The Perpetual Peace Puzzle

Introduction

Kant’s theory of international relations has been enormously influential; and yet in the key text, his Perpetual Peace (1795), there appears to be a contradiction that threatens to undermine it entirely. Kant claims that sovereign states, like individuals in the state of nature, are obliged to leave that dangerous situation by submitting to a common external, coercive authority. Yet he also says that, because states have in principle achieved a situation of rightful justice domestically, they cannot legitimately submit to a ‘state of states’ over and above them. In Kant’s own words, every state, ‘for the sake of its security, can and ought to require the others to enter with it into a constitution similar to a civil constitution, in which each can be assured of its right’. ¹ That is to say, a single legitimate international authority ought to decide on aspects of right – which he elsewhere defines as ‘the authorisation to use coercion’ ² – bearing upon interstate relations. He continues:

In accordance with reason there is only one way that states in relation with one another can leave the lawless condition, which involves nothing but war; it is that, like individual human beings, they give up their savage (lawless) freedom, accommodate themselves to public coercive laws, and so form an (always growing) state of nations (civitas gentium) that would finally encompass all the nations of the earth. ³

It has therefore become almost obligatory for commentators immediately to point out that Kant’s theory is a botch. A generally sympathetic reader, for example, argues that ‘Kant contradicts himself, proposes standards that he later drops, and at several points refuses to follow his own logic towards uncomfortable conclusions’. ⁶ Another critic, not as well-disposed to the Old Jacobin, says that his argument is ‘transparently bad’, and that it is ‘telling testimony to the power of piety over intelligence that so many have for so long thought Perpetual Peace an impressive work’. ⁷

The Perpetual Peace Puzzle, then, is the puzzle of how we should account for the ostensible contradiction in Kant’s argument. As we shall see a little later, most commentators on Kant think that he did contradict himself, but they disagree on the question of why he did so. Some regard Kant as having made one argument about what is the case and another about what ought to be the case, which he did not discriminate as clearly as he might have done. Others argue that Kant got confused because the concepts with which he was working simply were not up to the task of helping him to formulate a coherent theory. By contrast, I think that Kant’s argument in Perpetual Peace is consistent, both with itself and with the arguments about moral agency that he developed elsewhere. The key to making sense of it, though, is his description of the state as a moral person. It is not my intention in this article to make an argument about what Kant should have said. ⁸ Rather, I think that the resources for understanding Kant’s argument have to be excavated from his wider philosophical project.

The first part of the article examines what Kant meant when he wrote about moral personhood, in a general sense, and contextualizes this in terms of some of the core concepts in the Kantian hinterland, namely will, reason and the moral law, which are essential for understanding the work that moral personhood did in his writings. In the second and third sections my focus turns to Kant’s theory of international relations. The moral personality of the state has moved arguments in opposing directions in the secondary literature that has dealt with the relevance of that characterization of the state for Kant’s international theory, so I focus on these contrary movements in some detail. On one recent reading – that of Katrin Flikschuh, dealt with in the second section – the juridical compulsion of states to join a state of nations or world republic is ruled out because this would compromise their moral personality. ⁹ On another – that of B. Sharon Byrd, examined in the third section – when Kant called the state a moral person he intended to indicate that it possessed certain duties to itself and to others, for the sake of which it could be coerced to leave the international state of nature. ¹⁰ It is noticeable that Flikschuh and Byrd both fail to come to grips with the Perpetual Peace Puzzle, in different ways. Flikschuh stops before the Perpetual Peace Puzzle arises as a problem, by arguing that in fact Kant’s premises cannot strictly allow him to endorse any kind of authority above that of the state, so that both the State of Nations and the Negative Surrogate appear in the same light as mere wishful thinking. Byrd appears to deny that there is a Perpetual Peace Puzzle altogether, by arguing that Kant’s argument from the moral personality of states means that, whatever he may myopically have written in one place, states really must, on the basis of his premises, sanction the State of Nations. Clearly, both interpreters cannot be right, and nor can the textual problem which gives rise to the Perpetual Peace Puzzle be ignored by stopping short of addressing it (Flikschuh) or by resolving it by means of a fiction that Kant wrote something other than what he did in fact write (Byrd). Nonetheless, by pointing to the importance of the vocabulary according to which Kant limned the state, both scholars suggest something extremely important. I argue that Kant’s understanding of the state as a moral person permits him coherently to incorporate the two seemingly contradictory propositions quoted above in a single argument. We can solve the Perpetual Peace Puzzle and we do not need to ignore it.

