It is always a treat to read Pauline Kleingeld's work, and Kant and Cosmopolitanism is no exception.Footnote ¹ Kleingeld's vast knowledge of Kant's writings, her way of presenting Kant's argument and her engagement with other historical figures in order to elicit aspects of Kant's thought, all make for a deeply interesting, engaging and fun monograph. Furthermore, Kant and Cosmopolitanism covers an impressive number of philosophical topics in relation to Kant's philosophy of right, and even includes a separate chapter that contrasts Kant's philosophy with those of Rawls and Habermas. Both Kleingeld and I belong to the republican interpretative tradition of Kant's legal-political philosophy, so there is much in the book we agree on. As is customary, however, I will focus my comments on points of disagreement: Kleingeld's discussions of patriotism, poverty and global justice.

Patriotism is the main topic of the first chapter of Kant and Cosmopolitanism. Kleingeld quickly clarifies her focus by explaining that

At stake, then, is explaining how Kant makes room for patriotism understood as citizens’ special allegiance to their own particular state. In the remainder of the chapter, Kleingeld defends the claim that according to Kant proper patriotism is ‘essentially connected to a just political system, not to a cultural or ethnic community in the nationalist sense’ (20). More specifically, according to Kleingeld, the just political system to which one can legitimately have special allegiance is the ‘constitutional democracy’, as it involves ‘collective self-legislation’ (27). Because we participate in the democratic self-legislation of our own country only, she argues, we have special allegiances or duties towards it that we do not have in regard to other countries. Kleingeld clarifies that this account of patriotism entails a ‘negative (perfect) duty not to pay no special attention to the civic affairs in one's own state as a matter of principle, and it supports an equivalent positive (imperfect) duty to adopt the maxim to have some special concern for the state in which one is a citizen’ (31). Citizens have, for example, positive imperfect duties towards preserving their own state, to help it flourish sufficiently to secure justice for all, and to help improve its institutions, such as political and educational ones (31). Kleingeld also emphasizes that the special, perfect duty of citizens ‘not [to] renounce [their own political institutions] … as a matter of principle’ is not an ‘original’ duty ‘to support one's compatriots’. Rather, it is a duty to promote ‘the institutionalization of justice’, and ‘the [imperfect] duty of civic patriotism does not prohibit one from trying to promote just states elsewhere’ (32). Discerning which obligations take precedence at any particular time in a person's circumstances is seen as requiring the individual's moral judgement.

Starting on the assumption that Kleingeld's interpretation of Kant is correct, I first worry that the account she provides of special duties in terms of perfect and imperfect duties does not explain how this distinction corresponds to the difference between enforceable duties of right and non-enforceable duties of virtue. Kleingeld seems to imply that the perfect duties are enforceable and the imperfect are not. But this position is controversial: for example, it seems that for Kant the duty not to lie is a perfect duty, and yet he holds that lying should not be illegal as such. Nor can her position, in its current form, solve other relevant philosophical puzzles, such as why treason is characteristically deemed a public crime. Moreover, if, on Kleingeld's account, our duty to improve our legal-political institutions (including educational ones) is an imperfect duty, and imperfect duties are not enforceable, then it would seem that states could not tax their citizens in order to develop these institutions, which appears a philosophical cost of following this interpretation of Kant. Finally, it remains unclear to me exactly what Kleingeld means by saying that citizens have a perfect duty ‘not to pay no special attention to the civic affairs in one's own state’.

