In recent years, ever since the intervention in Kosovo in 1999 and the failure to intervene in Rwanda in1994, much attention has been paid to the question whether either the United Nations or indeed states or alliances like NATO have either a duty or a right to intervene militarily for humanitarian reasons in countries where human rights are being violated. A landmark decision of the United Nations was the Resolution passed in 2005 on The Responsibility to Protect. The authorisation of military intervention under the clause ‘we are prepared to take collective action, in a timely and decisive manner … should peaceful means be inadequate and national authorities are manifestly failing to protect their populations’ is only a small part of a wide range of measures accepted to reduce the incidence of human rights violations. But it has attracted a large amount of critical attention.
This article does not attempt to assess the legal issues that are raised about whether the UN or states have either legal duty or legal right to engage in military intervention. Nor does it examine the kinds of criticism often made that the use of it to justify interventions is a mask for realist agendas and motives. It simply takes R2P as a convenient peg on which to hang some ethical questions concerning the justification for military humanitarian intervention. It suggests three kinds of difficulties. First, if one adopts a consequentialist way of reasoning, it is plausible to argue that such reasoning is unable to establish a case, given a sufficiently rich account of the kinds of consequences that need to be considered. Second, consequentialism does not take the right approach to the relationship between means and ends; a non-consequentialist account that recognises that the means ought to be ethically consistent with the ends pursued is to be preferred.
Third, between these two positions consideration is also given to a way of thinking which gives military intervention a prima facie ethical plausibility. It is a common human response to human suffering that is caused by other human beings to try to stop it and to believe it is one’s duty to stop it if one can, and to believe that there is more moral urgency about so doing compared with when the human suffering – maybe be just as bad – is caused by other factors, such as natural disasters, or other human practices, such as economic oppression or systemic violence. A more robust form of this argument is actually expressed in rights language when it unjustifiably privileges the preventing or stopping of the deliberate violation of human rights by other agents over the promotion of human rights and protecting human rights in the face of suffering caused by natural causes or the impacts of human institutions. I shall suggest that the plausibility of privileging responses to human ills caused deliberately by other humans is illusory. I shall focus on the rights version because on the face of it the non-consequentialist rights account may seem to make the case in a way that consequentialist thinking does not for humanitarian military intervention.
So I am going to identify three difficulties with humanitarian military intervention: first, it involves inadequate consequentialist reasoning; second it unjustifiably privileging of the duty to respond to the violation of human rights by other humans over promoting human rights fulfilment generally; and third. It fails to harmonise means with ends.
I consider these arguments from a cosmopolitan point of view, since any serious case for humanitarian intervention would appear to be based on cosmopolitan premises – though this may not always be made explicit – that it is concern for the suffering of human beings anywhere in the world that motivates it. If someone prefers an ethical case for military humanitarian intervention which is claimed not be cosmopolitan in the relevant sense, then my argument may not apply to this case.
These arguments should be seen as supplementing the kinds of just war considerations which need to be applied to any military intervention, humanitarian or otherwise, such as last resort, proportionality and non-combatant immunity. How far these arguments apply to all other cases of war is an open question. Furthermore these considerations are expressed in such a way that, whilst they do not provide an absolutist ban on all military intervention, they certainly raise the moral stakes.
But the way I discuss these moral issues, which I regard as belonging to the more abstract end of normative moral theory, is only per accidens to do with the ethics of war, since they have a bearing on how we do ethics quite generally. That is, how we think of consequences, whether we think there is something special about responding to human ills caused directly by other human beings and what we think of the relationship between means and ends, all have a bearing on ethics generally, and indeed have relevance, whether or not my conclusions concerning humanitarian intervention are accepted.
Background to R2P and the cosmopolitan turn
The NATO intervention in Kosovo in 1999 provoked a lot of controversy partly because of the lack of clear UN Mandate, but it certainly reflected a considerable amount of support for the idea that the international community ought to try and stop human rights violations inside countries. Such action was not simply condemned as earlier interventions had been e.g. by India in East Pakistan in 1971 and by Tanzania in Uganda in1979 because they violated the UN Charter provisions (especially Articles 2.4. and 2.7) about not intervening in the internal affairs of members states without their consent.
