Melbourne Journal of International Law
LEGAL-POLICY CONSIDERATIONS AND CONFLICT CHARACTERISATION AT THE THRESHOLD BETWEEN LAW ENFORCEMENT AND NON-INTERNATIONAL ARMED CONFLICT
When characterising a conflict situation as an international armed conflict, states and other analysts traditionally consider the ‘facts on the ground’. When determining whether a situation is one of civil disturbance and riot, or has risen to the level of a non-international armed conflict (‘NIAC’), there is much greater latitude for legal-policy considerations to influence, and indeed direct, the characterisation decision. This article explores three aspects of legal-policy concern for states dealing with conflict characterisation at this lowest law of armed conflict (‘LOAC’) threshold between less-than-NIAC law enforcement and NIAC: a general outline of three elements of legal-policy discretion that are clearly assumed and inherent within LOAC; legal defensibility, a discourse that is fundamentally governed by the tension between applicable law and policy objectives; and utility, a concern that focuses upon the balance to be struck between the legal argument employed to justify conflict characterisation and the capacity of the state to retain some degree of context control.
When faced with violence rising above the level of localised or sporadic, the black letter law answer as to whether a situation is one of armed conflict or not is that it is simply a question of fact. The rationale behind this seemingly clear and simple principle is well expressed in Jean S Pictet’s most humanitarian explanation: ‘A wounded soldier is not more deserving, or less deserving, of medical treatment according to whether his Government does, or does not, recognize the existence of a state of war’. What is occasionally overlooked, however, is that Pictet’s focus was upon art 2 of Geneva Convention I, and was thus concerned with international armed conflict (‘IAC’), where the ‘facts’ that would generally speak as to characterisation are overwhelmingly overt and highly visible — uniformed militaries of two or more states engaged in hostilities. In this sense the focus on ‘facts’ is fundamentally a call to use an objective test with a narrow range of fairly self-evident indicators. The complications and disagreements one faces with respect to the ‘facts’ in such situations are rarely as to whether there is an armed conflict afoot, but rather as to whether it is purely IAC, or has been transformed by some complication into some form of hybrid or mixed armed conflict, such as ‘internationalised internal armed conflict’ or concurrent IAC and non-international armed conflict (‘NIAC’).
But this inference that the ‘facts on the ground’ will determine conflict characterisation affects both branches of the law of armed conflict (‘LOAC’) — IAC and NIAC. This assertion to the effect that an objective assessment of what you see before you will always set you right thus appears (consciously, it could be said) to leave no room for ‘policy’ or ‘interpretation’ by a state considering its options when faced with a threshold situation. As Andrew Clapham acutely asks, ‘[s]o if the law applies should we worry about the attitude of the government?’ The answer is, if the law clearly applies, ‘no’. The problem, however, is that the jurisdictional ‘facts’ that inhabit the threshold between NIAC/less-than-NIAC law enforcement (that is conflict and violence, but not an armed conflict such as to bring into operation the relevant LOAC), which are accorded the power to determine ‘if the law applies’, are significantly less objective than in prospective IAC situations. Indeed, they are fundamentally subjective and fluid as opposed to the more fixed and readily identifiable criteria common to IAC. As Geoffrey Best observes of this conundrum for the negotiators of the Geneva Conventions:
They had known what an international war was, but how were they to know a non-international armed conflict when they saw one? How were they to tell it from mob violence, riots, and banditry? … These were not silly or necessarily non-humanitarian questions.
For IAC, the relevant facts are generally widely and easily verifiable — the presence of one state’s forces on another state’s territory, the adversarial posturing and positioning by two states’ forces in international waters or airspace followed by the clash of arms between these forces, and so on. However, the ‘facts’ relevant to determining which side of the law enforcement–NIAC threshold a situation falls involve highly flexible concepts such as violence, banditry, terrorism and threat. As Kenneth Watkin has noted of the ‘difficulty in categorizing conflict at the lower end of the conflict spectrum’, with reference to the mixed law enforcement and/or NIAC options available in ‘stability operations’: ‘The degree of uncertainty regarding what law applies … should raise significant concern’. This is of intense practical importance given the fact that the ‘standard’ post-1945 armed conflict is (by a very large margin) NIAC, not IAC. One reason why this is such an intensely practical concern is the consequences that characterisation around this lower threshold manifest with regard to use of lethal force: as many practitioners and scholars have noted, use of lethal force in a law enforcement context is governed by very different rules to use of lethal force in a NIAC context.
My aim in this analysis is to explore three aspects of legal-policy concern which are legitimate considerations for states dealing with conflict characterisation at the lowest LOAC threshold — between less-than-NIAC law enforcement and NIAC. The first is a general outline of three elements of
legal-policy discretion which are clearly assumed and inherent within LOAC. Following this, I will examine two further, discrete aspects of legal-policy concern with respect to conflict characterisation, which mix the prudential and the categorical. The first is legal defensibility, a discourse which is fundamentally governed by the tension between applicable law and policy objectives. The second is utility, a concern which focuses upon the balance to be struck between the legal argument employed to justify conflict characterisation, and the capacity of the state to retain some degree of context control. This concern extends to issues as fundamental as facilitating the state’s desired course of action, and characterising, or assigning status to, the ‘other’ (or the ‘adversary’) within the conflict context.
Genuflection to the objective finality of ‘facts’ has never been, and still is not, the full picture in conflict characterisation at the law enforcement–NIAC threshold. Indeed, I think it is safe to assert that this threshold not only anticipates state discretion and policy influence, but essentially requires it. LOAC lawyers should not shy from this, nor obfuscate it as if it is heresy or in some way unlawful or unethical — recognition of this requirement is built into the law. W Michael Reisman’s analysis of the ‘relations between mission design and international legal and institutional environments’ in relation to Afghanistan, is clearly built around this honest and upfront recognition. Dale Stephens also assumes this truism, pointing to the ‘reality’ of policy overlays upon law, and the legal feedback loop into policy, which governments routinely utilise. There are, I believe, three aspects of legal-policy discretion that are legitimately available to states considering this threshold, and that are inherent within their decisions as to conflict characterisations. Each aspect deserves a brief exegesis.
The first aspect of this legal-policy discretion is that states retain some degree of control over the ‘facts’ and how they are labelled, presented, and interpreted. This occurs on many levels. It is generally states which hold the primary role in deciding, or at least influencing, what ‘facts’ are presented, how they are presented, and whether they are considered sufficient to tip a situation across the threshold. The targeted release by states of threshold-relevant intelligence and other information is a prime source of ‘facts’, and state responses to assertions of ‘fact’ by other actors are fundamental to determining the contextual colouration and appreciation those facts are given. It was, for example, predominant mastery of the ‘facts’ that (paradoxically) allowed the United States Government to present such a confused narrative when reporting the Osama bin Laden ‘kill/capture’ mission to the public.
Similarly, the manner in which a state officially or formally labels core jurisdictional ‘facts’ is highly influential in assessing their threshold relevance. States have a certain degree of labelling choice when describing a violent incident — the language of ‘a military-style offensive’, or ‘coordinated attack’ tends to support a finding of NIAC, whereas descriptions such as ‘terrorist incident’, ‘isolated incident’ or ‘murder’ tend to support a disposition to characterise the situation as less-than-NIAC. In his speech to the House of Commons, announcing the 1993 Joint Declaration on Northern Ireland (‘1993 Downing Street Declaration’), Prime Minister of the United Kingdom, John Major, was at pains to describe ‘the Troubles’ as ‘murder and destruction’, a situation that was amenable to a law enforcement response as an ‘internal UK matter’, which was precisely the characterisation the UK had pursued for more than three decades. This labelling was maintained even though the proposed solution was cast in terms of language more traditionally associated with NIAC contexts, such as ‘self-determination’, ‘political settlement’ and so on. The UK maintained this labelling throughout the entire operation in Northern Ireland, including through peaks of violence such as, for example, in 1970–71, when the Joint Security Committee was concerned to ensure ‘balanced reporting’ of what was ‘a near-war situation’. This conscious colouration of the threshold defining facts continues, with concern that the Consultative Group on the Past in Northern Ireland might recommend a re-classification of the Troubles from ‘law enforcement’ to ‘war’ the subject of heated debate in many forums, including at the Democratic Unionist Party of Northern Ireland assembly in 2008, and in submissions and evidence to the Consultative Group itself. Indeed, it is not a long leap between Charles Garraway’s astute observations about the possible application of Common Article 3 to this conflict, to an argument that, if the Troubles were reclassified as an armed conflict, it is far from clear (noting the information that is now available regarding support and sanctuary networks which crossed the Republic of Ireland–Northern Ireland border) that NIAC would have been the appropriate characterisation.
The second aspect of legitimate legal-policy discretion that LOAC confers upon states engages with the ‘constructive ambiguity’ inherent in many core NIAC threshold terms and concepts. LOAC expressly anticipates and permits scope for discretionary assessment by its principal implementers, states.
As Pictet points out in Commentary I:
All international Conventions, including the one with which we are concerned, are primarily the affair of Governments. Governments discuss them and sign them, and it is upon Governments that the duty of applying them devolves.
