The non-intervention principle

According to the non-intervention principle, any exercise of authority by one state within the jurisdiction of another constitutes interference in its internal affairs and is impermissible. This prohibition on the right to intervene proceeds from the conventional view that international society is without any definitive and supreme governing authority.Footnote ² Mirroring the quasi-epistemological state of nature, states are taken to be the principal actors within this environment of anarchy, endowed with the status of political equality and invested with the right to rule within their territories. The state possesses ‘final and absolute political authority’ precisely because ‘no final and absolute authority exists elsewhere.’Footnote ³

Although non-intervention is the corollary of state sovereignty, certain seventeenth-century natural law philosophers argue that because ‘unpunished violations of natural law by one sovereign’ within the state of nature ‘harm every other sovereign by undermining natural law’, every sovereign possesses the right to ‘enforce natural law against any other government that is guilty of violating it’.Footnote ⁴ One prominent philosopher who makes this argument, Hugo Grotius, asserts that under natural law, intervention is permissible if a sovereign's atrocity against its citizens is conspicuous.Footnote ⁵ Likewise, Samuel Pufendorf contends that ‘a War may be made by a Person, not only for himself, but for another’ and ‘where there is no other Reason, the common Relation alone of Men to Men, may be sufficient, when the Party imploring our Aid is unjustly oppressed, to engage our Endeavours, as far as with Convenience we are able, to promote his Defence.’Footnote ⁶

Despite its admirable intentions, granting every sovereign the discretionary prerogative to prosecute and correct the alleged misconduct of other sovereigns through intervention weakens the principle of sovereignty, the security that states enjoy under that principle and the right of each community of citizens to self-determination. A system of international law that affords every state judicial and enforcement rights pertaining to the domestic affairs of a foreign state erodes domestic jurisdiction and authority secured by sovereignty. Through a historical dialogue centred on these considerations, non-intervention arguably developed as a direct response to the right of every state to compel obedience in accordance with its discretionary perception of another state's behaviour.

Realism provides a deeper insight into the logic of non-intervention. It proposes that international society shares a remarkable resemblance to the Hobbesian state of nature, where human selfishness within an environment of anarchy invariably enables egoism to assert itself ad libitum. This compels governments seeking to guarantee the life, liberty and general well-being of their states to act prudentially. Moral guidelines, illustrated through divine command or dispassionate reason, fail to be authoritative because they provide no assurance for the continued existence of states under conditions of anarchy. As Steven Forde describes it, ‘the essence of international realism is its belief in the primacy of self-interest over moral principle […] international conditions compel states to defend their interests by frequently immoral means, and this compulsion of self-defense dissolves moral duties’.Footnote ⁷ In essence, states act under the instruction of self-interest with little or no regard for moral prescriptions.

The concern here is that states might use humanitarian concern as a pretext for interfering in the domestic affairs of another state.Footnote ⁸ This ambiguity ‘pits the virtues of humanitarian rescue against the horror of having expanded opportunities for aggressive war’Footnote ⁹ and questions both the moral and political authenticity of intervention under the categorical belief that international action is initiated through considerations of self-interest. Although the example of Rwanda suggests that widespread atrocity will remain unchallenged without the provisions of exclusive benefits for the rest of the international community,Footnote ¹⁰ Iraq demonstrates how the ideals of humanitarian intervention can easily be misused to justify force emanating from private political interests.Footnote ¹¹ Hans Kochler's assertion that humanitarian intervention ‘has become one of the key terms to legitimize what otherwise would have to be called “act of aggression” or “interference in internal affairs”’ expresses the fundamental scepticism surrounding the permissibility of intervention.Footnote ¹² It is in this need to prevent self-interested intervention that the non-intervention principle finds one of its strongest justifications. Advocates of non-intervention argue that adhering to this principle guarantees the sovereignty of states and preserves every state's right to self-determination.

The case for non-intervention is undoubtedly persuasive, but justifying it without exception remains contentious. Any claim to ‘justification’ necessarily appeals to the concept of ‘justice’ and against the excessive broadness of recent scholarship on its meaning and application; it might be argued that justice derives its meaning from morality.Footnote ¹³ Justice implies an inherent commitment to certain moral principles and to judge in the context of what is ‘just’ or ‘right’ is to make a moral assessment and evaluation. Concluding that non-intervention is justified without exception necessarily commits us to the position that intervention is always unjust. Yet as Cicero writes in De Officiis, ‘of injustice there are two types: men may inflict injury; or else, when it is being inflicted upon others, they may fail to deflect it even though they could.’Footnote ¹⁴ Standing idly by in faithful adherence to the non-intervention principle while governments perpetrate or permit violence hardly justifies the principle on moral grounds. To persist with that notion is to suggest the counterintuitive claim that ‘morality permits immorality’. Asserting that morality justifies its own negation forwards the impression that morality is inherently incoherent and exists with an element of chaos at its core.Footnote ¹⁵

Rousseau, morality and sovereignty

Rousseau's reputation as a realist within the canon of international political thought might leave most readers surprised at his inclusion as an intellectual resource for discussing humanitarian intervention. Most interpreters claim that his writings on International Relations firmly reject the idea of an inherently stable and peaceful international society.Footnote ¹⁶ Substantiating this realist designation is Rousseau's apparent dismissal of the possibility for a confederal solution to the problem of anarchy.Footnote ¹⁷ Stanley Hoffmann offers such a reading when he writes that for Rousseau, ‘the very intercourse of nations breeds conflict […] there can be no assurance that each nation will be able to remain its own master, for the competition itself may always become the overriding tyrant.’Footnote ¹⁸ Similarly, F. H. Hinsley notes that in Rousseau's view, ‘since the international struggle was the automatic consequence of the international system nothing less than a federation of all states would eliminate war; but nothing less than the international system prevented the conclusion of a federation.’Footnote ¹⁹

