“ . . . a people in arms for liberty is the only [case] of any nicety, or which, theoretically at least, is likely to present conflicting moral considerations.”
—John Stuart Mill“Rebellion is not always evil.”Footnote 1 So the English republican Algernon Sidney once declared to his great personal cost. When he wrote this in the 1680s, it was a deeply provocative statement and, in fact, the manuscript in which it appeared was called as a “witness” for the prosecution when he was tried and executed for treason in 1683.Footnote 2 Today, for many who would describe themselves as democrats of one kind or another, some such idea is more or less axiomatic. Often it reflects a commitment to what we may call the “Rebel Principle,” which holds, roughly, that if rebellion against one's government is necessary as a means of resisting its oppressive policies, then it might be morally justifiable. Likewise, it might be morally justified as part of a more ambitious attempt at revolution in circumstances of profound and systemic political injustice.Footnote 3
In spite of the prevalence of this belief, however, we currently lack an adequate understanding of its normative implications. Moreover, recent attempts by the United States and other democracies to respond to oppressive political regimes in the Middle East demonstrate that this lacuna is a problem not just for political theory but also for practical politics. Three prominent recent cases reflect its importance.
Following the failure of the United States to induce internal rebellion in Iraq via covert efforts of the CIA during the 1990s, the George W. Bush administration toppled Saddam Hussein by means of military invasion in 2003.Footnote 4 Whether “ending tyranny,” as Fernando Tesón has described it, was intended as an end in itself or purely as a means of addressing purported security threats to the United States is a matter of controversy. Regardless of intent, the invasion brought about a revolutionary outcome without awaiting the trigger of a domestic rebellion. As this article makes the case, the catastrophic results may be attributed in part to the failure to take into account the need for popular political agency within Iraq when calculating the odds of any replacement regime succeeding.Footnote 5 In the second case of Libya, Western powers exploited the ambiguities of a UN mandate concerned only with the prevention of war crimes as cover to assist rebels in bringing down the Qaddafi regime, a move that has so far only sown further civil conflict and created a power vacuum, leaving the country vulnerable to the ambitions of ISIS. And yet in the third case, the contrasting, more conservative approach taken by the United States and its allies toward Syria has only seen things go from bad to worse as moderate rebels have sought alternative sources of military support from increasingly radical Salafi-Islamist groups. The humanitarian crisis reached a point such that even Michael Walzer, who had warned against intervention in all three states, publicly considered whether theorists might need to revisit the principle of nonintervention and debate anew the ethics of external military assistance to rebels.Footnote 6
These three contrasting cases indicate that not only do we lack a clear idea of how to pursue liberal-democratic commitments against foreign tyrannies but we also lack a pragmatic default position in its absence.Footnote 7 In this article I argue that our ability to judge such cases effectively has been impeded by a flaw in the normative architecture that arises from its failure to take account of what I call the Rebel Perspective. This refers to the theoretical point of view from which we can envisage the sort of demands those truly suffering from domestic tyranny might reasonably make, and from which we can specify appropriate ethical principles to guide rebels in a struggle to secure them. By specifying what those suffering oppression could justifiably do to resist, we also help to define what they might justifiably claim to need by way of assistance. The Rebel Perspective, therefore, has a crucial role to play in determining how outside parties ought to describe, evaluate, and interact with rebel groups.
In this article, I will use the terms “rebel” and “rebellion” to refer to (those contributing to) an attempt, at a minimum, to challenge and overturn the existing government in a state and, at a maximum, to bring about a revolution that challenges the political and social order of the state as such. This latter sense includes not only, say, cases of national liberation movements struggling against foreign or colonial rule and seeking to change the territorial range of the state but also attempts at radical change to the social and political order within a state. My concern, however, will chiefly be with armed rebellion.
Elsewhere, I have set out what I think the rules and principles governing rebellion should be.Footnote 8 My aim here is to show how the failure to take account of the Rebel Perspective not only constitutes a significant problem in its own right but also constitutes a source of potential instability within the wider global normative architecture governing the use of force: in other words, the cluster of doctrines, laws, and vocabulary that is generally used to regulate and evaluate it. There is therefore an urgent need to think about how insights from emerging literature by political theorists on revolution and rebellion may be used to correct this flaw.Footnote 9
Three Pillars
While human rights define fundamental values and specify claims that may be made against societies, my focus is on three further pillars of the global normative architecture that define the terms by which actions and judgments ought to be guided when these rights are under threat of violence. All three pillars have a particular bearing on rebellion:
-
(1) The concept of “terrorism” and the commitment to coordinated efforts to discourage, degrade, and defeat it;
-
(2) The International Law of Armed Conflict and International Humanitarian Law (collectively referred to in this article as “Law of War”) as a framework through which to interpret, evaluate, and guide the behavior of actors in armed conflict and through which to guide reactions to it by the international community; and
-
(3) The Responsibility to Protect (RtoP) as a doctrine to guide the reactions of UN members to human rights violations by states against their own subjects, and the complementary principle of nonintervention defined by the immunity of states from interference as guaranteed under the UN Charter when RtoP has not been triggered.Footnote 10
Directly or indirectly, all three pillars reflect specific understandings of the possibility of legitimate (armed) resistance and how to respond to it. They suggest ways of interpreting and naming instances of nonstate violence, criteria by which to appraise them morally, and policies or actions that might be undertaken when they occur.Footnote 11 The implied responses are not merely theoretical. They are also connected with legal rules, reinforcing their practical salience by specifying penalties and sanctioning enforcement.