This important principle heralds three significant process implications. First, the primary duty for implementation in any given situation rests upon states, and therefore the views of states must be accorded (prima facie, at any rate) a degree of primacy. This must remain the case unless and until a binding judicial body has provided a more definitive assessment; one which has properly taken into account the views of the states concerned. Secondly, a duty to implement not only assumes a process of decision-making with respect to the mechanics of implementation — what, when, how, where and why — but also assumes a subsidiary requirement to settle the internal definitions and contextual appreciations necessary to properly frame and inform that decision-making process. As Michael Schmitt notes in the LOAC context,‘[i]n arriving at sound decisions regarding what course to pursue, the key lies in the process of decision making’, and this is equally true of state decisions as to the applicability of LOAC in the first place. The inherent process obligations that LOAC imposes upon states under the ‘implementation’ and ‘application’ banners inevitably involve preliminary shaping judgments as to parameters, inputs and relevance, which in turn inescapably (and legitimately) confer a degree of legal-policy discretion upon the state. Thirdly, the primacy of states as subjects of the duty to implement and apply LOAC clearly carries with it an inherent duty to interpret LOAC, and the susceptibility of key lower threshold concepts — such as ‘armed conflict’ (Common Article 3), and ‘internal disturbances and tensions’ (Additional Protocol II, art 1(2)) — to colouration and contextualisation is (intentionally) very high. This high susceptibility continues to permeate the legal penumbra surrounding the lower thresholds for LOAC regardless of the fact that there have been considered judicial attempts to further define these threshold terms (with the stated intention of assisting assessors to distinguish less-than-NIAC from NIAC), such as the International Criminal Tribunal for the Former Yugoslavia’s (‘ICTY’) ‘intensity of the conflict and the organization of the parties to the conflict’ analysis. Yet threshold conceptual uncertainties continue — as indeed they must if states are to retain their mandated primacy in implementation and interpretation. As Constantin von der Groeben noted in relation to the German Prosecutor’s investigation into the Kunduz Tanker incident in Afghanistan in 2009:
The ambiguity in the facts follows an ambiguity in the applicable laws. The problem with the air strike is that it was unclear whether it had been performed as part of a non-international armed conflict in Afghanistan or just as part of a stabilization mission below the threshold of ‘armed conflict’.
There is, therefore, a clear — and clearly lawful — expectation within LOAC at the lower thresholds that states will retain the primary initial role in interpreting and applying the critical threshold terms that either enliven or dismiss the application of LOAC in situations which can potentially be characterised as either NIAC or as less-than-NIAC law enforcement.
The third aspect of legitimate legal-policy discretion which states still possess, in terms of LOAC at the lower threshold, is that the context of violence hovering at or around that lower threshold is by its very nature often uncertain. There is nothing fundamentally improper with states imposing some interpretive order upon such contexts, and thus relying upon that interpretive order as a key driver in selecting the applicable legal paradigm — less-than-NIAC law enforcement or NIAC — as international law ultimately requires that they must. The first indicator of the legitimacy of this discretion is the very fact that LOAC actually evinces multiple thresholds. This naturally complicates the already much more subjective assessment (in comparison to IAC) required when determining whether a situation should be characterised as one remaining subject to the ‘normal’ law enforcement paradigm, or one that has crossed the Rubicon into NIAC. The para 1.1.1 definition of NIAC adopted by Michael Schmitt, Charles Garraway and Yoram Dinstein in The Manual on the Law of Non-International Armed Conflict is an express attempt to impose some interpretive order upon the issue of ‘threshold’ in light of the variety of NIAC thresholds found in LOAC. One of LOAC’s most important paradigmatic features — indeed the feature that underpins its continuing ‘fitness for purpose’ — is this very flexibility, or adaptive capacity. Geoffrey Corn’s observation (in relation to conflict characterisation with respect to Afghanistan, the Taliban, and al-Qaeda) that ‘diverging conclusions’ can ‘stress’ the paradigm but are ‘in no way fatal to the ability of the law to adapt to the … changing nature of warfare’ is indicative of the high degree of contextual discretion built into LOAC. As Corn continues:
All law is adaptive, but this is particularly true with regard to the LOAC, a conclusion illustrated by the fact that this law has endured for centuries. This area of international legal regulation has been historically resilient precisely because the law has always responded to the changes in the nature of warfare. Perhaps more importantly, these responses have been implemented in a manner considered credible by States and the armed forces called upon to execute military conflicts.
In no component of LOAC is this more clearly evident than in the broad discretion and flexibility afforded to states grappling with the complicated contexts they inevitably encounter at the NIAC/less-than-NIAC threshold for paradigm applicability.
With this brief outline of some of the myriad paths by which legal-policy discretion is (in general) actually built into, or assumed by, LOAC at the lower thresholds, it is now possible to examine in greater detail some of the substantive concerns that might be said to shape and influence the exercise of these discretions.
The legal defensibility of a state’s decision as to conflict characterisation will always be a first order concern for that state. It is obvious that advancing an unsustainable legal argument as to why a state has determined that a situation has or has not crossed the lower LOAC threshold between law enforcement and NIAC will create problems for that state on many levels — from difficulties in coalition building; through to the political, policy, and reputation costs incurred when a state becomes the subject of widely mobilised criticism; to possible susceptibility to judicial review. Characterisation needs to be readily defensible in terms of international legal coherence and compliance, and must, to some extent at least, reflect the ‘facts’. However, this simple assertion disguises a legal-policy assessment process which itself contains significant scope for state discretion.
The immediate post-1949 view of thresholds, adopted by many commentators, focused upon applying LOAC as widely as possible in internal situations, and was premised upon the assumption that LOAC offered better and broader protections than non-LOAC based responses to internal violence. As Pictet notes in Commentary I:
We think, on the contrary, that the Article [Common Article 3] should be applied as widely as possible. There can be no reason against this. For, contrary to what may have been thought, the Article in its reduced form does not in any way limit the right of a State to put down rebellion. Nor does it increase in the slightest the authority of the rebel party.
It is arguable, however, that this humanitarian imperative to apply LOAC was based upon a once justified, but now perhaps outdated (and cynical), general view of state aims in denying LOAC application. The analyses of Rogier Bartels and others regarding pre-1949 attitudes to use of force in the suppression of rebels is instructive in this regard. Martin Luther, as Bartels notes, was not alone amongst theologians and lawyers in advocating, and justifying, actions that we would describe as extrajudicial punishment and summary execution of rebels. For Emmerich de Vattel, the law permitted a sovereign to ignore the laws of war when dealing with rebellion that had not risen to the level of formal ‘belligerency’ — a situational status through which rebels could secure recognition, a degree of international status, and an obligation (upon both parties) to comply with the laws of war. W F Finlason, writing from a British perspective in 1867 and 1868 (on the legal consequences of a declaration of ‘martial law’ in response to rebellion), was similarly disposed to observe that the traditional approach to rebellion was that ‘normal’ rules were suspended and arbitrary detention and (in some circumstances) summary execution could be accommodated.
At the 1949 Diplomatic Conference, many states objected to both the proposals of the International Committee of the Red Cross (‘ICRC’) to apply IAC LOAC to situations of internal conflict, and proposals that appeared to set, too definitively, the threshold between law enforcement and LOAC-governed situations. Regardless of public statements and positions, the reason that states objected to any broader and more detailed application of LOAC at this threshold — or so the traditional discourse asserts — was that they wished to retain the freedom to deal with such violence as a purely internal matter, not subject to the constraints of LOAC. What arguably underpinned this assessment at the time was reflection upon the previously dire legal position of rebels below the belligerency threshold. But the shadow lying behind this assessment appeared to be a perhaps overly cynical assumption that states would seek to characterise a situation as below the LOAC threshold specifically to deny the application of LOAC, in the belief that this would legally confer upon the state a much greater freedom of action to ruthlessly and violently suppress the rebellious conduct. An echo of this conception of the state motivated by a desire to be able to kill, detain, restrict, and destroy, at levels and in situations beyond what LOAC would permit, is recognised by James Stewart (writing on the 1979 Soviet invasion of Afghanistan, and internationalised internal armed conflict more generally):
Consequently, States involved will seek to deny the very existence of insurgent forces or the possibility that such forces have entitlements or liabilities, thereby affecting the humanitarian law applicable between State and internationalized insurgent and, perhaps more importantly, the law protecting civilians in such circumstances.
This is but one discrete manifestation of that broader traditional assumption — what John Ip describes as ‘the conventional account’, and a theme also identified by Daniel Abebe and Eric Posner as a common thread in ‘foreign affairs legalism’ — that, faced with a threat, the executive will generally overreact, employ excessive means, and seek to veil its conduct within the least constricting legal paradigm so as to facilitate maximum freedom of action.
At the same time as the human rights project has fundamentally altered and limited the less-than-NIAC law enforcement paradigm, any supposition that a state may instead opt for NIAC because it imposes only a light legal touch is blind to the reality that is international law, and in particular IAC LOAC and human rights reach-down into NIAC contexts. At one stage, as Bartels notes, the laws of war were such that if a state ‘recognised’ what today would be characterised as a NIAC, it opened the path for the next step — recognition of ‘belligerency’ — to be taken out of its control. Such recognition brought with it obligations in line with what is today the LOAC as applicable to IAC, thus opening the state to greater levels of informed critique and international accountability. This possibility still haunts the NIAC threshold in a number of ways. First, as some commentators have suggested, many states (particularly developed states) are already effectively held to higher standards of compliance with LOAC. Secondly, as Stewart has observed, customary international law has at any rate certainly (although contestably) reduced the gap between the IAC and NIAC regimes. Thirdly, as Schmitt has argued, ‘[a]pprehension over condemnation certainly influences … policy choice’. Today, states regularly, and often very publicly, opt for more onerous limitations on use of force than are required by LOAC — that is, policy is used to constrain otherwise lawful action. The Commander of NATO’s International Security Assistance Force Tactical Directive is but one manifestation of this tendency. Fourthly, art 1(4) of Additional Protocol I expressly provides a pathway for IAC LOAC to govern certain self-determination related NIAC contexts. The sum of these many pressures can clearly lead a state to consider it less problematic to characterise a situation as law enforcement. In taking this path, the state can avoid a number of significantly escalated and mandated risks, such as: unconstrained international assessment and comparison of its conduct, which the NIAC regime both anticipates and facilitates; the risk of the NIAC being conflated with IAC (through concepts such as ‘internationalisation’, or ‘internationalised internal armed conflict’), which reach-down implies and the space for political controversy and debate affords; and the greater risk of ‘atrocity’, and thus condemnation, which LOAC’s more permissive approach to use of force can invite.
