The Laws of War: Under Siege or Gaining Ground?
The Laws of War: Under Siege or Gaining Ground?
Ten Conflicts to Watch in 2024
Ten Conflicts to Watch in 2024
Speech / Global 20+ minutes

The Laws of War: Under Siege or Gaining Ground?

Speech by Louise Arbour, President & CEO of the International Crisis Group, on the occasion of the Kirby Lecture at the Australian National University, Canberra, 23 June 2011.

I am delighted to have been invited to deliver this lecture, named in honour of Michael Kirby, a great jurist, judge, humanitarian, human rights defender, friend, and, in the judicial parlance that we once shared, I could have said brother.[fn]I wish to thank Jon Shephard for his research assistance and insights in the preparation of this paper.Hide Footnote I am also particularly happy to be doing so in the presence of the distinguished Chancellor of The Australian National University, a great politician, policy guru, trail blazer, lawyer malgré lui, my friend Gareth Evans. The fact that I am here, connected to both of them, allows me to use my favourite word in the English language, a word that has no real French equivalent: serendipity.

My connection to Michael brings me back to our years on the bench, geographically far apart, but I like to think united by a shared vision of the law and the judicial function. Empowered by a judicial mandate anchored in the Canadian Charter of Rights and Freedoms, I lived in an era denounced by some as having encouraged the legalisation of politics, celebrated by others as having allowed the rule of law to truly flourish. Either way, where I come from, the law mattered.

My connection to Gareth reflects the latest chapter of my entry into the world of armed conflict. Not surprisingly, I came to it from a legal perspective, but under his influence I broadened my horizons to embrace the murkier waters of political analysis, particularly after I took over from him the direction of International Crisis Group. So it is today through this mostly political lens that I want to share with you a few observations on the relevance, or lack thereof, of the law of armed conflict. It is with no false modesty that I disclaim any sophisticated expertise in the current state of that body of law. What I once knew is dated, and after a few years away, I don’t even trust the skills I once had to figure it out. But I am interested not so much in the state of the law as in its role and impact in the prevention, management and resolution of armed conflict.

Allow me then to start with a recent statement by one who can claim to have some insight into the politics of it all. The laws of war are increasingly outdated, and threaten to “handcuff all states fighting terrorism”.[fn]‘PM Netanyahu addresses the Saban Forum’, Israel Ministry of Foreign Affairs, 15 November 2009.Hide Footnote  So claimed Israeli PM Benjamin Netanyahu shortly after the publication of the Goldstone Report, which sharply criticised Israel for alleged violations of humanitarian law. Netanyahu is not alone in his criticism. In May of this year, Sri Lanka’s representative to the UN told the Security Council that the “one-size-fits-all” approach to humanitarian law fails to take into account the changed nature of today’s conflicts and imposes unfair burdens on states struggling to rid themselves of terror and insurgency.[fn]Security Council meeting 6531, 10 May 2011, p. 31.Hide Footnote  This in the wake of the bloody end to Sri Lanka’s civil war, which saw an estimated 40,000 civilians killed, many by government forces, in just a few months.

Israel and Sri Lanka are far from being the only critics of humanitarian law. In recent years, a growing number of voices have questioned the contemporary applicability of the laws of war. Citing, amongst other things, the failure of terrorist and insurgent groups to abide by the Geneva and Hague Conventions, the increasingly unclear line between civilians and combatants and the use of what has been termed “lawfare” to challenge states’ military options, they have called for a radical rethink. Another group of critics has taken the opposite line: far from tying states’ hands, they argue, the lex specialis of humanitarian law paves the way for states to inflict huge civilian casualties with impunity. Moreover, they suggest, even those aspects of the law that could provide some protection regularly go unheeded, and violators go unpunished.

The debate has been largely fuelled and mediatised by the politics of the War on Terror. But it is also rooted in the renewed relevancy of a body of law that, until the mid 1990s, was largely unimplemented, invisible, and the preserve solely of normative discussions amongst specialists. The growth of international criminal law, first in the launch of the ad hoc tribunals for the former Yugoslavia and Rwanda, then in the daring moves of national magistrates operating under the framework of universal jurisdiction, and finally in the creation of the International Criminal Court, have made Milošević, Pinochet and Bashir if not household names, at least the recognisable faces of international crime. In my view, the recent criticisms mounted against the laws of war reflect less their inadequacy than their increased effectiveness. And, in addressing below the emerging concept of lawfare, I will further suggest that implementation comes as much from compliance as from coercion, and that this is as it should be.

