One can describe a foreign policy as libertarian in either of two senses. In the first and stronger sense a foreign policy is libertarian if it is implied by libertarian principles—if libertarians must follow it because it can, and alternative policies cannot, be carried out without violating anyone’s rights. One thesis of this chapter is that there is, in that sense, no libertarian foreign policy or at least none whose consequences many libertarians are willing to accept. The second thesis of this chapter is that there is a libertarian foreign policy in a second and weaker sense—a policy that libertarians would expect to work better than alternative policies for some of the same reasons that they expect a libertarian society to work better than alternative societies.

In discussing foreign policy I will, for the most part, ignore the question of who conducts it and how it is paid for. Those libertarians who believe in limited government may think of it as the foreign policy of such a government. Those who believe, as I do, in some form of society without government may think of it as the foreign policy of whatever institutions within that society are responsible for defending it from foreign governments or as the foreign policy that we should urge our government to follow until we succeed in abolishing it.

I find it is useful to start by considering two broad classes of foreign policy: interventionist and non-interventionist. Under an interventionist foreign policy a nation defends itself by a network of alliances. It supports those powers and those political forces that it believes will be useful allies in the future; it opposes those it regards as likely enemies. Under a non-interventionist policy a nation makes few or no alliances and takes little or no interest in what the governments of other nations are doing. It defends itself by shooting enemy soldiers who try to cross its border or firing nuclear missiles at any country that fires nuclear missiles at it.

Some might argue that an interventionist policy is non-libertarian because, by intervening in the internal affairs of other nations, we are violating their freedom to rule themselves. This argument confuses the independence of nations with the freedom of individuals. Whether my nation is independent and whether I am free are two quite different questions. That my nation is independent merely means that I am ruled by people who happen to live near me. I know of nothing in libertarian theory that makes coercion morally legitimate merely because the coercers and their victims live in the same part of the world, speak the same language, or have the same color skin.

A better argument against an interventionist policy is that such a policy almost inevitably involves allying with oppressive governments. There are, after all, not many libertarian governments available to ally with. Even if we allow alliances with governments similar to our own we are still locking ourselves out of much of the world and so gravely handicapping any serious attempt at an interventionist policy. In practice, an interventionist policy almost inevitably involves alliances with the Shah of Iran, or Joseph Stalin, or Ferdinand Marcos, or, in the case of the actual policy of the U.S. over the past 70 years, all of the above.

Allying with unattractive governments does not merely mean offering to help them against our common external enemies. Oppressive governments have internal enemies as well. If we are not willing to provide such governments with the assistance they need to stay in power, they will find other allies with fewer scruples. So, in practice, an alliance with the Shah cannot be limited to defense against a Russian invasion. It also includes arming and training the secret police.

If we are supporting, training, arming, subsidizing the forces which a government uses to coerce its people, we are in part responsible for that coercion. If, as libertarians, we believe that we cannot initiate coercion, it would seem to follow that we cannot help other people initiate coercion. It follows from that that we cannot have an interventionist foreign policy, or at least not much of one. Even if the best way of defending ourselves against coercion by the Soviet Union is by allying with the Shah of Iran or the Chinese Communist Party, we are not entitled to buy our defense at the cost of the Iranians and the Chinese.

I find this a persuasive argument. Unfortunately, it can be carried one step further. The obvious alternative to an interventionist policy is a non-interventionist policy. Under such a policy we defend ourselves not by a network of foreign alliances but by a large number of missiles equipped with thermonuclear warheads. The missiles are pointed at the Soviet Union; if the Soviet Union attacks the U.S., we fire them. The result is to kill something between fifty million and two hundred million inhabitants of the Soviet Union. While a few may be high ranking party officials, most will be innocent victims of the Soviet system, no more guilty for the sins of their government than are the Iranians or Chinese.

Both interventionist and non-interventionist foreign policies involve, for libertarians, the same moral dilemma. Under an interventionist policy we defend ourselves, when it seems necessary, by helping the governments we ally with to oppress their citizens. Under a non-interventionist policy we defend ourselves, when it seems necessary, by killing innocent citizens of the governments we are fighting against.

In both cases it is tempting to justify our actions by treating countries as if they were people. We would like to say that if the Russians attack us we are justified in killing them in return, just as, if John Smith tries to kill me, I am entitled to kill him in self-defense. But the Russians, unlike John Smith, are not a person. Speaking the same language or living in the same country as someone does not make me responsible for his crimes. Similarly, we would like to say that, whatever sort of aid we give to the Iranian government, we cannot be guilty of coercion since the Iranians asked for the aid. But the Iranians who asked for the aid and the Iranians against whom it is used are different people.