Rousseau's ‘realism’ begins with humanity's movement out of the original state of nature and its innately peaceful setting into what Hoffmann calls ‘de facto society’; marking humanity's descent into an environment of moral and political anguish similar to Hobbes' state of nature.Footnote ²⁰ To remove themselves from the hostilities of de facto society, individuals resolve to organise and establish civil associations that guarantee security through membership. While this provides a solution to domestic disorder, it reflects not an ‘end to competition’ but ‘merely a displacement of it’ into the international arena.Footnote ²¹ As Ian Clark suggests, ‘while [the state] solves one problem of order at the domestic level, it immediately creates another at the international: the institution of the state creates domestic order but initiates international anarchy.’Footnote ²² Michael Doyle adds that ‘even just General Wills, each reflecting true long-term national interests, can have competitive interests’ that will eventually lead to ‘rivalry’.Footnote ²³

The establishment of distinct political communities projects the identification of the self and other internationally, transforming the state from a device to restrain competition into the very ‘source of international chaos’.Footnote ²⁴ Reinforcing Rousseau's realist status is his apparent pessimism that international disorder can be overcome. Despite endorsing the concept of a federation as the only logical answer to international chaos, Clark insists that Rousseau ‘rejects this as a solution on the grounds that there is absolutely no hope of its realisation’.Footnote ²⁵ That Rousseau denies the prospects of resolution from the conditions of ‘war’ strikes Clark as confirmation of his realist identity and beliefs.Footnote ²⁶

Rousseau's political thought cannot help explain the moral grounds of humanitarian intervention if it is indeed realist. Yet, as Torbjørn Knutsen astutely remarks, ‘although realist themes can be found in Rousseau's thought, they do not make him a realist […] they are the themes of the age, not the themes which made Rousseau unique in the age. He distinguished himself not by his reiteration of these realist themes, but by his transcendence of them.’Footnote ²⁷ His restatement of seemingly realist ideas must be understood as ‘characteristic of a particular epoch of human development’ that should not be confused with universal validity.Footnote ²⁸ When Rousseau speaks of the politics of power, self-interest, mistrust and disorder, he refers specifically to his political environment and the historical processes leading up to its development. It is not a fixed theoretical position that imposes its principles eternally – if history brought this condition into existence, history will surely bring about its ruin. Any proper understanding of Rousseau's commentary on International Relations therefore requires sensitivity to the context and historical location of his work.

Like Knutsen, Michael Williams contends that understanding Rousseau's international political thought demands reading his arguments about international politics within the broader context of his political writings and their historical setting. Rousseau's criticism of the modern state – its historical development and concurrent corruption – must be understood to unravel the complexities of his international thought. As Williams argues, Rousseau's ‘international politics depends upon both the particular historical relationships between states and the nature of these states and citizens themselves.’ These categories cannot be ‘rendered conceptually distinct or historically static; each is crucial to understanding any specific situation and the political judgements to be made in those situations.’Footnote ²⁹

Rather than submit to the realist position of political despair that insists on the inescapable chaos of international society, Rousseau clearly attempts to transcend the grid of power and self-interest by establishing his republican ideals on the concept of moral freedom. This entails both recognising individual rights and reciprocity in the recognition of rights. Williams correctly notes that a social contract ‘founded on this principle’ is merely the expression of this ‘universal principle’ in a ‘particular form’.Footnote ³⁰ Because it is universal, this principle is applicable ‘not only to citizens but to all humanity’.Footnote ³¹ Bearing in mind the relationship between the nature of states and International Relations, Williams maintains that for Rousseau ‘no true sovereign’ founded on the basis of moral freedom can ‘violate the right of another without contradicting the grounds of’ its ‘own right’ and that the ‘rule of the strongest must be superseded by the law of reason and the principles of political right.’Footnote ³² This aspect of Williams' reading does not suggest that he holds Rousseau to be a utopian.Footnote ³³ Rather, it demonstrates that Rousseau's ideas cannot be confined within the narrow space of ‘realpolitik’ without injuring the expansive intricacies of his international political thought.

Admittedly, this brief account of the debate surrounding Rousseau's apparent realism does justice to neither realist nor anti-realist readings of Rousseau. But I do believe that it is accurate to read Rousseau beyond the limits of realism. From both a historical and a theoretical perspective, Rousseau is not a realist, and the importance of morality in his political thought clearly indicates the accuracy of this conclusion. As we shall see, Rousseau's discussion of morality and sovereignty clarifies how an argument for humanitarian intervention can be made on moral grounds. His approach clarifies the relationship between reason and conscience, enhancing our understanding of a philosophical basis for humanitarian intervention.

According to Rousseau, any historical or cross-cultural study of morality necessarily reveals common moral principles that both express and determine our judgements of justice and injustice. But unlike Kant and most thinkers associated with the tradition of natural law, Rousseau does not appeal exclusively to ‘reason’ for the deliverance of moral precepts. Rather, he affirms that the principles of morality are discernible not only through an introspective consultation with reason but also with the natural and ‘innate principle of justice and moral truth’ to which he gives the label ‘conscience’.

