Monday, November 14, 2005

My initial reactions to the ICRC Customary International Humanitarian Law Study

(Important update, 24 January 2006. Jean-Marie Henckaerts, who along with Louise Doswald-Beck are the authors of the Customary Law Study, has been kind enough to send along a response to my comments on the study. It was extraordinarily nice of him to take time to respond to what is, after all, a blog post - my thanks for taking it with such care and seriousness. I have put his response in two places on this blog. It appears at the end of this post, following my comments. I have also put it as a separate post under January 24, 2006, here.)


This past spring, the ICRC issued a major new study in the laws of war, a massive three volume treatise and reference work titled Customary International Humanitarian Law (Cambridge UP 2005; various materials related to it can be found at the ICRC website, here.) I have been slowly - very slowly - working my way through its 5,000 some pages and am delighted to say I have finally finished it. Of course one doesn't really read such a study - it is a reference work, used like a dictionary - but this kind of work requires a total reading if you are looking to understand its underlying structure and suppositions. I plan to write a review essay on it for Policy Review, and perhaps for some law review as well (my Spies report that Anthony Dworkin is reviewing it for the Times Literary Supplement, and I look forward to reading his review). Because the Study is so new, very little review material or academic reactions have yet been published, although the ICRC has some important conference materials on its website from the initial launch, and the ICRC book tour around various law faculties (including an important one in September at my law school) will eventually produce some important scholarly commentary. At this point, the one law review article I am aware of is found in the German Law Review, by Malcolm MacLaren and Felix Schwendimann, which I discussed in an earlier blog post.

What follows are edited remarks taken from the briefing memo I drafted giving initial reactions to the ICRC Study in preparation for a small discussion meeting on Wednesday that Tod Lindberg and I are sponsoring on behalf of the Hoover Institution. The meeting is very small, just a couple of people for lunch, but I believe it will be useful in thinking through my review essays. I will also find an occasion to talk with the ICRC, and I hope with the Study's authors, Jean-Marie Henckaerts and Louise Doswald-Beck, who have certainly devoted years of their lives to this work. But here are my initial reactions:

The Study arises from a mandate given to the ICRC in 1995 by the 26th Annual Conference of the Red Cross and Red Crescent, and has been some eight years now in preparation. It was written by two highly respected specialists and members of its legal division – Jean-Marie Henkaerts and Louise Doswald-Beck, both of whom had served at the ICRC, although the ICRC considers the work primarily scholarship and so “respected the academic freedom of both the report’s authors and of the experts consulted.” (Kellenberger introduction, at xi.) (See the acknowledgments in order to get an idea of the various national and international experts - all leading names in the field - who assisted in the Study’s preparation.)

The overall intent of the Study is to “capture the clearest possible ‘photograph’ of customary international humanitarian law as it stands today.” (Kellenberger introduction, at xi.) In seeking to make that both clear and comprehensive, however, the result is extremely voluminous, running three volumes and five thousand or so pages. It is daunting, by any measure, and yet outstandingly organized and edited to make its pronouncements and supporting documentation remarkably clear. The writing is plain, clear, and, within the inherent limits of the subject matter, rigorous. By any measure, the Study counts as a prodigious accomplishment in scholarship, ranking alongside and, in many respects, even surpassing as a scholarly achievement the ICRC’s Commentaries to the Geneva Conventions and the Commentary to the 1977 Protocols in scholarly and academic terms. This is high praise indeed.

What does the Study mean in practical terms for matters of international humanitarian law? The view of the ICRC is that the Study “does indeed present an accurate assessment of the current state of customary international humanitarian law.” The ICRC therefore intends to “take the outcome of this study into account in its daily work.” (Kellenberger introduction, at xi.) The Study’s view of customary law will thus form the basis, for example, of ICRC challenges to the US practices regarding detainees. Beyond the ICRC, it will almost certainly be absorbed as authoritative by other non-governmental actors, international tribunals, and others. It will thus be cited as essentially binding authority in a wide variety of venues, ranging from courts to treaty negotiations. Because of the fact that customary law is, in principle, binding on all states even without formal consent, the implications of the Study being accepted as authoritative are, to say the least, weighty for international law. Finally, it should be noted, the Study may well have a very significant practical effect within US domestic legal practice, as it will surely be widely cited and, absent some clear dissent from the US, be accepted as the authoritative standard for the content of customary international law in Alien Tort Statute and other domestic cases that turn on international law.