However, while each pillar casts a certain amount of light on the possibility of legitimate rebellion, none illuminates it fully. Even when they highlight one facet of the problem, they often simultaneously throw others into shadow. All three efface or exclude—and thus render it impossible to think clearly about—certain forms of legitimate armed rebellion. Moreover, as legal frameworks these pillars include a degree of indeterminacy: gaps in normative guidance that still need to be filled by means of practical judgment and perhaps further theoretical input.Footnote 12 On the other hand, each is also valuable to some degree, focusing our attention on a particular kind of threat to human rights and providing a framework for evaluating possible responses.Footnote 13 So although the three pillars may be inadequate to the problem of nonstate violence taken as a whole, I believe we are compelled to retain them in some form, even if reasserting the Rebel Perspective might require us to adjust and renegotiate certain aspects of them in order to fill the gap in the global normative architecture regarding the use of force.
Pillar I: Terrorism
Whereas it is possible to use the word “terrorism” to make non-normative distinctions, the term is mainly pejorative, condemning those it is used to describe. And while many scholars have argued that the term ought to be used equally to challenge the practices of states as well as nonstate actors, it remains a fact that it is more widely used in reference to the latter.Footnote 14 It is a key label by which those participating in public debates often refer to nonstate political violence that they regard—or want others to regard—as wrongful. And it is also a central concept in legal measures both nationally and internationally aimed at degrading the ability of nonstate actors to threaten human rights. Insofar as “terrorism” thus refers to cases that have crossed a moral line within the category of nonstate political violence, it implies the possibility (a contrario) that nonstate political violence might also take forms that are not wrongful. We might therefore be tempted to turn to the concept of terrorism for some guidance in this regard, but I think there are two major obstacles to doing so.Footnote 15 As I will detail in the following sections, the first is that the term is perennially susceptible to redefinition in ways that swing wildly between over- and under-inclusion; the second is that it might be understood to imply a variety of contrasting wrongs and social evils.
Terrorist and Nonterrorist Nonstate Violence
Yasser Arafat's speech to the UN General Assembly in November 1974 is a classic example of under-inclusion when it comes to the terrorist label. Arafat tried to challenge what he perceived to be the overextension of the category by appealing to the sympathies of people who felt that some armed resistance movements must be morally legitimate and therefore nonterrorist. His method was to define as “revolutionaries” those who fight for a “just cause” while reserving “terrorist” for foreign occupiers and colonial powers, a move that attempted to draw attention away from the nature of a group's tactics and refocus it on the nature of its aims. Like the more recent arguments by representatives from the Organization of the Islamic Conference at the committee tasked with formulating a UN Comprehensive Convention on International Terrorism, Arafat's speech implied that those committed to political liberation were immune from the charge of terrorism. The pursuit of a just cause, he suggested, was all that mattered when it came to deciding the difference.Footnote 16
While I believe that the ethics of armed resistance and rebellion are much more complicated than the picture Arafat painted, I sympathize with the sentiment that defining terrorism over-inclusively can cause laws and norms to be prejudiced against nonstate actors. On the other hand, it is also true that under-inclusion can cause laws and norms to be prejudiced against the security of states. Writing soon after 9/11, for instance, Alan Dershowitz argued that the threat from al-Qaeda was itself an indirect result of under-inclusiveness in defining and responding to the wrongs of terrorism during earlier decades.Footnote 17 He adopted an expansionary approach, arguing that the problem of terrorism was overwhelmingly urgent, presenting an extraordinary danger to Western secular states. It therefore demanded the most robust efforts to ensure that terrorist methods were deterred to the greatest extent possible. To achieve this end, it was essential that all those engaging in terrorism were prevented from gaining any reward for doing so, political or financial. Dershowitz also recognized that most people would regard at least some cases of nonstate political violence as legitimate, and that reasonable people would always be prone to disagree about genuine exceptions since there was no objective or universally accepted way to adjudicate particular cases. Given the urgency of addressing the common threat, however, he concluded that citizens and political leaders must agree to condemn all cases of nonstate political violence as terrorism. In other words, even if this meant applying the prohibition on terrorism in a way that was over-inclusive, the dangers of under-inclusiveness were too great to let it get in the way of determined counterterrorism.Footnote 18
I have little sympathy for the specific definitional proposals implied by either Arafat or Dershowitz, but both show how legitimate concerns motivate contestation, rendering the first pillar quite unstable and incapable of guiding political debate in an authoritative and determinate way. The problem is only intensified if we then turn to recent attempts to settle the wider controversy by appeal to customary law and legal treaty.
