A. Authorized intervention

It is generally though not universally agreed that the Security Council has the authority, under Chapter VII of the UN Charter, to conduct or authorize humanitarian intervention.¹⁶

When authorizing humanitarian intervention, the Security Council typically determines that a humanitarian crisis poses a threat to the peace. Since the end of the Cold War, the Security Council has interpreted the term ‘threat to the peace’ broadly.¹⁸

Article 2(7) of the UN Charter states that the Charter does not authorize the United Nations “to intervene in matters which are essentially within the domestic jurisdiction of any state.”¹⁹

B. Unauthorized intervention

If humanitarian intervention is not authorized by the Security Council, its status under the Charter is considerably more dubious. Most attention focuses here on Article 2(4), the Charter's prohibition of the threat or use of force in international relations. Article 2(4) states: “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.”²⁵

Some interpret Article 2(4), in the context of the Charter as a whole, as prohibiting all use of force in international relations, with only two exceptions: (1) as authorized by the Security Council, and (2) in exercise of the right of self-defense recognized in Article 51 of the Charter.²⁶

With regard to the specific terms of Article 2(4), some interpret the term ‘territorial integrity’ broadly; in their view, every territorial incursion is a violation of territorial integrity. This interpretation of ‘territorial integrity’ would once again turn Article 2(4) into a general prohibition. Under a more narrow interpretation of ‘territorial integrity’, a state violates the territorial integrity of another state only if it seizes part of the second state's territory.

It should be noted that even the more narrow interpretation of ‘territorial integrity’ poses a problem for humanitarian intervention on behalf of secessionist minorities. The Kosovo intervention ousted the state of Yugoslavia from its Kosovo province for the indefinite future, and the Northern no-fly zone over Iraq ousted the state of Iraq from the Kurdish areas in Northern Iraq for the duration of the Saddam Hussein regime. In both cases the intervenors ultimately disclaimed the intention of redrawing international boundaries, but some find it hard to see these interventions as anything other than the use of force against territorial integrity.

The term ‘political independence’ in Article 2(4) is also contested, particularly in the context of pro-democratic intervention. Under a broad interpretation of the term ‘political independence’, intervenors violate the political independence of a state when they change its political path in any way—for example, from dictatorship to democracy.²⁸

The import of the third prohibition in Article 2(4)—proscribing force otherwise inconsistent with the purposes of the United Nations—is also contested. Article 1 of the Charter lists “The Purposes of the United Nations.” The first-listed purpose, and by nearly all accounts the primary purpose, is “To maintain international peace and security.”³⁰

Also relevant to the legality of humanitarian intervention is Article 2(3) of the Charter, which states: “All Members shall settle their international disputes by peaceful means in such a manner that international peace and security, and justice, are not endangered.”³²

In addition to Charter law, the legal status of humanitarian intervention may be affected by the somewhat mysterious body of law known as customary international law. It is generally though not universally agreed that international custom can rise to the level of law if (1) it is sufficiently widespread, and (2) it is regarded as binding by states. Article 38 of the Statute of the International Court of Justice lists several sources of international law. The first-listed source is “international conventions” (including, of course, the UN Charter). The second-listed source is “international custom, as evidence of a general practice accepted as law.”³⁴

A number of commentators believe that the Charter can, in effect, be amended through the processes of customary international law.³⁷

There are many opinions as to the meaning of the Charter and the content of customary international law. While I have expressed some of my own opinions, I am not primarily interested in whether humanitarian intervention is legal or illegal in the abstract. I am interested in the possible legal consequences of humanitarian intervention. I will explore these consequences by considering what would happen if a target of intervention attempted to challenge the intervention as a violation of international law. The target state might try to press its case in a number of different forums, but each of these forums would present problems for it.