A fundamental legal-policy purpose which influences a state’s approach to characterisation at the lower thresholds is the degree to which the law serves or undermines that state’s ability to exert a measure of control over the broader strategic context of the conflict. Characterisation decisions can thus reflect concern with a range of legal-policy consequences of the second and third order. Determining that a situation is one of NIAC as opposed to less-than-NIAC law enforcement carries with it intensely practical, and political ‘therefores’, such as the need to commit heavier (more expensive) combat capabilities as opposed to lighter (less expensive) reconstruction or stabilisation forces. Combat capabilities are significantly more expensive in terms of required enablers — 3:1 through to 10:1 support:combat ratios are standard across most militaries. Committing combat forces to combat roles also entails a higher risk of own casualties, and of inflicting civilian casualties, with all of the domestic and international political and social consequences that flow from inflicting suffering and sustaining casualties in war. Further, the political commitment (and cost) required for engaging in combat operations is radically higher than for non-combat roles: the political capital required to assemble a fighting coalition is much greater than that required to create a stabilisation coalition; the potential to be tainted by association with the combat conduct of other troop-contributing nations is much greater; and the legal and capability interoperability issues that arise when LOAC-based use of force enters the equation are sharper. States have long been acutely aware, for example, of the potential consequences of adopting LOAC authorisations in urban contexts. As the Final Report to Congress: Conduct of the Persian Gulf War (submitted in 1992) observed:
The concentration of Iraqi military personnel and vehicles, including tanks, invited attack. CINCCENT [Commander-in-Chief, Central Command] decided against attack of the Iraqi forces in Kuwait City, since it could lead to substantial collateral damage to Kuwaiti civilian property and could cause surviving Iraqi units to decide to mount a defense from Kuwait City rather than depart. Iraqi units remaining in Kuwait City would cause the Coalition to engage in military operations in urban terrain, a form of fighting that is costly to attacker, defender, innocent civilians, and civilian objects.
There are, therefore, significant political, social, operational, reputational and economic consequences that flow from a characterisation decision, and so it should be no surprise that states will have such concerns — which fundamentally affect their ability to maintain some degree of control over their strategic context — in mind when making that decision. A complete surrender to black letter law is thus not only a legal nonsense, but also a practical nonsense. This is evident with regard to several levels of state concern.
The first legal-policy dimension which reflects the logical concern of states to adopt a characterisation strategy which best serves their ability to maintain a level of control over the strategic context is the very uncertainty that surrounds not only which threshold is called into play, but the consequences of doing so. Determining that a situation is one of NIAC at the lowest level does not, for example, insulate a state from the possibility that others will take that ‘admission’ and then seek to re-characterise the conflict as IAC. This carries with it, of course, very significant legal consequences for the state. There are several mechanisms by which a state’s decision that a NIAC exists — at the lowest threshold — may spin the characterisation debate out of its control. First, if the ‘other’, or a series of states who may support the other, are able to re-badge the situation as self-determination, the Additional Protocol I regime and assimilation to full IAC might follow. Secondly, admission that a situation is a NIAC brings with it the potential for application of the ICTY’s Prosecutor v Rajic test. In essence, this test envisages that, regardless of how correct and limited a state’s determination as to NIAC might be, intervention by another state (which, if it is on the side of the ‘other’ is something the first state cannot control) will de facto internationalise the conflict. Once again, any state facing a legitimate NIAC or less-than-NIAC characterisation decision will always seek to maintain one eye on its ability to retain a modicum of strategic control over the situation. But in doing so, it will need to assess the potential for third party actions to unilaterally internationalise the conflict (in legal terms) along with all of the law, scrutiny and avenues for greater condemnation that accompany IAC and internationalised NIAC. Thirdly, a concession on the part of the state that a situation is one of NIAC almost always carries with it the potential for the ‘other’ to internationalise the conflict through inviting other actors to engage on its side — a fear undoubtedly of concern to the al-Assad regime in Syria. Finally, precisely because ‘principles’ — which are designed to take much of their colour and substance from their context, as this is what underwrites their flexibility and fitness for purpose — play such a large role in LOAC, the scope for differential interpretations by actors other than the centrally concerned state is very broad. The heavy reliance that LOAC places upon applying general principles in a contextually sensitive manner is an inescapably open invitation to others to apply the same principles, in the same context, and arrive at legitimate and equally defensible, but different, legal conclusions. The gulf between the fundamentally different conclusions of the US/NATO and Human Rights Watch reports as to the lawfulness of the air strikes on Azizabad on 21–22 August 2008 is a case in point.
Each of these risks — all of them irremediably inherent within any decision to characterise a situation as NIAC — may in fact convince a state that the most contextually useful approach is to adopt a less-than-NIAC law enforcement characterisation. Law enforcement carries with it a much stronger reliance upon and recourse to the ‘normal’ domestic legal paradigm within the affected state, placing the state in a significantly enhanced position in terms of control over and exegesis upon the applicable law, as compared to others who will seek to utilise international law as the substance and mechanism of critique. A domestic ‘law enforcement situation’ is generally the very definition of an ‘internal matter’, carrying with it the stronger sense of legal insulation that this foundational principle of the international order implies, when compared to the more active, overt, forcible and effective interventions available to the international community with respect to LOAC. And, because domestic law in relation to law enforcement authorisations is often much more oriented towards ‘black letter law’ than LOAC, the state generally holds a very strong hand in interpreting its own law and explaining/defending — in legal terms — the way in which it is applied in any given context. This is not to say that a law enforcement situation is opaque to external scrutiny and criticism — it is not, thankfully, and the European Court of Human Rights provides clear evidence of this (the Chechnya cases, for example). But it is nevertheless clear that for most states, the scope for critique and external scrutiny, using the law as the mechanism and measurement of criticism, is much reduced as compared to that afforded under LOAC.
An example of the seductive contextual utility which adopting a law enforcement characterisation as opposed to a NIAC characterisation may afford is found in the United Nations Security Council’s (‘UNSC’) approach to use of force during the International Force for East Timor (‘INTERFET’) and United Nations Transitional Administration in East Timor (‘UNTAET’) operations in East Timor/Timor Leste in 2000–01.
The context is well summarised by Stephens, who states that during the UNTAET mission in the early 2000s, it was the duty of the Peacekeeping Force (‘PKF’)
to disarm and detain members of the militia groups who were seeking to stymie the would-be transition to East Timorese independence. Such groups were openly displaying weapons, including semi-automatic/automatic weapons and grenades and conducting tactical patrols within East Timor. They were exploiting the UN rules of engagement and had fired upon PKF troops without warning, attacked PKF positions, ambushed PKF patrols, and responded to verbal challenges issued by PKF troops by opening fire. After the PKF had experienced a number of casualties, the Department of Peacekeeping Operations ultimately authorized an addition to the rules of engagement that deemed certain tactical behaviour by militia groups as threatening for the purposes of activating a heightened right of self-defence. This in turn gave an individual soldier the capacity to act according to a subjective belief that he or she was threatened when faced with such behaviour and to employ lethal force in self-defence.
The attraction of simply declaring the situation to be one of NIAC, thus opening up the LOAC attack rules, was certainly considered at the time. However, the ability to retain the heightened level of contextual control which a law enforcement characterisation afforded — for example, control over their own contingents’ use of force, and control over the strategic and political fallout with Indonesia that would have accompanied such a change in conflict characterisation — was considered more important by the states concerned, and by the UNSC. The resulting rules of engagement (‘ROE’) addressed this issue by ‘expanding’ the concept of self-defence:
Following the passage of [UNSC Resolution] 1319 (2000), and on the basis of continuing concerns with respect to the threat environment, the Deputy Force Commander, the Force Commander and, ultimately, the SRSG [Special Representative of the Secretary General], requested that the UN change the ROE to authorise a more assertive military posture. In response, the UN modified the definition of ‘hostile intent’ and the previously mandated escalated approach to the use of force. The amended ROE meant that force could be initiated in self-defence in circumstances in which militia were positively identified, were carrying arms and behaving in a manner that indicated a hostile intent, and a member of the PKF formed the subjective assessment that a threat was imminent. In such circumstances, there was no requirement that PKF apply non-lethal measures before resorting to force.
The overall consequence was that contextual utility was best served by adopting a legal-policy decision as to characterisation which (on the one hand) retained the levels of contextual control inherent in a less-than-NIAC law enforcement characterisation, whilst approving (on the other hand) extended ROE based around individual self-defence — an innovation not without a certain level of domestic legal risk for the military members involved.
A second example of the mixture of law and policy in conflict characterisation decisions by states may be found in Australian concerns (in relation to the INTERFET operation in East Timor) that if Geneva Convention IV was applied, it could imply reciprocity. This in turn, it was feared, could ‘have the legal consequence either of, rendering ADF [Australian Defence Force] personnel “lawful” targets, making Australia party to any conflict, or brining [sic] into effect the other Geneva Conventions of 1949’. Ultimately, a legal-policy decision determined that
the Convention [Geneva Convention IV] would not make [Australian] troops a party to a conflict who could then be targeted ‘as of right by other parties to the conflict’. … If the Fourth Convention applied and armed elements attacked [Australian] troops this would be illegal unless it was part of an organised armed force with a responsible command structure.
The possibility of affording anything resembling reciprocal rights to the adversary was thus clearly an issue of policy concern which informed the legal decision as to conflict characterisation. As a law enforcement operation below the NIAC (or indeed IAC) threshold, the adversary had absolutely no paradigmatic claim to LOAC authorisations in relation to the INTERFET force.
A third example of the legal-policy sub-decisions implicit in a state’s decision-making in relation to the broader question of conflict characterisation concerns the extraterritoriality of domestic law. The role which this factor plays in state decision-making as to conflict characterisation is highly variable and reflects, in many ways, a number of fundamental legal and policy assumptions which states bring to their deliberations. Some states are loathe, as a matter of principle, to extend the reach of their domestic law to their nationals whilst abroad. Although US military forces take the Uniform Code of Military Justice with them everywhere, John Knox has characterised the general US position as one of reluctance to apply law extraterritorially, with a number of exceptions for particularly egregious, or fundamental, matters such as (respectively) child sexual exploitation and tax offences. Other states are willing to apply broader swathes of their criminal law to their nationals — including military members — when abroad. This was a fundamental issue in the German analysis of the legal regime against which the Kunduz tanker incident was to be assessed.