If libertarian principles rule out both interventionist and non-interventionist foreign policies, are there any alternatives left? The answer, I think, is yes, but not very attractive ones.

One strategy supported by a few libertarians is to defend ourselves with guerrilla warfare and propaganda instead of either alliances or missiles. I doubt it would work. So far as I know, guerrilla movements without external support have been uniformly unsuccessful against regular armies. Further, guerrillas generally pay no more regard to the rights of innocent parties than do the government armies they are fighting against. If we choose guerrilla warfare in order not to violate any individual rights, our guerrillas will fight under severe restraints. They may never explode a bomb where it would damage private property. They may never use automatic weapons if there are civilians in the background who are likely to get hit. They are, in effect, fighting with one hand behind their backs.

It is sometimes argued that one advantage to defending a libertarian society in a libertarian fashion is that the Soviets cannot conquer us if there is nobody to surrender to them. Perhaps, if we have no state, the Soviets will find that constructing a puppet government starting with nothing is simply more work than it is worth. Where, after all, will they find enough Communist bureaucrats who speak English?

Unfortunately, as I pointed out in Chapter 34, there is a simple solution likely to occur to the Soviets or any other conqueror. All they need do is pick out a medium-sized city of no great importance and announce how much tribute they expect and when it is due. They also announce that if the tribute is not forthcoming by the deadline, the city will be used as a test site for a nuclear weapon. The organization of the government that will provide the tribute can safely be left to local initiative. If the tribute is not paid the Soviets drop the bomb, film the result, and send the film on tour. The next city pays.

If my arguments so far are correct, it appears that we have only two choices. Either we follow a policy which makes it easy and profitable for any powerful nation to conquer us or we defend ourselves by means that are at least questionable in terms of libertarian principles. If we make the latter choice, we are taking the position that, if the only way to defend ourselves involves injuring innocent people, we are entitled to do so. Our moral position is then similar to that of an armed man who is attacked in the middle of a crowd and shoots back at his attacker, knowing that he may well hit one of the bystanders. It seems unfair to the bystanders to make them bear the cost of his defense, but it also seems unfair to say that his only moral alternative is to stand there and be killed.

If we are not willing to impose costs on others in defending ourselves, then there is a libertarian foreign policy: surrender. That is not a policy that very many libertarians of my acquaintance are willing to accept. If we are willing to impose such costs, then libertarian principles do not tell us whether we should adopt an interventionist policy and impose the costs on the citizens of oppressive governments with whom we ally or adopt a non-interventionist policy and impose the costs on the citizens of our enemies. In that sense, there is no libertarian foreign policy. On one interpretation of libertarian principles neither alternative is acceptable, on the other interpretation both are.

I believe, however, that there is a libertarian foreign policy in another sense, a foreign policy that libertarians would expect to work better than its alternatives for some of the same reasons we expect a free society to work better than its alternatives. To show why, it is convenient to start with the argument for an interventionist policy and the problems with that argument.

The case for an interventionist policy can be summed up in one phrase: the lesson of Munich. It has been widely argued that if only the British and French had been willing to stop Hitler at the time of the Munich agreements, he would have backed down and World War II would never have happened. Many people conclude that the appropriate way to deal with potential enemies, especially enemies aiming at world conquest, is to fight them before they get strong enough to fight you, to prevent their expansion by allying with the nations they want to annex, to ally with any government willing to join you in opposing them.

If the Nazis attack Czechoslovakia, the Czechs will fight in their own defense as long as they see any chance of winning. If we help them, we fight the Nazis, in large part, with Czechoslovakian blood and treasure. If we let Czechoslovakia go, five years later we find ourselves fighting against the products of the Skoda arms works in the hands of the German army. It is a persuasive argument. It seems to have persuaded U.S. policy makers and much of the U.S. public, with the result that we have tried to follow such a policy in dealing with the Soviet Union.

The weak point in the argument is its assumption that the interventionist foreign policy will be done well—that your foreign minister is Machiavelli or Metternich. In order for the policy to work, you must correctly figure out which countries are going to be your enemies and which your allies ten years down the road. If you get it wrong, you find yourself unnecessarily blundering into other people’s wars, spending your blood and treasure in their fights instead of theirs in yours. You may, to take an example not entirely at random, get into one war as a result of trying to defend China from Japan, spend the next thirty years trying to defend Japan (and Korea, and Vietnam, …) from China, then finally discover that the Chinese are your natural allies against the Soviet Union.