C. UN Security Council

Under Article 24(1) of the UN Charter, the Security Council has primary responsibility for the “maintenance of international peace and security.”³⁹

During the Kosovo intervention, the Security Council was presented with a draft resolution that would have condemned the NATO bombing campaign against Yugoslavia.⁴⁰

In the case of Kosovo, the veto power of permanent members was not necessary to the defeat of the proposed resolution condemning intervention. Things were different in the case of Nicaragua in 1984. The United States was then seeking to overthrow the Sandinista regime in Nicaragua. The motives of the United States were geopolitical rather than directly humanitarian, though one could say that the geopolitical goal of containing Communism was indirectly humanitarian in that Communism causes suffering.

In seeking to overthrow the government of Nicaragua, the United States mined Nicaraguan ports and conducted several naval attacks on Nicaraguan ports, oil installations, and a naval base. These operations were conducted or organized by the CIA. In addition, the U.S. armed and trained the Nicaraguan “Contra” rebels who were seeking to overthrow the government of Nicaragua.⁴¹

Nicaragua sought to challenge U.S. intervention as a violation of international law, and its efforts still hold lessons for any target of intervention, humanitarian or otherwise. Nicaragua first sought relief from the Security Council against American military intervention. All of the members of the Security Council except the United States voted for (or abstained on) a mild resolution in favor of Nicaragua, one that did not take any enforcement action against the United States. However, the United States exercised its veto, and the resolution did not pass.

Nicaragua next sought relief from the ICJ. It achieved somewhat better results there, as discussed below.

D. International Court of Justice

The ICJ is the court of the United Nations; its statute is a part of the UN Charter.⁴²

Article 96 of the Charter provides that the General Assembly or the Security Council may request an “advisory opinion” from the ICJ on “any legal question.” Conceivably, even if the ICJ could not exercise jurisdiction over a case against the intervenor state, the intervenee could persuade the General Assembly to request an advisory opinion on legal questions relating to the intervention. It is not certain that the ICJ would agree to deliver such an opinion, however.

There is very little likelihood that the ICJ will ever brand as illegal a humanitarian intervention authorized by the Security Council. There is a very large likelihood that the ICJ will brand as illegal any humanitarian intervention not authorized by the Security Council, if the ICJ ever squarely addresses the legality of such an intervention.

In the Kosovo dispute, Yugoslavia sought relief from the ICJ. Yugoslavia brought cases against ten of the NATO countries, seeking, among other things, a provisional measure (what in the United States would be called a preliminary injunction) ordering the NATO countries to cease their attacks. The ICJ refused to order the NATO countries to stop bombing. The NATO countries had raised jurisdictional objections, and the ICJ decided that it was not sufficiently persuaded that it had jurisdiction over any of the cases brought by Yugoslavia.⁴⁴

The ICJ also retained all but two of the cases on its docket (it dismissed the ones against the United States and Spain). This left open the possibility that Yugoslavia could ultimately prevail on the merits after convincing the ICJ that it had jurisdiction. In the interim, of course, the NATO campaign was successful in ousting Yugoslavia from Kosovo, and there was a change of government in Yugoslavia. As of March 2003, the eight remaining cases filed by Yugoslavia are in a kind of limbo; possibly Yugoslavia will withdraw them.⁴⁶

It is interesting to speculate on what would have happened if the ICJ had ordered the NATO countries to stop participating in the attack on Yugoslavia. My own view is that some of the European countries would have obeyed the ICJ's order.

In the Nicaragua case that began in 1984, the ICJ did issue a preliminary injunction and then a final judgment against the United States.⁴⁷

The ICJ's 1986 decision on the merits of the Nicaragua case contains a number of passages that are hostile to the concept of humanitarian intervention:

Under the ICJ statute, decisions of the ICJ are theoretically nonprecedential; a decision “has no binding force except between the parties and in respect of that particular case.”⁴⁹

Despite the Nicaragua decision, advocates of unauthorized humanitarian intervention might still entertain some hope that the ICJ will some day drop or moderate its opposition to such intervention. First, the world has changed since 1986. At that time, two contending political systems were in equipoise: democracy and Communist dictatorship. Now democracy is ascendant, and this development may influence the ICJ (more of whose judges now come from democratic countries). Second, the government in Nicaragua was by no plausible account a particularly horrible regime, in the context of its time and place. Possibly a truly repellent regime would receive less sympathy from the ICJ.