The above questions presuppose the correctness of Kleingeld's interpretation of Kant on the issue of patriotism, and I would now like to question this assumption. To start, I am worried about Kleingeld's portrayal of Kant's account of patriotism as one in which good patriotism only tracks states characterized by democratic self-legislation. Presumably, this entails attributing to Kant the view that good patriotism is not possible until a liberal, representative democracy is in place, because, I think, Kleingeld reads Kant as arguing that political obligations and legitimacy arise only within the context of such a state. But why link good patriotism (and political obligations and legitimacy) to a particular form of government (i.e. democracy)? For one thing, it is not clear to me that Kant links political obligations and legitimacy to a particular form of government in the way this interpretation presupposes. Alternative interpretative positions are defended by Kantians like Katrin Flikschuh, Arthur Ripstein and myself. Like Kleingeld's, these interpretative positions belong to the republican tradition, but they are included in the main competing non-democratic interpretative strand of thought within that tradition. These positions do argue that political obligations and legitimacy presuppose representative institutions of the liberal kind, but not necessarily representative democracies (even though, on these positions also, the ideal is to transform these institutions into democratic ones). These alternative non-democratic republican interpretations maintain that the minimally just state (the legal-political institutional structure that is politically legitimate and can issue political obligations) must be liberal and representative, which requires that it be a tripartite public authority and secure certain basic private and public rights for all its citizens. But the minimally just state need not be a democracy. It could also be, for example, a liberal aristocracy or a liberal monarchy. Flikschuh, Ripstein and I argue that political obligations arise, to borrow Flikschuh's phrase, once ‘self-governance’ (a set of liberal representative, or public institutions) is in place (Flikschuh Reference Flikschuh2009: 424). In contrast, on Kleingeld's position, political obligations presumably do not arise until ‘self-legislation’ (a constitutional democracy) exists. Of course, Kleingeld may be right that Kant views democracy as constitutive of the minimally just state. But I do not see her refutation of the competing liberal republican interpretations, and Kant's own texts seem to yield serious resistance to her democratic interpretation. He seems to resist this interpretation in the Doctrine of Right, and also seems to reject it explicitly in the Anthropology where, for example, he says: ‘One sees that only … [the republic: force with freedom and law] deserves to be called a true civil constitution; by which, however, one does not have in view one of the three forms of state (democracy), but understands by republic only a state as such’ (Kant Reference Kant2011; 7: 331).Footnote ²

Let me now return to the issues raised by applying Williams's famous ‘one-thought-too-many’ objection to the issue of patriotism, which will lead to the more general issue of the extent to which it is best to view, as Kleingeld does, Kant's conception of good patriotism as identical with practical or moral patriotism. As we will see, I do not think that Kleingeld's interpretation of Kant on patriotism can answer Williams's infamous love objection to universalist theories (when made relevant to the issue of patriotism), or fully captures Kant's own view. Starting with the former problem, I will first show that Kleingeld's interpretation of Kant seems to saddle him with ‘one thought too many’.

The point of Williams's ‘one thought too many’ objection to universalist theories, I take it, is that although these theories can explain why the potential lifesaver must save his wife, they can only explain this through an argument that involves making an exception to the general rule. Universalists such as utilitarians and Kantians, the objection goes, will have to argue that, although each of the drowning persons has a claim on the lifesaver to be rescued, the lifesaver should make an exception in this case because the drowning person is his wife. Universalists must appeal to the special obligations or values involved in relations between wives and husbands – and those special obligations or values are why the lifesaver can and should rescue his wife. The universalist reasoning involves ‘one thought too many’, however, because the lifesaver should have only one thought in his mind, namely, to save his wife because ‘it is she’, because he loves her. In such a situation, having first the thought (#1) ‘I should save all the drowning persons’, and then the thought (#2) ‘but I should make an exception for my wife’, is to have one thought too many. Correspondingly, in the patriotism case, if we follow Kleingeld's interpretation of Kant, then it seems Kant reasons as follows: I have an obligation to support all just institutions (thought #1), but I should make an exception for my own state because of my special obligations to it (thought #2). But if this is Kant's view, then it seems that his account also here involves one thought too many.

In addition to the problem of there still being one thought too many, the highly moralized nature of Kleingeld's account of patriotism worries me, both as a matter of Kant interpretation and as concerns the resulting philosophical position. In fact, I believe this is also part of Williams's worry about universalist theories: their inability to recognize the normative importance of our affective, social relations. In an effort to show quickly why I believe Kant himself may have a better response available, let me draw attention to a germane passage from The Metaphysics of Morals. Kant says:

Here, Kant makes two points of particular relevance for an account of patriotism. First, we must not let moral anthropology set the parameters for a metaphysics of morals (an account of freedom), since it must always be the other way around. Second, we cannot do without moral anthropology, which means that we cannot, without it, give a full account of how embodied, social beings of our kind realize freedom. Moral anthropology identifies the subjective conditions in human nature that hinder or help the realization of freedom.

Kleingeld's account of patriotism appears to me to be an account of freedom without any necessary admixture of moral anthropology, and so seems to be a thoroughly moralized account of good patriotism. It concerns only the metaphysics of morals, including a suggestion regarding which perfect and imperfect ethical duties patriotism involves. On her account of good patriotism, our special allegiance to our country is not our affectionate love for our particular state or country, but only our ethical duties (practical love) towards its just institutions – the institutions through which we self-legislate and thereby realize rightful relations. Since ethics (virtue) aims to make sense of our special allegiance, this account of good patriotism leaves no necessary role for ‘moral anthropology’.