International jurists and others devoted a lot thinking to how the UN’s position might be adjusted within an international law framework, and a report was produced by the International Commission on Intervention and State Sovereignty called Responsibility to Protect. This eventually led to the World Summit Outcome Document in 2005. This was a broader and more politically oriented document than the document of 2001. It contained three pillars of action, and it is in the third pillar that the possibility of military intervention is stated. A Report of the Secretary General on Implementing the Responsibility to Protect in 2009 gave further consideration to implementation issues.
The move towards accepting the general idea of the responsibility to protect represents an important example of a shift in thinking about the UN as having cosmopolitan goals. The UN is of course in international relations theory an example of the Westphalian model of global political relations, since it is premised on nation-states continuing to be the key actors in global affairs. It is an inter-national organisation. Nevertheless part of its rationale is to promote universal goals and improve the life-conditions of all human beings – exactly what the cosmopolitan presses for.
It may be thought that there is nothing new here since the UN right from the beginning accepted global goals, e.g. in the Universal Declaration of Human Rights (1948) and the subsequent Covenants of 1966. However there is a significant shift. The earlier Human Rights documents certainly asserted universal values but it was assumed that the responsibility to protect and promote human rights lay primarily within nations-states, with the international community only coming to states’ assistance with the consent of those states. What R2P represents is a recognition that the responsibility to protect human rights is a stronger one than merely assisting states.
Cosmopolitanism is stronger than global ethics as such, if global ethics is only about universal values. Certainly it is one thing to assert that there are values which are universal, that is either universally accepted or universally applicable (even if not fully accepted by all), it is quite another to assert that those responsible for protecting and promoting these rights lie outside particular states and their members. Universal value does not in itself entail transboundary responsibilities. However most people nowadays who favour global ethics do generally assume the latter. Cosmopolitanism makes it explicit.
Pogge for instance characterises cosmopolitanism as follows:
Three elements are shared by all cosmopolitan positions. First, individualism: the ultimate units of concern are human beings, or persons – rather than, say, family lines, tribes, ethnic, cultural or religious communities, nations, or states. … Second, universality: the status of ultimate unit of concern attaches to every living human being equally – not merely to some sub-set, such as men, aristocrats, Aryans, whites, or Muslims. Third, generality: this special status has global force. Persons are ultimate units of concern for everyone – not only for their compatriots, fellow religionists, or suchlike. (Pogge World Poverty and Human Rights (2002): 169)
It is the third feature – generality or global force – that is the crucial one from our point of view. So what happens to people’s rights is in principle of concern – practical action-guiding concern – to people anywhere.
Cosmopolitans have good reason to welcome the general idea behind R2P, since it is quite an example of the cosmopolitan idea of transboundary responsibility. So it may seem natural that cosmopolitans would welcome the idea of military intervention to stop human rights violations – from a cosmopolitan point of view such military action would seem justified in a way that many military actions would not. Indeed there is no reason why a cosmopolitan might not endorse such military intervention. Cosmopolitanism in itself does not point to one position or another position; it depends on the detailed ethical norms accepted and how one reads the consequences of actions. What follows is a particular interpretation of cosmopolitanism.
There is a further interpretative issue which needs stating: how strong is this transboundary obligation? Cosmopolitanism may be presented in a strong sense: we ought to do all that we can to promote human well-being anywhere, by avoiding harming others, by helping them positively and through opposing harm done by others. Cosmopolitanism may be presented in a weaker sense: beyond the sphere of reasonable self-interest, care for particular others and other affiliations, one has significant obligations to be concerned with the well-being of others including others anywhere in the world, by avoiding harming them, by helping them positively and through opposing harm done by others.
The above formulations bring out a consequentialist feature of cosmopolitanism: cosmopolitanism is about promoting good consequences in the world in various ways. This does not mean that a cosmopolitan is a consequentialist in the full-blown form taken by some ethical theories like utilitarianism. It may be that other moral norms e.g. about justice or how we do things (the means) are important as well, as we shall see. But there is no doubt that cosmopolitanism is oriented towards the question ‘how can we create a better world for all?’ so the question ‘what kinds of actions should we perform as cosmopolitans?’ turns to a large extent on the question ‘what kinds of action will indeed advance human well-being in the world?’ And it is to this aspect of cosmopolitan thinking that I turn first. Clearly advocates of military humanitarian intervention believe that it will achieve some good or a greater good than not doing so.