The methodology of the study, at the most practical level, is to state a rule considered by the Study to be a binding rule of customary international humanitarian law, and then discuss it, its boundaries, interpretation, and internal and external features by reference to existing treaty law and what the Study refers to as state practice. It offers, in Volume I, 161 such rules, organized in a coherent order beginning with the most basic principle of distinction, working through specially protected persons of various kinds, methods of warfare, weapons, treatment of civilians and other persons hors de combat, and finally implementation.

State practice. The nature of “state practice” is discussed in detail in the Study’s introduction, at xxxi-xlii. It is evident, both from that introduction, as well as from examination of the rules themselves, that state practice in the sense of the Study has a particular emphasis on the verbal acts of states – state pronouncements, internal military manuals, announced procedures, etc.

Therein, however, lies a chief methodological question for consideration: does the Study privilege the written and verbal aspects of state practice over what states, and in particular militarily significant states, actually do in actual wars? My own view is that notwithstanding the stated emphasis on state practice in totality, the actual tendency of the Study is to rely primarily on written expressions by states, and often to rely on written views of states that may or may not have much practice of contemporary warfare. Thus, for example, the views of Germany through its military manual – a model of writing and clarity, it must be acknowledged – are a very important source of state practice for the Study. Yet the fact remains that however well developed in theory the views of the German government, its actual participation – its physical state practice – in war since 1945 is approximately zero.

Likewise, the exhaustive cataloging of the views of many smaller countries on these issues, while admirably inclusive from the standpoint of acknowledging the “international community,” might be thought to muddy the waters of what actual state practice is in the world of war today. How useful is it, really, to consult the supposed state practice, in matters of war, of Trinidad and Tobago? The completeness produces a prodigy of scholarship – but one might be concerned that the effort might better be regarded as an edifice of scholasticism, massively researched and yet not actually reflecting the world of war as it is.

The flip side of this exhaustive inclusivity is that the views of larger and military powerful states are inevitably downplayed. Unsurprisingly, the views of the United States are the most downplayed – not from any ideological motivation, but simply as a result of a method that emphasizes including everyone’s views, and regards “views” as equally important if not, in the aggregate, more important than the facts of how parties fight. There is, in other words, a certain sense in the Study that in the battle over legal rules, he who writes the most memos wins. (This concern is highlighted by the Malcolm MacLaren and Felix Schwendimann article included in the materials which, given the newness of the study, is one of the few published academic discussions of the Study.)

Protocol I. At the substantive level, there is an elephant lurking in the living room, so to speak. This is simply that at critical junctures, the Study proceeds as though Protocol I really had been accepted by the whole world, and that customary law can be largely predicated on the language of Protocol I and, in important respects, go beyond it. This is a very, very serious problem for the Study, and certainly for the US view of it. One can see, however, how such a view might arise, from the sheer inclusivity of approach to state practice – after all, Protocol I has been accepted by 163 countries. With such wide participation, how can it not essentially be customary law? The problem, as George Aldrich points out in his comments to the Study, is that twenty nine countries have not ratified Protocol I. Lest one think they are of minor import, they include India, Indonesia, Pakistan, Iraq, Iran, Israel, Turkey, and the United States. (And, it should be added, the reservations made even by ratifying countries are dauntingly large.)

The states that have not ratified Protocol I, in other words, include some those most involved in, or likely to be involved in, significant warmaking. As Aldrich says, substantive dissent “from certain provisions of the Protocol by at least some of those States makes it impossible to suggest that the Protocol should, like the 1949 Conventions, be considered to be, at present, a codification of customary law.” Indeed, as he points out – and I fully concur – had Protocol I obtained

“near universal acceptance ... there would be little interest in the question of what constitutes customary international humanitarian law in international armed conflicts. And I suspect that the massive research project that resulted in this book would either not have been undertaken or would have concentrated on internal armed conflicts. It is the failure of Protocol I to achieve complete acceptance of States that makes the substance of customary law both important and sensitive.”

The implication of this is plain and profound with respect to US views of the Study. Quite bluntly, the Study amounts to an attempt to create customary law status for many of the provisions of Protocol I – including, however, both provisions which surely merit such status but also provisions that, if Protocol I has not achieved customary status, do not merit that status on their own. In very important, very disturbing ways, the Study represents a sort of customary law “end run” around the failure of Protocol I to achieve near universal or customary status itself. And indeed, as an examination of various of the Study’s rules indicates, the Study proposes to carry certain very significant rules beyond even where Protocol I’s negotiated outcomes rested them.

