The Humanitarian Intervention Discourse: A Debate on the Edges of the Law


The Humanitarian Intervention Discourse: A Debate on the Edges of the Law

Ainsi, la plupart des mauvaises actions des hommes sont venues au-devant d’eux, déguisées sous la forme spécieuse de la nécessité; puis, la mauvaise action commise dans un moment d’exaltation, de crainte, de délire, on voit qu’on aurait pu passer auprès d’elle en l’évitant. Le moyen qu’il eut été bon d’employer, qu’on n’a pas vu, aveugle qu’on était, se présente à vos yeux facile et simple; vous vous dîtes: comment n’ai-je pas fait ceci au lieu de faire cela?*


THE 1999 NATO (North Atlantic Treaty Organisation) intervention in Kosovo was contemporaneously accompanied by a period of stark silence from international lawyers concerning the strict legality of the operation.1 Many of those who proffered comment upon the matter couched their analyses in terms of ‘classical international law’2 or ‘traditional international law’.3 The rationale that brought about this reticence to affirm the legality of NATO’s actions may be easily explained: any such affirmation could be immediately demonstrated to be untrue.

The United Nations Charter, the primary legal instrument concerning the use of force by States, effectively outlaws war. Article 2(4) of the Charter represents a general prohibition upon the use of force by States at international level. There exist two exceptions to this rule, however, namely collective action authorised by the UN Security Council under Chapter VII of the UN Charter, and an inherent right enjoyed by each State, to (individually or collectively) defend itself, described in Article 51 of the Charter. However, humanitarian intervention may not be held to fall within either of these exceptions.

For the purposes of this discussion, I shall define humanitarian intervention as: (1) the threat or use of force; (2) across State borders; (3) by a State or group of States; (4) aimed at preventing or ending gross human rights violations; (5) where such violations are being committed by the violating State against individuals other than the citizens of the intervening State(s);4 (6) without the permission of the State within the territory of which force is applied.

The above corresponds with the definition of JL Holzgrefe and Robert O Keohane,5 but to it, I add one further tenet: (7) the intervention is not authorised by the UN Security Council. This final criterion shall limit our discussion to interventions which are not sanctioned by the Security Council, thereby avoiding confusion between authorised and unauthorised ‘humanitarian’ interventions.

It is clear that this type of operation, since it is not of a self-defensive nature, and since it has not received the prerequisite Security Council go-ahead, is outside of either of the two designated exceptions to Article 2(4). The above, in tandem with the fact that customary international law has not accepted a right of humanitarian intervention as lawful, renders such activities illegal, both under customary international law and Charter law.

However, in the aftermath of the NATO Kosovo operation, condemnation of the intervention as downright illegal was relatively scarce. Instead, a broad range of theories attempting to justify the legality, or at the very least, the justice of the intervention, surfaced. The reasons for this may be traced in the historical and political Zeitgeist and shall be discussed herein.

This work will examine a great deal of the scholarship which has been undertaken attempting to justify humanitarian intervention, particularly since 1999. It will focus on leading authors and the justificatory models and tests which they have proffered. The goal of this work is to argue that all of the scholarly models which have been identified are inadequate to do the job for which they were devised. None of the arguments resolve the conflict which arises between law and justice, and many of them do damage to international law and order. In addition, several theories which in 1999 were utilised to justify NATO’s involvement in Kosovo have since been employed to justify other interventions where the humanitarian motives of the interveners were less apparent, notably the 2003 invasion of Iraq. The tone of this work will necessarily therefore be rather negative, seeking as it does to criticise the broad majority of scholarship in this field. This may be easier to understand if I offer some brief explanation as to the content of the later chapters of this volume. Based upon the stated deficiencies of the various proposed legitimising formulae, I shall undertake a study of whether the third source of international law – that is, general principles (almost completely ignored by scholarship in this field until now) – may hold the possibility of a third exception to the prohibition upon the use of force. It is my position, based upon my research thus far, that such an exception may indeed exist. However, the need for exploration of such a theory is only justified if the models proposed by other authors in this field are faulty. It is my contention that they are, and hence my wish to undertake the critical examination contained in this volume.


Following upon persistent threats and warnings of its imminent resort to the use of force, NATO, on 24 March 1999, commenced air strikes against the Federal Republic of Yugoslavia (FRY). A veritable plethora of reasons were proffered justifying this operation. The NATO operation, given its timing in the wake of the horrors of Srebrenica and Rwanda, was broadly welcomed. Someone at last seemed to be standing up for human rights. However, the legality of the operation was, on the face of it, at least, extremely suspect. This fact presented a quandary. Human rights protection is essentially enshrined within the normative framework of international law. Indeed this protection constitutes one of the three fundamental tenets of international law which underpin international relations.6 Many legal and political commentators found themselves vexed by how to treat an ostensibly illegal military operation at least purportedly undertaken in the name of human rights protection. How could one break international law in the name of international law?

The responses composed to the NATO operation were diverse, between commentators as between intervening States and third party States. Some condemned the operation as downright illegal, while others attempted to excuse this illegality on the basis of an ‘international exception’. Still others drove the debate away from the strict legality of the operation and attempted to assess its ‘legitimacy’, while there were also those who claimed a new norm of customary international law was developing. A great many assessments – some contradictory – were advanced regarding what Kosovo could mean for international law. However, if one thing has clearly emerged from the debate, it is that there is no overall consensus. This is disquieting, in that it leaves the future uncertain.

In this chapter, I shall attempt to shed some light on a selection of the principal component arguments within the humanitarian intervention debate, post-Kosovo. In doing so, I shall attempt to remain within the legal sphere as far as possible. However, it is evident that a full treatment of the debate may not be undertaken without at least a passing reference to historical debate, political comment, and the facts of the Kosovo crisis itself. Furthermore, I shall endeavour to cast a critical eye over many of the arguments adduced, in order to show the relative weaknesses of many of the proffered justifications for the Kosovo operation, and indeed for humanitarian intervention in general. Obviously, due to the breadth of scholarship which events post-Kosovo have generated, it is impossible to be exhaustive here, but I shall treat the opinions of leading commentators and indicate when opinions may be treated as exemplary of broader trends within the discourse.

An exhaustive treatment of the literature in this field would be all but impossible in a work of this length. However, I shall nonetheless endeavour to examine the principal schools of argument in this field through a thorough treatment of their usage by leading authors. The authors chosen for in-depth analysis represent some of the leading exponents of the various theories, though necessarily choices had to be made to include some rather than others for the sake of brevity. In addition, and in order to furnish a more complete picture of the doctrinal debate, I shall furnish brief commentaries via annotations and footnotes on where these authors find concurring and dissenting opinion both within and without their schools of thought.

I shall begin with the convincing argument proposed by, inter alia, Ian Brownlie, that the Kosovo intervention was illegal. Brownlie maintained this position for many years, and his research here is second to none. Thereafter, I shall treat the analysis of Bruno Simma, who epitomised the school of thought that the Kosovo action, although illegal, was only separated by what Simma dubs a thin red line’ from international legality. I shall then refer to the position argued by Antonio Cassese and others, that while humanitarian intervention is illegal, a customary norm justifying such activities may soon evolve, and will compare this with the ‘Solidarist’ position defended by Nicholas Wheeler. I shall then describe a ‘turn to ethics’ amongst international lawyers, who have attempted to employ ethical and political rationales to bolster legal arguments, and shall demonstrate the theoretical repercussions which this process has had upon the humanitarian intervention debate. I shall also examine efforts to legitimise the intervention in a way keeping closer to the strict letter of the law, particularly the analysis of Thomas M Franck, and shall enquire whether the logic of the debate, and particularly the parallels which may be drawn between it and the long-defunct Just War doctrine are dangerous, insofar as they may be employed by the unscrupulous for their own designs. Finally, I shall attempt an in-depth appraisal of the most visible concerted and international effort to provide a new discourse combined with a legal framework for humanitarian intervention post-Kosovo, namely the Responsibility to Protect initiative.


In a memorandum treating the international law aspects of the Kosovo crisis, Ian Brownlie and CJ Apperley present a cogent case that the operation was illegal.7 I shall deal with this treatment in some depth, as it may later be used in order to refute many of the apologist/justificatory theories submitted by other scholars. This article shall represent my major source for the ‘illegality’ position within the debate. It is chosen for such purposes due to its timing – shortly after the Kosovo intervention – and due to its strength of argument, as well as the fact that it typifies those who argue for the illegality of humanitarian intervention. I shall, however, refer to further concurring opinion from Brownlie himself and indeed from other sources in order to demonstrate the strength of this position. The position also represents the ‘standard’ positivist reading of international law that has been prevalent throughout international legal scholarship, seeing the general prohibition of the use of force and its two exceptions as a watertight, closed, regime.

