aristotle or Plato. If it takes divergences and group theory to compute the position of a single quark, what could it possibly mean to speak of strategies for foreign and security policy — strategies that involve millions of human beings in situations of enormous complexity, where each person is a vast colony of cooperating cells and each cell is a vast collection of quarks? On the face of it, there seems to be no way out of the maze. Reality is too complicated, and therefore there is nothing intelligent we are capable of saying about it — least of all at levels as elevated as foreign and security policy.

Reality is obviously complicated. Yet the foundation of disciplines such as economics, history and political science is that despite the astonishing complexity of reality, stable statistical patterns can be identified within it at the macroscopic levels of individuals and societies — and those patterns are the conclusions we are able to draw in such fields. No reasonable person would dispute this, so I will not dwell on it.

If we do accept the conclusion that intelligent things can be said about large entities such as people and states, notwithstanding the great internal complexity of those entities, then the question arises: must foreign and security policy be complicated? The surprising answer is no. There is a foreign and security policy that is in many respects ideal — and, remarkably, also extremely simple.

How shall we go looking for this miraculous policy? The foundation lies in understanding the constants in relations between states. There is no doubt that states gain from mutual cooperation and lose from its absence, whether one-sided or mutual. Note that the gains in question accrue to the citizens, not to this or that ruler. This is a property of the natural world around us, and of the fact that more can be extracted from it by joint effort. The property is not transient, and evidence for it exists throughout the blood-soaked history of the human race. Another interesting property of these relations between states is that as long as none of the states disappears, the states are condemned to make these decisions — cooperation versus hostility — again and again, without end.

Is there a mathematical model that accurately captures these properties of relations between states? There is indeed! It is the famous Prisoner's Dilemma, extensively studied in mathematical game theory for economic, political, biological and other purposes.

In the Prisoner's Dilemma, the police catch two criminals. If both inform on each other, each incriminates the other and both go to prison for long terms (both lose). If one informs and the other keeps silent, the informer receives only a light sentence and his friend goes to prison (one gains and the other loses). If both keep silent, both go to prison for short terms (both gain).

The question arises: what is the preferred strategy when we play the Prisoner's Dilemma? The unequivocal answer is to inform. The reason is that a strategy of silence is a strategy dominated by the other side. That is, if we keep silent we are leaving our fate to be determined by the other side, which is trying to maximize its own gain and will therefore choose to inform. Since this reasoning is clear to both sides, both will inform.

A different game is the iterated Prisoner's Dilemma. In this game the same two players play the Prisoner's Dilemma against each other a great many times. If those same two criminals find themselves in the same situation over and over, the question becomes: what long-term strategy should they adopt?

In his fascinating book The Evolution of Cooperation, the political scientist Robert Axelrod describes an experiment in which he ran a tournament between different strategies for the iterated Prisoner's Dilemma, submitted as computer programs by some of the best minds in the field. The strategies were played against each other in every possible pairing. The winner by a considerable margin in the first round was the strategy "Tit For Tat." This strategy holds that you should do to your opponent whatever he did to you in the previous round. If he cooperated with you in the previous round, then cooperate; informing is repaid with informing. After Axelrod informed the tournament participants of the results of the first round, he asked them for a second attempt — and remarkably, even though all the professors knew that Tit For Tat had won the first round, the result was identical.

The main reasons Tit For Tat won the tournament are: its simplicity, and hence its clarity to the other side; its consistent punishment of non-cooperation; its consistent rewarding of cooperation; and the plain conclusion drawn by everyone who played against it, namely that the only way to gain in a game against it is by cooperating with it.

In fact there is a remarkable correspondence between the iterated Prisoner's Dilemma and the relations described above between states, and so the result is striking: there exists a foreign and security strategy that is simple, clear and ideal. From here the conclusion is trivial: you do not need to be a great strategist in order to conduct foreign and security policy! The only thing required is the courage and the governing capacity to make hard decisions in order to implement the Tit For Tat strategy — in conflict as much as in peace.

This conclusion supports the position that what matters in government is the combination of making the right but simple decisions, such as Tit For Tat, and the governing capacity to carry them out. It would appear that what we need is a dictator who is not especially clever but is capable of making hard decisions, such as going to war or making peace.

On the other hand, the shift in the system of government from dictatorship to representative rule (or, as it is today mistakenly — or deliberately — called, democracy) brought about a reduction in the power of the central government. It is hard to dispute the claim that this change of regime was a change for the better, at least for most of the population. From this it would seem that reducing the power of the central government is an advantage. As for the dilemma we set out, this position appears to support the view that courage and the capacity to make decisive choices are not a critical component of a successful system of government, and that intelligence is apparently the more important component.

How are we to reconcile the contradiction between Axelrod's surprising conclusion and the plain fact that representative government looks like a success? The resolution is simple. Tyrannical rule is so bad that it fails our first requirement — "making the right decision" — even though it is strong in its capacity to decide and to execute. Tyrannical rule is problematic to such a degree that the loss involved in reducing its decisiveness in the move to representative government is small compared to the reduction in its power to make very bad decisions. This is the principal reason for the relative successes of representative government. The conclusion, then, is that Axelrod was right, but we are still far from the foreign and security strategy that follows from his findings, because we first had to deal with the catastrophic decision-making of dictators. We do not need a ruler with sophisticated strategies, since the required strategy is simple. But first we must make sure that the central government's decisions are sane.

Is it worth going further still and reducing the decisiveness of representative government even more? The answer emerging in recent years seems to be no. Representative government appears relatively well suited to small- and medium-scale decisions, but impotent when it comes to fateful ones. It is enough to recall the Israeli government's inability to force a decision in the First Lebanon War through a decision in principle to eliminate Yasser Arafat, and then its inability to withdraw its forces for eighteen years. Did the government here show that it controlled reality, or that reality controlled it? Representative government appears dangerously weak, and so talk of reducing its power is beside the point. In fact, whenever we sense that the people want a strong leader, we find that the reason is a strong leader's ability to make decisions that representative government cannot make. Unfortunately, increasing the power of government is problematic in its own right. The number of people holding power shrinks, and we risk sliding down the slippery slope that leads us back into the bear hug of the dictator.

Is there no way out of the maze? Is there no system of government with a dictator's power to implement strategies like Tit For Tat, but without the dictator's tendency to make decisions that are disastrous for the population? Representative government is too impotent to meet these requirements, and dictators who at first appear to be all smiles turn out to be enthusiasts of enriched uranium.

Never fear — the solution is already here!