An apparently unsolved philosophical puzzle in Kleingeld's account also points to part of the reason why, I suspect, Kant thinks we need both a moral anthropology and a metaphysics of morals (a theory of freedom) in order to provide a complete philosophical account of a phenomenon like patriotism. Patriotism seems to involve an irreducible notion of affectionate love of one's particular country, and highly or purely moralized accounts of patriotism struggle to make good sense of this.Footnote ³ Hence, although I believe Kleingeld is quite right to say that morality matters in our love relations – especially as we encounter difficult loving relationships – I suspect we should not be too quick to moralize love itself, or settle for too moralized an account of love (including our love of country). For instance, when I learned on 22 July 2011 of the bombing of the government buildings in Oslo and the shooting of the children on a nearby island, my spontaneous, unreflective reaction was grief. This, it turns out, was also the reaction of most other Norwegians. In contrast, non-Norwegians’ around the world reacted with moral horror, moral shock, sadness and sympathy for, and empathy with, Norwegians. I believe the responses after 9/11, and after the terrorist bombings in Spain and the United Kingdom, were parallel: Americans, Spanish and British people (respectively) grieved, while others around the world were morally horrified, morally shocked and filled with sadness, sympathy and empathy. In my view, this reveals that love for one's country involves an unreflective level of affectionate, social attachment – a ‘subjective condition’ or orientation (to speak with Kant) – that, of course, is very much accidental. We happen to be born in different countries, with different cultures, and so different ways of being socially and affectively directed or identified in the world. Yet these differences are constitutive of who we are. We develop particular, affectionate attachments to the countries we stand in such an identifying relation to: our countries. Hence, when our own country is attacked, our implicit affectionate love for that country is revealed to us through our immediate, spontaneous grief.Footnote ⁴ Moreover, I believe that sometimes, though of course not always, it is people's love for their countries – this aspect of their patriotism – that pushes them to try to make their states better, despite their states’ difficult pasts or currently anarchic or deeply unjust institutions or social movements. Out of love, they refuse to let the past or the present determine what the country amounts to, and rightly so.

This non-moralized aspect of patriotism allows that I can be motivated to risk my life to save my unjust country because I love it, because it is mine and my fellow citizens’ (including ours to fix) – not anybody else's, even if we might need external help from friends at times. Of course, we can also be motivated to risk our lives for justice, to make our states more just, for instance, by fighting current racist or sexist laws, policies and attitudes. So in response to Williams, I argue that for Kant the determination ‘to save’ one's country can, at times, involve only one thought or motivation. In the first case I mentioned the thought or motivation is ‘because I love it’, and in the second it is ‘because it is right’ (one's duty). Analysing patriotism as including this unreflective, affective aspect, which in itself is not inherently morally problematic or in conflict with morality, also seems in keeping with relevant remarks Kant makes about love, or what he calls ‘self-love’. For example, Kant says in the second Critique that ‘Pure practical reason merely infringes upon self-love, inasmuch as it only restricts it, as natural and active in us even prior to the moral law, to the condition of agreement with this law, and then it is called rational self-love’ (5: 73). The full Kantian account of patriotism therefore necessarily includes both elements: the prior, affective, natural or social love (what Kant simply calls ‘self-love’), as well as practical or moral self-love (what Kant calls ‘rational self-love’). Moreover, affective love for one's country is, as such (so most of the time for socially healthy people), morally unproblematic and proper; often it does not even need to be restricted. After all, a healthy love for one's country presupposes that the country loved is a country of social, affectionate persons living together. Viewing the natural or social, affective elements of self-love (and, so, also of patriotism) as in themselves morally unproblematic, as prior and necessary to, and as part of, a full account of self-love (and, so, also of patriotism), does not imply that loving oneself (or one's own country) cannot be difficult, impossible, and sometimes wrong. An account of how to deal with those difficult, impossible, or wrong loves, I believe, will require moral considerations – and explaining this is part of a full account of patriotism. For example, patriotic love may become difficult, impossible or wrong when only a smaller fraction of citizens actually have a healthy love for their country (a love affirmative of personhood), rather than a pathological one (a love marked by self-conceit or arrogance towards the other citizens or non-citizens). Here, I believe Kant argues, morality comes in and corrects this corruption of our affectionate, natural or social love for our country by striking down self-conceit (5: 73).