Finally, if one is speaking as bluntly as I have just done, it bears noting that although the research teams indubitably are a selection of the world’s finest scholars in the laws of war (see xlvi), it cannot be said, so far as I am aware, that many, if indeed any of them, bring to the table any significant skepticism about the desirability of an expanding reach for customary rules, or even, so far as I am aware of their writings individually, any reservations about the desirability of seeing Protocol I become universally applicable. This is by no means to suggest that they have not brought their full scholarly knowledge and objectivity to bear; I am sure they have – the problem is that this field of law is inherently not measurable entirely objectively. Thus it remains that, within the expert pool, there are apparently no skeptical voices with regard to a methodological preference for carrying the “customification,” if you like, of the laws of war, forward, and this seems to me a difficulty.

Examples from the Study's Rules
Here are a couple of examples of the rules, in order to see how they are developed and what state practice is purported to mean.

Rule 54, covering objects "indispensable to the survival of the civilian population." As Aldrich points out, the formulation of the rule in the Study admits no exceptions - and it is therefore vastly more restrictive than the equivalent rule in Protocol I, article 54, from which it is derived, which qualifies the rule with several important layers of exceptions.

Rule 106, requiring that combatants distinguish themselves from civilian population. Again, as Aldrich points out, the Study rule is significantly simpler than the equivalent Protocol I rule or the Article 4, Geneva III rule. Aldrich points out that the Protocol I rule, unlike the Study rule 106, limits the reach of POW status, without which “any terrorist band could argue that they were combatants provided that they distinguish themselves from civilians in the limited circumstances stated in Rule 106.” It hardly bears stating that the United States, which has significant problems with the formulation in Protocol I, would hardly agree with the still more generous formulation of Rule 106.

Rule 81-82, Landmines. These two rules acknowledge that despite the 140 ratifications of the Ottawa Convention banning antipersonnel landmines outright, the failure of several states to ratify in addition to usage by a dozen or so non-party states “means that it cannot be said at this stage that the use of anti-personnel landmines is prohibited under international law.” (At 282). It is encouraging that despite the ardent desire of much of the international community to have antipersonnel landmines banned under customary law, the Study resists the conclusion. (I should add my own ardent desire to see antipersonnel landmines banned – by treaty.) However, it is also striking – and this is fairly consistent throughout the Study – that the reason offered is not that the dissenting states are large, militarily important, or characterized in such a way as to distinguish, for example, the importance of US non-participation from that of another, but far smaller non-participant, Finland. It is, rather, simply that “several States” have not ratified – perhaps suggesting that in a list that includes the United States, China, India, Pakistan, South Korea, and Russia, as well as Finland, the military importance of those countries speaks for themselves.

The US Response and Opinio Juris
The final question, then, is how the United States should respond to the Study. Should it respond formally? One of the key aspects of the Study, something that goes beyond the particular subject matter of the Study itself, is the implication of the Study’s assumptions that “verbal” state practice is as if not more important than physical practice. The principal focus of the Study is not how warfare has actually been conducted, but, to the extent that the practices of states engaged in warfare matter, what they have said about it. This puts pressure on the United States to do more than simply presume that the fact that it consistently behaves in a certain way is sufficient to establish that as the state practice of a leading military state; it puts the United States under pressure to declare itself as a matter of opinio juris.

Yet the United States seemingly has had great difficulty getting out a consistent, formal, clearly enunciated statement of opinio juris on matters of the laws of war since the end of the Viet Nam war at least. The United States has only fitfully responded to the challenge of Protocol I – refusing to ratify, but never spelling out in a formal way its objections. More importantly, it has not issued its own revised Army laws of war manual that would be the natural place in which the United States would make plain what it agreed with in Protocol I and what it did not. Whatever the virtues of the US approach in the early 1980s to allow customary law rules to evolve out of Protocol I, that approach seems quite mistaken – wholly reactive – today. It has the effect of shifting the “ownership” over the laws of war out of the hands of a combination of states and other actors such as NGOs and the ICRC, but very strongly influenced by leading military states, into the hands of NGOs, nonmilitary states and, in the case of the Study, the ICRC itself. At this point in time, that cannot be a good situation either for the United States or for the development of a consistent, genuinely widely held law of war.

This discussion not intended to lessen the enormous achievement of the Study. It is a magnificent and pathbreaking work of scholarship. And very large parts of it are both right in substance and framed in such a way as to make it accessible in practical legal contexts. But taken as a whole work, in the way that the ICRC has announced that it views it and intends to use it, it cannot be seen, in my estimation, as an authoritative statement of law. It cannot be seen as a statement of law which settles the legal issues of customary law for international tribunals, for US courts, for disputes within the US government. The reasons for that lie in the nature of the study – first, its too-great reliance on verbal state practice and, second, its fraught relationship with Protocol I.

