From ISIS to ICISS: A critical return to the Responsibility to Protect report

Introduction

Since the forces of the Islamic State of Iraq and the Levant (ISIS) captured the city of Derna in eastern Libya in November 2014, the parallels between Iraq and Libya have become increasingly irresistible. In both cases, we had the overthrow of a dictator by Western military intervention leading directly to civil war, state collapse, massive refugee flows and both countries disintegrating into their constituent ex-Ottoman provinces. While advocates of the Responsibility to Protect doctrine (R2P) have vociferously argued that the invasion of Iraq did not authentically embody the principles of R2P (e.g. Thakur, 2005), this is harder to argue in the case of Libya. The United Nations (UN) resolution 1973 that authorised the 2011 NATO campaign against Colonel Gaddafi’s regime explicitly referenced the failure of the Libyan state to protect its people as justification for NATO powers to protect Libyan civilians menaced by Gaddafi’s forces.

For many advocates of R2P, the Libya intervention redeemed hopes that R2P had not been compromised by the fact that humanitarian arguments were at least partially used to justify the 2003 invasion of Iraq (Moses, 2013). Yet today Libya is already a mini-Iraq: the outcome of military intervention has been remarkably similar in both countries. Despite this, resolution 1973 is still widely regarded as embodying normative progress in the development of R2P doctrine. Given that so much scholarly discussion of R2P today has become absorbed by tracing and even strengthening the various branches of the doctrine as it has developed over the last ten years, it is also worth examining the doctrine as a whole – root and branch – in light of the post-intervention disaster that is Libya today.

The ICISS report remains the single most succinct and elegant theoretical tap-root of R2P doctrine as a whole. The notion of international responsibility has become increasingly present in the rhetoric surrounding R2P. The Secretary-General’s 2009 report on ‘Implementing the Responsibility to Protect’ returned to themes found in the ICISS report (Ki-moon, 2009). These themes of responsibility have been reinforced by scholars insisting that members of the international community have a responsibility at least to try to implement R2P (e.g. Bellamy, 2015; Welsh, 2013). Even one of the most innovative twists to R2P doctrine – the proposed Brazilian ‘Responsibility while Protecting’ idea – has taken inspiration from the ICISS report. Indeed, the 2011 Libyan intervention even conformed to the least likely prescription of the whole 2001 document – the plea for permanent members of the Security Council to refrain from exercising their veto when vital national interests were not at stake (ICISS, 2001: 51). In the case of Libya, both China and Russia refrained from exercising their veto powers with which they could have blocked UN authorisation for the NATO air campaign. Thus it is worth revisiting the core doctrine of R2P as expressed in the 2001 report of the Canadian-sponsored International Commission on Intervention and State Sovereignty (henceforward ICISS). What was it all about?

In this paper I provide a normative critique of the ICISS report and argue that far from embodying hopes for progress, the reframing of state power and political practice that lies at the core of R2P perversely renders the exercise of power less accountable, and therefore less responsible. In other words, I find ICISS doctrine wanting on its own terms: R2P ultimately dilutes the political responsibilities that states owe their peoples. To that extent, I suggest that the post-intervention crisis in Libya is consistent with what we might expect from the doctrine even in its earliest, classical forms. The report proposes paternalistic principles that are ultimately at odds with representative government and popular sovereignty, and this can only be a bad thing for anyone interested in political accountability and the responsible exercise of political power.

Thesis 1: R2P inverts the burden of justification for intervention

The ICISS report seeks to check interventionism by acknowledging its destructive effects and admitting the political primacy of the state. But it also shifts the onus of justification for intervention from the intervening state to the state being intervened in. ²

Historically, the norm of non-intervention has been honoured in the breach as much as the observance. On the face of it, the most honest appraisal of international politics would be to acknowledge intervention as a timeless fact of international life. Accommodating intervention may appear a more straightforward and honest means of restraining it, rather than rigidly insisting on abstract rights that offer no substantive protection against foreign powers’ military might. It could even be argued that such clear-sighted pragmatism helps to ensure that our political and legal norms do not drift too far from the facts of political life. But integrating an exceptional circumstance into an extant normative order has a recursive effect on the normative structure in question, changing the very coordinates within which ‘political facts’ themselves are established. Several consequences flow from this.

