2007

Towards a Modern, Realist Philosophy of International Law (*)

Danilo Zolo

1. A vetero-European doctrine

It would be a mistake to underestimate the growing importance of the Western doctrine of 'legal globalism' over the last decade and a half, after the collapse of the Soviet empire and the end of the division of the world in two blocks. Its international weight certainly derives from the economic and military hegemony of the West, as well as from the influence of globalization processes, which seem to require a planet-wide unification also of normative and jurisdictional structures. However, it depends as well on the fact that it is a legal philosophy aimed at legitimizing present-day international institutions, particularly their more recent 'humanitarian' and judicial developments.

In my opinion, however, this success, while undeniable, does nothing to attenuate the weakness of a theory which, in spite of its cosmopolitan pretensions, remains anchored in the culture of old Europe, which is to say, classical Christian doctrine of natural law. The idea of international law that it advances is inseparable from a theological-metaphysical view - reflected in the notion of civitas maxima - which founds the international legal community on a twofold belief in the moral nature of man and the moral unity of mankind.

This legal philosophy is dominated by the idea, Kantian and neo-Kantian, that the progress of humanity is possible only if certain ethical principles are shared by all men and validated by super-national powers that transcend the 'polytheism' of existing ethical convictions and normative systems. Indeed, the individualistic, liberal doctrine of the human rights - which likewise, as Kelsen acknowledged, bears the imprint of the natural law doctrine - is proposed nowadays obsessively to non-Western cultures as the paradigm for a world political constitution. Even a phenomenon such as war is attributed to the situation of 'anarchy' which, according to this monistic philosophy, has marked no less than three centuries of relations among states. A just world order and a stable, universal peace can be guaranteed, as natural law theorists never cease to repeat, only by a super-national power hierarchy, a universal legislation (inspired by Judaic-Christian ethics) and a compulsory international criminal jurisdiction.

2. A modern, realist philosophy of international law

If what I have said has any basis and significance, then it can be maintained, echoing a judgment of Martin Wight's, that Western culture lacks a modern philosophy of international law. Obviously, I do not deny that there are various theories of international relations, some of them a far cry from the Kant-Kelsen line. One need only recall the tradition of international political realism, harking back to Hans Morgenthau, and its more recent neo-realist developments, from Kenneth Waltz to Robert Gilpin. A certain weight must also be acknowledged to the Grotius line - as Martin Wight calls it - which rejects the idea of an international community which includes all men since all men are moral subjects. This current proposes as an alternative the model of a legal society of sovereign states or in any case of political and social organizations, not of single members of the human species.

My intent is to assert something rather limited, but also quite specific: I hold that what is lacking is an international philosophy of law able to withstand the hegemony of the tradition of thought going back to Kant's metaphysical ethics and the cosmopolitan interpretations of Zum ewigen Frieden. In other words, a modern legal philosophy should be capable of advancing proposals for legal and institutional organization that are different and alternative with respect to the 'Holy-Alliance model' that has found its most fulfilled expression in the United Nation Charter (1).

A modern theory of international law would have to construct its categories inductively, starting from an analysis of the political, economic, and military problems on the international agenda. It should also take into account the normative expectations emerging from diverse civilizations of the planet - Samuel Huntington has identified about ten - which are for the most part in conflict with the 'universally legislative' ambitions of the West. Such a theory should not start out, as Kelsen's normativism does, from an idea of reason, Platonically assumed as the condition for law to be thought of in scientific terms. It should instead give up any prerogative of cognitive objectivity and deontological universality as far as its enunciations are concerned, and thus not attempt to anchor the validity of international law in a presupposed 'truth' or, still less, a universal ethics.

From the standpoint of a bounded-rationality epistemology (Herbert Simon), a legal theory of this sort ought to limit itself to proposing certain normative instruments as being functional for certain purposes hypothesized as worthy to be pursued. Moreover, in setting out these instruments and purposes, it should take into account historical and social variables that would limit the effectiveness of the legal norms, especially those of international law. The latter, as Hersch Lauterpacht cautioned, is so exposed to the contingencies of politico-military equilibriums and market turbulence as to run a greater risk than other forms of law that the effectiveness of its norms be eroded.