This second point means two things, each of great importance. First, the Study in a significant way represents the attempt by large part of the international community to find a way to reach essentially the law of Protocol I without having to go through what are now, after 25 years, stalled treaty mechanisms. Since the United States, among others, believes that it has sound, substantive grounds for continuing to reject important parts of Protocol I, this approach will hardly do. Second, the Study also represents an attempt to move international humanitarian law beyond the status quo, such as it is, of Protocol I – simplifying and removing, for example, important qualifying language in particular rules. At the grand level, the perceived movement forward of international humanitarian law is stalled because of the inability to get through and then past Protocol I, and seen in this respect, the Study is an attempt to move forward on customary law grounds beyond the shoals of a stalled treaty.

But the divisions in the world over the substance of Protocol I – and not merely somewhat peripheral issues such as “colonial regimes” and the like – and especially between the United States and Western Europe and elsewhere make this either a utopian task, doomed to failure, or else one which tries to move forward by fiat, ignoring the deeply held views of one group or the other. Customary law is therefore the least likely vehicle by which to move forward in such circumstances, because it purports to be – as a statement of law presumably so deep and so wide that it binds even without consent – the most universal.

(Below is Jean-Marie Heckaert's reply to the above comments:)

Comments on Professor Anderson's initial reaction to the ICRC Study on Customary International Humanitarian Law

Geneva, 19 January 2006


With respect to your question whether the Study privileges the written and verbal aspects of state practice over what states actually do in actual wars, the following points can be raised:

A. The study has not looked only at verbal acts of State practice.

1. Official reports on the conduct of actual wars have been included to the extent that they were available, e.g. the report by the US Department of Defence to the US Congress on the conduct of the Gulf War.

2. In addition, numerous instances of what appear to be verbal acts of State practice do in fact describe practice in actual wars. This is particularly the case for military orders, instructions and manuals which reflect what armed forces are instructed and trained to do and what they end up doing most of the time in actual wars. We often lose sight of this fact because the media and others focus so much on the alleged violations of those documents. For example, the best way by which to describe the many instances of humane treatment of prisoners by US troops in its recent wars in Afghanistan and Iraq is either by reference to an official report on those conflicts (see A.1) describing the required treatment of prisoners or by reference to a military order or a manual prescribing such treatment. Other reports, for example those relating to ill-treatment of prisoners, have to be treated differently, namely as descriptions of violations (see below).

B. An accurate snapshot of customary IHL does not merely require a description of what parties "actually do in actual wars". This is so because customary international law is formed by a general practice accepted as law. The latter element is often referred to as "opinio juris". The requirement of a general practice, combined with evidence that this practice conforms to a legal obligation has two important implications:

1. To the extent that field practice is not official and does not represent the legal conviction of the State concerned, it does not count. This implies, in particular, that if actual practice is generally seen as a violation of existing rules, this practice is not of a nature to modify existing rules. That is why notwithstanding numerous reports of, for example, attacks against civilians, pillage and sexual violence these acts are still prohibited under customary international law. The conclusion that these acts are considered to be violations of existing rules can be derived i.a. from a number of verbal acts, such as legislation, case-law and official statements. These verbal acts are therefore important and have to be considered to get the full and correct picture of customary international law.

2. The opposite is also true. Although States may in practice abstain from engaging in certain behaviour, through their verbal acts, qualifying their abstention as mere policy not based on a legal obligation to abstain, they can reserve their right to engage in those acts in the future. That is why attacks against works and installations containing dangerous forces – although seldom or never resorted to in the last twenty years – would still be considered lawful by States in case they constitute military objectives and sufficient precautions are taken. If mere battlefield behaviour were examined such targets would have to be considered off-limit. The same is true for nuclear weapons: on the basis of a mere consideration of battlefield practice, nuclear weapons would have to be considered unlawful – a position clearly not shared by the nuclear powers. This shows that verbal acts have an important impact on how battlefield practice has to be looked at.

In conclusion, while opinio juris alone cannot create custom, practice alone cannot create custom either. Both elements are required and were looked for in the Study. It seems to us therefore that on methodological issues we are actually more in agreement than in disagreement.


With respect to some of the substantive comments, the following remarks can be made, which indicate again that we are more in agreement actually than in disagreement:

Rule 54. Attacking, destroying, removing or rendering useless objects indispensable to the survival of the civilian population is prohibited.

Contrary to what is asserted, this rule is subject to exceptions. They are addressed in the commentary but admittedly they should have been mentioned in the rule ("in principle" or "subject to exception"). The rule was never meant to be absolute or to go further than Additional Protocol I. George Aldrich first raised this comment at the Hague launch and we agree with him on this issue.