Brownlie and Apperley point to the excuses proffered by the UK Government for military action, couched in: (1) UN Security Council Resolutions 1199 and 1203; (2) the assertion of a legal right ‘to prevent a humanitarian catastrophe’; and (3) the fact that military action might only be undertaken in the event of Belgrade’s non-acquiescence to the Contact Group’s demands (the Rambouillet ‘Agreement’).

Further, in the British House of Commons, a customary right to humanitarian intervention was espoused, although as the authors point out, none exists. A tendency on the part of the British Government to avoid reference to any exact legal sources (such as treaties, the UN Charter, etc) was further remarkable.

Ab initio, Brownlie and Apperley appear palpably hostile to the idea of Kosovo finding legal justification. They note that the novel phrase’ ‘humanitarian catastrophe’ is more often employed than the traditional nomenclature of humanitarian intervention,8 and refer to evidence that the actual views, as opposed to those publicly expressed, by NATO legal advisors, were unfavourable to action.9 However, this evidence, although it may point to disingenuous comportment on the part of the aforementioned actors, is simply not pertinent to a legal argument, and I struggle to grasp why Brownlie and Apperley mention it. After all, following from the Eastern Greenland case of 1933, government officials are estopped from denying the value of unilateral declarations when they have given others cause to believe they are to be relied upon.10 Therefore, the fact that such declarations are not necessarily truthfully representative of the sentiments of the officials which impart them is legally irrelevant.11

The UN Charter, Article 2(4) prescribes that:

All Members shall refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.

Brownlie and Apperley point out that the travaux préparatoires indicate that this phrase ought to be afforded a strict interpretation, and that the second part of the phrase – against the territorial integrity and political independence of any state – may not be construed to refer to a particular threshold which must be reached by threats and uses of force in order to render them illegal.12 As Anne-Sophie Massa has pointed out, far from restricting the scope of the prohibition, these words ‘were inserted as a guarantee for small states to reinforce the impermissible character of recourses to force against a state’, a pleonasm of sorts.13 Although not referred to by Brownlie, this assertion is further evidenced by the rejection of a more supple interpretation of this phrase by the International Court of Justice (ICJ) in the Corfu Channel case (1949).14 Hersch Lauterpacht adopts a similar attitude.15 This position has been confirmed 25 years after the Charter’s inception in the ‘Friendly Relations’ Declaration,16 which reaffirmed the principle ‘concerning the duty not to intervene in matters within the domestic jurisdiction of any state, in accordance with the Charter’. Therefore, the ‘qualifying clause’ was intended as an extra bulwark against force, rather than ‘the key to the padlock’. Indeed, as shown from the ICJ’s Nicaragua judgment, a restrictive interpretation will be applied to any of the narrow exceptions which do exist to the prohibition upon the use of force, and thus both under customary international law and the law of the UN Charter, States do not have a (unilateral) right of ‘armed response to acts which do not constitute an “armed attack”’.17 Therefore, any attempts at a broad teleological interpretation of Article 2(4) are, per Brownlie and Apperley, misguided.

Regarding the legal status of humanitarian intervention, Brownlie and Apperley state that no evidence exists of such a legal right and that such a right cannot be compatible with the UN Charter. Indeed, in a number of earlier texts, Brownlie had noted the fact that while humanitarian intervention had not induced express condemnation within the Covenant of the League of Nations, nor in the Briand–Kellogg pact (and therefore its status inter-bellum was more discutable) it remained ‘highly doubtful’ that such a prohibition had survived the general interdiction regarding armed force enunciated in the UN Charter.18 Further, as noted by Francioni, the Draft Articles on the Responsibility of States for Internationally Wrongful Acts stipulate that State responses to grave violations of international obligations owed erga omnes – including human rights obligations – must be lawful and consistent with the UN Charter.19

The argument that humanitarian intervention is without legal basis finds support from a wide selection of other authors. Akehurst, for example, is unequivocal in stating that the evidence indicates that humanitarian intervention is ‘incompatible’ with international law.20 Oscar Schachter voices a similar view, citing the probable rationale behind this:

[G]overnments by and large (and most jurists) would not assert a right to forcible intervention to protect the nationals of another country from the atrocities carried out in that country …

The reluctance of governments to legitimise foreign invasion in the interest of humanitarianism is understandable in the light of past abuses by powerful states. States strong enough to intervene and sufficiently interested in doing so tend to have political motives. They have a strong temptation to impose a political solution in their own national interest.21

However, the rationale advanced by Schachter does not stand up to scrutiny. As Rosalyn Higgins J has pointed out:

Many writers do argue against the lawfulness of humanitarian intervention today. They make much of the fact that in the past the right has been abused. It undoubtedly has. But then so have there been countless abusive claims of the right to self-defence. This does not lead us to say that there should be no right of self-defence today …

We delude ourselves if we think that the role of norms is to remove the possibility of abusive claims ever being made. The role of norms is the achievement of values for the common good …

Claims which may in very restricted exceptional circumstances be regarded as lawful should not a priori be disallowed because on occasion they may be unjustly invoked.22

Schachter’s view therefore is right insofar as most States view humanitarian intervention as illegal, but wrong inasmuch as he justifies their position with reference to the possible abuse of such a right. The illegality of humanitarian intervention was indeed upheld by the British Government, precedent to the Kosovo intervention, as evidenced by a 1986 Foreign Office Policy Document:

[T]he best case that can be made in support of humanitarian intervention is that it cannot be said to be unambiguously illegal … the overwhelming majority of contemporary legal opinion comes down against the existence of a right of humanitarian intervention.23

The above position perhaps explains the British Government’s aforementioned reticence toward adopting common parlance and dubbing the Kosovo mission ‘humanitarian intervention’, instead choosing to refer to a ‘humanitarian catastrophe’. To do otherwise would have been to admit the illegality of the operation, as previously affirmed by their own Foreign Ministry. Coining an alternative nomenclature, as it were, at least afforded the UK the opportunity to avoid seeming openly hypocritical, and possibly to avoid appearing to be openly breaking international law.24 In addition, the use of the word ‘catastrophe’, an emotive term to the ears of the masses, suggests urgency and a sense of exceptionality that makes immediate and effective action seem warranted and even indispensable to the public at large.

Based upon the above, in addition to a wide selection of further commentaries, Brownlie and Apperley are able to conclude that: (1) the careful wording of Article 2(4) was precisely intended to allay fears of smaller States in light of conduct in the inter-war period (it having been a favourite excuse of Hitler to put forward mistreatment of minorities as a rationale for military invasion). (2) The overwhelming body of legal opinion considers humanitarian intervention illegal. The authors cite 18 separate authorities, spanning 12 nationalities, and including the views of three former Presidents of the ICJ (Schwebel, Jiminez de Arechaga and Ruda). (3) Only a small minority of international lawyers have adopted the position that humanitarian intervention may be considered lawful (as distinct from excusable).

The views of Brownlie and Apperley find concurring opinion elsewhere.25 JS Watson and Alfred Rubin, for example, both condemn humanitarian intervention (as well as attempts to enforce human rights norms through the operation of international criminal tribunals) as illegal acts masquerading under the guise of legality.26 Scholtz stresses the importance of Article 2(4), without which, he argues, international relations would be governed by the use of force, and worries that illegal humanitarian intervention may pose a risk to it.27 O’Connell is quite unequivocal on this point, stating that There really is no normative basis for non-compliance’.28

However, not all international lawyers are in accord with Brownlie and Apperley’s views. It is, I feel, particularly germane to note that many of the sources quoted in support of their position predate the invasion of Kosovo.29 Only one cited source, that of Bruno Simma30 is contemporaneous with the initial threats of force (undoubtedly as illegal as the use of force itself) made by NATO Member States against the FRY, and of the cited sources, Simma’s is perhaps the most equivocal and least convincing, holding that ‘only a thin red line separates NATO’s action on Kosovo from international legality’.31 The position adopted by Simma represented a portent of things to come. A move away from a strict legal analysis was in the offing in the wake of the Kosovo operation. Brownlie was to find his positivistic posture was no longer en vogue.

I shall now discuss how in the aftermath of the NATO intervention in Kosovo, a new discourse, or rather several new discourses, developed regarding the concept of humanitarian intervention.


Professor Bruno Simma, while admitting the illegality of humanitarian intervention, adds a caveat to this assessment, making the seemingly innocuous proposal that the NATO operation in Kosovo was not far away from being legal, and therefore ‘almost legal’, as it were. Simma is the leading proponent of such a position, and hence, his shall be the principal analysis referred to in this section. Simma’s contention, while superficially innocent looking, was to serve as a starting point for many scholars who used this pretended ‘proximity’ to legality to introduce philosophical theses to excuse or legitimise the Kosovo intervention. I shall examine these efforts later, but for the moment it is germane to assess the import of Simma’s own position, which has itself been subsequently mirrored by other scholars.