One highly significant attempt to settle the definition of “terrorism” as a matter of international law was set out in the “Interlocutory Decision on the Applicable Law” (February 16, 2011) issued by the Appeals Chamber for the Special Tribunal for Lebanon (STL). The STL, established by UN Security Council Resolution 1757 on May 30, 2007, is an international criminal tribunal to try those accused of killing the former Lebanese prime minister, Rafik Hariri, and twenty-one other people in a bomb attack in 2005. As such, its pronouncements on such matters had great potential significance, with the prospect of providing “long-anticipated answers to legal questions formerly only subject to academic debate,” and thus to indicate “the general direction” of subsequent jurisprudence.Footnote 19 The Decision argued that a “widely accepted definition of terrorism . . . has gradually emerged” across a wide range of national and international laws and treaties.Footnote 20 The “customary rule of international law regarding the international crime of terrorism, at least in time of peace,” argues the STL, is defined by the following components:
(i) the perpetration of a criminal act (such as murder, kidnapping, hostage-taking, arson, and so on), or threatening such an act; (ii) the intent to spread fear among the population (which would generally entail the creation of public danger) or directly or indirectly coerce a national or international authority to take some action, or to refrain from taking it; (iii) when the act involves a transnational element.Footnote 21
Since element (iii) is a prerequisite only for recognizing terrorism as an international crime, its terrorist character specifically is captured by elements (i) and (ii).Footnote 22
Far from settling the controversy as viewed from the Rebel Perspective, this definition exacerbates it by drawing a line between terrorist and nonterrorist that precludes the possibility of legitimate nonstate armed actions against an oppressive regime, at least prior to such time as the rebels “exhibit such features as to enjoy international legal personality.”Footnote 23 Legitimate rebels will fall afoul of the consensus definition and face the sanctions that follow if they resort to armed force, an act that will presumably be considered illegal by the governments they oppose, thus satisfying condition (i). Further, by using force to try to compel an oppressive state or government to change its policy or step down they will also satisfy condition (ii).Footnote 24 Crucially, rebels who have not yet secured recognition as belligerents in a state of war will be treated as if they are committing these actions in “peacetime,” a difficulty to which I will return below.
Thus, the consensus definition offers no criterion by which to recognize the salient difference between those who ought (morally speaking) to be denounced as “terrorists” and those who have a rightful claim to recognition as legitimate “rebels.” The STL decision suggests that such actions are “under no circumstances justifiable”Footnote 25 and asserts the “social necessity” and obligation to punish those who engage in them.Footnote 26 The STL assertion of a consensus definition thus instantiates and renders legally salient the problem of over-inclusion, thereby casting a deep shadow obscuring the ethics of legitimate rebellion.
The Wrongs of Terrorism
The second problem with guiding political reaction to potentially legitimate rebellions using the concept of terrorism is that it might be understood to imply a variety of contrasting wrongs and social evils. While these are all legitimate sources of moral concern, as recent philosophical debates suggest, they do not necessarily comprise a coherent, unified vision. My claim is that these wrongs will only begin to make complete sense when viewed as part of a wider account of the rights as well as the wrongs of nonstate political violence.
The first wrong that the term “terrorism” implies is the one identified by Arafat: the use of violence by nonstate agents for political ends that are not of the right kind or magnitude to justify it. This is the nonstate equivalent of a war that fails to meet the ad bellum standard of just cause. In contrast with unjustified violence, the second wrong is when nonstate agents use violence illegitimately (another ad bellum issue) in the narrower sense of the word, meaning that they lack authority or political standing. Thus, even if there is a just cause, there might still be a valid objection where nonstate groups are seen as unjust actors if they usurp the right of a political community to make collective decisions about what causes to pursue, the best methods of securing them, which risks are and which are not bearable in pursuing them, and so forth. Even people suffering from political oppression might have a well-founded preference not to be represented by self-appointed paramilitaries claiming the right to fight on their behalf. The word “terrorist” is sometimes used to reflect this aspect of the paramilitaries’ violence.Footnote 27
Another wrong frequently implied by the term “terrorism” is the deliberate harming of people with immunity. The relevant categories are often designated as “civilians,” “noncombatants,” or “the innocent.” But using this wrong to pin down a definition of terrorism has proven difficult. If there is no recognized state of war, then the combatant vs. noncombatant distinction is not relevant. The criterion of innocence vs. non-innocence has triggered debates both about whether combatants are always non-innocent in a way that renders them liable to attack and about the degree to which civilians whose claims to moral innocence are open to doubt can insist on immunity.Footnote 28 And if we were to define terrorism as the use of force against civilians, then the definition would exclude, by implication, soldiers and other members of the armed forces who are not civilians. Sometimes members of these groups might be liable to attack by legitimate rebels, and we might prefer not to designate such attacks as “terrorist.” But this should not be taken to imply that it is always more acceptable for nonstate groups to attack soldiers than to attack civilians.Footnote 29
Thus even if our sole concern were with specifying the meaning of “terrorism” as a concept, I do not think it would suffice only to enumerate the different wrongs with which it is identified and try to identify connections between them.Footnote 30 In many cases, the wrongs of terrorism are defined by exceeding the limits of the rights of rebellion. We would therefore also need an account of the conditions under which soldiers or others might become liable to harm by rebels. What sorts of cause would be of sufficient importance to merit such measures and under precisely what sorts of conditions? What conditions would rebels have to satisfy to be able to speak as the legitimate agents of those they claim to represent? And what principles of jus in bello should they follow? For convenience, I will refer in the remainder of this piece to rebels who satisfy all ethical requirements (having a just cause, possessing appropriate political standing, and adhering to an appropriate standard of jus in bello) as “legitimate.” The full complexity of the ethics of nonstate political violence cannot be grasped fully with reference to the notion of terrorism alone any more than the morality of war can be derived entirely from the notion of war crimes: crimes of war are partly defined against the rights of belligerents, and the wrongs of terrorism are likewise defined by the limits of legitimate rebellion.
