1. The two cultures
A military lawyer once joked that at professional meetings you can always tell which are the military lawyers and which the civilians. Military lawyers refer to the laws of war as ‘LOAC’ – law of armed conflict – while civilians from the world of non-governmental organizations call the laws ‘IHL’ – international humanitarian law. ‘They're in the business of saving lives,’ he cheerfully explained. ‘We're in the business of killing.’Footnote 1 I did not take his black humour as callousness about killing, nor did he intend it that way. He takes the law and morality of war seriously; he simply saw no need to mince words about what weapons are for. His observation about professional idioms was, in fact, largely accurate. The idioms dramatize what Eyal Benvenisti has rightly labelled a ‘cleavage between two visions of the law’, one that of armies aiming to ‘conciliate the necessities of war with the laws of humanity’, the other ‘a manifest of humanitarian fraternity’.Footnote 2 The LOAC vision of the law begins with armed conflict. It assigns military necessity and the imperatives of war-making primary, axiomatic status.Footnote 3 In this vision, the legal regulation of warfare consists of adjustments around the margins of war to mitigate its horrors. Those adjustments occupy a noble and important role that must be honoured and that militaries in fact want to honour. But it is logically secondary, and it yields to the force majeure of military necessity. The law of war dwells in the interstices of warfare.
The IHL vision begins with humanitarianism, and assigns human dignity and human rights primary status. It measures the progress of civilization in the enhanced protection of human dignity, views law as an indispensable instrument for advancing human dignity, and regards peace as the normal condition for human life. This vision regards war as a human failure – no doubt inevitable, in the way that poverty and injustice are inevitable, but, like poverty and injustice, not something that deserves legal priority over the protection of rights and dignity. This might sound like pacifism, but the IHL vision is not pacifist. It recognizes that human dignity and human rights need muscle to secure them – they are worth fighting for, and it is noteworthy that legal doctrines favouring humanitarian military intervention (doctrines like the responsibility to protect and the obligation to prevent genocide) grew out of IHL culture, not LOAC culture.
I will have much more to say about these two visions, but this is enough for now. The important point is that the difference in visions largely corresponds with a cultural divide among lawyers concerned with the laws of war – a divide between what I shall call ‘humanitarian lawyers’ and ‘military lawyers’. The terminology is stereotyped. I do not mean to imply that every lawyer in the miltary accepts the military version of the law of war (which I will call the LOAC version or LOAC vision) on all issues, nor that every lawyer working for human rights organizations accepts the humanitarian version (the IHL version, for short) down the line. Some renowned experts have at different times in their careers worked for both military and humanitarian organizations, and some renowned scholars whose views represent the LOAC version of the laws of war are civilians.
Nor, for that matter, do military lawyers always say ‘LOAC’ rather than ‘IHL’ – but the military lawyer I quoted was not being wholly facetious. The two visions of the law of war closely track organizational cultures. That, at any rate, is my conjecture as well as my concern. The concern, which I raise in the shadow of the ‘lawfare’ debate, is that in the last decade military and humanitarian lawyers increasingly see themselves as competing teams whose goal is to ensure that their vision of law prevails. It may not always have been so. David Kennedy rightly notes that the International Committee of the Red Cross (ICRC) ‘has prided itself on its pragmatic relationship with military professionals’, and ‘military lawyers and lawyers from the Red Cross are often able to find common ground with surprising ease’.Footnote 4 At the same time, Kennedy warns, both military and humanitarian professionals often suspect each other of advancing legal positions for tactical reasons, and ‘the vernacular that promised a common conversation has produced a dialogue of the deaf’.Footnote 5
A recent address by the eminent scholar Yoram Dinstein illustrates the two cultures problem in action. Professor Dinstein delivered the closing remarks at a conference on International Law and the Changing Character of War at the US Naval War College. After congratulating the participants on an excellent conference, he warned of the ‘menace’ posed by ‘human rights zealots and do-goodniks, whom I shall call “human rights-niks” for short’.Footnote 6 Professor Dinstein explains:
Far be it from me to suggest that every human rights scholar or activist falls under this rubric . . . . But all too often today we encounter the unpleasant phenomenon of human rights-niks who, hoisting the banner of human rights law, are attempting to bring about a hostile takeover of LOAC. This is an encroachment that we must stoutly resist.
The human rights-niks in back are by no means to be confused with the barbarians in front: far from endorsing the methods of barbarism, the human rights-niks would prefer a non-violent solution to every conflict. Nevertheless, the danger that the human rights-niks pose is equally acute, since they threaten to pull the legal rug from under our feet.Footnote 7
The ‘barbarians’ are terrorists and militants who wilfully violate the rules of warfare, and it is startling that Professor Dinstein regards the danger ‘human rights-niks’ pose as ‘equally acute’. Concluding his speech, Professor Dinstein offers advice:
(a) keep up the good work on the application and interpretation of LOAC; and
(b) keep poachers off the grass.Footnote 8
This is a particularly blunt statement of the two-cultures problem; but I have heard enough less blunt statements to become convinced that the problem is real.Footnote 9 I do not mean to suggest that this adversarial stance comes only from the military side. On the contrary, I believe that the phenomena that bother Professor Dinstein are real, even if his rhetoric is exaggerated and belligerent. The parallel phenomena on the side of humanitarian lawyers include an aggressive overreading of humanitarian protections in IHL, a tendency to presume that civilian casualties are probably unlawful, and occasional double standards in who gets singled out for criticism.
2. Lawyers as advisers
The cleavage between visions of the law not only mirrors the cleavage in professional cultures, the two reinforce each other. Organizational cultures are interpretive communities; lawyers in them develop their own lore about what the law means and how to read it.
That generates a problem for the professional responsibilities of lawyers. Among lawyers’ most important roles is that of legal adviser and what is often labelled ‘compliance counselor’ who reviews client plans to ensure their legality. In military organizations, the adviser's role appears in several disparate guises. In the upper echelons, military lawyers write manuals, directives, and operational handbooks that are, in effect, book-length advice on what the law means. Within combat units, military lawyers help train soldiers in the laws of war, a rather different form of legal advice. And in operations, lawyers provide case-by-case oral advice to commanders.
The legal adviser's basic ethical obligation is to offer independent and candid advice on what the law requires – not frivolous advice, not far-fetched advice, and above all not advice that simply tells the client what he wants to hear.Footnote 10 A lawyer who blesses whatever transactions the client wants to undertake, and in the process writes a get-out-of-jail-free card for the client, is an unethical lawyer.
The problem is that the two visions of the law of war reflect a practical indeterminacy in the law, in which military lawyers and humanitarian lawyers sometimes systematically disagree about what independent, candid legal advice would say.Footnote 11 I call the indeterminacy ‘practical’ in that it would remain a stumbling block even if, like Ronald Dworkin, you believe on theoretical grounds that legal questions have unique right answers. An answer that cannot persuade someone because her interpretive community rejects your basic assumptions will be practically indeterminate even if in some ultimate sense it is right.Footnote 12 Interpretive communities set the inarticulate boundaries of legitimate legal disagreement, beyond which a legal opinion will seem frivolous or even outrageous. But what if the interpretive community itself has split into two interpretive communities? Then the views of each may seem outrageous to the other. The natural impulse of each will be to reject the other's conclusions more or less wholesale. That creates a problem of what role the alternative vision plays in the advice they offer their clients.