First, once the formal possibility of legitimate intervention is conceded, not only do standards for intervention become more permissive, but the onus of justification shifts. Instead of an interloping state having to defend its violation of the non-intervention principle, it is equally (if not more) incumbent on the state being intervened in to defend why it is entitled to be free from external interference: ‘States will be forced to substantiate the claim that they are upholding certain norms and standards of governance vis-à-vis their own population when they invoke sovereignty as a defence against external intervention’ (Stahn, 2007: 199). The analogy in the domestic sphere is encroachment on those so-called ‘negative’ rights (such as civil liberties or the presumption of innocence) designed to preserve a sphere of individual autonomy separate from state power.

It could be argued that the ‘domestic analogy’ is inappropriate here (Suganami, 1986), as the burden of justification is being shifted onto a state rather than an individual. The problem remains however that it is not only the state (in the narrow sense of the state apparatus) that would be limited, but also the people that the state represents. Once states must also justify their political authority to other states, this means they no longer have rights in the international sphere that derive solely by virtue of their representing a given people. This can only degrade the rights of the people in question, whose political claims are effectively made dependent on international license. In other words, to suppress the sovereign supremacy of any state is to suppress its people’s right to unfettered representative government. The rights of a state cannot be called into question by other states without also calling into question the rights of the people within that state. This is one key point that defenders of intervention consistently fail to address. We shall return to this theme below.

Lest it be thought that these concerns apply only to peripheral states, it is worth noting that changing the burden of justification for intervention has anti-democratic consequences for the intervening state as well. Michael W. Doyle draws attention to ‘indirect’ reasons for avoiding intervention: ‘Interventions foster militarism and expend resources [within intervening states] needed for other national and international goals’ (Doyle, 2003: 217). What is more, normalising intervention at the international level provides states with yet another weapon with which to squash domestic resistance to military adventures. Legal theorists and constitutional experts have argued that US president Barack Obama exceeded executive prerogatives and flouted the US constitution when he claimed that UN authorisation for US air operations in Libya obviated the need for congressional approval (e.g. Fisher, 2012). Following the British parliament’s vote on 14 September 2014 to intervene militarily in Iraq against ISIS, British prime minister David Cameron made clear that he was willing to extend the air campaign into Syria without seeking parliamentary approval again if there was a ‘humanitarian catastrophe’ (Swinford and Graham, 2014).

Thesis 2: the responsibility to protect usurps self-determination in favour of paternalism

A keystone of the ICISS report is the claim that states are duty-bound to protect their peoples. The principle that states bear responsibilities for their people seems as banal as to verge on tautology. After all, it would be ridiculous to argue for irresponsible sovereigns. Does the idea that states bear responsibilities for their peoples mean anything beyond the platitudes that accompany international pronunciamentos on human rights? I argue that the ICISS report is in fact giving the (otherwise banal) idea of state duties a different emphasis – one which privileges the state at the expense of popular sovereignty.

As argued above, the possibility for coercive intervention raised by the ICISS report is in fact a secondary effect of the way in which ICISS conceives of sovereignty. By tying the provision of internal security to the state’s international obligations, the ICISS report makes the provision of security a component of state legitimacy. Hence, the rights traditionally associated with sovereignty are rendered contingent on satisfying international requirements. According to Amitai Etzioni this gouges open ‘a gaping hole’ in the very ‘foundation of democratic theory’:

The given nation’s people would … have at best limited representation: an international forum could rule that their government was acting irresponsibly without having to give the affected citizens the opportunity to affect the forum’s judgements and positions. (Etzioni, 2006: 72) ³

If states are required to uphold human protection for their citizens by virtue of their international commitments and (ultimate threat of coercion by other powers), this can only mean states are in the position of holding responsibilities for their people rather than to their people. In a practical sense, forcing states to orient themselves not only around their people’s demands but also around the gaze of other states has the effect of making a state’s citizens less central to its political choices. Perversely, being forced to take greater account of the international community could weaken a state’s commitment to its own people. For a doctrine that invites us to regard state power with scepticism, it is peculiar on the face of it that it also offers such a clear opportunity for states to slither out of the grasp of their own people.