One problem with an interventionist foreign policy is that you may intervene unnecessarily or on the wrong side; that, arguably, is the history of much of our China policy. A second problem is that, even if you are on the right side, you are frequently involved in conflicts which are much more important to the other players, with the result that you end up paying the cost of intervention but not achieving very much.

One of the striking things about the Vietnamese war is that the Vietnamese on both sides continued to fight after taking casualties which, relative to their population, were immensely larger than the casualties which drove the U.S. out of the war. That is not, if you think about it, very surprising. Vietnam is worth a great deal more to the Vietnamese, North or South, communist or anticommunist, than it is to the Americans. Even though we were much larger and more powerful than the other forces involved in the war, we found that the price of winning was more than we were willing to pay. The Soviets seem to have learned a similar lesson in Afghanistan.

The problem with an interventionist foreign policy is that doing it badly is much worse than not doing it at all. Something which must be done well to be worth doing is being done by the same people who run the post office—and about as well.

To say that our foreign policy is badly run is in a sense misleading. Perhaps when we support dictators who contribute very little to the defense of the U.S., the reason is that they contribute instead to the profits of American firms who do business in their countries and the American firms in turn contribute to the politicians who make our foreign policy. If so, what we are observing is not the incompetence of the people making our foreign policy but their competence at achieving objectives other than the defense of the U.S., most notably their own wealth and power.

But exactly the same thing can be said of the Post Office. One of the reasons it appears badly run is that postal jobs are political plums used to reward faithful supporters of the party in power. When one describes government as incompetent to achieve its objectives, one is speaking metaphorically; the government is not a person. It does not have objectives any more than it has hands or feet or ideas. What I mean by saying that government does a bad job of running the Post Office is that one consequence of many individuals using the government to achieve their own objectives is that the mail gets delivered infrequently and late. What I mean by saying that government does a bad job of running our foreign policy is that another outcome of individuals using the government to achieve their own objectives is a foreign policy poorly designed to defend the U.S. Whether the reason is incompetence or corruption is irrelevant.

There is a lesson to be drawn from Munich, but it is a different lesson than is usually drawn. At the time of the Munich agreement, England and France had interventionist foreign policies; that is why Hitler made sure he had their permission before he invaded the Sudetenland. If they made the wrong decision and missed their opportunity to prevent World War II, that is evidence of what is wrong with the usual argument for such a policy. One should not base decisions about what kinds of things a government should do on the assumption that it will always do them well.

This argument suggests that libertarians ought to be skeptical of an interventionist foreign policy. It is difficult to run a successful interventionist policy and, as libertarians, we do not expect the government to do difficult things well. Even if foreign policy were conducted by some private organization funded along the lines suggested in Chapter 34, many of the same problems would exist. Such an organization, although private, would be more like the Red Cross than like an ordinary private firm since it would have neither competitors nor an easy way of measuring performance.

If an interventionist policy can be expected to work badly, the obvious next question is whether a non-interventionist policy can adequately defend us. If the answer is no, then, however skeptical we are of the government’s ability to conduct an interventionist policy well, we may have no alternative.

The case against a non-interventionist policy starts with the observation that Western Europe and Japan possess a large part of the world’s resources. By resources I do not mean natural resources. In the modern world, natural resources have very little to do with world power; that is why Australia, Canada, Kuwait, Zaire, and Zimbabwe are not world powers and Japan is. When I say that Japan and Western Europe have a large part of the world’s resources I mean that they have skilled workers, machines for those workers to use and political and social institutions which result in those workers and machines producing lots of useful things. It seems likely that if those areas were conquered by the Soviet Union, the Soviet Union would become a more dangerous enemy than it now is. It would seem to follow that the U.S., in its own interest, must defend Japan and Western Europe.

But the same things which make those countries worth conquering also make them capable of defending themselves. West Germany, France, and Japan have each about half the GNP of the Soviet Union—Japan somewhat more, West Germany and France somewhat less. The combined GNP of the Western European countries, their ability to build tanks and fighters and missiles, is greater than the GNP of the Soviet Union and its satellites.

Of course, the Europeans may not be able to get together to defend themselves—but they do not have to. If West Germany had half the army of the Soviet Union and half the missiles and half the airplanes, the Soviets would be very unlikely to invade West Germany. The Soviets have a long border with China to worry about. They have a collection of fraternal allies whose friendship is causally related to the availability of Soviet troops. And besides, it would not be much of a victory if they annihilated West Germany and lost fifty percent of their own population.

