2007

The domestic analogy and the Kantian project of Perpetual peace (*)

Chiara Bottici (**)

One of the objections raised against the Kantian tradition of international relations (1) is that of its reliance on an essentially flawed analogy between domestic and international order. In Hedley Bull's words, the "domestic analogy is the argument from the experience of individual men in domestic society to the experience of states, according to which states, like individuals, are capable of orderly social life only if, to use the Hobbesian expression, they stand in awe of a common power" (Bull 1977: 46). According to Bull, we must abandon the domestic analogy altogether, not only because the attempt to understand something by means of something else is a sign of infancy in a subject, but also because the international society is unique and owes its nature to qualities that are peculiar to the situation of sovereign states (Bull 1966a: 45).

This criticism of the domestic analogy is often used as an argument for rejecting any supranational federal arrangement. (2) According to Michael Hardt and Antonio Negri, for instance, not only the Hobbesian hypothesis, which emphasises the contractual process giving rise to a new unitary and transcendental supranational power, but also the so-called Lockean conception, focusing on the counter powers that animate the constitutive process, must be rejected. In both cases, according to this criticism, the necessity of a global power would be drawn from an analogy with the experience of the domestic society, i.e. from the idea that peace can be reached only through the subjection to a common power.

The aim of this paper is to assess the validity of such a critique. Do we have to reject the Kantian argument and the other world order proposals following the same line of reasoning because they rest on a naive analogy - as Bull suggests? To answer this question, I will first try to evaluate the domestic analogy as a form of analogical argumentation, by analysing the conditions of its application. Second, I will apply this analysis to Kantian thinking, arguing that the domestic analogy here plays only a rhetorical and heuristic, but not a justificatory role.

We can identify in western thinking two mainstream meanings of the term "analogy": the first, ranging back to Plato and Aristotle, points to the equality of relationships - like the one existing among the members of a mathematical proportion (x : y) = (k : z) ; the second, which is closer to the common use, points to the extension of knowledge from one situation to another on the basis of the similarities existing between the two situations. (3) Despite their being distinguished, these two meanings are interwoven, insofar as the similarity which supports the extension of knowledge from one situation to another may be conceived of as a similarity of relationships. Other possible meanings of the term might be found, but they all seem to refer to one of the two semantic areas identified above. For instance, one of the four terms of a proportion may be identical with another one - as in what Aristotle calls the "geometrical proportion", where the second term is identical to the third. (4) However, I will deal with all these meanings as special cases of one of the two semantic areas described.

Let us move to the "domestic analogy". In its classical formulation, this type of reasoning consists in a proportion that can be expressed in the following way: "states : international relations = individuals : domestic realm". Here, the unknown factor is the second, whereas all the other terms - as well as the relationship between the last two - are assumed as given. On the basis of a similarity of relationship between the two conceptual pairs, on the one hand, and on the basis of other similarities between the first terms of each pair, on the other, the relationship between the terms of the first pair can also be derived.

Moving one step farther, I will now analyse this way of reasoning by asking what the conditions for its application are. First, since the analogy is always drawn between two similar, but different, situations, whoever argues for the domestic analogy implicitly assumes that the two domains - domestic and international - are distinct. This thought has not yet been sufficiently emphasized and deserves more attention. Up to a point, a heuristic analogy between two domains can be so perfect that the differences between these domains diminish to the point of vanishing. But at the limit of this diminution, one can argue, the analogy itself disappears. It follows that there is a negative moment in any analogy which cannot be removed, without removing the analogy altogether thereby. According to C. Perelman and L. Olbrechts-Tyteca, the use of analogy in law, for instance, is much less widespread than is commonly held. In their view, the use of analogy is limited to the comparison of particular features of positive rights that are distinguished in space or in time. On the contrary, when looking for similarities between systems, these latter are thought of as examples of the general category of law, and, even more, when arguing for the application of a norm to new cases, we are moving within one single field. In fact, inasmuch as we recognise that the new cases fall within the same principle, there is no longer any separation between the two domains of the analogy, and therefore the analogy itself is thereby overcome (Perelman, Olbrechts-Tyteca 1958: 395).

Now, as the terms inter-national and domestic underline, what marks the difference between the two realms is national sovereignty, with its double face, internal and external. Sovereignty itself marks the borderline between the two domains by separating, inwards, the space for the exercise of the sovereign power (which is therefore domestic) and, outwards, the anarchic space of the clash of equally sovereign nation states (inter-national). Beyond this distinction there is, then, the tendency of modern political theory to conceive of the universe of politics in terms of a dichotomous view, separating its internal domain, subject to sovereignty, from the external one, domain of mere anarchy. We can talk of a crucial role played by the sovereignty versus anarchy dichotomy in modern political thinking. (5) By this, I mean that the two spaces, for the exercise of sovereign power, on the one hand, and for the anarchical clash of sovereign nation states, on the other, tend to be conceived of as both mutually exclusive - because what is included in the first cannot be part of the second - and jointly exhaustive of the whole universe of politics - so that what does not belong to the domain of sovereignty is usually assumed to belong to that of anarchy.

The previous remarks lead to the second condition for the application of the domestic analogy, namely, the fact that the knowledge extended to the international domain must derive from the experience of individuals in domestic society. Again, what is extended to the international domain is a knowledge that is drawn from the modern "domestic" experience. Be it the whole experience of the sovereign state - as in world state proposals - or some kind of partial knowledge - as in the idea of a limited social contract among states, in both cases, in order to have a fully-fledged domestic analogy, the idea that the existence of a supreme power is the condition of social order must be derived from the specific modern experience of individuals. Therefore, those arguing according to this line of reasoning, without recognising its historical roots, run the risk of making an implicitly universal paradigm out of a historically situated experience, whereas any analogy drawn from historical experience should always be previously recognised and assessed as such.

