LEGITIMACY AND HUMANITARIAN INTERVENTION: WHO SHOULD INTERVENE?
ABSTRACT: In this article, I examine who should undertake humanitarian intervention. Should we prefer intervention by the UN, NATO, a regional or sub-regional organisation, a state, a group of states, or someone else? To answer this question, I first determine which qualities of interveners are morally significant. I highlight in particular the importance of an intervener’s effectiveness and, in doing so, develop a particular conception of legitimacy for humanitarian intervention. I then consider the more empirical question of whether (and to what extent) the current agents of humanitarian intervention actually possess the morally relevant qualities identified, and therefore should intervene. In the last part of the article, I consider ways of improving agents’ willingness to intervene and, ultimately, the legitimacy of humanitarian intervention.
KEYWORDS: humanitarian intervention, legitimacy, consequentialism, jus in bello, the responsibility to protect
CITATION: James Pattison (2008) “Legitimacy and Humanitarian Intervention: Who Should Intervene?”, International Journal of Human Rights, 12 (3), pp. 395–413.
Since the end of the bipolar, divisive international system of the Cold War, the United Nations and its Security Council have been reinvigorated, and this has been reflected in the number of its peace operations. Military force sponsored by the UN was used only 22 times from 1946 to 1990, but 56 times from 1990 to 2000.1 There has been a similar proliferation in peace operations by non-UN actors, such as regional organisations.2
The events of 11 September 2001 and the 2003 War on Iraq, however, risked undermining this new-found willingness to undertake humanitarian action. First, in the aftermath of the terrorist attacks on the U.S., a common – and perhaps well-founded – fear was that states would turn inwards, focusing their interests on national security and the ‘War on Terror’ rather than on humanitarian intervention to protect the human rights of individuals in far-off places. Second, the U.S. and U.K.-led operation in Iraq threatened to damage the credibility of humanitarian intervention irrevocably, since one of the justifications offered by George Bush and Tony Blair was essentially humanitarian: to end the violation of human rights by the Ba’athist regime and to bring freedom and democracy to Iraq. That this war had seemingly questionable motives, used force indiscriminately, involved the abuse of civilians, and has led, in effect, to civil war, could have created an unrelenting cynicism and rejection in the international community of any international action for apparently humanitarian purposes. The risk of world public opinion and elites being against any future international action with a purported humanitarian justification was increased further by the degree of worldwide attention on – and condemnation of – the war.3
Although these two events have led to a degree of reluctance on the part of western states to participate in peacekeeping operations and perhaps to conduct controversial humanitarian interventions in the future, there have still been a number of humanitarian interventions and peacekeeping missions since.4 Examples include: intervention in the Ivory Coast by France, the Economic Community of West African States (ECOWAS), and the UN; the ECOWAS and UN action in Liberia in 2003; the EU and UN intervention in the Democratic Republic of Congo (DR Congo); the Australian-led peacekeeping mission in the Solomon Islands in 2003; African Union peacekeeping in Darfur and Burundi in 2003 and 2004 respectively; and the U.S. intervention in Haiti in 2004. In addition, there continue to be calls for humanitarian intervention to be undertaken in a number of other places where the violation of basic human rights currently goes unchecked. More generally, despite recent opposition to the war in Iraq, there has been a growing consensus in the international community that humanitarian intervention can be morally acceptable on occasion. Indeed, it is much harder to find someone who completely supports non-intervention nowadays. The lack of action in Rwanda (or, more accurately, lack of effective action) and the subsequent genocide had a massive impact on the theory and practice of intervention. Even those who are deeply suspicious of humanitarian intervention and deeply sceptical about its prospects of success will probably still admit that it might, in theory, be justified when a humanitarian crisis is sufficiently serious.
