The first issue addressed by the panel was that of “naked” self-defence, a concept defined as “resorting to force in self-defence, but in ways in which the means and levels of force used are not part of an armed conflict, as a matter of the technical law of war. Those circumstances include self-defence uses of force against non-state actors, such as individual terrorist targets, which do not yet rise to the NIAC [non-international armed conflict] threshold”.⁴ In March 2010, US State Department legal adviser Harold Koh endorsed this notion when he stated that the legal standards of necessity, distinction and proportionality apply in this resort to self-defence.⁵ Anderson says Koh arrived at this via customary international law rather than from obligations stemming directly from the technical laws of armed conflict.⁶

This particular notion of “naked” self-defence has been met with criticism, both outside the symposium⁷ as well as during the panel discussion. For example, the point was raised that such a concept or interpretation of self-defence would be a misreading of international law and that the use of the self-defence paradigm does not mean that international humanitarian law (IHL) or international human rights law (IHRL) frameworks can be escaped. When state consent is questionable (e.g., Pakistan’s consent to drone strikes, where consent is not clear), self-defence might permit going past Article 2(4) of the UN Charter; however, it would not justify a disregard for IHL and IHRL—these frameworks remain applicable with their corresponding provisions (e.g., threshold for armed conflict and imminence). Additionally, this particular notion or interpretation of self-defence as the US purports to use, raised important questions regarding the roles and obligations of non-state actors, problems with state sovereignty, and the role of human rights within the armed conflict context as well as outside of it. One, perhaps controversial, remark was that the US may have felt as though it needed to use the law of armed conflict paradigm after 9/11 in order to detain and kill people who were not related to a conflict, and therefore the “war on al Qaeda” framework was implemented and complemented by this idea of “naked” self-defence in order to fulfil policy goals. This recalls Terry Gill’s opening remarks of the symposium regarding his fear that without a legal consensus, policymakers would act out of expediency.

One panellist expressed regret that concepts specifically related to jus ad bellum were continually being mixed with those under jus in bello, while the two paradigms are meant to be kept separate. In his view, that mixture convolutes the discussion.⁸ He opined that “naked” self-defence was purely theoretical—to really get to the crux of the matter, one must look at the facts on the ground to see whether IHL applies (i.e., whether there is an armed conflict).

Another issue is that regarding whether force can be used against terrorist suspects in anticipatory or pre-emptive self-defence before an armed attack has taken place; and, in the case of pre-emptive self-defence, even as a response to a persistent threat under which it is unclear when the attack will precisely take place but is unlikely to take place imminently.⁹ Under the Caroline doctrine, anticipatory self-defence is recognised but limited to those cases in which an armed attack is imminent.¹⁰ This is the case when “the necessity of that self-defence is instant, overwhelming, and leaving no choice of means, and no moment of deliberation”. This is the standard of imminence traditionally accepted in customary international law.

However, John Brennan, then Assistant to the US President for Homeland Security and Counter-Terrorism, has argued that there is increasing recognition by the international community that, when fighting terrorism, a more flexible understanding of “imminence” is appropriate.¹¹

The issue to be debated is at what point a terrorist threat is “imminent”. Brennan argues that a flexible approach must be taken towards the imminence concept in the terrorism context, for a terrorist threat differs from traditional conflicts.¹² On the one hand, al Qaeda does not have a traditional command structure, its members do not wear uniforms or carry arms openly, leading one to say they do not meet the requisite organisational criteria outlined in IHL. On the other hand, al Qaeda is capable of attacking unexpectedly and causing significant civilian and military damage, likening their acts to hostilities, which could be governed by IHL. According to Brennan, this calls for a broader possibility to strike against terrorists out of self-defence and therefore a broader concept of imminence.¹³

