2. Despotism and republicanism

Kant’s most cited definition of despotism appears in Toward Perpetual Peace, where he requires that ‘the civil constitution in every state shall be republican’ (TPP, 8: 349) and then makes a distinction between the republican and democratic constitution. While democracy, as a form of sovereignty of a state, is distinguished from monarchy and aristocracy ‘according to the different persons who have supreme power within a state’, republicanism, as a form of government, is opposed to despotism ‘according to the way a people is governed by its head of state’ (TPP, 8: 352). As Kant states, ‘Republicanism is the political principle of separation of the executive power (the government) from the legislative power; despotism is that of the high-handed management of the state by laws the regent has himself given, inasmuch as he handles the public will as his private will’ (TPP, 8: 352). A state in any form of sovereignty can be either despotic or republican; therefore, there are six forms of state based on the combination of these two categories: republican monarchy, despotic monarchy, republican aristocracy, despotic aristocracy, republican democracy, and despotic democracy. Surprisingly, Kant claims that

This critique of democracy has naturally sparked controversy. Wolfgang Kersting, for example, claims that ‘Kant has ruled out the possibility of a republican-led democracy’ in Toward Perpetual Peace (Kersting Reference Kersting2007: 326). However, some interpreters believe that Kant criticises merely direct democracy without a representative system (Byrd and Hruschka Reference Byrd and Hruschka2009: 179; Maliks Reference Maliks2014: 103).

There is only one organisational principle of a republican constitution in Toward Perpetual Peace, namely, the principle of separation of powers. Kant does refer elsewhere to a representative system in which delegates exercise legislative power for the people in parliament, and, immediately following the critique of democracy quoted above, he refers to a nonrepresentative form of government as well:

However, here, the representatives are not those in legislation, but rather another person outside the sovereign that represents the sovereign in executing laws. A representative system is, thus, not a matter of elected delegates in parliament but concerns only the relationship between the legislative and executive powers. Kant does not propose a new principle in addition to the principle of separation of powers. Therefore, ‘democracy in the strict sense of the word’ refers only to a democracy without a system of separation of powers. A government without this system is despotic because the legislator and the executor are the same person, and even if there are publicly declared laws, they can be modified or abolished at any time by the executor, who is also the legislator. Special and arbitrary private will, therefore, replaces laws with formal certainty. Kant believes that it is most difficult for a democracy to republicanise its constitution precisely because, the number of sovereigns being so large, they can directly govern society simply by exercising their power. In contrast, in monarchies and aristocracies, no matter how much the sovereign wants to take overall control, because the number of sovereigns is limited he or she must rely on others outside the sovereign to govern society. The fewer the sovereigns, the more representatives are needed; therefore, more public rules need to be promulgated to regulate and guide representatives to execute the will of the sovereign. Even if this will is only the private will of the sovereign at the beginning, it obtains a certain formal certainty and is thereby transformed into a public will, making further reforms possible. Specifically, Kant claims the following:

Certainly, we can argue that precisely because the number of sovereigns is too large in a democracy, not all affairs can be resolved by a general assembly. Therefore, it is necessary to delegate executive power to some people. In this sense, executive power must be separated from legislative power. However, for Kant, the key issue is whether there are different persons in such separation, that is, does the executor exercise power as an independent person or merely as part of the legislature? Only in the former case are there representatives because representative relations exist only between different persons. Such ‘a democratic constitution in a representative system’ (V-ZeF, 23: 166; my translation) is certainly not a despotism. In the latter case, the person of the executor is still confused with that of the legislator. Just as a committee in Congress directly exercises administrative power, in this case, the formal certainty of laws is under constant threat; this is despotism in Kant’s view. Therefore, Kant only criticises democracy without a system of separation of powers, a specific and historical form of democracy. He calls this ‘democracy in the strict sense of the word’ because it is the original and ancient form of democracy. As Kant states, ‘None of the ancient republics, so called, knew this system, and because of this they simply had to disintegrate into despotism’ (TPP, 8: 353). During Kant’s era, a system of separation of powers was still a new concept in practice, and the transition from despotism to republicanism was still a problem of the times. Kant believes that compared with the transition from a despotic democracy to a republican democracy, it is more likely for a despotic monarchy to adopt the principle of separation of powers by gradual reforms and thereby transform itself into a republican monarchy, providing a new basis for further reforms.