Underlying this apparently increased acceptance of humanitarian intervention has been a gradual change in the concept of sovereignty. As traditionally conceived, the principle of sovereignty emphasises a state’s freedom from external interference, so that it can pursue whatever policies it likes within its own boundaries. Although this notion of sovereignty as authority provided a legal and normative barrier that weaker states could use to fend off the interference of larger states, it presented the leaders of certain states with what was essentially a free hand to violate their citizens’ human rights with impunity. Humanitarian intervention, from this perspective, is unjustifiable. Indeed, a key aspect of the traditional notion of sovereignty is the non-intervention principle.
This notion of sovereignty as authority, however, is no longer sacrosanct.5 The concept of sovereignty has been gradually changing to one of sovereignty as responsibility, the responsibility to uphold citizens’ basic human rights. The report by the International Commission on Intervention and State Sovereignty’s (ICISS) in 2001, The Responsibility to Protect (generally referred to as ‘R2P’), has been a key development in this context.6 R2P argues that if a state does not protect the human rights of its citizens, such as in cases of genocide, war crimes, ethnic cleansing, and crimes against humanity, sovereignty is temporarily suspended, and there is an international responsibility to respond by undertaking humanitarian intervention. This notion of a ‘responsibility to protect’ has, to a certain extent, caught on in policy-making circles. Most notably, at the 2005 UN World Summit, states agreed that there exists a universal responsibility to protect populations and indicated their preparedness to undertake humanitarian intervention ‘should peaceful means be inadequate’ and when ‘national authorities are manifestly failing to protect their populations from genocide, war crimes, ethnic cleansing and crimes against humanity.’7
Despite this agreement, it remains unclear which particular agent in the international community should act in response to the mass violation of basic human rights. This question arises because, as things stand, there is not an obviously legitimate institution to undertake humanitarian intervention. The UN might, at first glance, appear to be the most appropriate agent, yet its failures in Rwanda and Bosnia have cast significant doubts over its credibility as a humanitarian intervener. There are a number of other potential agents of humanitarian intervention, but, as I will discuss, none of these other agents are stand-out choices.
Yet the moral stakes could not be much higher. Which particular agent undertakes humanitarian intervention has substantial implications for, firstly, those suffering the humanitarian crisis. Thousands of peoples’ lives, security, and future depend on which particular agent intervenes. Who intervenes also has significant implications for those individuals who collectively form the intervener. These individuals may face increased taxation, decreased spending on public services, military casualties, but, at the same time, may enjoy an improved international standing. Moreover, there are significant implications for the international system as a whole as well. For instance, an illegitimate intervener might weaken international law and order and destabilise certain regions and areas, such as by creating refugee flows.
As a result, this article will answer the question, ‘who should intervene?’ Should it be the UN, NATO, a regional organisation (such as the African Union), a state or group of states (perhaps with the authorisation of the UN Security Council), or someone else? The first part of the article (I) discusses the qualities required for an intervener to be legitimate. I argue for a particular normative conception of legitimacy that asserts the moral significance of an intervener’s effectiveness. The second part (II) considers which of the current agents of humanitarian intervention (if any) have the qualities identified and should therefore be the preferred choice to undertake humanitarian intervention. In the final section (III), I consider ways of improving the preferred agents’ willingness to act.
But before beginning, I need to clarify four points. First, my use of the concept of legitimacy draws on Allen Buchanan’s account of political legitimacy.8 This account is normative in that it is concerned with the morality of political power, and, as such, differs from sociological legitimacy (or ‘perceived legitimacy’). As Daniel Bodansky notes, there is ‘a conceptual difference between saying, “the Security Council is legitimate”, and “the Security Council is accepted as (or perceived as) legitimate”’.9 For Buchanan, ‘[w]hether an entity is politically legitimate depends on only whether the agents attempting to wield political power in it are morally justified’.10 So, legitimacy pertains to agents – it is an ‘agent-justifiability question’. Rather than the focus being on whether a particular action is justified, the concern is with the justifiability of the agent undertaking the act. Hence, we need to know the qualities of an agent (i.e., intervener) that would mean it could justifiably wield power (i.e. to undertake humanitarian intervention). My aim is to determine what these features are.