There are undoubtedly gaps and ambiguities in humanitarian law which need to be addressed; this much even the International Committee of the Red Cross has recognised. But the Geneva Conventions remains sound, and should not be too quickly dismissed. Critics allege that international law fails to take into account many issues pertinent to today’s conflicts; in fact, however, many of these issues are expressly addressed – just not in the way critics would like.

Current criticisms of the laws of war come in the context of significant changes in the conduct of conflict. Most humanitarian law was designed at a time when armed conflict usually meant a declared war between two or more states with identified territories and armed forces; today, such conflict is the exception, rather than the rule. Today’s conflicts are usually internal, and often involve one or more amorphous non-state actors. These changes have provoked a number of intersecting and cross-cutting criticisms, which go both to the scope and the substance of humanitarian law. The common thread of these is that the laws of war, at least as currently constituted, are no longer valid. I will attempt to address, very summarily, the core criticisms. They relate, largely, to the definition of armed conflict, the fundamental principles of distinction and proportionality, the issue of lawfare and, to a lesser extent, the notion of reciprocity.

Armed Conflict

International humanitarian law is a specialised body of rules and principles activated only when there is an armed conflict. When it does apply, it imposes a framework that is at the same time both more permissive for, and, some suggest, more constraining on state action. I think it is important to first test this premise. In the current context, the debate is essentially focussed on the question of protection of civilians. The body of law we are interested in offers, essentially, a quid pro quo: it requires the protection of civilians as an exception to a granted right to kill. In contrast, domestic criminal law, if compliant with international human rights standards, would prohibit extra judicial executions and restrict the right of law enforcement officers to kill only in self-defence, or in the reasonable use of force in performing an arrest, for instance.

Simply put, the law of armed conflict contains exceptions to the right to kill, while domestic law enforcement contains exceptions to the right to life. However, both are governed by non-excluded laws of general application, such as the prohibition of genocide or the absolute prohibition of torture. This point is important because the arguments about the special needs of the war on terror often seem to assume that it can only operate in a legal black hole; it is not only IHL that is seen as too constraining, but also a large part of international human rights law, including the Torture Convention and the Convention of the Rights of the Child, for example, which imposes a special regime for the treatment of child soldiers, not to mention a good part of domestic criminal law.

With that understood, the question still arises as to when there is an armed conflict sufficient to activate the laws of war. For international armed conflicts, the situation is relatively clear. The issue is more complex, however, in relation to non-international – or, more accurately, non-interstate – armed conflict. Additional Protocol II to the Geneva Conventions, which governs such conflicts, has a high threshold for application: parties must have, inter alia, territorial control, formal organisation and responsible command.[fn]Additional Protocol II to the Geneva Conventions of 1949, Article 1(1).Hide Footnote  Even Common Article 3 to the four Geneva Conventions, which is designed to apply as broadly as possible, requires, according to the influential Pictet Commentary, elements of formal structure and territorial control. Perhaps the most useful guidance on this issue is found in the ICTY’s Tadić decision, in which it suggested that the two operative conditions for determining the existence of an armed conflict are (i) the intensity of the conflict and (ii) the organisation of the parties.[fn]Tadić jurisdiction decision, ICTY Case No. IT-94-1-T, 1995, paragraph 70.Hide Footnote  But even this leaves much room for dispute.

Ultimately, no overarching determination as to whether the laws of war should apply to fights against rebels, terrorists or insurgents will ever be possible: a case-by-case approach will always be necessary. Certain episodes in the ‘war on terror’, for example, might reach the level of armed conflict, as might the repression of new illegal armed groups in Columbia or the military action undertaken by the government of Mexico against organised criminal gangs, despite the warning in the Tadić decision that “banditry, unorganised and short-lived insurrections, or terrorist activities … are not subject to international humanitarian law”.[fn]Tadić trial judgment, ICTY Case No. IT-94-1-T, 1997, paragraph 562.Hide Footnote

One must be wary of calls for an expansion of the application of the laws of war. As indicated before, the rules applicable during armed conflicts are far more permissive than those applicable during times of peace – some stringent human rights standards are displaced by the far less strict provisions of humanitarian law.