This is not to say that any analogical argument relying on the modern experience is essentially flawed - as some critics of the domestic analogy seem to argue (6). This can be held only either by rejecting analogical reasoning altogether or by holistically discarding the modern experience as a whole. In reference to the first strategy, one might object that analogical reasoning, after having been rejected altogether by neopositivist thinkers, is now the object of great debate, suggesting its importance as a heuristic tool and as a means for argumentation. Whereas almost everybody recognises the validity of the analogy as a heuristic tool, its value is hardly held to be more than that of a mere starting point. I argue that analogical reasoning must also have some argumentative value, since it can potentially count in favour of one hypothesis over others. However, I recognise its precarious status, which is due not only to the fact that an overly successful analogy may disappear, as I stressed above, but also to the fact that the force of the analogy may always be denied by pointing to its negative moment, namely, to the difference between the domains of the analogy.

Therefore, an analogy may be too little to uphold a world order proposal, but it is certainly not only the sign of the infancy in the subject - as Bull argues. On the other hand, I am not willing to reject the whole modern experience either, as some postmodern thinkers seem to do. With the previous remark, I want only to point out the historical character of the experience of the modern state, which does not mean taking a position either in favour of or against it.

Now, what can justify the extension - whole or partial - of this experience to the international realm? This question leads us to the third condition for applying the domestic analogy, namely, the existence between the two domains of a sufficient number of similarities - which, in contrast to the differentiating side of the analogy, may be called the positive moment of the analogy. (7) As clearly appears from the proportion that I have described above, the main similarities between the two domains are, first, that between states and individuals and, second, the similarity of relationships assumed to exist between states and the problem of international order, on the one side, and individuals and the problem of domestic order on the other.

The assumption of a similarity between states and individuals, despite its ancient origins, received a new irresistible force in the modern age through the emergence of the concept of sovereignty. (8) However ancient the idea of the personality of the state may be, ranging back to the Justinian Roman law idea of the state as an abstract subject of rights, (9) it is in modern times, with the consolidation of the concept of a power superiorem non recognoscens that the idea of the state as a unitary actor, equipped with a single will and rational behaviour emerged. It is no coincidence then that the metaphor state-person continually recurs in the texts of the earliest theorists of sovereignty, like Hobbes and Bodin. As Carl Schmitt has pointed out, these authors transpose the concept of sovereignty from the field of theology to that of political theory (Schmitt 1922). A "political theology" emerged thereby, because they conceived of the modern state in terms of secularised theological concepts. In this way, God's superiorem non recognoscens power became the state's sovereign power and the omnipotent God became the omnipotent lawgiver.

Once states started to be conceived as unitary actors endowed, like individuals, with a sovereign will, it was easy to go one step further and apply to the international domain the same category of state of nature which has been used by authors belonging to the natural law tradition to justify the institution of a sovereign power. Indeed, the idea of an international state of nature played a crucial role in the representation of the international realm up to the present. (10) Having drawn this parallel between the individuals' state of nature and the international state of nature, the difference then emerges between those who maintain that this condition is unchangeable and those who go further and recommend for states the same social contract which has been applied to individual men: among the former is certainly Thomas Hobbes, whereas l'Abbé de Saint Pierre, who explicitly constructs his project for perpetual peace on the full extension to international relations of the Hobbesian argument, belongs to the latter.

In the current literature, Kant is also commonly held to be one of the major supporters of the domestic analogy. (11) According to this interpretation, his project Perpetual Peace rested ultimately on the mere analogical application of the conceptual apparatus of the natural law doctrine to the international realm. There are many passages which seem to support such an interpretation, as the following from Second Definitive Article of his Perpetual Peace: "People who have grouped themselves into nation states may be judged in the same way as individuals in a state of nature, independent of external laws, for they are a standing offence to one another by the very fact that they are neighbours. Each nation for the sake of its own security, can and ought demand of the others that they should enter along with it into a constitution, similar to the civil one, within which the rights of each could be secured" (Kant 1795: 102). (12)

The following parallel "civilised nations of Europe = savages of America" (Kant 1795: 103), which had already been proposed by Hobbes (Hobbes 1651: 187), highlights the suggestive power of the idea of an international state of nature, as well as the rhetorical strength of a comparison with "savage peoples". As in the Hobbesian Leviathan, we found in this passage not only a parallel between states and individuals, but also a proportion, which I held to be constitutive of analogies: "states : international relations = individuals : state of nature". However, in contrast to Hobbes, who - believing a change of this condition as ultimately unnecessary - (13) did not push the analogy any further, Kant seems to draw from this proportion the conclusion that states, like individuals, must exit from the state of nature through a social contract and give rise thereby to a juridical condition (Rechtszustand) - where for Kant, "the right (das Recht) is the sum total of those conditions within which the will of one person can be reconciled with the will of another in accordance with a universal law of freedom" (Kant 1797: 133). (14)

The problem now arises of evaluating how far Kant went with this analogy. In fact, the passage quoted above continues by saying that the constitution "similar to the civic one", to which states should give rise, must be a federation of states (Völkerbund) (15) and not a world state (Völkerstaat), because this latter would contradict the initial assumption of considering the right of nations in relation to one another insofar as they are a group of separate states which are not to be welded together as a unit (Kant 1795:102) and, moreover, because laws progressively lose their impact as the government increase its range, and a soulless despotism would finally lapse into anarchy itself (Kant 1795: 113; 1797: 171).

One could now object that these arguments, precisely in a Kantian perspective, seem not to be decisive. The last two arguments that he puts forward against the hypothesis of a world state are mostly drawn from experience, whereas in Kant's view, as we will see, experience cannot provide a certain guide for the future. One may also argue that the Constitution of the United States of America (1787), to which Kant did not devote enough attention, acts as counterexample to both these arguments: in this latter, whereas the impact of laws is guaranteed through a clear division of competencies between federal and national levels, the dangers of despotism are avoided by a complex counterbalancing between both federal and national powers as well as between legislature, executive and judiciary. Indeed, Kant by despotism means a form of government (forma regiminis) that is counterpoised to republicanism because it does not respect the separation between the government and the legislative power, and therefore it prevails each time the laws are made and arbitrarily executed by one and the same power (Kant 1795: 101). This is a point on which many interpreters have recently insisted: the arguments mentioned above hold only against the hypothesis of a despotic world state - what Kant calls "universal monarchy" (Kant 1795:113) - but not against the idea of a republican world state (Weltrepublik), which would fight the dangers of despotism through a clear division of the legislative from the executive. (16) On the other hand, one can add, the argument moving from the idea of a ius gentium seems to be a petitio principii and it contradicts the passage from the Metaphysics of Morals where he says that the ius gentium concerns not only relations among states but also among individuals (Kant 1797:165).