This focus on the agents of humanitarian intervention distinguishes my discussion from other accounts of humanitarian intervention, which tend to concentrate on whether, when, or why a particular intervention is justifiable. For the purposes of this article, I largely assume that humanitarian intervention is justified in certain circumstances, that is, when a serious humanitarian crisis arises. My concern instead is who should intervene in these circumstances. And whilst there is some overlap between the two questions on certain issues, my approach concentrates much more on the institutional questions – the qualities needed for an intervener to be legitimate – than most other accounts, which tend not to address directly these questions and which instead devote their energy to the question of just cause.
Second, it is important to make clear that my use of the term legitimacy does not necessarily imply legality. It might be true that those interveners who can legally undertake humanitarian intervention are also morally legitimate, but we should not assume that this is the case. Although some international lawyers automatically equate legality and legitimacy, this usage is not consistent with an apparently growing trend in international law to distinguish between the two. The most famous example of this is the Independent International Commission on Kosovo’s assertion that ‘the intervention was legitimate, but not legal, given existing international law’.11
Third, I take legitimacy to be scalar, that is, a matter of degree. A number of different qualities contribute to the legitimacy of an intervener. To be fully legitimate, an intervener needs to possess all the relevant legitimating qualities. But an intervener does not have to possess all of these qualities in order to have an adequate degree of legitimacy. Any combination of qualities is acceptable, as long as they each contribute enough legitimacy so that, when added together, the intervener has an adequate degree of legitimacy. Hence, this approach is cumulative: the legitimacy of an intervener depends on the combined contribution of the various qualities that it possesses. This differs from a categorical approach. On a categorical approach, an intervener would need to possess all of the relevant qualities in order to be legitimate. If it were to lack even one quality, it could not be legitimate.12 By contrast, on the scalar approach I adopt, an intervener that lacks one quality could still have an adequate degree of legitimacy, depending on the other qualities that it possesses. Notwithstanding, to be fully legitimate, an intervener will need to possess all of the relevant qualities.
Lastly, I do not presuppose that, when a humanitarian crisis arises, there will be a large pool of interveners ready and willing to undertake humanitarian intervention from which we can select. For varying reasons, there is often an unwillingness and a lack of commitment to undertake humanitarian intervention. My aim is first and foremost normative: to indicate whose intervention we should prefer when a humanitarian crisis arises. But I will also indicate who would be the next best choices, if the first choice decides not to intervene. In addition, towards the end of the article, I will consider ways of achieving these goals, that is, of improving the willingness and commitment of most legitimate interveners, so that in the future we will have more willing – and better – interveners to choose.13
I. Legitimacy and humanitarian interveners
What are the most important factors for an intervener’s legitimacy? Or, to put it another way, what are the morally relevant qualities when deciding who should intervene? There are a number of potential answers to this question. In what follows, I outline and defend one particular consequentialist approach, the key assertion of which is that an intervener’s effectiveness is the primary (and a necessary) determinant of its legitimacy. When deciding who should intervene, this conception focuses on the intervener that will be the most effective. But it does not hold that this is the only determinant of an intervener’s legitimacy. Other non-consequentialist factors, such as an intervener’s internal and external support and fidelity to the principles of jus in bello, matter to a certain degree, although they are less important than effectiveness.
Before going any further, it is worthwhile spending some time considering why we should take the consequences of an intervener’s action seriously. The notion that an intervener should be effective is intuitively appealing. Indeed, in the normative debates surrounding humanitarian intervention, one subject that continually arises is the effectiveness of humanitarian intervention. The discussions on how, when, and most notably, whether, humanitarian intervention should be undertaken all revolve around this issue. As Eric Heinze argues, consequentialist thinking on humanitarian intervention is the dominant and common-sense moral approach, since it identifies the most morally relevant concerns.14 Those who are sceptical of intervention can cite the failure of the 1991 UN and U.S. interventions in Somalia and UN action in Bosnia as examples of the ineffectiveness of intervening to save lives. John Stuart Mill, for example, defends the principle of non-intervention because humanitarian intervention is unlikely to be successful, given the importance of self-determination.15 Those more favourable to intervention, on the other hand, can highlight the successes of NATO’s 1999 action in Kosovo, UN-authorised Australian-led action in East Timor, and Indian action in Bangladesh.
