Outline of the two possible legal innovations

It is worthwhile looking at the law governing inter-State relations for a possible way out. Here a similar tension exists between the goal of preventing the illegal use of force in the first place and the need to provide humanitarian protection if this first goal has not been reached. There is reason to doubt that this tension can be completely dissolved, but current international peace and security law tries to moderate this tension in the following manner: it not only prohibits the use of force (save for a limited number of exceptionsFootnote ³⁹), but also criminalizes, though the crime of aggression, individual participation in serious instances of the illegal use of force.Footnote ⁴⁰ Yet the threat of international punishment is confined to those “in a position effectively to exercise control over or to direct the political or military action of a State”.Footnote ⁴¹ Within that legal system, the privilege of combatancy retains its potential also for the armed forces of the aggressor State to comply with IHL. As regards relations between States, international law therefore addresses the State leadership's responsibility to prevent a war and, if such prevention fails, provides the armed forces with an incentive to wage war as humanely as possible.Footnote ⁴²

This inter-State model suggests how the privilege of combatancy could be extended to rebels while maintaining a powerful normative barrier against their taking up arms in the first place. A reformed international law of NIAC could rest on two pillars, similar to those of its inter-State counterpart: it could embrace a privilege of combatancy in the jus in bello interno and at the same time prohibit any non-State organization from launching an armed attack against its government through a new jus contra bellum internum. Footnote ⁴³ Within such a model, international law would no longer (save for the intervention by the UN Security Council) remain indifferent to the outburst of a NIAC, but would make the prevention of civil war its direct concern.Footnote ⁴⁴ Within the reform model under consideration, the State monopoly on the use of force would therefore not be weakened by the introduction of the privilege of combatancy in the law of NIAC; this monopoly would in fact be strengthened through the introduction of a ban on armed rebellion under international law.Footnote ⁴⁵ The latter prohibition, however, would have to be confined to the leadership level of the rebels, in accordance with the definition of the crime of aggression.Footnote ⁴⁶ In this way the incentive, provided for the individual rebel soldier through the new privilege of combatancy, to fight the NIAC in conformity with IHL could be maintained.Footnote ⁴⁷

Whether a jus contra bellum internum could be upheld without exception is a most delicate question. It bears emphasizing that considering the human suffering in NIACs (and this includes above all the suffering of those not participating in the fighting), there are compelling reasons to prohibit not only regime change by force but also the taking up of arms against a government which does not act in full conformity with international human rights standards. But what if internal politics become cancerous and a government-orchestrated genocide or crime against humanity of murder or extermination is imminent? In light of such a scenario, there is a strong case for an (extremely limited) jus ad bellum internum – that is, a right of self-defence of the civilian population under lethal attack, “until the UN Security Council has taken the measures necessary to maintain international peace and security”.Footnote ⁴⁸ As of yet, such a strictly circumscribed jus ad bellum internum has only received extremely limited scholarly attention.Footnote ⁴⁹ The underlying problem has rather been debated from the perspective of an intervening foreign State. This has been done either by reference to the (subsidiary) international responsibility to protect or, more controversially, under the heading of a possible right of States to forcible humanitarian intervention.Footnote ⁵⁰ This is not particularly satisfactory from the perspective of the law's internal consistency. At a stage of the development of international law where non-State actors are no longer necessarily relegated to being indirect beneficiaries of State rights, the justification for an exceptional right to use force in the case of an impending humanitarian catastrophe should be rooted in the individual right of self-defence of the civilian population under attack, and any right of a foreign State forcibly to come to the assistance of the targeted population should be conceived of as a form of collective self-defence to the benefit of the population concerned and derivative of the civilian population's right. In that context, it is extremely interesting to note that, in its declaration at the March 2013 Doha Summit, the Arab League explained the legality of supplying weapons to the Free Syrian Army on the basis of the right to self-defence of the Syrian people.Footnote ⁵¹

In its essence, a fully symmetrical legal regime for NIAC was already known in the nineteenth century, but at this time it had to be activated by the recognition of belligerency, whose occurrence could never be taken for granted. A completely symmetrical law of NIAC would, however, have to be complemented by a new international law against civil war. By virtue of such a jus contra bellum internum, the prohibition on the use of force in international law would be extended to non-State organizations with a capacity to initiate massive violence. The new jus contra bellum internum could build on the now consolidated practice of the UN Security Council to consider a NIAC as a threat to international peace and security,Footnote ⁵² and it would enrich international law through the introduction of a hitherto unknown discriminatory concept of civil war. It is therefore conceivable in theory to compose a new international law on NIAC, which would reach yet another stage of its development in Antonio Cassese's spirit and which would even extend slightly beyond Cassese's last word on the subject through the distinction between an asymmetrical jus contra bellum internum and a completely symmetrical jus in bello interno, mirroring to an extent the law governing the relations between States. But what about possible fundamental objections to such a new law, and if such objections could be addressed, what about the workability in practice of such a new law to the benefit of humanity?