Legal advising is the most important thing that lawyers do, and lawyer advice, rather than judicial decisions, defines the law.Footnote 13 Each year, in millions of confidential lawyer–client interactions that are almost never reviewed by anyone else, lawyers advise their clients about what the law requires of them. These conversations are the mosaic tiles that make up the law. Theorists err when they focus on judicial decisions as the heart of the law. Think of Holmes's famous dictum that ‘prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law’.Footnote 14 Judge-centred theorists focus on ‘what the courts will do in fact’, as though the courts pervade the law – which they do not, especially in the law of war where litigation is extraordinarily sparse. A better reading of Holmes's dictum notices that, literally, he is saying that ‘prophecies of what the courts will do in fact’ are what he means by the law. Holmes makes clear that the prophecies he has in mind take place in the lawyer's office, not the courtroom, whenever the lawyer discusses legality with her client. This comes closer to recognizing the law-defining role of lawyer advice, but even Holmes focuses too much on predicting what judges will do and not enough on the lawyer's own choices – a point that will prove important in what follows, where the focus will be very much on lawyers’ own choices.
Given the central importance of legal advising to the law itself, the ethics of legal advice turn out to have major significance for the integrity of law.Footnote 15 The crucial requirement is one I mentioned earlier: the legal adviser is supposed to be candid and independent, and the ethical requirement of candour and independence is the same for military and civilian lawyers. Candour implies at the very least that the adviser gives the client her best interpretation of the law. It may require more than that, because if the lawyer's best interpretation of the law is controversial, and particularly if it represents a minority view, the lawyer must alert the client of that fact, for that too is part of candour. Independence reinforces the demands of candour: it means not tailoring the advice to what the client wants to hear.Footnote 16 The adviser has to have the guts to say no to clients, to give them unwelcome legal news. That makes the adviser's approach to legal interpretation very different from the advocate's. For those of us accustomed to thinking of lawyers as courtroom advocates, who resolve every uncertainty in favour of their clients and who argue forcefully for whatever interpretation of the law helps the client prevail, candour and independence from the client seem like unusual requirements.Footnote 17 Yet without them, the law will be twisted and subverted; it will barely exist.
That was the problem in the notorious ‘torture memos’ written by lawyers in the US Department of Justice between 2002 and 2005. They were subsequently withdrawn and condemned by the department's Office of Professional Responsibility. The torture memos dealt with a host of legal issues connected with a dozen brutally ‘enhanced’ interrogation techniques, and resolved every issue in favour of permissibility. The legal reasoning by which the lawyers conjured away the obligation not to torture was bizarre. Yet these opinions became the secret law of the land for several years, and they created an ironclad legal defence for the CIA interrogators who waterboarded detainees, slammed them into walls, and kept them awake for as long as a week, naked and hanging in chains – as well as for the officials who authorized these interrogations in full knowledge of what they would involve. Ultimately, a prosecutor looking into the torture allegations dropped the investigation in all of the 101 cases.
Candid, independent legal advice can lead to awkward moments, because the lawyer works for the client and understands all too well that strong-willed clients do not like Cassandras. In frequently quoted words, the American lawyer and diplomat Elihu Root wrote, ‘The client never wants to be told that he can't do what he wants to do; he wants to be told how to do it, and it is the lawyer's business to tell him how.’Footnote 18 Yet Root also famously remarked: ‘About half the practice of a decent lawyer consists in telling would-be clients that they are damned fools and should stop’.Footnote 19 I am not sure why Root restricted himself to would-be clients. The same is true of actual clients, when they are eager to push boundaries and cut corners and the lawyer's job is to tell them no.
The laws of war are, at bottom, constraints on warriors and war-fighting. Kennedy disagrees, observing that the laws of war also license violence and war-fighting.Footnote 20 That is true, but without them, violence would need no legal license; for that reason, I disagree that LOAC's licensing function is as fundamental as its constraining function. Lawyers who interpret and enforce the laws of war are basically agents of constraint. If they are military lawyers – I will use the American expression JAGs (Judge Advocate General's Corps) – that places them in an unusual and sometimes awkward position. As military officers, JAGs share the military mission of their service. But as lawyers occupying the role of compliance counsellor, their job is often to put on the brakes, not press the accelerator. The tension is obvious.
JAGs’ double role as military officers and lawyers amplifies the tension. Both roles are quintessentially partisan: they demand loyalty to us, to our side. It is hard enough for a civilian lawyer with a civilian client to comply with the ethical requirement of candid, independent advice. How much harder, then, for a military lawyer to veto a tactic or a targeting choice that a superior would like to use. And yet, sometimes, that is the JAG's job. To be sure, good institutional structures can make it easier – for example, the ‘stovepipe’ structure in the US JAG Corps, which permits an operational JAG to take issues directly to higher JAG officers. But the tension inherent in the role never goes away.
To make matters worse, military lawyers must at times overcome the suspicion that they are not really part of the band of brothers – or, more insidiously, they must deal with their own suspicion that the combat soldiers regard them as ‘jobniks’ rather than warriors even when that is not true.Footnote 21 That provides additional temptation for them to over-identify with their clients even when their professional duty calls for independence.
One question is what countervailing resources are available to JAGs to resist these multiple pressures. That, unfortunately, is a topic for a different paper. My question is whether they should resist the pressures at all, if the laws of war provide a valid (if controversial) case that what commanders want to do is legal.
It is here that the cultural divide within the law of war and the lawyers who interpret it becomes significant. The LOAC version of the laws of war is, as we shall see, less restrictive of combat activities than the IHL version, and systematically so. A military lawyer will find that the LOAC version lessens the strains of commitment. Even better, it does so without veering into the moral twilight zone of the torture lawyers, because the LOAC version is not frivolous, and in fact not particularly extreme. The professional community of military lawyers is not a band of zealots pushing an ideology. Its jurisprudence is more conservative than that of humanitarian lawyers, but it does not rest on eccentric or fantastic premises.
I have said that what makes this a real question is practical indeterminacy in the laws of war, making them susceptible to systematically inconsistent interpretations. But is there such indeterminacy? To explore this question, it will be useful to describe the two visions of the law of war in more detail.Footnote 22 Before doing so, let me preview the argument to follow. After examining the two visions of the law of war to explain their systematic differences, I highlight one feature of the IHL vision, which I call the ‘civilianization’ of the law of war, by which I mean the assumption that civilian interests matter as much as military interests and that, as a consequence, humanitarians have as much claim to the laws of war as militaries. In the final section, I argue that military necessity, properly understood, is itself an inherently civilianized concept, in the sense that it takes civilian interests into account. This section traces the evolution of the concept of military necessity, and argues that claims of military necessity are implicit proportionality claims that the marginal military advantage of an option over alternatives that offer more protection to civilians outweighs the marginal damage to civilians. This creates the possibility of convergence between the two visions, or so I suggest in the conclusion.