Despite their differing empirical judgments, what both sides agree on is the importance of intervention being successful. If humanitarian intervention is not successful, then it should not occur; but if it is, perhaps it should. The basis of this highly plausible notion is a certain consequentialist logic: if intervention in another political community is to be undertaken in order to achieve a humanitarian outcome, it matters that it should achieve that humanitarian outcome.
From this intuitive notion follows another: those that undertake humanitarian intervention should be successful. If the UN, for instance, is to intervene in Burundi, it should do so effectively. This is a frequent requirement made of interveners both in the academic literature and by those involved with the practice of humanitarian intervention. For instance, in his address to the 54th session of the UN General Assembly, Kofi Annan called upon member states to unite in the quest for more effective policies to stop egregious violations of human rights.16 Furthermore, the Just War tradition typically requires war to have a reasonable prospect of success and to be proportionate. These two criteria can be interpreted as requiring that those undertaking war should be expected to be effective and, when applied to humanitarian intervention, that interveners should have a good prospect of success.17
At the very least, then, a degree of consequentialist thought on humanitarian intervention is appealing. This helps to provide some initial backing for my claim that an intervener’s effectiveness is the primary determinant of its legitimacy.
Three types of effectiveness
To see more clearly why an intervener’s effectiveness is such an important consideration for its legitimacy, it helps to distinguish between three types of effectiveness. The first type of effectiveness most clearly demonstrates the significance of an intervener’s effectiveness and why effectiveness is the primary determinant of an intervener’s legitimacy. This is an intervener’s ‘local external effectiveness’, which depends on whether an intervener is likely to promote or harm the enjoyment of basic human rights of those in the political community that is subject to its intervention. In other words, to be locally externally effective, an intervener needs to be successful at tackling the humanitarian crisis. For instance, if the UN is to intervene in Darfur with the purpose of helping the Darfurians, it is vital that its intervention should benefit the Darfurians. If it were likely to make the situation even worse, then it would be locally externally ineffective and (in all probability) would not be legitimate.
Why is this a highly significant factor for an intervener’s legitimacy? If an agent is to go to the extreme of undertaking military action in another state, with the risk of great harm to the citizens of this state, in order to end, decrease, or prevent a humanitarian crisis and assist (some of) these individuals, it seems paramount that it will actually assist these individuals. More specifically, my reasoning is as follows. The degree of human suffering typically involved in the violation of basic human rights is perhaps the greatest moral wrong, more morally urgent than other moral concerns. That is, we tend to think, generally speaking, that rape, torture, and murder are more morally significant than repression, inequality, etc. A humanitarian crisis usually involves the mass violation of basic human rights. As such, it involves the worst moral wrong on a massive scale: mass killing, mass rape, mass torture, and so on. Accordingly, it is of the utmost moral importance that the humanitarian crisis is effectively tackled. It follows that an intervener’s local external effectiveness is vital. This is because a locally externally effective intervener will tackle the humanitarian crisis and therefore prevent, reduce, or halt the worst moral wrong on a massive scale – the mass violation of basic human rights.
The second type of effectiveness is ‘global external effectiveness’. This depends on whether an intervener is likely to promote or harm the enjoyment of basic human rights in the world at large. The significance of an intervener’s global external effectiveness is best seen in its negative aspect: an intervener that undertakes humanitarian intervention that severely harms the enjoyment of basic human rights in the world at large loses legitimacy. For instance, an intervener could destabilise the neighbouring states of the target political community (perhaps by creating a large refugee flow) and therefore severely harm the enjoyment of basic human rights of those in neighbouring states.