Thesis 3: victims’ rights provide license for the exercise of power

We therefore arrive at the strong claim made for the doctrine – that there is a ‘duty to protect communities from mass killing, women from systematic rape and children from starvation’ (ICISS, 2001a: 17). Whereas previous discussions of intervention were formulated in terms of the ‘right’ of one state to interfere in another for the sake of human rights, the ICISS report bracketed the ‘rights’ of states in favour of the rights of victims.

ICISS commissioner Ramesh Thakur enjoins us to leave behind the ‘doctrine of national sovereignty in its absolute and unqualified form, which gave rulers protection against attack from without while engaged within in the most brutal assault on their own citizens’ (in Weiss, 2007: 24). It seems straightforward to interpret the claim to ‘sovereignty’ in this way, and by extension to welcome its demise. If sovereignty is that power which has no higher constituted power above it, then supreme power does indeed mean doing as one pleases. According to the ICISS report, even the ‘strongest supporters’ of state sovereignty do not ‘claim … the unlimited power of a state to do what it wants to its own people’ (ICISS, 2001a: 8).

But this is a tendentious reading of sovereign supremacy that writes off centuries of political history (Cohen, 2004: 12). The historical process through which popular sovereignty supplanted absolutist monarchy and empire, via the liberal and national revolutions of the last three centuries, bears out the theoretical point that a claim to absolute power is a precondition for the exercise of popular will. As Martin Loughlin makes clear, absolute power should not be confused with the tyranny of the majority: ‘general will, although absolute, has nothing in common with the exercise of an arbitrary power’ (Loughlin, 2003: 73). By articulating a framework of political universalism that is supreme but not reducible to majoritarianism, the concept of the general will offers the political possibility of unifying the interests of minorities with those of majorities (Rousseau, 1947). The category of general will shows that the conceptual resources exist in political theory that allow us to reconcile the interests of majorities and minorities without sacrificing political universalism or necessitating extra-sovereign checks and balances. The means of preserving the sovereign rights of a people is through the citizenry exercising their collective will through the unfettered supremacy of their representative, the state. Thus, the right of the sovereign to act ‘with impunity’ is testimony not to tyranny but to the central importance of the will of the people in modern politics. This does not mean sovereignty can only exist as a hyper-centralised monolith – on the contrary, sovereignty can be sustained even when articulated across a range of internal institutions and centres of political power (see further Loughlin, 2003: passim, and Loughlin, 2000). The US Civil War, for example, established the sovereignty of a federalised (i.e. decentralised) Union over centrifugal (confederal) secessionist states. But sovereign impunity vis-à-vis all other states in fact helps preserve the role of the people as the sole arbiters of their collective (sovereign) will within their state. Accepting that sovereignty is not a synonym for arbitrary tyranny does not, to be sure, provide us with any easy or pre-formatted answers to the dilemmas posed by intervention. It does however draw attention to the dangers that might flow from confusing claims to supreme authority with tyranny, and alerts us to the fact the R2P has significant ramifications for structures of political representation.

Nor does this view of sovereignty commit us to jettisoning the basic constructivist insight that sovereignty is a flexible institution that evolves and adapts over time, incorporating different standards and reflecting wider international norms: ‘sovereignty is itself a malleable principle that has been constructed in different ways over time and can be reconstructed in the future’ (Bellamy, 2003: 334). While it is obviously true that political institutions change and adapt over time, in the context of R2P the problem with this line of argument is its teleological character, which ends by quietly dropping its own insight.