As von der Groeben notes:
German Criminal Law is applicable to acts committed abroad by a German national. Any German soldier in Afghanistan or elsewhere in the world is therefore generally responsible under German Criminal Law. In addition to the criminal responsibility under German criminal law, a German soldier can be responsible under the Völkerstrafgesetzbuch (German Code of Crimes against International Law) if his act was committed in relation to an international or non-international armed conflict and if it amounts to an act of genocide, crime against humanity or a war crime.
Determining whether it was the highly transportable general criminal law, or the more specific German law regulating conduct in armed conflict situations, which applied to this incident in Afghanistan was thus of fundamental importance because these two regimes approached the facts differently, applied different standards to their assessment, and could come to radically different conclusions. Australia is in a similar position. Section 15.3 of the Australian Criminal Code (extended geographical jurisdiction category C) permits many offences governed by this category of jurisdiction to apply to nationals ‘whether or not the conduct constituting the alleged offence occurs in Australia; and whether or not a result of the conduct constituting the alleged offence occurs in Australia’, but with a defence which applies if the conduct was entirely outside Australia and the accused is not an Australian national or body corporate and the offence is not an offence under the applicable local law. These laws apply to Australian military members on law enforcement operations overseas. Where the situation is one of armed conflict, however, priority in assessing conduct should generally default to the war crimes, grave breaches and crimes against humanity offences within div 268 of the Criminal Code, with all of the differential tests and contextual appreciations required when assessing conduct against LOAC as opposed to general criminal law. The characterisation question, at the NIAC and less-than-NIAC law enforcement threshold, is thus an important one for determining the extraterritoriality and/or extra-jurisdictionality of domestic law in relation to a state’s forces. It is obtuse to presume for one moment that this legal-policy factor does not enter into consideration when states — with one eye on the contextual utility and control they will retain or surrender with respect to their own forces — make a conflict characterisation decision.
A second legal-policy discretion implicit in state decision-making as to conflict characterisation at the lower thresholds is the degree to which the selected legal paradigm serves or undermines access to certain courses of action desired by the state. Just as any given situation may prompt a state to seek to characterise a situation as law enforcement rather than NIAC (so as to ensure it retains the maximum degree of context control), it is equally possible that a state may utilise its discretion in conflict characterisation to facilitate a preferred course of action. Two illustrative examples will suffice. The first is Germany’s long denial of the belligerency of its forces within the NIAC in Afghanistan. This served an important domestic political purpose related to constitutional arrangements for the deployment of German forces, and a general public reluctance as to the involvement of German forces in offensive, LOAC-governed, operations. This translated into a bipartisan determination to maintain a non-LOAC-based conflict characterisation, in relation to the German contingent of the International Security Assistance Forces at any rate, and was clearly a consequence of domestic political concern as opposed to either fine legal analysis or a truthful assessment of the ‘facts on the ground’. The German Government maintained this stance until the facts were overwhelming, and investigation of the Kunduz Tanker incident invited independent review of this characterisation. Indeed, had Germany maintained its less-than-NIAC involvement characterisation, it is difficult to see how the German commander at the centre of the Kunduz incident could have avoided further criminal investigation in relation to general German criminal law, as opposed to that specialised body of German criminal law which applies in situations of armed conflict.
The second example can be illustrated through reference to internal US bureaucratic and political manoeuvrings as to the nature of operations in Libya in 2011. Legal debate surrounding conflict characterisation (with respect to US involvement) centred upon the intensely practical issue of whether President Obama was required to seek Congressional approval for the continued deployment of US forces in support of the NATO Libya operation, Operation Unified Protector. Media reports have indicated that the prospect of Congress ‘playing politics’ with this issue for purely domestic and partisan ends was so strong that President Obama broke with convention so as to avoid this outcome, and rejected the normally ‘trump’ Department of Justice Office of Legal Counsel (‘OLC’) advice. The OLC view — apparently supported by the Department of Defense General Counsel — was contradicted by the Department of State Legal Adviser (and, reportedly, the White House Counsel), who sought to insulate the President from any requirement (under the War Powers Resolution) to return to Congress within 60 days of committing forces, for approval to continue engagement in ‘hostilities’. That both sides of the debate accepted that the situation was one of armed conflict, and that LOAC would apply, was arguably implicit: the OLC opinion refers to the ‘conduct [of] offensive operations, including attacks on civilians’ by Gaddafi forces, and the targeting of military objectives by NATO forces. Legal Adviser Harold Hongju Koh’s statement before the Senate Foreign Relations Committee similarly refers to a joint 1975 Department of State–Department of Defense legal opinion to the effect that even some situations ‘in which units of the US armed forces are actively engaged in exchanges of fire with opposing units of hostile forces’ (highly LOAC-laden language) might still not rise to the level of ‘hostilities’. He further cites the Reagan Administration’s view that the landing of US forces in Grenada in 1983 and the ‘combat’ operations that followed — clearly a situation of armed conflict — nevertheless did not amount to ‘hostilities’ for domestic constitutional purposes.
The critical legal term at issue — ‘hostilities’ — was therefore analysed from an entirely domestic perspective, and (in the case of Legal Adviser Koh’s advice) clearly with policy and political concerns in mind. Koh’s legal advice is frank on this point, emphasising the need to determine this particular ‘legal’ question ‘more by interbranch practice than by a narrow parsing of dictionary definitions’. Koh concluded that the legal debate
should not distract those of us in government from the most urgent question facing us, which is one not of law but of policy: Will Congress provide its support for NATO’s mission in Libya … ensuring that Qadhafi does not regain the upper hand against the people of Libya? … However we construe the War Powers Resolution, we can all agree that it serves only Qadhafi’s interest for the United States to withdraw from this NATO operation before it is finished.
Thus whilst the OLC advice characterised US involvement in the Libya operation as ‘hostilities’ for the purposes of the War Powers Resolution, the view of the Department of State Legal Adviser’s view (ultimately accepted by the President) denied that the operation met this threshold. Consequently, the ‘60 day pullout’ rule did not apply, and the President was not required to return to a hostile and highly partisan Congress to seek its authorisation to continue the deployment. The course of this debate thus provides yet another clear illustration of the tendency for ‘legal’ conflict characterisation decisions — not only in LOAC terms, but also in domestic law — to be driven (often overtly) by policy concerns.
The interaction of law and policy in terms of context control is nowhere more critical, nor more evident, than in relation to assigning a legal status to the adversary (the ‘other’) in any given conflict situation. Indeed, a desire to assign a legally ‘useful’ status to the adversary could, in some circumstances, drive the broader question as to conflict characterisation as a whole. The default position, according to Pictet, was traditionally that ‘[i]n a civil war the lawful Government, or that which so styles itself, tends to regard its adversaries as common criminals’. But this seemingly bland recognition that decisions as to the legal status of the adversary are driven by both legal and policy concerns disguises a good deal of legitimate, and long acknowledged, decisional latitude and flexibility. There are at least three intricately linked, but clearly mixed, legal-policy concerns that can agitate this ‘legal’ decision.
The first of these concerns is to retain some control over, or indeed to exclude, the capacity of third parties to legitimately challenge a state’s conflict characterisation decision. If a state wishes to exclude the legal potential for a third state to physically intervene on behalf of the other in a potential NIAC context, the surest course to steer is to maintain that the situation is one of mere riot and banditry (simple criminality) as opposed to having crossed the threshold into NIAC. Criminality is an internal state law enforcement matter and there is no general regime which acknowledges the right of a third party to forcibly intervene in the situation. For states, accepting that there is a NIAC afoot thus runs two potential risks that do not present when the other is successfully assigned the status of simple criminal. The first risk relates to the fact that states hold justifiable fears that recognition of the other as an adversary in a NIAC — regardless of the fact that this other may still be treated as a criminal if apprehended — can accord this adversary an internationally acknowledged ‘quasi’ status. Clearly, NIAC LOAC does not afford the rebel any actually altered legal status expressible as a ‘positive’ for the adversary. It does not change the dominant paradigm for assessing LOAC-compliant rebel acts against state forces from criminality to combatant immunity, but does, however, render those same rebel forces liable to lethal targeting operations by state forces, as opposed to keeping state forces limited to law enforcement based apprehension options for action. Neither of these is an unambiguously ‘positive’ consequence for the other; indeed, these consequences are more properly characterisable as positives for the state.
Despite these positives, however, conceding even the sheen of a status beyond simple criminality is often a trump states are unwilling to cede. Pierre Hauck and Sven Peterke, discussing analogies between organised crime and organised armed groups, argue that the essence of criminal law is that it seeks to deal with individuals for what they did, not who they are: ‘The focus on illegal activities is clearly favoured by the fact that modern criminal law does not punish individuals for what they are (eg, members of a criminal organization) but for what they do (eg, application of violence)’. Thus any shift away from a routine criminal law based focus on conduct (violence) can invite an unwelcome focus on status (and, hence, aims and objectives), potentially generating attention, empathy and even sympathy. As Pictet’s Commentary I on the purpose of the last sentence of art 3(2) of Geneva Convention I records:
It meets the fear — always the same one — that the application of the Convention, even to a very limited extent, in cases of civil war may interfere with the de jure Government’s lawful suppression of the revolt, or that it may confer belligerent status, and consequently increased authority, upon the adverse Party.
To compel the Government of a State in the throes of internal convulsions to apply to these internal disturbances the whole body of provisions of a Convention expressly concluded to cover the case of war would mean giving its enemies, who might be no more that a handful of rebels or common brigands, the status of belligerents, and possibly even a certain degree of legal recognition. There was also a risk of common or ordinary criminals being encouraged to give themselves a semblance of organization as a pretext for claiming the benefit of the Conventions, representing their crimes as ‘acts of war’ in order to escape punishment for them.