Luigi Caranti has recently proposed a new interpretation according to which Kant’s critique of democracy includes all forms of democracy. This is because in a democracy, whether direct or representative, people are allowed to pursue private or partisan interests in legislation, which is not allowed in monarchies and aristocracies. Therefore, there is a significant structural defect of democracy that will cause its legislation or decision-making to deviate from the general will. However, it is untenable to believe that decision-making in monarchies and aristocracies is closer to the general will than in democracies simply because ‘while in the case of non-barbaric monarchies or oligarchies the ruling class must explain how decisions are inspired by the general will, in a democracy there is no obligation of this sort’ (Caranti Reference Caranti2023: 175). Legislative proposals in democracies must appeal to public discourse to win majority support. Claiming to represent the general will and truly representing the general will are two different things. In this regard, monarchies and aristocracies are not necessarily better than democracies; often they are worse. Caranti also acknowledges this and believes it is why Kant is more supportive of democracy in the Doctrine of Right. In this later text, Kant no longer mentions the despotic danger of democracy. Instead, he claims that ‘with regard to right itself’, autocracy is ‘the most dangerous for a people, in view of how conducive it is to despotism’ (MM, 6: 339). Nonetheless, Caranti still opposes the suggestion that Kant changes his attitude towards democracy in the Doctrine of Right. He believes that Kant only reveals the structural defect of democracy ‘from a purely theoretical point of view’ in Toward Perpetual Peace, while in the Doctrine of Right, he focuses mainly on ‘the concrete risk of a degenerating potential’ of different forms of state based on a series of empirical considerations (Caranti Reference Caranti2023: 180). This interpretation is also untenable in addition to being contrary to the more usual impression of these two works of Kant. While Toward Perpetual Peace is a work that contains many empirical considerations, the Doctrine of Right, as a part of the Metaphysics of Morals, should be a purely theoretical text that excludes all empirical considerations.

I propose a new interpretation here to defend Kant’s consistency between these two texts. In my view, the mistake of all previous interpretations with respect to the question of consistency lies in confusion between the public will and the general will in Kant’s definition of despotism. When Kant distinguishes between despotism and republicanism based on the principle of separation of powers in Toward Perpetual Peace, he warns that the private will of the executor would replace the public will if the legislator and the executor were the same person. The public will should be understood here as the legislative will embodied in public laws with formal certainty, regardless of whether the content of this will is particular or general. A constitution without a system of separation of powers is despotic simply because there are no public laws with formal certainty; therefore, ‘any form of government which is not representative is, strictly speaking, without form [eine Unform]’ (TPP, 8: 352). Kant does mention the general will when explaining why democracy in the strict sense of the word is necessarily despotic; however, this has nothing to do with the content of the will. This kind of democracy is despotic because there are no public laws with formal certainty; thus, the executive power that directly addresses the freedom and rights of individuals in particular cases is based on private, not public, will. Here, every decision made by the sovereign against specific individuals, regardless of whether its content is beneficial or harmful to the latter and regardless of whether the latter agrees or disagrees, is based on private will, which is neither public will nor general will. The general will of legislation must be public will, but the public will is not necessarily general will, which must be not only public in form but also general in content.

The principle of separation of powers is, therefore, only the formal standard of a republican constitution. Based on the distinction between public and general will, we can further distinguish between two levels of republicanism and two levels of despotism. A constitution without a system of separation of powers is necessarily despotic because there are no public laws with formal certainty. However, a constitution with this system only meets the formal standard of a republican constitution. A higher and more substantial standard of republican constitution requires not only the formal existence of legislative public will but also that the legislation conforms to the general will in content. As Kant states in Toward Perpetual Peace:

This is a higher standard of a republican constitution, which further demands that all rightful legislation be based on the idea of the original contract. The content of legislation should be something upon which everyone can rationally agree; in other words, legislation should conform to the general will. According to this standard, a constitution with a system of separation of powers can still be despotic if public laws merely reflect private and particular will. From this, we can derive two levels of despotism: despotism in the sense of the absence of public will and despotism in the sense of the absence of general will, both of which are characterised by the domination of political power by private will.