A “realizable utopia”?Footnote ⁵³

The idea of achieving full symmetry in the law of NIAC gives reason to pause and to have a fresh look at the legitimacy of the privilege of combatancy in the law of IAC in the first place. In this respect, the starting point is to recognize that, in the age of a prohibition of force in international relations, the legitimacy of a permission to use lethal force for the individual soldier of the aggressor State must be open to question. A couple of years ago, Jeff McMahan subjected the privilege of combatancy for the “unjust warrior”Footnote ⁵⁴ to an elaborate criticism in terms of moral philosophy. He argues that neither the idea of a mutually valid consent,Footnote ⁵⁵ nor a general presumption in favour of an unavoidable error of law, nor the reliance upon an allegedly binding superior order to wage war given by the leadership of the aggressor State, constitutes a good reason to justify such a privilege.Footnote ⁵⁶ It is important to note McMahan's argument because it carries even more weight with respect to a non-State “unjust warrior” in a NIAC. Yet, McMahan has explicitly refused (while adding the caveat “at present”) to argue for an asymmetrical construction of the law of IAC as a consequence of his moral argument. In that context, he refers inter alia to the incentive which flows from their privilege of combatancy for “unjust warriors” to comply with the law of armed conflict.Footnote ⁵⁷ In a recently published, thoughtfully written book chapter, Frédéric Mégret takes a different position and questions the symmetrical privilege of combatancy in the law of IAC. This position is (among other things) based on the disbelief that the privilege of combatancy in fact works as an incentive for compliance. Mégret denounces the widely held opposite view as being based upon no more than the unreflected psychological assumptions of laypersons.Footnote ⁵⁸ These doubts deserve to be treated seriously, and they call for another round of reflection on the incentive for compliance potential of a privilege of combatancy for the “unjust warrior” with respect to both the current law of IAC and a possible future law of NIAC. Thought should also be given to the feasibility of an empirical study on the subject. As regards the law of NIAC, which is at the heart of this essay, one thing is clear: the introduction of a privilege of combatancy for non-State fighters in a NIAC imperatively requires the realistic prospect of thereby significantly improving compliance with IHL by the non-State party to the conflict.

The impossibility of justifying such a prospect would, however, not render the debate about a new jus contra bellum internum moot. It is true that the idea of such a new law has been presented here as a complement to a fully symmetrical jus in bello interno. But it bears emphasizing that a new jus contra bellum internum (including an exceptional jus ad bellum internum in case of an impending humanitarian catastrophe) should be a matter for debate in any event. In other words, a move towards complete symmetry in the law of NIAC is dependent on the introduction of a new jus contra bellum internum, but this statement does not necessarily hold true the other way round.

The idea of a new jus contra bellum internum also gives rise to important questions. First of all, careful thought must be given to the question of whether it is possible to formulate a new international prohibition (under the threat of international punishment) as well as a limited exception therefrom, which is both satisfactory in substance and sufficiently precise.Footnote ⁵⁹ Confirmation is also needed on whether a non-State party leadership level can be identified in most instances of an emerging NIAC so that the new international rule could indeed work as the intended legal barrier against the unleashing of a civil war. Furthermore, the workability in practice of the new legal regime under consideration would be exposed to serious doubt if the outbreak of a NIAC was generally preceded by an escalation dynamic that could not be disentangled with sufficient clarity to reliably attribute responsibility. Sadly, the situation of Syria (since 2012) provides us with a fresh example of such dynamics. Yet another challenge results from the fact that there are national constitutions which provide for a right to armed resistance under certain conditions.Footnote ⁶⁰ There would thus be a need to bring both levels of legal regulation in harmony with each other. Finally, there can be no doubt that the introduction of a jus contra bellum internum would constitute an ambitious legal reform. One wonders whether it is a bridge too far for international law, at least for the time being.

In view of all these important and as of yet barely discussed questions, the reform model for an international law on NIAC is not here presented for immediate adoption. Instead, the model is submitted in a scholarly spirit for debate as to whether it could be seen as a “realizable utopia” in the sense of the late Antonio Cassese. It would be a multifaceted and challenging debate, but as civil wars are one of the main evils of our time, an improved, if modest, contribution by international law to deal with that evil at its roots would therefore justify the most demanding intellectual effort.

The hard-to-defend asymmetry of the humanitarian regimes of international and non-international armed conflicts

To begin with, that more could be done to improve compliance with the humanitarian regime of NIAC is fairly evident. In terms of humanitarian policy, finding ways to incentivize both States (repressing non-State armed groups with a vengeance because they feel it is their sovereign prerogative to do so) and non-State armed groups (unsocialized in the ways and values of international law) to respect certain humanitarian limitations has become a sort of Holy Grail of international humanitarianism. A distinct and influential strand of humanitarian thought has for quite some time argued for an upgrading of the regime applicable to NIAC by reference to that of IAC. Kreß's analysis in this respect lies firmly within the current progressist canon that sees the need to develop the law relating to the conduct of hostilities in NIACs.