To consistently follow through on the insight about the changeable character of sovereignty, then we must recognise that what is malleable about sovereignty is that it is an institutional expression of political will, and political will is by its nature self-defining. What is seen as a fixed barrier to sovereignty in one period may be dissolved in the next – witness the widespread nationalisation of banks during the economic crisis of 2008–2009. Michael Oakeshott characterises this as one of the defining features of sovereignty, namely that it retains ‘the authority and procedures to emancipate itself continuously from its legal past’ there being ‘no law so ancient and so entrenched that it could not be amended or repealed’ (in Loughlin, 2003: 59). ‘Sovereignty as responsibility’ is often tendentiously presented as the crowning achievement of centuries of political evolution. Sovereignty is malleable – but only in the past: today, the only option for recalcitrant states is to fall into line with ‘modern’ ideas, i.e. R2P doctrine. As a result, the sovereignty of today is restricted to what is (in actuality) a historically arbitrary standard of global human rights norms and technical measures of institutionalised ‘capacity’.

To fix sovereignty to a historically arbitrary standard is to limit the scope of self-determination. The sovereign, by virtue of its sovereignty, must always retain the right to renounce any fetter and dissolve any compact by which it is bound. If we are to retain the valid constructivist insight about the historically changeable character of sovereignty put forward by the partisans of R2P, then we must also accept that these standards and norms will be overthrown in future – and this too, will be entirely consistent with sovereignty as political practice.

This gives us another perspective on the value of non-intervention. Louis Henkin attacks the caricatured portrayal of non-intervention that depicts it as if it exists solely to enable states to perpetrate genocide (Henkin, 1999: 825–825). Rather the ‘law against unilateral humanitarian intervention may reflect, above all, the moral–political conclusion that no individual state can be trusted with authority to judge and determine wisely’ (Henkin, 1999: 825; emphasis added). Henkin reminds us that intervention inescapably means privileging the power of some states over others, whatever the talk of victims’ rights. In other words, the rule of non-intervention could plausibly be read as entrenching suspicion of state power in general, even if this comes at the expense of privileging the rights of an incumbent state against external claimants, however benevolent. ⁴ Even oppressed peoples have some degree of political relations with their state. They have no institutional relations (and therefore no basis) on which to hold to account external powers. The norm of non-intervention preserves the link between people and state. The democratic slogan ‘Power to the people’ is meaningless in the absence of sovereignty.

What is telling about the reformulation of the debate in terms of victims’ rights is that the condition for enacting R2P is the reduction of its recipients to passivity defined by suffering. This is a direct result of the ‘ideology of victimisation’ that separates questions of oppression and conflict from questions of institutions, representation and self-determination. The net effect is described by Slavoj Žižek (1999) in relation to NATO’s 1999 Kosovo campaign:

it’s perfectly fine to help the helpless [Kosovo] Albanians against the Serbian monsters, but under no circumstances must they be permitted to throw off this helplessness, to get a hold on themselves as a sovereign and independent political subject – a subject that doesn’t need the kindly shelter of NATO’s ‘protectorate.’ No, they have to stay victims … The other will stay protected so long as it remains the victim. ⁵

It is not difficult therefore to see how the R2P proposition can be inverted: by their very nature the weak are less capable of holding their self-appointed benefactors to account. Victims do not take their political destiny into their own hands, but instead exist to exemplify the virtues of their rescuers.

Perhaps this focus on victims’ rights helps us understand why R2P has won such widespread support, as exemplified by the Outcome Document from the Millennium World Summit of 2005. The image offered in the ICISS report is that of benevolent states caring for their populations’ needs and security, crowned by a great power directorate in the Security Council. Small wonder such a vision won support from China (Bellamy, 2009: 87). China’s acquiescence to R2P is more inadvertently revealing than any R2P supporters care to admit. It only needs a moment’s reflection to realise how the R2P vision of world order may appeal to an authoritarian government – a world where states care for their peoples more than they have to answer to them, a world where the role of the major powers is enshrined as global policeman hovering above the law that applies to lesser states. On this much at least, China, Russia and the West seem to agree.