I now turn to the question of whether the second pillar, the Law of War, can accommodate the normative parameters and consequences of the Rebel Principle in a sufficiently comprehensive and coherent way such that it remedies the inadequacies of the Terrorism pillar.
Pillar II: The Law of War
Michael Ignatieff has argued that while human rights are essentially nonviolent, enjoining individuals to settle differences through peaceful means, they might also justify a resort to armed rebellion against oppressors who refuse to recognize and respect them. If violence proves truly unavoidable, he maintains, rebels must then revert to the second-best ethical frame of reference designed for those circumstances: the Law of War.Footnote 31 Ignatieff's view reflects the common practice of debating ethical performance in armed conflict with reference to legal principles, particularly noncombatant immunity. And, indeed, insofar as it may be applied to conflicts involving legitimate rebels, the Law of War has considerable merits. Above all, it offers action-guidance for combatants, both state and nonstate, by defining the immunities of noncombatants and the limits of permissible force. By identifying direct attacks against noncombatants as criminal acts, the Law of War provides an incentive structure and a heuristic for combatants as well as a road map for those not engaged in combat to predict and avoid risks. Insofar as it thereby shapes the behavior of combatants, it may thus limit the destructiveness of armed conflicts. And importantly, the Law of War provides a normative frame of reference through which others can evaluate and contest the conduct of belligerents. But while the Law of War might offer a useful normative framework to interpret the use of force in some armed conflicts involving rebels, it is unlikely to do so for all such cases. This leaves users with two fundamental problems.
The first arises from sources of normative indeterminacy in international law when it comes to defining the political status of rebels. Whereas the boundaries and strictures of the Law of War are relatively simple when warring parties are states and they fight openly, things become much more problematic when one party is a nonstate actor. Central to the normal operation of the Law of War is the principle of equality according to which each side reciprocates in playing by the same rules: both abide by noncombatant immunity and both respect the right of opposing combatants to use force discriminately. The problem for justified rebels is that, while they may well be subject to the prohibitions specified by the Law of War, they might not be seen as benefiting from the permissions that are also associated with it—in particular, the “belligerent privilege,” which bestows both the right to participate in hostilities and the immunities enjoyed by combatants.Footnote 32 On what Jens David Ohlin calls “the orthodox view,” rebels cannot claim recognition and belligerent privilege under any circumstances.Footnote 33 And even on views that challenge this orthodoxy by suggesting that rebels might be able to claim recognition in some cases, the circumstances that trigger this right do not necessarily track the morality of rebellion very closely.