The third type of effectiveness is an intervener’s ‘internal effectiveness’, which depends on the consequences for the intervener’s own citizens. Like global external effectiveness, the importance of internal effectiveness is also typically best seen in its negative aspect. In most cases, we would not expect an intervener to make an improvement in its own citizens’ enjoyment of basic human rights, given the costs of humanitarian intervention in terms of lives and resources. Instead, humanitarian intervention is likely perhaps to decrease some of its citizens’ enjoyment of basic human rights, e.g. its soldiers who are wounded and killed in action. But this decrease must not be excessive: an intervener that undertakes reckless humanitarian intervention, which will severely decrease its own citizens’ enjoyment of basic human rights (perhaps by incurring heavy casualties among its own forces or by bankrupting the state), loses legitimacy (i.e. it is ineffective overall). Thus, an intervener’s legitimacy depends also on its internal effectiveness and typically how internally ineffective it is.
Given the importance of an intervener’s being effective in these three senses, it follows that an intervener’s overall effectiveness is a necessary condition of its legitimacy. If, when combining its local external effectiveness, global external effectiveness, and internal effectiveness, an intervener is ineffective overall, it cannot be legitimate. If an intervener’s effectiveness were not a necessary condition of its legitimacy, an intervener could be legitimate even though it (1) failed to make an improvement in the humanitarian crisis – and so lacked local external effectiveness, (2) undertook intervention that was excessively costly to human rights worldwide – and so was extremely globally externally ineffective, and/or (3) undertook intervention that was excessively costly to its citizens – and so was extremely internally ineffective. Accordingly, an intervener must be likely to make an overall improvement in the enjoyment of basic human rights to be legitimate. A similar point is made by Jane Stromseth:
‘humanitarian intervention should have a reasonable prospect of success in stopping the atrocities that triggered intervention in the first place. Otherwise, the interveners will simply be exposing their soldiers and the target population to life-endangering situations without the hope of success that justifies the risks to be borne’.18
The standard way that the intervener will be effective overall is by being substantially locally externally effective, that is, by successfully tackling the humanitarian crisis. It follows that, in most cases, an intervener’s local external effectiveness is a necessary condition of its legitimacy. An intervener cannot be legitimate if its intervention is likely to worsen the situation of those suffering the humanitarian crisis.
How should we measure the effectiveness of an intervener? The expected increase in the enjoyment of human rights for all three types of effectiveness should be considered in the long-term and should be judged by comparison with the counterfactual of nonintervention. Assessing intervention in the long-term does not mean that short-term results are of lesser importance. Where possible, the intervention ‘must be tailored to suit these long-term objectives, though… securing an immediate cessation of hostilities will, in some cases, trump other objectives’.19 If a state’s intervention is expected to save 50,000 lives in the short-term but cost 40,000 lives in the long-term, this is still a positive outcome in the long-term (10,000 lives have been saved).
To be effective in these three ways, interveners need to have a number of characteristics. These include: the required military and non-military resources; a suitable strategy and mandate to be able to use successfully these resources; the necessary commitment to ensure a lasting resolution to the humanitarian crisis; and the ability to intervene in a timely manner, i.e., quickly when the situation is ripe for humanitarian intervention.
However, an intervener’s success is not determined fully by the degree to which it possesses these characteristics. Circumstances can affect an intervener’s effectiveness in two ways. First, an intervener will have a different expectation of success in different circumstances. For instance, there may be more local resistance to the intervention by State A in State B than in State C. Hence, the probability of success varies according to the situation. Second, an intervener will have greater opportunity to achieve a large-scale success in some situations than in others. Where there is a terrible humanitarian crisis and the potential for great harm to a large number of individuals, such as genocide, there is more scope for an intervener to achieve extremely beneficial consequences by tackling the crisis and preventing the harm. Other less (although still) serious situations, such as the oppression of political opposition, present less scope for an intervener to achieve extremely beneficial consequences. Hence, the magnitude of the potential success varies according to the circumstances.