E. International Criminal Court

A target of intervention might also seek relief from the new International Criminal Court. The ICC came into existence in July 2002, despite opposition from the United States.⁵⁶

Under its constitutive treaty (the “Rome Statute”), the ICC is supposed to exercise jurisdiction over four kinds of international crimes: genocide, crimes against humanity, war crimes, and the crime of aggression.⁵⁷

One outstanding issue is how to define the crime of aggression. Most definitions proposed for inclusion in the Rome Statute are based on the General Assembly's 1974 “Definition of Aggression” resolution, which in turn is based on Article 2(4) of the Charter, the Charter's prohibition on the threat or use of force.⁵⁹

Another outstanding issue regarding the crime of aggression is whether, as a precondition to any prosecution for the crime of aggression, the Security Council must determine that an act of aggression has occurred. Some believe that such a prior determination by the Security Council is required by the UN Charter; they point to Article 39 of the Charter, which states that the Security Council “shall determine the existence of any threat to the peace, breach of the peace, or act of aggression.”⁶⁰

If there can be no prosecution for the crime of aggression until the Security Council has determined the existence of an act of aggression, then the ICC will not be of much help to targets of intervention. However, there is great resistance, among parties to the Rome Statute, to allowing the veto of one permanent member of the Security Council to preclude a prosecution for the crime of aggression. Several proposals now under consideration would give the Security Council an initial opportunity to determine the existence of aggression, but in the case of a veto would allow final determination to be made by the ICC, or the ICJ, or the General Assembly.⁶¹

If there could be any proceedings before the ICC on a prosecution for the crime of aggression, without the prior approval of the Security Council, targets of intervention would have a major new legal avenue. It would be very embarrassing for leaders of an intervening state to be the subjects of an ICC prosecution, even if the likelihood is small that they would ever fall into the custody of the ICC. It would also be easier for intervenees to confer jurisdiction on the ICC over a claim of aggression than it is for intervenees to bring a claim of unlawful force that is within the jurisdiction of the ICJ.⁶³

F. Charter law is not plastic

I previously noted the view of some scholars that Charter prohibitions on the use of force could be abrogated through the processes of customary international law. The Charter plasticity view has widely different normative implications, depending on whether force prohibitions are assumed to be (1) vulnerable but not yet abrogated, or (2) already abrogated. If force prohibitions are assumed to be vulnerable but not yet abrogated, Charter plasticity could be an additional reason to avoid humanitarian interventions, lest they lead to a general abrogation of prohibitions on the use of force between states. If force prohibitions are assumed to be already abrogated, Charter plasticity could remove international-law impediments to humanitarian intervention.

In any event, neither version of the Charter plasticity view is realistic, given the institutional framework I have described above. Anyone can make a grand pronouncement that the law is now “Y” rather than “X.” But the ICJ, which is governed by the Charter, will never say that the Charter's prohibitions on the use of force are void. All of its decisions tend in the other direction; since the Nicaragua case, the ICJ has issued a number of decisions demonstrating that it takes a narrow view of the permissibility of the use of force under the Charter. There is not the tiniest cloud hanging over Article 2(4) in the ICJ's decisions.

The Security Council, similarly, will never give up its role in enforcing Article 2(4). All of the Security Council's resolutions on Iraq, from 1990 to 2002, grew out of Iraq's violation of Article 2(4) by its invasion of Kuwait. While the veto of permanent members prevents enforcement against them and their allies, the Security Council will never agree that Article 2(4) has been nullified so that all states are now free to use force in international relations.

And similarly, if the crime of aggression is ever defined as part of the Rome Statute, the ICC will never determine that the provisions of the Rome Statute as to the crime of aggression have become instantly obsolete, and that aggression is not a crime after all. As noted, the definition of the crime of aggression is likely to be based on Article 2(4) of the Charter.

The idea that Charter law can be modified in a way that would never be accepted by the ICJ or the Security Council reflects an anachronistic, pre-Charter view of international law. At one time, the behavior of states was almost the exclusive material of international law, but we now have respected international institutions. It is almost pointless to suggest that something is or could be law if these institutions will never accept it as law.