On the first of these circumstances, recognition might be expected to occur once the intensity of conflict and the degree of organization shown by the rebels passes a certain threshold, thus basing judgments about status on the terms of common Article 3 of the 1949 Geneva Conventions. Or, second, recognition might be based on the even more demanding criterion that nonstate parties must have control over territory before the conflict is necessarily treated formally as a “Non-International Armed Conflict,” based on the 1977 Additional Protocol II to the Geneva Conventions (AP2), Article 1.Footnote 34 Or, third, if we follow Ohlin's view on the deep logic of the Law of War, it might be that rebels can gain recognition for their combatants on the basis of satisfying the “functions” of belligerency, including the use of fixed insignia, the establishment of a responsible chain of command, abiding by the principles of the Law of War, and so on. But there are problems with all these approaches.Footnote 35
The first is that intensity will not necessarily track justification. Some illegitimate rebels with objectionable aims that are at odds with human rights—such as, historically, the fascists who rebelled against the Spanish Republic in 1936 or, indeed, the rebellious states in the American Civil War—might be able to achieve recognition on this basis faster than some legitimate ones. And on any such approach, morally legitimate rebels would suffer from nonrecognition and the associated penalties when the revolt is still incipient. A classic example is that of the American revolutionaries early in the war of 1775–1783, when the British Crown declared them to be traitors and denied them the customary rights of prisoners of war. Second, as Anthony Cullen comments, a “plain reading” of common Article 3 leads to the conclusion that applying its final clause “has no effect on the legal status of nonstate actors and as such does not in any way prevent a de jure government from treating them as criminals for their participation in a non-international armed conflict.”Footnote 36 So, while the recognition of a condition of (non-international) armed conflict might be obligatory on the basis of either common Article 3 or AP2, it is less clear that recognition of the rebels follows. In such a case, rebel fighters might find themselves liable to prosecution for war crimes under the Law of War if they violate noncombatant immunity and to a domestic criminal charge of murder even for ostensibly discriminate acts of war against state soldiers.Footnote 37
The third problem is that even if Ohlin is correct about the deeper logic of the law or if either basis in the Law of War—common Article 3 or AP2—could compel states to recognize rebels as having belligerent rights, it should not necessarily lead us to expect recognition in the right cases considered from the Rebel Perspective. Rebels without just cause might abide by the functional criteria of belligerency that Ohlin identifies; rebels with a just cause might sometimes be unable to.
This takes us to the second major problem with the Law of War as a normative framework for guiding and judging rebels. One reason rebels might not be able to satisfy Ohlin's criteria is due to the problems of wearing a fixed insignia or an identifiable uniform in a conflict against the domestic state, and the effect this has on the rebellion's chances of success. Though the 1977 Additional Protocol I (AP1) permits guerrilla forces to maintain the appearance of civilians (unless deployed for attack) during “international” conflicts against colonial powers, alien occupation, or domestic racist regimes, it does not extend to non-international, purely domestic rebellions (and some states reject this clause of AP1 altogether). It is certainly true that during some conflicts rebels may eventually command some territory within the state and their forces may reach a scale at which direct engagement with state forces using regular methods in conventional confrontations becomes feasible. In such cases the requirement that combatants wear uniforms or fixed insignia visible at a distance may then be comparatively easy to satisfy. And the moral rationale for doing so is likely to be very strong: Distinguishing clearly between rebel soldiers and the surrounding population on whose behalf, in theory, they fight is consistent with the aim of liberating those people and trying to seek protections for their human rights. But during the early phases of such a struggle, satisfying those requirements without jeopardizing all possibility of success might not be realistic. When state forces have command of the territory as a whole, their technological military dominance likely means they will be able to monitor and track down rebels much more easily than rebels can monitor and track down them. If guerrilla fighters take care to distinguish themselves by openly carrying arms, wearing combat uniforms, and the like, they risk exposing themselves to air strikes, for example.Footnote 38
The difficulty of complying with the uniform convention is just one of several possible ways in which the Law of War might prove too demanding for justified rebellions. As I argue elsewhere, other practices prohibited by the Law of War, such as the use of civilian disguise as a means of ambush or the targeting of civilians, might be justifiable in certain cases, say, when those civilians contribute in significant ways to state oppression. The pressure to adopt alternatives to conventional military methods is greatest where asymmetries of technology and military force between poorly armed rebels and well-equipped state forces are most severe. Such irregular methods may be justifiable particularly when pro-government forces refuse recognition to rebel fighters and systematically violate the Law of War.Footnote 39 Bracketing for a moment the question of just cause, this pressure can be seen in conflicts with insurgent forces across a wide range of contexts since World War II, from the Algerian War of Independence, to the Israel-Palestine conflict, to the occupations of Afghanistan and Iraq.
In summary, the Law of War is unlikely to map neatly and directly onto the morality of justified rebellion across all possible cases. This is partly a result of the very limited range of morally permissible tactics available to rebels. But above all, indeterminacy in the criteria governing the recognition of nonstate belligerents renders rebels uncertain about their status and leaves them vulnerable to the worst of both worlds: liability to punishment for crimes defined by the Law of War without immunity for discriminate fighting.