The US argues that a pattern of behaviour over several years can form an imminent threat when alleged terrorists had previously planned, conspired and perhaps acted in other places.¹⁴ One panellist opined that the US government has used this standard but defined it beyond recognition. What John Brennan has said, and what a recently leaked White Paper from the US Department of Justice reiterates, is that its use needs to be broad and flexible—but this has led to a situation of too broad and too flexible that is beyond any Caroline manifestation of imminence. Another panellist stated that the exception formed in Article 51 of the UN Charter is for an imminence likened to tanks massing on the border of one country aimed at another. In this panellist’s view, the post-9/11 approach has been to de-couple imminence from the idea that an armed attack would happen by supplanting it with the idea that great harm might happen and the concept of this occurring was so bad, the temporal element of self-defence must be relaxed—essentially for security reasons. The majority of the panellists agreed that when it came to the imminence requirement, a careful, measured and strict interpretation was advised in order to prevent attacks and reprisals.

The question becomes whether the broadening of the principle of imminence has, to an extent, indeed become recognised by the international community or if it is forecasted to do so in the (near) future. This segues nicely into the third main issue discussed in the panel: temporal considerations of armed conflict.

When trying to define the beginning or ending of an armed conflict, many challenges arise, especially regarding conflicts between a state and a non-state actor. Derek Jinks has outlined guidelines about the initiation and cessation of armed conflict as well as defining what an armed conflict actually is.¹⁵ He writes that regarding the initiation, international armed conflict (IAC) is more straightforward.¹⁶ In case of a NIAC, one can turn to Common Article 3 of the Geneva Conventions, which provides that in armed conflicts not of an international character, minimum standards apply to each party. One integral issue identified in the literature is that there is no authoritative definition of armed conflict. The International Law Association’s Use of Force Committee concluded a five-year study into this issue and delivered its findings in the 2010 Committee Report. In that report, the committee confirmed that at least two characteristics are found with respect to all armed conflicts: (1) the existence of organised armed groups that are (2) engaged in fighting of some intensity. In addition to these minimum criteria respecting all armed conflict, IHL includes additional criteria so as to classify conflicts as either international or non-international in nature.¹⁷

With regard to the end of the application of IHL, in both IACs and NIACs, the general rule is that IHL applies until the “general close of military operations”.¹⁸ However, the point at which that occurs, especially with regard to a conflict between a state and a non-state actor, is not clear. A peace treaty (though uncommon) is the clearest method of ending a conflict.¹⁹ But even in the absence of such a peace treaty, a complete cessation of hostilities and a de facto resumption of normal relations between the parties may occur.²⁰

During the symposium, one panellist remarked that it is never easy to draw a line at the beginning or the end of armed conflict but the correct law nevertheless must be applied.²¹ Another thought the relevant question was not necessarily about the beginning or end, but rather whether there is an IAC or a NIAC in any given situation.²² Yet another reiterated that the need for organisation criteria applies and though the question is difficult regarding when to apply IHL, it is an important one that should not be abandoned.²³ Finally, the discussion with respect to the temporal dimension refocused on the US, demonstrating that the US government cites situations that occurred pre-9/11 in its justification of this current conflict with al Qaeda, going back to 1996 with Osama Bin Laden’s fatwa.²⁴ Regarding the end of hostilities, some officials say “when Al-Qaeda has degraded to such an extent it is much more difficult to carry out an attack”.²⁵ The US and NATO plan to drawdown troops in Afghanistan by the end of 2014,²⁶ and it will be interesting to see how the rest of the conflict on-going between the US and al Qaeda morphs once that occurs. Without the most solid link until now to an existing IAC, it remains to be seen what kind of legal framework the US will try to use once the IAC with Afghanistan draws to a close.²⁷

The idea behind this panel was to discuss issues related to the geographic scope of armed conflict, such as: where can a war be fought? Where is the battlefield in an armed conflict (i.e., does it have a territorial scope tied to a nation state or a geographic region)? Is an armed conflict related to a “hot battlefield” or does the conflict follow the participants wherever they may go? Does the consent of a territorial state matter in the use of force when a member of a non-state actor group finds him/herself there? The use of analogy between IAC and NIAC was also explored. Three main issues were identified in the discussion during the symposium: (1) the implication of territorial state consent, (2) the issue of the “hot battlefield”, and (3) combatant status and location of hostilities.