3. The LOAC vision
3.1. Taking necessity seriously
The LOAC vision begins, innocuously enough, with the premise that in fact an armed conflict is going on – the regulation of armed conflict presumes armed conflict. Armed conflict is a serious enterprise, and the law of war presupposes that states and organizations engage in it for serious reasons. That will not always be true, and wars have often been waged for despicable reasons. But the law of war presupposes the opposite. Otherwise, there would be little point in regulating war, and no point in making rules for war fighters who are not assumed to be serious moral agents. The LOAC vision presupposes that the war fighters, as the primary audience of the laws of war, are deeply serious moral agents. That requires taking their job seriously and not simply denouncing the destruction it causes.
Warfare imposes its own physical and spiritual demands, what the 1880 Oxford Handbook (quoting Jomini) called ‘inexorable necessities’.Footnote 23 If the enterprise is a morally serious one, its inexorable necessities are equally serious, and the law of war cannot wish them out of existence. Whatever restraints the law imposes must accommodate themselves to military necessity and regulate around its margin.
That implies that any legal regime that would make it impossible, or even inordinately difficult, for fighters to wage war, cannot be serious. If taking the law at face value guarantees defeat for one side or the other in advance, then taking the law at face value cannot be the right way to interpret it.
This is not a merely hypothetical point. To take a current important example, the existing law on human shielding arguably violates the stricture of taking military necessity seriously. On the one hand, it forbids the use of human shields or the siting of military forces among civilians – which, from the point of view of guerrillas or partisans, amounts to requiring the fish to leave the water voluntarily, the better to become prey for the fishermen. On the other hand, the law requires militaries confronted by an enemy that (illegally) uses human shields to maintain the principle of distinction, which would very likely require them to refrain from attacking partisans deeply embedded in a civilian population. The law may have looked like a plausible compromise to those who negotiated the treaties, but in reality it creates requirements that neither side can honour without accepting defeat.
Terrible things happen in wars. The point of the laws of war cannot be to abolish those terrible things. The point can only be to shrink them to what is necessary, where, awful as it is, necessity always means someone else's tears.Footnote 24 Any sane person wholeheartedly wishes that war itself could somehow be made unnecessary, but if that ever happens we will no longer need to talk about the law of war. So long as the law of war remains a subject matter, it must take necessity seriously.
3.2. Treaty interpretation
Taking Necessity Seriously carries implications for the way legal language gets interpreted. Treaties are agreements among states, and the LOAC vision presumes (perhaps only as a rule of thumb or a guiding hunch) that states would never give away powers that enable them to wage war successfully.Footnote 25 Thus, treaty language should be interpreted narrowly and formalistically. Interpreters must not fill gaps in humanitarian protections, because gaps are signs that the contracting parties, the states, never reached agreement on the subject and chose to retain their own unfettered discretion.Footnote 26
Legal theorists often contrast narrowly formalist interpretation with so-called purposive or teleological interpretation – interpreting language through the lens of whatever purpose the legal instrument was supposed to serve.Footnote 27 One might guess that lawyers adhering to the LOAC vision will be sceptical of purposive interpretation, and they sometimes are, when the interpreter appears to be going off on a humanitarian tear, filling in treaty gaps that (military lawyers believe) the states parties left in the treaties for a reason. However, nothing in the LOAC vision is in principle hostile to purposive interpretation. The real difference lies in what you take the purpose of laws of war to be. Is it to protect civilians, even at cost to military effectiveness, or is it to give full sway to military necessity and protect civilians (only) against military excess? Military lawyers assume the latter; it is part and parcel of Taking Necessity Seriously. So even using purposive interpretation, they will be reluctant to fill treaty gaps in civilian protections.
For example, consider the differences between Additional Protocols I and II of the Geneva Conventions (1977). AP I was a milestone in the legal regulation of war, because it codifies the basic jus in bello rules for protecting civilians (the principles of distinction and proportionality). However, by its terms AP I applies only in international armed conflicts and anti-colonialist struggles (Article 1). AP II applies to non-international armed conflicts like civil wars, and it too prohibits making civilians the ‘object of attack’ (Article 13). Unlike AP I, however, it provides no detailed rules protecting civilians, for example, rules requiring warnings and prohibiting excessive collateral damage. Overall, AP II is briefer, leaner, and far more perfunctory than AP I. Furthermore, AP II applies only to a subset of non-international armed conflicts, those where the non-state party actually controls territory.
Clearly, the states negotiating the two protocols were much less generous in protecting civilians during civil wars than in international armed conflicts; it seems likely that they wanted great latitude to put down rebellions. A treaty interpreter in the LOAC vision would be sceptical of efforts to import the civilian protection rules of AP I into AP II; and she would emphasize that neither protocol applies to an uprising where the rebels control no territory. In a conflict not explicitly covered by either Protocol, anything goes except for gratuitous savagery. If the jus in bello rules in the Additional Protocols are taken to represent customary international law, or if their widespread acceptance can be said to constitute it, the same difference would exist between the customary rules applicable in international and non-international armed conflicts.
The IHL vision, by contrast, is deeply reluctant ever to concede that anything goes, and is more likely to blur the treaty lines that LOAC lawyers highlight. The authors of the ICRC's massive study of customary international humanitarian law complain that ‘[c]ommon sense would suggest that such rules [as those in AP I], and the limits they impose on the way war is waged, should be equally applicable in international and non-international armed conflicts’.Footnote 28 They continue: ‘The fact that in 2001 the Convention on Certain Conventional Weapons was amended to extend its scope to non-international armed conflicts is an indication that this notion is gaining currency within the international community.’Footnote 29 A lawyer in the LOAC vision would cite the same evidence to reach the opposite conclusion: the fact that states amended the Convention on Certain Conventional Weapons but not the Additional Protocols shows that they were not prepared to import AP I standards into AP II, or into parallel rules of customary international law. The UK Ministry of Defence's Manual of the Law of Armed Conflict recognizes that LOAC applies in internal as well as inter-state armed conflicts, but immediately notes: ‘Different rules apply to these different situations.’Footnote 30
3.3. Customary international law
Besides treaties, customary international law is the main source of the laws of war. The definition of customary international law is ‘a general practice accepted as law’ or, in another influential formulation, ‘a general and consistent practice of states followed by them from a sense of legal obligation’.Footnote 31 Beginning law students are taught that customary international law therefore has two elements, state practice and opinio juris (the sense of legal obligation). Those who claim that a rule belongs to customary international law must demonstrate both elements.
The tricky part of customary international law is that no automatic way exists to tell if a rule has attained customary status. Interpreters may disagree about whether non-unanimous state practice has achieved sufficient ‘density’ to be labelled a general practice, or whether states undertake niceties out of a sense of legal obligation or for other reasons. More than any other part of law, customary international law exists (or not) in the pronouncements of experts reading the tea leaves of diplomatic practice. The issue is, one might say, interpretive all the way down. That it makes it ripe for duels between experts.