And if we look at the behavior of states, we would have to conclude, once again, that Charter law on the use of force is alive and well: it has considerable weight as a political norm. When the United States was moving toward a war with Iraq in 2002, most states expressed the view that such a war, if not authorized by the Security Council, would be a violation of the UN Charter. Faced with this opposition, the United States drew back, temporarily, from launching an unauthorized attack on Iraq. There could hardly be any greater proof of the viability of an international norm than if the most powerful state in the world prepares to violate that norm and then draws back, even temporarily, in the face of opposition from other states. Certainly there have been times, since the founding of the United Nations, when permanent members of the Security Council launched nondefensive military interventions with far less concern for the approval of the Council.

As long as the United Nations lasts, international law will never again permit the free use of force by states. There may be a small possibility that an exception to the law on force will develop for unauthorized humanitarian intervention—an exception some claim already to find in international law. Such an exception is possible because there are plausible interpretations of Article 2(4) that permit it. Given the ICJ's decisions on the use of force, however, it seems unlikely that the ICJ will smile on unauthorized humanitarian intervention in the foreseeable future.

I hasten to add that international law should not be the last word in international political morality. Perhaps states should undertake unauthorized humanitarian intervention even if such intervention is contrary to international law. I now turn to such issues of morality and policy.

A. Alternatives to utilitarianism

Many others also take a generally utilitarian approach to humanitarian intervention, whether or not they use the term ‘utilitarian’. However, there are alternatives to a utilitarian approach. For example, a number of thoroughgoing deontological or rights-based positions are possible. Such positions are sometimes suggested by the rhetoric of some participants in the debate over humanitarian intervention.

It might be argued, on deontological grounds, that intervenors should be willing to pay any price to secure human rights abroad. Thus, the United States and NATO should have intervened in Chechnya, even if the result were a bloody war with Russia—even if the result were a bloody nuclear war with Russia.

A deontological position with nearly opposite consequences would be that intervention is prohibited unless the motives of the intervenor are pure: if the intervenor is partly or primarily motivated by geopolitical rather than humanitarian considerations, it must hold back. This position would rule out most if not all interventions, whether or not they are authorized by the Security Council. As political philosopher Michael Walzer and others have suggested, intervenors rarely if ever have purely humanitarian motives; most beneficial interventions occur when the geopolitical interests of the intervening state fortuitously coincide with the interests of humanity.⁶⁵

Another deontological position would be that intervenors must avoid inconsistency and hypocrisy with respect to the seriousness of human rights violations that they are seeking to remedy. This position would in effect require that we intervene everywhere (e.g., war with Russia over Chechnya, war with China over Tibet) or nowhere. And the likely choice of nowhere would be bad for all victims of oppression, not just those whom the intervenor otherwise would have helped. Humanitarian intervention shines a light on abuses that are similar to those committed by the intervenee, especially if failure to suppress these similar abuses seems hypocritical. The Kosovo intervention spawned charges of hypocrisy as to NATO's failure to intervene in Turkey (treatment of the Kurds), Indonesia (East Timor), and India (Kashmir). All three of these states have now improved their treatment of their respective minorities or breakaway groups. There are, of course, local factors in all three cases, but perhaps a part was played by additional publicity and additional pressure from allies generated as a result of the Kosovo campaign. To the old saw that hypocrisy is vice's tribute to virtue, we can add: hypocrisy shines virtue's light on vice.

There are also egalitarian alternatives to a utilitarian position on humanitarian intervention. Egalitarians would like to give more weight to the interests of those who are worse off than to the interests of those who are better off (at least, that is how utilitarians view the matter).⁶⁶

In the area of humanitarian intervention, egalitarianism (at least, of the cosmopolitan variety) would require intervenors from wealthy countries to accept more casualties than would any variant of utilitarianism. Thus, on an egalitarian analysis, it might be right for the United States to sacrifice the lives or limbs of one thousand of its soldiers in order to save the lives or limbs of five hundred badly off young people in another country, even without considering the other benefits of intervention. This would be sacrificing a greater amount of utility for a more equal distribution of utility, a trade-off that egalitarians typically claim to favor.

