Thomas Hobbes had the ability to shock. The most famous statement in his Leviathan (1651) was that human life in the natural state would or could become "solitary, poor, nasty, brutish and short"; this was a deeply disturbing claim to make at a time when most people believed in a God-given natural order of things. According to Hobbes, there was a natural disorder of things, and the only way to keep disorder at bay was to set up an artificial institution, the state, endowed with enough power to deter violence and promise-breaking among its subjects. As for the relationship between one state and another, this was similar in some ways to the relationship between individuals in the "state of nature": Order could not be guaranteed, because there was no overarching authority to maintain it.

Today, Hobbes's theory of the state is intensively studied and is found to be not so shocking after all. His analysis of the basis of political authority--of the implicit engagements that bind human beings in a political community--is complex and intriguing and has been studied sympathetically by conservatives, liberals, Oakeshottians, Kantians, game theorists and historians of natural law. But there is one area of his theorizing that is still regarded as somehow crude and extreme: his account of international relations. Here is an aspect of Hobbes's thought that has been constantly referred to or exclaimed against but hardly studied at all.

The people doing the referring and the exclaiming have been general writers on international relations theory. Open almost any standard international relations textbook, and you will find Hobbes described as an archetypal ultra-realist. This is a tradition of interpretation to which both realists and anti-realists have contributed. E. H. Carr called Hobbes the second great realist, after Machiavelli; Michael Walzer located realism "at its source and in its most compelling form" in the works of Thucydides and Hobbes. Yet the theory attributed to Hobbes by modern writers on international relations is so crude and simplistic as to make one wonder why anyone should regard it as great or compelling--or worth bothering about at all.

According to Hans Morgenthau, it is "Hobbes's extreme dictum" that "the state creates morality as well as law" and that there is no morality outside the state; according to Stanley Hoffmann, therefore, relationships between states can only be a matter of "simple amorality." The influential theorist Martin Wight claimed that Hobbes, like Machiavelli, regarded politics as nothing more than the art of "obtaining and preserving state power as an end in itself"; both Carr and Morgenthau saw Hobbes as an advocate of a kind of international power politics that must necessarily involve wars of expansion and aggression; Hannah Arendt even identified him as a forerunner of imperialism and racism.

One common ploy--not confined to the international relations textbooks--is to pair Hobbes off against Kant in a Manichaean contrast between darkness and light. Thus Robert Kagan observes, on the first page of his Paradise and Power (2003): "Europe is moving beyond power . . . [into] a post-historical paradise of peace and relative prosperity, the realization of Immanuel Kant's 'perpetual peace'. Meanwhile the U.S. remains mired in a Hobbesian world where international laws and rules are unreliable." Kagan is unusual, of course, in defending the Hobbesian side of the argument. But in making his defense (on the grounds that "Europe's new Kantian order could flourish only under the umbrella of American power exercised according to the rules of the old Hobbesian order"), he does not challenge the assumption that these are two fundamentally different approaches to international relations--or that the Kantian system would, of course, be morally preferable if only it could be made to work. It seems that we are just stuck with Hobbesianism as a description of the world--and that if descriptive realism is true, then some sort of prescriptive realism must follow.

But is the simplistic Hobbesian view, as set out in the international relations textbooks, convincing either as a description of reality or as a set of prescriptions? The answer has to be "no"--and this should not surprise us, given the assumptions of most of those textbooks' authors. Hobbes the ultra-realist has long served as a theoretical straw man, designed to demonstrate the inadequacy of realism as such. Those who, like Kagan, defend the "Hobbesian" view are thus riding into battle on a horse that has already been declared broken-backed by most of the leading experts in the field.

The disproof of textbook Hobbesianism goes like this. Hobbes's argument depends on an analogy between individual human beings in the state of nature on the one hand and sovereign states in the international arena on the other. But this analogy fails in three important ways. First, Hobbes argues that all individuals are equal, because even a strong person can, in some circumstances, be killed by a weak one. This cannot apply in the case of states. Second, Hobbes's whole argument about individuals is based on self-preservation--in other words, the avoidance of death, the summum malum (greatest of all ills). But there is no such clear-cut summum malum in the case of a state: Terrible destruction can take place while the state formally survives as an entity, while on the other hand a state can cease to exist (thus failing the self-preservation test) when its members peacefully and cheerfully vote to be merged with another state.