Rule 106. Combatants must distinguish themselves from the civilian population while they are engaged in an attack or in a military operation preparatory to an attack. If they fail to do so, they do not have the right to prisoner-of-war status.

It has been suggested that (a) this rule goes beyond Protocol I or that (b) it elevates the relaxed standards of distinction found in Article 44(3), second sentence, to the status of customary law. Neither suggestion is correct, or at least not as such intended:

(a) Beyond Protocol I: This suggestion is based on the fact that in Protocol I, Article 44 has to be read in conjunction with Article 43. In other words, not any combatant who distinguishes him/herself is entitled to POW status, but only members of (State) armed forces. The term "combatants" in Rule 106 is limited in this sense as it is used in the sense defined in Rules 3 and 4 but admittedly this could have been stated more clearly. It was never the intention to go beyond Protocol I.

(b) Relaxed standards customary: This cannot be the conclusion of anyone reading the commentary. The rule is phrased in general terms and does not indicate exactly how a combatant has to distinguish. It is beyond doubt that a combatant complying with the standards in Article 4 of GC III does properly distinguish. Whether the relaxed standards that have been introduced in Article 44(3), second sentence, of Protocol I for exceptional situations (i.e. for resistance and liberation movements) are customary is explained in full detail in the commentary and that discussion is inconclusive. In other words, it is not concluded that it is customary.

For these and other reasons the study does not support the assertion that all of Protocol I is customary. Other examples which indicate that the study does not support the customary nature of Protocol I as a whole include the following. First, the study does not say anything about the customary nature of a number of provisions, as they are not as such addressed in the study:

- Article 1(4)
- Article 36 (new weapons)
- Article 45 (presumption of POW status)
- Articles 61–67 (civil defence)

Secondly, the study actually concludes that a number of rules are not customary:

- Articles 24–31 (medical aircraft)
- Article 50(1), second sentence (rule of doubt)
- Article 51(6) (reprisals against civilians)
- Article 56 (works and installations containing dangerous forces)

The fact that numerous rules of Protocol I are customary today should not be surprising, as many of them were already customary in 1977. In important respects, Protocol I was a codification of customary norms. It is true, on the other hand, that a number of provisions in the Protocol were new in 1977 but have in the mean time become customary because they have been extensively and uniformly accepted in practice. And as the study clearly shows, a number of provisions have not become customary because they are not uniformly accepted in practice.

We therefore think that the Study has approached Protocol I, and for that matter any other treaty, in a cautious manner and has not assumed that a rule is customary merely because it is contained in a widely ratified treaty. As stated in the introduction:

"This study takes the cautious approach that widespread ratification is only an indication and has to be assessed in relation to other elements of practice, in particular the practice of States not party to the treaty in question. Consistent practice of States not party has been considered as important positive evidence. Contrary practice of States not party, however, has been considered as important negative evidence. The practice of States party to a treaty vis-à-vis States not party is also particularly relevant."


In conclusion, we would like to underline that the ICRC was requested by States to undertake this study, in order to assist them in the complicated task of identifying custom. The organisation took this mandate very seriously and spent nearly 10 years on research and consultation. This study represents, therefore, our best possible effort in providing a snapshot of customary international humanitarian law that is as accurate as possible.

Jean-Marie Henckaerts
Legal Adviser
ICRC Legal Division


Anonymous said...

I think it's great that you're going to discuss this at the Heritage Foundation. Please think carefully about your comments on Rule 106. Despite the U.S. stated objections to relaxing the standards for POW qualifications, which as a practical matter also define the requirements for lawful belligerency, the US in fact supported a resistance movement in Afghanistan for a decade that didn't wear uniforms, more recently supported and coordinated operations with a number of diverse elements of the "Northern Alliance," many of which had no distinctive emblem, and employed a number of US special forces and CIA paramilitary personnel dressed in local native attire.

To my reading there are only two possibile results of this pattern -- either U.S. leaders including Ronald Reagan, both Presidents Bush, all current senior U.S. military leaders, etc., etc., are liable for prosecution as war criminals under an aiding and abetting/joint criminal liability scheme or else actual U.S. state practice is to now accept Rule 106 as declaratory of customary international law. I'd be very curious to know if Heritage leadership is willing to go on record calling these folks, who I thought it held in general esteem, as war criminals.

Anonymous said...

Oops - please change "Heritage Foundation" to Hoover Institute in my comment above - I'm a bit brain dead today!