Another possible alternative to utilitarianism is ethical nationalism, the view that states should give absolute priority to the interests of their own citizens. Under this view, it would be wrong to sacrifice the welfare of even one citizen in order to save the lives or relieve the suffering of any number of foreigners. An advocate of ethical nationalism would see humanitarian intervention as an immoral impulse.

Of course, it is also possible to have a mixed view, one that combines utilitarianism with other approaches. Legal scholar Fernando Teson, for example, delves very deeply into the rhetoric of deontology and trots out some stock arguments against utilitarianism, but ends up concluding that “other things being equal, humanitarian interventions that are likely to cause substantially disproportionate additional suffering should not be initiated.”⁶⁷

I will not defend here my utilitarian approach to humanitarian intervention (other than through the foregoing discussion of the alternatives). In what follows I will view the problem of unauthorized humanitarian intervention through the lens of utilitarian moral theory, but some of what I say may be of interest to devotees of other approaches.

B. Act or rule?

The debates about humanitarian intervention recall debates in moral theory about rule-utilitarianism, act-utilitarianism, and the place of institutions in utilitarian theory. Most discussions of humanitarian intervention are rule oriented: Under what conditions should a state (such as the United States) intervene? The most prominent possible rule, and one that arguably already exists, is that states should intervene only with the authorization of the Security Council. I will refer to this as the ‘authorization rule’.

Suppose one thought, as the Kosovo intervention was about to be launched, that it would have good effects overall. One might still oppose the Kosovo intervention, based on the authorization rule. This would be similar to a rule-utilitarian approach, and if one's reasons for supporting the authorization rule were utilitarian, it would in fact be a rule-utilitarian approach.

Rule-utilitarians are vulnerable to embarrassment by situations in which obeying a generally beneficial rule would have disastrous consequences. Some such hypothetical examples conjured up by moral theorists are a little silly, but in the area of humanitarian intervention, the problem seems more realistically acute. Should states observe the authorization rule at the cost of allowing hundreds of thousands of people to die?⁶⁸

The same problem can beset adherents of other rules that have been suggested in the area of humanitarian intervention. Numerous writers have attempted to specify the conditions under which humanitarian intervention should take place. If these conditions take the form “Intervene only if X, Y, and Z,” they are also vulnerable to attack by way of difficult realistic examples. For example, three worthy suggestions are that (1) intervening states should act through regional organizations; (2) intervening states should subject themselves to the jurisdiction of the ICJ and the ICC for all claims arising out of their intervention; and (3) intervening states should not bomb from high altitudes if doing so makes it difficult to distinguish between military and civilian targets. Suppose now that the United States had been prepared to intervene against genocide in Rwanda, but had not been able or willing to meet these and other worthy conditions. After all, the humanitarian impulse is a fragile thing, likely to be overridden by national interest, and such conditions would give the United States additional reasons not to intervene. Would it be better not to intervene in a Rwanda-like situation than to intervene without meeting the conditions?

Many theorists, following philosopher David Lyons, believe that rule-utilitarianism is always in danger of collapsing into act-utilitarianism. In the area of humanitarian intervention, such a merger would be effected by the rule: intervene so as to cause the best consequences, or intervene so as to minimize suffering overall.

Lyons suggests that while utilitarians can support institutions that limit by rules the direct pursuit of utility, they will behave in an act-utilitarian manner once they are inside those institutions.⁶⁹

In the area of humanitarian intervention, one's commitment to rules, and in particular to the authorization rule, is likely to be affected by one's attitude toward the United States. The United States is the country most able to project force beyond its borders. Is the United States a benevolent, huggable hegemon or a malevolent, reckless rogue? Doubtless the truth is somewhere between these two poles. I am disposed to believe that the effects of American intervention will likely be good; I know that others disagree.