Critiquing this assumption, to which we turn first, cuts across the question whether one adopts a strong or a weak view of the extent of obligation. Whichever view one takes, the problem we identify arises. Even if one thought theoretically that the stronger view ought to be accepted and applied, the reality is in any case that the representatives of states, insofar as they are motivated by cosmopolitan considerations at all, will only accept the weaker thesis. So the practical question is: to the extent we think we ought to act for the good of others in the world generally, is military intervention the appropriate way of realising that goal?
Problems with cosmopolitan consequentialist reasoning
There are actually three types of problem with cosmopolitan consequentialist reasoning justifying military intervention: it has too narrow a focus; it fails to see that the use of force is counterproductive; and it is inconsistent with the ethics of the means. The last objection, which I consider later is of cure a rejection of a simple consequentialist way of ethical reasoning. But the first two constitute internal criticisms: that is, even if you accept that considering the best outcomes is all that matters ethically, the reasoning is unlikely to be successful. These problems as I noted earlier arise whether one takes cosmopolitanism in a strong sense or in a weak sense.
What if states were to do everything in their power (strong version) to promote good/reduce evil (or promote human rights), or (weak version), to the extent that they ought to be concerned with the good of humanity generally, were to promote the best balance of good over bad outcomes?
Military intervention has in fact to survive two consequentialist tests: First, is it counterproductive? Second, in the general spread of things to be done, is it the best use of resources (finance, infrastructure capacity, expertise etc.)? First there is the general argument familiar to anti-war theorists that violence begets violence usually, and that even when there appear to be short-term successes, the longer-term consequences may be more be negative, both is respect to the fact that the conclusions of wars often sow the seeds of future conflicts, and in respect to the fact that the resort to war perpetuates a culture in which the resort to war is seen as an acceptable way of solving conflicts. Henry Ballou the 19th pacifist vividly catalogues human woes perpetuated by the resort to war. Three recent interventions in Afghanistan, Iraq and Libya, all partly justified by humanitarian considerations, hardly paint a picture pointing in the opposite direction.
But the point I want to focus on is a rather different one. Even if the resort to war (or more accurately a reasonable assessment of a decision to go to war) survived the first test, it needs to face the question: in the general spread of things to be done, is it the best use of resources? That is, if one took a genuinely cosmopolitan approach to what would advance human well-being most successfully, would one focus on coercive intervention, or rather on a whole range of other measures to promote human rights fulfilment? There are several considerations that favour the latter approach.
Generally speaking cooperative strategies are more effective than coercive strategies; apart from wasted energy involved in confrontation and in coercing many who are unwilling and oppose what one is doing, using military force of course involves killing and maiming people and destroying property and infrastructure central to human development. The alternatives of not just greater and better quality of aid but changes in the international and economic order that lead to less inequality are likely to be more productive to human well-being. Apart from the immediate benefits of better development programmes, the result will be that the occasions for armed conflict are reduced. If we take cosmopolitanism seriously it should be concerned not merely with tackling extreme poverty because extreme poverty undermines human well-being but also promoting the general conditions of peace, since lack of peace in the form of violence or war undermines human well-being in different but equally significant ways. (In practice of course violence and poverty are often causally linked.) This duty to promote the conditions of peace anywhere I have called ‘cosmopolitan pacificism’. As an aspects of this it seems pretty clear, if one reasons consistently as a cosmopolitan, that maximum efforts to promote human well-being include reducing both the current levels of armaments and the arms trade: as Noel-Baker once said ‘the arms race is killing’ (because of the diversion of resources from live-saving and other socially useful activities). The proliferation of arms through the arms trade is a major factor in widespread military violence, and from a cosmopolitan point of view the idea of profit-making in the manufacture of arms is ethically questionable.
The general point here is that a consistent cosmopolitan who wishes to do what helps people anywhere achieve well-being and avoid what impedes it is unlikely to favour military intervention, both because there are so many other things to be done and because as an effective use of whatever resources one is prepared to commit to a better world, the military use of resources scores low. So even if a cosmopolitan consequentialism were accepted, it is doubtful if consequentialist reasoning works in favour of military intervention.
But it is in fact inadequate theoretically because of two features of ethical thinking that ought to be incorporated into a cosmopolitan theory, namely that we need an adequate theory of global justice and human rights, and a proper account of the relationship between means and ends. Cosmopolitanism should be seen as a theory not merely about promoting human well-being anywhere (including avoiding harming people and helping people harmed in various ways), it is about doing these things justly with a proper respect for human rights and about observing certain moral norms concerned with how things are done. In saying that these other parts of cosmopolitan thinking are important, I am not suggesting that all cosmopolitans think this; rather this is my preferred account of g cosmopolitanism.