Simma begins by reminding us that the obligation weighing upon States to respect and protect human rights is owed erga omnes.32 Hence, in case of material breaches of such obligations, any other State has locus standi to consider itself injured and may therefore resort to countermeasures (not involving force) against the perpetrator. The fact that no forcible countermeasures may be adopted in such cases is further apparent due to the comprehensive nature of Article 2(4), which Simma admits is intended to be ‘watertight’,33 and that such countermeasures would not fall under either of the exceptions to the general prohibition allowed under the Charter.

Simma further states that Article 2(4) represents a peremptory norm of international law, having acquired ius cogens status. It can therefore only be modified by another norm of a similar status. Surprisingly, having made this observation, Simma fails to expand upon it. It seems unusual that when wishing to express a contention that the NATO operation in Kosovo was proximate to legality that one would not observe that at least a minimum modicum of human rights norms are themselves ius cogens, and may therefore themselves conceivably alter other ius cogens norms, such as Article 2(4).34 However, Simma does note the ICJ position, albeit one predating the end of the Cold War, concerning the relationship between the two: ‘the use of force could not be the appropriate method to monitor or ensure … respect [for human rights]’.35

However, Simma argues that humanitarian interventions undertaken in the spirit of preventing recurrence of past atrocities ought to meet a friendlier reaction. Such interventions may breach international law. ‘But such a general statement cannot be the last word’.36 Rather, argues Simma, one must weigh concrete circumstances against the relative illegality of any operation, taking into account the efforts of the interveners to get ‘as close to the law’ as possible. Per Simma, such analyses will influence both legal and moral judgments. It is on the basis of this test that Simma contends that only a thin red line separates the Kosovo intervention from legality. A similar position, citing a ‘trend’, epitomised by the Kosovo intervention, which pushes humanitarian intervention into the domain of international legitimacy and ‘close’ to international legality is proposed by Paolo Picone.37

Simma notes Security Council Resolutions 1160 and 1199 of 1998, adopted under Chapter VII, which determined that ‘Kosovo’ represented a threat to peace and security in the region and which promised consideration of additional measures if a political solution was not found. The knowledge that Russia would veto any Resolution authorising force precluded any pursuant actions. ‘At this point’, Simma tells us, ‘NATO took over, as it were … The principal legal basis for such action was to be the concept of “humanitarian intervention”, linked as closely as possible under the circumstances to the UN Charter in order to further gain legitimacy’.38 This is a curious contention, outlining the principal legal basis for an operation which Simma himself admits was illegal, and bringing to mind Truepenny CJ’s contradictory dictum on adherence to law.39 Further, the confusion of the terms ‘legality’ and ‘legitimacy’ is not a technique limited to Simma. In fact, Simma borrows this terminology from NATO Secretary-General Javier Solana, who determined that the concrete circumstances – the non-compliance of the Federal Republic of Yugoslavia (FRY) with Security Council Resolutions; the impending humanitarian catastrophe; the impossibility of procuring Security Council authorisation to use force; and Council’s determination that a threat to the peace existed – constituted ‘legitimate grounds for the Alliance to threaten, and if necessary, to use force’.40 Of course, it is worth mentioning the fact that ‘legitimacy’ is a far more flexible and far less precise notion than law, precisely because there is no undisputed authority mandated to evaluate legitimacy, and no agreed procedure for doing so.41 Perhaps it is for this reason that it was to become a key term in the humanitarian intervention debate.

According to Simma, citing the Resolution on Recasting Euro-Atlantic Security, adopted by the North Atlantic (Parliamentary) Assembly,42 NATO sources acquired a clear message from the Kosovo crisis, namely that if it transpires that Security Council authorisation for future NATO non-self-defensive missions involving armed force is unobtainable, NATO must still be able to push ahead with such enforcement. Simma parallels the language used in the Resolution with that contained in the ‘Uniting for Peace’ UN General Assembly Resolution of 3 November 1970, but points out that ‘Uniting for Peace’ was an effort to fill gaps in the Charter during the darkest era of the Cold War. He admits ‘such ersatz constructs might have had a certain legitimacy at the time. But today, things are different: since the end of the Cold War, the Security Council is functioning precisely in the manner envisaged in 1945’.43 The language of legitimacy creeps into Simma’s rhetoric again here, and since he fails to define the concept, we are left to wonder as to what he means. However, as to the legality of the operation, we are left in no doubt. Simma affirms the primacy of the UN Charter over the Washington Treaty, adding that the ‘status of the Charter should not be prejudiced by NATO’,44 (except, of course, by the circumlocution tactics that he himself proposes).

Nonetheless, it is evident that Simma reaches this conclusion with a heavy heart. His ‘legitimacy’ criterion, based loosely around that proposed by Javier Solana is never completely explained. Perhaps it is as well that this is so, as it can only clash with the law. Simma finishes by dubbing Kosovo a ‘hard case’, which it undoubtedly is, which involves terrible dilemmas, which may ‘appear to leave no choice but to act outside the law’.45 He argues for a singular interpretation of the circumstances that led to the Kosovo intervention, and that such actions should be examined ad hoc and case-by-case, in order to reduce their power to erode the tenets of international law. This may be done by indicating the concrete circumstances which led to the decision, and in this regard, he commends NATO for proffering a convincing case. Resort to illegality may be permissible in such circumstances, but not as an instrument of general policy. It is for this reason that only a thin red line separates the Kosovo operation from legality.

Simma’s idea of relative legality, or near-legality represents a valiant attempt to wrestle with the unyielding prohibition of Article 2(4).46 However, problems remain with Simma’s thesis. For one thing, as Francesco Francioni has pointed out, describing such a massive use of force as the two months of bombardment which took place in Kosovo as a ‘minor’ breach is hardly accurate.47 While Simma does finish by avowing the legal worth of Article 2(4), which serves to render humanitarian intervention illegal, his usage of the legal/legitimate dichotomy and his avowal that interventions such as that undertaken in Kosovo may be excusable so long as not undertaken as an instrument of general policy is curious. If anything, the NATO position makes a great deal more sense. At least the NATO position, while also couched in illegality, creates a set of concrete guidelines for circumstances under which NATO may act even in the absence of Security Council authorisation. It is plain that Simma does not like this idea, as it weakens international law,48 but surely, if ‘another Kosovo’ rears its head tomorrow, NATO should act again. Nehal Bhuta echoes this proposition, stating that ‘one way of understanding humanitarian intervention is not so much as a “right”, but as illegal conduct which is tolerated by the international community in extreme circumstances’.49 What is the sense in an ad hoc decision, based upon ‘concrete circumstances’, if similar circumstances will always lead to a similar result (as would be the case for my hypothetical ‘another Kosovo’) then surely we are in the process of creating a rule-based system.50 Further, considering the role of State practice in developing international law, consistent similar action without broad condemnation must surely lead to the gradual emergence of a legal rule. As Anthea Roberts has pointed out, Simma’s approach, shifting the focus away from legality can undermine the coherence of international law, and raises the risk of any ‘legitimacy test’ being manipulated.51

If Simma’s ideas will inevitably lead to a rule-based system (inside or outside the strictures of international law), it merely rests to design such rules. Antonio Cassese is amongst those who have attempted to do so, and I shall deal with his contribution, devised as a direct response to the arguments of Professor Simma.


Cassese agrees with Simma that the NATO intervention falls outside the scope of the UN Charter and is, ergo, illegal, but goes on to note that ‘My agreement with Simma ends, however, when he contends that “only a thin red line separates NATO’s action in Kosovo from international legality’’’.52 Per Cassese, the breach cannot be termed minor. Also, the simple fact that the Security Council had determined that a threat to the peace existed does not constitute a legal ground for an attack on a foreign State. Cassese seems therefore to adopt a trenchant position that such breaches are not merely the ‘borderline’ affairs that Simma attempts to dub them. Furthermore, Cassese notes that ‘once a group of powerful states has realised that it can freely escape the strictures of the UN Charter and resort to force without any censure, except for that of public opinion, a Pandora’s box may be opened’.53

It seems from the above formulation that Cassese is not prepared to adopt the same apologist undertone to his illegality argument that Simma proposes before him. However, after the bullish rhetoric initially apparent in his article, Cassese goes on to adopt a vastly more equivocal position.

Cassese affirms that in the current international community framework, three sets of values underpin inter-State relations, these being peace, human rights, and self-determination. Cassese expands upon this, opining that when tension arises between these three core values, peace must be given precedence. Such an affirmation – regardless of the fact that it is over-simplistic in that it ignores sovereignty – lends further credence to the contention that humanitarian intervention has no legal basis. Human rights, we are told, must ceder place to peace.