Consent provided by the territorial state (state A) to the state seeking to use force (state B) against non-state actors present in state A precludes the violation of the sovereignty of state A.²⁸ In other words, the use of force by one state on the territory of another is allowed when that state thereto consents and the violation of Article 2(4) of the UN Charter’s prohibition on the use of force is precluded.

However, whether consent has been given can be difficult to determine. Posner sets out what he calls “coercive consent”.²⁹ The US justifies its use of drone attacks in Pakistan against terrorists there through Pakistan’s consent. Publicly and officially, however, Pakistan has opposed the use of drones on its territory.³⁰ Nonetheless, the US claims such consent was given, and infers further and continuing consent from the fact that the “Pakistani military continues to clear airspace for drones and doesn’t interfere physically with the unpiloted aircraft in flight”.³¹ To do otherwise would be risky, according to Posner, and Pakistan is not in the position to actually do something about the attacks.³² Recently, the UN Special Rapporteur on the Promotion and Protection of Human Rights and Fundamental Freedoms while Countering Terrorism, Ben Emmerson, conducted a country visit to Pakistan and after meeting with several government representatives concluded: “The position of the Government of Pakistan is quite clear. It does not consent to the use of drones by the United States on its territory and it considers this to be a violation of Pakistan’s sovereignty and territorial integrity”,³³ though this statement has been met with some critical reception.

A further issue then becomes whether consent can be inferred and what risks come with it. Regarding consent in international law, in the Armed Activities case before the International Court of Justice (ICJ), JudgeTomka stated that when the use of armed force is a lawful exercise of the right to self-defence,the force used falls outside of the scope of Article 2(4)’s prohibition. Andre de Hoogh remarks that “this view appears to mischaracterise the relationship that exists between Articles 2(4) and 51, which is one of general prohibition and justification. Any use of armedforce necessarily falls within the scope of the prohibition, more so when territorial integrity is interpreted to mean territorial inviolability”.³⁴ And he points to the Court’s conclusion that Uganda violated the prohibition, to illustrate that “Article 2(4) does not exclude certain specific armed measures or activities from its scope”.³⁵ Therefore, this kind of use of force needs some kind of justification (e.g., either self-defence or Security Council authorisation). This is only relevant when considering consent. As De Hoogh points out, “if the territorial State agrees to the use of armed force by another State on its territory, e.g., to suppress armed bands or pursue terrorists, there will not be a violation of Article 2(4) because such force will not be against the territorial integrity or political independence of the former State, nor inconsistent with the purposes of the United Nations”.³⁶

During the symposium, the case-by-case basis approach regarding classification of conflict was a theme that returned to this particular topic in order to assess the type of conflict, and analyse where it is occurring rather than trying to assign a “blurry” concept of a battlefield to it. It was posited that the concept of consent in IHL takes away the possibility of classifying the conflict as an IAC, but then the intensity and organisation of the parties must be examined in order to see whether a situation rises to the threshold of a NIAC—where IHL would govern—or not—where IHRL and the law enforcement paradigm would be more appropriate. In conclusion, it was offered that it is not the consent, but rather the facts on the ground that determine the classification of an armed conflict.

Regarding the issue of the “hot battlefield”, John Brennan has addressed the US’ position regarding the geographic scope of armed conflict. He made reference to al-Qaeda’s leadership base as being in Pakistan and that the “affiliated forces” are “in places like Pakistan, Yemen, and countries throughout Africa”.³⁷ This serves to concretise some ideas about how the US perceives the battlefield (i.e., naming particular countries where operations have already happened or may yet occur). Brennan directly addressed the geographic scope by stating that the US was not “restricted solely to ‘hot’ battlefields like Afghanistan” as the armed conflict with al-Qaeda allows the US to use force against these non-state actors under a self-defence regime, but without the requirement to do “a separate self-defence analysis each time”.³⁸