David Kennedy’s assessment that ‘international law only rarely offers a definitive judgment on who is right’ is highly apposite. Whilst the less-than-NIAC law enforcement paradigm offers the state the conditional, but useful, trump of interpretive supremacy regarding ‘criminality’ in domestic law, the introduction of an applicable international legal regime, such as LOAC, brings with it a degree of inherent interpretive uncertainty and internationalised debate over which the state has limited control. Indeed, some non-western legal traditions afford an acknowledged and ‘higher’ status to rebels, while other traditions (including, arguably, the concept of ‘belligerency’) provide for concessionary ‘additional’ status for certain forms of rebellion, and recognise that such concessions carry with them increased legal rights, and thus status. As Dawn Steinhoff and many others have observed, even this sheen of ‘higher’ status, as illegitimate as it may be, can to some extent ‘internationalise’ the conflict and reduce a state’s capacity to maintain its treatment of captured fighters of the adversary party as criminals, rather than as quasi-prisoners of war (where those individuals have engaged only in conduct that would have been LOAC-compliant had it been undertaken by state forces). Affording even the quasi-status of ‘adversary’ under NIAC LOAC creates the opportunity for that adversary to exploit, and be heard within, the international context which LOAC bequeaths to all armed conflict, including NIAC. Annyssa Bellal, Gilles Giacca and Stuart Casey-Maslen appear surprised at the ‘almost visceral rejection’ by states of any suggestion that non-state actors in Afghanistan owe human rights obligations, arguing that states need to ‘accept that the world has transformed into one where a variety of non-state actors potentially have a range of international human rights law obligations’. I, on the other hand, am not at all surprised by this rejection of any assertion that other armed groups may have IHRL obligations, as this would be a major concession by states as to ‘status’. Keeping the adversary firmly within the category of mere criminal can reduce the opportunity for legitimate and sanctioned external interference and international pressure, albeit to a decreasing extent as IHRL continues its welcome evolution into an internationally available measure for assessing law enforcement-based conduct.
This is not, however, a unilateral consideration — states may equally seek to characterise a situation as a NIAC precisely so that they can hold rebels to account under LOAC, often for presentational as well as legal reasons. Characterising a conflict context as a NIAC also legitimately opens up for the state the right to conduct lethal targeting operations (in accordance with NIAC LOAC) unhindered by the usual requirements of a law enforcement based response (with its emphasis upon apprehension and its expectation that lethal force is generally permitted only in situations of individual or third party self-defence) as required in less-than-NIAC situations. Similarly, states may wish to maintain a less-than-NIAC characterisation out of a desire to keep their internal security forces in check. By maintaining a law enforcement characterisation, and thus removing from security forces any legal capacity to utilise LOAC-based means and methods — such as lethal targeting rather than apprehension based operations — states can maintain a tight leash on use of force. At its best, such self-limitation might be the result of a consciously ethical decision by a state: welcome, but, perhaps less laudably, it might simply be recognition of the fact that ‘“barbarism” is a self-defeating strategy, especially if the long-term goals are to achieve a political outcome such as peace and security or regime change, because it serves to increase political and military resistance’. Ultimately, it might simply reflect hard-earned wisdom in terms of control of security forces with a history of excess. As Guatemala: Memory of Silence, a report into the long NIAC in that state reports: 93 per cent of acts of violence were perpetrated by ‘agents of the State, especially the Army’, whilst 3 per cent were committed by the guerrillas. Such experience offers as sound a reason as any to limit a potentially heavy-handed response by state security forces to narrower law enforcement authorisations, as opposed to LOAC permissions.
A second reason why states may seek to retain a less-than-NIAC, law enforcement based characterisation of the ‘other’ is that states have significant freedom to criminalise forms of organisation and association which they can label as being for criminal (most particularly, profit-based) ends. However, recognising a conflict context as a NIAC radically limits a state’s capacity to set the definitional and descriptive conditions for identifying and criminalising the ‘other’. LOAC has a good deal to say about status and identification in NIAC: civilians taking a direct part in hostilities, and, within this category, organised armed groups in particular. There is almost no leeway for states to unilaterally rewrite or depart from the relatively formal criteria that LOAC has already established for assigning this status, which leaves the state with little political or legal freedom to proscribe contextually convenient and specific forms of organisation and conduct — an option still available via general criminal law, where a situation remains below the NIAC threshold. During the Troubles in Northern Ireland, as noted previously, the UK Government was at regular pains to employ the language of criminality. In one late 1971 assessment of the situation, the language employed emphasised the increasing association of ‘professional criminals’ with ‘political criminals’: namely cooperation between the Irish Republican Army (‘IRA’) or the Ulster Volunteer Force, and ‘the hooligan element of the working class in Catholic/Republican and Protestant urban areas’. In the Northern Ireland context, this requirement for paradigmatic discipline in terms of language was evident on many levels. Clive Walker and Russell Weaver have analysed, for example, the attempted exploitation by the Republican movement of the previously noted reference to ‘self-determination’ in the 1993 Downing Street Declaration to internationalise the conflict (likely by reference to the very situation envisaged in art 1(4) of Additional Protocol I as being one where LOAC applies). Kieran McEvoy’s fascinating study of legal character and status in relation to prisoners in Northern Ireland provides many further examples of the ways in which a criminal group seeking international recognition may attempt to leverage LOAC. As he perceptively notes, ‘[i]n a violent political struggle, the treatment of prisoners is a mirror to the state’s view of the conflict’. Paramilitary prisoners (particularly Republicans, records McEvoy) often engaged in violence, escapes, hunger strikes and other conduct ‘in the assertion of their status as “political” rather than “ordinary” prisoners’. IRA prisoners often avoided using post-incarceration reintegration services so as to maintain their status as political prisoners, ‘lest they be seen as acquiescing to the label of “criminal”’. Indeed, as McEvoy himself seems to evidence, the linguistic slip from ‘political prisoner’ to ‘former combatant’ is one easily made, but one that carries with it significant consequences in terms of legal characterisation, given that ‘combatant’ is a legal term of art only in LOAC.
However, as noted previously, this consideration also has policy implications in reverse; that is, third party states can also achieve policy aims by characterising a conflict as a NIAC (in which they might be able to intervene) as opposed to a less-than-NIAC law enforcement matter (in which they generally may not). Darryl Li has argued that the Bush Administration cultivated the term ‘foreign fighter’ for precisely this interventionist purpose. One recent illustration of a policy pull on conflict characterisation is the Libya intervention by NATO, where it was implicit that the US, UK and France (amongst others) accepted that there was an armed conflict afoot (a NIAC, at least at the start), as they could not otherwise have lawfully conducted LOAC-based targeting operations. This characterisation was politically and legally convenient as it allowed for LOAC-based targeting interventions in support of rebel groups, an option not available to a third party (even with a UNSC Chapter VII ‘all necessary measures’ mandate) where the context has not been recognised as having risen to the threshold of an armed conflict, and thus remains a law enforcement matter. It is curious indeed to compare the LOAC-laden language of NATO in relation to Libya (where they wanted to intervene), with the assiduously LOAC-free language of the international community in relation to Syria (where there appears to be little appetite to intervene), given that this is likely a conscious policy characterisation choice, rather than simple rhetorical accident.
At the 1949 Diplomatic Conference to settle the four Geneva Conventions, many states objected to both the ICRC’s proposals on applying IAC LOAC to situations of internal conflict, and to any proposal that appeared to set, too definitively, the threshold between law enforcement and LOAC-governed situations. For some states, this highly pragmatic approach may well have been prompted by their desire to preserve some legal capacity to characterise conflict situations as being below the LOAC threshold — and thus to deny the application of LOAC — in the belief that this would confer upon them a much greater freedom of action to ruthlessly and violently suppress rebellious conduct. Other states were clearly more concerned with the legal practicalities, and potential unintended consequences, of attempting to impose too narrow an interpretive order upon the lower threshold of NIAC, given that it was (and remains) conceptually and practically very difficult to do so. Regardless of motivation, however, the result was, and remains, that LOAC affords states a bespoke and broadly tailored discretion to fashion characterisation decisions at the NIAC/less-than-NIAC law enforcement threshold of internal conflict. Operating in a discourse space that extends radically beyond rhetorical genuflection to the threshold defining ‘facts on the ground’, this discretion very consciously anticipates a wide margin for legal-policy influence. It is for this reason that assessing, and understanding, ‘the law’ as it relates to the NIAC/less-than-NIAC conflict threshold remains such a fundamentally challenging proposition with LOAC.
[*] Rob McLaughlin is Associate Professor of Law at ANU. He has served as Director of Operations and International Law (Department of Defence), Director Naval Legal Service, Strategic Legal Adviser, and Fleet Legal Officer (Royal Australian Navy). He has served in East Timor, Iraq, and on maritime law enforcement operations. He is a consultant to the United Nations Office on Drugs and Crime and to the International Maritime Organization on counter-piracy law, and holds a PhD from Cambridge University. The author wishes to thank the anonymous reviewers for their helpful comments and suggestions.
 Jean S Pictet (ed), Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field: Commentary (International Committee of the Red Cross, 1952) 29 (‘Commentary I’).
 Geneva Convention for the Amelioration of the Wounded and Sick in Armed Forces in the Field of 12 August, 1949, opened for signature 12 August 1949, 75 UNTS 31 (entered into force 21 October 1950) (‘Geneva Convention I’).
 See, eg, Annyssa Bellal, Gilles Giacca, and Stuart Casey-Maslen, ‘International Law and Armed Non-State Actors in Afghanistan’  881 International Review of the Red Cross 1, 5–7; James G Stewart, ‘Towards a Single Definition of Armed Conflict in International Humanitarian Law: A Critique of Internationalised Armed Conflict’  850 International Review of the Red Cross 313. Stewart cites the views of many authors as to the international armed conflict (‘IAC’)/non-international armed conflict (‘NIAC’) distinction ‘frustrating’ the humanitarian purpose of the law of armed conflict (‘LOAC’) in the vast bulk of modern armed conflict situations: at 313.