However, a much stronger argument against the hypothesis of a world state seems to be that again put forward in the Perpetual Peace: "while natural right allows us to say of men living in a lawless condition that they ought to abandon it, the right of nations does not allow us to say the same of states. For as states, they already have a lawful internal constitution, and have thus outgrown the coercive right of others to subject them to a wider constitution in accordance with their conception of right" (Kant 1795: 104). Here, no way out seems to be suggested: having argued according to the analogy states-individuals, a fundamental qualification is added.

If there is a domestic analogy in Kantian thinking, it must therefore be significantly limited. In fact, Kant does not simply apply to the international realm the domestic model of social order and therefore he cannot be accused of having overlooked the specificity of international order, as Bull in his criticism was suggesting. Kant's proposal of a free federation of states, as well as his perplexities on the hypothesis of a world state, stems precisely from the awareness of the difference existing between states and individuals, i.e. from the fact that states already have a constitution. In this sense, the Kantian project cannot be charged with naively applying to the international realm the domestic model of social order.

Furthermore, according to the first definitive article of Perpetual Peace, this constitution must be republican (Kant 1795:99). Republics, in Kant's view, are not inclined to move war because - as is the case under this kind of constitution - the consent of the citizens being required to decide whether or not war is to be declared, it is natural that they will have great hesitation in embarking on so dangerous an enterprise, for this would mean calling down on themselves all the miseries of war (Kant 1795: 100).

Even if this is a thesis that can be isolated from the others and is subject to criticism, (17) it should not be overlooked. By saying that states must have a republican constitution in order to achieve the ultimate end of perpetual peace, Kant clearly shows his awareness of the differences existing between states and individuals. Put in the language of international relations theories, Kant combines the systemic level of analysis to an approach which focuses on the units of the international system, and in so doing he is open to catch the specificity of international order.

At this point one could, however, object that the Kantian project of Perpetual Peace still ultimately rests on the domestic analogy - however significantly qualified this analogy might be. Indeed, as I stated above, the negative moment of the analogy is constitutive of it and, moreover, one can still argue that the necessity for states to give rise to a common power, even if this latter is not intended as a world state, is still derived from the experience of individuals in domestic society. In fact, there are in Kantian thinking at least two of the three conditions for the application of the domestic analogy. There is a separation between the domestic and the international domain, a separation that seems to acquire systematic value through the division of the concept of public right (öffentlicher Recht) into right of the state (Staatsrecht), international right (Völkerrecht) and cosmopolitan right (Weltbürgerrecht) (Kant 1795:98; 1797:136) (18). Furthermore, as we have seen, there is also the statement that similarities between the two domains exist - similarities that are ultimately expressed through the proportion "states : international realm = individuals : state of nature". But what role do all these elements play? Once it has been stated that important elements of the domestic analogy are present, the real question must focus on the role that these play. I argue that, despite their being present, they do not play any decisive founding role, because they are necessary but not yet sufficient conditions of the domestic analogy. In order to have a fully-fledged domestic analogy, all the three conditions must be respected. And this does not happen in Kant's project because, as I will try to show, he explicitly rejects the second condition of the domestic analogy, i.e. to derive the necessity of a social contract among states from the domestic experience of individual human beings.

In other words, going back to the two meanings of the term analogy described at the beginning of this paper, one can certainly say that there are in Kant some elements of domestic analogy, because there is a proportion "states : international realm = individuals : state of nature", and there is therefore an analogy according to what I called the first meaning of the term. But can we say that he derives from such a proportion an extension of knowledge from one domain to another - as would be required according to the second meaning of the term? Is his reference to the domestic experience the basis for saying that, since individuals successfully resolved the problem of the social order through a social contract, states should do the same?

In my view, it could never possibly be so because for Kant the reference to a fact, be it the experience of individuals in the domestic society, or any other kind of historical experience, could never be the basis for saying states should follow this example. This is to say that the reference to whatsoever past experience could never rise to the range of philosophical justification. As he patently puts it in the Metaphysics of Morals when talking of the constitution necessary in order to achieve the perpetual peace, "the rule on which this constitution is based must not simply be derived from the experience of those who have hitherto fared best under it, and then set up a norm for others. On the contrary, it should be derived a priori by reason from the absolute ideal of a rightful association of men under public laws. For all particular examples are deceptive (an example can only illustrate a point, but does not prove anything, so that one must have recourse to metaphysics). And even those who scorn metaphysics admits it necessity involuntarily [...]" (Kant 1797: 174). (19)

In other words, there is nothing in the experience of individuals that might tell us why states should do the same, because this experience is at best an example which does not prove anything and might even be misleading. We will came back later to the reasons why according to Kant particular examples are deceptive. Here it is sufficient to say that, according to him, in order to reach perpetual peace, which is the ultimate end of the whole theory of right, we cannot rely on the past experience, because this is contingent and can therefore at best illustrate (erläutern) a point, but does not demonstrate (beweisen) anything (Kant 1797: 174).

More so: according to Kant it could not possibly be experience that makes the public coercion of right necessary, because even if experience would tell us that human beings or states are extremely benign to each other, even if we imagined them to be as benevolent and law-abiding as we please, the a priori rational idea of a non-lawful state will still tell us that, before a public and legal state is established, individual men, peoples and states can never be secure against acts of violence from one another, since each will have his own right to do what seems right and good to him (Kant 1797:137). Therefore, independently from what it actually happens, states should abandon the state of nature because reason tells us a priori that this, as a lawless state, is unjust.