This legal theory should also define, at least implicitly, the relationship to be established between the functions of the international legal system and those of the legal systems of single states. Furthermore, in making this choice, it should refer, at least implicitly, to a legal philosophy that defines the functions to be attributed to law as a specific normative technique - one that interacts with other normative modalities to contribute to the stabilizing of political and social organizations very different among themselves.

Of course, this is not the place to attempt such challenging and complex theoretical operations. I shall limit myself here to maintaining that an alternative conception of international law to the one I have called 'globalist' could be developed by taking as its point of reference a neo-Grotian and realist line of thought that goes from Martin Wight to Hedley Bull and, at least in part, to Andrew Hurrell. This line of thought has been expressed in its most authoritative form in a very valuable work, Hedley Bull's The Anarchical Society, published in London in 1977. The task would be - this is my 'reconstructive' proposal - to develop, within the specific terms of a realist philosophy of international law, what has been proposed, often only on the level of intuition or general statements, by this Oxonian current of non-globalist thought.

3. A few theoretical 'starting points'

On the basis of these premises I shall try in the following pages, without any claim to making a systematic presentation, to indicate a few 'starting points' which, in my opinion, may be useful for the elaboration of an alternative theory of international law to Kantian-Kelsenian metaphysics.

One may object that my 'starting points' are not neutral, indeed are laden with philosophical assumptions and value judgments. I am not going to deny this; my 'starting points' presuppose - let me put it this way - a starting place. This place - to paraphrase a famous anecdote by Hedley Bull (2) - is precisely the place where I find myself in this moment (and where others find themselves here together with me). The place cannot be chosen to one's liking.

In the formulation of my 'starting points' there are implicit references to a general epistemology, an anthropology, and a politico-legal theory that are fairly well-defined. These are the places - the tabulae inscriptae - whence my reflections start, and which condition my attempts to take my first steps. However, what I do claim, with respect to the Kantian-Kelsenian tradition I criticize in these pages, is the non-speculative character of my philosophical premises and the non-dogmatic nature of my approach to the problems of politics and law. All I intend to do is to propose an interpretative pattern - not an 'explanation' - of normative phenomena on the international level: a pattern which, thanks to a certain coherence and plausibility of its own, allows for the understanding of more significant problems and a certain possibility to foresee their developments.

As far as concerns theologico-metaphysical assumptions as to the moral qualities of the human species, I prefer the results produced by ethological and anthropological research (without claiming that they have any absolute value). These studies present the homo sapiens to us as an 'animale superior' much exposed to environmental risks and thus deeply insecure and always on the outlook for reassurance, protection, and power. Indeed, 'producing security,' according to a realist conception, is the specific function of the political system and its legal system. These social structures guarantee individuals a threshold level of controllability and predictability for the social environment in which they live, protect them selectively from risks, and, in fine, 'reduce fear' (3). According to this vision, politics and law do not have the task - which instead the neo-Kantian contractualists assign to it, starting with John Rawls - of carrying out ideal objectives such as justice or the 'common good,' not to mention universal 'brotherhood.' These would seem to be aspirations, rebus sic stantibus, for the realization of which all premises are lacking; one might say that what is lacking is the starting points.

These being some of my philosophical 'prejudices,' it is natural that I, along with Hedley Bull and Stephen Toulmin, should look with diffidence on a politico-legal Cosmopolis as the goal. It is no less natural that I should mistrust the Kelsenian idea of a civitas maxima, understood as an ideal place in a (Western) vision of reason where a universal morality, a universal law, and a universal state are supposed to converge; in other words, a place wherein to realize just that 'optimal political system' that Bull criticizes in The Anarchical Society, to which he says he prefers the more realist project of a 'minimal political order.'

As will soon be clear, this is the model that I quite clearly prefer, in the name of pluralism, complexity, and cultural diversity, all of which I do not conceive of as stumbling blocks to the progress of legal and political rationality but, quite the contrary, as a precious, anthropological heredity, rich with potential for development. My preference comes, among other sources, from the general theory of systems: in highly complex situations with variable environmental turbulence it is more prudent to get along with even a very high degree of disorder rather than to try to impose a perfect order.