Many of the arguments for assimilating the regimes of IAC and NIAC are compelling morally if not always strictly in positive law. In many respects, the regime of IAC is more comprehensive and thoroughly protective than that existing in NIAC. It is often argued that limitations of the humanitarian regime in NIAC derive from sovereign concerns that have more to do with protecting a domaine réservé than creating the best of humanitarian worlds. What difference should it make, after all (except for sovereignty-obsessed States whose motivations we have increased reason to be wary of), whether as a victim of war one is situated in an IAC or a NIAC? Is not a lot of energy wasted in arcane issues of conflict classificationFootnote ⁶² that could be usefully redirected to more productive pursuits? Auspiciously, then, many international lawyers would argue that, as a result of what Kreß refers to as the “Tadić dynamic”, the laws of war in the last twenty years have moved towards a greater humanization, taking their cue from international human rights law.Footnote ⁶³ It is only a small step from there to effectively argue for a“unified” law of war that would apply equally regardless of the conflicts,Footnote ⁶⁴ and the trend towards convergence at times seems unstoppable.Footnote ⁶⁵

The danger of over-application of the laws of war

In addition to this proactive commitment to the development of the law, Kreß is also wary of something that almost goes in the opposite direction, namely the risk of what one might call the “over-application” of the laws of war. This may be counter-intuitive since we are used to States seeking to deny the application of the laws of war, but evidently the opposite is also true. It is perhaps the natural tendency of any body of law to seek to apply to an ever-increasing range of situations, spurred by good intentions and fear of a legal void. That may be particularly the case when that body of law is associated with humanitarian virtues. A fortiori, if this expansion is backed by powerful and well-organized institutional agendas, the case for it may be overwhelming.

However, as Kreß notes in one of the most clear-sighted passages in his article, there is also a danger of displacement of other bodies of law where arguably they should remain applicable or continue to be controlling in the background. In particular, a comprehensive application of a conduct of hostilities regime in NIAC might be said to objectively displace the ordinary application of human rights. In IAC, that departure from ordinary expectations about core rights (to life, to physical integrity) is perhaps less intensely felt because of the highly anomalous status of the international system normatively, and the way structured assumptions about war have led us to a state of permanently degraded expectations about what can be achieved. The problem is that protection is relative: what is most humanitarian (in the laws of war sense) is not necessarily what is most protective. The regime of IAC may be more protective than (i) nothing and (ii) the regime of NIAC, but it is certainly not as protective as normally applicable human rights standards. The issue of what exactly is being displaced should therefore be front and centre of any attempt to import a legal regime originally designed to address very different kinds of situations.

In this respect, I would add a further point to signify my profound agreement with Kreß on this issue. The humanitarianism inherent to non-international violence starts from fundamentally different premises than the humanitarianism inherent to inter-State violence. With regard to the latter, the starting point, as exemplified perhaps a little stereotypically by the founding myth of Solferino, is a situation in which almost no rules prevail in the waging of violence. Humanitarianism, then, is a significant victory over barbarism. Domestically, however, the situation is quite different. Indeed, the starting point is one that is ideally characterized by the pursuit of human rights. Humanitarianism, therefore, does not intervene in a void but rather displaces a thick pre-existing network of obligations, duties and norms. It significantly downgrades our ordinary expectations of justice. Understanding this asymmetry between the humanitarian projects in IAC and NIAC is crucial. Hence the international humanitarian solicitude, however well-intentioned, is one that the human rights community may understandably be wary of in principle.Footnote ⁶⁶ In practice, for example, whilst we should rightly be wary of States who refuse to acknowledge armed conflicts on their territory, we may also have reason to find problematic States who too readily find conflicts occurring there.

Now it might of course be argued that the applicability of the laws of war is better than nothing and may introduce a modicum of restraint in a context that is fast spiralling out of control. That may be a price worth paying for pragmatic reasons (an issue further addressed in the final section). A time of “armed conflict” may not be the time to be picky about rights, and one should seek to save what can still be saved in deeply compromised moral circumstances. This is the characteristic, almost foundational motive of the humanitarian project in IAC as well as in NIAC: humanitarianism is preferable to unrestrained violence if violence is inevitable. But we should also always keep in mind that even “humanized” laws of war may be little consolation for a broader flight from rights. The irruption of the laws of war in the domestic context interferes with a relationship structured around radically different assumptions. It displaces dearly held ideals about the sanctity of human life. We should at least recognize that concession to humanitarianism for what it is, namely a deep bowing to the inevitability and to a degree the legitimacy of war as an institutional practice in international society.Footnote ⁶⁷ Whether that is a move we are ready to make as confidently in NIAC as in IAC is the real normative question. It is thus very much to Professor Kreß's merit to have raised it.

The need for a domestic jus contra bellum

Finally, in addition to an amplified humanitarian regime, one of the more intriguing points suggested by Kreß is that a form of non-international jus contra bellum should be developed. There is no doubt that this is quite a striking and radical suggestion, and one which is bound to be very sensitive with States who traditionally have considered that the repression of violent challenges to their power is in principle (means aside, and even then, grudgingly) entirely within their sovereign prerogative. That is of course because behind any jus contra bellum is an implicit claim of jus ad bellum, one that could operate to the benefit of the State (say, in the pursuit of its legitimate maintenance) but also to the benefit of non-State armed groups. Very briefly, I broadly concur with three types of arguments made by Kreß for internationalizing the question of the legal use of force domestically.