Thesis 4: the doctrine of the Responsibility to Protect imposes imperfect duties

Although the ICISS report clearly says that states are responsible for people on their territory in the first case, the centrality of this duty of care is not an absolute question but one of degree. As states may be called upon to rescue humanity anywhere in the world, states’ duties diffuse across the globe. But a responsibility that belongs to everyone also belongs to no-one: it is an imperfect duty – in the words of Michael Walzer, ‘a duty that doesn’t belong to any particular agent’ (Walzer, 2000: xiii). ⁶

ICISS tries to disguise this bald fact by cataloguing the various agencies that jointly bear the burden of R2P (states, groups of states, regional and international organisations, the UN, non-governmental organisations, etc.). In grafting R2P on to as many different organisations as possible ICISS presumably seeks to brace the doctrine with as many buttresses as they can. But diffusing R2P as widely as possible cannot compensate for the fact that R2P does not obligate any specific agent to act in an emergency. There is no escaping the fact that in any crisis a state must assess the situation in light of its own aims and interests (Walzer, 2000: xiv).

It is the imperfect character of humanitarian duties which helps explain how the doctrine can win such widespread support without having to manifest any practical examples of actual intervention (Arbour, 2008: 450). The potential appeal to states, particularly powerful ones, is clear: those states that have both the proclivity and ability to intervene are offered greater possibilities to do so, without having to limit themselves to any prior commitments. This was at least tacitly acknowledged by ICISS as the commissioners recognised the need to off-set this potentiality by erecting ‘precautionary’ and ‘threshold’ criteria to regulate interventions.

Putting to one side problems with the ‘precautionary criteria’, we can untangle the problems with ‘imperfect duties’ by focusing on ‘threshold criteria’. ‘Threshold’ criteria restrict intervention to ‘extreme cases only’ that ‘genuinely “shock the conscience of mankind”’ (ICISS, 2001a: 31). These are specified as ranging from ‘overwhelming’ natural catastrophes to genocide (ICISS, 2001a: 33). Regardless of how these crises are defined, even the most stringent criteria cannot eliminate the need for an outside state to make a politically-driven decision as to whether or not a crisis merits intervention. In a word, threshold conditions will always be politically understood – ‘subject to interpretation and manipulation’ (Welsh, 2002: 514).

Alex J. Bellamy understands this ‘problem of indeterminacy’ as a question of controlling the application of the norm: ‘there is no guarantee that when confronting a humanitarian emergency, states would agree that a just cause threshold has been crossed, or the precautionary principles satisfied’ (Bellamy, 2006: 148). But the problem of securing multilateral agreement in such a case is only a secondary effect of the prior problem – namely, that it is not clear who is entitled to control the application of the norm (Welsh, 2004: 66–67). By claiming that it is only in exceptional circumstances that states forfeit their right to non-intervention, the ICISS report has at once suggested some vague moral limit on states’ rights, while simultaneously affirming their place as key political actors. As a result, ICISS has obscured the relationship between the exercise of rights and political agency. As argued above, the range of actors invoked by ICISS as carrying the burden of R2P testifies to the fact that the duty cannot be clearly attributed to any single agent. To that extent, the enforcement of R2P would be as arbitrary as the humanitarian interventions mounted in the 1990s: the ‘international community’ may intervene to halt a government offensive against rebels (Libya) while standing aloof as an entire country is convulsed by bloodshed (Syria). The difference between what we may call the 1990s’ system of intervention and that of the ICISS report consists partly in that R2P strengthens the idea that intervention exists as a ‘discretionary entitlement’ (Berman, 2007: 161), even if this entitlement is no longer claimed in the language of states’ rights but of victims’ rights.

It is important to be clear about the negative consequences flowing from the imperfect duties embodied in the ICISS report. Many have argued that the norms around sovereignty neither constrain nor drive intervention: what matters is the political will to intervene, and this is more or less independent of international norms (Bellamy, 2006: 33; Weiss, 2007: 47). ⁷ While this is undoubtedly true, embedding intervention as a ‘discretionary entitlement’ in the system of international norms has perverse effects even if it does not generate military expeditions ex nihilo. It contributes to what Jean Cohen calls the ‘deformalization of international law’ whereby clear procedures and standards become increasingly ‘indeterminate’: powerful states can intervene when and as they please (Cohen, 2006: 496).