Cassese then proposes that while it is true that it is difficult to sit idly by while thousands are massacred, this shall only serve to justify the Kosovo intervention from an ethical viewpoint, not a legal one. This represents a precursor of a large-scale ‘turn to ethics’ which has been treated by Martti Koskenniemi, and to which I shall myself devote attention in due course. However, one may at this point make the preliminary observation that in a legal discussion, it is of dubious merit to have recourse to moral and ethical considerations in order to change one’s perceptions of the legal validity of a legal proposition. Above all, the question must be posed as to which barometer of morality we should use, as no morality is all-ensconcing and objective.

Despite his moralising, however, Cassese still admits that he ‘cannot avoid observing in the same breath that this moral action is contrary to current international law’ (emphasis added).54 The use of the word ‘current’ is paralleled by a trend amongst international lawyers to avoid discussing the strict legality of the Kosovo operation, instead prefixing their frame of reference as ‘common perceptions of international law’, ‘traditional international law’, or, as Cassese would have it, ‘current international law’.55 The explanations for this trend are plain, allowing the discourse to also take extra-legal factors into account, but such an approach serves only to devalue international law, insofar as it plainly implies that one may take a broad range of interpretations thereof. Surely ‘traditional’ and ‘modern’ international law must be different. If not, why bother affixing the prefix in the first place?

Nonetheless, Cassese broadly concedes the illegality of the NATO operation. He goes on to contend, however, ‘that as legal scholars, we must stretch our minds further and ask ourselves two questions’.56 These are: (1) whether the action was partially rooted in contemporary trends of the international community; and (2) whether parameters were set in this operation that might lead to gradual legitimisation of forcible humanitarian countermeasures by a group of States outside any Security Council authorisation.

In order to answer the above questions, Cassese commences by examining the ‘basic premise or root’ of the NATO intervention. He makes six points: first, it is a ‘truism’ that human rights are no longer the sole concern of the State where infringed. Secondly, obligations to respect human rights are owed erga omnes. Any State may take non-forcible steps to ensure respect. Thirdly, an idea is emerging that large-scale atrocities may give rise to aggravated State responsibility, entailing increased countermeasures. Fourthly, the international community is increasingly intervening through international bodies when human rights are in jeopardy. Fifthly, peaceful dispute settlement measures are increasingly regarded as crucial, and the idea is firmly rooted that these must take precedence. Finally, it has been claimed by some NGOs and governmental officials that under exceptional circumstances, ‘where atrocities reach such a large scale as to shock the conscience of all human beings and indeed jeopardise international stability, forcible protection of human rights may need to outweigh the necessity to avoid friction and armed conflict’.57

As Cassese puts it, ‘positive peace’, that is the realisation of justice, should prevail over ‘negative peace’, that is the absence of armed conflict.58 Anne-Sophie Massa is fervently against such a contention, arguing that the substantive provisions of Charter law clearly privilege peace over justice, and that the travaux préparatoires demonstrate that this was a conscious choice, fuelled indeed to some degree by a belief that allowing a right of humanitarian intervention might not be the best way to protect human rights, owing to the gap that this would create in the prohibition upon the use of force and the inevitable deaths and human rights abuses which would result from the further wars generated by this exception.59 This rather sensible conclusion is shared by other commentators.60

It is clear, then, that ‘negative peace’, as Cassese dubs it, represents the UN Charter position (the primacy of peaceful international relations over subjective justice), and is therefore the international legal position. An appeal to a ‘positive peace’ based on justice represents the effective transposition of the old justice versus law conundrum – or the natural law (represented here by ‘positive peace’) – versus positivism (represented here by ‘negative peace’, that is the UN Charter) debate – onto the international sphere. Cassese’s formula is all very superficially attractive, but the problem is, it has no legal basis. Cassese’s ‘negative peace’ ought to properly be called peace simpliciter. His ‘positive peace’ is effectively a euphemism for effective (and presumably forcible) enforcement of human rights.61 This would represent an about-face of Cassese’s earlier ‘values’ formula, with human rights taking precedence over peace. Giving the same concepts different names does not mean they have different levels of value. Either peace is paramount or it is not, tout court.

Cassese is not alone in employing such a formula, however. Kok-Chor Tan notes with reference to liberal cosmopolitan views that ‘the more demanding a conception of global justice is with respect to how states treat their own citizens, the more readily a rationale is provided for states to intervene against each other in the name of upholding justice’.62

Furthermore, it is questionable whether all of Cassese’s six points stand up to in-depth scrutiny. It is perhaps telling that concerning the increasing trend of the international community to intervene via international bodies where human rights are in jeopardy, Cassese furnishes a generous list of examples, whereas for points three and six, which cite the emergence of opinion trends, he fails to furnish any authority whatsoever to buttress his argument. I do not for a moment contend that none exists, but rather wish to highlight the fact that Cassese deliberately chooses not to use what authority there is in this area, perhaps due to the fact that the arguments evinced by such authority may not be convincing.

In any case, ‘Based on these nascent trends in the world community’, Cassese contends that under strict conditions, resort to armed force may become justified absent Security Council justification (although this has yet to happen).63 These strict conditions are: (1) gross egregious human rights breaches, thousands of innocents killed and crimes against humanity; (2) the central authorities are unwilling/unable to put an end to the abuses; (3) the Security Council is unable to act due to discord or use of a veto; (4) all peaceful avenues which may be explored consistent with the urgency of the situation have been exhausted; (5) a group of States (not one power) decides to halt the atrocities with the support or non-opposition of a majority of UN members; and (5) armed force is used exclusively to stop the atrocities.

Cassese’s test, while substantially based upon ‘nascent trends’ and other criteria he fails to adequately justify, contains some elements which are commendable, including the ‘last resort’ condition, meaning that such a test may only be contemplated when the usual legal avenues – namely Security Council action – have been effectively exhausted. It is perhaps worth mentioning here that my own test for legal humanitarian intervention based upon equity and general principles of law, which will be set out in detail in chapter five, contains many similar or analogous elements to that of Cassese. It is not with the logic of his test that I wish to quibble, therefore, but rather the fact that it is insufficiently justified, and is effectively grounded in legal nothingness.

Cassese contends that Kosovo broadly fits into the criteria proposed by his test. This is hardly surprising given that the embryonic customary norm to which he alludes is itself a product of the Kosovo operation. However, the intervention remains unlawful, as no new customary norm has yet developed. Such a customary rule would constitute an exception to the UN Charter system of collective enforcement, similar to that encapsulated in Article 51.64 This formulation is echoed by Olivier Corten, who, unlike Cassese argues for a straightforward and orthodox reading of the UN Charter. Corten, however, admits that Kosovo and subsequent events may yet serve as precedents which could engender a change in customary international law, effectively forming a third exception to Article 2(4) in the future. Per Corten, ‘Doit-on … conclure que le droit international n’est pas susceptible de s’adapter aux évolutions de la société dans laquelle il s’inscrit? Certainement pas’.65

Cassese refines his stance in a ‘follow-up’ article published in the same edition of the European Journal of International Law,66 in which he effectively admits to having been premature in drafting parameters for the emergence of a new customary norm, admitting it may be more sensible to await repetition of such actions under the same conditions and exigencies. However, a customary rule of international law also requires a second element in addition to conduct – the mens rea equivalent of statist international law – opinio iuris, that is, the belief that the conduct was legal. To this, Cassese devotes renewed attention.

Cassese opines that States have stopped short of viewing the Kosovo action as lawful per se. However, a phenomenon known as opinio necessitatis has been widespread and appears to be crystallising. Cassese makes reference to the relative paucity of persistent objectors, and the overwhelming feeling in the international community that while the intervention itself was not legal, it was felt to be ‘justified’. The idea that the action was necessary and proper does not yet render it legal, however. Cassese holds that it is premature to speak of the emergence of a customary rule as the element of usus or diuturnitas is clearly lacking.67 Cassese posits that while there may indeed be instances where a single episode of some magnitude combined with the reactions of other States may suffice to form an ‘instant custom’ (the Truman Declaration and the laws of Outer Space are perhaps prime examples);68 more is required in the case of Kosovo.69


Cassese’s position is broadly reflective of one of the major trends of opinion concerning Kosovo. There seems to be little prospect of terming humanitarian intervention legal in the short term, and yet, most feel reticent about condemning those who endeavoured to do what they thought was right. As Habermas put it, this was ‘A war on the border between legality and morality’.70 How far to move away from the law in this regard in order to achieve the desired result was, however, a hotly debated topic, and one which brought international lawyers into competition with political scientists and ethical philosophers. Nicholas J Wheeler is at the forefront of those from outside the legal discipline who have exercised an influence upon the debate and it is to his influential ideas that I now turn.