First, NIACs represent today the vast majority of armed conflicts, and they have significant transnational and international consequences (regional destabilization, refugee flows, escalation to inter-State conflict). In other words, sovereigns and non-State actors can hardly claim that they can keep domestic violence within their borders, and the international community is within its rights when it seeks to scrutinize the conditions under which violence is triggered. Such an argument has already been invoked quite successfully in the context of constructing a humanitarian regime for NIAC. Second, as international law becomes normatively denser and more hierarchical, it simply cannot be expected to blandly accept that the protection of sovereignty through repression of non-State violence is always its ipso facto right. If nothing else, there are extreme cases – for example, that of an ongoing genocidal campaign – where it is inconceivable that international law would stand on the side of the sovereign oppressor, denying the would-be victims the right to protect themselves.Footnote ⁶⁸ Third, in effect there is a long tradition of international legal thinking and practice that has foregrounded the existence of legitimate practices of violence in fighting tyranny, radical oppression and criminality. This tradition has re-emerged more strongly in recent decades as part of an increasingly influential discourse of human rights. International law is, in other words, not starting from nothing, and there is merit in seeing a discussion about domestic ad bellum as a form of continuation of long-existing debates rather than a sudden proposed innovation.Footnote ⁶⁹

In concluding this first section, I would like to point out that it is not clear which of Kreß's three concerns, outlined above, is foremost in his mind. Kreß's article opens with an emphatic evocation of the humanitarian project and is, after all, published as part of a special issue on the scope of the law in armed conflict in the International Review of the Red Cross. The shadow of Cassese looms large in the article and orients it towards the humanitarian dimension of NIAC. However, towards the end, the article becomes much more interested in the jus contra bellum issue, almost as if such a bold move as the conferral of a POB needed to be compensated by something that would reassure States. Both concerns, however, might be in tension with the risk of overextension of the laws of war. This is why the connection between these concerns needs to be the object of further investigation.

The privilege of belligerency is not in itself humanitarian progress

A certain confusion is created by the fact that the POB is largely contained in laws of war instruments.Footnote ⁷² However, there are practical and political reasons why this may be so which do not necessarily lead to an inference that it is an intrinsically humanitarian issue. In fact, the POB's presence in humanitarian law instruments appears to be largely accidental. Even though it may be a political condition of feasibility of the humanitarian project, that does not in itself make it a humanitarian provision. Rather, if anything the POB pertains to a sort of proto jus ad bellum, somewhere in a grey area between the jus ad bellum and the jus in bello, which identifies who, a priori, ought to be considered a privileged combatant for the purposes of the laws of war, and therefore who can be considered as a (relatively) legitimate participant in the particular form of violence internationally known as “war”.Footnote ⁷³ Its Statism (State troops almost always benefit from it, where non-State armed groups only do so under strict conditions) is not accidental but very much a recognition that sovereigns are clearly the historically preferred participants in armed conflict.Footnote ⁷⁴

Let us first consider the juridical significance of the POB in and of itself (not in some broader, instrumental sense). The POB effectively functions as an immunity from prosecution – or even a permission – for what might otherwise be considered a criminal act (killing and wounding other human beings, provided they are combatants). This much at least should be clear, therefore: the POB is not per se a form of humanitarian protection, except in the twisted sense that it protects combatants who kill other combatants.Footnote ⁷⁵ It is generous towards the very act of war, but it does exactly the opposite of what humanitarianism normally does and is about: limiting violence. It may be that it is part of a broader, realistic trade-off that makes the humanitarian project more palatable to States (by saying “we recognize your unique privilege to kill combatants to encourage you not to kill non-combatants”), but that is not the same thing as saying that it is in itself humanitarian.

In fact, one could argue that the fairly wanton killing of combatants is the one irreducible non-humanitarian blind spot of humanitarianism, the bare definition of war that the laws of war have never dared to challenge. It is true that prisoner-of-war status – which is the logical consequence, if one has been captured, flowing from the POB – is humanitarian at least in the sense that it proposes that the wounded or those who have surrendered should not simply be killed (which might otherwise make sense in war). However, this limited humanitarian protection could still be obtained without the conferral of a POB. Indeed, wounded or surrendered combatants could be entitled to certain forms of safe detention and yet be prosecuted for having killed other combatants.Footnote ⁷⁶ At any rate, the idea that combatants can kill each other legally is an abnormal exception to the notion present in most legal systems that one cannot consent to one's own death at the hands of others. Rather than limiting war, the POB makes the very practice of war, understood as a legalized activity, possible.

The privilege of belligerency is problematic from the point of view of human rights

In addition to not being intrinsically humanitarian, the POB is arguably one of the crucial ways in which the ordinary promise of human rights is downgraded under the banner of humanitarianism. Internationally, this is perhaps less evident because of the paucity of States' extra-territorial legal obligations and the broad positive-law deferral to the laws of war as the lex specialis.Footnote ⁷⁷ The POB has become such an assumed part of the laws of war that lawyers typically do not question the legality of killing other combatants – surely this must be the incompressible minimum on which the very existence of the laws of war depends. Nonetheless, from a rights perspective informed by moral and cosmopolitan thinking (as opposed to positive international human rights law), the POB still means that individuals can kill or be killed as part of the mere existence of an objective state of affairs between sovereigns – an “armed conflict” – at the expense of the maximization of their rights. Specifically: (i) enemy combatants fighting a war of self-defence can be legally killed, which is a hard-to-justify encroachment on their right to life;Footnote ⁷⁸ and (ii) combatants ordered to engage in a war of aggression can essentially be sacrificed by their own State for an illegal end, again in violation of their right to life.Footnote ⁷⁹

In domestic contexts, this downgrading of the ordinary promise of human rights through the granting of a POB would be even more evident because, as has been pointed out, the starting point is the applicability of internationally protected human rights. The downgrading would manifest itself in the simultaneous undermining of the existing framework of domestic, particularly criminal law, and of human rights. When it comes to domestic law (whose protections are indispensable to human rights), the POB undermines the idea that killing or wounding others without legitimate and legal authority or permission (e.g., individual self-defence) is, straightforwardly, a criminal act (in addition to being in contradiction to the State's monopoly on the legitimate use of violence). It thus dramatically weakens the rights of those who are targeted in armed conflict, whose expectation that their right to life will be upheld in all but the most exceptional circumstances now gives way to a situation in which they can be killed legally.