A noteworthy illustration is the controversy between the United States government and the ICRC over the latter's customary international humanitarian law study – a milestone of the IHL vision of the laws of war. The study took ten years and involved hundreds of researchers canvassing state practice and opinio juris. The final product includes a volume of rules accompanied by two volumes of supporting documentation, and it weighs in at 4500 pages. Soon after it was published, however, the US State Department's legal adviser John Bellinger and the general counsel of the Defense Department, William Haynes, wrote a sceptical letter to the ICRC. They complained that the state practice cited by the ICRC is ‘insufficiently dense’; that it places too much emphasis on written materials like military manuals, ‘as opposed to actual operational practice by States during armed conflict’; that it gives ‘undue weight to statements by non-governmental organizations and the ICRC itself’; that it does not give due regard to the practice of ‘specially affected States’ and ‘tends to regard as equivalent the practice of States that have relatively little history of participation in armed conflict and the practice of States that have had a greater extent and depth of experience . . .’.Footnote 32
Now, the last of these complaints might be discounted as special pleading for American exceptionalism. But overall, the Bellinger–Haynes article reflects the concern by military lawyers that humanitarian lawyers overreach in announcing customary international law and ignore actual military practice in favour of documents – and this concern is hardly special pleading.
3.4. The very nature of international law
Military lawyers have states as clients, and the LOAC vision of international law is statist in the traditional fashion represented by the 1927 Lotus dictum: ‘rules of law binding upon States . . . emanate from their own free will as expressed in conventions or by usages generally accepted as expressing principles of law’.Footnote 33 Period. The LOAC vision is sceptical of claims that states’ wills can be constrained by supposedly higher standards – ‘progressive’ trends and tendencies, generalizations and extrapolations, jus cogens, and especially pronouncements by activist tribunals and NGOs.Footnote 34 Thus, the UK Ministry of Defence recognizes only two sources of LOAC, customary law and treaty law, both of which depend on state consent.Footnote 35 Under the LOAC vision, a state that has not bought into a standard remains unbound regardless of what the NGOs or the IHL experts say.
3.5. Deference and discretion
One important consequence of Taking Necessity Seriously is that military lawyers want to leave the judgement calls about what is necessary to military commanders, without after-the-fact second-guessing by courts, investigating commissions, or Human Rights Watch, whose proper role should be condemning clear-cut violations. Consider proportionality calculations, which under the law require weighing the concrete and direct military advantage of an operation against the anticipated ‘incidental’ damage to civilians and civilian objects. Everyone understands that ‘calculations’ in a literal sense are out of the question. Different observers will assign different weights to military advantages and evaluate risks to civilian objects differently. For military lawyers, it is self-evident that commanders enjoy broad discretion and deserve significant deference. Military lawyers vigorously object to ex post second-guessing of command decisions.Footnote 36
Here is an example of the kind of second-guessing that bothers military lawyers: In the ICTY's 2011 Gotovina decision, the Trial Chamber found that a Croatian artillery officer named Rajčić indiscriminately shelled a city by firing at the enemy commander's apartment in a civilian neighbourhood. General Ante Gotovina, who had ordered Rajčić to shell the city, drew a 24-year sentence for war crimes.
The Trial Chamber has found above that firing at Martić's apartment could disrupt his ability to move, communicate, and command and so offered a definite military advantage. Rajčić recognized that the chance of hitting or injuring Martić by firing artillery at his building was very slight. Rajčić testified that the HV [Hrvataska Vojska, the Croatian army] sought to harass and put pressure on Martić and that the HV took the rules of distinction and of proportionality into account when deciding whether to target the apartment block. The Trial Chamber considers that Martić's apartment was located in an otherwise civilian apartment building . . . . At the times of firing, namely between 7:30 and 8 a.m. and in the evening on 4 August 1995, civilians could have reasonably been expected to be present on the streets . . . . Firing twelve shells of 130 millimetres at Martić's apartment and an unknown number of shells of the same calibre at the area marked R on P2337, from a distance of approximately 25 kilometres, created a significant risk of a high number of civilian casualties and injuries, as well as of damage to civilian objects. The Trial Chamber considers that this risk was excessive in relation to the anticipated military advantage of firing at the two locations where the HV believed Martić to have been present. This disproportionate attack shows that the HV paid little or no regard to the risk of civilian casualties and injuries and damage to civilian objects when firing artillery at a military target . . . .Footnote 37
For military lawyers, it is outrageous for the Trial Chamber to second-guess a field commander's risk–benefit assessment on such a fact-intensive and situation-sensitive decision. Who were the judges to decide how important it was to disrupt Martić's ‘ability to move, control, and command’ – which even the Trial Chamber calls a ‘definite military advantage’ – or how likely the shelling was to injure civilians and civilian objects, or how to weigh the military advantages against the risks?Footnote 38 Reversing Gotovina's conviction, the Appellate Chamber partly agreed, complaining that the Trial Chamber's analysis ‘was not based on a concrete assessment of comparative military advantage, and did not make any findings on resulting damages or casualties’.Footnote 39 Military lawyers would add that this should come as no surprise, because a trial court is badly situated to offer a concrete assessment of comparative military advantage.
It goes almost without saying that under the LOAC vision, external accountability for law-of-war violations is undesirable; state militaries should be relied upon to police their own. Universal jurisdiction is particularly deplorable, not to mention inadequately supported by state practice and opinio juris. This outlook represents more than military lawyers’ professional protectiveness of their clients, and adoption of the clients’ point of view (although that is undeniable). The mistrust of external accountability rests on principle, and flows from the conviction that military commanders deserve deference and discretion for decisions they make in the fog of war. The fog of war is one of the ‘inexorable necessities’ that the law cannot wish out of existence.
3.6. Human rights
In the LOAC vision, human rights do not have much to do with the laws of war. Historically, the laws of war evolved along a separate track that pre-dates human rights law. That separate track comes from two sources: centuries-old traditions of chivalry and martial honour, and humanitarianism, meaning the desire to alleviate suffering. Humanitarianism has little to do with human rights; its source is compassion and pity, not recognition of individuals as rights bearers. Its philosophical motivation is the Benthamite desire to diminish pain and suffering, not the Kantian injunction to respect human dignity. Human rights law is for peacetime, and the LOAC vision emphasizes the laws of war as lex specialis, ‘special law’ that displaces more general law under the rule of speciality.Footnote 40 That means it displaces human rights law whenever peace gives way to war.
Military lawyers hewing to the LOAC vision also emphasize the stringent jurisdictional restrictions in major human rights instruments, which underline that human rights are meant to bind governments within their own territory, not in foreign wars. For example, Article 2 of the International Covenant on Civil and Political Rights reads: ‘Each State Party to the present Covenant undertakes to respect and to ensure to all individuals within its territory and subject to its jurisdiction the rights recognized in the present Covenant . . .’ (emphasis added). The Covenant does not obligate a belligerent state outside the territory and jurisdiction of the state.Footnote 41 The point is not that military lawyers denigrate human rights and human rights law. They do not. It is that they doubt its relevance in armed conflict, and see little conceptual or historical connection between human rights law and LOAC.
To summarize, the LOAC vision takes necessity seriously, favours wide discretion and deference to military commanders in judgement calls, reads treaties narrowly and formalistically under the assumption that states gave up as little as they could when negotiating them, interprets customary international law through a ‘hermeneutics of suspicion’ about humanitarian overclaiming, adopts a statist view of international law, and doubts the relevance of human rights law to the laws of war. Military necessity is the keystone of the arch: all the other elements make sense because the necessities of combat are inexorable, and military lawyers regard LOAC as interstitial and narrow – a body of law designed to infuse the ‘principle of humanity’ into the conduct of war, but only in those places where states are willing to let it impinge on military necessity.