Justice and Human Rights
My point about justice and rights can be introduced via a consideration of an argument that quite independently of the consequentialist argument we considered, suggests that military intervention is justified. There is a special obligation of justice to stop the injustices done to others by third parties. Even if it was not good from a consequentialist point of view, it might still be the right thing to do. This might be seen as a corollary of the more familiar case of a country coming to the aid of an ally that was attacked. However whether or not coming to the military aid of an ally is justified, it involves special features, and I do not pretend to consider them (though I should note that from a cosmopolitan point of view, the answer is not self-evident).
The general case of intervening to stop injustice or to prevent (the continuation of) human rights violations may seem like a peculiarly strong claim of justice or a special implication of talking about rights being actively violated by other human beings. But this is only because it is contrasted with and privileged over whatever else justice or concern for human rights might require of us from a cosmopolitan point of view. I suggest that a proper regard of what justice and human rights might require does not privilege such intervention, and so one source of the appeal of military intervention is removed. When one adds the thought that such intervention fail the earlier consequentialist test, then the case lapses.
I continue the discussion by looking at human rights in particular. An anti-consequentialist may then say: surely there is something about rights that requires the privileging of responding to the violation of human rights? Surely that takes priority over the general promotion of human rights fulfilment? Arguably this is not the case. This indeed is the key issue: is ‘A ought to stop others violating the rights of others’ parallel to ‘A ought not to violate the rights of others’? Or it is parallel to ‘A ought to help others fulfil their human rights whatever the cause of their rights not being unfulfilled’? It looks parallel to the first because they are both about violating rights. But in fact it is like the second because it is about our response to the failure to realise rights. In both cases we are not doing any violating. Indeed ‘A ought to stop others violating the rights of others’ is a special application of ‘A ought to help others fulfil their human rights whatever the cause of their rights not being unfulfilled’. Once this is understood, we need to consider what kinds of response are most appropriate, and we are back to the general considerations I mentioned earlier.
What however about the objection: how can we let people die or suffer at the hands of others when we could intervene and stop it (or reduce it)? The reply is: how can we let people die when we could intervene in all sorts of ways, whether because of natural causes or because of systemic human causes? We do let people die and suffer in both kinds of case, and when we do intervene in one way or other, we do so in a highly selective and limited way. Of course from a cosmopolitan point of view, generally speaking we ought to be more willing than we usually are to take appropriate action to promote or tackle the impediments to human rights fulfilment. It does not follow from this that we ought to engage in all the different kinds of intervention there are – as that would include justifying military intervention – but rather that what we ought to do should be effective and not counterproductive. All these considerations shows is that if we were simply wrong to let others die at the hand of others, then it would be simply wrong to let other die when we could intervene in all sort of other ways. Even the most generous of supporters of charities should acknowledge that they do not do all that they could.
In any case, if we accept (as most moral theorists would acknowledge) that it is more important to avoid harming people/violating their rights than to help them (to realise their rights), then any story that emphasises rights fulfilment should place an emphasis on that. It may be thought that it is easy to avoid harming people/violating their rights. But in fact if one recognises all the complicated and indirect ways in we may be contributing towards harming others (e.g. being part of economic systems that harm, having a large environmental footprint), it will be apparent that there is plenty to do to reduce our negative impacts, before we zoom in on one particular kind of response to human rights violation by others. This is not to suggest that we do not do appropriate positive things to help in the world – far from it – but it helps to make the appropriate response to human rights an immensely complicated one, and saves us from a ‘holier than thou’ attitude that since we – people in the rich North –are not harming people, we are entitled to impose our solutions elsewhere.
Means and ends
Another kind of reason why military intervention is morally problematic turns on the nature of the relationship between means and ends. A saying of Gandhi’s much quoted by Quakers reads: ‘The means are the ends in the making’, and this captures what I want to suggest, which is that the means we take should reflect the value(s) in the ends we are promoting. If we promote human rights we should do so by respecting human rights in the actions we take to that end and so on. In fact this quotation has two interpretations which whilst expressed differently come to much the same general insight. First there is what may be called an existential/metaphysical claim: our real ends are already revealed in the means we take. If I pursue my ends by deceit then I intend deceit/I am the kind of person who wills deceit in the world. Second, there is a normative claim: my means ought to be consistent with the ends I am pursuing: I ought not to pursue justice unjustly, promote peace unpeacefully etc. (if one were to put it in a quasi-Kantian formulation one might say ‘So act that your means are value-consistent with your ends’!)