One panellist remarked that with non-state actors, it is easy for conflicts to spread and therefore we should link the physical footprint (i.e., where the non-state actors find themselves) to the on-going conflict.³⁹ Panel discussion further identified the fact that there is a widespread use of the term “hot battlefield” without it even being mentioned in the Geneva Conventions. When describing the permissible boundaries, one panellist remarked that the question really revolved around the distance allowable from the hot battlefield and thought that a training facility located, for example, in Somalia, ought to be legitimately classified as part of the battlefield, mainly based in using the law of neutrality by analogy.⁴⁰

Another issue that arose relates to the status of the combatant and the location of hostilities. The crux of the issue is whether the conflict follows a participant wherever he may be found. The US position is that killing suspected members of al-Qaeda in today’s conflict is, by analogy, just as legally defensible as killing Japanese General Yamamoto in the Second World War.⁴¹ “For the United States (and others that adopt this position), once a state is in an armed conflict with a non-state armed group, that conflict follows the members of that group wherever they go, as long as the group’s members continue to engage in hostilities against that state (either on the ‘hot battlefield’ or from their new location.)”.⁴² This has been challenged by some scholars because applying IAC standards in a NIAC by analogy is inherently problematic.⁴³ Additionally, according to the International Committee of the Red Cross (ICRC), the US theory that “a person directly participating in hostilities in relation to a specific ongoing NIAC ‘carries’ that armed conflict with him to a non-belligerent state by virtue of continued direct participation (the nexus requirement) and remains targetable under IHL” is a novel view in contrast with the underlying object and purpose of the Geneva Conventions.⁴⁴ This is demonstrated in the same ICRC report, given that the legal expansion of this theory allows for an application of the rules governing the conduct of hostilities to a globally limitless battlefield.⁴⁵

In discussion, one panellist found the idea that the conflict follows the participant very problematic given that this construct was not envisioned by the Geneva Conventions or the subsequent protocols. Going back to Common Article 3 of the Geneva Conventions (which states that each party to the conflict shall be bound to apply, as a minimum, certain provisions, “[i]n the case of armed conflict not of an international character occurring in the territory of one of the High Contracting Parties”), he argued, it says “in the territory of one of the High Contracting Parties” not two or three. Additional Protocol II stipulates “in the territory of a High Contracting Party” which is slightly more flexible. In looking at the Afghanistan/Pakistan situation, spill over is geographically problematic. The panellist urged the audience to look at attacks on a case-by-case basis, so those in Pakistan remained separate from those in Afghanistan. The next step is to then ask the requisite questions: is this a new IAC? If there is consent from the territorial state (see discussion on consent on pp. 7–8), this might make a difference in classification of the conflict (NIAC vs. IAC), but again, it is not the consent, it is the facts on the ground that determine the decision involving classification of the armed conflict.

Another panellist offered three relevant different operational lenses to view the issue through: prospectively (where can a soldier go?), real-time (where can a soldier find his/her enemy?) and post hoc accountability (what law applies?), while also emphasising that the answers to these questions may depend on who is trying to answer them. A military perspective might say that a global battlefield is preferable—an easy answer, but not necessarily the right one. The level of threat perceived will also have a bearing on what decisions are made, as you cannot divorce law from policy. Looking to the fundamentals of the law can be helpful when IHL is not clear in that lawyers must balance principles of IHL.⁴⁶ One conclusion may be that “geographical limitations cannot be found on a map” as one panellist suggested. Another panellist reiterated an idea already discussed: the real question is whether there is an IAC or a NIAC, and there is a need for careful analysis on a case-by-case basis, rather than an overly broad application of IHL across the entire globe. If there is a NIAC, is it indigenous and where is the NIAC located? Only once you have answered this can IHL be applied, and it is applied, geographically, across the entire territory of hostilities.⁴⁷ Yet another panellist thought that it was not up to IHL to govern the geographical scope of armed conflict but that this was rather a jus ad bellum question about where force was being used and whether the amount of force rose to the requisite intensity and organisation, in which case IHL follows the hostilities rather than determines them.