The laws of war are a compromise, designed to offer a minimum level of protection to non-combatants, and to combatants who have laid down their arms, in terrible circumstances. The price for the protections offered is the recognition of a circumscribed right for states to kill, to wound, to imprison without trial and so on. To extend this waiver to all situations where violence is used would be to distort both the spirit and the letter of the law. On the other hand, when a true situation of armed conflict exists, as it clearly did in the last few months of the battle between the government of Sri Lanka and the LTTE in early 2009, to characterise it instead as an internal campaign against terror in an attempt to escape the requirements of distinction and proportionality in the law of armed conflict is, at its core, a call for lawlessness.

Regardless of the justification for war in the first place (jus ad bellum), the laws of war (jus in bello) require that states respect the rights of enemy combatants, limit the damage inflicted by their forces and take steps to preserve the lives of civilians, even as their opponents flout those same rules to gain a military advantage. This lack of reciprocity, in particular, is often advanced as an argument for rejecting the applicability of this body of law. This argument must be rejected. Non-state actors are bound by legal rules, whether under domestic law or, arguably, under international customary law. And humanitarian obligations undertaken by states by treaty are not dependent on reciprocity, as they would otherwise be eviscerated. If one violates legal obligations, the answer is in the enforcement of the law – through personal criminal responsibility for example – not in the lifting of all rules for all sides. I will come back to this idea under my discussion of the concept of “lawfare”.

At the heart of current criticisms lies the issue of civilian protection, governed by the principles of distinction and proportionality. In modern conflicts, critics argue, these principles are simply untenable.

Let me now turn briefly to these two legal requirements.


Essentially, the law permits belligerents to attack each other, but not civilians. Article 51 of the First Additional Protocol to the Geneva Conventions states that “ … the civilian population as such, as well as individual civilians, shall not be the object of attack … ” The rule also applies to non-international armed-conflicts, by virtue of Article 13(2) of the Second Additional Protocol. Under the terms of the Protocols, civilians are those persons who are not members of State armed forces or of organised armed groups belonging to a party to an armed conflict. Such civilians are protected against direct attack unless and until they “directly participate” in hostilities. The issue becomes even more complex under the principle of proportionality which expands the protection of civilians beyond the prohibition against direct attacks, as we will see below.

Detractors argue that the line between civilians and combatants is increasingly blurred, as civilians lend their support to insurgents and rebels on the one hand, whilst states make increasing use of civilian and military contractors on the other.[fn]See, for example, Gross, M, Moral Dilemmas of Modern War: Torture, Blackmail and Assassination in an Age of Asymmetric Conflict (CUP, 2009: Cambridge), Ch. VII.Hide Footnote  The formulaic approach of the Geneva Conventions – which envisages a clear distinction between members of standing armies or militias and civilians – is, they argue, anachronistic and confusing. Today’s conflicts involve a spectrum of actors from ‘civilians’ to ‘combatants’, by way of terrorists, insurgents, political sympathisers, supporters and private contractors. This complexity means that it is increasingly difficult for parties to conflict to determine who may, or may not, be legitimately targeted.

In particular, tactics adopted by guerrilla and terrorist groups – using human shields, dressing as civilians, blending into the general population – have driven some states to claim that the principle of distinction unfairly benefits non-state groups. As a result, some have sought to expand the category of legitimate targets to include civilians even when not directly participating in hostilities. During the 2009 Gaza War, for example, an IDF spokesman declared that “[o]ur definition is that anyone who is involved with terrorism within Hamas is a valid target,” a clear departure from the current standard based on direct participation on hostilities.[fn]‘Gaza conflict: Who is a civilian?’, BBC, 5 January 2009.Hide Footnote

Criticism of the current system undoubtedly carries some weight. The approach to determining rights and obligations under Convention law places too much weight on considerations of status and group membership, and not enough on a functional test focussing on the activities of the individuals or groups involved. The formal requirements for combatant status laid out in Article 4 of the Third Geneva Convention – which include openly bearing arms and wearing a uniform – are simply no longer appropriate to today’s conflicts. But this does not justify abandoning the principle of distinction.