Now to the crux of the matter: when an intervener has a high probability of achieving a success with a large magnitude, effectiveness may be sufficient for it to have an adequate degree of legitimacy. An intervener may be legitimate, for instance, simply because it is highly likely to prevent genocide. This is the case even if it lacks other qualities. On a scalar approach, a legitimate intervener does not need to possess all of the morally relevant qualities; it need only have enough of these factors in order to possess an adequate degree of legitimacy. An intervener can have an adequate degree of legitimacy by achieving hugely beneficial consequences. The likely achievement of these extremely beneficial consequences means that it is likely that extreme levels of human suffering will be prevented. The good achieved by this intervention is likely to outweigh any other moral problems that come from the intervener’s not possessing other qualities. Suppose, for example, if in the beginnings of the genocide in Rwanda, the U.S. had been willing to intervene and was highly likely to do so effectively. Given that this could have saved hundreds of thousands of lives, the fact that it lacked other qualities (for instance, it might not have consulted with the Rwandans and its soldiers might have committed abuses) would not have undermined its general legitimacy as an intervener.
So, effectiveness can, in unusual circumstances where hugely beneficial consequences are more than likely, be sufficient for an adequate degree of legitimacy. In most cases, however, effectiveness will not be sufficient because an intervener will not have a very high probability of achieving a very sizeable success. Normally, then, an intervener’s legitimacy will also depend on the degree to which it possesses other, non-consequentialist qualities, such as fidelity to the principles of jus in bello and internal and external support (I establish the importance of these qualities below). Moreover, even where hugely beneficial consequences are more than likely, and effectiveness is sufficient for an intervener to have an adequate degree of legitimacy, the intervener will not be fully legitimate unless it has all the relevant qualities. Hence, effectiveness can, at most, be a sufficient condition for an adequate degree of legitimacy. In the majority of circumstances, it is not even sufficient for this.
Effectiveness is, then, a substantial consideration when deciding who should intervene. It is a necessary condition of legitimacy and even occasionally sufficient for an adequate degree of legitimacy. But why is effectiveness only sufficient for an adequate degree of legitimacy in exceptional cases? Why is it not sufficient for full legitimacy in all cases? On an extreme consequentialist approach, effectiveness is not a primary determinant of legitimacy. It is the only determinant. This approach therefore presents a different sort of challenge to the conception of legitimacy for humanitarian intervention that I am proposing. It claims that holding that effectiveness is the primary determinant of an intervener’s legitimacy does not go far enough. However, in placing all moral weight on consequences, an extreme consequentialist approach disregards other moral qualities, which are also significant for an intervener’s legitimacy. The legitimacy of an intervener also depends on three other qualities that have intrinsic as well as instrumental value: the intervener’s fidelity to principles of jus in bello; its internal support, and its external support.
Fidelity to the principles of jus in bello
The first intrinsically valuable quality is the conduct of the intervener during intervention. This is often mentioned as an important consideration.20 Some insist that an intervener must follow principles of international humanitarian law.21 Others frame this requirement in terms of Just War Theory, and, in particular, with reference to the principles of jus in bello, principles of just conduct in war.22
There are several principles of jus in bello that can be applied to humanitarian intervention. For our purposes, the three main principles are: a strict rule of noncombatant immunity which maintains that civilian casualties are impermissible; a principle of proportionality which limits the harm that the intervener can cause to combatants; and a restriction on the sort of soldiers the intervener can use (e.g. not child soldiers).