 See, eg, UK Ministry of Defence, Manual of the Law of Armed Conflict (Oxford University Press, 2004) (‘LOAC Manual’). The LOAC Manual states, in relation to IAC, that ‘[t]he law will apply … provided an armed conflict is in fact in existence’: at 28–9 [3.2.3]; and, in relation to ‘internal armed conflicts’, that ‘[i]f the situation in a country amounts to an armed conflict, the law of armed conflict applies’: at 31 [3.5]. In relation to a particular conflict scenario, see, eg, Laurie R Blank and Benjamin R Farley, ‘Characterising US Operations in Pakistan: Is the United States Engaged in Armed Conflict?’ (2011) 34 Fordham International Law Journal 151, 160–2. Blank and Farley state that ‘determination of the existence of an armed conflict does not turn on … how the participants characterise the hostilities — but rather on the facts of a given situation’: at 161.
 Andrew Clapham, ‘Human Rights Obligations of Non-State Actors in Conflict Situations’  863 International Review of the Red Cross 491, 510.
 With respect to the more complicated IAC situation of occupation under Geneva Convention IV there is admittedly greater scope for states to legitimately challenge characterisation on the facts: Geneva Convention relative to the Protection of Civilian Persons in Time of War, opened for signature 12 August 1949, 75 UNTS 287 (entered into force 21 October 1950) (‘Geneva Convention IV’). For example, ongoing debate as to the nature of Israeli presence in the West Bank/Palestine. Another example is early United States and United Kingdom uncertainty, or perhaps obfuscation, as to their status as occupying powers in Iraq: for example, the US reaction to United Nations Secretary-General Kofi Annan’s reference to its status as an ‘occupying power’ on 24 April 2003: see, eg, Jonathan Fowler, ‘US Bridles as UN’s Kofi Annan Calls it “Occupying Power”’, Associated Press Newsfeed (online), 24 April 2003 <http://www.commondreams.org/headlines03/0424-06.htm> . This uncertainty continued until the joint US and UK letter to the UN Security Council (‘UNSC’), indicating that they would ‘strictly abide by their obligations under international law’ (Letter Dated 8 May 2003 from the Permanent Representatives of the United Kingdom of Great Britain and Northern Ireland and the United States of America to the United Nations addressed to the President of the Security Council, UN SCOR, UN Doc S/2003/538 (8 May 2003) 1) was referenced by the UNSC in support of US and UK acceptance of their formal status as occupying powers: SC Res 1483, UN SCOR, 4761st mtg, UN Doc S/RES/1483 (22 May 2003) 2. See also Evidence to the Iraq Inquiry, London, 28 January 2010 (Sir Michael Wood) <http://www.iraqinquiry.org.uk/media/44055/wood-statement-occupying-powers .pdf> Martin Zwarnenberg, ‘Existentialism in Iraq: Security Council Resolution 1483 and the Law of Occupation’  856 International Review of the Red Cross 745.
 Geneva Convention I; Geneva Convention for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of the Armed Forces at Sea, opened for signature 12 August 1949, 75 UNTS 85 (entered into force 21 October 1950); Geneva Convention relative to the Treatment of Prisoners of War, opened for signature 12 August 1959, 75 UNTS 135 (entered into force 21 October 1950); Geneva Convention IV (collectively, ‘Geneva Conventions’).
 Geoffrey Best, War and Law since 1945 (Clarendon Press, 1994) 174.
 Kenneth Watkin, ‘Stability Operations: A Guiding Framework for “Small Wars” and Other Conflicts of the Twenty-First Century?’ in Michael N Schmitt (ed), The War in Afghanistan: A Legal Analysis (US Naval War College Press, 2009) 411, 417.
 Ibid 420.
 Andreas Wenger and Simon J A Mason, ‘The Civilianisation of Armed Conflict: Trends and Implications’  872 International Review of the Red Cross 835, 841–2: summarising statistics compiled by the Uppsala Conflict Data Program (‘UCDP’). The UCDP analysis for 2009 records 29 ongoing intrastate armed conflicts, 7 ongoing internationalised armed conflicts, and 0 interstate armed conflicts: Department of Peace and Conflict Research, UCDP Conflict Encyclopedia (UCDP Database), Uppsala Universitet <http://www.pcr.uu.se/research/ucdp/database/> .
 See, eg, Kenneth Watkin, ‘Controlling the Use of Force: A Role for Human Rights Norms in Contemporary Armed Conflict’ (2004) 98 American Journal of International Law 1, 10; Blank and Farley, above n 4, 187. My own view is expressed in Rob McLaughlin, ‘The Legal Regime Applicable to Use of Lethal Force when Operating under a United Nations Security Council Chapter VII Mandate Authorising “All Necessary Means”’ (2007) 12 Journal of Conflict and Security Law 389; Rob McLaughlin, ‘The Use of Lethal Force by Military Forces on Law Enforcement Operations — Is There a “Lawful Authority”?’ (2009) 37 Federal Law Review 441.
 There are, of course, several other equally important aspects which would repay close attention. What tradeoffs between strategic policy and tactical needs must be considered? To what extent is the status and character of the party standing behind the adversary (or pulling the adversary’s strings) important in characterisation decisions? Does previous UN engagement on the ground in a conflict situation tend to drive characterisation decisions down a particular path?
 W Michael Reisman, ‘International Legal Dynamics and the Design of Feasible Missions: The Case of Afghanistan’ in Michael N Schmitt (ed), The War in Afghanistan: A Legal Analysis (US Naval War College Press, 2009) 59, 60.
 Dale Stephens, ‘Blurring the Lines: The Interpretation, Discourse and Application of the Law of Armed Conflict’ (2009) 12 Yearbook of International Humanitarian Law 85, 86.
 See, eg, James Zogby, ‘A Bungled PR Job Keeps the bin Laden Controversy Alive’, Al Arabiya (online), 8 May 2011 <www.alarabiya.net/views/2011/05/08/148269.html>.
 United Kingdom, Parliamentary Debates, House of Commons, 15 December 1993, vol 234, col 1071.
 Ibid col 1071–82.
 Joint Security Committee (UK), Conclusions of a Meeting of the Joint Security Committee Held on Thursday 4 February 1971 in Stormont Castle at 11 30 AM (PRONI Reference No HA/32/3/5, Public Records Office of Northern Ireland) <http://cain.ulst.ac.uk/proni/1971/proni_HA-32-3-5_1971-02-04.pdf> . For examples of the UK Government’s stress upon a law enforcement approach to the situation, see also Prime Minister’s Meeting with Home Secretary — Wednesday 4 February 1970 (PRONI Reference No CAB/9/G/91/2, Public Records Office of Northern Ireland) ,  <http://cain.ulst.ac.uk/proni/1970/proni_CAB-9-G-91-2_1970-02-04.pdf> .
 ‘Troubles “Not War” Motion Passed’, BBC News (online), 18 February 2008 <http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/7249681.stm> . See, eg, Northern Ireland Affairs Committee, The Report of the Consultative Group on the Past in Northern Ireland, House of Commons Paper No 171, Session 2009–10 (2009) annex (‘Evidence’) 57 (Written Evidence from Jim Allister) (citations omitted) (emphasis in original):
The use of the term ‘active service’ to describe terrorists engaged in murderous criminality in a press release issued by OFMDFM [Office of the First Minister and Deputy First Minister of Northern Ireland], when a member of the Victims’ Commission was discussing how her brother lost her [sic] life while attempting to murder, has already provoked huge outrage among innocent victims in Northern Ireland. Its use suggests that terrorists were military personnel engaged in a war rather than the reality — vile terrorists bent on murder and mayhem. I deeply regret that the CGP [Consultative Group on the Past] has, by employing the term, sought to sanitise the murders of Republican terror squads.
 Common Article 3 refers to art 3 which is common to all four Geneva Conventions.
 Charles Garraway, ‘Afghanistan and the Nature of Conflict’ in Michael N Schmitt (ed), The War in Afghanistan: A Legal Analysis (US Naval War College Press, 2009) 157, 164. Garraway notes that it is possible that ‘the fact that elements of the Irish Republican Army operated from across the border in the Irish Republic excluded the application of Common Article 3’.
 Pictet, above n 1, 39, on Common Article 3.
 Michael N Schmitt, ‘The Law of Armed Conflict as Soft Power: Optimising Strategic Choice’ in Michael N Schmitt and A Ralph Thomas (eds), International Law across the Spectrum of Conflict: Essays in Honour of Professor L C Green on His Eightieth Birthday (US Naval War College Press, 2000) 455, 456 (emphasis in original).
 Pictet acknowledges that the concept of ‘armed conflict’ in Common Article 3(1) was deliberately left undefined in order to assuage the significant objections of many states as to enumerating conditions or further detail. After outlining the types of conditions which various delegations had at one time or another suggested, he concluded that
[h]owever useful, therefore, the various conditions stated above may be, they are not indispensable, since no Government can object to respecting, in its dealings with internal enemies, whatever the nature of the conflict between it and them, a few essential rules which it in fact respects daily, under its own laws, even when dealing with common criminals.
Pictet, above n 1, 50 (emphasis added). However, as Laurie Blank and Benjamin Farley, considering both the Commentary IV and the International Criminal Tribunal for the Former Yugoslavia jurisprudence which has placed some flesh upon the concept, observe, none of the conditions elaborated in either source is ‘dispositive’: see Jean S Pictet (ed), Geneva Convention relative to the Protection of Civilian Persons in Time of War (International Committee of the Red Cross, 1958) (‘Commentary IV’); Blank and Farley, above n 4, 163–4.