So here are my 'starting points'.

3.1. An 'impure' theory of international law

The first of my points is the idea (a legal realist idea) that any theory of international law must be an 'impure theory' of law. I am not going to take up the general theme of the autonomy of legal method here, nor do I intend to linger over any further criticism of the 'pure theory of law' proposed by Kelsen. I do intend to claim that it is in the international arena, more than anywhere else, that the impurity of legal theory seems to be imposed inevitably, just because of the nature of its object. The relationship between law and power is so close and so ambiguous in the international sphere that a philosophy of international law would be nothing more than normative speculation if it did not pose at its theoretical core the many variables that make the relationship between the law in books and the law in action problematic: that is, if it did not study, as the specific object of 'legal science,' the network of political, economic, and social transactions through which the principles and rules of law become an effective discipline for single concrete cases. Think only of the complex relationships of thematic intersection and normative interdependence that link legal prescriptions among themselves on the one hand, and, on the other hand, with the religious and cultural traditions of various continents, the politico-military strategies of the superpowers, political terrorism, and international crime.

A theory of law as a science of pure normative propositions - hence, separate from any political theory or any sociology of agents and normative policies - would be, if this be possible, still less sustainable in the international than in the domestic sphere. A modern, realist theory of international law, then, should start by thematizing the relationship between forms of law and the, so to speak, deformities or absence of form of the arcana imperii. It should likewise translate this relationship into a systematic theoretical contamination between law and power and between power and violence. This would mean, in other words, acknowledging the impossibility of 'purifying' (international) law by splitting it away from (international) politics, and the parallel impossibility of sterilizing politics, in all its forms, from particular interests and the conflict between mutually hostile symbolic universes in which those particular interests nest and find their justification.

3.2. The structural differentiation of legal systems

A second, methodologically very significant point is the thesis of the plurality of legal systems and their structural differentiation. Even systems that are as different from a morphological standpoint as - say - canonic law and, at another extreme, the International Tennis Federation, are worthy, with equal justification and without exception, to be considered 'legal systems.' This is legitimate with the sole condition that such systems serve, albeit with varying degrees of effectiveness, to regulate single controversies or single cases in point. Thus, one must reject the monistic, normativistic idea that, on the basis of a presumed logical necessity, there can be one and only one all-including legal system. In particular, the tendency must be criticized, that we have seen operative deep in Kelsen's writings, to gauge the degree of historical evolution of international law according to the 'evolutionary stages' of the law of states. Apart from the doubts one may have as to the existence of evolutionary laws governing the structure of legal systems, the Kelsenian idea that international law is a 'primitive system' due to the decentralized nature of its coercive functions is devoid of interest from the legal realist point of view.

Kelsen holds that at the historical-evolutional level, a legal system is all the more perfect to the degree in which the wielding of force is taken away from individual initiative and centralized in specialized organs, such as governments and courts. In this sense the modern nation state is, for Kelsen, as we have seen above, a perfect legal system, because within it collective order is guaranteed by means of a high concentration of the use of force. However, in the realist perspective, the legal quality of the international system would be evaluated not from the standpoint of its 'structure' but from that of its 'functions,' that is to say, its capacity to make its normative claims effective in relation to the purposes socially assigned to it.

Once the hypothesis has been accepted that international law has the task of guaranteeing 'collective security' worldwide - this is the task the later Kelsen assigns to it, giving up his previous reference to peace - it is in relation to its capacity for fulfilling this task that its legal authority must be measured, or, which is the same thing, its degree of regulatory effectiveness. Hence, there is little sense in holding that international law, to overcome its condition of structural primitiveness, must tend toward centralizing its coercive functions and making its penal jurisdiction obligatory. An international system with a centralized structure but unable to regulate international conflict with any degree of effectiveness - a system which, say, contributed to the spread of terrorism or the secession of powerful international criminal associations - would not deserve to be appreciated as the 'superior' legal system that Kelsen formalism claims it must be.