The law recognises that anyone who is directly participating in hostilities is a legitimate target for attack, regardless of their broader status. The ICRC has recently produced guidance on this issue, which holds that an individual is so participating whenever their actions are “likely to adversely affect … military operations”, where there is a direct causal link between their actions and the resulting harm, and where the harm is intentional.[fn]Nils Melzer, Interpretive Guidance on the Notion of Direct Participation in Hostilities Under International Humanitarian Law (ICRC, 2009: Geneva), p. 16.Hide Footnote  As a result, insurgents and terrorists are not protected as civilians when they are preparing for, carrying out, or returning from attacks. The fact that such actors recover their protection as civilians when they are no longer participating in hostilities does not mean that action cannot be taken against them through usual law enforcement methods.

Undoubtedly, debate will continue as to who really warrants protection as a civilian. This is very welcome: discussion and argument are the best ways to develop our understanding of the law. Guidance from NGOs, academics and international tribunals will help to develop and refine our understanding over time. But the fundamental principle of distinction remains valid. The current formulation does allow for the targeting of ‘civilian combatants’ when they are engaged in hostilities. To expand humanitarian law to allow the targeting of those civilians not directly involved in hostilities would be a dangerous step, and would entirely undermine the rationale of civilian protection.

I should add, in passing, that the expression “innocent” civilians confuses the debate. There is no requirement of innocence in the determination of status or function. But the use of the expression is revealing. When fighting rebels or insurgents, for example, it is very tempting for states to expand their view of the enemy to not only enemy combatants but also to political sympathisers or passive supporters or even, ultimately, to all members of the community that the combatants purport to represent. It is in that sense that the state would not consider these civilians “innocent” civilians, and would therefore prefer that they be denied protection.


Proportionality is the second vexing question in the law of modern internal armed conflict. Proportionality has two distinct aspects. The first belongs to the jus ad bellum. It requires that even when acting in self-defence, a state should use a measured, proportionate military response. It is the second aspect that concerns us, as it relates to the conduct of hostilities (jus in bello). Under Article 51 of the First Additional Protocol, an otherwise legitimate target may not be attacked where harm to civilians would be “excessive in relation to the concrete and direct military advantage anticipated.” Whilst no equivalent provision exists for non-international armed conflicts, relevant authorities all hold that the principle nonetheless applies, as it forms part of customary law.

There is great scope for argument as to what constitutes an “excessive” number of civilian casualties in any given situation. On the one hand, one can argue that the standard is purely a proportional one: civilian casualties could be, theoretically, unlimited, provided that the military advantage was suitably important. Sir Christopher Greenwood, for example, argues that

“the crucial question is whether other weapons or methods of warfare available at the time would have achieved the same military goal as effectively while causing less suffering or injury.”[fn]Greenwood, C, Command and the Laws of Armed Conflict (SCSI, 1993: Camberley), p. 24.Hide Footnote

On the other hand, the reference to “excessive” casualties might be taken to impose an absolute limit, regardless of the value of the military target. This is the view taken by Pictet Commentary, which holds that “incidental losses and damages should never be extensive.”[fn]ICRC Commentary to the Additional Protocols to the 1949 Geneva Conventions, p. 626.Hide Footnote  Both of these approaches have some merit, and discussion as to the correct standard will long continue.

A troubling trend in recent years, however, has been the proliferation of more extreme views regarding the requirement of proportionality. At one end of the spectrum are those who argue that any sort of proportionality is overly restrictive and should be done away with. In the aftermath of the 2006 Israel-Lebanon war, Israeli former general and current National Security Adviser Yaa’kov Amidror proclaimed that

“[n]ow those around Israel understand that Israel has certain red lines and that if these lines are crossed, Israel’s retaliation will be intentionally disproportionate. As a small country, we cannot allow ourselves the luxury of reacting proportionally.”[fn]Amidror, Y, ‘Lebanon II – the Fallout’, Jerusalem Post, 17 January 2007.Hide Footnote