And third, if the Hobbesian parallelism between individuals and states really held, Hobbes's own argument would require sovereign states to do what his individuals do when they extricate themselves from the disordered state of nature--gather together and subject themselves, collectively, to a higher authority. Hobbesianism would thus require the creation of a world state, something Hobbes never recommended and from which all "Hobbesian" realists instinctively recoil.

These objections, like almost everything else to do with the straw man version of Hobbes, fall apart on closer inspection. When Hobbes says that a strong person can be killed by a weak person, he refers, among other things, to scenarios in which the weak person acts in an alliance with other people--a situation that clearly also applies in the case of states. But there is a larger point here. When Hobbes argues for the equality of individuals in his state of nature, he does so because he wishes to prove that this "equality of ability" will lead automatically to such a degree of competitiveness and conflict as to turn the state of nature into a constant existential threat against the people who are in it. Once a civil state has been formed, the level of existential threat drops away, and the people in that state have much less motivation to threaten the existence of people who live in other states. The famous analogy between individuals in the state of nature and states in the international arena is never a strict parallelism in Hobbes's argument; those disproofs of Hobbes that assume that it is are therefore bogus.

Other aspects of the standard view of Hobbes are equally mistaken. Take, for example, the claim that he regards morality as the mere creation of the state and that relations between states must exist in a realm of "simple amorality." Hobbes does in fact devote many pages in the first part of Leviathan to demonstrating that the rules of morality are "laws of nature": These laws, he says, are "immutable and eternal", and "the science of them is the true and only moral philosophy." In the state of nature these laws exist and are knowable. The only problem is that circumstances may often make it impossible to act on them, because doing so would incur risks to the actor's own life. But even in the state of nature such circumstances do not always apply. Hobbes describes scenarios--for example, a contractual arrangement in which the other side has performed his part of the bargain first--in which there is a duty to obey the natural laws. And on this point, at least, there is a rough parallelism between the state of nature and the international arena. There can be no doubt that Hobbes envisages natural-law duties applying to international affairs, since the various formulations he gives of the natural laws include "that all men that mediate peace be allowed safe conduct" and "that men allow commerce and traffic indifferently to one another." At one point Hobbes even summarizes his argument with the simple formulation "the Law of Nations, and the Law of Nature, is the same thing."

What about Hobbes as the proponent of power politics, wars of aggression and imperial expansion? It is true that he comments that "force and fraud are in war the two cardinal virtues"; but that observation flows from his argument that the state of war is the worst possible state for mankind. It is also true that he declares: "I put for a general inclination of all mankind a perpetual and restless desire of power after power." Yet, as the rest of his argument shows, his notion of "power" here is abstract and strictly instrumental. (He defines it as "the present means to some future apparent good.") He is simply pointing out that whatever aims a person may have--for example, a peaceful and beneficent life--that person will necessarily desire, at any given moment, the means towards those ends.

The popular idea of Hobbes as a believer in the principle that "might is right" is thoroughly mistaken. Hobbes insisted, emphatically, that might did not make right. Even in the case of conquest by force, he argued, it is the consent of the conquered, not the force of the conqueror, that creates the new rights of government.

Hobbes was certainly not an advocate of wars of aggression. He ridiculed monarchs who "affect war for itself, that is to say, out of ambition, or of vain-glory." Listing the "diseases" of a state in Leviathan, he included "the insatiable appetite, or bulimia, of enlarging dominion; with the incurable wounds many times received from the enemy; and the wens, of un-united conquests, which are many times a burden, and with less danger lost, than kept." And although he regarded colonization as a permissible solution to the problem of an excess labor force in the mother country (he had been, for several years, a shareholder in the Virginia Company), he argued that the colonists were under a moral duty to treat the native people humanely and to encourage them to use greater productivity to compensate for the loss of territory. Hannah Arendt's attempt to portray Hobbes as a forerunner of imperialist "racism" was peculiarly ill conceived. Hobbes poured withering scorn on the Aristotelian claim that some native peoples were "natural slaves", and it is hard to think of any early modern thinker who argued more robustly against the idea that any one group of human beings was naturally superior to any other group.

Thomas Hobbes was, then, neither a fool nor an ogre. But was he just a middle-of-the-road natural law theorist advising people to be virtuous? The Hobbesian theory is more complicated, and more distinctive, than that. And it does have some significant affinities with the tradition of so-called realism, even though Hobbes was far from being the ultra-realist described in the textbooks. But in order to understand his theory of international relations, it is necessary to go back to his concept of natural law and his account of the formation of the state.