In the Doctrine of Right, Kant sets forth the principle of separation of powers in §45, which naturally constitutes a premise for his later discussion of the forms of state in §51. This discussion concerns only the legislature and the content of legislation. As Kant states, in an autocracy ‘only one is legislator’, while in an aristocracy, it is the nobility ‘as legislator’ (MM, 6: 339). Here, it is unnecessary to discuss despotism in the sense of the absence of public will. What needs to be discussed is the relationship between legislation and the general will in various forms. In this regard, Kant tends to believe that the smaller the proportion of legislators in the whole population, the greater the danger that legislation will deviate from the general will and infringe on individual freedom and rights in the form of public laws. In this sense, autocracy is ‘the most dangerous for a people, in view of how conducive it is to despotism’. The despotism discussed here is not the absence of public will in form but the absence of general will in the content of legislation. Under the premise of separation of powers, democracy is the form of state with the least danger of despotism and is most in line with the idea of the original contract. ‘Any true republic is and can only be a system representing the people, in order to protect its rights in its name, by all the citizens united and acting through their delegates (deputies)’ (MM, 6: 341). This does not contradict the assertion that, among the forms of state without a system of separation of powers, the despotic danger is highest in democracies. In Toward Perpetual Peace, Kant attributes the reason why a republican state is least likely to start a war to the fact that ‘the consent of the citizens of a state is required in order to decide whether there shall be war or not’ (TPP, 8: 350). Thus, we should realise that Kant’s attitude towards democracy does not change between these two texts; he merely discusses the relationship between democracy and despotism at different conceptual levels according to the difference in context.

Certainly, we can also argue that Kant changes his terminology without changing his ideas. This is evident in Anthropology from a Pragmatic Point of View, where Kant defines four conditions in terms of freedom, law, and force: ‘A. Law and freedom without force (anarchy). B. Law and force without freedom (despotism). C. Force without freedom and law (barbarism). D. Force with freedom and law (republic)’ (Anth, 7: 330). Despotism, in the sense of the absence of public will, is redefined here as barbarism; this concept is thus limited to the absence of general will in public laws. Therefore, we can assume that Kant only discusses despotism at the second level in the Doctrine of Right because he sets the principle of separation of powers as a premise, or we can assume that he had already adopted the distinction between barbarism and despotism at that time. This is only a terminological issue; Kant does not change his ideas. Although in a footnote on the republican constitution in Toward Perpetual Peace, Kant also defines freedom as ‘the warrant to obey no other external laws than those to which I could have given my consent’ (TPP, 8: 350), freedom in this sense is only derived in his system. The starting point of his philosophy of right is such individual freedom: ‘Freedom (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, is the only original right belonging to every man by virtue of his humanity’ (MM, 6: 237). The concept of freedom through which Kant distinguishes between despotism and republicanism in Anthropology from a Pragmatic Point of View must be understood only in this way; otherwise, it is difficult to imagine that there is also freedom in anarchy. This kind of freedom precedes any political community and all actual legislative procedures and limits the possible content of the general will. The external laws to which everyone could have given his or her consent must conform to the original right. Therefore, contrary to Maus’ claim that ‘Kant’s highest priority is the structure of democratic self-organisation within which state tasks can first be decided upon’ (Maus Reference Maus2006: 472), Kant is essentially a liberal. Democratic decisions are not necessarily consistent with the general will; they may also infringe on individual freedom. Therefore, both barbarism and despotism express the infringement of individual freedom by the power dominated by private will. When Kant asserts in On the Common Saying and Toward Perpetual Peace that a world state is in greater danger of leading to ‘the most fearful despotism’ or ‘soulless despotism’, he has certainly not yet distinguished between barbarism and despotism, so the concept of despotism still contains two levels. From a formal perspective, a state in which there are no public laws or laws that have no formal certainty is the most despotic. From a substantive perspective, a state where political power most seriously infringes on individual freedom is the most despotic. In practice, these two are often the same: a state without public laws with formal certainty is often a state where individual freedom is most seriously infringed upon.