As we also read at the beginning of Perpetual Peace, where Kant is about to present the Definitive Articles for perpetual peace, these articles do not stem from an a posteriori experience, but they are based an a priori postulate of reason. Given that human beings, or peoples, living in a state of nature, even if they do not actually (de facto) injure each other, they are each other a threat for the lawfulness of their state (statu iniusto), all these articles must be based on the postulate of reason according to which "all human beings who can at all influence one another must adhere to some kind of civil constitution" (Kant 1795: 98). (20) This is not to say that these articles have nothing to do with the real world. On the contrary, it is only to say that the way in which things actually are in this world cannot be the basis for saying how they should be. But once we have derived from pure reason the principle in question, the next step consists in applying it to the possible cases provided by experience. Indeed the passage quoted above continues by saying that "any legal constitution, as far as the persons who leave under it are concerned, will conform to one of the three following types: 1) a constitution based on the civil right of individuals within a nation (ius civitatis); 2) a constitution based on the international right of states in their relationship with one another (ius gentium); 3) a constitution based on a cosmopolitan right, in so far as individuals and states, coexisting in an external relationship of mutual influences, may be regarded as citizens of a universal state of mankind (ius cosmopoliticum) (Kant 1795: 98-99).

As it clearly appears from this and analogous passages, the structure of the argument involves two fundamental steps: in first place there is the justification of an a priori postulate of reason, and secondly there is its application to experience. It is not merely that these steps are distinguished; their order cannot be inverted too, because experience cannot be the basis for the justification of an a priori principle. Therefore Kant could never have moved from the domestic experience for saying that, since this has been successful, states should follow it and also enter a lawful condition - as Bull in his interpretation suggests. On the contrary, he moves from an a priori postulate of reason and then applies it to all the possible cases: individuals within a nation, states in their relationships with one another, and finally individuals and states as far as human beings are regarded as citizens of a universal state of mankind.

Indeed it is also counterintuitive to think that Kant moves from the domestic realm to the global one: given Kant's cosmopolitan outlook, we would on the contrary expect that he moves the other way. (21) I will come back to this point later. It is sufficient to stress here the importance of the distinction between the two steps of philosophical justification of a principle and its application. I argue that it is only because this distinction has been overlooked that Kant could have been interpreted as a supporter of the domestic analogy. In fact, even if there are some elements of the domestic analogy in Kantian thinking, they could never play the decisive founding role because in his view the reference to the past experience cannot provide the basis for a philosophical justification.

This distinction can also help to make sense of otherwise unsolvable tensions of his thinking. For instance, according to Otfried Höffe, there would be a contradiction in the thesis put forward by the Second Definitive Article of the Perpetual Peace, namely, between the final proposal of a federalism of free states, which implies the rejection of a world state, and the "founding analogy states-individuals, which would instead call for statehood" (Höffe 2001: 225). However, as I have shown, the analogy states-individuals does not necessarily imply the proposal of a world state, because, in order to have this further passage, it is also necessary to accept the second condition of the domestic analogy, which, as we have seen, Kant explicitly rejects.

To put it in a different way, there is no contradiction, because in order to have a contradiction we must have two contrasting propositions which are held to be true at the same time and under the same respect. But here we have two propositions, one asserting that states like individuals should give rise to a lawful condition, and the other saying that a free federation of states, under given circumstances, might be better suited to realise this end, whose truth holds from different points of view. Indeed even if the text of Perpetual Peace with its structure of international treaty suggests that Kant is addressing these issues to policy makers, it would be misleading to read the text as if Kant would have been here aimed only at providing definitive fully-fledged institutional projects. Of course there is also the attempt at identifying the institutions better suited to implement the a priori postulate of reason that tells us that whoever find himself in the condition of influencing one another should enter a lawful condition, but this discussion situates itself at the level of application of a principle, and must not be confused with the philosophical justification of such a principle. In fact, in a Kantian perspective, whereas there might be doubts on what institution is better suited to the intended end, no doubts can emerge over an a priori principle of reason.

In my view, it is only the distinction between these two levels of argumentation, sometimes intermingled in the texts, that can explain why Kant fluctuates between the proposal of a world state and that of a free confederation of states. Many have noticed this tension without explaining how can it possibly be so, sometimes even within few pages - as it is the case of Second Definitive Article of the Perpetual Peace. (22) I think it can only be so because according to Kant, whereas at the level of the philosophical justification of a principle we can expect a knowledge that is certain, when moving on to the level of the application of such a principle, we have to take into account the contingency and multifaceted variety of the empeiria and we cannot expect the same degree of certainty.

This is a crucial point which deserves further attention. We have seen that, using the language of the contractualist tradition, Kant asserts that states can be judged as individuals in the state of nature and must therefore, like individuals, enter into a lawful condition trough a social contract. But how did Kant conceived of this latter? What kind of necessity does it implies in his view? What kind of knowledge is he talking about when he says that states should give rise to it? I argue that this necessity is neither drawn from the past experience, nor from a simple rule of prudence. In fact, in Kant's view the social contract is neither an historical fact which explains the origin of power, nor an helpful means to obtain a given end, as would have been if he had conceived of it according to the mainstream contractarian tradition. Against the first interpretation, Kant explicitly argues that an inquiry into the origins of (state) power is not only useless for the purpose of perpetual peace, but also dangerous for it could be used as an excuse for breaking the law (Kant 1797:162). On the other hand, in contrast to the prudentialist reading, he argues that political maxims must not derive from the prospect of any benefit or happiness which might accrue to the state if it followed them, i.e. by an end which each state takes as the object of his will; on the contrary "they should stem only from (ausgehen) the pure concept of rightful duty, i.e. by an obligation whose principle is given a priori by pure reason" (Kant 1795:124). Indeed, as we read in On the common saying: "this may be true in Theory, but it does not apply in Practice", "we need by no means to assume that this contract (contractus originarius or pactum sociale) based on a coalition of the will for the purposes of rightful legislation, actually exist as a fact, for it cannot possibly be so [...]. It is in fact merely an idea of reason, which nonetheless has undoubted practical reality, for it can oblige every legislator to frame his laws in such a way that they could have been produced by the united will of a whole nation..." (Kant 1793:79).