I

The best place from which to start to make headway on solving the Perpetual Peace Puzzle is Kant’s theory of the state. As we shall see, this theory itself is made intelligible only by careful scrutiny of the legal and ethical philosophies in which it is embedded. Kant described the state as a moral person, and he did so when dealing with the subject of interstate relations or the right of nations. In Perpetual Peace, he wrote that the state, like ‘a trunk…has its own roots; and to annex it to another state as a graft is to do away with its existence as a moral person and to make a moral person into a thing’. ¹¹ In part I of The Metaphysics of Morals (1797) on the Doctrine of Right, Kant again claimed that ‘a state, as a moral person, is considered as living in relation to another state in a condition of natural freedom’. ¹² For all the interest in Kant’s theories of politics and international relations, the locution according to which he characterized the state has received very little attention. ¹³ And yet it is crucial, I will argue, to his characterization of the proper international politics that should subsist between states, and thus to unravelling the Perpetual Peace Puzzle.

Kant may well have taken the nomenclature from the textbook on natural law published by the pioneering statistician Gottfried Achenwall. ¹⁴ At any rate, Kant’s marginal notes on his copy of Achenwall’s Ius naturale (1763), where the latter described the state as persona moralis, show Kant taking note of the terminology. ¹⁵ But Achenwall was only the latest in a line of theorists who had, since the natural jurist Samuel Pufendorf (1632–94) in his The Law of Nature and Nations (1672), called the state a moral person. In the 18th century, the most significant writers before Kant to maintain that states were moral persons were the polymath Christian Wolff (1679–1754) and the diplomat Emer de Vattel (1714–67). Pufendorf had contrasted moral persons to physical ones, where moral persons were roles or institutions – such as ‘citizen’ or ‘hospital’ – that were intersubjectively constituted and recognizable, to which were attached sundry rights and responsibilities, and which moreover were patterned on physical persons so that they could also be said, like individual physical human beings, to have reason and will. ¹⁶ The state itself, said Pufendorf, was just such a moral person. ¹⁷ When Wolff described the state as a moral person he exploited the physical–moral distinction in order to argue, against Hobbes in chapter XXX of Leviathan, that the law of nations could not be ‘the same thing’ as the law of nature applied to sovereigns. ¹⁸ This was because moral persons were different from physical persons, and therefore the law of nations would have to be worked out with due attention to this fact. ¹⁹ Vattel maintained that, just as physical persons have a duty to do what they can to perfect themselves, so do the moral persons that are states. ²⁰

As these thinkers did, Kant began by writing about moral personhood in general before he came to address the moral personhood of the state in particular. However, Kant contrasted moral personality not to physical personality but to what he called ‘natural’ personality or personhood. A natural person, he wrote in The Metaphysics of Morals, is someone to whom we may impute action based at least in some part on ‘an inner determining ground’ that ‘lies within the subject’s reason’. ²¹ In other words, if we may impute actions to an entity because we attribute to that entity some power of free will, then that entity is a natural person. A moral person, by contrast, is someone or something to which we impute not only actions but also ‘deeds’, or those actions by which we judge the actor capable of the power to choose between right and wrong. In Kant’s view, ‘imputation in the moral sense is the judgment by which someone is regarded as the author (libera causa) [free cause] of an action’ understood as a ‘deed insofar as it comes under obligatory [moral] laws’, whereas natural personhood is about the imputation of actions which are morally indifferent. ²² Moral personality, for Kant, implies ‘nothing other than the freedom of a rational being under moral laws’, and the actions of a moral person are such that ‘the agent is regarded as the author of its effect, and this, together with the action itself, can be imputed to him, if one is previously acquainted with the law by virtue of which an obligation rests on these’. ²³

Kant says, then, that ‘natural’ persons are those beings to which we can attribute ownership of their actions on the basis of some exercise on their part of reason and will, while moral personhood involves in addition that we recognize in such a being the capacity freely to perform deeds whose obligation falls under the law of right and wrong. In order, therefore, to grasp what Kant means when he calls the state a moral person, we must understand what he means by the terms will, reason and moral law.

Let us begin with will and reason. For Kant, will is not a faculty separate from intellect or reason but what he calls in the Groundwork ‘a kind of causality of living beings’ that they possess to the extent that they are in command of reason. Will, for him, is not so much a power to intervene in a chain of causes but rather ‘a causality in accordance with immutable laws, but of a special kind’. Will does not redirect a physical causal process but acts to bring the rational agent under the direction of causal laws of a different type. For the will is a causality, and since ‘the concept of a causality entails that of laws…it follows that freedom is by no means lawless’. ²⁴ The best starting point for understanding what Kant means here when he describes the will as a cause operating in accordance with law while still potentially being free is his essay of 1786, ‘What Does it Mean to Orient Oneself in Thinking?’ As the title indicates, in this text Kant considers the activity of thinking, and one of the questions that he addresses in it is, when is it appropriate to designate thinking as being free? His answer: ‘Freedom in thinking signifies the subjection of reason to no laws except those which it gives itself; and its opposite is the maxim of a lawless use of reason’. ²⁵ When a person thinks, Kant argued, she may do so in accordance with laws of nature and therefore not freely: when a person’s thoughts are passional, for instance, her thinking is in line with natural causality, because desires and emotions can be explained in terms of physical causes. But she may also think by means of what Kant called the lawless use of reason: by means of the human power that we call will, the free agent may think without such thought processes being determined by natural causes. However, Kant asserted that such lawless thinking would in fact be contingent on specific and variable circumstances, including those that are intimately a part of the agent, such as her passions. Thus this person would be led back to being determined by alien determinations as she came under the sway of other people and institutions that played on the ungoverned force of her passions. ‘The natural consequence’ of all this, he wrote:

…is that if reason will not subject itself to laws it gives itself, it has to bow under the yoke of laws given by another; for without any law, nothing – not even nonsense – can play its game for long. Thus the unavoidable consequence of declared lawlessness in thinking (of a liberation from the limitations of reason) is that the freedom to think will ultimately be forfeited. ²⁶

In that essay, the public character of the principles of reason is emphasized, but the kind of freedom addressed is the freedom of thought. But Kant’s principle of autonomy – ‘to choose only in such a way that the maxims of your choice are also included as universal law in the same volition’ – became the principle not only for thinking but also for acting, and specifically for acting morally. In The Metaphysics of Morals, Kant indeed calls all the universal laws of reason that apply to individuals in respect of their interactions ‘moral laws’. Moral laws can apply, he argues, in two different ways. First, when the moral laws are themselves ‘the determining grounds of actions’, then we call the moral laws ‘ethical laws’. ²⁹ Ethics relates primarily to the freedom of agents in the sense described above, namely their capacity to think and act in accordance with the principles of pure practical reason, but in particular ethical laws comprehend the things that human beings may do to themselves and to each other, directly or indirectly, that are relevant to questions of morality. It is in the realm of ethics that moral persons, in Kant’s sense, step into the picture, for the actions that we are now concerned with are actions bearing on good and evil. When Kant comes to probe the topic of the morality appropriate to moral persons in the Groundwork, his famous first formulation of the Categorical Imperative barely registers a change from the principle of autonomy: ‘act only in accordance with that maxim through which you can at the same time will that it become a universal law’, although the law here was explicitly a moral one. ³⁰ The Categorical Imperative, then, is the fundamental principle of morality of the ethical order governed by the principle of autonomy as the relevant law of internal freedom. The fact that it is an imperative is important. It is a command addressing the problem of moral law for individuals who are finitely autonomous and rational, and who might oppose the principle of their own legislative activity because of their various sensuous, all-too-human, inclinations. ³¹ But it is not an alien command; it lays upon people a duty which reason and therefore autonomy dictates, in order to respect the autonomy of all. It is in fact Kant’s claim that a ‘finite rational being, whose will is inevitably affected by, though not inevitably determined by, sensuous impulses, experiences objectively valid principles as imperatives or commands of reason’. ³² The human being ‘brings it about that he becomes either good or evil, according as he either incorporates or does not incorporate into his maxims the incentives contained in that predisposition’ to the good which is the mark of his humanity. ³³ The experience of the command makes a moral duty out of a moral principle. ³⁴

Reason is concerned with principles, according to Kant. The first formulation of the Categorical Imperative, given by reason, is a principle of morality governed by a principle of autonomy. Moral persons, however, are not only reasonable but also rational. By this, Kant means that when moral persons act, they act not only on principle but also for ends. Reason will thus have to discover a formulation of the Categorical Imperative appropriate to individuals who act also for ends. As with the first formulation, this must be universalizable in order to count as a moral law. Kant’s second formulation of the Categorical Imperative thus states that when human beings act they ought always to act so as to treat all of humanity as an end and not as a mere means to an individual’s own end: ‘So act that you use humanity, whether in your own person or the person of any other, always as the same time as an end, never merely as a means.’ ³⁵

The other part of Kant’s two-sided conception of the laws of freedom deals with what he calls the ‘juridical’ aspects of freedom. The major difference is that while ethical laws are the ‘determining grounds’ of action in the ethical sphere, ‘juridical laws’ are ‘directed merely to external actions’ or the ‘external use of choice’ and concern only the ‘conformity to law’ of these actions and choices. ³⁶ When we deem one another as capable of owning actions and deeds, says Kant, and thus see one another as moral persons, we also judge one another as being capable of external freedom, or the ability to make choices corresponding to our desires. Kant claims that in virtue of each person’s humanity each has an ‘innate right’ to make such choices, and that other people’s choices should not interfere with or take away this fundamental right. ³⁷ Kant then appeals to the distinction between subjective ‘rights’ (which we possess) and objective ‘right’ (which posits the proper order of relationships between individuals). The concept of right, he holds, guides the ‘form’ of the external relations between individuals with respect to their choices:

The concept of right…has to do, first, only with the external and indeed practical relation of one person to another, insofar as their actions, as deeds, can have (direct or indirect) influence on each other. But, second, it does not signify the relation of one’s choice to the mere wish (hence, also to the mere need) of the other…but only a relation to the other’s choice. Third, in this reciprocal relation of choice no account at all is taken of the matter of choice, that is, of the end each has in mind with the object he wants. ³⁸