Where killing State agents lawfully exercising their authority is normally considered one of the gravest crimes, it would suddenly leave such agents legally defenceless against attacks. This exemption to the most cherished domestic laws, then, would be a boon to rebels, suddenly normalizing the completely illegal into a mere act of war. Perhaps more importantly, when it comes to human rights law, the POB is particularly relieving of obligations that would normally apply to State agents. In cases where a State's forces would normally be liable for human rights violations for killing individuals (except in narrow circumstances such as during an arrest, while keeping order, or self-defence), they could now avail themselves of the considerable violence implicit in the laws of war to kill those who would henceforth be considered as combatants. A POB for State troops in NIACs is a formidable licence to kill.

The privilege of belligerency is problematic in terms of jus contra bellum

Kreß distinguishes the issue of granting a POB in NIAC from that of developing a domestic jus contra bellum and presents these as largely irreducible issues. The reinforcement of the domestic jus contra bellum is offered up as a sort of quid pro quo for having introduced a POB in NIAC. What this seems to acknowledge yet simultaneously neglects, however, is the considerable objective role that the POB would have in legitimizing resort to force in the first place. It is as if Kreß's jus in bello proposal has simultaneously and discreetly undermined his own jus contra bellum commitment. One cannot help getting the impression, in other words, that the proposal to grant a POB to actors involved in NIACs removes one of the clearest disincentives for resorting – including illegally – to the use of armed force domestically in the first place. Whilst the concern with granting a POB to non-State armed groups is one that Kreß associates with traditional States, it is also crucially important to note that the goal of deterring an illegal rebellion or, for that matter, an illegal war by a State against its population should rank as a major human rights concern in its own right.Footnote ⁸⁰

Again, I will turn in the next section to the ways in which the POB is nonetheless often understood as part of a broader system of trade-offs which characterizes the laws of war, so that even its unsavoury aspects are considered redeemable. For the time being, let us note that the POB in IAC not only protects soldiers engaged in war in general, but also specifically protects soldiers involved in illegal wars from what might otherwise be the consequences of such participation. That this is essentially a sunk cost of doing international warfare does not relieve us from highlighting it for what it is, especially at the moment of extending it to another, very different type of warfare. The POB effectively concedes the existence of a broad, unconditional right to kill combatants, even for individuals who clearly know that they are engaging in actions that are illegal – even criminal – in the first place. It thus removes what might have been a particularly strong safeguard against aggression, namely the sure knowledge that one cannot kill combatants when one is knowingly engaging in a war of aggression.

Of course, this result is powerfully reinforced by the fact that the international criminal law of aggression has itself conceded that only senior commanders can be liable for the crime.Footnote ⁸¹ Low-ranking officers and soldiers (in practice, the vast majority of the military) who implement orders to carry out an aggression fall outside the penal reach of the jus ad bellum. It is thus logical that they are given the protective cover of the jus in bello (it would be paradoxical, of course, if the jus ad bellum let them off the hook but the jus in bello did not). It is perhaps no surprise as a result that individual soldiers do not feel particularly compelled to challenge the violations of the jus ad bellum that they are ordered to take part in, given how the jus ad bellum and jus in bello conspire to fashion them into largely irresponsible actors. Still, even allowing for that specific limitation, the POB also applies to those commanders located at the top of the chain of command who may be held liable for the crime of aggression, but would nonetheless not be liable for killing the combatants they knew they had no ad bellum right to kill in the first place. The point here is not as such to criticize this facet of the regulation of IAC that is often seen as unassailable even by the majority of moral theorists writing about the topicFootnote ⁸² (I will return briefly to this issue in the conclusion). Rather, it seeks to warn about what is exactly proposed to be imported in NIACs.

Indeed, there is no reason to think that these very significant concerns with the POB would be redeemed in NIAC; quite the contrary. Inserting a POB in NIAC may import precisely one of the most problematic features of the regulatory apparatus of IAC. Even though a POB would presumably not prejudge the ad bellum legality of the rebels' acts, it would at least provide them with a significant legal cover regardless of that legality. Rebels might not have a just or legal cause, but they would nonetheless be recognized as being a relatively legitimate actor for the purposes of engaging in war. No longer risking prosecutions for the mere fact of attacking State troops, they would find themselves significantly empowered, and a very big step would be taken in the direction of at least partly protecting unjust or illegal rebellions. For the State, such a formal and automatic recognition of a POB to rebels might be exactly the sort of danger of overextension of the laws of war against which Kreß seeks to guard us: one that would deprive the State of its full ability to maintain public order and discourage sedition, in the name of prioritizing a humanitarian objective above all else. It would also incidentally mean that non-State armed groups would have an objective interest in bringing their campaign to the level of an armed conflict so as to benefit from the POB and, as a consequence, launder their individual behaviour. Needless to say, this is a morally and legally paradoxical outcome.