If this argument is right, then the parts of the world worth defending are parts that can defend themselves. We are left only with a problem of transition. Given that the Germans and the Japanese do not currently have the military forces to defend themselves, how do we persuade them to acquire those forces and make sure that they do not get conquered before they do so?

The first step is to make it clear that the U.S. is moving towards a non-interventionist policy, that at some point in the near future we will stop defending the countries that have been our allies. A possible second step, to shorten the transition period, is to sell our allies some of the weapons—including the warheads—with which we are presently defending them.

One advantage to having West Germany and Japan defended primarily by Germans and Japanese is that it should substantially reduce the possibility of war by miscalculation. Suppose that, under the present system, the Soviets are considering an invasion of Western Europe. They will ask themselves whether the U.S. is willing to risk its own nuclear destruction in order to save its allies. They may decide the answer is no, and invade. Whether they are right or wrong, the result, from the standpoint of both Americans and Europeans, is an unfortunate one.

The Soviets may reasonably doubt whether the U.S. is willing to start World War III in order to defend Germany or France. There is much less doubt that Germany or France would be willing to. So a world in which major countries are responsible for their own defense is likely to be a good deal safer than one in which they depend on us.

There is a second reason why the world produced by a non-interventionist foreign policy might be safer than the world we now live in. Since World War II we have had a two-power world—historically an unusual situation. It seems likely that a two-power world is inherently less stable than a many-power world. If there are only two great powers and one of them manages to defeat the other without being totally wiped out in the process, it has won the whole game. If one of the two powers has a temporary lead it may be tempted to attack, since if it does not the situation might reverse. If, on the other hand, there are five or six great powers, then a successful war by A against B lets C through F pick up the pieces. That is a good reason for A not to attack B.

My conclusion is that the U.S. should move towards a non-interventionist policy. This is not, in any sense, a principled conclusion; it is the result of balancing what I judge to be the relative advantages of the two alternatives. In order to simplify the discussion, I have put it in terms of polar alternatives, interventionist and non-interventionist. While my arguments suggest that we should prefer a policy near the non-interventionist end of the spectrum, they do not imply that the U.S. government, or some libertarian successor, should have nothing at all to do with foreign governments. One can easily imagine particular cases—a treaty to permit U.S. radar stations in Canada to give early warning of an attack over the pole, for instance—where the advantages would outweigh the disadvantages.

I started this chapter by asking whether there was a libertarian foreign policy. In one sense my answer is no. Any foreign policy that is likely to be successful in defending us involves serious moral problems for libertarians. That is one example of a point I made in an earlier chapter, the difficulty of defining individual rights in a way that does not at least occasionally lead to conclusions we are unwilling to accept. In another sense, I believe that there is a libertarian foreign policy—a foreign policy which libertarians can expect to work better than alternative policies. That policy is to defend ourselves by fighting those who actually attack us rather than by maintaining a global network of alliances. The argument is a simple one. An interventionist policy done badly is very much worse than one not done at all, and we can be sure that an interventionist foreign policy run by the U.S. government will be done badly.

The great rule of conduct for us, in regard to foreign nations is, in extending our commercial relations to have with them as little political connection as possible. . . . ‘Tis our true policy to steer clear of permanent alliances, with any part of the foreign world.

GEORGE WASHINGTON, FAREWELL ADDRESS TO THE PEOPLE OF THE UNITED STATES,
SEPTEMBER 1796.

[This chapter was written for the second edition back when the Soviet Union was still a going concern. Since then the U.S. has repeated in Iraq and Afghanistan the lesson it should have learned in Vietnam.

Also since then I have come across, in Winston Churchill’s history of the Second World War, additional evidence in support of my argument. The first time that Hitler attempted to annex Austria it was Mussolini who stopped him, announcing that Italy would not tolerate such an annexation and making his point by moving Italian divisions into the Brenner pass.

What changed that was the response of the western powers to the Italian invasion of Abyssinia. In Churchill’s view they should either have ignored the invasion in order to keep Italy as an ally or intervened with naval force to prevent it, bringing down Mussolini’s government. What they did instead was to voice their opposition without doing much of anything. Mussolini concluded, first, that they were not his friends, and second that they were not very dangerous enemies. The next time Hitler moved against Austria, it was with Mussolini’s permission. The interventionist policy of France and England had given Hitler his first significant ally.]