Regarding the notion of status, in a NIAC only those who directly participate in hostilities (DPH) might be targeted under international law. The ICRC published a study on guiding the interpretation of DPH in 2009 and posits that DPH implicates “individual (civilian) involvement of a person in hostilities (i.e., the resort by the parties to the conflict to means and methods of injuring the enemy)”.⁴⁸ But it does not stop there. In this interpretive guidance, the conclusions drawn about civilians taking direct part in hostilities have the following three constitutive elements: (1) a threshold of harm must be reached, (2) there must be direct causation by the direct participant to have reached this harm, and (3) there must be a belligerent nexus between one party causing the harm to another party to the conflict.

In terms of other statuses assigned to individuals involved in hostilities, it must be said that membership within a particular group, such as al-Qaeda, cannot be based on “abstract affiliation, family ties, or other criteria prone to error, arbitrariness or abuse”.⁴⁹ Instead, it must depend on whether one’s “continuous function corresponds to that collectively exercised by the group as a whole, namely the conduct of hostilities on behalf of a non-State party to the conflict”.⁵⁰ This continuous combat function role is crucial to distinguishing those who fight in an organised armed group from those who directly participate in hostilities on a merely spontaneous, sporadic or unorganised basis, or who assume exclusively political, administrative or other non-combat functions.⁵¹

When there is an armed conflict, the question arises about which law is applicable. This raises additional questions such as the extraterritorial applicability of certain human rights treaties, principles of non-derogation, and ideas about whether IHL is the lex specialis and therefore always controls within an armed conflict or if there can be concurrent application of IHRL in certain situations or times where IHRL fully controls in an in bello framework. Two main issues were explored in-depth in the panel session: (1) IHL as lex specialis or IHRL as lex generalis in counter-terrorism, and (2) the complementarity of IHL and IHRL via the discussion of the killing of Osama Bin Laden. However, as the issues were handled with such synchronicity, the overlap dictates that they are handled concurrently in this paper.

In terms of the US conflict with al-Qaeda, Ohlin has set out the discussion in US federal courts on the application of the Authorization to Use Military Force (AUMF) by Congress for the President to fight the war on terror.⁵² The Justice Department of the Obama Administration claims that the AUMF should be interpreted in a manner consistent with international law, that the AUMF gives the President authority in accordance to what is allowed under international law and IHL specifically. Within the support of this view another disagreement is apparent: there are those that claim that the AUMF should be interpreted both in light of IHL and IHRL. For Ohlin this is an impossible position because IHL is a lex specialis “in the sense that it displaces other bodies of law, including domestic criminal law, with a set of radically different norms based on reciprocity, namely that combatants can kill each other with impunity but must protect civilians and others hors de combat”.⁵³ He claims further that the ICRC is of the opinion that both IHL and IHRL can apply at the same time based on the Israeli Supreme Court’s Targeted Killings case. Ohlin dismisses this by pointing to the fact that there is no other precedent.⁵⁴

In response, Gabor Rona states that international jurisprudence accepts the logic and necessity of applying IHRL in times of armed conflict, while the explicit terms of both instruments are in accordance with each other.⁵⁵ Ohlin’s response was that the rule exclusion means that if there is an applicable rule of IHL on a specific issue, then IHL applies and IHRL does not,⁵⁶ pointing again to the Israeli Supreme Court Targeted Killings case.⁵⁷ The Court there applied a rule of IHL that allows the targeting of civilians taking DPH. However, the Court went further and concluded that a civilian taking DPH cannot be attacked when other less harmful means can be employed. The Court seems then to be reading IHRL norms on proportionality into the IHL proportionality norm applicable in that case. To Ohlin this kind of co-applicability, where IHRL and IHL apply to the same rule, is strange because, in his view, IHL is lex specialis and it always displaces IHRL when there is overlap in armed conflict situations. Concluding, he agrees that there exist lacunae in the relative scope of IHL and IHRL application to be filled by international law, but this is no reason to scrap the idea of concurrent application altogether.