Here it may be worth bringing attention to a specificity of NIACs when it comes to the ad bellum that cautions against any broad introduction of the POB. One of the typical arguments for denying that individual soldiers can engage in aggression (and therefore ought to normally benefit from a POB in IAC) is that it may be hard for them to know that the war they are asked to fight is, indeed, an aggression. Whatever the merits of this argument in IAC, it should be noted that it is particularly weak in the context of NIAC. The argument that individual rebels “could not know” whether an attack on the State was legal or not is clearly less well taken than its international variant: individual rebels are not members of stable and institutionalized military structures (“private armies”) waiting to be given orders, and who can therefore be forgiven for going with their leaders' choices. Rather, they are typically cause-oriented participants in armed conflicts that they chose, and should thus less easily be able to claim innocence from such choices.

Moreover, the international confirmation of a broad POB for State troops in NIACs, regardless of the legality of their campaign, would launder behaviour that might be deeply problematic from the point of view of the jus contra bellum and even human rights. Imagine for example that the State is fighting against a non-State actor that is an emanation of a civilian population trying to protect itself from genocide: would it make sense, legally or morally, to consider that the POB always applies to State troops effectively helping a project of annihilation of a people? Kreß's indication for a more humanitarian regime of NIACs, therefore, may come at the cost of being less humanitarian vis-à-vis fundamentally unjust forms of repression. Non-State aggressors would, for most purposes, be treated in the same way as non-State self-defenders. The POB (at least in the dominant form in which it is conceptualized) is a great humanitarian equalizer, precisely where the jus contra bellum seeks to draw stark differences.

The privilege of belligerency in international armed conflict was never particularly thought of as an incentive for humanitarian compliance

One might formulate the fundamental humanitarian reason for having a privilege of belligerency in the following manner: (i) we recognize that the POB is not in itself a humanitarian feature of the laws of war, but (ii) its conferral is part of a humanitarian “grand bargain” in which a limited type of combatant activity is defined as legal (killing combatants) in exchange for many other activities being defined as illegal (killing non-combatants). However, there are two ways of assessing this grand bargain as it is understood at least in IAC. The first one suggests that at a very general level States, as international law's principal norm bearers, would not accept the very idea of the laws of war if they did not grant State combatants a POB in NIACs. Even if this were true, it is also true that States are unlikely to see any problem with rebels not being granted the POB. In addition to this general sense, in NIAC the concession of a POB is typically proposed as more directly and tactically a factor in respect for the laws of war. Indeed, this idea that the POB is an incentive (and its absence a disincentive) is very prominent in the discourse and constitutes one of its most unquestioned assumptions.Footnote ⁸⁶

I believe, however, that there are many reasons to take this assumption as a descriptive statement with a salutary pinch of salt. Since Kreß is gracious enough to point to some of my own inchoate attempts at grappling with the question,Footnote ⁸⁷ this may be an opportunity to further scrutinize humanitarian assumptions about the uses of the POB. Indeed, too much is at stake to loosely assume that these, which are often heavily indebted to an international model of armed conflict, hold true always, often, or ever. The view is premised on the quite widespread but in my opinion wrong view that the POB in IAC (from which most lawyers take their cue) was itself historically part of the humanization of IAC. But there is no sign in the negotiations of the Geneva Conventions that the POB was to be granted as a form of incentive for troops to comply with the laws of war. In other words, it is not the case that the parties to the Geneva negotiations were initially reluctant to recognize a privilege of combatancy for those involved in an armed conflict but were nonetheless swayed to do so by its humanitarian promise. Rather, they cast the privilege of combatancy as one of the founding, unquestionable and axiomatic starting points of the laws of war, for reasons that were themselves antecedent to and distinct from the humanitarian project, and to which I have already alluded. It is probably only later that, as a key feature of the regime of IAC, it became casually assumed that it must surely be central to the Geneva Conventions' humanitarian promise.

The problem is that had the recognition of a POB really been granted as a way of humanizing war, it should have been granted conditionally. In light of the fact that it is given “for free”, as it were, to State forces, it is hard to accept that it is really an incentive as opposed to simply a privilege that flows from the identity of certain actors (i.e., States) in war. If something is granted ab initio based on (sovereign) status and cannot be lost even if abused (and this is certainly the case for State actors, who always retain it, regardless of how many civilians they massacre), it can hardly be said that its granting is part of doing everything possible to ensure that actors restrain from acting badly. In other words, if there was really a quid pro quo between humanitarian obligations and the privilege of combatancy, then obviously one should not be allowed to have the benefits of the latter without any of the constraints of the former. Conversely, a quid pro quo would mean that the POB could be withdrawn from those having committed grave offences. Instead of being an incentive to comply with the laws of war, the automatic conferral of the POB is in that respect a disincentive, or at least not as much of an incentive as it could be.

During the negotiations of Additional Protocol II to the Geneva Conventions, a proposal was made to grant non-State armed groups a POB. Although such a proposal probably had loose humanitarian underpinnings (but may have also been linked to the need to achieve peace in NIACs, as is the requirement for amnesties), the arguments against it failed to take its humanitarian rationale seriously. At no time then or since has this been identified as a particularly problematic feature of Additional Protocol II and its ability to deliver on its humanitarian promise. Rather, the automaticity of the POB with regard to State troops and its exclusion when it comes to non-State armed groups reflects a much more conventional bias in favour of States' monopoly on the legitimate use of force.