However, Kant was clear that the external use of choice was still subject to the moral laws. He thus formulated a universal principle of right that resembled the Categorical Imperative except that it governed the external use of choice of persons rather than their ethical freedom:

Any action is right if it can coexist with everyone’s freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone’s freedom in accordance with a universal law. ³⁹

Kant then introduces an interesting analogy to the physical laws of nature. The ‘freedom of everyone under the principle of universal freedom’ can be understood, he writes, ‘by analogy with presenting the possibility of bodies moving freely under the law of the equality of action and reaction’. Rights-relations pertain to embodied persons coexisting on the enclosed surface of the earth, and the concept of right comprehends a spatially bounded system. The law of the equality of action and reaction in the physical realm, Kant argued, might also be called a ‘law of reciprocal coercion’, and so the same might be said of the form of right. ⁴¹ Reciprocal coercion, in the moral realm, means that the universal principle of right can be enforced coercively but legitimately. When people co-exist together in a spatially circumscribed world, Kant argues, each person’s exercise of her or his capacity of external freedom inevitably affects everybody else’s exercise of that capacity. After all, having external freedom must mean having the means with which to set and pursue our ends, and these will involve at least in part external objects of our choice that we ‘have the physical power to use’, objects over which others might also claim rights. ⁴² The universal principle of right expresses a criterion for the compossible exercise of the external use of choice of each individual. But before individuals have left the state of nature nobody can enforce this principle; given the ‘innate equality of each’, each person lacks any authority by which to prescribe coercive law for everybody else. This is what the Kant literature calls ‘the assurance problem’. A ‘unilateral will’, he proclaims, ‘cannot serve as a coercive law for everyone’, because a private will would be contingent on one person’s judgement alone, and would not put that private will under the law in the same way as all others. Only ‘a will putting everyone under obligation, hence only a collective general (common) and powerful will’ is a legitimate agent for imposing a universal coercive law that can uphold the universal principle of right. ⁴³ This is why the state is necessary: in entering into a civil condition, coexisting and legitimate rights claims against one another can be justifiably managed by a legitimate authority whose public will can pronounce valid coercive law.

The laws of freedom or morals are thus to be policed in different ways, according to Kant. Ethical freedom is to be self-enforced by autonomous reason in line with the stipulations of the Categorical Imperative. External freedom is to be protected by a public authority in line with the universal principle of right. The final incentive for the proper exercise of ethical aspects of freedom, for real persons with their various sensibilities, is the idea of duty, while the final incentive for the proper exercise of juridical aspects of freedom appeals to baser motives:

But that lawgiving which does not include the incentive of duty in the law and so admits an incentive other than the idea of duty is juridical. It is clear that in the latter case this incentive which is something other than the idea of duty must be drawn from pathological determining grounds of choice, inclinations and aversions, and among these, from aversions; for it is lawgiving, which constrains, not an allurement, which invites. ⁴⁴

II

We are now in a position to address Kant’s discussion of the moral personality of the state, for which the above account of the moral law has prepared us. Kant explicated his theory of the state as one aspect of his analysis of the concept of right, and more specifically of public right, nicely summarized by Mary Gregor as ‘the sum of laws that need to be publicised in order to produce a rightful condition, one in which individuals, nations and states can enjoy their rights’. ⁴⁵ States come into being when individuals leave the state of nature in order that their external freedom be protected by an authority that upholds the universal principle of right. But, of course, this universal principle is not thereby upheld universally. This is because, according to Kant in Perpetual Peace, states themselves ‘can be appraised as individuals, who in their natural condition (that is, in their independence from external laws) already wrong one another by being near one another’. ⁴⁶ States together inhabit the enclosed spherical space of the earth. Analogously to the physical law of equality of action and reaction, when one state makes use of its legitimate external freedom, it impacts on all the others; and the ‘community of nations of the earth has now gone so far that a violation of right on one place of the earth is felt in all’. ⁴⁷ Kant goes further still in The Metaphysics of Morals, in which he maintains that if rights claims between states cannot properly be adjudicated and enforced, then the capacity of states to sustain the universal principle of right at the domestic level is also curtailed, for ‘if the principle of outer freedom limited by law is lacking in any’ possible form of rightful condition, then the ‘framework’ of the others ‘is unavoidably undermined and must finally collapse’. ⁴⁸