3. Conditions of the possibility of political reform

If we understand despotism in the sense of the infringement of individual freedom by the private will of the ruler, it can be said that this despotic danger exists in any form of state because it is rooted in human nature. As Kant states in Idea for a Universal History:

The legal order of coexistence requires an irresistible supreme power, which, however, can be held only in the hands of human beings as animals. Not only is there an inclination to misuse this power in human nature but also the ‘possession of power unavoidably corrupts the free judgement of reason’ (TPP, 8: 369); therefore, this despotic danger exists in any form of state, be it an individual state or a single world state, monarchy, or democracy. How can this despotic danger be eliminated, and how can the principles of right be realised in turn? For Kant, this problem is ‘the most difficult of all’ (IUH, 8: 23). Nonetheless, he still believes that even an individual monarchy can be expected to republicanise its constitution and continuously realise the principle of right through gradual reforms. This is also why we can expect that the human race is constantly progressing. Where does his confidence come from?

In Idea for a Universal History, Kant argues that three conditions are necessary to establish a perfect constitution: ‘correct concepts of the nature of a possible constitution, great experience practiced through many courses of life, and beyond this a good will that is prepared to accept it’ (IUH, 8: 23). While the principles of a republican constitution can be derived a priori from practical reason, the latter two conditions are beyond the scope of the metaphysics of right. They express two indispensable elements of political practice: the political judgement that can integrate the principles of right with specific empirical situations and the political will to realise these principles in experience. The disagreement between politics and morals (more precisely, the principles of right) discussed in Toward Perpetual Peace largely stems from the absence of these two conditions. Due to his distrust of individual morality in politics, Kant would certainly not, as Caranti claims, appeal to ‘the morality of individual politicians’ (Caranti Reference Caranti2017: 242). Nor does he place his hopes merely on the establishment of a republican constitution, as this should be a result, rather than an initial condition, of political practice.

In Toward Perpetual Peace, Kant proposes a solution to the disagreement between politics and morals, that is, the principle of publicity. He suggests that ‘every claim to a right must have the capacity for publicity’ and claims that ‘it can yield a criterion to be found a priori in reason’. (TPP, 8: 381) However, it should be noted that publicity is not an a priori principle in the sense that it must ‘abstract from the empirical’ (Caranti Reference Caranti2023: 117). It is also not a principle for politicians to judge the justice of laws or policies in their self-reflection. As Allen Wood points out, the a priori here cannot be understood in a metaphysical sense; rather, it is in the sense ‘in which through experience a man might know a priori that if he undermines his house, then it will collapse’ (Wood Reference Wood, Formosa, Goldman and Patrone2014: 78). Kant believes that if an unjust law is publicly declared, it ‘can derive this necessary and universal, hence a priori foreseeable, resistance of everyone to me [the ruler] only from the injustice with which it threatens everyone’ (TPP, 8: 381). In The Conflict of the Faculties, he also claims that ‘such a public declaration would rouse all of his subjects against him’ (CF, 7: 86–7). However, a ruler can foresee the possible reaction of the public to the declaration of a law or policy through experience only if the public has already expressed its general support or opposition on similar issues. The principle of publicity, to reconcile the disagreement between politics and morals, requires that the public not only have the ability to make public use of its own reason to judge the justice of laws and policies but also the courage to publicly oppose unjust laws and policies, which means that the public must be politically enlightened. The principle of publicity, thus, presupposes the general enlightenment of the people. As Kant states, ‘This enlightenment, however, and with it also a certain participation in the good by the heart of the enlightened human being who understands the good perfectly, must ascend bit by bit up to the thrones and have its influence even on their principles of government’ (IUH, 8: 28).