Rousseau argues that ‘by reason alone, independent of conscience, no natural law can be established’ and that ‘the entire right of nature is only a chimera if it is not founded on a natural need in the human heart.’Footnote ³⁴ Responding to natural law philosophers, he asserts that ‘even the precept of doing unto others as we would have them do unto us has no true foundation other than conscience and sentiment; for where is the precise reason for me, being myself, to act as if I were another, especially when I am morally certain of never finding myself in the same situation?’Footnote ³⁵ Reason cannot autonomously inform us of the principles of morality; neither can it independently guarantee reciprocity and compliance to this maxim. ‘If natural law were written only in human reason it would hardly be capable of directing the majority of our actions, but it is also engraved in man's heart in indelible characters and it is there that it speaks more strongly than do all the precepts of Philosophers.’Footnote ³⁶

Yet, this does not suggest that reason is not required for moral knowledge. On the contrary, Rousseau believes that reason is partially necessary to discern moral principles. Alone, conscience is inadequate for persons to either discern or follow moral principles. Although the cognisance of moral principles rests on an honest deliberation with the principle of conscience initiated by looking into the depths of our hearts, reason must be consulted to discern and enlighten these ‘true affections of the soul.’Footnote ³⁷ In return, conscience ensures that the faculty of reason is never ‘without principle’.

Appealing to both reason and sentiment, Rousseau believes that the principles of morality are common universally. Regardless of distinct cultural aphorisms, moral judgements are adjudicated according to the ‘involuntary rule’ of conscience that exists within us prior to socially informed prejudices. As Rousseau writes, ‘a rule prior to opinion exists for the whole human species. It is to the inflexible direction of this rule [conscience] that all the others ought to be related. This rule judges prejudice itself, and only insofar as the esteem of men accords with it ought this esteem to be authoritative for us.’Footnote ³⁸ This rule is, then, not determined by the customs and mores of particular societies but is instead an objective principle authoritative for all members of the human species. Rousseau's poetic disquisition on this theme is worth quoting in full.

It must be emphasised that the element of commonality within Rousseau's moral discourse is not grounded on universal agreement. Common moralityFootnote ⁴⁰ does not draw its content from the fortuity that certain moral principles are universally endorsed. Neither is it derived from the identification and collation of similar mores or customs from the breadth of societies. Rather, its universality stems from the fact that its principles apply to all members of humanity.Footnote ⁴¹ As Rousseau puts it, only if we regard common morality as universal in this way are we able to ‘judge our actions and those of others as good or bad in spite of our own maxims.’Footnote ⁴²

The corollary of this universal applicability is that the principles of common morality are authoritative for all moral agents regardless of their endorsement. For Rousseau, it is erroneous to conclude that the principles of morality cannot be applied universally simply because there exists ‘in a corner of the world one custom opposed to the concepts of justice’, ‘as if all the inclinations of nature were annihilated by the depravity of some individuals and as soon as there are some monsters the human species is no longer anything.’Footnote ⁴³ In spite of exceptions, Rousseau certainly considers common morality to be ‘this evident and universal agreement of all nations’ that demonstrates the ‘striking uniformity of men's Judgment.’Footnote ⁴⁴ Nevertheless, common morality is not common because it is widely endorsed; it is widely endorsed because its principles are authoritative for all.

By bringing Rousseau's appeal to sentiment within the discourse of humanitarian intervention, the basis of the claim that intervention is justified when acts ‘shock the moral conscience’ of humanity is made more intelligible.Footnote ⁴⁵ Because the ‘love of the good and hatred of the bad are as natural to us as our own existence’, discernible simply by listening to our natural inclinations, even ‘the most corrupt souls cannot lose this first inclination completely.’Footnote ⁴⁶ Although morality compels us to aid innocent victims of violence, its guidance does not result solely from a dispassionate consultation with reason. It also results from the provisions of moralsentiment. As Rousseau writes,

The turn to moral conscience as the partial basis for intervention can thus be seen to rest on a coherent and objective moral principle that is not arbitrary despite its appeal to sentiment. It might be said, however, that the moral conscience of humanity is not a singular entity and that not all of humanity feels shocked by the same situations. It is simply tenuous to ground intervention on conscience. But Rousseau's account of the cognisance of moral principles does not rest solely on conscience. It relies on both reason and conscience. More importantly, intervention is not grounded on conscience but on moral principles discerned through consulting both reason and conscience.

The claim that not all of humanity feels shocked by the same situations might also be restated as the claim that conscience does not speak universally. Rousseau does, in fact, raise this issue when he asks ‘if it [conscience] speaks to all hearts […] then why are there so few of them who hear it?’ His answer is that it ‘speaks to us in nature's language which everything has made us forget.’Footnote ⁴⁸ For Rousseau, individuals in modern societies are corrupt. Ruled by inflamed amour-propre, they direct their choices and preferences in accordance with the prejudices of public opinion in order to be esteemed. This desire for esteem leads to competition and vice and turns individuals into adversaries, a relationship concealed only by their affected manners. Civil engagement results, then, in a loss of authenticity. And ‘as a result of being dismissed, it [conscience] finally gives up, it no longer speaks to us, it no longer responds to us, and after such long contempt for it, to recall it costs as much as banishing it did.’Footnote ⁴⁹

Two points are worth noting here. First, common morality is grounded on universal applicability, not agreement. Our prejudices might sometimes corrupt our recognition of common moral principles, but this does not mean that these principles are not common, authoritative and binding on all. Neither does it mean that universal standards of justice do not exist. On this view, it may be said that moral conscience is a singular entity made to appear otherwise by our prejudices. Second, fundamental conceptions of moral principles do converge across diverse religious, cultural, social and political institutions.Footnote ⁵⁰ Moral disagreement arises not from diverse understandings of moral principles but from different interpretations of common principles as they are applied to particular situations. The mistake lies in attributing these disagreements to contrary definitions of the principles of morality.