Wheeler, in his acclaimed work, Saving Strangers, puts forward what he dubs ‘a Solidarist theory of humanitarian intervention’. Wheeler bases his argument upon the notion that the United Nations legal model governing the use of force is manifestly inadequate to deal with the issue of humanitarian intervention, and places the debate firmly into the domain of political theory, founding his thesis (again) upon notions of international legitimacy. Rejecting the classic dictum of EH Carr that notions of international morality or legitimacy are merely manifestations of the interests of dominant groups of nations at a particular time,71 Wheeler posits that the ‘key point of notions of international legitimacy is that they are not within the control of individual agents’.72 This is perhaps a good point, but Wheeler fails to take stock of the fact that the logical consequence of such a statement is that measuring humanitarian intervention against the yardstick of legitimacy rather than that of legality gives every actor the right to shape his yardstick differently.73

Per Wheeler, humanitarian intervention exposes the conflict between order – that is, international law – and justice at its starkest. He adjudges that the Charter’s prioritisation of these two values must be inverted, and hence, sees fit to move outside the realm of law.

Wheeler looks approvingly upon Michael Walzer’s definition of humanitarian intervention as a justified response with a reasonable prospect of success to acts that shock the moral conscience of mankind.74 He echoes Walzer’s sentiment that States should be denied the protection of sovereignty in the extraordinary circumstances where their governments are guilty of crimes against humanity.75 However, it is not without hesitation that Wheeler decides to depart from international law, citing Hedley Bull’s assertion that any institutional order which is provided by international law forms a necessary condition for the protection and promotion of individual well-being.76 He is highly cognisant of the fact that international law (as it stands) is in and of itself a bulwark against human rights atrocities. Nonetheless, as regards the use of force by States, Wheeler is of the view that the problem of abuse of Article 51 of the Charter confirms the Realist school of international relations’ assertion that talk is cheap, a public disguise which masks the true rationales for States’ actions, which they will find a way to accomplish in any case.77 In this way, he is able to dismiss Franck and Rodley’s contention that adding another ‘escape clause’ covering humanitarian intervention alongside that of Article 51 will only increase opportunities for the unscrupulous to exploit.78

Although clearly concerned with balancing order and justice, Wheeler is convinced that selfish State interests will prevail over notions of altruism in humanitarian emergencies, and therefore rejects the assertion that any test for the legality or legitimacy of humanitarian intervention must rest upon the primacy of humanitarian motives for said intervention. Wheeler takes cognisance of the Realist objection that States have no business risking their soldiers’ lives to save strangers, but notes that pushed to its logical extreme, ‘the implication of the statist paradigm is that governments should not risk the life of even one soldier to save hundreds of thousands, or even millions of non-nationals’.79 Perhaps this may be dubbed a reductio ad absurdum, but his sentiments find favour amongst other commentators, especially relating to the relative chances of intervention by democratic polities.80 It is also worth noting the comments of Walden Bello in this regard, who remains highly sceptical about the purportedly humanitarian rationale underlying the Kosovo operation. Per Bello, while the humanitarian rationale was undoubtedly the purpose of some advocates of the Kosovo intervention, particularly those in the academic community and those holding positions in human rights NGOs, NATO’s execution of the operation itself did not prioritise such concerns and merely furthered Washington’s geopolitical designs (as well as, to some extent, the fear that refugee flows might destabilise other European States). ‘The lasting result of the Kosovo air war was not a stable and secure network of Balkan States but NATO expansion’.81 This sceptical position as regards the position of human rights in NATO’s planning is shared by Mient Jan Faber,82 and is borne out by the comments of one NATO spokesman, who, paraphrasing Otto von Bismarck,83 admitted that he did not consider the arrests of senior commanders in Bosnia-Herzegovina for war crimes to be ‘worth the blood of one NATO soldier’.84

Therefore, per Wheeler, since there is virtually no chance of States acting entirely unselfishly, a test for the legitimacy of humanitarian intervention must rest upon other criteria, these being: (1) just cause, constituting a ‘supreme humanitarian emergency’; (2) that the intervention be undertaken as the last possible resort; (3) that the means used be proportionate as regards the ends to be achieved; and (4) that there be a high probability of a positive humanitarian outcome.

Leaving aside the clumsy use of the superlative (‘supreme humanitarian emergency’), Wheeler’s test is primarily concerned with practicality rather than normativity, and in this respect makes for interesting reading. It certainly addresses Anthea Roberts’ concerns that in legal discourse surrounding humanitarian intervention there has been ‘a consistent failure to address the question of the methods used in such interventions’.85 This would seem logical, since the separation between ius ad bellum and ius in bello is hardly a new phenomenon. Wheeler’s is not the only model which calls this separation into question, but it is worth recalling that the parity and equality of the parties in the conflict is hugely undermined if they are not governed by the same norms of international humanitarian law, and if one side must accept additional obligations in this regard in order to legitimise their conduct. Were Wheeler to have his way, such conduct would even serve to render humanitarian interventions legal as well as legitimate. Disagreeing with Franck and Rodley’s assertion that humanitarian intervention, while morally permissible, should not be legalised,86 Wheeler posits that accepting such moral incompatibility with the law fatally weakens the international legal regime. As Verwey has pointed out, such a position ‘would imply the recognition – no more, no less – that international law is incapable of ensuring respect for socially indispensible standards of morality’.87 As a result, according to Wheeler, human rights should rank alongside peace and security in the hierarchy of UN Charter principles. In adopting this ‘counter-restrictionist’ viewpoint, he posits that the logical consequence should be the eventual recognition in international law of a unilateral right of humanitarian intervention.88

Wheeler’s assertion of the position of human rights alongside international peace and security in the Charter purposes is undermined somewhat by an examination of the travaux préparatoires. These make it clear that the Charter could have potentially allowed for sanctions for gross human rights violations, but deliberately chose not to do so.89 Therefore Wheeler’s argument that human rights should or ought to be accorded hierarchical parity with international peace and security takes on an aspirational tone, the call of one arguing for future law reform, rather than one arguing for a new, so-called ‘counter-restrictionist’ interpretation of the positive law at hand. Further, I am not convinced by his argument that States will voice whatever justification they can to act in a way which suits their selfish interests and that opening the door to a possible legal right of humanitarian intervention would not herald a new series of abuses of such a right. One need only examine the justifications proffered by the Bush administration in retrospective defence of the Iraq war to see how, even over a short space of time, a vague notion of legitimate intervention may be stretched out of all proportion. A new legal casus belli that does not require primacy of humanitarian motives could be extremely dangerous in the hands of unscrupulous governments, a point which has been raised by Noam Chomsky and to which I shall return shortly.90


Simma, Cassese and Wheeler are indicative of the broad trends within international scholarship in the years following 1999. Dissatisfaction with Charter law was palpable. The differences in approaches reflected different tacks on how to deal with the difficulties created by the Kosovo conundrum, but all resorted to justificatory techniques based upon a moral examination of an apparently illegal act.

The essential question here was: whether the legally prudent option (abstention from the use of force), or the option which involved illegal means (humanitarian intervention) was the ‘right’ choice.

Of course, moral philosophers will be quick to highlight that there exists no objective ‘right’. However, certain situations provoke strong feeling in the majority of the populace, including international lawyers. The efforts to justify a plainly illegal action provoked a ‘turn to ethics’, which has been stylishly discussed by Martti Koskenniemi.91 The title of Koskenniemi’s article, ‘The Lady Doth Protest Too Much’, is an allusion to William Shakespeare’s Hamlet (1601), to which Koskenniemi repeatedly refers. The line in question describes the attempts of a piece of staged theatre within the play to prick the conscience of its audience in order to engender a desired response from the murderous King Claudius – ‘the play’s the thing, wherein I’ll catch the conscience of the King’. However, Koskenniemi’s choice of Hamlet as a descriptive analogy is useful on more than this level. Indeed the central theme of the play – whether to continue to tolerate an injustice or to take action although this may be dangerous and plainly illegal – mirrors perfectly the Kosovo conundrum.

The essential question here was: whether the legally prudent option (abstention from the use of force), or the option which involved illegal means (killing the murderous King) was the ‘right’ choice.