This general enlightenment can, in turn, be achieved only in the public sphere. As Kant states in What is Enlightenment?, it is difficult for single individuals to break free from the care of others and independently make use of their own reason, ‘but that a public should enlighten itself is more possible; indeed this is almost inevitable, if only it is left its freedom’. Therefore, what is needed for enlightenment is only the ‘freedom to make public use one’s reason in all matters’ (WIE, 0: 36). On the one hand, people can continuously enlighten themselves through the public use of their own reason and thereby reach rational consensus in the public sphere. On the other hand, the increasingly unified public opposition in political affairs can also force the ruler to promote political reform. The public sphere is, therefore, the formative domain of political judgement and political will at the same time. For this reason, Kant repeatedly emphasises that freedom of speech is ‘the sole palladium of the people’s rights (TP, 8: 304) and the ‘single gem remaining to us in the midst of all the burdens of civil life, through which alone we can devise means of overcoming all the evils of our condition’ (WOT, 0: 144). Therefore, ‘the prohibition of publicity impedes the progress of a people toward improvement, even in that which applies to the least of its claims, namely, its simple, natural right’ (CF, 7: 89). Adjectives such as ‘sole’, ‘single’, and ‘alone’ fully demonstrate that Kant places his hope of resisting the despotic danger of political power in the formation of a rational public sphere, and he believes that this formation and the gradual enlightenment of the people are inevitable, as long as there is a certain degree of freedom of speech. This general enlightenment, which makes continuous political reform possible, does not rely on the progress of individual morality but is ‘partly out of love of honour, partly out of well-understood self-interest’ (CF, 7: 92). There is an antagonistic propensity in human nature, that is, ‘the unsociable sociability of human beings’ (IUH, 8: 20), which drives people to pursue their interests in society and interact with the laws and policies that directly affect their interests. As a result, an increasing number of people will actively or passively participate in the public discourse and thus have a better understanding of the relationship between their interests and the state: only by participating in shaping the state can one better defend his or her interests. Well-understood self-interest is, thus, the most reliable driving force for people to participate in public affairs and thereby continuously promote political reform in the public sphere.

Byrd and Hruschka claim that in the Doctrine of Right, ‘Kant abandons his principle of publicity’ (Byrd and Hruschka Reference Byrd and Hruschka2008: 602). This is a misconception of this principle. Publicity is not a principle of the metaphysics of right but rather a principle of political practice aimed at reconciling the disagreement between politics and morals in practice. Nonetheless, in the Doctrine of Right, Kant still emphasises that people can ‘legally resist the executive authority and its representatives (the minister) by means of its representatives (in parliament)’ (MM, 6: 322). This is a typical scenario of democratic politics in the contemporary sense, where people supervise executive power through their representatives in parliament. However, this system is insufficient to curb abuses of power. What if the representatives do not properly supervise the executive power or even abuse the legislative power against the general will? Kant recognises this possibility and claims that if such a situation persists, ‘this would be a sure sign that the people is corrupt’ (MM, 6: 322). Kant does not give significant consideration to party politics, likely because, in his view, regardless of whether it involves competitive elections or party-based governance, political power cannot be effectively restrained without active public participation and individual self-enlightenment. If the people are already corrupt, party politics may devolve into what he terms ‘furtive politics’ (TPP, 8: 386). Therefore, the hope of curbing the abuse of political power ultimately lies with the people themselves. If the people have lost their courage and ability to publicly express their opinions and opposition, and to supervise their representatives in properly performing duties, even if the state has formally established a republican constitution, it is still in danger of degenerating back into despotism. In the contemporary context, we can certainly regard the public sphere as a part of democratic culture in a broad sense. However, for Kant, the most important mechanism for preventing despotism is not the democratic procedure itself but an enlightened public sphere that makes well-ordered democracy possible and can even effectively criticise political authority within other forms of state. Thus, the enlightened public sphere is not a complement to well-ordered democracy but its foundation. In this sense, exploring the differences between individual states and a world state regarding despotic danger in Kant’s context requires focusing on the critical function of the public sphere.