How, in view of these sentiments, are we to understand a Rousseauian challenge to the principle of non-intervention? This ‘more powerful duty’ that Rousseau speaks of, based on laws that ‘deprive us of the right to protect innocence’, when read back into the discourse of humanitarian intervention, alludes unmistakably to that principle, which, as provision within international law, denies and restricts the authority of the morality that inspires in us an inclination to protect. As Rousseau argues, morality is the basis for justifying intervention. It compels us to ‘take up the defense of the oppressed person’ and to do so reflects justice or ‘right’. Because it is common in the sense that its rules are applicable universally, any state or government engaging in morally repulsive acts remains culpable, making intervention aimed at deflecting these acts permissible. Clearly stated, morality makes intervention on humanitarian grounds right. In this respect, the moral vacancy of the non-intervention principle is exposed.

Although it is clear that the non-intervention principle turns on the concept of sovereignty, many who rely on that concept are unclear about the identity and character of the sovereign. According to Rousseau, the sovereign is a moral being ‘to whom the social pact gave existence, and all of whose wills bear the name of laws.’Footnote ⁵¹ Sovereignty, then, lies with the people – the body politic composed of citizens joined in a social compact. The government is merely the executive branch of the state, tasked with the interpretation and execution of the laws defined by a ‘general will’ legislated through the constitutional decision-making powers of the sovereign.Footnote ⁵² To declare war on a sovereign is to ‘attack the public convention [social compact] and everything that results from it; for the essence of the state consists only in that’.Footnote ⁵³ For Rousseau, the social compact is ‘under an engagement to provide for the preservation of the humblest of its members with as much care as for all the others’ and a citizen's well-being is never ‘less the common cause than that of the whole state’.Footnote ⁵⁴ It is impossible that the sovereign ‘would allow any member of the state, whoever he might be, to injure or destroy another’.Footnote ⁵⁵ When sovereignty is properly understood, it becomes clear that humanitarian intervention neither infringes on it nor denies a community its right to self-determination. Rather, it protects sovereignty by coming to the aid of the very essence of the state that is invested with the character and identity of the sovereign. It compensates for the definitional impropriety of associating sovereignty with its representative. And by violating the principle of non-intervention, it paradoxically achieves the aim of that principle: the protection of state sovereignty. In short, if the principle of non-intervention is justified on the grounds that it protects sovereignty – the sovereignty of the people – then in times of humanitarian crises this ground must justify its negation. The principle must turn on itself if it is to hold true to its purpose.Footnote ⁵⁶

Kant, justice and intervention

Moving from Rousseau to Kant expands the philosophical basis of the right to intervene in instances of widespread atrocity. Although the argument remains within the domain of morality, knowledge of its principles is now discernible solely through rational inquiry. By relying exclusively on reason, Kant departs from the general discomfort of having moral principles partially emanate from what some would deem a vague and philosophically inadequate notion of conscience. As Alan Donagan describes it, morality as Kant understands it consists of moral principles that are binding on all rational agents ‘by virtue of their rationality’.Footnote ⁵⁷ These principles are to be acknowledged by all rational agents as universally lawful. Like Rousseauian morality, Kantian morality is common not because it is universally endorsed, but rather, its universality rests on the notion that its laws apply to all rational creatures. Kant's moral project is often extolled as the first to develop, ground and authenticate a systematic theory of morality through ‘autonomous reason.’Footnote ⁵⁸

There are of course, difficulties in using Kant's work to support intervention on humanitarian grounds. They begin with Kant's apparent commitment to the principle of non-intervention. In the Fifth Preliminary Article of Perpetual Peace, Kant declares that ‘no state shall forcibly interfere in the constitution and government of another state’.Footnote ⁵⁹ Regardless of ideological differences, states are to have complete authority over their domestic affairs. A ‘sense of scandal or offence which a state arouses in the subjects of another’ should serve only ‘as a warning to others, as an example of the great evils which a people has incurred by its lawlessness’.Footnote ⁶⁰ As Burleigh Wilkins writes, ‘Kant believed the duty of non-intervention in the affairs of a state by other states is a perfect, i.e., exceptionless duty […] No state has the right to intervene in the affairs of other states, and, although Kant does not explicitly say so, states should presumably be left to deal with human rights problems, as with all their problems, through self-reform.’Footnote ⁶¹ Similarly, Georg Cavallar maintains that ‘the obligation of non-intervention is unconditional. Allowing some states (the good ones) to intervene in others (the bad ones) violates the universal concept of justice. As a consequence, Kant did not allow any exceptions to the principle established in the Fifth Preliminary Article.’Footnote ⁶² It must be noted, however, that Kant's Fifth Preliminary Article is preliminary and therefore provisional. Contra Wilkins and Cavallar, this is not a moral point but one dictated by the necessities of bringing states gradually out of the condition of nature into a lawful condition.