 Protocol Additional to the Geneva Conventions of 12 August 1949 and relating to the Protection of Victims of Non-International Armed Conflicts, opened for signature 8 June 1977, 1125 UNTS 609 (entered into force 7 December 1978) (‘Additional Protocol II’); Yves Sandoz, Christophe Swinarski and Bruno Zimmermann (eds), Commentary on the Additional Protocols of 8 June 1977 to the Geneva Conventions of 12 August 1949 (International Committee of the Red Cross, 1987) 1354–6. ‘No real definitions are given. The concept of internal disturbances and tensions may be illustrated by giving a list of examples of such situations without any attempt to be exhaustive’: at 1354. Indeed, less-than-NIAC can cover situations where ‘there exists a confrontation within the country, which is characterised by a certain seriousness or duration and which involves acts of violence’: at 1355 (citations omitted).
 See, eg, Prosecutor v Tadić (Judgement) (International Criminal Tribunal for the Former Yugoslavia, Trial Chamber III, Case No IT-94-1-T, 7 May 1997) . The Tribunal, however, specifically noted that interpretation of factors affecting the threshold between NIAC and less-than-NIAC law enforcement should nevertheless refer back to the Commentary I which, as we have seen, is explicit to the effect that there are no dispositive conditions, and (tacitly) that a wide field for interpretive discretion thus exists at this lowest of armed conflict thresholds: see also Michael John-Hopkins, ‘Regulating the Conduct of Urban Warfare: Lessons from Contemporary Asymmetric Armed Conflicts’  878 International Review of the Red Cross 469, 473; Bellal, Giacca and Casey-Maslen, above n 3, 7–9; Gary D Solis, The Law of Armed Conflict: International Humanitarian Law in War (Cambridge University Press, 2010) 153.
 Constantin von der Groeben, ‘Criminal Responsibility of German Soldiers in Afghanistan: The Case of Colonel Klein’ (2010) 11 German Law Journal 469, 475.
 Michael N Schmitt, Charles H B Garraway and Yoram Dinstein, The Manual on the Law of Non-International Armed Conflict: With Commentary (International Institute of Humanitarian Law, 2006) 2–3.
 Geoffrey S Corn, ‘Making the Case for Conflict Bifurcation in Afghanistan: Transnational Armed Conflict, al Qaida and the Limits of the Associated Militia Concept’ in Michael N Schmitt (ed), The War in Afghanistan: A Legal Analysis (US Naval War College Press, 2009) 182, 184.
 Pictet, above n 1, 50.
 See, eg, Rogier Bartels, ‘Timelines, Borderlines and Conflicts: The Historical Evolution of the Legal Divide between International and Non-International Armed Conflicts’  873 International Review of the Red Cross 35.
 Ibid 43. See generally Diarmaid MacCulloch, A History of Christianity: The First Three Thousand Years (Penguin Books, 2009) 604–21. Luther’s views on this issue were traumatically shaped by the ‘Farmers’ War’ — a series of revolts across Central Europe in 1525, which in some ways owed its genesis to Luther’s radical advocacy of liberty, and his denial of several core elements of (Roman) Catholic theology, including the authority of the ecclesiastical hierarchy. ‘The revolts were brutally crushed — and Luther, terrified by the disorder, applauded the rulers’ brutality’: at 614.
 E de Vattel, The Law of Nations or the Principles of Natural Law: Applied to the Conduct and to the Affairs of Nations and of Sovereigns (Charles G Fenwick trans, Oceana Publications, 1964) bk III, ch XVIII [trans of: Le droit des gens ou Principes de la loi naturelle appliqués à la conduite et aux affaires des nations et des souverains] (first published 1758).
 W F Finlason, Commentaries upon Martial Law, with Special Reference to its Regulation and Restraint: With an Introduction, Containing Comments upon the Charge of the Lord Chief Justice in the Jamaica Case (Stevens and Sons, 1867).
 W F Finlason, Review of the Authorities as to the Repression of Riot or Rebellion, with Special Reference to Criminal or Civil Liability (Stevens and Sons, 1868).
 Diplomatic Conference Convened by the Swiss Federal Council for the Establishment of International Conventions for the Protection of War Victims, Geneva, 21 April – 12 August 1949 (‘1949 Diplomatic Conference’).
 Bartels, above n 33, 58–9.
 See generally Bartels, above n 33; Stewart, above n 3; Dawn Steinhoff, ‘Talking to the Enemy: State Legitimacy Concerns with Engaging Non-State Armed Groups’ (2009) 45 Texas International Law Journal 297.
 Stewart, above n 3, 343 (citations omitted).
 John Ip, ‘The Supreme Court and the House of Lords in the War on Terror: Inter Arma Silent Leges?’ (2010) 19 Michigan State Journal of International Law 1, 2–3.
 Daniel Abebe and Eric A Posner, ‘The Flaws of Foreign Affairs Legalism’  VirgJlIntLaw 1; (2010) 51 Virginia Journal of International Law 507, 518.
 See, eg, Eighth United Nations Congress on the Prevention of Crime and the Treatment of Offenders: Report prepared by the Secretariat, UN Doc A/CONF.144/28/Rev.1 (7 September 1990) annex (‘Basic Principles on the Use of Force and Firearms by Law Enforcement Officials’).
 Hans-Peter Gasser, ‘Acts of Terror, “Terrorism” and International Humanitarian Law’  847 International Review of the Red Cross 547, 566.
 See, eg, Gasser, above n 45, 560.
 Bartels, above n 33, 51.
 Gabriella Blum, ‘On a Differential Law of War’ (2011) 52 Harvard International Law Journal 163, 166.
 Stewart, above n 3, 321.
 Schmitt, above n 24, 459.
 For the publicly released portions of the Tactical Directive, see: International Security Assistance Force, COMISAF Tactical Directive (Revised) (6 July 2009) North Atlantic Treaty Organization <http://www.nato.int/isaf/docu/official_texts/Tactical_Directive_90706.pdf> . See generally Stephens, above n 15, 86.
 Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts, opened for signature 8 June 1977, 1125 UNTS 3 (entered into force 7 December 1978) (‘Additional Protocol I’).
 Stewart, above n 3, 342.
 United States Department of Defense, Final Report to Congress: Conduct of the Persian Gulf War (Report, April 1992) 721 <http://www.ndu.edu/library/epubs/cpgw.pdf> .
 Additional Protocol I art 1(4). See, eg, Steinhoff, above n 40, 318; Stewart, above n 3, 218.
 Prosecutor v Rajić (Review of the Indictment pursuant to Rule 61 of the Rules of Procedure and Evidence) (International Criminal Tribunal for the Former Yugoslavia, Trial Chamber, Case No IT-95-12, 13 September 1996) –. For example, at , the Trial Chamber determined that
[t]he materials described above constitute prima facie evidence … that units of the Croatian Army were present in central Bosnia during the period from late 1992 to March 1994 and that these Croatian Army troops were sent to Bosnia by the Croatian Government and were engaged, alongside the Bosnian Croat forces, in fighting against the forces of the Bosnian Government. There is therefore enough evidence to establish for the purpose of the present proceedings that, as a result of the significant and continuous military intervention of the Croatian Army in support of the Bosnian Croats, the domestic conflict between the Bosnian Croats and their Government in central Bosnia became an international armed conflict, and that this conflict was ongoing at the time of the attack on Stupni Do in October 1993.
 Stewart, above n 3, 229.
 Ibid 248.
 Stephens, above n 15, 100–1.
 See the US/NATO report (also known as the ‘Callan Report’) and the Human Rights Watch critique of that report: Memorandum from Michael W Callan, Brig Gen, USAF, to Acting Commander, United States Central Command, 1 October 2008; Human Rights Watch, ‘Afghanistan: US Investigation of Airstrike Deaths “Deeply Flawed”’ (15 January 2009) <http://www.hrw.org/news> .
 See, eg, Isayeva v Russia (2005) 41 EHRR 847.
 See generally Michael Kelly et al, ‘Legal Aspects of Australia’s Involvement in the International Force for East Timor’  841 International Review of the Red Cross 101; Dale Stephens, ‘The Use of Force in Peacekeeping Operations: The East Timor Experience’ (Monograph, Asia Pacific Centre for Military Law, 2005); Dale Stephens, ‘Military Involvement in Law Enforcement’  878 International Review of the Red Cross 453.
 Stephens, ‘Military Involvement in Law Enforcement’, above n 62, 462–3 (citations omitted).
 Australian Defence Force Military Law Centre, ‘Law and Military Operations in East Timor (UNTAET) Feb 2000 – May 2002’ (Report, Asia Pacific Centre for Military Law, 2003) 23 (citations omitted).
 Australian Defence Force Military Law Centre, ‘Law and Military Operations in East Timor Sept 1999 – Feb 2000’ (Report, Asia Pacific Centre for Military Law, 2003) annex E
(‘Defence Legal Office Policy Statement on the Application of the Fourth Geneva Convention of 1949 to Possible Deployment of the ADF to East Timor’) i .
 Ibid 4 .
 10 USC ch 47 (2012).
 John H Knox, ‘A Presumption against Extrajurisdictionality’ (2010) 104 American Journal of International Law 351.
 von der Groeben, Criminal Responsibility of German Soldiers in Afghanistan, above n 28, 470.
 Criminal Code Act 1995 (Cth) sch 1 (‘Criminal Code’).
 Ibid. Offences subject to this category of jurisdiction include making unwarranted demands (div 139), obstruction of Commonwealth public officials (div 149), criminal associations and organisations (div 390), and postal offences (div 471). Category D jurisdiction goes even further and applies the same broad scope as category C jurisdiction, but without the built-in category C ‘defence’ — that is, the offence applies to all persons, everywhere, regardless of whether the offence is also an offence under local law. This category of jurisdiction applies to certain offences relating to espionage and similar activities (div 91), terrorism (div 101), terrorist organisations (div 102), financing terrorism (div 103), theft from the Commonwealth (div 131) certain fraudulent conduct (divs 134–7), bribery (div 141) and torture (div 274).
 For a broad analysis of the domestic law which a number of militaries take with them on operations, see Rain Liivoja, ‘Service Jurisdiction under International Law’  MelbJlIntLaw 12; (2010) 11 Melbourne Journal of International Law 309.