The defeat of the LTTE in Sri Lanka – with its extraordinarily high civilian cost – is perhaps the most appalling example of what happens when states abandon proportionality. The killing of 40,000 civilians in the last few months of the Sri Lankan conflict shows what can happen when all thoughts of proportionality are abandoned. Whatever definition of “excessive” one adopts, it is beyond question that slaughter on this scale, inflicted through the indiscriminate shelling of supposed safe zones, the targeting of hospitals and the post defeat execution of prisoners, is out of all proportion to any potential military gain. And yet the Sri Lankan approach is touted by some as the future of anti-rebel operations. Troublingly, Sri Lanka hosted a conference at the start of this month entitled “Defeating Terrorism – the Sri Lankan Experience”, at which the Minister of External Affairs claimed that “the entire body of international law must be revisited,” presumably to be more favourable to the Sri Lankan approach to conflict.[fn]‘Sri Lanka seeks new international laws to fight terrorism’, Reuters, 31 May 2011.Hide Footnote

At the other end of the spectrum, there are those who believe that the idea of “collateral damage” actually legitimises civilian deaths, and allows parties to conflict to slaughter civilians with impunity. Marc Herold writes:

“civilian casualties are not accidents or mistakes. They result from careful calculation by … commanders and military attorneys who decide upon the benefits of an air strike versus the costs in innocent civilian lives lost.”[fn]Herold, M, ‘Truth as Collateral Damage’, Guardian, 22 October 2008.Hide Footnote

The implication is that, by making allowances for civilian deaths, humanitarian law allows states to be lax in their targeting decisions. Herold and others therefore suggest that the standard should become far stricter, closer to the human rights standard of the inviolability of life. This view has its own difficulties: calls for a far stricter standard are unrealistic, and risk alienating states and fostering further resistance to the entire humanitarian regime.

Debate as to what humanitarian law truly requires in this regard will undoubtedly continue. Like all jurisprudential norms, clearer standards will evolve on a case-by-case basis. Ideas of proportionality, excessiveness, reasonableness, fairness, forseeability, recklessness, negligence, actual or constructive knowledge and the like are not foreign to other areas of legal thinking. They are applied daily in courtrooms as well as in the day-to-day management of human affairs. They are perfectly suitable standards by which a proper balance can be achieved between military and humanitarian imperatives. As with many other aspects of humanitarian law, the development of jurisprudence and other guidance will provide greater clarity over time.


A final issue relates to the recent growth in so-called ‘lawfare’. The term has been used, over the last decade or so, to mean the use or abuse of the laws of war as a military tool. In short, it means “the use of law as a weapon of war”.[fn]Dunlap, C, ‘Law and Military Interventions: Preserving Humanitarian Values in 21st Century Conflicts, Working Paper (Boston: Harvard Kennedy School, 2001), p. 5.Hide Footnote  This can manifest itself in two ways: either through threatening the prosecution of senior figures allegedly involved in violations – U.S. and Israeli officials have most often been targeted in this way – or by highlighting abuses in the court of public opinion. In that sense it is akin to propaganda as a military tool, or as others might put it, a form of ‘hearts and minds’ initiative.

Lawfare is, in and of itself, neither good nor bad. If violations have been committed, then it is both right and proper that they should be exposed and that attempts be made to address them, through whatever forum is available. If they have not, the law provides a means by which to rebut the accusations. When they have acted in accordance with the law, states should not be afraid to justify their actions. Of course, lawfare can have a negative impact. Spurious or outright false claims threaten to bring the entire concept of humanitarian law into disrepute, and can create the erroneous perception that it is the law itself – and not its deliberate misapplication – which is at fault. But crying “lawfare!” in an attempt to discredit those who make accusations creates the appearance that the accused cannot justify their actions, and simply compounds the negative impac

The emergence of this concept of lawfare is reminiscent, for me, of the claims of legalisation of politics and the denunciation of judicial activism that were prevalent in the early days of my judicial career, and which will be familiar to Michael Kirby. The resistance to constitutionalism and implementation of fundamental legal norms, particularly in the field of human rights, expressed itself in attacks against the legal and judicial system itself, just as the claims of obsolescence of the laws of war mask the discomfort with its more robust recent implementation. Yet the reality is that international humanitarian law today reflects the best of international law, and of law generally, in that it fosters increased compliance with a minimum of coercion. The relevance, and the potency, of the law of armed conflict, was best expressed by Charles J Dunlap Jr, Deputy Judge Advocate General, Headquarters US Air force who said, in a 2009 article:

“If anyone doubts the role of law in 21st century conflicts, one need only pose the following questions: what was the US military’s most serious setback since 9/11? Few knowledgeable experts would say anything other than the detainee abuse scandal known as “Abu Ghraib”. That this strategic military disaster did not involve force of arms, but rather centred on illegalities, indicates how law has evolved to become a decisive element – and sometimes the decisive element – of contemporary conflicts.”[fn]Dunlap, C, ‘Lawfare: A Decisive Element of 21st-Century Conflicts?’, Joint Forces Quarterly 54 (2009), p. 34.Hide Footnote

Later in the same article, the author claims that lawfare “has emerged as the principal effects-based air defence methodology employed by America’s adversaries today”, referring in particular to the demonization of air strikes “through the manipulation of the civilian casualties” they can produce.[fn]Ibid., p. 36.Hide Footnote

Whilst we should always be wary, therefore, of attempts to manipulate and misuse the law, we should also embrace its potential to foster greater compliance.


This need for modernisation and clarification is not unique to the laws of war. It is a natural part of any legal system. It will come from greater public awareness, greater compliance and greater enforcement. Ultimately, however, any interpretation of the Geneva Conventions must take into account those Conventions’ prime objective: the protection of non-combatants.

The humanitarian objective of civilian protection has become so central to the management of armed conflict in the last decade that it has in fact become itself a casus belli. Resolution 1973 of the UN Security Council has authorised “all necessary measures”, including military action but excluding occupation, to protect civilians in Libya. Not only, therefore, must civilians be protected in war; they may also be protected by war.

With this in mind, any ‘re-interpretation’ of the laws of war that seeks to limit civilian protection seems absurd. Radical solutions – which might involve granting states more leeway when fighting terrorists, or making a distinction between aggressors and defenders so as to give greater scope of attacks to the “good side” – should not be considered. In fact, I suspect, calls for these sorts of revisions stem more from the desire of certain states to legitimise their recent violations of humanitarian law than from any real concern with bringing the law up to date.

Despite some progress in compliance and in enforcement, allegations of serious violations continue to surface regularly, and even the most shocking often remain unaddressed. Defiance knows no limits, a point perhaps best illustrated by Sri Lanka’s recent assertion that “ensuring zero civilian casualties was an overriding priority” in its conflict with the LTTE.[fn]‘Military conference to share methods of war’, BBC, 1 June 2011.Hide Footnote  But the momentum has already shifted against violators. The challenge now is to put in place a universal mechanism of enforcement. The reach of the International Criminal Court is expanding, but is unlikely to become universal any time soon. And its jurisdiction remains ex post facto, addressing only the most serious violations. Calls have been made for a universal monitoring body. The current push for investigations is always ad hoc, and not always successful.

As Charli Carpenter has pointed out, other international regimes have well-established observation mechanisms: the IAEA for non-proliferation, the WHO for international medical standards and the WTO for the rules of free trade. She concludes that,

“[w]ithout an independent monitoring mechanism capable of making informed, systematic, nonpartisan claims about what has happened on the ground, it is all too easy for countries to exploit the gray areas in humanitarian law.”[fn]Carpenter, C, ‘War crimes reporting after Goldstone’, Foreign Affairs, 9 May 2011.Hide Footnote

Developing such a mechanism could serve to provide empirical data on humanitarian abuses, to make it more difficult for states to brush off claims of abuses and provide the global public with a stronger basis for action.

In the end, if the laws of war are under siege, the siege has only served to highlight their continued relevance and the need for their stronger implementation. The reduction in the number of armed conflicts since the end of the cold war is in part a reflection of increased international involvement in the prevention of conflict, including through the work of organisations like International Crisis Group. The reduction of civilian casualties in armed conflict is due, in large part to the moral, political and operational impact of the law. And it’s gaining ground. 

Subscribe to Crisis Group’s Email Updates

Receive the best source of conflict analysis right in your inbox.