To sum up, in Kant's view, the social contract is an a priori idea of reason - and not a means to obtain a given end, and even less the explanation of the historical origin of power. And reason, in Kantian terminology, is the faculty that, given its transcendent character, is the practical faculty (23) par excellence, i.e. it is the faculty of determining the will only through the idea of duty itself, without taking into consideration what experience has to say. (24)

Therefore, in Kant's thinking, the necessity to abandon the state of nature could never have been drawn from experience - as it would have been if Kant had conceived the exit from the state of nature as a rule of skill for finding means to one's purposes, because it is the a priori determination of the will that is at stake here. In fact, a rule of skill, in Kant's view, would have involved a theoretical necessity, since, as we read in the Critique of Practical Reason, the rules of skill only point out the manifold of the possible action that is sufficient to produce a certain effect, and are thus as theoretical as any proposition that asserts the connection of a cause with an effect: whoever approves the effect must also be willing to approve the cause (Kant 1788: 159). On the contrary, what is at stake in the exit from the state of nature is the practical necessity, that is to say, the necessity stemming from an a priori determination of the will that is not subject to the approval of any further end, but that is an end in itself, namely from the duty in the sense of a categorical imperative.

We can also face the issue from the point of view of the Kantian distinction between the "moral politician" and the "political moralist". This is a distinction that Kant introduces in the Appendix to the Perpetual Peace to counterpoise two possible approaches to the problem of peace. After having said that there cannot be any conflict between politics, as an applied branch of right, and morality, as a theoretical branch of right (Kant 1995: 116), a clear division is made between the political moralist who follows maxims from experience - such as the sophistries "Fac et excusa", "Si fecisti nega", "Divide and impera" - and treats the problem of peace as a mere technical task (Kunstaufgabe), and the moral politician who rightly conceives of it as a moral task (sittliche Aufgabe). Only the second approach - he continues - is appropriate, since to bring about perpetual peace is not just desirable as a physical good, but is desirable precisely as a state of affairs which must arise out of recognising one's duty (Kant 1795: 122). Furthermore, according to Kant, to the solution of the first problem (that of Staatsklugheit) not only a great deal of knowledge of nature is required, so that one can use its mechanism to promote the intended end, but - nevertheless - "all this is uncertain so far as its repercussions on perpetual peace are concerned, no matter which of the three departments of the public right one considers. On the contrary, the solution of the second problem, that of the Staatsweisheit, presents itself as it were automatically; it is obvious to everyone, it defeats all artifices, and leads straight to its goal, so long as we prudently remember that it cannot be realised by violent and precipitate means, but must be steadily approached as favourable opportunities present themselves" (idem).

As clearly emerges from this passage, the solution to the problem of peace not only should not be drawn from experience, but it also could not possibly be so. Indeed, any attempt at doing so is doomed to failure, because, as we have seen, such a knowledge would be uncertain. This knowledge is indeed unreliable because, as we read in the same passage, "history gives examples which go in opposite directions" in reference to the means by which to reach this goal (Kant 1795:122).

There is no space here to discuss at length the reasons why, according to Kant, such a knowledge is uncertain. I will only briefly point out that in my view this happens because Kant considers human beings as equipped with what he sometimes calls the "transcendental freedom", namely, the faculty of choosing and determining a new course of actions, independently from natural causality. This kind of freedom is also sometimes called "causality as freedom" (Kant 1788: 215) and must be kept distinguished from freedom in the sense of the autonomy of will. (25) Thus, whereas we can predict what falls under the category of natural causality, in the realm of history and human actions we have to take into account humans beings' independence from natural causality.

This is not to say that we cannot have any secure guide in this field, but only that the necessity at stake is not theoretical, but rather practical. In fact, if experience cannot provide any secure guide to the end of perpetual peace, on the contrary, pure reason, as the highest legislative moral power, absolutely condemns war and prescribes the exit from the state of nature as an immediate duty (Kant 1795:122). Over the necessity of this duty no doubt can be raised, precisely because it is prescribed a priori by reason and does not rely on contingent experience.

Therefore, to conclude on this point, only an empiricist reading which neglects the distinction between the philosophical justification of an a priori and its application to the multifaceted cases provide by experience, could at the end have attributed to Kant what I called the second condition of domestic analogy. On the contrary, as I have tried to show, this latter is not only patently rejected in the texts, but also contrary to philosophical outlook of the Kantian approach to the problem of peace.

However, even if Kant cannot simply be considered a supporter of the domestic analogy because he refuses one of its condition, at different points he refers, as we have seen, to the analogy states-individuals and, furthermore, he endorses the statement of the proportion "states : international relations = individual : state of nature", thereby applying the language of the contractarian tradition to international domain. The problem therefore arises to assess the way in which these elements of domestic analogy should be read.

In first place, I think we should read them as rhetorical devices aimed at persuading rulers to apply those principles in their policy-making. This is true in particular for the Perpetual Peace, which, with its structure of international law treaty seems to be aimed at persuading policy makers of the desirability and feasibility of this project (Kant 1795: 115). This aim is made clear in the section entitled Secret Article of Perpetual Peace, where Kant explicitly invites rulers to listen and even consult philosophers on the conditions under which peace is possible. It is here interesting to note that, when introducing the Second Definitive Article, to which interpreters usually refer when attributing to Kant the domestic analogy, Kant argues that since we regard as barbarism, coarseness, brutality the way in which "savages" cling to their lawless freedom, we would expect that the European "civilised peoples" would hasten to abandon so degrading a condition as soon as possible (Kant 1795: 103). Here the series of dichotomies state of nature/lawful condition, barbarism/ civilisation, freedom of folly/ freedom of reason are clearly intended to play a rhetorical role. This seems also to be confirmed by the order of the arguments, since Kant begins by insisting on the general and uncontested contempt for the savages' lawless freedom, and only in a second step he turns the accuse of being like savage peoples to European sovereigns, introducing thereby the parallel "international relations"= "state of nature".

On the other hand, I think that the elements of domestic analogy that are present in the text play a further role, which I would define as heuristic. By this I mean that for instance the analogy states-individuals helps to formulate the problem at the beginning. This is to say that it enables to formulate the problem by addressing the question "can states be judged in the same way as individuals living in a state of nature". Indeed, as we have seen, there are many passages where this and similar formulations can be found. However, this is only the starting point of the formulation of the problem: once it is found that to solve the problem what is needed is not the reference to the experience of individuals, but the a priori justification of a principle, those formulations can no longer help. In fact, as we have seen, the final structure of the whole reasoning upon which the Kantian project ultimately rests is that of justification and application of a principle.