The killing of Osama Bin Laden in May 2011 by US Navy Seals raised issues on the applicability of IHL and/or IHRL standards. The issue was whether the legality of the killing depended on the question whether Bin Laden could have been captured through non-lethal means rather than killed and if that would have been the preferred measure.⁵⁸ Starting from the premise that both IHL and IHRL apply to this killing, Milanovic sets out in his article the relationship between both legal regimes. Under IHL, targeting takes place on the basis of status, meaning that Bin Laden, either as a combatant or a civilian taking DPH, could be attacked at any time while the status persists as long as he is not hors de combat. Under IHL there is no necessity requirement for attacking a target that has such a status so there is no obligation to first use non-lethal means or to capture or detain before going for the kill. This is different under IHRL, which proscribes the use of non-lethal means primarily, and only if those means are not practically feasible, can lethal use of force be lawful. Depending on the facts, the killing of Osama Bin Laden could be lawful where the risk to the life of others, including that of the US soldiers, in attempting to capture him alive and the risk of escape outweighed his right to life. The killing would not have been lawful only on the basis that it was vastly easier to kill him than to capture and prosecute him.⁵⁹

The real disagreement now lies in how norms of IHL and IHRL should interact.⁶⁰ Models of co-application exist according to Milanovic. One is based on the Nuclear Weapons Advisory Opinion of the ICJ and would be that any IHL-compliant taking of life is by definition not arbitrary for the purpose of Article 6 of the International Covenant on Civil and Political Rights (ICCPR).⁶¹ IHL is then used to interpret Article 6 ICCPR as a norm of IHRL, though other IHRL norms do not leave such interpretative space.

Another model asks the question whether IHRL sets additional requirements for the lawfulness of a killing under IHL and whether these requirements can be less stringent than those developed in human rights jurisprudence in and for times of peace.⁶² Milanovic would cautiously answer yes to both questions. The question whether we can expect militaries to abide by more stringent and humane rules than what is strictly necessary under IHL depends on treaty interpretation. The Israel Supreme Court Targeted Killings decision is an example where the Court held that a capture-before-kill requirement of IHRL was needed because of the degree of control the Israeli military exercises over the occupied territories.⁶³ This is the preferable approach according to Milanovic because it shows a reflection of the object and purpose of IHL and IHRL treaties in their best light, while at the same time reflecting the demands of universality of human rights and practical considerations of effectiveness. The question remains: how far should IHL allow IHRL into its domain without compromising itself?⁶⁴

During discussion at the symposium, one panellist stated that the international legal community has reached the point in 2013 that the answer the ICJ gave in its 1996 Nuclear Weapons Advisory Opinion is no longer sophisticated enough. In IHRL, killing is a last resort. IHL is actually about killing people in order to win battles and wars and there are cases where the two bodies of law will contradict. The killing of Bin Laden is one prime example. If we assume this killing took place within an IHL conflict, and that he was targetable because of his status (e.g., continuous combat function), he can still be killed. In IHRL this is not the case. Another panellist stated that the lex specialis has always been a rule of interpretation to establish priorities enacted by the same legislator, which is not the case for these two distinct sets of laws. Yet another wondered what the “gaps” were in IHL. If they are present, would they be filled with IHRL? These questions and more were analysed and discussed before concluding that more work must be done with the development of the application of IHL and IHRL.

Against these discussions and theoretical background, this panel looked at a specific and concrete case study on targeted killings and drones.

First of all, it was observed that drones, as a weapons system, are not inherently unlawful: they are not indiscriminate and do not cause unnecessary suffering. In fact, drones could be more precise as they can stay longer in the air surveilling the target and can gather more information before attacking the target. Their accuracy was also mentioned, although this point was also questioned.⁶⁵

However, their use in specific situations may be unlawful and needs to be considered on a case-by-case basis. The three main problems identified with respect to targeted killings and drones concern (1) the jus ad bellum, (2) the personal dimension (i.e., the person that is being targeted)—again something that has to be considered on a case-by-case basis, and (3) accountability and transparency.⁶⁶

Concerning the topic of the jus ad bellum, the point was made that the US is blurring the lines between the jus ad bellum and the jus in bello when it uses both paradigms at the same time (“we are in an armed conflict and