If instead, as I would be inclined to propose, the international legal system were not only charged with the task of guaranteeing collective security but also and above all with that of ritualizing the use of force, then it is in relation to this parameter that its regulatory effectiveness and ultimately its legal value would be measured. Of crucial significance in this case would be its ability - clearly not absolute - to subject international power to the forms and procedures of law, thus excluding arbitrariness and privilege and in particular harnessing the tendency of the great powers to readily use their military might to carry out their geopolitical strategies.

Adopting this approach, one might have doubts whether the legal model designed by the United Nations Charter is very 'advanced.' Kelsen and Bobbio applaud this model because they interpret it as a system of international security with a high degree of centralization of coercive functions (which in their opinion the League of Nations totally lacked). However, we have seen that, using a realist parameter, it is not the most centralized legal-institutional model that should be considered superior but the one able to be most effective in its regulatory aims, and from this point of view one finds it hard to recognize any superiority of the United Nations with respect to the League of Nations. The concentration of highly significant military functions in the Security Council - which distinguishes the United Nations from all previous international institutions - has surely not coincided over the past decades with any greater capacity to subject the power of states to the rules of international law.

From this hypothetical point of view, it could indeed be maintained that the League of Nations should be seen as the superior legal form. Its Covenant did not bring international law to bow to the supremacy of a few great powers, bestowing them with the 'legal surplus value' of veto power and a series of other privileges and arbitrary faculties; nor did it mortify one of the fundamental principals of modern law, namely the formal equality of the subjects of rights. Thus, in spite of its serious limitations, the League of Nations did not lend itself to an almost automatic legal legitimization of the politico-military strategies of the great powers, as the United Nations has repeatedly done.

3.3. A generalization of the theory of 'legal regimes'

My third point has to do with the problem of the sources of international law. Normativism stresses the necessity of having specialized, centralized organs handle normative activity. A legal realist approach, although it does not deny the need for an explicit production of norms, should, in my opinion, lay stress rather on the role that customary, consensual, and conventional processes play in forming international law. In the sectors where it is more effective - one thinks for example of diplomatic and consular protocols, reciprocal protection of citizens abroad, or political asylum - modern international law has formed very slowly over the centuries. Even classical ius gentium, as Friedrich von Hayek has maintained in direct disagreement with Kelsen, was formed not as a result of legislative decrees but on the basis of time-honored trade conventions, observed in ports and trade fairs. According to Hayek, these conventions, and not bourgeois revolutions, should be credited with having favored the rise of free, open societies in the West.

As against the legislative voluntarism arising from Enlightenment legal culture and nineteenth-century reliance on codification from Jeremy Bentham to John Austin and down to Kelsen, a legal realist philosophy should consider law in general, and international law in particular, as a - so to speak - 'systemic' phenomenon. What is finally the effective discipline for each concrete case is the result of normative interactions between actual behavior and expectations as to what that behavior would be (and expectations as to expectations as to that behavior). Every agent pursues particular ends without at all seeking to contribute to the stability of a 'legal system'; indeed, he tries to get around its formal rules whenever he has something to gain from doing so. In other words, the actual practice governing social relations very selectively assimilates, by means of complex re-interpretive and distorting transactions, the impulses that come to it from formal norms, and this holds especially for the international legal system, which is perhaps the most complex.

For this reason, rather than accredit the idea of a lex mundialis to discipline international relationships from the heights of a monolithic normative hierarchy, a sort of planetary Stufenbau, a theory of the sources of international law might appropriate and seek to generalize, as I have said above, the theory of 'international regimes' developed by Robert Keohane and Stephen Krasner. (4) Keohane and Krasner have shown that the lack of any imperative prescription and of a centralized, binding jurisdiction does not rule out the possibility that most international agents are capable of unitarily disciplining important questions. The deviant behavior does not need to be sanctioned by using military might. For instance, this holds in issue-areas like international trade, the monetary exchange system, deep-sea fishing, space research, meteorology, disciplining human activity in the Antarctic, and many other sectors as well. 'International regimes' establish frames of legal responsibility, consolidating systems of general norms, specific rules, and procedures whose purpose is to discipline the interaction among agents, define their rights, and regulate their behavior in various ways, not including the use of military force.