This also exposes how little is actually on offer to the people suffering extreme duress. Beneath the gushing talk of suffering humanity, all the ICISS report really offers to ‘humanity’ is a vague assurance that mighty foreign powers may intervene to some extent, to alleviate suffering to some degree, if it happens to be convenient for them to do so. Speaking of the Darfur crisis in 2005, Alex J. Bellamy and Paul D. Williams lament the ‘apparent contradiction between governments’ use of “sovereignty as responsibility” language to enhance their own humanitarian credentials and their unwillingness to take “responsibility-based” action’ (Bellamy and Williams, 2005: 29). But this ‘contradiction’ disappears once it is understood that discretionary entitlements by their very nature enhance power and moral authority without responsibility. The only thing guaranteed here is, as Aidan Hehir puts it, ‘the permanence of inconsistency’ (Hehir, 2013). Thus the ICISS report not only fails to turn imperfect duties into perfect ones, but in its struggle to do so enhances the scope for the arbitrary exercise of international power.

Beyond the problem of whether the doctrine may generate perverse incentives, there is also the question of whether R2P may endanger non-combatants by encouraging them to remain in war-zones (Luttwak, 1999: 38). This could be either under the shelter of inevitably insufficient protection activities that UN peacekeepers are now expected to provide (UN, 2008: 22), or in the expectation of more robust ‘human protection’ operations that are more likely to be promised than delivered. Indeed, drawing on the insights provided by Mark Duffield (2007) and Jacob Stevens (2006) we could radicalise this point. By offering people the hope of military intervention as well as material and political support to linger in war- and disaster-zones, does R2P become complicit in and productive of a highly restrictive global migration regime that imprisons people in conflict, inhibits the free movement of peoples and helps secure the borders of wealthy states?

Building on this point, and in light of the on-going Libyan refugee crisis, some hard questions have to be asked about how R2P doctrine ranks humanitarian priorities and responses. Given the ostensible spread of humanitarianism in the dense multiplication and thickening of R2P commitments since 2005, why is it that military intervention still seems to be so much politically easier than, say relaxing border controls, or devoting resources to providing for and hosting refugees as Jordan and Turkey have done (and with significantly less resources than European states)? How would the proposed European Union intervention in Libya to interdict people-smuggling sit within the development of R2P doctrine, given that it is clearly a response to the prior application of R2P doctrine? To be sure, some advocates of R2P have suggested that intervening powers have a responsibility to provide for refugees fleeing the country that was intervened in (although the fact that people are still fleeing in the first place is not taken as a criticism of R2P; see for example Ralph and Souter, 2015). But even here, absorbing refugees is seen as one option on a menu that extends to military intervention, rather than being presented as the prime and most authentically humanitarian response to conflict.

Thesis 5: the responsibility to rebuild diversifies post-conflict responsibilities

The responsibility to rebuild is a crucial aspect of ICISS doctrine. The ICISS report seeks to ensure that attention devoted to ending human suffering is not episodic, while also emphasising the need to restore self-rule as rapidly as possible (ICISS, 2001a: 45). Perversely, by making the conditions for intervention more permissive, the dilution of responsibilities for post-conflict governance and reconstruction has been sanctified by the ICISS.

As Michael Ignatieff observes, in the new peacebuilding operations direct political control emanating from a single metropole no longer exists: ‘In the new humanitarian empire, power is exercised as a condominium, with Washington in the lead, and London, Paris, Berlin and Tokyo following reluctantly behind’ (Ignatieff, 2003: 17). Michael W. Doyle and Nicholas Sambanis argue that the institutions and norms of the UN ‘make the quasi-colonial presence that a multidimensional peace operation entails not only tolerable but effective’ (Doyle and Sambanis, 2006: 318). If multilateralism makes it more difficult to individually profit from peacebuilding operations, it also makes it easier to evade responsibility. If the classic critique of the League of Nations’ mandates system was that it provided an international fig-leaf for the extension of the Franco-British colonial empires, at least that system had the benefit of making clear which state was actually in charge of those territories.