The conception of legitimacy that I have been defending leaves room for some non-consequentialist values by claiming that effectiveness is the primary (rather than sole) consideration. By contrast, an extreme consequentialist approach cannot fully account for these principles of just conduct in war in its conception of legitimacy and is therefore unpersuasive. By placing all moral weight on an intervener’s effectiveness, it marginalises the importance of an intervener’s expected fidelity to these principles of jus in bello. In this context, Heinze claims that a purely consequentialist account ‘has serious problems when employed as part of a theory of the morality of war based on human rights, because it suggests that aggregate human suffering is the only moral concern that need be addressed’.23 He claims that if, for instance, a purely consequentialist principle alone were used to determine proportionality in NATO’s war in Kosovo, it would have been permitted to pursue its primary end of the capitulation of the Milosevic regime unconditionally, regardless of civilian casualties.24
That said, there a consequentialist defence of these principles of jus in bello can be made – an intervener that follows these principles will, generally speaking, face less resistance from the local population.25 But this defence is not wholly reliable. An intervener will sometimes be more successful if it abandons these principles and uses the most efficient means. Yet that an intervener follows these principles of jus in bello matters intrinsically.
Indeed, there seems to be something more to the importance of an intervener’s expected fidelity to the principles of jus in bello than simply whether this improves its effectiveness. This is the distinction between doing and allowing. That is, there is a morally relevant distinction between what one does oneself and what one allows others to do.26 In addition to any instrumental justification, a reason why an intervener’s likelihood of following the principles of jus in bello is important when deciding who should intervene is that an intervener should not itself do harm (specifically, harm that is impermissible according to these principles).27 It would, to a certain degree, be better if an intervener were to allow harm, perhaps thereby being less effective, than for it to target civilians, to use chemical weapons, or to rely on child soldiers. Indeed, it seems more important that an intervener has a satisfactory degree of fidelity to the principles of jus in bello when it is using force for humanitarian purposes than for any other reason. When going to the lengths of using military force for humanitarian purposes, it matters intuitively that an intervener should be likely to do so in a way that is itself humanitarian.
One reason why the doing and allowing distinction matters is because when one does the action, it is oneself that is violating the right, whereas when one allows the action, it is someone else that is violating the right. There is a difference between the government of state A violating state B’s citizens’ rights and the government of state A not intervening to stop the government of state B violating its own citizens’ rights. To be sure, I am not claiming that the difference between doing and allowing is of overwhelming moral significance. In fact, on an absolutist, deontological position according to which the difference between doing and allowing is of absolute moral significance, an intervener could never be legitimate because intervention almost always involves some harm that is impermissible.28 My point is rather that there is, at least, some moral significance in the distinction between doing and allowing. When deciding who should intervene, it matters, to a certain degree, that an intervener will not violate innocent individuals’ rights itself, even though this may ultimately allow more rights to be violated. Thus, who undertakes humanitarian intervention should be determined in part by the non-instrumental importance of an intervener’s following principles of jus in bello.
That said, according to the scalar approach to legitimacy adopted in this article, an intervener can be sufficiently legitimate, even though it does not have a satisfactory degree of fidelity to the principles of jus in bello. As long as the intervener is able to make up in other ways the loss of legitimacy that comes from its not following closely the principles of jus in bello in other ways, its overall level of legitimacy may still be sufficient for it to have an adequate degree of legitimacy. One clear way in which an intervener can make up this loss of legitimacy is if there is a high expectation of achieving extremely beneficial consequences, for instance by preventing genocide. Suppose that there is mass ethnic cleansing – genocide – in Benin. Tens of thousands of civilians of a certain ethnic group are being slaughtered, maimed, and raped every day by government troops and militias. Suppose further that Nigeria intervenes in Benin to stop this ethnic cleansing, and does so very effectively, but in doing so uses conscripts, a number of whom kill and sexually assault the non-combatants they are supposed to be helping. Although Nigeria’s intervention would far from being fully legitimate, the fact that it is effective at preventing genocide means that it would have an adequate degree of legitimacy overall. Hence, if hugely beneficial consequences are highly likely, then effectiveness may be sufficient for an adequate degree of legitimacy and the importance of an intervener’s following these principles of jus in bello can be trumped by the consequences that it will achieve.
This is not equivalent to endorsing an extreme consequentialist approach. Given the intrinsic importance of an intervener’s following the principles of jus in bello, the intervener’s expected effectiveness is sufficient for an adequate degree of legitimacy only in particular circumstances.29 As discussed above, these are circumstances in which the intervener has both a high probability of achieving a success with a particularly large magnitude – in short, when highly beneficial consequences are more than likely. In other cases, effectiveness is not sufficient for an adequate degree of legitimacy because of the non-instrumental significance of an intervener’s following these principles.