Despite arguing that Kant firmly prohibits intervention, both Wilkins and Cavallar leave room for interpreters to justify intervention on Kantian principles. If we take removing individuals from the state of nature as a minimum benchmark for states to possess moral personality, states that fail to do so cannot claim immunity from intervention. This does not, however, grant liberal states the right to intervene in the affairs of non-liberal states for the purposes of political reform. Rather, only states that can do no better than the state of nature remain susceptible to external interference. As Wilkins suggests, ‘one possible reading of Kant would be that intervention could be justified only in cases where a state creates a life for its citizens no better than they would find in a state of nature. Human rights violations would have to be egregious before intervention would be acceptable to Kant, and then if intervention were the only available way to protect these rights.’Footnote ⁶³ He notes that ‘in cases where segments of a population of a state are imprisoned or exterminated on the basis of race, class, or ideology, it is hard to see how such a state could be said to have a moral personality at all.’Footnote ⁶⁴

Likewise, Cavallar remarks that ‘in grave cases such as massacres, genocide, or acts ‘that shock the moral conscience of mankind’, humanitarian intervention is permissible or justified’ precisely because ‘the totalitarian regime relapses into the state of nature and thus loses its moral personality. Other states no longer have duties towards this regime, only towards its people.’Footnote ⁶⁵ Here, Cavallar assumes incorrectly that a state that engages in massacre must be totalitarian and his claim that the ‘regime’ literally relapses into the state of nature is also questionable. Nevertheless, he is quick to emphasise, and I believe Wilkins will agree, that such an argument for the permissibility of intervention ‘certainly goes beyond Kant's texts, though it could be argued that it follows and applies Kantian principles.’Footnote ⁶⁶

In making a case for intervention, Cavallar cautions against the elementary classification of all non-liberal states as candidates for liberal intervention. The absence or loss of moral personality must be narrowly conceived and cannot be imposed on non-liberal states by virtue of their non-liberal constitution. Criticising Doyle for implying that liberal states possess the right to intervene in non-liberal states, Cavallar also faults him for claiming that such a position is ‘supported by Kant’.Footnote ⁶⁷ This criticism is mistaken, however, for Doyle maintains that according to Kant an important liberal principle is ‘extending nonintervention by force in [the] internal affairs of other states to nonliberal governments’.Footnote ⁶⁸ Similarly, Fernando Tesón has been challenged for his alleged Kantian argument that ‘sovereignty is conditional’ – only states that respect human rights reserve the right of non-intervention.Footnote ⁶⁹ Tesón's argument for the conditionality of sovereignty might sound appealing, but it effectively justifies liberal intervention in non-liberal states on the basis of their domestic political arrangement.Footnote ⁷⁰ Grounding this excessive allowance for intervention on Kantian principles appears to be questionable at best.

Resolving this exegetical debate on the appropriate application of Kantian ideas to support intervention requires a more extended argument than can be presented here. I believe, however, that Kant's theory of right can easily be made to explain the moral foundations of humanitarian intervention.Footnote ⁷¹ To understand how a Kantian argument for the moral grounds of humanitarian intervention might proceed, it is first necessary to understand Kant's idea of ‘Recht’ or ‘right’.Footnote ⁷²

‘Right’, according to Kant, applies ‘to the relationships between one person and another’ and ‘is concerned only in the form of the relationship between the two wills.’Footnote ⁷³ It is not concerned with ‘the relationship between the will of a person and the desires of another’, nor is it concerned with the possible attainment of the good to be derived from any relationship.Footnote ⁷⁴ No discussion of ‘interest’, ‘desire’, or any other non-moral conception of ‘good’ can be relevant to the concept of ‘just’ or ‘right’. In short, these concepts are fundamentally distinct from the notion of ‘right’ and to argue that an ‘interest’ can be ‘just’ is to commit a blatant conceptual error. For Kant, ‘right’ is simply the ‘conditions within which the will of one person can be reconciled with the will of another in accordance with a universal law of freedom.’Footnote ⁷⁵

In working out the implications of this concept of right, Kant arrives at the seemingly counterintuitive proposition that right ‘entails the authority to use coercion.’Footnote ⁷⁶ Clearly needing to explain this proposition, he begins by asserting that any contrariety to the ‘hindrance of an effect’ necessarily assists its advancement and must therefore be ‘consonant with it’.Footnote ⁷⁷ According to universal law, the antithesis of right imposes itself as a ‘hindrance to freedom’ while coercion is suitably understood as being incompatible with the requisites and pursuit of freedom. Interestingly, if coercion is applied to acts that serve as a hindrance to freedom, then according to the understanding that the contrariety to the hindrance of an effect actually promotes that effect, coercion will in fact serve to promote freedom. Put simply, coercion applied to the hindrance of freedom ‘will be a hindrance to the hindrance of freedom, and will thus be consonant with freedom in accordance with universal laws – that is, it will be right.’Footnote ⁷⁸ This clever piece of philosophical reasoning has profound implications for the debate surrounding humanitarian intervention. If governments turn against their own citizens and persist in genocide, mass murder or torture, their acts constitute a hindrance to the freedom of the community of citizens. Forcefully intervening to thwart or remove this hindrance to freedom experienced by the local community vindicates freedom by coercively impeding its hindrance and must therefore be considered as concordant with the principles of right and justice.

Holding on to the position that states are motivated only by self-interest, the cynic or realist might respond that this justification for humanitarian intervention is problematic at its core. Stripping away the pretensions of altruism, humanitarian intervention can be seen to be essentially motivated by the morally vacant or immoral instruction of personal gain.Footnote ⁷⁹ To justify intervention on moral grounds when it is in fact motivated by immoral or amoral concerns forwards the impression of inconsistency and incoherence. This harsh assessment neglects, however, the distinction between Kant's doctrine of right and his doctrine of virtue. In the Metaphysics of Morals, Kant dismisses attaching the ethical motivations of the willing agent with the just quality of an action. For Kant, ‘the only object of right in general is the external aspect of actions, right in its strict sense, i.e. right unmixed with any ethical considerations, requires no determinants of the will apart from purely external ones.’Footnote ⁸⁰ The just quality of an action does not have to coincide with the moral character of the agent performing that action for the action to be recognised as just. Neither does the just quality of an action depend on the motivation behind its performance. It is the external performance of that action, not the internal motivation to perform that action that determines whether an action is ‘right’.