So what did Kant advocate in order that states could extract themselves from this lawless and dangerous condition? It is here that we come to the apparent discrepancy, which I noted in the Introduction, between the universal principle of right – which dictates that all states ought to enter into a world republic – and Kant’s abrupt declaration that this will not happen and that the best that can be hoped for is the ‘negative surrogate’ of a voluntary league of nations. Different commentators have tried to account for the Perpetual Peace Puzzle in different ways. Some have pointed to the peculiarity of Perpetual Peace, with its peace-treaty structure and satirical and ‘ironical tone’, and have contended that ‘Kant himself did not take [it] too seriously’, and that contradiction is to be expected among such ‘reveries’. ⁴⁹ For others it is still to be considered as a serious work, but they have supposed that there is a ‘tension between heart and head’ on display, and that the capitulation is evidence of the head winning out. ⁵⁰ ‘Kant the philosopher of right…in the tortured course of Perpetual Peace lost the battle with Kant the practical politician.’ ⁵¹ Some other expositions focus on the apparently pragmatic concerns about a world state adumbrated by Kant, such as his warning that ‘as the range of government expands laws progressively lose their vigour, and a soulless despotism, after it has destroyed the seeds of good, finally deteriorates into anarchy’. ⁵² But others have argued persuasively that Kant’s objection in this respect was only to universal monarchy. ⁵³ Fewer scholars have focused on conceptual issues. Of those who have, most have begun by arguing that Kant did not mean to draw an analogy between individuals in the pre-civil condition and states in the international system. Chiara Bottici, for example, argues that the analogy is merely ‘rhetorical’. ⁵⁴ She quotes Kant:

…the difference between the state of nature of individual human beings and of families (in relation to one another) and that of nations is that in the right of nations we have to take into consideration not only the relation of one state towards another as a whole, but also the relation of individual persons of one state towards the individuals of another, as well as towards another state as a whole. ⁵⁵

Similarly, Arthur Ripstein and Kjartan Koch Mikalsen independently argue that the analogy does not hold because, unlike individuals in the state of nature, states do not have the external objects of choice that gave rise to the assurance problem in the state of nature and for which the state was the solution. This is because ‘territory is just the spatial manifestation of the state’ rather than its property, which it possesses necessarily and for which it needs no omnilateral authorization in the same way that persons in the state of nature needed authorization of their own acquired property. ⁵⁷ Territory ‘counts as embodiment and not property’, so any analogy between individuals and states is partial. ⁵⁸ (Mikalsen also considers that Kant’s argument, that in the international state of nature states ‘wrong one another by being near one another’, points to the fact of territorial contiguity and does not bear on the problem of property.) But the trouble with this argument is that the contrast is not so clear: a state’s borders are not natural in the way that a person’s body is. As Helga Varden points out, there ‘can be no rightful borders determined by individual states in the [international] state of nature for reasons similar to those explaining why there cannot be rightful private property boundaries in the state of nature’. ⁵⁹ We know that there is a problem of indeterminacy with regard to state borders that does not exist when it comes to individual persons and their bodies. Ironically, then, this difference between states and individuals gives us reason to take seriously Kant’s commitment to an analogy between persons and states.

Perhaps the most compelling explanation of Kant’s reticence to follow through on the premises of his analogical argument from individuals in the state of nature to states in the international system has been put forward recently by Katrin Flikschuh. I want to focus on her account in some detail, because its limitations reveal why Kant’s description of the state as a moral person is so important for his international theory and for making sense of the Perpetual Peace Puzzle. ⁶⁰ Flikschuh recognizes that for Kant states have ‘moral personality’ and that this ‘warrants the…analogy between individuals and states’. But, she argues, there is an important limitation of this analogy, which is that ‘individuals’ wills are juridically non-sovereign, whereas states’ wills are juridically sovereign’. ⁶¹ The state itself is the very principle of the morally justified coercion of external freedom in accordance with the universal principle of right. It is this that justifies its sovereignty. Now, in Kant’s words, every state by definition ‘involves the relation of a superior (legislating) to an inferior (obeying…)’. ⁶² But, says Flikschuh, under the idea of an international civil condition for states akin to the civil condition between individuals, ‘the concept of state is made to occupy the position of superior and inferior, ruler and ruled, simultaneously’, and this is a contradiction. ⁶³ She considers that Kant had moral reasons for hedging his analogy between individuals and states. ‘Non-sovereign individuals can be compelled into the civil condition as the only condition under which their [rights] claims become enforceable. Their being so compelled in no way affects their status as (non-sovereign) moral persons.’ But it is morally impermissible to compel states themselves to enter into a state of states, because, she holds, ‘in contrast to non-sovereign individual persons, their possession of sovereign authority is the very mark of their distinctive moral personality as states’, and thus ‘the juridical compulsion of states would compromise their moral personality’. ⁶⁴ The force of Kant’s argument, then, is conceptual. Too many commentators have, in her view, latched onto Kant’s assertion, quoted above, that states simply do not want to enter into a civil condition with one another and thus have attributed to him a declaratory conception of sovereignty, according to which the state’s ‘capacity to act on its own terms serves as warrant for its refusal to recognise any higher juridical or indeed moral authority’. ⁶⁵ Rather, she writes, Kant’s conception of sovereignty is juridical, as he himself made clear:

What holds in accordance with natural right for human beings in a lawless condition, [that] ‘they ought to leave this condition’, cannot hold for states in accordance with the right of nations (since, as states, they already have a rightful constitution internally and have hence outgrown the constraint of others to bring them under a more extended law-governed constitution in accordance with their concepts of right). ⁶⁶