To be, or not to be, – that is the question: –
Whether ‘tis nobler in the mind to suffer
The slings and arrows of outrageous fortune,
Or to take arms against a sea of troubles,
And by opposing end them?92

As Freud has noted, throughout Hamlet, ‘the play is built up on Hamlet’s hesitations over fulfilling the task of revenge that is assigned to him’.93 However, the role that Freud holds to be assigned to Hamlet, that of an avenger – or at the very least, of someone who can bring a murderer to justice – is not the role of a Crown Prince of Denmark, but rather that of magistrates and police. It is the audience and its moral conscience that dictate that someone must do something, since Claudius appears invulnerable to official powers. This person seems to be Hamlet. So too with humanitarian intervention. Since a deadlocked Security Council is unable to act to quell an unstable situation in Kosovo, it falls to NATO. Why Hamlet? Why NATO? It seems that the simple answer is because they can. ‘Who can, should’, as Michael Walzer puts it.94

However, such reasoning plainly takes us into a sphere, which, although doubtless interesting, lies outside the law. Reconciling moral intuitions with professional competences has therefore been a difficult endeavour for international lawyers. Koskenniemi makes reference to this, in addition to the fact that reconciling international law with cold political realities, particularly involving the use of force. Indeed, Koskenniemi opines, in the Nuclear Weapons Opinion (1996), the ICJ came very close to conceding that no law could govern self-defence when the State’s very existence is at stake.95 This has led to disputes over whether a ‘closing rule’ may be held to exist in international law, and whether there may be instances that are so far outside positive international law that the law may simply be indifferent to them.96

Koskenniemi is firmly in favour of such a ‘closing rule’ as regards the use of force, and is blunt in stating that ‘NATO was either entitled to bomb Serbia or it was not. Tertium non datur. Surely it is an essential part of the Rule of Law that society contains no corner of outside-the-law?’ Yet he admits shortly thereafter that ‘most lawyers – including myself – have taken the ambivalent position that it was both formally illegal and morally necessary … Kosovo has invited international lawyers to throw away dry professionalism and imagine themselves as moral agents on a mission civilicatrice’.97

Koskenniemi outlines an intricate and relatively cogent process, in eight steps, through which international lawyers are transformed into moralists by the logic of the argument for humanitarian intervention, passing from (1) formal law stricto sensu, to (2) formal law lato sensu, to (3) instrumentalism (international law being an activity oriented toward a particular human purpose), to (4) utilitarianism, to (5) rights as trumps (per Dworkin), to (6) legislative discourse, to (7) law as procedure, and finally to (8) the turn to ethics, as Koskenniemi brands it.

To recapitulate the full logic of Koskenniemi’s discourse would be over-burdensome in a work of this length. In addition, I have a problem with it. The analogies with Hamlet which I have attempted above were inspired by parallels drawn by Koskenniemi himself. The central lesson of the play is that one should tackle problems, even if the consequences entailed in doing so may be difficult to countenance. If not, and if one procrastinates, so much the worse. One does not need to be a literary critic to garner this lesson from Hamlet. Similarly, one does not need to be an international lawyer to feel one’s conscience pricked by the Kosovo crisis and to decide that someone must do something to ameliorate the situation, whatever it takes. Koskenniemi’s eight steps are effectively unnecessary. The turn to ethics occurred due to an ‘instinctive’ reaction from States and commentators and a decision that in this decision, justice was paramount over law and human rights needed to be upheld by force. Koskenniemi’s reasoning reads like a judicial opinion justifying a verdict which the judge reached instinctively. However, if one treats it as such, it becomes vastly more useful, demonstrating why so many theories on the Kosovo intervention read differently and yet reach the same result. The feeling evoked by the action was instinctive. The turn to ethics was a mask to shield the fact that for many, the impossibility of reconciling their consciences with the law was something that needed to be overcome.

In any case, regardless of how the turn to ethics is reached, Koskenniemi continues to describe its mise en oeuvre. Per Koskenniemi, even if law is not just formal texts, its informality cannot be reduced to utilitarian calculations, absolute rights, or procedural techniques either. A decision has to be made, and, as Carl Schmitt would opine, this decision is borne out of legal nothingness.98 Per Koskenniemi, ‘what counts are the experience of the decision-maker and his or her sensitivity to the demands of the situation. The problem is not about criteria or process, but about something that might be called “wisdom”’.99 The merit of this approach is clearly that it discounts Kosovo as a precedent. However, this is cancelled out by the subjectivity of the decision-makers, that is, the intervening powers. Here we come full circle in our circumvention of the Security Council. After all, its Permanent Members are also subjective decision-makers. Neo-Realism, the anathema of international lawyers, rears its head.

Therefore, the turn to ethics may also be a turn to politics, a politics by those who have the means to control others. The Kosovar Albanian is worthy of support as long as he remains a helpless victim, but woe betide him if he attempts to liberate himself from NATO guardianship.

Morality and ethics clearly play a role in legal reasoning within the Kosovo debate, as does politics. Koskenniemi is correct in this respect, but the question remains, (how) can ethics aid in resolving the legal problems brought up by the Kosovo crisis?


Allen Buchanan is a prominent example of those scholars who have attempted to chart a practical application for the use of ethics in reforming international law. Buchanan posits that the law surrounding humanitarian intervention is in need of reform, citing the deficiencies of existing international law to support this contention.100 He poses the question as to under what circumstances it may be morally defensible (and ergo excusable) to violate international law in an effort to reform it.101 As D’Amato has stated, if we live in a system of immoral laws, ‘we will spend our time thinking of ways to act illegally and subvert the system so as to establish a new system where the new laws will be entitled to respect’.102 Noting that the Kosovo operation was justified on two bases – primarily as a necessary response to an impending humanitarian disaster but also in the opinion of some leaders as an important first step toward establishing a new norm of customary international law103 – Buchanan assesses whether the latter contention is valid. Investigating whether such an operation, which violated international law as it then existed, was justified in order to initiate an improvement in the international legal system, Buchanan devises a series of rules to be used in determining the moral justifiability of illegal acts of reform. He is not alone in such ideas.104

Buchanan tells us that these guidelines are not intended to provide comprehensive conditions for the justification of humanitarian intervention, but rather are to be applied to proposals for armed interventions once the ‘familiar and widely acknowledged’105 conditions for justified intervention are already satisfied. Amongst the most important of these, Buchanan tells us, is proportionality. Unfortunately, he neglects to enumerate the others. Given that he dispenses some 20 pages previous to this discussing moral justifiability and subjectivity, failing to let the reader know what basic conditions ‘justify’ (objectively, one would presume) intervention as a precursor to his ‘guidelines’ seems a glaring omission. Leaving this lacuna aside, however, Buchanan tells us that eight guidelines may provide help in assessing the justifiability of illegal action directed towards system reform.

Very briefly, Buchanan’s guidelines run thus: (1) the closer the system approximates the ideal of the rule of law, the greater the burden of justification for illegal acts; (2) the less seriously defective the system is in terms of justice, the greater the burden of justification; (3) the closer the system is to legitimacy, the greater the burden. Buchanan opines further that (4) illegal acts that violate fundamental morally defensible principles of the system carry a greater burden, and that (5) the greater the improvement that would result from reform, the stronger the case for carrying it out. Further, (6) illegal acts likely to improve system legitimacy are more easily justified; (7) illegal acts likely to improve substantive justice are more easily justified; and (8) illegal acts likely to make the system more consistent with its most morally defensible fundamental principles are more easily justified.

Buchanan’s rationale for the guidelines is twofold. First, they provide a means of weighing whether any illegal acts can be accurately described as being directed toward system reform, and secondly, if this is the case, whether committing said act is compatible with a sincere commitment of bringing international relations under the rule of law.

Buchanan’s guidelines appear unnecessarily complicated to my mind. Effectively, he cites four desiderata, namely ‘the rule of law’, ‘substantive justice’, ‘legitimacy’ and ‘morally defensible fundamental principles’, and opines that legal system reform should tend toward these, not away from them. This in itself is a simple contention, apart from the fact that he fails to define these terms. Indeed, Buchanan himself admits that ‘different agents may have different views about what justice requires. Thus the guidelines are intended to provide concrete guidance without presupposing a particular theory of justice’.106 This is where Buchanan’s thesis ceases to be of any practical use. Although he brands his guidelines ‘concrete’, they are far from it, and are in fact merely nebulous. ‘The rule of law’ may, I suppose, find an adequate definition upon which lawyers, philosophers and States can broadly agree. However, the other desiderata are, I fear, far from clear. Their content is open to a broad range of interpretations, and this being the case, even supposing that States were to take cognisance of Buchanan’s guidelines in regulating their illegal conduct, one State’s interpretation of what is simultaneously an improvement in the fields of ‘legitimacy’, ‘substantive justice’, and ‘morally defensible fundamental principles’ may differ wildly from that of its neighbour. Francioni has further noted that such models too quickly translate the ‘exigencies’ of justice into positive exceptions to the UN Charter system.107 This is so, even discounting bad faith on the part of one State and the potential for such – abusing the vagueness of the operative terminology to attain a selfish end – is clear. Buchanan as much as admits this himself:

The guidelines proceed on the assumption that content can be given to the idea of improving the system morally and they employ the notion of justice. However, they are not intended to provide a comprehensive moral theory nor to supply content for the notion of justice.108

The question remains, however, without a frame of reference as regards terminology, how can these guidelines become practically useful? Buchanan attempts to demonstrate their potential utility in examining the NATO intervention in Kosovo as a ‘test case’.