This emphasis on action over motive does not mean that Kant dispenses with the need to consider motivation. Considering the ethical value of an action, however, requires us to move from action to agent, that is, from performance to motivation. The motivation to perform an action falls within the domain of virtue and does nothing to diminish or enhance the just quality of an action. According to Kant, ‘if it is not our intention to teach virtue, but only to state what is right, we may not and should not ourselves represent this law of right as a possible motive for actions.’Footnote ⁸¹ It follows that humanitarian interventions can be justified even if they are motivated by personal gain. On this account, it is possible to perform an intervention that is humanitarian and just, yet morally inadequate. It is not immoral because the act itself is just; it is morally inadequate only because it is not motivated by virtue. This seemingly counterintuitive conclusion occurs only if we continue to insist on collapsing and understanding the motivation and action of humanitarian intervention as entirely inseparable when passing moral judgement. Justification for humanitarian interventions should be understood as justifications for action, not motivation.

The duty to protect

Having established that from the standpoint of a Kantian theory of justice there is in principle a right to intervene, the mode of inquiry must now shift to considerations of obligation, duty or responsibility.Footnote ⁸² A much-cited report by the International Commission on Intervention and State Sovereignty argues that the ‘responsibility to protect’ consists in ‘the idea that sovereign states have a responsibility to protect their own citizens from avoidable catastrophe – from mass murder and rape, from starvation – but that when they are unwilling or unable to do so, the responsibility must be borne by the broader community of states.’Footnote ⁸³ Yet, as Kok-Chor Tan correctly observes, the Commission fails to demonstrate why the international community must bear the duty to protect. The findings of the Commission simply assume that the international community has a responsibility to protect, taking the permissibility of humanitarian interventions as the only obstacle to be overcome.Footnote ⁸⁴

Any demonstration of a duty to protect must begin with a conceptual treatment of ‘duty’. According to Kant, duty ‘is that action to which someone is bound. It is therefore the matter of obligation’.Footnote ⁸⁵ Its universal imperative proceeds from acting ‘as if the maxim of your action were to become by your will a universal law of nature.’Footnote ⁸⁶ Demanding this axiomatic transformation of individual action into a universally authoritative law, Kant tests the applicability of this law in a hypothetical situation fundamentally similar to the deleterious conditions of a humanitarian crisis. He inquires if it is indeed possible to stand idly by while fellow men suffer ‘great hardships’ despite being in a position to help, and if willing this as a law of nature will impede the subsistence of the human race.Footnote ⁸⁷ Despite concluding that the ‘human race could very well subsist’ under this maxim of parochial apathy imposed as universal law, Kant challenges the possibility of willing this principle as a law of nature.Footnote ⁸⁸ He insists that a will that reaches such a resolution will necessarily engage in ‘conflict with itself’, for many occasions will come to pass that it will itself require aid.Footnote ⁸⁹ Yet because it has willed apathy as law, any hope for assistance will be all but extinguished in these times of exigency. For Kant, we have a duty to assist others in their time of need; we simply cannot will – as universal law – to do otherwise.

Before proceeding further, it is necessary to confront the issue of agency. As it is often suggested, ‘we’ have a right and obligation to intervene when governments permit or perpetrate atrocity domestically. This ‘we’ presumably appeals to all moral persons, although states perform the actual intervention. This raises the question – why do states have the right and duty to intervene when massive human rights violations are taking place?

There are two possible ways to answer this. First, it could be argued that the authority of common morality applies to states as well as individuals. This requires that we consider both as agents endowed with moral personality. Although the argument for individuals hardly requires explanation, categorising the state as a moral agent requires clarification. The assertion that ‘universal moral principles cannot be applied to the actions of states’ reflects a widely held position among theorists of International Relations.Footnote ⁹⁰ This realist position appears to rest on two puzzling premises. The first is that although individuals are moral agents, states – institutions formed and governed by moral agents – are not. It appears that by virtue of their association, moral agents are able to cast off the authority of morality when interacting with other similar associations. Because morality is dismissed on the grounds of association, an association of two could claim exemption from morality. Realists will be quick to respond that it is not solely the notion of association that diminishes or discards moral authority; rather, it is the absence of any formal governing authority coupled with the distinct interests of each association that renders recourse to moral principles implausible. The suggestion here is that morality is suspended in the absence of government and depends on a coercive institutional apparatus for its authority. But this is an error, for morality cannot be suspended and does not depend on any government for authority. Within international society, states are artificial persons and hence moral agents because they are constituted and governed by rational and moral individuals. If natural persons are moral agents subject to the authority of moral principles, states – as associations of these moral agents – must necessarily find morality authoritative as well. The moral duty to protect applies equally to individuals and states.