Today, by contrast if intervention can be legitimately pursued for non-national (i.e. humanitarian) gains, it equally makes sense to pursue so-called transitional administration in a cosmopolitan fashion. Further conceptualisation is needed in this field before we can begin to more sharply define the questions concerning the locus and character of political and legal responsibility in these situations. By way of contributing to this task, here I will suggest some avenues for further research by identifying some key problems.

Part of the rationale for sovereignty is the establishment of a clear locus of political authority through such means as endowing the state with legal personality (Loughlin, 2003: 56). By contrast, Roland Paris draws attention to the ‘decentralized’ character of ‘international governance structures’ that lack ‘a single corporate identity’ in peacebuilding. Such institutions ‘lack clear lines of accountability, meaning that even if we … disapproved of [their] actions …, there is no single mechanism through which we could demand a change of peacebuilding policy. Nor is there a single actor whom we could collectively hold responsible for the outcome of particular operations’ (Paris, 2000: 43). Paris’ points about political accountability are well made and well taken. But there are other problems resulting from the networked character of these ‘international governance structures’.

The flipside of a lack of accountability is political coercion. As described by David Singh Grewal, under networked forms of power ‘aggregate outcomes emerge not from an act of collective decision-making, but through the accumulation of decentralized, individual decisions that, taken together, nonetheless conduce to a circumstance that affects the entire group’ (Grewal, 2008: 9). Second, when particular outcomes can be clearly attributed to particular agents, not only is there a basis for accountability, but social order is perceived as the result of political decisions and individuals. When governance structures are diffuse and incoherent, political order and social life is correspondingly more opaque and intractable. It will be less clear where political initiatives originate, or which political claims have genuine precedence: those of the local government or those of internationally-appointed viceroys?

The five theses above offer a critique of the ICISS model for regulating interventionist practices in international relations. The various means proposed to restrict the practice of intervention by talking up the duties of states, outlining various pre- and post-conflict responsibilities and so on – are offered up as a means of sheltering the state from predation by intervening powers. In fact, such a move can only be predicated on a greater acceptance of intervention in the first place. This gives the lie to the conciliatory ethos of ICISS. I have also argued that the consequences of the greater acceptance of interventionism is disturbing not only because it resurrects the spectre of empire, but also, and perhaps more importantly, it speaks to the restructuring of political authority in a more paternalistic direction.

From ICISS to ISIS

What light might this critique of ICISS doctrine shed on the aftermath of the intervention in Libya, locked in simmering civil conflict over the last four years since the NATO campaign and increasingly a haven for jihadi insurgency? At least three possible perspectives suggest themselves.

First, in light of the refugee crisis around Libya (and even more so, Syria) we can build on theses four and five above to say that the fact that the doctrine makes military intervention easier also suggests that it has contributed to the scrambling of humanitarian priorities. Relaxing border controls would seem a more obvious and immediately humane response to conflict than intervening to stop the fighting – not least because opening borders does not involve killing even more people. Yet asylum for refugees has found few advocates among humanitarian defenders of military intervention, and has proved to be politically more difficult to implement than war – and this despite the fact that recent wars in Iraq and Libya have turned out so badly. Second, and linking back to the first thesis offered above, by giving greater international legitimacy to military intervention R2P has implications for democratic control over executive power, particularly with regards to the use of force and the ‘imperial presidency’ in the US. Third, as per the fifth and final thesis above, the failure of peacebuilding and rapid disintegration of the Libyan state may reflect how R2P diffuses post-conflict responsibilities.

Needless to say, further research would be necessary to elaborate these links. Nonetheless, the criticisms of R2P offered above give us plenty of reasons to suspect that a paternalistic doctrine based on flexible, discretionary standards may boost executive power and lead to decentralised and ineffective post-conflict reconstruction. At the very least, testing the theories of R2P for their internal consistency and against political outcomes offers a more stringent test of the doctrine than getting lost in the ever-thickening brambles of global R2P norms, whose rapid growth across UN resolutions and reports always provides easy evidence for international normative progress, regardless of what is actually happening in the countries left behind after intervention.