Buchanan begins by dubbing the NATO intervention illegal. Moreover, he describes it as clearly meeting the threshold of an illegal act directed at international legal reform. While this claim is dubious (as it was not claimed by all intervening parties that this was an element of the rationale for launching the operation), Buchanan continues to test how this intervention fares when judged against his guidelines. In light of the deficiencies inherent in the international legal system concerning the rule of law, justice and legitimacy, Buchanan notes that the burden of justification for such an attempt at reform is not overly demanding. However, since such interventions as that in Kosovo infringe the fundamental principle of sovereignty, as articulated in Articles 2(4) and 2(7) of the UN Charter, expressly forbidding such armed interventions, the burden would appear to be much higher in this regard. Nonetheless, Buchanan informs us, it is only morally defensible fundamental norms which are to be protected, and in this respect, he adjudges that State sovereignty need not necessarily provide a definitive barrier to reform if the reform is carried out in the name of human rights.

However, Buchanan finishes by concluding that humanitarian intervention à la Kosovo does not qualify under his guidelines. Such a practice would be too much of a destabilising influence to the rule of law, and would not constitute a moral improvement to the international legal system. In his own words,

[a] military alliance such as NATO is not the sort of entity that would be a plausible candidate for having a right under international law to intervene without UN authorisation. The chief difficulty is that such a norm would be too liable to abuse.109

Buchanan’s contention seems to be that while the current rule is inadequate, the manner in which the NATO intervention was carried out cannot be held to represent a template for future actions, as this would not represent a moral improvement in the legal system.

Thus, Buchanan’s ‘ethics in action’ failed to meet its billing, it seems. In providing us with guidelines couched in nebulous terminology, he is unable to tell us anything concrete beyond the fact that the manner in which the Kosovo intervention was undertaken is outside the realms of his desiderata. Perhaps here I am being harsh on Buchanan. His utopian vision of law reform via illegal action tending to buttress commendable goals such as legitimacy and the rule of law is doubtless a noble cause. However, what is required in this context is more than a utopian philosophy. Concrete criteria, rooted if possible in the law itself, are required, allowing us to distinguish the justified/legitimate/legal (depending on which discourse one follows) interventions from the abusive ones. Finally, it is perhaps germane to note the opinion of the World Court in the Nicaragua case, which stated that where a breach of a legal rule is accompanied by a justification of said breach referring to the rule in question, ‘the significance of that attitude is to confirm rather than to weaken the rule’.110 Hence, it is dubious whether illegal international law reform as conceived of by Buchanan is even possible. This sentiment is echoed by Corten, who explicitly states that ‘A breach of international law, as such, cannot generate an evolution of the rule prohibiting force’.111


Buchanan’s analysis is by no means unique. However, it is indicative of just how far away from international law the debate on humanitarian intervention has managed to travel in a remarkably short time, and it somewhat bears out Koskenniemi’s theory of a ‘turn to ethics’. However, this tendency was not universal. Moving so far away from the law, and vesting judgment of the legitimacy and permissibility of conduct in international relations in uncertain notions of ethics opens the door for abusive expansion of whatever framework one uses to define such legitimacy. This was realised by some scholars who attempted to couch their analyses in something more concrete, and more akin to the law itself. Thomas M Franck is a member of this group.

Franck adopts an honest, direct approach and reduces the debate on humanitarian intervention to one concerning the conflict between peace and justice, rather than legality versus legitimacy. This alternative branding of the relative positions represents somewhat more than a cosmetic readjustment of the same debate, as Franck displays a strong predisposition toward working with the positive law rather than trying to circumvent it.112 He quotes with approval Lauterpacht’s dictum that ‘It is impossible, in the scheme of things devised to secure the reign of law, to provide machinery calculated to disregard the law’.113 Franck’s position in this regard certainly sets him apart from the bulk of those scholars attempting to separate the legality of humanitarian intervention from the ethics of the action. Branding humanitarian intervention as a legal exception to the prohibition on the use of force will necessarily require legal justifications, as distinct from the moral, political, linguistic and ethical justifications proffered in defence of ‘legitimacy’ theories.114 He begins with a discussion of the idea of legal ‘loopholes’, lacunae within the law allowing one to circumvent the general everyday rule. His view toward such phenomena is ambiguous, stating that loopholes can be bad, but can also help to ‘save the law from itself’.115 As to the current state of positive international law, Franck wonders whether the drafters of the UN Charter tried to ‘plug’ too many loopholes, and in attempting to forge a just international society unintentionally created situations of grave injustice.

In undertaking a teleological analysis of the Charter regime, Franck begins by examining its origins, beginning with the Covenant of the League of Nations and the 1928 Briand–Kellogg Pact. Upon examination of the Charter itself, Franck notes the peripheral role accorded to human rights provisions, particularly when compared with the overriding importance attached to the provisions outlawing the use of force. Peace, therefore, Franck concludes, has definitively been prioritised over justice.116 Nonetheless, he notes that the historical context surrounding the Charter had also demonstrated the need ‘to guard against the sacrifice of justice for peace, as in the craven Anglo–French surrender to Hitler’s “humanitarian” demands on Czechoslovakia in the name of the Sudeten-Germans’.117 With this in mind, the French, amongst others, had attempted to amend the Charter; legitimising intervention when clear violations of human rights rose to the level where they themselves could be perceived as a threat to the peace. However, such proposals were rejected.118 Despite such failures, Franck maintains that the drafters of the Charter realised that ‘To preserve peace … would also require an effective response to massive injustices of the kind perpetrated by Nazi and Fascist governments against their own and other populations’.119 Here, Franck’s reasoning begins to come unstuck somewhat. It is clear that a right of humanitarian intervention – which transpires to be Franck’s ‘effective response’ of choice – was discussed at the San Francisco conference and was rejected, as were the French proposals linking human rights with a ‘threat to the peace’. While such a nexus has arguably, at least to some limited degree, been established by the custom and usages of the Security Council since then,120 to suggest that the Charter’s drafters believed that that a sine qua non of the preservation of peace would entail meddling in the internal affairs of a foreign State is utopian and unrealistic. Such a position would have been anathema to any State that set high stock by its own sovereignty, and would have caused a seismic shift in international law. Further, it is perhaps worth noting that the drafting process of the Charter was contemporaneous, not subsequent, to the evidence gathering activities for the war crimes trials at Nuremberg. The full extent of Nazi atrocities was not yet completely clear. Even had it been, it is perhaps doubtful that the attitudes of many States would have been different. The numbers exterminated by Nazi genocide reached shocking proportions, but they themselves were dwarfed by the numbers killed as a result of the Second World War itself. In such circumstances, prioritising peace over justice might well have been seen as rational.

This somewhat skewed account of the drafting process aside, Franck’s analysis of the international legal regime on the use of force is quite thorough. He comes to the same conclusion as Oscar Schachter in holding that the chief deficiency of international law is its lack of effective enforcement, particularly as regards the absence of compulsory jurisdiction and the limited ability of any international institution to impose effective sanctions upon a violator.121 Per Franck, this may force recourse to self-help on the part of disgruntled States. Franck opines that if the wrong being perpetrated within a State against a portion of its own population is of a kind specifically prohibited by international agreement, then ‘humanitarian intervention against those prohibited acts may be thought of as a subspecies of self-help’.122 This is an interesting position, but raises the question as to the limits of such self-help. What about other international agreements which do not have a human rights component? Only allowing humanitarian intervention and not other intervention as a means of self-help to enforce other types of international accords presupposes a normative hierarchy of sorts with human rights sitting atop the pyramid of norms. Since nothing of the sort exists in international law (unless we take fundamental human rights norms to be ius cogens, in which case it is highly salient to point out that the prohibition on the use of force is also ius cogens) the hierarchy must be of a moral, rather than a legal nature. This brings us a long way from self-help as a mooted legal doctrine, and into the realms of moral, rather than legal, remedies.

Despite such problems, Franck makes a conscious effort to make use of the law, to stretch it rather than disregard it, and to base his model for humanitarian intervention on something more concrete than subjective conceptions of ‘legitimacy’. He notes the discussion of the possible extension of any right of intervention by the ICJ in the Nicaragua (Merits) judgment, which stated that ‘Reliance by a State on a novel right or an unprecedented exception to the principle might, if shared in principle by other States, tend towards modification of customary international law’.123 While Franck notes that practice itself cannot amend a treaty per se, the Court has stated in the past that the practice of a UN organ may be seen to interpret the text and thereby shape our understanding thereof.124 Franck makes the oft-repeated observation that hard cases may make bad law, and that there is a strong tension between legal consistency on the one hand, and justice on the other. However, Franck argues, slavish adherence to a rule which is unjust in the circumstances can also undermine law’s legitimacy, albeit in a different way. In order to resolve this problem, Franck proposes the transposition of the concept of necessity and the plea of mitigation from domestic law to international law.