This view suggests that moral agents within a moral community cannot stand idly by while governments permit or perpetrate widespread atrocity and that all moral agents have, in principle, a duty to protect. But it does not follow that each person or state is individually responsible for rescuing the victims of violence in a humanitarian crisis. Relying on the metaphor of a drowning swimmer, Tan argues that not ‘everyone ought to rush into the ocean to attempt the rescue; indeed, a nonswimmer certainly does not have the obligation to attempt the actual rescue.’Footnote ⁹¹ Rather, all ‘parties are to contribute to the rescue in ways commensurate with their ability and the needs of the rescue operation.’Footnote ⁹² Likewise, having a duty to thwart genocide does not require all to physically intervene in the target state. The actual process leading up to the intervention and the intervention itself might impose different demands on different people. The claim that moral agents have a duty to protect does not mean that all have similar roles to play in the intervention process. Having a duty to act simply means that we can be reasonably required to participate in the action, or to support it, if necessary.

On this view, every agent is obligated to protect the innocent, although this duty is expressed in different ways. Individual persons, acting on their own, are entirely incapable of performing an intervention. Their duty entails supporting legitimate instances of humanitarian intervention, either through pressing their governments to take action or to demand that action be taken by other states or by international organisations. Support could also take the form of advocating or accepting their government's use of tax dollars to finance a humanitarian intervention. At an absolute minimum, they would have an obligation not to oppose efforts to rescue the victims of violence, assuming rescue was feasible. But even if we understand the duty to intervene as falling on states rather than on their citizens individually, this does not imply that any particular state must carry out the intervention. It does imply, however, that every state has in principle a duty to participate in, support, or at least not obstruct interventions aimed at ending widespread violence. States with the capability to intervene militarily may have a duty to make use of that capability or make it available to international agencies.

The second answer to the question of why states have a duty to intervene differs slightly from the first. This argument, which turns on Kant's principle of respect, is compatible with the first but unlike the first it does not require that the state be considered a moral agent. In essence, it mandates that all moral persons carry a duty to oppose violence committed on innocent persons. In cases where governments perpetrate or permit widespread violence against innocent persons, it becomes impossible to discharge this duty to protect individually. It is only possible to discharge this duty collectively. As Carla Bagnoli writes, ‘not all individual duties are to be discharged individually […] Delegation to the appropriate institution […] can be the only appropriate way to satisfy some individual duties. In other words, some duties we bear as individuals may and sometimes ought to be discharged collectively.’Footnote ⁹³ In this way, although the moral obligation to protect falls on individuals, individuals are only able to discharge this duty collectively. The state – as a collective association of individuals equipped with the necessary capabilities needed to perform an intervention – is the appropriate institution through which individuals are able to discharge their moral obligation to protect.

According to this view, the duty to protect essentially lies with individual persons. In cases where widespread atrocity requires a humanitarian intervention, however, the moral obligation that individuals have to intervene can only be discharged collectively through the institution of the state. Although the moral obligation of humanitarian intervention rests with individuals, the performance of this duty can only be undertaken collectively through states.Footnote ⁹⁴

For most commentators who think we have a duty to protect, the problem lies in their understanding that this duty is an ‘imperfect duty’, that is, it is a duty of beneficence that we cannot be compelled to perform. It is meritorious to perform these duties but ‘failure to perform them is not in itself culpability.’Footnote ⁹⁵ A duty is also considered imperfect when no performer is specified. This has led Michael Walzer to claim that ‘the general problem is that intervention, even when it is justified […] is an imperfect duty – a duty that doesn't belong to any particular agent. Someone ought to intervene, but no specific state in the society of states is morally bound to do so.’Footnote ⁹⁶

Bagnoli disagrees with Walzer. She argues emphatically that humanitarian intervention can be classified as a perfect duty – a duty we can be required to perform. Beneficence might admittedly be the moral basis for assistance, but it also bears a deep relationship to the ‘principle of respect’ for it encompasses advancing the fundamental conditions necessary for ‘human rights’ and ‘self-respect’ to exist and flourish.Footnote ⁹⁷ Forcefully challenging extreme human rights violations, although beneficent, stems from the moral obligation to respect the rights of others which is itself a perfect duty that all moral agents are bound to perform.Footnote ⁹⁸ Kant himself acknowledges that the principle of respect remains a perfect duty we owe to others.Footnote ⁹⁹ By shifting the focus of humanitarian interventions from beneficence to respect for humanity and arguing that this respect is ‘owed to others and can therefore be “legitimately demanded by them”’, Bagnoli relocates humanitarian intervention from the dutiful uncertainty of an imperfect duty to the obligatory certitude of a perfect duty.Footnote ¹⁰⁰

It must be emphasised and made clear that the duty to intervene is grounded on Kantian right rather than Kantian ethics and it is thus a perfect duty that we can be reasonably required to perform. According to Kant, duties are to be understood either as ‘duties of right (official iuris), that is, duties for which external lawgiving is possible, or duties of virtue (official virtutis s. ethica), for which external lawgiving is not possible.’Footnote ¹⁰¹ Put differently, duties of right are duties that individuals may rightly be compelled to perform while duties of virtue are duties that do not authorise coercion. By virtue of their humanity, all individuals possess an innate right of freedom – the independence to pursue self-determined ends without ‘being constrained by another's choice’ insofar as this freedom is able to coexist ‘with the freedom of every other in accordance with a universal law’.Footnote ¹⁰² Regarding an individual as a person demands that the individual not be ‘valued merely as a means to the ends of others or even to his own ends, but as an end in itself.’ The individual ‘possesses a dignity (an absolute inner worth) by which he exacts respect for himself from all other rational beings in the world.’ The individual's innate right of humanity is ‘the object of respect which he can demand from every other human being’.Footnote ¹⁰³ In other words, human beings are owed respect and must be treated as non-instrumental ends. Kant restates this thesis in a slightly different manner when he writes that

For Kant, a ‘failure in the duty of respect infringes upon one's lawful claim’.Footnote ¹⁰⁵ Understanding respect to be a lawful claim and a duty owed to others implies that it is a duty of right for which external lawgiving is possible. Duties of right, it must be recalled, relate to justice. Respecting others as persons is thus a question of justice, not beneficence. If respect for persons is a matter of justice, ending widespread atrocity on the grounds of respect must be understood as a duty of right, that is, a perfect duty. Humanitarian interventions that aim to thwart hindrances to freedom, although beneficent, should not be seen as acts of beneficence but as attempts to uphold justice by submitting to its demands.