The moral personality of the state, on this reading, is nothing but a ‘publicly constituted juridical will’ that is itself the ‘condition of legitimately enforceable [external] freedom claims between private persons’. ⁶⁸ States are therefore subject only to the morality of right: specifically, that share of public right which Kant calls international right. ‘The idea of individual autonomy’, Flikschuh argues, ‘plays no role in Kant’s philosophy of Right.’ ⁶⁹ States are not subject, then, to ethical laws of freedom. And so Kant must be left to flail: he wants states to enter into a civil condition with each other, but his own argument indicates that this is morally impermissible from the standpoint of the concept of right. He gestures at such a proposal, but it is little more than meaningless hand-wringing. Hence the confused recommendation of a non-coercive league where each state self-enforces principles of international right, even though self-enforcement is irrelevant to the morality of right. Kant bequeaths to the liberal international order founded by those he inspired, according to Flikschuh, an international society of uncoercible agents, which on his own terms is a situation of injustice. In the final analysis, Kant’s is a ‘failed argument from an analogy between individuals and states’, and we are all heirs to his failure. ⁷⁰

Flikschuh’s interesting critical exegesis of Kant, however, goes wrong at one point. The major problem with such an account of Kant’s notion of the moral person of the state is that it equates the moral personality of the state to legal personality in its contemporary sense. To be sure, the history of the concept of moral personhood has often seen it inscribed in a legal context. For example, corporations had long been described as personae fictae or fictional persons when Kant wrote, and medieval French canon lawyers translated persona ficta as personne morale or ‘moral person’ in their discussions of how it was that corporations such as hospitals and monasteries could acquire property or receive gifts. ⁷¹ The point was to mark that corporations were independent legal subjects, as real in law as its individual human subjects, although they were subjects by legal fiction, real subjects only within the framework of the law. ‘Moral person’ was a juridical notion; and when Flikschuh records Kant as describing the state as a moral person she says that this means that the state possesses a ‘juridical will’ and is therefore subject only to public right. ⁷² However, Kant’s use of the concept of moral personality is discontinuous with this legal heritage, just as Pufendorf had used the notion of moral personality to denote an institutional identity and centre of willing and action in a social, political and economic world served by law but by no means existing for the sake of law. When he takes up the term, Kant also does something quite different with it. We saw above that Kant intends moral personhood to convey that a ‘natural’ person – one to whom we could attribute ownership of any actions at all – also owns those of her or his actions that fall under moral laws, including ethical laws as much as juridical ones. Might it not be the case, therefore, that by calling the state a moral person, Kant explicitly inscribes it within the ambit of the ethical as much as the juridical? States might be as much a part of the kingdom of ends as they are part of the system of right. They might be duty-bound to adhere to the maxims of the moral law.

III

I think that it is the case that the wider frame of reference for Kant’s use of the term ‘moral person’ is thoroughly germane to his philosophy of international relations. Indeed, the portrayal of the state in terms of the more complete set of commitments involved with the concept of moral personhood has not gone entirely unnoticed by Kant scholars. B. Sharon Byrd has probed Kant’s label more systematically and fully than anyone else. ‘Central to the notion of a moral person’, she points out, ‘is that the person is self-legislating, or autonomous, is subject to moral laws, is endowed with freedom of choice, and therefore is someone whose actions are capable of imputation.’ ⁷³ Unlike Flikschuh and the other interpreters of Kant surveyed above, Byrd argues for a strong analogy between human beings and states in Kant’s thought, secured by the moral-person description. She argues that Kant uses the notion of the state as being a moral person precisely so as ‘to indicate that a state has perfect and imperfect duties to itself and to other states corresponding to [his] portrayal of these duties for the individual’. ⁷⁴

Where Flikschuh asks us to read Kant’s international theory in light of his Doctrine of Right, Byrd asks us to read it in light of the second part of The Metaphysics of Morals, his Doctrine of Virtue. It is here that Kant elaborates a distinction between perfect and imperfect duties. ⁷⁵ This is a complex division that Kant expounds in different ways. First, perfect duties prescribe an action that we are to take (or omit), while imperfect duties prescribe ‘only the maxim of the action…not the action itself’. ⁷⁶ That is to say, ‘perfect duties impose requirements on actions; imperfect duties on ends’. ⁷⁷ Second, perfect and imperfect duties are obligatory in different ways. The only kind of latitude permitted in respect of perfect duties is to decide whether a given moral principle is relevant at all, and if the policy of conduct prescribed by the duty is disjunctive in form – if it can be satisfied by doing A or B or C – then to decide on how best to discharge the duty. Perfect duties specify act-types (‘Repay Jane the £10 you owe her’), and some act-token (paying her with 10 £1 coins, or 2 £5 notes, or a cheque) of the act-type specified by the duty must always be observed. ⁷⁸ Because perfect duties prescribe policies of conduct rather than the form of conduct, then, they admit of little ‘playroom for free choice’. ⁷⁹ By contrast, imperfect duties, while obligatory, admit of much greater latitude. The Metaphysics of Morals is ambiguous about this latitude. The designation of a duty as imperfect, says Kant, ‘is not to be taken as permission to make exceptions to the maxim of actions but only as permission to limit one maxim of duty by another (e.g. love of one’s neighbour in general by love of one’s parents)’. ⁸⁰ Thus, Marcia W. Baron argues that the latitude of imperfect duties permits an individual to forgo doing an action if she is already performing another act that falls under the principle of imperfect or perfect duty; whereas Thomas E. Hill argues that the latitude extends to allowing people to forgo performing an action as long as they will do so at another point in time: people have a pro tanto reason to discharge imperfect duties on some occasions, only because otherwise they would be too demanding. ⁸¹ Thirdly, Kant explains that perfect duties in principle admit of being perfectly attained, whereas imperfect duties enjoin ends which are only approachable and which we will never absolutely be able to attain. ⁸²