In conditions wherein international factors are highly complex and interrelated, multilateral negotiation is a decentralized source for producing and applying law and it is efficacious in spite of the absence of jurisdictionally centralized normative and jurisdictional functions. The nature, for the most part spontaneous, of the phenomenon shows, among other things, that a disciplining of international relations does not necessarily mean a loss of sovereignty by single states, although it obviously requires a degree of agreed-upon self-limitation.

3.4. Modern war: legibus soluta

My fourth point takes up the issue of how modern war can be qualified legally, and the thematically related issue of the role that recently constituted or soon to be constituted super-national judicial institutions can play in a strategy of peacekeeping.

The main thesis I intend to propound - in opposition to one I do not hesitate to consider an out-and-out Kelsenian theoretical accident - is that modern war can in no wise be interpreted as a legal sanction available to international law. In another essay I have already set out the reasons which, in my opinion, corroborate this thesis (5). Here I would only add a general consideration: a modern, realist philosophy of international law ought to stand askance from any version of the doctrine of 'just war,' whether in an ethical version, as is the case of the tradition of iustum bellum lately reproposed by Michael Walzer (6), or in a more strictly legal version (as in Kelsen's re-elaboration).

If the primary function of international law is, as I think it is, to subject the exercise of international force to the respect of general rules and hence, above all, to criteria of proportionality, discrimination, and measure in the use of force, it follows that in our nuclear age - after Hiroshima and Nagasaki - war must be considered a phenomenon incompatible with law, being legibus solutum. Bobbio himself advanced this thesis, as we have seen, in direct polemic with the modern favorers of the doctrine of bellum justum. Modern war - Bobbio wrote in an essay from the 1960s - is beyond any possible criterion of legitimization and legalization, uncontrolled and uncontrollable by law, like an earthquake or a tempest. After having been first thought of as a means for enforcing law (from the theory of the 'just war') and then as a possible object to be regulated legally (in the evolution of ius in bello), today war has returned, according to Bobbio, to be the antithesis of law (7).

Modern war, waged with weapons of mass destruction (particularly if nuclear or quasi-nuclear, as in the Gulf War of 1991) is, ethically and legally, an incommensurable event, because, by its very nature, its function is to destroy - without proportion, discrimination, or measure - the life, belongings, and rights of thousands, even hundreds of thousands of human beings, regardless of any consideration of responsibilities deriving from their behavior. (8) Therefore, quite like terrorism, modern war is outside the purview of any normative technique, whether legal or - even more so - moral, since, on the moral level, the theory of 'objective responsibility,' as infelicitously revalued by Kelsen, is even less acceptable. Modern war is incompatible with international law, just as a police operation to capture or kill a criminal holed up in a city neighborhood would be in contradiction with the legal system of a democratic state if it were decided, in the name of law and order, to carpet-bomb the whole neighborhood, killing the inhabitants along with the criminal.

As far as super-national judiciary institutions are concerned - from the Nuremberg and Tokyo Tribunals to the more recent Criminal Tribunals for former Yugoslavia and for Rwanda - suffice it to cite Hedley Bull's critique of such institutions already in the 1970s. (9) Here I would only add one general consideration: what cannot but be surprising today is the lack of almost any philosophical reflection on the judiciary activities of these new penal organs. The fact that behind the Nuremberg Tribunal there was an evident desire for political vengeance, scantily draped in the robes of international justice, was so clear that even Kelsen could not but denounce it forcefully. However, the new international tribunals are based on the idea that inflicting severe sentences on a few dozen people - in the future it might be a few hundred - a sort of 'humanitarian criminal justice' is achieved suitable to promote human rights and guarantee world peace.

One of the most recurrent slogans mouthed by the supporters of the new International Criminal Court, is "No peace without justice!" In my opinion, apart from its propagandistic emphasis, the slogan reflects a simplistic view of the relationship between justice - understood in a singular fashion as merely judiciary - and world peace. Worse, it denotes a sort of penal fetishism, naively applied to international relationships, almost as if we did not have behind us several centuries of theoretical reflection on the dubious preventive efficacy, both with individuals and in general, of penal afflictions - especially detention - and there were not a growing perplexity as to the rehabilitation value of prison terms.