 See, eg, James D Bindenagel, ‘Afghanistan: The German Factor’ (2010) 1 Prism 95; Timo Noetzel, ‘Germany’s Small War in Afghanistan: Military Learning amid Politico-Strategic Inertia’ (2010) 31 Contemporary Security Policy 486; ‘German Foreign Minister Calls Afghanistan “Armed Conflict”’, Deutsche Welle (online), 10 February 2010
<www.dw-world.de>; Siobhán Dowling, ‘New Evaluation on Afghanistan Long Overdue’, Spiegel Online (online), 2 November 2010 <www.spiegel.de/international>; Matthias Gebauer and John Goetz, ‘German Officer Defends Controversial Afghanistan Airstrike’, Spiegel Online (online), 2 November 2010 <www.spiegel.de/international>; Matthias Gebauer and John Goetz, ‘German Army Withheld Information from US Pilots’, Spiegel Online (online), 2 January 2010 <www.spiegel.de/international>; Constantin von der Groeben, ‘German Federal Prosecutor Terminates Investigation against German Soldiers with respect to NATO Air Strike in Afghanistan’ on EJIL:Talk! (29 April 2010) <www.ejiltalk.org>; Nicolas von Kospoth, ‘Political Hypocrisy in a Matter of Facts and Nomenclature’, Defence Professionals (online), 26 June 2009 <http://www.defpro.com/daily/details/341/> .
 For media coverage, see, eg, Bruce Ackerman, ‘Legal Acrobatics, Illegal War’, The New York Times (New York), 21 June 2011, 27; Matt Spetalnik, ‘White House Defends Legality of Action in Libya’, Reuters (online), 15 June 2011 <http://www.reuters.com> . For the Office of Legal Counsel’s advice, see Memorandum Opinion for the Attorney General by Caroline D Krass, Principal Deputy Assistant Attorney General, ‘Authority to Use Military Force in Libya’ (1 April 2011) <http://www.justice.gov/olc/2011/authority-military-use-in-libya.pdf> . For the publicly available equivalent UK advice, which is substantially slimmer, see ‘In Full: UK Government’s Legal Advice on Libya’, BBC News (online), 21 March 2011 <http://www.bbc.co.uk> . For the most part, the UK advice merely reiterates SC Res 1973, UN SCOR, 66th sess, 6498th mtg, UN Doc S/RES/1973 (17 March 2011).
 War Powers Resolution, 50 USC §§ 1541–8 (2012).
 See Evidence to Senate Committee on Foreign Relations, United States Congress, Washington DC, 28 June 2011 (Harold Hongju Koh) <http://www.state.gov/documents/organization/167452.pdf> .
 Krass, above n 74, 3.
 Evidence to Senate Committee on Foreign Relations, United States Congress, Washington DC, 28 June 2011, 6 (Harold Hongju Koh) <http://www.state.gov/documents/organization/167452.pdf> , quoting Letter from State Department Legal Adviser, Monroe Leigh, and Department of Defense General Counsel, Martin R Hoffmann, to Chairman Clement J Zablocki, 5 June 1975.
 Evidence to the United States Senate Committee on Foreign Relations, Washington, 28 June 2011, 9 (Harold Hongju Koh) <http://www.state.gov/documents/organization/167452.pdf> .
 ‘War’ was not the key legal term at issue in this debate: the OLC provided advice (with which the Department of State agreed) to the effect that US engagement in Libya did not reach the threshold of ‘war’ in terms of US Constitutional requirements (in that only Congress may declare ‘war’). In this context, ‘war’, as the OLC opinion noted, would ‘generally … be satisfied only by prolonged and substantial military engagements, typically involving exposure of US military personnel to significant risk over a substantial period’: Krass, above n 74, 8 (commitments akin to Vietnam and Iraq, as opposed to more ‘limited’ commitments such as Haiti, Bosnia, and Libya).
 Evidence to Senate Committee on Foreign Relations, United States Congress, Washington DC, 28 June 2011 (Harold Hongju Koh) <http://www.state.gov/documents/organization/167452.pdf> .
 Ibid 13.
 War Powers Resolution, 50 USC § 1544(b) (2012).
 The NATO approach to characterising its involvement in the Libya conflict is opaque. In one sense, it is clear that NATO consciously chose to use LOAC-based language to describe its involvement. See, eg, NATO, ‘Statement on Libya following the Working Lunch of NATO Ministers of Foreign Affairs with Non-NATO Contributors to Operation Unified Protector’ (Press Statement, 14 April 2011):
A high operational tempo against legitimate targets will be maintained and we will exert this pressure as long as necessary and until the following objectives are achieved:
1 All attacks and threats of attack against civilians and civilian-populated areas have ended;
2 The regime has verifiably withdrawn to bases all military forces …
3 The regime must permit immediate, full, safe and unhindered humanitarian access to all the people in Libya in need of assistance.
See also NATO, ‘Statement on Libya following the Working Lunch of NATO Ministers of Defence with Non-NATO Contributors to Operation Unified Protector’ (Press Statement, 8 June 2011):
By maintaining a high operational tempo and carrying out precision strikes against legitimate military targets, we have seriously degraded the ability of the Qadhafi regime to attack civilians and relieved the pressure on civilian populated areas such as Misratah.
See also NATO, ‘Press Briefing on Libya by NATO Spokesperson Oana Lungescu and with Wing Commander Mike Bracken, the Operation Unified Protector Spokesperson’ (Press Briefing, 17 May 2011):
Q [FT Deutschland]: Sorry, just to clarify the command-and-control centre. It’s not important, let’s say, what technical installations or whatever are inside, but what matters is who is inside this so-called command-and-control centre. It doesn’t have anything to do with technical installation. These people could meet each other at any time in a tent, in a desert or whatever. It’s just important who meets.
A [Spokesman]: No. The activity, the planning and delivery of attacks onto the civilian population by a command-and-control centre would make it a legitimate command-and-control target.
The people, as we’ve said before, it’s not about individuals, it’s about the activity that’s being conducted. And if that is delivering armament onto civilians that would make it a legitimate target.
However, this arguably logical LOAC-based understanding of NATO’s involvement was confused by regular lapses into the language of ‘self-defence’ as the operating paradigm justifying use of force at the tactical level. See, eg, NATO, ‘Press Briefing by NATO Spokesperson Oana Lungescu and Group Captain Geoffrey Booth, Operations Division, International Military Staff’ (Press Briefing, 25 March 2011):
Question: You said the Rules of Engagement is right of self defence but after you start enforcing the NFZ [no-fly zone] it is very likely that the aggression will come from the ground. At that point self defence means bombing the ground. So what difference will be between the coalition activities and the NATO activities. You will still be doing a NFZ plus.
[NATO Spokesman]: I am not sure that is quite correct. Basically, if you have a threat to you specifically from the ground, not talking here about whether pro Libya forces, pro-Ghadaffi forces or non pro-Ghadaffi forces, are fighting. What I am saying is if you are threatened by say a surface to air missile system that’s stopping you enforcing the NFZ, you have the right of self defence. If air to air anti aircraft guns attack you as an aircraft enforcing the NFZ, you have the right to respond. The NFZ does not give you the right to just go if you like, as you put it, bomb targets on the ground.
See also NATO, ‘Press Briefing on Libya by Carmen Romero, the NATO Deputy Spokesperson and Colonel Roland Lavoie, Operation “Unified Protector” Military Spokesperson’ (Press Briefing, 26 July 2011):
We are basically monitoring the situation and acting where it is required to prevent attacks on civilians or to prevent a build-up of military capabilities that could stop humanitarian aid, for example.
So we are not a party in that conflict, and we have no intent to be.
 Pictet, above n 1, 39 (on Common Article 3).
 Pierre Hauck and Sven Peterke, ‘Organized Crime and Gang Violence in National and International Law’  878 International Review of the Red Cross 407, 410.
 Pictet, above n 1, 60.
 Ibid 43–4.
 David Kennedy, The Dark Sides of Virtue: Reassessing International Humanitarianism (Princeton University Press, 2004) 273.
 See, eg, Wenger and Mason, above n 11, 836; Clapham, above n 5, 491–3; Sadia Tabassum, ‘Combatants, Not Bandits: The Status of Rebels in Islamic Law’  881 International Review of the Red Cross 121.
 See, eg, Steinhoff, above n 40, 308–16.
 Bellal, Giacca, and Casey-Maslen, above n 3, 32.
 John-Hopkins, above n 27, 493 (citations omitted).
 Letter Dated 23 April from the Secretary-General Addressed to the President of the General Assembly, UN GAOR, 53rd sess, Agenda Item 44, UN Doc A/53/928 (27 April 1999) annex (‘Guatemala: Memory of Silence’).
 Ibid 29 .
 Hauck and Peterke, above n 86, 410.
 See above n 17 and accompanying text.
 Assessment of Operations in Northern Ireland — 1 May to 1 Nov 71 (PRONI Reference No HA/32/2/51, Public Records Office of Northern Ireland)  <http://cain.ulst.ac.uk/proni/1971/proni_HA-32-2-51_1971-nd.pdf> .
 See above nn 17–18 and accompanying text.
 Clive Walker and Russell L Weaver, ‘A Peace Deal for Northern Ireland? The Downing Street Declaration of 1993’ (1994) 8 Emory International Law Review 817, 822–5.
 Kieran McEvoy, ‘Prisoners, the Agreement, and the Political Character of the Northern Ireland Conflict’ (1999) 22 Fordham International Law Journal 1539, 1574.
 Ibid 1540.
 Ibid 1571.
 Ibid 1542.
 Darryl Li, ‘A Universal Enemy? “Foreign Fighters” and Legal Regimes of Exclusion and Exemption under the “Global War on Terror”’ (2010) 41 Columbia Human Rights Law Review 355, 359–60.
 Charter of the United Nations ch VII.
 See above n 39 and accompanying text.