Internal and external support
There are two other intrinsically valuable qualities that need to be taken into account in any conception of legitimacy for humanitarian intervention. These are that an intervener should have the support of, firstly, its domestic population and, secondly, those in the political community that is subject to its intervention. The former is what I will refer to as an intervener’s ‘internal support’, the latter an intervener’s ‘external support.’30
Let us begin with the case for internal support. To have internal support, an intervener needs to have the backing of its citizens on the proposed intervention. To be sure, the importance of an intervener having internal support can be justified instrumentally. An intervener that has internal support will, for political reasons, be more likely to commit the financial, military, and diplomatic resources needed for success. However, the instrumental justification does not fully capture why internal support matters. To that extent, there are two intrinsic reasons why internal support matters.
The first is Lockean. It is important that an intervener has internal support of its domestic population because these individuals provide the resources for humanitarian intervention. Generally speaking, it seems intuitive that an individual should have some freedom to determine how his own resources (property) are used. To be sure, there are moral constraints on how someone should use his resources. Notwithstanding, some degree of control over one’s own resources is intuitively attractive. This argument could be applied to any governmental action that uses its citizens’ resources, but it is much more convincing for humanitarian intervention because of the level of resources involved.31 Those individuals from whom the intervener is collectively formed ultimately have to foot the bill for humanitarian intervention, perhaps through significantly increased taxation or greatly decreased public spending elsewhere. It seems right, therefore, that intervention should have their support.32
The second reason is Rousseauian. Internal support is morally significant because of the importance of individuals’ having a voice in the running of their political institutions. Individual self-government here possesses significant value. In Robert Dahl’s words: ‘To govern oneself, to obey laws that one has chosen for oneself, to be self-determining, is a desirable end’.33 As a significant undertaking by the state, it is important that humanitarian intervention be responsive to the concerns of individual self-government by being representative of its citizens’ opinions on intervention. An individual’s freedom to choose whether there should be intervention matters. To be sure, individual self-government is not always an overriding value; rather, more individual self-government is by and large desirable. Occasionally, other moral factors (such as highly beneficial consequences) may trump the importance of individual self-government, but this is not to deny its value.
Having seen why it is important that an intervener as internal support, let us now turn to consider external support. Like internal support, the importance of this quality can be justified instrumentally, but this does not fully capture why an intervener should have external support. Again, there are two intrinsic reasons for the importance of an intervener’s external support.
The first highlights the potential burdens of humanitarian intervention. Those in the community subject to intervention might have to suffer civilian and military casualties, damage to vital infrastructure, increased levels of insecurity, and other costs involved with being in a war zone. Given that these individuals face these burdens, it seems important that an intervener should have their support. The underlying principle here is that an individual should have some choice in the burdens that he or she faces. Those subject to the humanitarian intervention should have their opinions on the intervention taken into account because intervention may have a negative impact on their basic human interests.34
The second argument again highlights the non-consequentialist importance of individual self-government. It is important that an individual should be self-governing even if his opinions, if realised, would not obviously promote his well-being. It follows that a state, coalition of states, or multinational organisation should not intervene to protect those who do not want their political community to be subject to humanitarian intervention. This is the case even if intervention would promote these individuals’ well-being in the short-term, for instance, by protecting them from being the victims of oppression and from the violation of their basic human rights.35
For these reasons, then, an intervener’s internal and external support have non-instrumental value. It follows, then, that the dominance of consequentialist thought on humanitarian intervention is further limited by the intrinsic value of internal and external support. Nevertheless, given that effectiveness can, on occasion, be sufficient for legitimacy, the value of these two qualities, like that of fidelity to the principles of jus in bello, is not absolute. They are not necessary, but still important, factors for the legitimacy of an intervener.