4. The IHL vision
Where military necessity forms the unshakeable foundation of the LOAC vision, human rights and human dignity lie at the basis of the IHL vision. As the ICTY put it in its Furundzija judgement,
The essence of the whole corpus of international humanitarian law as well as human rights law lies in the protection of the human dignity of every person . . . . The general principle of respect for human dignity is . . . the very raison d'être of international humanitarian law and human rights law; indeed in modern times it has become of such paramount importance as to permeate the whole body of international law.Footnote 42
That does not mean that the IHL vision fixates on the rather specialized and technical body of human rights law, although as we will see the IHL vision gives human rights law a far larger place than the LOAC vision accepts. The point is rather that the IHL vision begins with individual human beings caught up against their will in a war – human beings who simply want to go about their daily business and whose human dignity confers upon them a right to do so without being blown up, maimed, brain- damaged, driven from the ruins of their homes, or forced to bury their children. The aim of IHL is to secure them, to the maximum extent possible, against the violence and indignity of war – violence and indignity that the LOAC vision never minimizes. IHL offers a civilian's-eye view of war, and gives ground grudgingly to claims of military necessity. Where legal restrictions operate in the margins of military necessity under the LOAC vision, IHL strains to relegate war to the margins of peacetime rights. As a result, its mode of legal interpretation is maximalist in just those places – the restraints and obligations of warriors – where LOAC is minimalist, and minimalist in the places where LOAC is maximalist: in discretion and deference to the military. Largely, both sets of lawyers read and accept the same body of jurisprudence, and both of them admit that war and human dignity belong to the human world. But they reach different conclusions because they assign military necessity and human dignity different logical priority.
A paradigm example of the IHL vision appears in one of the first decisions of the ICTY, the Tadić case. ICTY had a problem. Its statute empowers it to prosecute grave breaches of the Geneva Conventions, but grave breaches are defined only in international armed conflicts, and the Bosnian conflict was a civil war. If ICTY accepted the non-international character of the Bosnian war, it would be unable to prosecute one of its own Statute's principal categories of crimes. A formalist reading of the Statute lends itself to that conclusion, and the principle that ambiguous criminal statutes must be construed in favour of the defendant reinforces that reading. Instead, ICTY read the law broadly rather than narrowly. It devised a novel legal test of state control of military forces to conclude that the war was really an international armed conflict between Bosnia and Serbia. More noteworthy than the conclusion (which, legal novelty or not, accurately described reality on the ground) was the reasoning. The Appeals Chamber explained that its conclusion is based not only on the letter but also on the spirit of the Geneva Conventions, and
is borne out by the entire logic of international humanitarian law. This body of law is not grounded on formalistic postulates. It is not based on the notion that only those who have the formal status of state organs, i.e., are members of the armed forces of a state, are duty bound . . . . Rather, it is a realistic body of law, grounded on the notion of effectiveness and inspired by the aim of deterring deviation from its standards to the maximum extent possible.Footnote 43
The mode of interpretation employs the so-called ‘Principle of Effectiveness’ holding that legal instruments aiming to protect people's basic rights must be interpreted to make them effective at achieving their object and purpose. In its 1989 Soering decision, the European Court of Human Rights wrote:
In interpreting the Convention regard must be had to its special character as a treaty for the collective enforcement of human rights and fundamental freedoms. Thus, the object and purpose of the Convention as an instrument for the protection of individual human beings require that its provisions be interpreted and applied so as to make its safeguards practical and effective. In addition, any interpretation of the rights and freedoms guaranteed has to be consistent with ‘the general spirit of the Convention. . .’.Footnote 44
Soering treats the European Convention on Human Rights as an evolving document, one that expands its protections to fill gaps. Tadić treats humanitarian law exactly the same way. It appeals to the ‘spirit’ and ‘logic’ of humanitarian law. This is precisely the kind of gap-filling interpretation of legal instruments that lawyers in the LOAC vision reject. They see treaties as agreements representing the will of their parties, nothing more, not clauses in a ‘living constitution’. But for the IHL vision, human rights are different, special, and constitutional in their status.
Recall that adherents to the LOAC vision assume that states negotiating law-of-war treaties would never have given away war-fighting powers that matter to them. The IHL vision understands the history of humanitarian treaties differently, as manifestations of states’ desire – or at least the desire of important state constituencies – to break with the horrifying past. To suppose that governments and their Geneva delegates were uniformly hardliners and realists (in the international-relations sense) begs the question, misunderstands politics, and falsifies a complex history.
The IHL vision of customary international law also differs dramatically from the LOAC vision. First of all, the roster of agents whose practice and opinio juris contribute to law formation has expanded to include international organizations like the UN Secretariat and the ICRC. As the ICRC points out, both have legal personality, and both have a military role: the Secretariat controls UN military forces, and the ICRC ‘has received an official mandate from States “to work for the faithful application of international humanitarian law applicable in armed conflicts and . . . to prepare any development thereof”’.Footnote 45 Formalism aside, the Secretariat and the ICRC have institutional commitments to protecting civilians, and in the IHL vision that commitment belongs at the table in the formation of customary international law.
But the UN Secretariat and ICRC are hardly the only non-state actors who contribute to customary international law. The IHL vision recognizes that the process of international law-making has evolved substantially from the narrow statism of Lotus. Law-making now rightfully includes many actors, including international organizations and tribunals and NGOs. States may have created the IOs and tribunals, but in creating them, states endowed their creatures with the power to speak law on their own.Footnote 46 So too, NGO arguments operate indirectly by pressuring states to respond to them, but this gives them a genuine role in the formation of opinio juris.Footnote 47
The IHL vision is therefore willing to include a variety of soft-law instruments from multiple sources as evidence of opinio juris; and it treats official state statements – verbal practice as opposed to practice on the ground – as law-generative state practice.Footnote 48 After all, the practice on the ground might represent law-breaking rather than law-making, so there is no reason to privilege it over states’ statements of the principles they stand for.Footnote 49 Of course states’ anodyne pro-humanitarian statements may be public-relations hypocrisy, but that makes them no less a form of state practice. The maxim that hypocrisy is the homage vice pays to virtue is never truer than in diplomacy, and paying homage to virtue is a form of state practice, as old as the state itself. The fact that a state practice is insincere does not weaken it as a source of customary international law.
In the same way that the IHL vision uses more expansive methods of treaty interpretation than the LOAC vision, it is more expansive in the domain of state practice and opinio juris it canvasses to identify customary international law. International courts and tribunals have ratified the methods favoured by the IHL vision, and that means the IHL vision has gained a foothold in positive law – it is not simply wish-fulfilment fantasies of humanitarian reformers.Footnote 50
To be sure, there has been pushback against giving too much credence to soft law as a source of customary international law. Almost 30 years ago, Prosper Weil warned that ‘the accumulation of nonlaw or prelaw is no more sufficient to create law than is thrice nothing to make something’.Footnote 51 A decade later, Bruno Simma and Philip Alston criticized ‘a cultured pearl version of customary international law’ that operates ‘through proclamation, exhortation, repetition, incantation, lament’.Footnote 52 But traditionalist warnings like these miss the jurisprudential argument behind the IHL vision, which is that international law rightfully responds to broader constituencies than states. Legal accountability has become more pluralist and more populist than in the traditional state-centred view of international law. Today accountability runs to multiple institutions – public, semi-public, and private – which serve as surrogates for accountability to people. Accountability implies obligation,Footnote 53 and obligation is the calling card of law. This is a point of decisive importance, to which I return shortly.