Pillar III: Nonintervention and the Responsibility to Protect
The third and final pillar to cast some light on rebellion is the Responsibility to Protect (RtoP). On the face of it, RtoP's modification of the Westphalian commitment to sovereign immunity from outside interference seems a promising way to justify responding to cases where legitimate but beleaguered rebels call for international assistance. But whereas its early formulations opened up a space within which to consider such cases, the political practice that has emerged since its promulgation in 2001 has closed much of it down again. Indeed, the resulting principle has much in common with the qualified principle of nonintervention sketched out by John Stuart Mill in his famous essay of 1859, “A Few Words on Non-Intervention.”Footnote 40
According to RtoP, states have immunity from intervention if they successfully discharge their responsibility to protect their own citizens and subjects, whereas sufficiently serious failures to do so may trigger a duty on the part of other states to intervene. The major change to the doctrine has been the redefinition of the thresholds above which the right to non-interference becomes moot and the duty to intervene is triggered. Early formulations specified only that human lives need be threatened on a “large scale” and constraints on intervention were stated in just war terms, chiefly with reference to the ad bellum principle of proportionality. So the upshot could be interpreted as a principle permitting State A to intervene militarily in State B if doing so could be expected to achieve a net benefit in protections against lethal violence as compared with not acting.Footnote 41 On the assumption that armed rebellion is similarly justifiable only in the face of a regime committed to using lethal repressive force, the doctrine might therefore have permitted intervention in a relatively wide range of cases where legitimate rebellion was under way and might have lent some assistance in defeating human rights–violating governments.Footnote 42 However, a later formulation in the 2005 World Summit Outcome document (endorsed by 191 countries) narrowed the triggers for armed intervention to “genocide, war crimes, ethnic cleansing, and crimes against humanity.”Footnote 43 Following the outcome document, the resulting convergence between RtoP and Mill's account of the principle of nonintervention was so striking that Michael Doyle found in Mill's arguments an effective account of the key normative parameters of the current doctrine.Footnote 44 And the central thrust of Mill's argument, of course, was to challenge the idea of intervening to assist rebels engaged in armed revolt against a domestic government.Footnote 45
Two key arguments for prohibiting assistance to legitimate rebels may be drawn from Mill's essay. The first we may call the “rebel learning” objection. Mill assumes that political freedom can survive only where citizens have certain necessary “feelings and virtues” to equip and motivate them to defend it. These include, above all, a profound appreciation for the value of liberty.Footnote 46 Despotism and slavery provide unpromising soil in which to cultivate these qualities, but there is some hope that the oppressed may acquire the necessary feelings and virtues in the “school” of political struggle. When this takes the form of open confrontation with the regime, then it is participation in the rebellion itself that deepens the roots of these values.Footnote 47 Intervening to support a revolution, the argument goes, therefore risks thwarting a learning process that is vital to political success. The second objection is epistemic. Mill assumes that only the successful defeat of domestic tyranny can demonstrate that people are ready for political freedom. This should be sufficient warning against interference: Helping rebels to secure an outcome that we cannot be sure they could have achieved by themselves might result in the establishment of political freedom for a people not yet ready to maintain it.Footnote 48 The force of this objection is strengthened by the assumption that intervention should presumably only be contemplated in the first place if the rebels seem likely to succumb without outside help.
There are, of course, exceptions to Mill's principle of nonintervention. He thought it might be permissible to assist secessionists fighting to secure self-rule against a larger, multiethnic state. The grounds for doing so are that intervention in such cases would not interfere with the self-determination of a people and, hence, with its learning processes; rather, it could facilitate self-determination of peoples by helping ensure that each has its own state and, consequently, a qualified immunity from intervention by others. Similarly, if one state wrongfully intervenes in a purely domestic rebellion, then the principle of nonintervention permits others to engage in counterinterventions aiming to offset any artificial distortions to the balance of domestic forces.Footnote 49 The chief type of case that permits intervention on grounds that override the rationale for nonintervention arise when, to use Walzer's influential expression, “acts that shock the conscience of mankind” cannot be stopped by any other means (or where intervention is the most proportionate means of stopping them). Such “severities repugnant to humanity,” as Mill calls them, are now specified as war crimes, crimes against humanity, genocide, and ethnic cleansing.
Like the formulators of the RtoP doctrine, Mill concludes that any material assistance to domestic rebels is prohibited unless it is directed against a state engaged in one or more of these crimes. This is a highly problematic conclusion when viewed from the Rebel Perspective. Unless Millians assume that the balance of military strength in a civil war always tracks the balance of political allegiances across the national population, there will be cases where rebellion is justified in principle but will be defeated unless it receives outside help. And, as critics of Walzer's appropriation of Mill's argument in Just and Unjust Wars have long argued, this assumption is tendentious, and the above cases likely.Footnote 50 A government that maintains the support of the regular armed forces (not to mention mercenaries) will usually enjoy major advantages in military strength even without outside help.Footnote 51 Its prospects for victory are therefore likely to be out of proportion to its legitimacy across the civilian population.
Insofar as RtoP narrows the exceptions to nonintervention even more drastically than Mill, it consequently casts a dark shadow that obscures cases visible from the Rebel Perspective. There are at least three hypothetical cases, each to a degree effectively concealed in a global order committed to upholding RtoP. In the first case, a just rebellion is precluded from the benefit of outside intervention. In the latter two cases, legitimate would-be rebels who adhere to just war principles but have no chance of receiving outside help are effectively prohibited from even beginning their struggle. They are therefore likely to go unnoticed.