C. Danger of abuse

Opponents of a humanitarian exception to the authorization rule fear that such an exception will be abused. It is important to distinguish between two different ways in which such abuse can occur. First, a perceived right of humanitarian intervention could lead states to label as ‘humanitarian’ wars that they undertake for other-than-humanitarian motives, and that they would have undertaken in any event. This would not be an unfortunate result. Of course, it is unfortunate when states launch bad wars, that is, those having bad consequences overall. But given that a state will launch a bad war, it is preferable that the state claim to be undertaking humanitarian intervention rather than offer some other pretext, such as self-defense. A state that claims to be undertaking humanitarian intervention has pledged, however hypocritically, to promote the well-being of people beyond its borders. Such a pledge can be held up to it by other states, and may even moderate its behavior.

The second possibility is that a perceived right of humanitarian intervention could lead states to wage bad wars that they would not have waged if they could not claim to be acting from humanitarian motives. This is the true danger of abuse. Consider a hypothetical brutal dictatorship in State X. The dictatorship, let us suppose, is hostile to the United States, which desires to replace it with a friendlier and more benevolent regime. A U.S. military intervention to overthrow the dictatorship would have humanitarian aspects. The people of State X would probably be better off under a replacement regime (at least, those who remained alive would be better off). However, military intervention would also be motivated by nonhumanitarian considerations, which might be paramount for American policymakers. Hypothetically, the nonhumanitarian objectives could lead the United States to start a war that would predictably cause far more suffering than it would cure, a war that the United States would not start if it could not claim a humanitarian objective and also would not start if it were acting solely from humanitarian motives.

Accordingly, those who mistrust the United States (and, presumably, every other state) could advocate adherence to the authorization rule even while conceding the argument I made at the beginning of this essay, that there is a deficit of humanitarian intervention because of the particularism of states. A perceived right of humanitarian intervention could lead states to launch interventions that have some humanitarian aspect, but that have bad consequences overall because they are primarily motivated by geopolitical considerations.

One does not need bad motives, of course, to get bad consequences. Mistakes can also lead to bad consequences, and there are plenty of mistakes in war.

Even if we trust our own country (e.g., the United States), we might mistrust other countries. In that event we might support the imposition of a globally effective authorization rule. Suppose we sat in a world parliament that could somehow enforce its laws. As utilitarians, we might support the authorization rule, even though we thought that our own country could produce good consequences by deviating from it, if we thought that all unauthorized force, considered together, would have bad consequences. And once we succeeded in passing the authorization rule, we would not be able to deviate from it.

Of course, we are not sitting in a world parliament. If our country refrains from using unauthorized force, this will not necessarily prevent other countries from using unauthorized force. Our country might just be giving up “good” wars only to see other countries continue to make “bad” wars.

If we are seeking political realism, it may even be a mistake to view the issue as whether unauthorized force by our country, or unauthorized force under a claim of humanitarian intervention, would have good or bad consequences overall. Citizens interested in a moral foreign policy may be able to affect some uses of force better than others. My own view, as I have said, is that the humanitarian impulse is fragile: the more purely humanitarian an intervention, the more likely it is to be a marginal case, one in which policy can be swayed. Conversely, where a state has a strong geopolitical interest in war, it is less likely to be held back. Therefore, if we support the authorization rule because of its generally beneficial effect, we may be more likely to impede good interventions than bad interventions.⁷⁰

D. The rules we have

We do not have an effective authorization rule that makes deviation impossible and so makes it unnecessary to calculate the consequences of deviation. But we do have a somewhat ambiguous authorization rule, as part of international law and international politics, and violating this rule can have some consequences.

As a political norm, the authorization rule varies in effectiveness depending on the political support that it receives in any given situation. In the period leading up to Security Council Resolution 1441 on Iraq,⁷¹

Sometimes the authorization rule has specifically legal consequences, which in turn have political consequences. Some states are reluctant to violate what they consider to be international law, even if it is naked law, unsupported by any judicial decision. And once law is applied by an international court, it gains greater and wider credibility. In the case of forcible intervention, the ICJ may brand the use of force as illegal and order the intervening state to stop, as in the Nicaragua case. And the ICC prosecutor may bring charges against the leaders of an intervening state, if the ICC ever begins to exercise jurisdiction over the crime of aggression.