LOAC lawyers are ‘legal centralists’, a name for the jurisprudential school that recognizes only state-made law as genuine law; and they are positivists, granting legal validity only to norms that can trace their pedigree back to state consent. In IHL jurisprudence, legal centralism and positivism give way to legal pluralism, the view that even the most advanced societies contain multiple legal systems, some state-made but others not. IHL lawyers view international law as a pluralist system where states have lost their exclusive authority over the rule of recognition for customary international law. In the current environment the rule of recognition for international law is itself up for grabs, and under the IHL vision a progressive trajectory in state verbal practice and opinio juris can be taken up by other articulators of law to demonstrate the emergence of new rules of customary international law.
Observers commonly attribute this change to the human rights revolution and its consequences. The shift away from the classical state-centred picture of sovereignty began when the Nuremberg Charter stripped away the act-of-state and obedience-to-law defences and penalized crimes against humanity ‘whether or not in violation of the domestic law of the country where perpetrated’. It continued with the formation of the United Nations and the emergence of modern jus cogens norms. More recently, it manifested itself in the proposal that sovereignty is conditional on protecting the human rights of the state's people, an interpretation suggested in 1999 by then-UN Secretary-General Kofi Annan.Footnote 54 Once we subordinate state sovereignty to human rights norms, the consensualist theory of international law becomes at best incomplete and at worst incoherent.
In contrast to the LOAC vision, the IHL vision of human rights law perceives a strong connection with the laws of war. As we have seen, the LOAC vision rejects the application of human rights law in wartime as inconsistent with existing law and untrue to the very different historical paths on which the two branches of law have evolved. To both objections IHL has a response.Footnote 55 To begin with, the major law-of-war treaties incorporate what are plainly human rights protections. The most obvious is common Article 3 of the Geneva Conventions, frequently and correctly described as a mini-human rights convention embedded in Geneva. The same goes for Article 75 of Additional Protocol I, which offers a schedule of human rights protections for ‘persons who do not benefit from more favourable treatment under the Conventions or under this Protocol’. The Geneva edifice of ‘more favourable treatment’ therefore rests on a conceptual floor of basic human rights.
Furthermore, IHL proponents can point to the ICJ's holding that the ‘protection of the [International Covenant on Civil and Political Rights (ICCPR)] does not cease in times of war’,Footnote 56 unless states expressly derogate from it – and they cannot derogate from the most basic human rights. To be sure, the ICJ also points out that the law of war, as lex specialis, defines the meaning of human rights in wartime: the rights against arbitrary deprivation of life and liberty will be weaker in time of war because killing and imprisonment that would be ‘arbitrary’ in peacetime may not be arbitrary under laws of war that take necessity seriously. But even with that important qualification, the ICJ conceptualizes the law of war as an interpretive guide to the law of human rights rather than a substitute for it. Human rights law may provide lesser protections in wartime than in peacetime, but its obligations do not go away, and the lex specialis never supplants them.
What about the explicit jurisdictional limitation on the ICCPR? IHL reads the treaty language differently from the LOAC interpretation, under which the Convention applies only within a state's own territory or jurisdiction. IHL reads it as indicated by the brackets I have inserted in the jurisdiction clause:
‘Each State Party to the present Covenant undertakes [to respect] and [to ensure to all individuals within its territory and subject to its jurisdiction] the rights recognized in the present Covenant’.
On this reading, the territorial and jurisdictional limitation applies only to the obligation to ensure human rights, not the obligation to respect human rights; that is, not to violate them. Nor is this a strained reading. It makes sense that states can be expected to ensure human rights only on territory they control, but it also makes sense that they can respect human rights (by not violating them) anywhere. On the alternative LOAC reading, the requirement to respect rights, restricted to a state's territory or jurisdiction, becomes redundant: why include an obligation to respect rights if you already have an obligation to ensure rights? You cannot ensure rights if you are violating them. The virtue of the IHL reading is that by giving the duty to respect a wider scope than the duty to ensure, it makes the duty to respect non-redundant and, as lawyers like to say, it gives effect to every word of the treaty language.Footnote 57 Thus, the IHL reading is better on textual grounds and entirely plausible on policy grounds. The ICRC and the European Court of Human Rights believe that human rights law applies whenever a belligerent state obtains ‘effective control’ over an area, regardless of formal tests of territorial jurisdiction.Footnote 58
Finally, the IHL vision does not defer to the judgement of military commanders. That would defeat the whole point of legal accountability. Deference to military commanders making life-and-death decisions in the fog of war signifies more than rightful acknowledgement of military necessity. It also means allowing whatever slack commanders cut themselves to define the common law for interpreting IHL standards. Deference also means trusting commanders to tell the truth about what guided their decisions, under circumstances where they have everything to gain by lying. Recently, for example, a US military court acquitted the last suspect in the 2005 killing of unarmed civilians at Haditha, Iraq, of the most serious charges. A news account offers an explanation gleaned from interviews with military-law experts:
The Haditha case also fits another pattern: Many cases involving civilian deaths arise during the chaos of combat or shortly afterward, when fighters’ emotions are running high; they can later argue that they feared they were still under attack and shot in self-defense.
In those so-called fog-of-war cases, the military and its justice system have repeatedly shown an unwillingness to second-guess the decisions made by fighters who said they believed they were in danger, specialists say.Footnote 59
The acquittal rate is high in US trials for war crimes committed in combat zones as compared with comparable conduct outside battle.Footnote 60
The phenomenon is by no means an exclusively American one: NATO has steadfastly refused to investigate civilian deaths in NATO bombings in Libya, including an attack that reportedly killed 34 civilians (wounding 38).Footnote 61 NATO claimed that all targets were military, but in a few cases a UN investigating commission was unable to find any evidence that this was so, and recommended an investigation – a recommendation that NATO ignored.Footnote 62
The Gotovina decisions quoted earlier help illustrate what is at stake. Even though the Trial Chamber appears to second-guess the artillery officer's proportionality decision, it turns out that the Chamber really based its decision on something less controversial: it simply did not believe Rajčić, and concluded that Gotovina had ordered Rajčić ‘to treat whole towns . . . as targets when firing artillery projectiles’.Footnote 63 The Appeals Chamber found that the Trial Chamber's analysis rested on faulty assumptions about the accuracy of artillery fire, and it reversed on that ground.Footnote 64 Had the assumptions been justified, though, it appears that the Appeals Chamber might have sustained the Trial Chamber's verdict; there is no hint in its opinion that it would defer to commanders in a fog-of-war case.Footnote 65
Humanitarian lawyers believe that international law is already too deferential to military commanders. As a case in point, consider the report of a committee established by the ICTY prosecutor to review the NATO bombing campaign in the Kosovo war for possible war crimes. The committee used a relaxed standard of evidence: ‘credible evidence tending to show that crimes within the jurisdiction of the Tribunal may have been committed in Kosovo’.Footnote 66 Under that standard, one would surely guess that borderline cases would be referred for further investigation. But one would guess wrong. In fact, the committee resolved every debatable case in favour of NATO, and in the end it recommended no investigations at all.