The first case is the simplest: rebellion might break out on behalf of a people that values liberty enough to take great risks and make enormous sacrifices for it, and yet it might be defeated if the armed forces remain loyal to the incumbent regime. The rebels might be hopeful to begin with and yet find that their chances start to fade in the absence of outside military support to counterbalance the strength of a regime artificially inflated not by popular political support but by more effective armed forces. And state forces might manage to defeat them by conventional military means, avoiding the sort of atrocities that could trigger a military intervention likely to help the rebels. Let us call this type of scenario a case of “simple military defeat.”Footnote 52
To illustrate the second case, consider the following hypothetical. The political regime in a foreign state is oppressive, persistently and widely violating human rights and prohibiting the exercise of civil and political freedoms. Its inherently violent nature renders it liable to justified rebellion in principle, but it avoids committing the sort of sudden, large-scale atrocities that might trigger intervention under RtoP. Security forces are loyal to the regime, but the civilian population is almost uniformly opposed to it. Community leaders might therefore be able to initiate popular protest and resistance, but they recognize that doing so is likely to trigger an escalating cycle of violence, leading toward a civil war that they would have little or no chance of winning on their own.Footnote 53 In such a case, the ad bellum condition requiring that those beginning a war have a “reasonable prospect of success” would seem to prohibit resistance in the absence of outside help. Lacking an all-things-considered justification, opposition leaders would therefore remain quiet and avoid provoking the regime. We may call this sort of scenario a “futility trap.”
Let us then suppose the rebels did believe they had a reasonable chance of success. The jus ad bellum makes further demands that might still stand in their way if they could not call on international assistance. Proportionality demands not only that the initiation of war be nonfutile but also that it be expected to secure enough good in terms of lives defended or rights protected to outweigh the harms it is likely to cause. If this means that when deciding whether to initiate a confrontation rebels must take into account the numbers of innocent casualties they can expect as a result of the war taken as a whole (inflicted both by them and their opponents), then it is highly likely they will find themselves caught within what we will call a “proportionality trap.”Footnote 54 Cases such as the Baathist regimes in Syria and Iraq illustrate the two types of trap, both having demonstrated not only a capacity to crush revolts but also a willingness to inflict devastating casualties while doing so. That a similar effect might be achieved while using methods that could be defended as technically discriminate under the Law of War is exemplified by the Israeli bombardment of Gaza in 2008–2009 and in 2014. In such cases, any attempts at armed resistance (whether legitimate or not) are demonstrably futile and incapable of satisfying ad bellum proportionality.
The proportionality trap scenario can arise for any party—state or nonstate—that has just cause for war, but rebels are more likely to face it due to inferior arms and organization. And their predicament is made worse by what we will call the “proportionality paradox.” On the one hand, the more oppressive the regime, the greater the urgency of defeating it and, hence, the more compelling the in principle justification for rebellion (in other words, the gravity of the justifying cause). On the other hand, however, it is also likely to be true that the more oppressive the regime, the more brutal its methods will be if confronted, and the greater the costs it will be willing to impose as a deterrent. If the regime is able to ratchet up sufficiently the levels of damage it inflicts fighting the rebels while avoiding clearly visible breaches of the Law of War, then it may be able to increase the costs of resistance enough so as to make it self-defeating. To do so, the regime must ensure that the levels of violence inflicted on the rebels and their supporters greatly exceed those it routinely inflicts when it is unresisted. I think it is possible for regimes to do this. The Law of War permits belligerents a degree of proportionate collateral damage, but it is very hard to determine precisely when in bello proportionality has been exceeded. This means that, whereas the force of an in principle justification for rebellion increases continuously in proportion to regime brutality, the likelihood of an all-things-considered justification might not. The more willing a regime is to inflict such harms, the harder it becomes to satisfy the conditions for an all-things-considered justified rebellion against it.
Insofar as it relies on the visibility and success of a rebellion as indicators to outsiders of the readiness of a people for domestic liberty, the Millian approach to domestic rebellion is therefore highly problematic from the Rebel Perspective. We expect not only that people who are prepared for political freedom will sometimes fail in their resistance but also that they will sometimes show no overt signs of resistance at all. Thus, neither of these results tells us anything definitive about potential justification of or support for rebellion. The futility and proportionality traps that the Rebel Perspective illustrates show how something that looks like peace from the outside perspective of RtoP can be purely superficial, masking buried conflict.
Conclusion
I have argued that the existing global normative architecture regarding the use of force is inadequate as a guide for responding to rebellion internationally. While it casts limited light on some facets of the problem, the pillars that uphold it obscure a great deal across a potentially significant range of possibilities. The terrorism pillar obscures the possibility of nonterrorist, nonstate political violence, rendering legitimate rebels liable to counterterrorist measures and criminal prosecution for terrorist crimes. The Law of War pillar is likely to exclude at least some cases of incipient or low-intensity rebellion, leaving rebels without privileges or combatant immunity, and is problematic as a framework for evaluating rebels who justifiably use irregular methods that the pillar prohibits.Footnote 55 Finally, the RtoP pillar prohibits intervening to assist rebels who can satisfy just cause requirements, but who struggle to satisfy the need for an all-things-considered justification for rebellion without external assistance. Taken together, we might say that the three pillars thereby create what Ian Clark calls “zones of acute vulnerability”—political spaces in which people are unprotected from severe human rights violations.Footnote 56
My chief aim has been to show that there is such a problem, to define it, and to show why it needs attention. To what degree we might need to alter the existing architecture to address this problem is a matter for wider debate. But I will conclude by sketching out some suggestions.