When the duty of humanitarian intervention is understood in this way, it is clear that Walzer's statement misses the point. If intervention stems from the obligation to respect the rights of humanity, the reason why ‘no specific state in the society of states is morally bound’ to intervene must be because every state ‘in the society of states is morally bound to do so’. This is indubitably grounded on the fact that we are all members of the international moral community. The issue at stake here is not therefore the absence of proper authority, for this consists in the moral community of humanity; the real issue involves the absence of a representative for this proper authority to coordinate efforts to redeem humanity from the injustices of certain morally appalling acts. Reducing intervention to an imperfect duty on the grounds that proper authority cannot be established fails to distinguish between proper authority and the absence of its executive representative. Humanitarian intervention should accordingly be recognised as a perfect duty that we could be required, even coerced, to perform.

To reinforce this perfect Kantian-inspired duty to protect, I suggest turning to Hegel. Although Hegel's objections to Kant's moral theory are often mentioned, their views of morality are not as divergent as many think. To see how Hegel complements the Kantian perfect duty to protect, although with slight variations to the ongoing debate, it is first necessary to explain his objections.

The Hegelian commandment of right is fundamentally similar to the Kantian principle of respect. According to Hegel, ‘personality contains in general the capacity for right and constitutes the concept and the (itself abstract) basis of abstract and hence formal right. The commandment of right is therefore: be a person and respect others as persons.’Footnote ¹⁰⁶ Despite this underlying convergence, Hegel objects to Kant's moral theory on the grounds that Kant's categorical imperative persists as an abstraction that provides a moral law with form but not content. It has form in the sense that it is a universal law that is demonstrably right, but it lacks content because it fails to provide us with a substantive and informative guide regarding our duties. For Hegel, the categorical imperative ‘would be all very well if we already had determinate principles concerning how to act.’Footnote ¹⁰⁷ This explains why he thinks that the content of moral laws must come from one's actual social environment. Only then can morality escape the arraignment of empty formalism caused by this absence of content. Although Kant scholars have defended Kant against Hegel's objections, for the purposes of this article I will accept Hegel's challenge to Kant.

What is most instructive in Hegel is his declaration that ‘a human being has rights insofar as he has duties, and duties insofar as he has rights.’Footnote ¹⁰⁸ If we take this to mean that individuals have rights only insofar as they have duties and duties only insofar as they have rights, then the veracity of this Hegelian-inspired principle within the discourse of humanitarian intervention becomes clear. According to common morality, individuals have certain rights as rational agents that must be respected by all members of the human community. These rights exist in the abstract and can only be transformed into ‘real’ non-abstract rights through the universal imposition of a corresponding duty to respect them. The absence of corresponding duties to respect these rights necessarily permits violent persecutions that could continue or persist without the possibility of remedy.

Under these circumstances, rights are duly exposed to be merely formal and empty for they do not serve or possess any corporeal significance and exist purely as abstractions. Non-abstract rights exist only on the condition that corresponding duties exist, and vice versa. The only way to guarantee the actuality of their existence is to acknowledge that rational agents have corresponding obligatory duties that come along with their rights. These corresponding duties must be perfect if we are not to fall into the same circularity that originally reduced rights to an abstract formality. To put it differently, if our corresponding duties to protect our rights are imperfect, that is, if we cannot be coerced into performing them, any violation could once again continue to persist without remedy or relief. Rights will again be reduced to a mere formality that exists without real force. The imperfection of duty prevents an authoritative transformation of the Kantian moral laws from abstraction to a socially potent and meaningful actuality. We can only guarantee that our rights exist as part of our moral being if the corresponding duties to respect these rights are indeed perfect and applicable universally. The content of these duties, which give the moral law its content, will be drawn and developed from the international social environment.

It must be emphasised that the existence of rights rests on the existence of corresponding duties to respect these rights. These duties cannot be imperfect if our rights as moral agents are to be authoritative. Individuals simply cannot stand idly by while governments allow or commit widespread atrocity if they are to prevent their rights from slipping into an empty formality. Similar to the provisions of natural law, because unpunished violations of rights by some moral agents harm every other moral agent by dissolving the moral law and the actuality of rights, every moral agent must therefore adhere to the perfect duty of intervention to thwart this violation of rights if so called upon.

As I argued above, however, not all individual duties are to be discharged individually. Although the existence of rights depends on corresponding duties to respect these rights, some duties can only be discharged collectively. In cases where humanitarian intervention is required to respect these rights, individual duties to respect them can only be discharged collectively through the institution of the state. States are therefore obligated to perform an intervention when other governments commit or permit the execution of widespread violence within their territories. This conceptual relationship between rights and duties implies that humanitarian intervention is a perfect duty whose performance can be demanded by moral agents under violent persecution. As Rousseau describes it, ‘it is an error to consider only one's claims upon others, and to forget those of others upon us.’Footnote ¹⁰⁹