The claim that the privilege of belligerency is a key to better compliance is less clear-cut than is often thought

Even though it was not specifically thought of as incentivizing compliance, the POB might have that effect more or less fortuitously. General wisdom, for example, seems to suggest that not granting the POB provides the parties to an armed conflict with a pretext for not respecting the laws of war. How this may disincentivize belligerents is never clearly spelled out in the literature. Before cautioning against any facile mechanistic assumptions in that area, it may be helpful to briefly sum up what seems to be the dominant consensus. The reasoning goes as follows: warriors involved in IAC can at least – for a range of reasons linked to the history of war and statehood – get away with killing other soldiers: in fact, theoretically, this is primarily what they are asked to do. They do not risk automatic prosecution and infamy for doing what is after all the core of their activity. There is thus a sort of rough quid pro quo involved: we will pursue you if you kill non-combatants, but (regardless of whether you are on the good side of the jus ad bellum or if you respect the jus in bello) we will never bother you for targeting combatants. If that realistic and tragic agreement were to be compromised, according to the dominant view, then “all bets would be off the table”. No longer capable of being even partly legally redeemable, combatants would descend into a further spiral of violence.

A similar argument is often made in NIAC. In this case, the idea is that failure to dignify non-State armed groups as combatants means that they are not socialized into international law – that they essentially get all of the obligations and none of the privileges of being in war. We ask them to honour humanitarian obligations but, at the same time, we tend to treat them as criminals if they are caught by the State. This is unfair and unproductive. There is a certain intuitive appeal to the idea of a normative system's success depending on a level of fairness towards its potential subjects. As a result, one strong argument for granting a POB to all in any armed conflict is symmetry. The idea is that the laws of war cannot function without symmetrical applicability (consider, for example, the traditionally rigid separation between the jus ad bellum and the jus in bello) and that this would be mortally wounded by the idea that only one side would have a POB – in other words, that the other side would objectively be under a much higher burden of obligation. The destruction of the equality of parties will, in a sense, shatter their sense of obligation in war. In particular, non-State armed groups who think that State forces are benefiting from the POB whilst they are being denied it might give up entirely on respecting their humanitarian obligations.

There are several issues with this theory as an explanation of what is problematic with the regulation of NIAC. First, it seems too dramatic; clearly, it is typically not thought to be the case when it comes to NIAC, where non-State armed groups have long been incited to comply with the laws of war with no prospect of recognition of a POB. In fact, there are many reasons to question whether symmetry of obligations and the equality of parties are as central to compliance with the laws of war as is ordinarily assumed, even in IAC. The conceptual emphasis of the laws of war themselves over the last decades has been on the non-synallagmatic character of humanitarian obligations as exemplified in the desuetude of such practices as reprisals. It is bizarre, in a context where the fundamental element of non-reciprocity at the level of principles is consistently emphasized, to seek to make symmetry the central plank of a policy proposal focused on implementation. If anything, non-State armed groups' desire to comply with the laws of war is less likely to be incited by the blanket grant of the POB than by State troops themselves complying with the laws of war in the conduct of hostilities and in relation to protected persons (including rebels when they qualify).

Second, whether a non-State armed group in a NIAC decides to engage in a policy of war crimes or not seems to depend on a whole series of complex psychological, social and political factors (values, goals, publicity, reputation, “lawfare”, etc). In other words, if a non-State armed group is determined to kill civilians and is taking the risk, for example, of vulnerability to international prosecutions, then I find eminently implausible the notion that it would not do so on the basis that it might also be prosecuted for the (presumably relatively more benign) crime of attacking combatants. Inciting actors not to commit a particular genus of crimes by giving them immunity to commit another genus hardly seems a straightforward gamble. Certainly, the idea that non-State armed groups who have not received the POB would “slam the door on international law” and consider that, having nothing to lose since they will also be prosecuted for killing combatants, they might as well violate the rights of non-combatants sounds improbable, at least without much more empirical evidence.Footnote ⁸⁸ Simply because one risks being prosecuted for attacking State forces does not mean that one will then necessarily and wantonly also, out of normative spite as it were, target civilians.Footnote ⁸⁹

It may be that non-State armed groups feel they are outsiders when it comes to the international game, but aside from the fact that some may relish this status, it might precisely bring them to seek to compete with the humanitarian performance of States. There are many examples of parties to an armed conflict honouring their obligations even when faced with an enemy that does not at all, or even vis-à-vis an enemy that is not bound by the same obligations. A situation in which the POB is not by and large on the table has not prevented many notable and novel initiatives to implicate non-State armed groups in the implementation of the laws of war, to which they have shown themselves to be sensitive. If we allow for the possibility that asymmetrical compliance is actually sustainable, it is only a small step to conceiving of asymmetrical obligations as also being sustainable.Footnote ⁹⁰ In effect, IACs have already become more like NIACs in that they have often proceeded along asymmetrical normative lines. At any rate, one rarely hears from non-State armed groups that the POB would significantly enhance their humanitarian performance; one is more likely to hear that it would be a deserved recognition of the legitimacy of their cause.