The enthusiastic supporters of this unprecedented judiciary and penitentiary internationalism - The Hague has built the first super-national prison in history - seem unaware of the conclusions reached by theoretical and sociological research on modern penitentiaries: the underlying significance of these structures continues to be one of retribution and vengeance, following an archaic sacrificial and victimizing logic that modern legal rationalism had sought to remove but which it actually reintroduced under new colors. (10) The 'judiciary globalists' ought to try at least to argue that the pessimistic conclusions as to the effectiveness of penitentiaries in various nations are not pertinent when it comes to the international environment. Without even this minimal effort to come up with an argument, it must be concluded that the new super-national jurisdiction that is about to take effect is not accompanied by any serious penal philosophy, and this reinforces fears that the new institutions will not have any substantial role but to confer, thanks to their halo of impartiality, a surplus of symbolic legitimization to present-day international institutions and the great powers controlling them.

3.5. A minimal, super-national legal system

My fifth and final point, which could be entitled 'a minimal super-national legal system,' rather sums up my remarks. As I have mentioned above, it was Hedley Bull who first advanced the idea that it would be preferable on the international level to settle for a 'minimal political order,' setting aside the ideal of an 'optimal political order.' In so doing, Bull expressed his mistrust of hypotheses of a world authority entrusted, according to the ideology of Western globalists, with the task of guaranteeing not only a stable, universal peace but also distributive justice, economic development, international protection of subjective rights, the ecological equilibrium of the planet, containing demographic growth, and so on. (11) A cosmopolitan authority of this sort, no matter how representative, could not but be intensely interventionist and overbearing, and hence would threaten the integrity and autonomy of civilizations and cultures.

Drawing inspiration from this neo-Grotian and realist option, I would suggest that international law should aim at constituting a 'legal society' able to coordinate the subjects of international politics according to a logic of normative subsidiarity with respect to the competence of state legal systems. That is to say that a minimal amount of super-national power would be yielded to centralized organs, allowing them a minimum of recourse to coercive interventions authorized case by case by the international community at large on the basis of an 'equal sovereignty' among all members.

It is in this sense that I propose the expression 'minimal super-national law': following a federalist logic applied to the relationship between the normative competence of national states and that of super-national organs, this law would leave plenty of room for the functions of domestic jurisdiction, without claiming to replace it or smother it with super-national normative or judiciary organisms. In other words, a 'minimal political system,' if indeed it is to be minimal, should be based on a sort of 'multi-centered regionalism' of international law, rather than on a hierarchical structure that would risk provoking a revolt of the 'peripheries.'

To my mind, this is a highly significant point both for the general reason of respect for diverse cultural, political, and legal traditions, and above all for the defense of subjective rights. In fact, there is no proof so far that the international tutelage of rights - a task that in the European, continental, and Anglo-Saxon tradition is jealously entrusted to the rule of law - can be usefully transferred to super-national organisms. What stands in the way of this prospect is the lack, both in the United Nations and institutions for regional integration such as the European Union, of a constitutional structure of super-national powers that can at all be assimilated to that of a rule of law.

For this reason, it would be contradictory and risky to entrust the protection of human rights to the exclusive, or even prevalent competence of other judiciary organisms than the national ones, even in the hypothesis that the political or judiciary organisms of a nation state might transgress against the rights of citizens. Indeed, it would be an illusion to think, except for irrelevant exceptions, that the protection of basic freedoms can be guaranteed internationally, if this protection is not guaranteed first of all domestically by the liberal and democratic institutions of rule of law, and this line of reasoning can also be applied, in particular, to justice as administered by the ad hoc International Criminal Tribunals.

In any case, a 'minimal super-national law' must not mean inertia in the international community with respect to many problems that have lately taken on a global dimension - starting from the problem of peace - and with respect to which the single states are ill prepared or not prepared at all. Nevertheless, I think it is important to make a clear distinction between the need for legal coordination and close political collaboration among international political subjects and the idea that centering power in super-national organisms can be a valid answer to the problems posed by globalization.

(Translated by Gordon Poole)

Notes

*. From Danilo Zolo, I signori della pace [the peace-lords], Rome: Carocci, 1998.