For Hobbes, natural law was objective, in the sense that universally true statements could be made about it. But at the same time, it was subjectively grounded, in the sense that it derived its force from the existential requirements of each individual. Although the desires of different individuals would always be different, it would be true in every case that staying alive was the condition of fulfilling those desires. Self-preservation was thus a systematic requirement, and peace was the systematic condition in which self-preservation could best be secured. Hobbes's natural laws were rules for the attainment of peace. They were universal not because they concerned some universal entity or value (humanity, the common good) but because they were duplicated in the case of every individual: Peace is of ultimate value for me, therefore I should seek it, and for you, therefore you should seek it. And so on.

Such natural duties are entirely self-regarding, which means that whenever my self-preservation is better served by breaking those natural laws, I have the right to do so--and that, in the state of nature, is the problem. Of course, people whose self-regarding rights and duties are in broad alignment can live together and cooperate; even in Hobbes's "state of nature" there are social formations and "confederacies" of various kinds. But without a surrounding framework of authority, there is no reliable way of settling the disputes that arise when people's rights and duties diverge. To put it another way, at this stage there are no duties to other people, merely self-regarding duties to behave, if possible, in certain ways towards them. (Similarly, one might say that my natural-law duty not to eat poisonous berries is a duty to behave in a certain way towards those berries, not a duty to them.)

My duties to other people, and my claims on them (and theirs on me), arise only in a civil state. Here I enter a "jural" sphere, a condition in which, thanks to the existence of a common political authority, my fellow citizens and I are joined in a web of mutual obligations. It is still true, of course, that I have no such obligations towards people outside the state. The only difference now is that my relations with outsiders, like those of all my fellow citizens, are managed on my behalf by my sovereign.

In one sense, therefore, the relationship between a sovereign state and its neighboring states does resemble--as Hobbes explicitly says--the relationship between individuals in the state of nature. They are under no common jural framework; they have no direct claims on or duties to one another. Instead, their behavior can and should be directed by the laws of nature, which tell them to promote peace because it is in their own interests to do so. But since they cannot be held to those natural duties by any higher authority, and since obedience to natural law can always be trumped, if circumstances require it, by the assertion of natural right, this state of international affairs is always potentially a "condition of war."

It is at this point, however, that Hobbes insists that there is no complete parallelism between individuals and states. Having admitted that sovereigns are "in the state and posture of gladiators, having their weapons pointing, and their eyes fixed on one another", he adds: "But because they uphold thereby the industry of their subjects, there does not follow from it that misery which accompanies the liberty of particular men." In other words, the factors that make it essential for individuals to extricate themselves from the state of nature do not apply in the case of sovereign states.

A further difference emerges as Hobbes describes the role or duty (what he calls the "office") of the sovereign. Individuals in the state of nature must subordinate everything to self-preservation. That is the one principle from which all their rules of action must be derived. In that situation, the promotion of any particular benefit will be utterly secondary; individual concepts of benefit will differ, and natural law is concerned with the one condition--peace, the optimum means towards the end of self-preservation--on which all must agree. But the case of the sovereign is different:

"The office of the sovereign (be it a monarch or an assembly) consisteth in the end, for which he was trusted with the sovereign power, namely, the procuration of the safety of the people. . . . But by safety here, is not meant a bare preservation, but also all other contentments of life, which every man by lawful industry, without danger or hurt to the commonwealth, shall acquire."

Mere self-preservation is not enough; the sovereign is to pursue policies that will conduce to the positive benefit of his (or its) subjects.

Many pages of Leviathan--ones that tend to be skipped by most modern readers--are devoted to describing the kind of internal policy that sovereigns ought to pursue in order to promote the benefit of their subjects. Hobbes devoted less space to the foreign policy equivalent, but there are enough comments scattered through his writings to show that he had given it some serious thought. He recognized, for example, the importance of international trade, "because there is no territory under the dominion of any one commonwealth (except it be of very vast extent) that produceth all things needful." It was therefore necessary to have agreements with other states about such matters as commercial law and the rights and duties of travelers.