Therefore, we can also add, once the principle in question has been applied, the analogy is thereby overcome. As we have seen at the beginning, according to Perelman and Olbrechts-Tyteca, when we are applying a norm to new cases, we are not, properly speaking, reasoning by analogy: an analogy may help at the beginning, but once we have found that the new case fall within the same principle, the analogy is thereby removed, because there is no longer a separation between the two terms of the analogy, which - as we have seen - is one of the necessary conditions of the analogical reasoning. Therefore, going back to the Kantian reasoning, once we have found that the a priori principle, according to which whoever find her-/himself in the position of influencing one another must enter into a lawful condition, holds for states as well as for individuals, the analogy is thereby overcome because there is no longer any separation between the domestic and the international domain: rather we are in the unique domain of right. This domain is unique and the difference ultimately consists only in the different points of view for considering it: namely that of individuals within a nation (Staatsrecht), of states in their relationships with one another (Völkerrecht) and, finally, of individuals and states as far as, since the earth surface is spherical, human beings must also be considered as citizens of a universal state of humankind (Weltbürgerrecht). These are ultimately only three possible forms of the unique concept of rightful state (rechtlicher Zustand) so that if only one of these lacks the whole structure is undermined and must finally collapsed (Kant 1797:137).

Thus, what I have called the first condition of the domestic analogy, namely the separation between the two domains, has an uncertain status, too. It is true that it plays a central role in the distinction between the three domains of public right, but this distinction must be read merely as an approximation to a system. In fact, as we read in the preface to the Metaphysic of Morals, the concept of right, despite its being a pure concept, is oriented towards practice, because it finds application in the multifaceted cases provided by experience; this is the reason why Kant says he has to call his theory of rights "The metaphysical elements of the theory of rights" (metaphysische Anfangsgründe der Rechtslehre) - and not simply "metaphysics of right". According to him, there cannot be a purely metaphysical system of right (aus der Vernunft hervorgehendes System) - what would have implied a perfect division of its parts which, having to take into account the multiplicity of the experience is impossible, but only an approximation to a system (nur Annährung zum System) (Kant 1797, AB, III, IV).

This may also explain why, in dealing with the crucial issue of the subjects of the jus gentium, Kant asserts that this latter involves not only the relationship between states but also between individuals. Quoting again the Metaphysics of Morals: "A state of nature among individuals or families (in their relations with one another) is different from a state of nature among entire nations, because international right involves not only the relationship between one state and another within a larger whole, but also the relationship between individual persons in one state and individuals in the other or between such individuals and the other states as a whole" (Kant 1797: 165). Therefore even in considering the relationship between states we cannot but consider simultaneously individual human beings.

This also confirms what we suggested before, namely that, the Kantian outlook is cosmopolitan from its inception, and it remains cosmopolitan even when he considers the ius gentium - it does not matter whether he takes a world state or a free confederation as the institution better suited to implement his cosmopolitan principles. But if the perspective is cosmopolitan, then it is counterintuitive to think that Kant could move from justice in the domestic sphere to its application to the global one (26). On the contrary, as we have seen, he moves from an a priori principle and then applies it to different forms of right, but always with the awareness of their interdependence. This was clear to him since 1784, when in the Seventh Proposition of the Idea for a Universal History he wrote that "the problem of establishing a perfect civil constitution is subordinate to the problem of a law-governed external relationship with other states, and cannot be solved unless the latter is also solved" (Kant 1784: 47). Since then Kant was well aware of the interdependence of the three levels of right Staatsrecht, Völkerrecht and Weltbürgerrecht, interdependence which he derives from the observation that, being the surface of the earth spherical, individuals and peoples cannot indefinitely spread, but have to reencounter. (27)

Precisely for this awareness Kant occupies a peculiar position in the modern political philosophy. And, as Katrin Flikschuh recently noticed, this position also counterpoises Kant to most contemporary liberals: whereas these latter conceive of justice at the domestic level and then apply it the global context, Kant does not distinguish between different theories of justice, but simply considers different levels of institutionalising his cosmopolitan conception of right (Flikschuh 2000: 170). Indeed many of those who consider themselves Kantian, recover Kant through an empiricist reading which neglects its dependence on a metaphysical approach, and this may finally explain why their reasoning often appears closer to the domestic analogy than that of Kant.

The best example of this tendency is Rawls' treatment of the Law of peoples: after having considered justice in the domestic sphere he applies the same devices (original position, veil of ignorance etc) to the international one, only by adjusting them (Rawls 1999). There is not space here to discuss Rawls' position in detail; suffice to note that his approach, which takes states and not individuals as the starting point and which entails a radical refusal of the hypothesis of a world state, is in my view much closer to the domestic analogy than Kant's cosmopolitanism, which appears, at times, to embrace this hypothesis.

Therefore, in conclusion, to adopt a cosmopolitan perspective does not automatically imply to support the domestic analogy. As this analysis of the Kantian thinking has shown, the domestic analogy finally rests on an analogical reasoning that is alien to Kant's approach, which is based, on the contrary, on the deduction and application of an a priori principle. On the other hand, one might object that it is precisely in adopting such an approach that Kant is wrong; but this is a another criticism, which calls into question a different set of problems.