1. As is well known, it was Hegel who sarcastically highlighted the correspondence between the cosmopolitan proposal made in Zum ewigen Frieden [for everlasting peace] and the institution of the Holy Alliance: "Thus, Kant proposed a league of sovereigns to settle disputes between states, and the Holy Alliance was meant to be an institution more or less of this kind" (v. G.W.F Hegel, Grundlinien der Philosophie des Rechts [1821], ed. by E. Gans, Stuttgart: Friedrich Frommann Verlag, 1964, pp. 434-5, Engl. trans. Elements of the Philosophy of Right, ed. by A.W. Wood, Cambridge: Cambridge University Press, 1991, p. 362). Moreover, it was Hans Morgenthau who wrote that the international government of the United Nations was actually the international government of the Security Council. The Security Council was, so to speak, the Holy Alliance of our era, and its five permanent members made up a Holy Alliance within the Holy Alliance (Politics Among Nations, New York: Knopf, 1960, p. 480).

2. "The story is sometimes told of the man who was lost somewhere in Scotland, and asked a farmer if he could tell him which was the way to Edinburgh. 'Oh sir - the farmer replied - if I were you I shouldn't start from here!' The doctrine that the states system does not provide the best starting-point for the pursuit of the world order has something of this quality" (H. Bull, The Anarchical Society, London: Macmillan, 1977, p. 295).

3. On the theme of the 'reduction of fear', suggested by both Niklas Luhmann and Arnold Gehlen, see my Democracy and Complexity, Cambridge: Polity Press, 1992, especially the third chapter.

4. As is well known, the notion of 'international regimes' was introduced in 1975 by John G. Ruggie in "International responses to technology: concepts and trends", International Organization, 29 (1975), 3, pp. 557-84. For a discussion, see R.O. Keohane, "The demand for international regimes", now in S.D. Krasner (ed.), International Regimes, New York: Cornell University Press, 1983, pp. 141-71; R.O. Keohane, "Lo studio dei regimi internazionali e la tradizione classica nelle relazioni internazionali" [the study of international regimes and the classical tradition of international relations], Rivista italiana di scienza politica, 17 (1987), 3, pp. 349-76. Krasner defines a 'legal regime' as a "set of implicit or explicit principles, norms, rules, and decision-making procedures, around which actors' expectations converge in a given area of international relations. Principles are beliefs of fact, causation, and rectitude. Norms are standards of behaviour defined in terms of rights and obligations. Rules are specific prescriptions or proscriptions for action. Decision- making procedures are prevailing practices for making and implementing collective choice" (see S.D. Krasner, "Structural causes and regime consequences: regimes as intervening variables," now in S.D. Krasner (ed.), International Regimes, New York: Cornell University Press, 1983, p. 2).

5. See my essay "Hans Kelsen: International Peace through International Law", European Journal of International Law, 9 (1998).

6. In a few pages of the third chapter of Cosmopolis, Cambridge: Polity Press, 1997, I seek to confute the main theses expounded by Michael Walzer in Just and Unjust Wars, New York: Basic Books, 1992.

7. See N. Bobbio, Il problema della guerra e le vie della pace [the problem of war and the ways of peace], Bologna: il Mulino, 1984, p. 60.

8. When I use the term 'modern war' here, I do not mean only a war fought with nuclear or quasi-nuclear weapons - e.g., the fuel-air explosives used by the United States in the Gulf War of 1991 - but also any military action employing weapons of mass destruction, which involves necessarily, and hence with foreknowledge, the killing and mutilation of people, the destruction of their goods, and the violation of their basic rights, regardless of any consideration of their individual responsibilities or of the reasons why they find themselves involved in the conflict. From this standpoint, it makes no difference whether they are civilians or military.

9. See H. Bull, The Anarchical Society, cit., p. 89. Bull specifically criticized the selectiveness of these sentences and the asymmetry between the judged and their judgers.

10. On this theme see my essay "Filosofia della pena e istituzioni penitenziarie" [criminal philosophy and penitentiary institutions], Iride, 14 (2001), 32, pp. 47-58; see also the contribution by E. Santoro, Carcere e società liberale [prison and liberal society], Turin: Giappichelli, 1998.

11. See H. Bull, The Anarchical Society, cit., pp. 284-95, 302-5.