The kind of point that Byrd that wants to make by picking up on Kant’s description of the state as a moral person is well illustrated in the following quotation:

[T]he state has a perfect duty to itself not to permit revolution and its own dissolution, just as the individual has the perfect duty to maintain himself and not commit suicide. The state has the imperfect duty to itself to reform the constitution, just as the individual has the imperfect duty to himself to develop his talents. The state has perfect duties to other states corresponding to the three Ulpian formulae honeste vive [live honourably], neminem laede [injure no one], and suum cuique tribue [give to each their due] similar to duties of law as applicable to individuals. Finally, the state has the imperfect duty to other states to render aid in times of sudden need, just as the individual has the imperfect duty to be charitable to others. ⁸³

However, Byrd’s own elucidation of the perfect/imperfect duties distinction draws out the contrast in a manner which is unwarranted by anything in Kant’s texts, but with important implications for her own argument in respect of the Perpetual Peace Puzzle. Imperfect duties, she says, are ‘ethical duties’, whereas perfect duties are ‘legal duties’. ⁸⁶ While ethical duties, Byrd holds, correspond to the determining grounds of choice and cannot be coerced, legal duties correspond to the law of external freedom and to the sphere of right, and are thus coercible. Here, however, Byrd collapses two different distinctions that Kant did indeed make into one. He certainly did distinguish between juridical duties (or duties of right) and ethical duties (duties of virtue), where juridical duties are those to which people can be held by coercion. ⁸⁷ But he did not equate all perfect and juridical duties, nor did he identify all imperfect duties with ethical duties. For instance, among perfect duties to oneself Kant included prohibitions on lust, drunkenness and gluttony (as misuses of one’s capacity for freedom), and on lying, avarice and servility (for these undermined one’s freedom of choice directly). ⁸⁸ Among perfect duties to others he enumerated proscriptions of arrogance and ridicule. ⁸⁹ None of these did he consider as giving rise to legally enforceable rights claims on the part of those injured. The distinction between perfect and imperfect duties comes out most clearly when these duties are indeed phrased as prohibitions. It is at base a distinction between prohibitions of activities that either destroy or compromise free choice (perfect duties) or activities that would erode the capacity to cultivate or facilitate end-setting and the pursuit of such ends (imperfect duties). As Paul Guyer argues, all follow in one way or a other from the Categorical Imperative to treat human beings as ends rather than means. ⁹⁰ Ethical duties are all either perfect or imperfect and none of them is necessarily coercible by dint of being so grouped, which is why Kant explains them all in the Doctrine of Virtue, which is about ethics, rather than in the Doctrine of Right, which is about justice.

Whatever Kant may have written about what states want, Byrd says, states are moral persons, and just as human individuals are obliged by perfect duty to leave the pre-civil condition, so states have a perfect – and coercible – duty to leave the international state of nature. ⁹¹ Byrd’s reading thus begins with the claim that as moral persons states are ethical agents capable of acting on principle, but it ends by suggesting that the commitment of states to enter into a world republic is a requirement of Recht. But Byrd’s interpretation is not only undermined by a textually unjustified folding of one distinction into another; it is also crippled by another fundamental lack of evidence. There simply are no documentary grounds for such an interpretation of Kant, who nowhere says that states are legally obliged to renounce their sovereignty and enter into a state of nations. She solves the Perpetual Peace Puzzle by extrapolating from an argument that Kant does not make in order to argue that he says something that he does not say.