For the privilege of belligerency to have a real impact, it should be conditional

To be clear, my concern is not with the possibility that the POB might be extended under some circumstances, but that it might be granted broadly and ex ante to non-State actors in all NIACs. Even though Kreß does not dwell on this point, central to his suggestion seems to be an almost automatic granting of the POB. This is implicit in the fact that he takes his cue from IAC, where the POB is automatically granted when it comes to State armed forces. At any rate, if there is to be an element of conditionality, it is not spelled out with any precision.

Given the doubts I have voiced that the granting of a POB is in itself rich with humanitarian promise, how could it be used as a lever to actually increase compliance? Simply because it has not been very effectively used in this way does not, assuredly, mean that it could not be reinvented as a more potent element of the policy mix that goes into obtaining compliance with the laws of war. Put simply, the problem is that the grant of the POB is either too narrow or too broad. On the one hand, it is currently too narrow because it is not granted to any non-State armed group, notwithstanding that there may be good reason to grant it to some. This means that effectively, a non-State armed group which wages a campaign that reaches a certain critical intensity against the State “creates” an armed conflict, which then empowers the State to deal with it through relatively more ruthless methods than would be available under human rights law. Simultaneously, however, the non-State armed group does not gain much, except a range of international humanitarian obligations that constrain its actions without any obvious added benefit. “Virtuous” rebel movements may have their own reasons for being so, but they are not particularly rewarded compared to those who make a profession of violating the laws of war. On the other hand, the granting of the POB it is too broad because it is implicitly always granted to State armed forces, whether in IAC or NIAC. As for non-State armed groups, according to Kreß's proposal the POB should be granted broadly to rebel movements regardless of whether they are systematic violators of the laws of war or responsible actors.

The incentive implicit in a broad grant of the POB in NIAC is one that would encourage non-State armed groups (and perhaps even the State) to bring violence to a certain critical level of intensity. By the time an armed conflict is recognized to exist, their actions would cease to be criminal and would be covered by the POB. If that were all there was to this particular calculation, it could create a perverse incentive. Violent actors would be encouraged to go beyond skirmishes and the odd armed attack to a full-blown military campaign. The more systematic violence would be legal, whereas the most benign would remain strictly sanctioned under domestic law. It is unclear in this context who the laws of war would reward, except perhaps that they would bow to the power of certain actors to create mischief on a “humanitarian” scale. This is, needless to say, not very appealing. If such a radical shift in the granting of the POB to non-State armed actors is to occur (and this is what is at stake), it should depend on something other than the fortune of arms.

Remedying one problem by creating another (i.e., from over-narrowness to over-breadth in granting the POB) is not the right solution. The problem, I believe, lies with the idea that the POB is in itself a cause of greater humanitarian compliance. I suggest that we must more clearly inverse the assumption and consider that the POB should be a consequence of a certain high level of compliance. This implies that, rather than broadly granting a POB to non-State armed groups in armed conflicts on a priori humanitarian grounds, we should make the grant of the privilege conditional upon a recognized humanitarian performance. The privilege would be something that is earned rather than simply granted ab initio once a non-State armed group has effectively created the fait accompli of a NIAC. Perhaps more crucially, it is something that could be lost (in contrast to State armed forces, who retain the POB regardless of compliance with IHL) if it became evident that a group which had once been granted the privilege had significantly failed to honour the obligations that come with it.

Now, it is true that Additional Protocol II already defines its material field of application as those conflicts confronting a Contracting Party's armed forces “and dissident armed forces or other organized armed groups which … exercise such control over a part of its territory as to enable them to … implement this Protocol”.Footnote ⁹¹ In other words, the ability to implement the Additional Protocol is a condition of there being a conflict in the first place. Yet this ability is not the same thing as actual performance and points merely towards a broad capability approach. It does not give us a strong indication of sustained commitment to the laws of war. Here, a conditional POB could be envisaged as a considerable leverage in ensuring respect for IHL. An appropriate formula would have to be found to describe the expected level of compliance with the laws of war of rebels, but the threshold could be set quite high and go beyond capability to include elements of a sustained record. In such a situation, the State would be (relatively) empowered by the emergence of an armed conflict, but so would the responsible non-State armed group. For some of those who have decided to engage in an armed conflict, holding out a promise of a POB could act as a significant incentive to bring their game up to international standards.

Concretely, it may be that rather than granting a POB ex ante or at least in the midst of a conflict, the better policy option would be to grant amnesties retrospectively to those groups that have broadly satisfied humanitarian exigencies over the course of a conflict.Footnote ⁹² This would be a specifically humanitarian-compliance oriented use of amnesties, whereas their grant is traditionally invoked in the context of peacebuilding and transitional justice.Footnote ⁹³ This would mean that the incentive would continue to operate on a consistent basis throughout an armed conflict and would thus consistently put rebels “on notice” that their acts may have consequences, and that a certain belligerent respectability is what is ultimately at stake. Even though a State might be reluctant to give amnesty to individuals for taking up arms against it, at least it might be forced to grudgingly recognize a particular rebel group as having behaved responsibly from a combatancy point of view. And the idea of an amnesty based on humanitarian performance would at least preserve the idea that for members of non-State armed groups to kill State combatants is excusable under certain circumstances, yet not strictly permissible.