References

  • Abbagnano, N. 1971, "Analogia" in Dizionario di filosofia, Torino, UTET, pp.37- 40
  • Ashley, R. 1988 "The Powers of Anarchy: Theory, Sovereignty, and Domestication of Global Life", in Der Derian (ed) 1995, International Theory - Critical Investigations, pp.94-128
  • Barakawi, T. and Laffey, M. 1999 " The Imperial Peace: Democracy, Force and Globalization", European Journal of International Relations, 5(4): 403-434, 1999
  • Beitz, C. 1979, Political Theory and International Relations, Princeton, Princeton University Press
  • Bull, H. 1966, " Society and Anarchy in International Relations", in Butterfield e Wight (eds)1966, pp.56-68
  • ---------1977, The Anarchical Society, London, Macmillan
  • ---------1981, "Hobbes and the international anarchy", in Social Research, n.17, pp 717-39
  • Bottici, C. 2002, Globalizzazione, sovranità e anarchia, in D'Andrea D., Pulcini E. (eds) Filosofie della globalizzazione, Pisa, ETS
  • Butterfield, H. and Wight, M. (eds) 1966, Diplomatic Investigations: Essays in the Theory of International Politics, London, Macmillan
  • Doyle, M. 1983, "Kant Liberal Legacies, and Foreign Affairs", Parts I and Part II, Philosophy and Public Affairs, 12 (3): 205-235 and 12 (4): 323-353, 1983
  • --------1997, Ways of War and Peace, Norton, New York
  • Edwards, P. 1967, "Analogy in Theology" and "Models and Analogy in Science" in The Encyclopaedia of Philosophy, Macmillan, London, pp. 94-97 and 354-359
  • Flikschuh, Katrin, 2000, Kant and Modern Political Philosophy, Cambridge University Press
  • Habermas, J. 1996, Kants Idee des ewigen Friedens - aus dem historischen Abstand von 200 Jahren, in Die Einbeziehung des Anderen, Suhrkamp, Frankfurt am Main
  • Hesse, M.B, 1970, Models and Analogies in Science, University of Notre Dame Press, Indiana
  • Hardt, M and Negri, A. 2000, Empire, Cambridge, Harvard University Press
  • Hobbes, T. (1651), 1985, Leviathan, London, Penguin Books
  • Höffe, O. 1995, (ed), Immanuel Kant - Zum ewigen Frieden, Akademie Verlag
  • --------- 2001 "Königliche Völker" Zu Kants kosmopolitischer Rechts und Friedenstheorie, Suhrkamp, Frankfurt am Main
  • Hurrel, A. 1990, "Kant and the Kantian Paradigm in international relations", in Review of International Studies, XVI, pp.183-205
  • Kant, I., 1781, Kritik der reinen Vernunft, in Gesammelte Schriften, hrsg. Königlich Preußischen Akademie der Wissenschaft, Berlin, 1902, AA.III-IV; Eng. tr. Critique of Pure Reason, in The Cambridge Edition of the Works of Immanuel Kant, 1998, Cambridge University Press
  • 1784, Idee zu einer allgemeinen Geschichte in weltbürgerlicher Absicht, AA VIII, 15-32; Eng. tr. Idea for a universal history with a cosmopolitan purpose, in Political Writings, 1991, Cambridge, Cambridge University Press, 41-53
  • 1788, Kritik der Praktischen Vernunft, Bd. V, 1-164; Eng. tr. Critique of Practical Reason, In The Cambridge Edition of the Works of Immanuel Kant, Practical Philosophy, pp.133-273
  • 1793, Über den Gemeinspruch: "Das mag in der Theorie richtig sein, taugt aber nicht für die Praxis", AA VIII 273-313, Eng. tr. On the Common Saying: "This maybe true in Theory, but it does not apply in Practice", Political Writings, 1991, Cambridge, Cambridge University Press, pp.61-93
  • 1793, Die Religion innerhalb der Grenzen der bloβen Vernunft, Bd. VI, pp. 1-202; Eng. tr. The religion in the limits of pure reason, in The Cambridge Edition of the Works of Immanuel Kant, 1998, Cambridge University Press
  • 1795, Zum ewigen Frieden. Ein philosophischer Entwurf, Bd. VIII, 341-386; Eng. tr. Perpetual Peace: a Philosophical Sketch, in Political Writings, 1991, Cambridge, Cambridge University Press 93-131
  • 1797, Metaphysik der Sitten, erster teil: Metaphysische Anfangsgründe der Rechtslehre, VI, 203-372; Eng. tr. Metaphysics of Morals, in Political Writings, 1991, Cambridge, Cambridge University Press 131-175
  • Kersting, W. 1992, Politics, freedom and order: Kant's political philosophy, in Guyer P. (ed) The Cambridge Companion to Kant, Cambridge University Press
  • --------- 1993, Wohlgeordnete Freiheit: Immanuel Kants Rechts und Staatsphilosophie, Suhrkamp, Frankfurt Am Main
  • Marini, G. 1998, Tre studi sul cosmopolitismo, Istituti editoriali e poligrafici internazionali, Pisa
  • Perelman, C., Olbrechts-Tyteca, L. O. 1958, Traité de l'argumentation. La nouvelle rhétorique, Presses Universitaire Francaise
  • Rawls, J. 1999, The Law of Peoples, Harvard University Press
  • Ritter, J. 1971, "Analogie" in Historisches Wörterbuch der Philosophie, Darmstadt, Wissenschaftliche Buchgesellschaft, Bd. I, pp. 214- 227
  • Rosenau, 1990, Turbulence in World Politics: A Theory of Change and Continuity, Princeton, Princeton University Press
  • Runcinman D., 2000, "What Kind of Person is Hobbes' State? A Reply to Skinner", Journal of Political Philosophy, 2, pp 268-278
  • Russett, B. 1993, Grasping the Democratic Peace: Principles of a Post-Cold War World, Princeton, Princeton University Press
  • Schmitt, C.1922, Politische Theologie: vier Kapitel zur Lehre von der Souveränität, Berlin, Duncker & Humblot
  • Skinner, Q. 1999, "Hobbes and the Purely Artificial Person of the State", Journal of Political Philosophy,7, pp.1-29
  • Suganami, H. 1989, The Domestic Analogy and World Order Proposals, Cambridge, Cambridge University Press
  • Barakawi, T. and Laffey, M. 1999 " The Imperial Peace: Democracy, Force and Globalization", European Journal of International Relations, 5(4): 403-434, 1999
  • Toulmin, S. 1990, Cosmopolis. The hidden agenda of modernity, New York, Free Press
  • Zolo, D. 1997, Cosmopolis. Prospects for World Government, Cambridge, Polity Press
  • ------ 1998, I signori della pace. Una critica del globalismo giuridico, Roma, Carocci
  • Wight, M. 1991, International Theory. The Three Traditions, London, Leicester University Press

Notes

*. This paper discusses arguments taken from my Uomini e Stati. Percorsi di un'analogia (2004, Pisa). A shorter version of this paper has been published in the Journal of Political Philosophy (December, 2003), while a German version is forthcoming in Merle, J.C. (Ed.), Globale Gerechtigkeit, (2005, Stuttgart). I am grateful to the "democracy in the age of globalisation" research group, directed by Furio Cerutti (University of Florence), for support and critical feedback. I am also grateful to Luca Baccelli, Katrin Flikschuh, Otfried Höffe, Thomas W. Pogge and Peter Wagner for their comments on earlier drafts of this paper.