All the same, Byrd is right that the imperfect duty of self-perfection that Kant ascribes to individuals also applies to states, and it is in this respect, I think, that Kant’s characterization of states as moral persons begins to make sense of the Perpetual Peace Puzzle. How so? As a moral person, the state is more than a natural person, and the ‘more’ indicates that we take it to have not only a broad will that may choose between options about how to act, but also a specific power of the will to recognize and choose between options which fall under the moral law of right and wrong. This is the whole moral law, not solely the juridical aspect of the law governing external actions but also the ethical aspect governing the proper use of autonomy. As a ‘moral person’, therefore, according to Kant the state must have a capacity to be moved to act by the idea of duty. Duty in the abstract decomposes into definite perfect and imperfect duties, and the moral person of the state has perfect and imperfect duties like any other. According to Kant, the fundamental imperfect duty to oneself is the cultivation of one’s own excellence in order to facilitate the pursuit of the ends that one has set oneself. Such efforts must necessarily be endless as self-perfection can never be realized, and therefore the duty to pursue it must be imperfect. Likewise, the moral person of the state has an imperfect duty to pursue its own self-perfection. And states will undeniably stand a greater chance of perfecting themselves if war is eradicated. Even though a state of states is, from a juridical perspective of the kind clarified by Flikschuh, a contradiction,

…the political principles directed towards perpetual peace, of entering into such alliances of states, which serve for continual approximation to it, are not unachievable. Instead, since continual approximation to it is a task based on duty…this can certainly be achieved. ⁹²

Conclusion

In this article, I have tried to make sense of the Perpetual Peace Puzzle, namely Kant’s curiously incongruous argument that states, according to the universal principle of right, should ‘give up their savage (lawless) freedom [and] accommodate themselves to public coercive laws’, but that, ‘in accordance with their idea of the right of nations’ that follows from this same principle, they cannot be expected to do this, and thus the best that can be hoped for is ‘the negative surrogate’ of a voluntary league in which member states agree to try to manage their relationships in order to forestall conflict. Why did Kant straightaway give up on the positive implications of his universal principle of right? What is it about the right of nations that means that the principle does not apply to states?

I have argued that we can get a handle on the Perpetual Peace Puzzle by attending to Kant’s argument that the state is one species of the genus ‘moral person’. Moral persons, according to Kant, are agents to whom we can attribute the power to discriminate and choose between actions that fall under the moral law. As we have seen, the moral law is for Kant two-pronged. In its juridical aspect, the moral law is directed to agents who can make choices in accordance with their desires and who must coexist in a physical space with other such agents. In its ethical aspect, the moral law is directed to agents who can make choices in accordance with the idea of duty. In terms of the juridical aspect of the moral law, states are both analogous and disanalogous to individual human beings. They are analogous because both states and individuals must take account of other such beings when they make choices based on their desires. They are disanalogous, however, because the state itself comes into being to bring about a rightful condition among individuals, that is, a condition in which each individual’s freedom of choice can optimally be accommodated to that of every other by putting all such individuals under a general will that upholds the universal principle of right coercively but legitimately. The disanalogy is very important, then, because, as Flikschuh demonstrates, it helps us to understand why Kant would argue that, on the one hand, states should enter into a world republic, but why, on the other, they must refuse to do this. In other words, the respects in which individual human beings and states are both analogous and disanalogous from the perspective of right explains in part the Perpetual Peace Puzzle.

But it is only a partial explanation, for it cannot explain why Kant continues to insist that states ought to commit to a non-coercive league of states instead. For if all it means for a state to be a moral person is that it possesses a ‘publicly constituted juridical will’ that is itself the ‘condition of legitimately enforceable freedom claims between private persons’, then it is hard to see how Kant can want the state to enter into any wider association that is not itself a stipulation of public right. ⁹³ The solution to this part of the puzzle is to persevere with Kant’s argument that states are moral persons. To be a moral person does not mean that the agent so described is subject only to the universal principle of right; a moral person is also subject to the ethical laws of freedom. We are entitled to impute to states, as moral persons, a capacity to regulate themselves according to the idea of duty. Ethical duty in the abstract finds form in concrete perfect and imperfect duties. As a moral person, therefore, the state has perfect and imperfect duties to itself and to others, and these duties explain why it is that states should enter into a voluntary league of nations, because such a league is a necessary condition for states to work towards their own betterment and that of others, which is the fundamental imperfect duty.

The greatest of thinkers who managed to wrestle systematic thought from contradictions, Hegel, famously chided the ‘empty formalism’ of Kant’s ethics, and argued that Kant’s ‘criterion that there should be no contradiction’ between duty in the abstract and the actual duties that follow from the principle ‘is non-productive – for where there is nothing, there can be no contradiction either’. ⁹⁴ I have argued that Kant’s philosophy of international relations is not as contradictory as it might appear to be and as it has appeared to others. I am sure that many contemporary ‘Kantian’ analysts of global ethics – who have tended to concentrate on the system of right, and for whom virtue has not been a major concern – will not think that the Perpetual Peace Problem can be resolved by an appeal to duty or virtue, which is what I have argued is in effect Kant’s own answer to the deep conceptual problem in his system of right. ⁹⁵ They may well be right to have serious concerns. But this is Kant’s answer nonetheless. Whether it is ‘productive’ can remain an open question. Now that the answer has been made clear I hope that it might illuminate further thinking as well as provide a better-lit target.