At this point, one might respond that a Bernard Williams sense of affectionate love of one's country is what Kleingeld understands by ‘cultural’ or ‘ethnic’ nationalism (which she is wary of), rather than ‘proper’ moral patriotism, which refers only to one's particular allegiance in the sense of how one's (perfect and imperfect) ethical duties (practical love) track enforceable political obligations towards one's country. But this response is unsatisfactory to me for at least three reasons: (a) it invokes a rather unintuitive, not common-sensical, technical use of the word patriotism (and not what I believe must be Kant's own); (b) it can at most capture the ethical-freedom or moral component of a full account of patriotism, while either missing or paying insufficient positive attention to the more affective or non-moralized love component (the moral anthropology part); and (c) it cannot, as we saw above, answer Williams's ‘one thought too many’ objection when applied to the issue of patriotism. Hence, the full Kantian account of patriotism, I believe, necessarily includes all elements – ethical duties, duties of right and social or affective love – and explains how the three operate together when one has a healthy, morally sound love for one's country.

In addition, in my view, an account of patriotism is, strictly speaking, not necessary to reject Simmons's famous objection to Kant, that is, the objection that Kant cannot explain why we are politically obliged to any particular state.Footnote ⁵ Kant's account of political obligations alone is sufficient to reject Simmons's objection, because it shows why we have special, enforceable obligations to our particular states (so neither an account of patriotism nor ethical obligations are needed for this). To reject Simmons, all we need is a non-voluntarist account of political obligations, and I believe all republican interpretations of Kant that provide ideal reasons for the necessity of the establishment of the state have this. Moreover, in my view, on Kant's account, the ethical obligations tracking our political obligations – including the duties to support the establishment of the legal-political institutions constitutive of the just state – are all perfect duties. That they are perfect duties is one reason why Kant concludes his discussion of private right in the Doctrine of Right by saying that choosing not to establish the state is not necessarily to wrong anyone else, but it is always to do ‘wrong in the highest degree’ (6: 307–8). Again, this is not to deny that we need a Kantian account of patriotism; it is to say, instead, that a Kantian account of patriotism seems only needed to respond properly to a Williams love-type objection to Kant, not a Simmons political obligations-type objection.

Let me now turn away from the issue of patriotism to the topic of poverty and global justice – in particular, to Kleingeld's idea of a world republic with coercive powers, starting with her discussion of global poverty relief. As republican interpreters of Kant, Kleingeld and I agree that it is only possible to arrive at the issue of global poverty relief after first addressing domestic poverty relief. Kleingeld's interpretation of poverty relief, however, rejects any reading that attributes to Kant the view that the minimally just state will guarantee unconditional poverty relief for all its citizens. Since Kleingeld specifically addresses this criticism to my interpretation of Kant, it will be useful first to clarify my position. On my interpretation of Kant's poverty relief argument, the state must secure unconditional poverty relief because this is the only way the state can reconcile its monopoly on coercion with the innate right to freedom of each of its citizens. The citizens’ innate right to freedom, Kant clarifies in the Doctrine of Right, is their right to ‘independence from being constrained by another's choice … insofar as it can coexist with the freedom of every other in accordance with a universal law’ (6: 237). I propose that Kant's argument for unconditional poverty relief is conceptual: without unconditional poverty relief, destitute citizens will find themselves without any legal access to means that are not subject to non-destitute citizens’ private choices, namely, the non-destitute citizens’ choices to employ or provide charity for the destitute citizens. And this is a situation the state cannot permit without failing in its very basic task to ensure that the total coercive framework of laws is consistent with each citizen's right to freedom. The state must represent each citizen, which means that it must make sure that the total legal-political institutional framework rules out the possibility of any one citizen's exercise of freedom being subject to another citizen's private choices. Each person's innate right to freedom gives each citizen a right to independence from being so subjected to others, while being subject, instead, only to the laws of the state. This is why Kant insists that the guarantee regarding poverty relief is part of ‘the right of the state against the people’, and why it must be secured ‘by public taxation, not merely by voluntary contributions’ (6: 236).