Franck draws attention to the two most celebrated of the ‘lifeboat’ cases, namely R v Dudley and Stephens125 and United States v Holmes.126 In both cases, the defendants were convicted of murder on foot of acts of cannibalism. However, due to extreme mitigating circumstances, whereby the defendants acted in the reasonable belief that they had no other choice and that their actions were in the last resort, sentencing was significantly lighter than would otherwise have been the case. In the former case, the two defendants were sentenced to death, but Lord Coleridge, for the unanimous Court, commended them ‘most earnestly to the mercy of the crown’, which duly commuted their sentence to six months imprisonment, most of which had already been served. In the latter case, the penalty of six months imprisonment was subsequently remitted. Franck notes that in neither case was necessity treated as an exculpatory defence to the charge of murder, claiming that the judges ‘went out of their way to ensure that murder remained a crime, even in circumstances of extreme necessity’. However, these circumstances were not ignored, and mitigated the penalties imposed upon those whose acts ‘were found to have been illegal but, in the extreme circumstances, justifiable’.127 Per Franck, there may be differences separating one national system from another as to whether necessity mitigates the consequences of a crime or excuses it completely, but all systems recognise the obligation of the law to do one or the other. It is perhaps worth noting that international law is not completely unfamiliar with the concept of ius neccesitatis, but that it, perhaps justifiably, has long enjoyed a ‘bad reputation’ in the international legal sphere.128 However, it is not without precedent and has been pronounced upon by international tribunals. In the early case of The Neptune, settled in 1797 by the international arbitral tribunal established by the 1794 Jay Treaty, the court held that a situation of necessity might allow for the setting aside of all written and customary law in certain circumstances, but that such necessity ‘must be absolute and irresistible’, thus establishing a very high threshold to be met, but nonetheless allowing for the possible operation of the principles of ius necessitatis contra legem strictum.129 This case was partly followed by an individual opinion of Anzilotti J in the Permanent Court of International Justice (PCIJ), where he noted that ‘necessity may excuse the non-observance of international obligations’, though it was clear that this was to be interpreted very restrictively and that necessity was to be understood as analogous to force majeure.130 The doctrine is rarely invoked by States. Ius necessitatis has the potential to open a ‘Pandora’s box’ of subjective invocations of such a doctrine, and therefore any invocation of necessity doctrine is regarded with utmost suspicion by States, and its normative status – beyond certain limited situations such as the doctrine of military necessity in international humanitarian law – is dubious, and is certainly not strongly-rooted enough to tamper with the UN Charter regime.131

Franck argues for the transposition of the criminal law plea of necessity into the international legal system, arguing that this ‘is not merely a summons to temper the law with considerations of moral legitimacy, but … also a reminder to consider the facts of a case before applying general normative principles’.132 The essence of mitigation, according to Franck, is that the law recognises the continuing force and validity of a given rule, such as the prohibition upon the use of force, while also accepting that in extraordinary extenuating circumstances, condoning a justifiable violation may do more to justify the law’s legitimacy than would its rigorous implementation. A similar model is proposed by Byers and Chesterman, as well as by De Nicola.133 As Roberts points out, the language in both cases is confusing. Franck in particular argues that interventions may be ‘justified’, but rather than using the tort defence of ‘justification’, employs the criminal law model of ‘mitigation’, which affirms that the intervention can never be (entirely) justified.134 The reason for this would seem to be the fear that Kosovo could otherwise be used as a precedent.

As to the inherent subjectivity of permitting a rule which allows States to disregard the law where they may judge it ‘necessary’ to do so, Franck argues for a novel system of ‘international jurying’, whereby the political organs of the United Nations may sit in joint judgment of the legitimacy of any given employment of the necessity ‘defence’. This would guard against unilateralist abuse of the necessity doctrine, as well as adding a Fuller-esque ‘social dimension’ to legal positivism.135 Franck argues that, to a degree, this is already happening, with the Security Council’s ‘post hoc approval of ECOMOG military action in Liberia, and implicitly, of NATO’s humanitarian intervention in Kosovo’.136 He acknowledges that he is, in reality, stretching the law quite some distance from its original intended purpose, but argues that ‘necessity’ and ‘international jurying’ are, in reality, simply examples of legal fictions, and that such legal fictions play a key role in any legal system, including the international legal system. He cites as an example the justification of the use of force by the Security Council to oust the Haitian military junta, where ‘it contrived the fiction that, by causing a flow of refugees, the junta was causing a threat to international peace and security’ such as to justify action under Chapter VII.137

Franck’s analysis has much to commend it on the face of it. Unlike many of his counterparts, he only resorts to stage three (instrumentalism) of Koskenniemi’s ‘turn to ethics’ model. In doing so, he remains firmly rooted in legal discourse. Nonetheless, Franck’s ideas reveal several unforgivable flaws which render his thesis wholly inadequate for dealing with the problem of humanitarian intervention. First, the aforementioned ‘moral hierarchy’, permitting intervention in furtherance of human rights obligations, but not in furtherance of other State obligations, is insufficiently rooted in law, and is a construct of Franck’s own devising. Secondly, Franck’s reading of the UN Charter and of other international legal instruments reflects a utopian bias. Thirdly, in repeatedly refusing to allow Kosovo to be used as a precedent, he fails to distinguish between act-utilitarianism and rule-utilitarianism. Fourthly, and importantly, his ‘international jurying’ model, while not without its merits, effectively completely unbalances the power structure within the United Nations system. A Security Council that fails to support armed intervention in a given State due to a split in voting is likely to be equally split when a post-hoc motion of censure for the action is tabled, as was the case in Kosovo. Furthermore, the General Assembly has no competence in this area. Awarding it such competence, as a post-hoc ‘jury’ would render the original power structures conceived by the drafters of the Charter defunct. The fact that Franck should feel the need to do so is also slightly surprising, given that he remains convinced that the Charter’s drafters were in any case aware of the need to protect human rights in order to ensure peace. In addition, ‘international jurying’ is unlikely to produce consistent results in consistent circumstances, due to the fact that the ‘jurors’ may have selfish interests in the outcome of any given crisis. Furthermore, public opinion is highly emotional and capricious, and completely lacks any evidentiary safeguards, necessary to ensure fair legal procedure, which also forms part of the human rights corpus.

Franck himself admits that ‘law derives part of its legitimacy from its consistent and equal application: the treating alike of like cases’,138 but admits this is unlikely using his model. He attempts to circumvent this problem by conceiving of humanitarian intervention ‘not as a new legal right … but as a mitigating circumstance that does not create law and which is recognised as purely circumstantial and discretionary relief’.139 He is not the only scholar to have toyed with such an idea, with Byers, Chesterman and even Ian Brownlie (many years earlier) having shown sympathy for such an approach.140 So Franck, despite purporting to stay within the law, nonetheless feels that legitimacy and due process requirements of a legal system cannot be applied to his remedy. This is to say nothing, of course, of the incongruity of transposing a necessity model shaped in common law criminal jurisdictions into an international system of inter-State obligations which more closely resembles tort law and which knows no such concept. Even were the concept less cumbersome to transpose into international law, it is worth recalling that within common law, in cases such as R v Dudley and Stephens and United States v Holmes, a plea in mitigation is but one of five ways in which a man may escape punishment for a given act which seems prima facie to be illegal, the others being: (1) determination by a judge that he did not commit a crime; (2) decision by a prosecutor not to seek an indictment; (3) acquittal by a jury; and (4) pardon or commutation of the sentence by the executive. Although none of the final three are held within a rigid formal framework of due process that prevents factual error, excludes emotional and personal factors and guarantees that all the formalities of legal procedure will be observed, the necessity defence grew up organically in an environment which knew these other phenomena, and it may well be germane to note that inter-State international law does not particularly mirror such a structure to any great degree, lacking, as it does, a proper prosecutor and executive.141 It is worth perhaps reminding ourselves of the difficulties which were involved in the admission by the Appeals Chamber of the International Criminal Tribunal for the Former Yugoslavia that a related defence of duress could exist in international criminal law when contemplating how complicated transposing the similar doctrine of necessity as a defence into general international law might be.142 One is also left wondering whether Franck could have chosen a slightly different case law based approach, resting upon the lenient punishment (a mere declaration of illegality with no sanctions) meted out to the United Kingdom by the ICJ in the Corfu Channel case, itself not a giant leap from the ‘illegal but justified’ model proposed by Franck, and resting upon much more contiguous jurisprudence.

In the end, then, Franck’s necessity/jurying model, after promising much, delivers little. While resisting the urge to veer away from law and into ethics and philosophy is commendable, it does not excuse the serious deficiencies inherent in Franck’s reasoning, which, much like the legality/legitimacy model, eventually loses its way by moving too far away from positive law.

Paolo Picone is representative of a select group of other scholars who attempt to root their analyses in or as close to the law as possible. Picone argues in a somewhat different vein for a case-by-case analysis, based upon what he dubs ‘alcuni parametri o principi di carattere più generale’.143 However, his parameters, although more general – bona fides; ius necessitatis or human catastrophe; proportionality; respect for ius in bello; and ‘effet utile’