4. The defects of a world state

Kant appears to have less confidence in the possibility of political reform and progress in a world state than within individual states. Therefore, he reiterates the despotic danger of a world state and rejects this model of global peace. Why can the conditions that promote continuous political progress within individual states not also function in a world state? The challenge for a world state is not that it is too large to maintain effective governance but rather that it is more difficult for individuals to maintain enough enthusiasm to participate in the public discourse of state affairs and thereby to politically enlighten themselves, let alone the difficulty in reaching consensus in the public sphere of a world state. This is because general state affairs are too remote from their personal lives in two senses. On the one hand, in a world state, individuals are so far away from political power that many do not pay much attention to what decisions political power makes; on the other hand, in a world state, political decisions or legislation on global scales are highly abstract or principled such that it is difficult for individuals to perceive the direct impact of such decisions or legislation on their interests.

As is shown by the fact that Kant regards a paternalistic government founded on the principle of happiness as ‘the greatest despotism thinkable’, it is not the specific policies that directly affect individual interests that Kant is concerned with but rather the basic principles that fundamentally determine the logic of the system. If the basic principles of the system have been quietly altered by ‘furtive politics’, it will be very difficult for individuals in particular cases to confront the systemic injustices of the whole system. While the problem of effective governance is largely based on Kant’s limited historical experience and can be solved by the development of technologies, such development may not have the same effect on solving the problem of general participation in a global public sphere. Individuals are usually more passive than the government or rulers, who have sufficient motivation and resources to actively use new technologies to strengthen their capacity to govern or maintain social control. The crux of the matter is not whether there are sufficient technological conditions to support individual participation in a world state but rather whether individuals are enthusiastic and motivated enough to participate in public discourse about laws and policies that are far removed from their personal lives. Kant argues that individuals are motivated to participate in political discourse in the public sphere ‘partly out of love of honour, partly out of well-understood self-interest’. However, it is difficult to expect individuals to maintain sustained attention to affairs that are so far removed from their personal lives that they can hardly discern any direct relevance, whether it is abstract legislation on a global scale or the concrete operation of executive power in distant places. Empirical research also confirms that political participation is significantly negatively related to the size of the community (McDonnell Reference McDonnell2020: 333). The larger a community is, the lower the sense of political efficacy individuals can gain from participating in community affairs, and the weaker their motivation to participate. Therefore, whether it is the love of honour or self-interest, it is difficult to motivate individuals to continue participating in public discourse about state affairs in a world state. As a result, the prevalence of political indifference will exhaust the public spirit and dissolve the critical function of the public sphere. This may be why Kant claims that a world state would destroy ‘the seed of good’; the danger of a soulless despotism lurks precisely here.

Here we can draw on de Tocqueville’s account to shed more light on this issue. Although de Tocqueville has a quite different theoretical attitude than Kant towards enlightenment and the public sphere, he correctly reveals that excessive egoism in modern society will lead to the exhaustion of the public spirit, which constitutes the spiritual basis for the emergence and permanence of despotism:

There is no essential difference between de Tocqueville’s and Kant’s definitions of despotism in the sense that they both understand it as the private will of the ruler infringing on individual freedom. De Tocqueville also claims that public participation is the key to cultivating public spirit and overcoming excessive egoism to counter the danger of despotism in modern society. However, what he values is mainly self-governance at the township level:

Who should care about ‘the destiny of the whole state’? De Tocqueville provides no explicit answers. Due to his vigilance against ‘the tyranny of the majority’, he holds an ambivalent attitude towards the public sphere (Jürgen Reference Jürgen1990: 209).

Kant holds a more consistent position insofar as he regards the critical function of the public sphere as the most important means of resisting the danger of despotism and promoting political progress, as ‘the sole palladium of the people’s rights’. He believes that to make gradual reforms possible, the people must continuously enlighten themselves in political affairs in the public sphere. However, that would be much more difficult at the world level. Even public participation at the local level does not help solve this problem. This is not only because local public participation cannot replace public participation at the state level but also because the sovereign state envisioned by Kant does not leave much room for local self-governance. According to Kant’s concept of sovereignty, the power of local governments only comes from the top-down authorisation of the supreme power (Pogge Reference Pogge1992: 59). As Kant states, ‘Even the constitution cannot contain any article that would make it possible for there to be some authority in a state to resist the supreme commander in case he should violate the law of the constitution, and so to limit him’ (MM, 6: 319). In this sense, public participation of individuals at the local level and their criticism of local governments can hardly inhibit the systematic degradation of a too-large state, nor can it counteract the systematic oppression caused by the whole state rather than by local governments as the enforcers of the supreme will. This, in turn, frustrates individuals’ sense of political efficacy in participating in public discourse on local matters and diminishes their enthusiasm for further participation.