Kleingeld is likely to respond to my interpretation in one of two ways. First, she may reiterate that my position is ‘extreme’ because it treats the right to poverty relief as unconditional, no matter the cause (141–2). I believe she would be wrong to make this objection. The right to poverty relief must be unconditional because it cannot matter why I ended up with nothing – whether it is because of accident, stupidity, the functioning of market forces, sickness or otherwise. The state must always relate to me as a citizen, as someone with an innate right to freedom – a right that cannot be alienated by my (immoral, stupid or the like) choices. Hence, from the perspective of the state, all that matters in such a situation is that I have nothing, and so no legal access to means that are not subject to another citizen's private choice (her or his consent). As a citizen, I have an irrevocable or inalienable right (a public right claim against the state) to be dependent only on the laws of the state and not to find the possibility of the exercise of my freedom subjected to other citizens’ private, arbitrary choices. The state cannot maintain a monopoly on coercion that is irreconcilable with my innate right to freedom (regardless of what I think or do), so it must guarantee the provision of unconditional poverty relief as a matter of public right. Obviously, poverty relief, for Kant, only gives me a right to exist somewhere and to sufficient means for survival – from there I have to work myself out of my bad condition.Footnote ⁶ Although extremely minimal, that right, I contend, must be unconditional given Kant's basic principles of justice. Even if Kleingeld and I disagree on this point as a matter of interpretation, I would insist that mine is the better interpretation of Kant's own position, since it has the virtue of being a conceptual argument that follows from Kant's ideas of the innate right to freedom and of the public authority as representing all citizens, which it does in part by institutionally securing legal access to subsistence means once it establishes a monopoly on coercion.

The second way Kleingeld might respond to my interpretation involves further elaborating on her claim that ‘the state does not have the duty to relieve the poverty of those who could relieve it themselves’ (141). So, if there are employment opportunities, then the poor do not have a right to poverty relief since they could relieve their problems themselves. But, in my view, this must be either incorrect or seen as entailing that the poverty relief must be unconditional. After all, it is not true that I can relieve my poverty myself if my only legal access to means goes through consensual access to someone else's means, namely, by that person consenting to give me such access (say, by employing me or giving me charity). The limits of my legal choices (what I can do for myself) are my rightful means; when I have nothing, I do not have means and hence I cannot myself relieve myself of my problems. Therefore, we see again that the guarantee of poverty relief must be unconditional, since only in this way does the state ensure that I am never so subjected to others’ choices and am instead subject only to law: I have a place to go (itself regulated by public right) where I can exist and where I will receive subsistence means. This is the guaranteed base from which I can work myself into a better position, including by applying for jobs. It follows that, once I have a job and means again, I can no longer claim subsistence means, but if I quit my job (for whatever reason), I still have somewhere to go once I have spent all my resources. The state must guarantee that I can exist somewhere safely when I have nothing: a safe place secured by its laws, from which I can start working myself out of my destitute situation once again.Footnote ⁷

Kleingeld uses her account of domestic poverty relief to justify global poverty relief, but I'm sceptical of this argument also. To see this, first note that according to Kleingeld there are two reasons why the republic must provide poverty relief: because the state must ‘maintain its citizens’ in order to maintain itself (140); and because poverty relief is connected to a ‘deeper’ justification for the state's right to levy taxes, namely, how ‘the state, through its legislation, is the condition of the possibility of there being lawful individual property and trade at all’ – reasoning that does not apply in the state of nature (141–2). Second, note that Kant emphasizes that the sovereign cannot possess land (cannot own private property) because then ‘he would make himself into a private person’ (6: 323), in which case the sovereign could not provide an ideal solution for the ideal problems of rightful interaction in the state of nature. The sovereign must be a public authority or person, not a private authority or person, and part of being a public person is not having private interests (including private property). The state does not act for itself, but acts only on behalf of its citizens – it represents them. The state is simply the means (the public person) through which private persons act in order to secure and enable rightful interactions, including the enabling of conclusively rightful, domestic private property relations. This also means, however, that in an important sense states are neither rich nor poor. States can have internally unjust economic policies, small tax bases, badly organized economies, be adversely affected by the world economy, and so on, but they cannot be rich or poor in the sense that private persons are rich or poor, since states do not have private property.

The disanalogy between private and public persons entails, I believe, that even if we set aside specific disagreements regarding the poverty relief argument at the domestic level, such an argument does not (for Kant) and cannot (for a Kantian) apply at the global level. Hence, I believe Kleingeld is wrong when she says