The most important aspect of foreign policy, in Hobbes's eyes, was the development of security alliances. "Leagues between commonwealths", he observed, "are not only lawful, but also profitable for the time they last." The emphasis here is--not surprisingly, given his warnings about the consequences of aggressive wars--on defensive alliances, the essential purpose of which is deterrence. But Hobbes was not thinking only of states banding together against threats from other sovereign states. In his historical treatise Behemoth (1679) he remarked: "It is methinks no great polity in neighboring princes to favor, so often as they do, one another's rebels. . . . They should rather, first, make a league against rebellion."

One particular threat from a non-sovereign entity obsessed him: the international or trans-national machinations of the Roman Catholic Church. Hobbes had (and expressed with tremendous vigor) philosophical reasons for his anti-Catholicism. But he was also a child of his time, a Protestant Englishman who had been 17 years old at the time of the Gunpowder Plot. Some of his arguments for international cooperation against Catholicism could quite easily be translated into modern terms by inserting the word "terrorism" in its place.

At its deepest level, however, Hobbes's vision of international cooperation went much further than that. He was acutely conscious of the fact that human actions are determined by beliefs and that the beliefs people have in their heads derive from a culture that crosses national boundaries. His long-term ambition was for a kind of cultural reformation--or, as later generations would call it, enlightenment--that would free human beings from false doctrine, priestcraft and thralldom to bogus intellectual authority.

If all such false opinions were dispelled, he wrote, people would understand their true interests much more clearly; indeed, "the human race would enjoy such secure peace that (apart from conflicts over space as the population grew) it seems unlikely that it would ever have to fight again." This was a striking statement, even an extraordinary one. The eminent Hobbesian scholar Richard Tuck may be exaggerating when he describes Hobbes as a "utopian", but we can at least say that this aspect of Hobbes's thought places him much closer to the optimistic ameliorism of the rationalist tradition than to the changeless pessimism of the realists. It also suggests that Hobbes possessed one of the key insights normally credited to liberal international relations theory--the idea that there is an essential link between the internal character of a state and its external behavior.

Was Hobbes, then, a liberal international relations theorist? No. That would be a classification too far. He was less illiberal than he has been made out to be--and he is certainly caricatured when portrayed as a crude ultra-realist. He believed in the pursuit of a constructive and cooperative foreign policy. He also thought that international relations were subject to some norms of behavior, the validity of which could and should be universally accepted. Yet there remained a deep difference between him and all the rationalist natural law theorists, because his natural law, unlike theirs, was based on ultimate self-interest: His philosophy allowed no teleology, no goal for humanity as such. For them, reason was a faculty that intuited universal values; for him, reason merely calculated the means, and individual interest supplied the end.

Hobbes thus stands far apart from the sort of liberal idealist who believes that rationality and legality have a force of their own and that it may therefore be possible to build a structure of human organization that so perfectly embodies rationality and legality as to make all subsidiary structures--such as sovereign states--irrelevant or obsolete. And the difference is not merely that Hobbes is distrustful of all abstractions, insisting commonsensically that "men rule, not laws." There is a deeper philosophical issue here.

For one of the most distinctive things about Hobbes is that he has a theory of the political realm as something sui generis, something essentially different from other forms of human interaction or organization. Many varieties of political theorist get by without any special sense of the peculiar nature of the political. Utilitarians, for example, analyze the state in terms of its utility, just as they would analyze any other level of human organization, higher or lower. An equivalent disregard for the peculiar nature of the state can be found in the case of Marxists, Neo-Thomist natural lawyers and many others. But Hobbes has a strong idea of what is special about the state--the unique nature of political authority, which transforms the human relations that take place under it, creating a jural community in the fullest (indeed, the only full) sense.

For Hobbes, then, more than for most other theorists, there is a finality about the political authority of the state. And yet, at the same time, he does not adopt the easy and crude position of supposing that if states are final, there can be nothing beyond them except sheer chaos and conflict. Instead, he has a vision of relations between states that allows for two types of ordered and positive interaction: formal agreements such as treaties (these are merely ad hoc jural relationships, not deriving their authority from any larger jural community), and conduct based on the principles of objective, universally knowable natural laws. Those natural laws are not a fig leaf or a fiction; they sum up, for Hobbes, the fundamental rules by which all our interests have to be managed. But he is also clear that, in accordance with those rules themselves, a special type of human authority has to be created--the state--and that it is only within a secure and properly functioning state that human life can flourish.