Ultimately, such legal consequences will be political consequences only to the extent that they have political credibility with states. But international judicial and prosecutorial decisions do have political credibility—at least, they have greater credibility than naked law. The ICJ's decisions in the Nicaragua case may have helped to end direct American military operations against Nicaragua, and if the ICJ had ordered NATO countries to halt their participation in the Kosovo campaign, such an order may also have had a political effect.

Thus, the political and legal force of the authorization rule can affect the behavior of states in the area of humanitarian intervention. As a matter of prudence, in light of legal and political consequences, states might not intervene because of the authorization rule we now have—though they would intervene if there were no authorization rule.

A different question is whether the legal consequences of the authorization rule should be morally significant beyond their effect on the national interest of potential intervenors. In act-utilitarian terms, this question becomes: Do the legal consequences of a particular humanitarian intervention make the total consequences of that intervention bad (not just for the intervenor, but for all people), where absent the authorization rule the total consequences would have been good?

If each humanitarian intervention eroded the authorization rule, making more likely the nonhumanitarian and nondefensive use of force, that would indeed be a negative consequence. I have already expressed my view that such erosion will not occur, and that it has not occurred as a result of previous humanitarian interventions. The ICJ will not hold that Charter prohibitions on the use of force have been abrogated, and the ICC will not hold that the Rome Statute provisions on the crime of aggression have been nullified before the ICC has even begun to exercise jurisdiction over that crime.

The idea that humanitarian interventions will lead to nonhumanitarian wars has been somewhat overtaken by events. In preparing the way for a war against Iraq, the United States advanced a very broad interpretation of the concept of ‘anticipatory self-defense’. Under this novel view of self-defense, the United States was entitled to attack Iraq and overthrow its government because Iraq might pose a threat to the United States at some time in the nebulous future. This doctrine, if universalized, truly would signal the end of any effective prohibition on the use of force in international relations: in every conflict, it is always possible that one side will someday pose a threat to the other. Here there is a possibility of like cases leading to like cases; this seems more realistic than the possibility, feared by opponents of unauthorized humanitarian intervention, that like cases will lead to unlike cases.

Humanitarian intervention will not erode the authorization rule so as to permit the free use of nonhumanitarian and nondefensive force. It is more likely that any change in the law caused by humanitarian intervention will relate specifically to humanitarian intervention. On the one hand, if repeated cases of unauthorized humanitarian intervention establish a humanitarian exception to the authorization rule, that could be a positive development. On the other hand, there is the danger that humanitarian intervention in borderline cases can strengthen the authorization rule, as applied specifically to humanitarian intervention, so that it becomes more difficult to intervene in the future even when intervention is enormously beneficial. The Nicaragua case, in which the ICJ announced that force is not an acceptable means of promoting human rights, was at the very best a borderline case for humanitarian intervention. While there was a risk that Nicaragua would slip into Cuba-style totalitarianism, the human rights situation in Nicaragua was far better than in many other countries in Central and South America. The Kosovo case, in which the ICJ remarked that the intervention “raises very serious issues of international law,” was also something of a borderline case. Yugoslavia confronted an armed secessionist movement, and its response, while brutal, was not so brutal as to set it apart from other states confronting armed secessionism.⁷²

Soft cases, I suggest, make bad law on humanitarian intervention. Probably the ICJ would not have been so quick to brand humanitarian intervention illegal, as in the Nicaragua case, if the intervenee was a state that was killing tens of thousands or hundreds of thousands of people. Fortunately, the ICJ has not yet addressed the merits of the Kosovo case and may never have to do so. If there is any possibility that the fall of Communism and other changed circumstances will cause the ICJ to moderate its position on humanitarian intervention, we should not foreclose that possibility by presenting it with another borderline case. So the broader legal consequences of unauthorized intervention may, after all, be a reason to err on the side of the authorization rule.