Perhaps most clearly, the way in which we approach the definition of “terrorism” seems profoundly inadequate, particularly the consensus view, since it precludes recognition of legitimate rebellion from the very start. Securing states against true cases of terrorism is important, but the definition must be rethought based on the recognition that there is a moral right to rebel against intolerable oppression, and that doing so should not be categorized as terrorism. Second, as desirable as it may be that all participants in armed conflict be persuaded to abide by the Law of War, it is necessary to recognize that the moral right of legitimate rebels to use force is likely to arise at some point before they gain recognition as having a legal right to fight. It is necessary, therefore, to give careful attention to the threshold at which recognition may be demanded, thus triggering the rights of belligerents. This is a problem that I think law alone will not entirely settle. Rather, it should be approached as a matter of political judgment in which both legal and moral principles are taken into account along with the concrete specificities of the case in question. And there may be other ways in which political judgment is required to adjudicate between morality and law in relation to jus in bello.Footnote 57
Finally, incorporating the Rebel Perspective requires revision of the principle of nonintervention. An assessment of the various normative traps facing legitimate rebels indicates that intervention might be considered justifiable across a wider range of cases than is currently licensed by RtoP. For instance, by specifying more clearly the points at which assistance could justifiably be requested from the Rebel Perspective, a prima facie case could be made justifying material assistance to help legitimate rebels satisfy the success and proportionality conditions. Of course, there might be a worry that doing so could increase the number of overt civil conflicts.Footnote 58 On the other hand, however, admitting such possibilities could also help address the problem of moral hazard identified with current doctrines of humanitarian intervention. Alan Kuperman maintains that RtoP already encourages rebels to provoke their enemies into committing the sorts of large-scale international crimes that are currently the only recognized triggers for intervention.Footnote 59 Were the threshold lower, then the degree of violence rebels would be tempted to provoke might be lower too. The proportionality-based threshold indicated in the earliest outline of RtoP might therefore be a more suitable framework in light of this issue. If it were applied to cases involving legitimate rebellion, it could be formulated so as to permit assistance only on strict terms that reflect judgments under just cause, legitimate authority, and right intention, while taking due account of the difficulties that rebels will experience in satisfying the success and proportionality conditions without assistance.
Another objection to widening RtoP is that it could license a greater number of misguided or unscrupulous interventions. I am doubtful it would, however. The unscrupulous will always abuse the rhetoric of counterterrorism or humanitarianism to mask their strategic aims, as witness the Russian bombing of rebels in Syria. Misguided interventions, too, are likely to happen even without revision to RtoP. In fact, if anything, the lack of clear principles to guide morally justifiable attempts to assist rebels constitutes an important doctrinal gap rendering ill-designed intervention more likely. Well-intentioned but poorly judged attempts at assistance are more likely when they have to take place “under the normative radar,” so to speak. And mission creep (such as when an intervention to prevent war crimes was used to help remove Qaddafi in 2011) and secrecy are more likely to occur when political leaders follow the intuitions of the Rebel Principle without an adequate and recognized normative framework to guide and legitimate their actions. Accountability might therefore increase under a properly formulated revision to rules governing intervention.
If so, then one key issue likely to receive more careful attention from interveners is the agency of those they seek to benefit. Whereas the current paradigm of intervention for humanitarian purposes tends to view those threatened with human rights violations chiefly as patients awaiting help, the Rebel Perspective reflects the capacity that they have for political agency. It therefore underlines the fact that intervening states must negotiate an appropriate relationship between two types of agent: interveners and rebels. And this in turn requires critical reflection on the political legitimacy of rebellion leaders as well as the justice of the ends they pursue. In a context like Syria, it would require taking special care to form alliances with the most promising of the factions within the wider rebellion, as well as working strategically both to guide and shape their political ambitions and to help them secure the strongest possible position within the rebellion as a whole.Footnote 60 Of course, these aims would have to be weighed in the balance against the enormous complexity of intervening in a conflict involving multiple parties, including those assisting the regime. While the Rebel Perspective might help formulate the best imaginable (I hesitate to say “ideal”) strategy, political wisdom demands careful consideration of matters such as the risk of wider conflict arising from attacks on international proxies, for instance. On the other hand, the presence of Russian weapons and forces on the side of Assad makes the Syrian case a more amenable one from a Millian point of view: as Mill himself declared, a “government which needs foreign support to enforce obedience from its own citizens, is one which ought not to exist.”Footnote 61