Certainly, we can also envision a global public sphere composed of world citizens in Kant’s context (Bohman Reference Bohman, Bohman and Lutz-Bachmann1997: 181; Cavallar Reference Cavallar2015: 142). However, Kant never optimistically envisions that an empirically possible global public sphere could still retain critical functions sufficiently effective to prevent the misuse and abuse of global power. In Toward Perpetual Peace, he is optimistic about the future of global peace:

Habermas regards this passage as a sign that Kant would admit political criticism of a global public sphere (Jürgen Reference Jürgen, Bohman and Lutz-Bachmann1997: 124). However, a global public sphere in this sense, similar to cosmopolitan right, is merely a supplement to the right of states within a world league of states. It may help to prevent wars among states in a league of states, but it is far from sufficient to prevent the abuse and misuse of supreme power in a world state. To do so, there must be a much stronger global public sphere. In a world state, what is more conceivable is that the general participation and political enlightenment of individuals would be more difficult (if not impossible); therefore, the participants in a global public sphere may be a few elite intellectuals or ideologists. However, for Kant, the critical function of the public sphere relies on the possible general opposition of the people. If the elite intellectuals’ criticism fails to arouse a general suspicion of the people against the ‘legislative renown (Ansehen)’ (WIE, 8: 40; my translation) of the ruler, public criticism could no longer exert enough pressure on the ruler; as a result, the common affairs of a world state would become ‘a furtive politics’. Habermas criticises Kant for failing to foresee the structural transformation of the public sphere, but he still admits that ‘supporting structures are needed to stabilize communication between spatially distant participants, who exchange contributions at the same time on the same themes with equal relevance. In this sense, there is not yet a global public sphere’ (Jürgen Reference Jürgen, Bohman and Lutz-Bachmann1997: 125). A global public sphere that can effectively support a world state is still an ideal and has not yet been validated by historical experience. The defects of a global public sphere are amplified rather than overcome in the contemporary world (Albrow and Glasius Reference Albrow and Glasius2007: 10–11).

Some readers believe that this problem can be expected to be solved by cultivating a cosmopolitan disposition in Kant’s context. Paul Formosa claims that the formation of a world republic requires a cosmopolitan disposition, which ‘will emerge over time through education, social interaction and economic, moral and political progress’ (Formosa Reference Formosa, Formosa, Goldman and Patrone2014: 55). For Kant, the cosmopolitan disposition is in fact a moral quality that makes us take ‘an interest in the best for the world’ (L-P, 9: 499); it should be an end of moral education and religious life. However, Kant never regards it as a condition of a world republic. Rather, he reiterates that the realisation of the principles of right cannot rely on the progress of individual morality; the political enlightenment of a citizen, thus, means only that he or she becomes ‘a good citizen even if not a morally good human being’ (TPP, 8: 366). Cosmopolitan right does require a kind of universal hospitality, which, however, means only ‘the right of a foreigner not to be treated with hostility because he has arrived on the land of another’ (TPP, 8: 357–8); it is therefore only a negative attitude of not being hostile to foreigners. Like de Tocqueville, Kant also realises that it is difficult for individuals to maintain sufficient enthusiasm for affairs that are too far away from their personal lives. In the Lectures on Ethics, he distinguishes three kinds of love for others: cosmopolitanism, patriotism, and love for a particular group. Unsurprisingly, he criticises the narrowness of the last kind, but he also concedes that