**. Dipartimento di Filosofia, Università di Firenze

1. I am here relying on Wight's classical distinction of three traditions of International Theory (Wight 1991). Instead of the label "revolutionist" I prefer to use the term referring to its founding father, since I do not find the former appropriate for a thinker so little inclined to sudden violent changes like Kant.

2. See, for instance, Rosenau 1990, Zolo 1997, 1998 and Hardt, Negri 2001.

3. See "analogia" in Abbagnano 1971, pp. 37-40; "Analogie" in Ritter 1971 pp. 214-228; "Analogy in theology" and "Analogy in science" in Edwards 1967, pp. 94-96 and pp. 354-359.

4. See Aristotle EN, V, 5 1131a and on the analogy as equality of relationships see Met. 9,6, 1047 b 35.

5. On the role played by the sovereignty versus anarchy dichotomy in modern political theory see Bottici 2002.

6. See for instance Ashley, whose criticism of the "double-voiced practice of domestication of the Western reasoning men" - a practice so called because it turns on the dichotomy of sovereignty versus anarchy -finally results in a holistic criticism of modernity (Ashley, 1988).

7. On the distinction between positive and negative analogy, see Hesse 1970.

8. See for instance Skinner's reconstruction of the nexus between the concept of sovereignty and the Hobbesian conception of the State as an artificial person (Skinner 1999, Runciman 2000).

9. I do not think that we can ascribe the idea of the state as a moral and juridical person, abstract subject of rights to classical political thought. The Platonic parallel polis - soul and the Aristotelian organicistic metaphor point instead much more to the organic composition of society, where the whole of the society is superior to the sum of its parts, but cannot be conceived of without them. On the contrary, with the concept of the state as a persona ficta, this latter is thought of as somehow independent from its members.

10. For a discussion of this point as well as a critical assessment of the international relations as a state of nature, see Beitz 1979.

11. Among the supporters of this interpretation in the field of international relations theories, see Bull 1977, Wight 1991, Suganami 1989. More recent readings in the field of Kantian studies are Höffe 1995, 1998, Marini 1998.

12. I will follow H.B. Nisbet's translation, Kant, Political Writing, Cambridge University Press, 1991, indicating the points where I am taking distance from it. Original passages are quoted from the Akademie Ausgabe, Gesammelte Schriften, hrsg. Königlich Preuβiche Akademie der Wissenschaft, Berlin, 1902.

13. After having stated the parallel savage peoples = sovereign states, and having described all the incommodities of such a "posture of warre", Hobbes then conclude that, since sovereign states "uphold thereby the industry of their subjects, there does not follow from it that misery which accompanies the liberty of particular men" (Hobbes 1651: 188). For a further discussion see Bull 1981.

14. The translation of Kant's noun "Recht" has always been a problem for translators, since, in contrast to the English term "right", Kant's usage rather points to a total condition of external lawfulness - as counterpoised to inner morality. In order to keep the symmetry with the correspondent adjective recht-right, I will also follow H.B. Nisbet's translation "Right", sometimes recurring to unusual terminology such as "international right".

15. What Kant meant by Völkerbund is a very controversial point. For instance in Die Religion innerhalb der Grenzen der bloßen Vernunft, we found two times the expression Völkerbund als Weltrepublik (B 30-31; B 182-183). In this passage, however, he seems to indicates rather what we would today call a confederation of sovereign states, i.e. a free union of states.

16. See Höffe 2001, Marini 1998. In the Second Definitive Article Kant defines the republican constitution as the constitution which is founded upon three principles: firstly, the principle of freedom for all members of a society (as men); secondly, the principle of the dependence of everyone upon a single common legislation (as subjects), and thirdly, the principle of legal equality for everyone (as citizens) (Kant 1795: 99). In the following passages, he qualifies this definition by saying that republics differs from democracies because republicanism is that political principle whereby the executive power (the government) is separated from the legislative power, whereas democracy, in the truest sense of the word, is necessarily a despotism, because the laws are made and arbitrarily executed by one and the same power (Kant 1795: 101).

17. This theory is today discussed under the heading of "democratic peace". Among his supporters see Doyle 1983, 1997, who explicitly refers to Kant, and Russett 1993, and more recently Rawls 1999. For a critique see Barakawi, Laffey 1999.

18. As we have seen above Kant uses the separation between Staatsrecht and Völkerrecht also as an argument against the Hypothesis of a world state (Kant 1795:102).

19. There are many passages which go in the same direction, for instance the following from the Metaphysics of Morals: (Kant 1797: 137.

20. See the whole passage of Perpetual Peace (Kant 1795: 98).

21. On this point I am indebted to Katrin Flikschuh (see particularly Flikschuh 2000:170).

22. See for instance Hurrel 1990 who carefully reassesses the balance between the statist and the cosmopolitan elements of Kant's thinking but without explaining why those elements can possibly coexist (Hurrel 1990)

23. In Kant's conception "practical" is what concerns the determination of the will, where the will, according to the Critique of practical reason, is the faculty of determining behaviour through the representation of rules (Kant 1788: 163).

24. According to Kersting the idea of a social contract can be interpreted as a variant of the text of universalisability, i.e. as the political counterpart of the categorical imperative: just as the categorical imperative as a moral principle allows for the evaluation of the lawfulness of maxims, so does the original contract as the principle of public justice serve to measure the justice of positive laws (Kersting 1993: 355).

25. These two meanings of the concept of freedom are to be kept separate, sometimes even against Kant himself. Kant clearly distinguishes between the two - both in theoretical and practical writings, but sometimes it is not so clear to which he refers. This can be due to the fact that, according to the conception put forward in the Critique of Practical Reason, (transcendental) freedom is the ratio essendi of the moral law, and therefore of the autonomy of the will, but this latter is the ratio cognoscendi of the former (Kant 1788, p. 140).

26. See Flikschuh 2000: 170.

27. The spherical surface of the earth is the central to Kant since the earlier writings on the philosophy of history (Kant 1784). It also continually recurs in the political writings, where it plays the central role in the foundation of the cosmopolitan right