Yet it seems to me that neither of Kleingeld's two domestic reasons for poverty relief mentioned above (‘maintenance’ and ‘the deeper’ reason for taxation of private property) applies at the global level. The aim at the global level is to enable rightful international and cosmopolitan relations – relations between states and between states and aliens – which is an argument that in its ideal form presupposes internally just states. Yet internally just states do not (ideally, in principle) need to be guaranteed poverty relief by the world republic. In addition, of course, states don't have private property, so the deeper reason does not apply at the global level; the global state is not a precondition of the possibility of rightful private property relations at the global level (as the state is at the domestic level). Of course, as states choose to become connected through interacting economies, the situation may change, but such changes are different from the poverty argument in the domestic case. To put the point differently, global poverty relief is not necessary to establish just relations between internally just states. The explanation for states’ economic assistance to each other through the global public authority will come from non-ideal, prudential or conditional (including prudential and conditional systemic reasons) rather than ideal reasons that issue from the initial establishment of the global public authority.Footnote ⁸

My final worry concerning Kleingeld's justification of global poverty relief relates to a more general concern about her claim that according to Kant we should aim to establish a world republic with permanent, coercive powers. I do not believe Kant did, or the Kantian should, defend a world republic with coercive powers beyond the UN Peace Corps Volunteers, which is justified only through non-ideal reasoning and the volunteers’ actual consent. Kant, I think, is committed to the view that ultimately the choice of enforcing international and cosmopolitan law lies with the particular citizens, though it should be exercised through their states and, if possible, through the global public authority. This view also follows from the position that the state does not have private interests (unless it is corrupt). Since the state does not have private interests (ideally), Kant's ideal arguments concerning so-called assurance and indeterminacy, which justify the public authority with a permanent monopoly on coercion in the domestic case, do not yield a corresponding argument for establishing a public authority with coercive powers in the global case. Let me elaborate on this point.

First, I will address why Kant's assurance argument doesn't seem to apply in the global sphere. Within the republican interpretative tradition, at the domestic level, the state serves as the means through which we provide assurance that we will respect each other's private property. Private property concerns possession of objects of choice that are distinct from us. Assurance by means of the state's establishment of its monopoly on coercion – the establishment of sufficient force that exists under our shared (public) control – is how we make it the case that our respect for each other's private property (which remains in effect even when we are not physically holding it) is not subject to anyone's choices (say, to keep one's promises). Since states do not have private property, their territory should not be understood as analogous to individuals’ private property. Consequently, the state's territory should be understood instead on the analogue of one's body – the extent of a state's territory is the extent of the shared, legal public person in a way comparable to how my body is the extent of my legal private person. Hence, there is never an ideal need for assurance in interstate relations since the relationship between the legal person of the state and its territory is analytic; the territory of the state is never distinct from the person of the state in the way that private property is distinct from the individual (private) person. So, whereas the relation between a state and its territory is analytic, the relation between an individual and her private property is synthetic. As a result, there is no need (as a matter of ideal theory) for assurance in interstate relations, just as persons interacting do not need assurance if we assume away all questions concerning their acquired, private right to property. Individuals’ persons and their force are analytically united in their bodies, just like states and their force are analytically united too. We might still need protection against evil attacks, but those would be non-ideal reasons for security, rather than the ideal reasons for assurance that Kant provides in the domestic case.Footnote ⁹

Next, and lastly, I will focus on why the so-called indeterminacy arguments do not apply in the same way in the global and domestic spheres. Here, too, I contend the analogy between the domestic and the global cases breaks down because the state doesn't have private property. Even reasonable border disputes are ultimately disputes between some particular citizen of state A and some particular citizen of state B.Footnote ¹⁰ These disputes do call for an international court of justice to find a rightful resolution, since neither state A nor state B is a rightful judge of the dispute. Yet since the purpose of the states is to enable rightful conditions for its citizens, and since it cannot be the legitimate judge of such disagreements, it will have no ideal reason not to enforce an international court's verdict with regard to its own citizens. Moreover, if it turns out that a state wrongly refuses to enforce a verdict, only war rectifies this. But such a decision to go to war cannot, according to Kant, lie with any state's leaders. Rather, it must rest with the particular citizens of that state, since the state does not own its citizens and cannot make decisions about their life and death on the citizens’ behalf. Finally, because no one can be forced to risk her life because of some others’ disagreement, citizens of countries C, D, E, F, etc. cannot be required to take part in enforcing such a verdict between countries A and B. Therefore, not only do we lack any ideal reasons to establish coercive powers at the global level, but we have ideal reasons against it: we cannot establish a permanent monopoly on coercion in the global sphere because doing so would be inconsistent with each person's basic right to freedom – a freedom that protects people against being used as mere means for someone else, including as mere means to win someone else's battles (i.e. the people's consent is required). Hence, the world republic must remain a voluntary enterprise for its member states, and it cannot establish a permanent monopoly on coercion.