At the level of theory, Kant undoubtedly holds a cosmopolitan position that runs through his philosophy (Höffe Reference Höffe2007: 179; Cavallar Reference Cavallar2015: 21). However, he does not support in practice an abstract cosmopolitan position that expects everyone to love all others in the world, because this is hardly possible in experience. Such overly abstract requirements would instead lead to ‘an indifferentism towards the human race which inhibits the dissemination of general human goodwill and prevents any communal participation for everyone’ (L-Eth, 27: 674). Therefore, Kant does not believe that a cosmopolitan disposition that completely negates love for an individual state is feasible in practice; rather, cosmopolitanism must be mediated by patriotism in practice. However, patriotism is different from the duty of political obedience; it is not a rational duty that can be demonstrated in the metaphysics of right but rather a practical attitude that individuals as citizens should adopt in political practice to realise the principles of right. Each individual can realise his or her freedom and rights only if he or she realises that his or her destiny, with respect to right itself, is closely linked to that of the other members of the community and only if he or she actively participates in the affairs of the community to fight for the general freedom and rights of every member of the community. A patriot is, thus, an enlightened citizen, and patriotic politics is the politics of enlightened citizens, which is intrinsic to a patriotic government that is opposed to a paternalistic government established on the principle of happiness. In this sense, those who regard patriotism as a rational duty and attempt to reconcile its conflict with cosmopolitan duty (Kleingeld Reference Kleingeld2003) have confused the levels of theory and practice in Kant’s context. In On the Common Saying, Kant claims that

In the metaphysics of morals, Kant explicitly rejects happiness as a principle of rational legislation. However, a government established on such a principle is ‘the greatest despotism thinkable’ not simply because happiness is an empirical concept that cannot be generalised but because such a government would dissolve the possibility of self-enlightenment and ultimately abrogate ‘all the freedom’ of the people. A world state, insofar as it would lead to ‘an indifferentism toward the human race which inhibits the dissemination of general human goodwill and prevents any communal participation for everyone’, would likewise destroy ‘the seed of good’ and abrogate all the freedom of the people. In this regard, ‘a soulless despotism’ or ‘the most fearful despotism’ of a world state is identical to ‘the greatest despotism thinkable’ of a paternalistic government.

A world state also lacks other conditions for continuous political reform, such as competition and threats from the outside. Kant reiterates that in modern international relations, the pressure brought about by competition between states is conducive to forcing a state to improve itself or at least to internally maintain a certain degree of freedom for the development of industry and commerce, by which the people can further enlighten themselves and promote political reform. In the Conjectural Beginning of Human History, he notes that ‘the danger of war is still today the sole thing that moderates despotism, because wealth is required for a state to be mighty, but without freedom, no enterprise that could produce wealth will take place’ (CB, 8: 120). A similar statement appears earlier in the Idea for a Universal History:

Kant views ‘the spirit of commerce’ (TPP, 8: 368) as an important factor driving human history towards perpetual peace. In modern international relations, competition between states ultimately relies on the development of domestic industry and commerce, as well as on the ‘internal culture’ related to education and enlightenment of the people. Kant takes the position of classical political economy and posits that industry and commerce cannot flourish without domestic freedom. He also believes that, because it ‘cannot coexist with war and sooner or later takes hold of every nation’ (TPP, 8: 368), the spirit of commerce will unite the world as a whole and, to some extent, prevent wars. Kant has thus proclaimed the arrival of an era of world history in which not only would the general communication driven by the spirit of commerce bring all states into a unified global system with identical commercial rules but also external pressure would force all states to reform or at least maintain a certain degree of domestic freedom necessary for the enlightenment of the people and, thereby, gradual reforms. A world state lacks such an external condition.

Therefore, not only ought one not to anticipate a universal monarchy formed by violent annexation, but because it lacks some crucial conditions for preventing despotic danger, even a world state established voluntarily and organised in a republican way is not desirable. The arguments that Kant uses to show that a world state is empirically impossible are not reasons for his rejection of a world state; rather, they express what he sees as the ‘purposiveness in the course of the world’ (TPP, 8: 361): fortunately, there is every indication that a world state that is not desirable in practice is at the same time impossible in experience. Kant may, of course, be wrong in this judgement, which is based on his limited historical experience. However, his real reason for rejecting a world state, namely, that such a state would be in greater danger of despotism, should still be taken seriously today.