In one instance, a pilot was assigned to bomb a bridge. Too late to stop the bomb, he realized that a train was rolling onto the bridge, and his bomb hit the train. He nevertheless decided to complete his mission by circling back to drop a second bomb, which also hit the train. At least ten passengers were killed and another 15 injured. The Committee ‘had divided views concerning the attack with the second bomb in relation to whether there was an element of recklessness . . .. Despite this, the committee is in agreement that . . . this incident should not be investigated’.Footnote 67 In another instance, NATO bombed a Belgrade television station, killing 10–17 people, in order to ‘interrupt broadcasting for a brief period’.Footnote 68 NATO argued that the bombing was justified in part because the station was broadcasting propaganda. The most the committee could say was that ‘if the attack on the RTS was justified by reference to its propaganda purpose alone, its legality might well be questioned by some experts in the field of international humanitarian law’.Footnote 69 It is hard to imagine a more timid assertion than this, which declines to draw any legal conclusion at all in the committee's own name. To IHL lawyers, the decision to open no investigations of NATO is a poster child of how badly deference to military decisions undermines accountability.
As for universal jurisdiction, IHL lawyers think it is an indispensable complement to states’ own spotty and self-serving efforts at accountability, and a powerful incentive to reluctant state militaries to investigate their own. Universal jurisdiction has a firm foundation in the Geneva Conventions, substantial state practice, and at least some support within the International Court of Justice.Footnote 70
At this point we can summarize the main features of the IHL vision in parallel to the summary offered earlier of the LOAC vision. The IHL vision takes human dignity seriously in the way that the LOAC vision takes necessity seriously: as a fixed parameter within which the law of war regulates. It grants commanders narrow discretion and little deference, reads treaties to fill gaps in the protection of civilians, expands the role of non-state actors and soft law in forming customary international law, favours legal pluralism over statism as the proper account of jurisprudence, and gives wide scope to human rights law even in armed conflict.
5. The indeterminacy problem
5.1. Indeterminacy and the ethics of legal advice
At this point, the indeterminacy problem should be clear. The problem is not that humanitarian lawyers and military lawyers often arrive at different answers to vexed legal questions, for example the permissibility of targeted killings or how to deal with voluntary human shields. The problem is that the modes of argument themselves differ systematically, because the LOAC and IHL visions represent coherent systems of jurisprudence that turn on different Archimedean points, military necessity or human rights and human dignity.
This is bad news for those who find one vision of the laws of war vastly more compelling than its rival. It is straightforward to argue with particular bits of legal reasoning (which does not mean you will win the argument), but much harder to argue with axioms and unarticulated background assumptions. It is hardly a secret that people who see the world differently are likely to see the law differently as well; and perhaps the cultural divide I am describing is nothing more than another manifestation of this well-known phenomenon. Even so, its contours are distinctive.
In the conflict between the LOAC and IHL visions of the law of war, every step of the argument rests on premises (about treaty interpretation, customary-law formation, the nature of international law, the role of human rights, and deference) that the other side rejects on principle; and the principles ultimately rest on the premises that I have called Archimedean points: military necessity and human dignity.
The result is practical indeterminacy – not in the sense that anything goes, that any legal answer is as good as any other, but in the more significant sense that the law can be understood through either of two structured systems that stand in opposition to each other. Both have ample support within recognized sources of law; neither is frivolous or tendentious on its face, although, of course, both can be used tendentiously (but all law can be used tendentiously).
The indeterminacy arises because seemingly discrete doctrinal issues turn out not to be discrete at all. Or rather, they are discrete only if you hold constant your background assumptions about treaties, customary law, military discretion, sources of law, and human rights. But if interpretive communities disagree about the background assumptions, there will simply be too many moving parts in play. Perhaps Dworkin's imaginary super-judge Hercules could nail down all the corners simultaneously; but Hercules is a fiction, unconstrained by the limitations of mortal men and women arguing in real time. In the non-Herculean world, indeterminacy can only be resolved if members of the other interpretive community give you the benefit of the doubt that your premises are reasonable enough that they should listen to your argument. Later in this article I will recommend exactly that as an approach to the two-cultures problem. But at this point it is important to see that granting the benefit of the doubt is a political and moral choice, not a choice compelled (or ruled out) by a fair reading of the law.
It follows that a military legal adviser who offers an opinion grounded entirely in the LOAC vision has not violated the ethical standards governing the adviser's role, or the craft values of the legal profession. (The same is true of a humanitarian lawyer and the IHL vision.) Such an opinion would not fall into the ‘torture memo’ category of frivolous, result-driven lawyer games. Hopefully, though, lawyers would set their sights higher than non-frivolity.
If the opinion comes in the form of an official military manual or other authoritative directive, it has the additional advantage of counting – by the IHL vision's own favoured criterion – as state practice and opinio juris, and in that way it contributes to a more LOAC-oriented version of customary international law, which the military lawyer also believes is the correct version. That sounds like a win–win solution, hoisting the IHL adversaries, as it were, on their own petard. Insofar as the military manual guides JAG advice in combat situations, it also shapes state practice on the ground, further embedding the LOAC vision in customary law. An experienced military lawyer recommends that military lawyers stay ‘involved in the negotiation and discussion of emerging legal norms precisely because it is so vital to maintain ownership in the field of humanitarian law’.Footnote 71 What better way to maintain ownership over humanitarian law than to actively shape it according to the LOAC vision?Footnote 72
5.2. Civilianization of the law of war
The problem, it seems to me, lies in the idea that military professionals ought to own the laws of war. (Recall Professor Dinstein's advice: ‘Keep poachers off the grass’.) That is like saying that Wall Street ought to own securities regulation. Militaries are not the only people in the battle space or the only ones war effects. It may be that in their origin the laws of war were an honour code for warriors who own the battle space by force of arms – in other words, a concession by the strong to the weak, which the weak have by grace and not by right because what matters is not the rights of the weak but the power of the strong.
But that is no longer how most people regard the laws of war. In the past decades, war-makers have had to come to terms with a political world in which their actions fall under the intense scrutiny of media and public opinion – and students of Clausewitz understand that public opinion makes or breaks military causes. The international tribunals give visible structure and focus to that public opinion, but it matters less that war-makers might come under the ICC's jurisdiction (they usually will not) than that they come under CNN's and YouTube's jurisdiction. Their deeds will be on the Internet within minutes of engagement, captured by cellphones.
This phenomenon has become a source of frustration to military lawyers, because audiences assume that gruesome scenes of dead civilians always mean war crimes even when they are nothing of the sort. The lawyers’ complaint is correct, but public scrutiny and accountability are a fact of life, and at bottom they represent a larger truth that the complaint overlooks: the civilian world has staked a claim to the laws of war that is not going to go away, and that should not go away. The laws of war are now common property.