Clearly in the case of humanitarian intervention action is taken to promote human rights or restore a just peace which involves violence and the violation of human rights in the process. This looks like a simple pacifist objection to the use of force which would apply to all kinds of organised violence, but I present it as a more general ethical test about the alignment of means to ends. A simple statement of the pacifist position often invites the statement of the non-pacifist position – end of story. What this higher level normative principle does is open up a reflection about how satisfactory the pursuit of ends is, if it involves serous departures from the values embedded in what we are trying to bring about. It is different from a pacifist position in three respects: it is about all the values we accept and express, not one value; it is about the relationship between means and ends, whereas pacifism is, at least generally, about the wrongness of killing, irrespective of what our ends might be; it is more open to a non-absolutist interpretation than pacifism is, at least as it is usually presented. It could indeed be treated as an absolutist principle, but it would be very difficult to live by completely. It is more useful as an ethical standard for assessing the ethical quality of our acts. If we accept its force we recognise in a general way that there is something unsatisfactory about courses of action which involve failing to live up to the values promoted. In the real world there may be circumstances where it is important to do so and indeed the right thing to do is to take means that are in tension with the end pursued. But it provides a test of how civilised a society has become. This of course is a persuasive definition of what being civilised means, but it does seem that to the extent to which its norm has been internalised and the extent to which in most normal circumstances it is seen as guiding moral decisions is one test of moral progress.
Cosmopolitanism is also, one can argue, a mark of progress in expanding the circle of whom one takes morally seriously. So there is a cosmopolitan twist to the argument about humanitarian intervention. This point about the consonance of means and ends should be particularly relevant to the understanding of a cosmopolitan ethic, and hence to what kinds of action it recommends. It is after all a civilising project.
 This article was originally given as a lecture at the conference in honour of Mikael Karlsson on 19th April 2013 in Akureyri. The idea for it came in response to a lecture I heard in January 2013 in Aberdeen when Alexandra Buskie, from UNA UK, gave a lecture on the Responsibility to Protect.
 See Dower, N., The Ethics of War and Peace, Cambridge: Polity Press, 2009.
 The UN Articles read:
2.4 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.
2.7 Nothing contained in the present Charter shall authorize the United Nations to intervene in matters which are essentially within the domestic jurisdiction of any state or shall require the Members to submit such matters to settlement under the present Charter; but this principle shall not prejudice the application of enforcement measures under Chapter VII. (UN, The Charter, New York, 1945)
 For more detail see http://www.who.int/hiv/universalaccess2010/worldsummit.pdf and http://www.un.org/en/preventgenocide/adviser/responsibility.shtml, and Buskie, A., ‘The Responsibility to Protect and the prevention of mass atrocities’, UNA-UK, 3, Whitehall Court, London, February 2013. I include the analysis she presented of the three pillars:
“Each individual state has the responsibility to protect its populations from genocide, war crimes, ethnic cleansing and crimes against humanity. This responsibility entails the prevention of such crimes, including their incitement, through appropriate and necessary means.”
“the international community should, as appropriate, encourage and help states to exercise this responsibility”…”we also intend to commit ourselves, as necessary and appropriate, to helping States build capacity to protect their populations … and to assist those which are under stress before crises and conflicts break out”
“the international community…also has the responsibility to use appropriate diplomatic, humanitarian and other peaceful means…to help protect populations…we are prepared to take collective action, in a timely and decisive manner…should peaceful means be inadequate and national authorities are manifestly failing to protect their populations”
 Ballou, A. (1866), ‘Christian non-resistance’ in S. Lynd (ed.), Nonviolence in America: a Documentary History, NY: Grove Press, p.31. Quoted in Wasserstrom, R. (1970), ‘On the morality of war: a preliminary enquiry’, in Wasserstrom, R. (ed.), (1970), War and Morality, Belmont, CA: Wadsworth Publishing Co.
 Dower, N., The Ethics of War and Peace, Cambridge: Polity Press, 2009.
 Noel-Baker, Ph. (1958), The Arms Race, London: Atlanta Books.