The implication for military legal advisers is both obvious and easy to misunderstand. It is obvious that legal advice is supposed to keep the client out of trouble, and that might mean counselling against courses of action that are ‘legal’ under LOAC but will look horrific. But I put ‘legal’ in scare quotes for a reason. Even though the lawyer might think that cautionary advice is a concession to prudence or politics and not a matter of law, the fact is that the court of public opinion cares about civilians and it demands that the law governing battle spaces also care about them, as much as it cares about military concerns. To the degree that the IHL version of the laws of war articulates concern about civilians that the LOAC version downplays, the IHL version sets the legal standard, not only the political standard.
Earlier I remarked that states have lost their exclusive authority over the rule of recognition, and what I meant was the legal pluralists’ point: states no longer have a monopoly over law creation (if they ever did), and the rule of recognition in the law of war has broadened beyond criteria set by state officials alone. Law recognition is a social practice among broader audiences who care about civilian interests as well as those of the military. They too must be able to – in Hart's terminology – take the internal point of view on the law.Footnote 73 They are audiences who will hold militaries accountable and who shape the political environment of military action.
The phenomenon I am describing might be called the civilianization of the laws of war, by which I mean simply that in the laws of war civilian interests matter as much as military interests. The phenomenon – 20 years ago, Theodore Meron called it the ‘humanization of humanitarian law’ – is unmistakable, and it is a legal phenomenon. It might equally be called the legalization of public opinion.
Of course, the civilian–military schema is too simple. The people of the state fielding the army are also civilians – and they cheer for their own team, care little about the enemy's civilians, and stand firmly against holding their own army accountable for legal violations. So the civilianization of the laws of war will be uneven and inconsistent. On the other side of the ledger, it would be a disastrous mistake to assume that the military client does not care about the human dignity of civilians, so the term ‘civilianization’ might be a misnomer that wrongly denigrates the consciences of soldiers.
I will nevertheless stay with the name ‘civilianization’, which singles out the important fact that the laws of war have passed from exclusive ownership by warriors to joint ownership by the civilians whose fate they determine. As I now argue, the civilianization of the laws of war is not simply poaching on LOAC's fiefdom.
The reason is that, properly understood, military necessity itself requires taking civilian interests into account. Before offering that argument, it is worth previewing what it implies: that the indeterminacy may not be as complete as I suggested. To the extent the LOAC version treats military necessity as a strictly technical limit on humanitarian concerns, it misunderstands military necessity; and, understanding it correctly, the possibility exists for convergence between the two cultures.
6. The career of military necessity
If the LOAC vision of the laws of war derives from the keystone principle that I have called ‘Taking Necessity Seriously’, it is worth our time to examine what military necessity means. As the following discussion demonstrates, the concept has evolved over time, and different formulations vary dramatically on key issues: whether military necessity is a legal concept or, on the contrary, an extra-legal fact that sets the boundary on the legal regulation of warfare; whether ‘necessity’ means what the literal language suggests – something unavoidable if war is to be waged successfully – or merely whatever the military finds convenient; and, above all, whether judgements of military necessity must take civilian interests into account.
6.1. Legal and extra-legal necessity
In medieval just-war theory, the principle of necessity ‘meant that the just side was permitted to use whatever degree of force was strictly necessary in the particular circumstances of the case to bring about victory. Beyond that point, all force became unlawful’.Footnote 74 Under this conception, necessity serves both a prohibitive and a licensing function. It prohibits gratuitous, wanton violence – unnecessary violence. But it licenses all non-gratuitous violence; that is, violence essential to the military goal. In Suarez's words, ‘if an end is permissible, the necessary means to that end are also permissible’.Footnote 75
Medieval theory understood its just-war doctrines as natural law, so it makes sense to describe both the prohibitive and licensing sides of military necessity as legal doctrines, although positivists might put scare quotes around ‘legal’. But starting with Hobbes another way of thinking about the relationship between law and military necessity appears. In war, ‘[t]he notions of right and wrong, justice and injustice have . . . no place. Where there is no common power, there is no law: where no law, no injustice’.Footnote 76 Military necessity is not a legal concept because in war there are no legal concepts.Footnote 77 On the other hand, nothing prevents states from writing rules of warfare and agreeing to enforce them either directly or through the indirect mechanism of reciprocity. The resulting laws of war are positive, not natural, law, because states write them on a Hobbesian legal blank slate. And military necessity represents the limit of rational regulation – Jomini's ‘inexorable necessities’ that states cannot expect each other to forgo. Military necessity therefore lies outside the law.
The latter view represents the regulation of war in the nineteenth century. Consider the St Petersburg Declaration of 1868, which explains its aim as follows:
[H]aving by common agreement fixed the technical limits at which the necessities of war ought to yield to the requirements of humanity, the Undersigned . . . declare as follows: . . .
That the progress of civilization should have the effect of alleviating as much as possible the calamities of war;
That the only legitimate object which States should endeavour to accomplish during war is to weaken the military forces of the enemy;
That for this purpose it is sufficient to disable the greatest possible number of men;
That this object would be exceeded by the employment of arms which uselessly aggravate the sufferings of disabled men, or render their death inevitable;
That the employment of such arms would, therefore, be contrary to the laws of humanity.Footnote 78
The argument proceeds from a technical analysis of the necessities of war to a state agreement to prohibit arms that exceed necessity, as ‘contrary to the laws of humanity’. Here, plainly, ‘the necessities of war’ lie outside the purview of the laws of humanity, and set the ‘technical’ limits to which ‘the requirements of humanity’ must ‘yield’.
The most famous expression of this conception of necessity as an extra-legal limit to the law is the Prussian military maxim ‘Kriegsraison geht vor Kriegsmanier’: the necessities of war (Kriegsraison) take precedence over the rules of war. The translation loses the Enlightenment flavour of the maxim, the contrast between raison (reason) and manier (manner, style). Military necessity represents reason, which uncovers immutable scientific laws – the technical side of military science exemplified most famously by Clausewitz – whereas rules of war are merely stylistic. Under this doctrine, the licensing function of military necessity remains grounded in natural law, but natural law means scientific laws, not moral laws. Ought implies can: if it is technically impossible to win the war under a given prohibition, the prohibition has no force. As for the prohibitive doctrine of military necessity, it has no grounding in natural law at all. In effect, the doctrine of Kriegsraison dismisses law in the name of science.
6.2. The Hostages formulation
The main feature of Kriegsraison on which I am focusing is that it conceives military necessity as a limit to the law that stands outside it, not as a doctrine within the law. This outlook changed in the wake of the Second World War, where victors and vanquished alike recoiled from the seemingly limitless violence the war inflicted. The post-war formula for military necessity appeared in the second round of Nuremberg trials, in the Hostages case:
Military necessity permits a belligerent, subject to the laws of war, to apply any amount and kind of force to compel the complete submission of the enemy with the least possible expenditure of time, life, and money.Footnote 79
Call this the ‘Hostages formulation’.
As Dinstein notes, ‘The key words here are “subject to the laws of war”.’Footnote 80 Now, instead of law being subordinated to military necessity, the law comes first.Footnote 81 If a rule in the law of war provides an explicit exception for military necessity, well and good: the Kriegsraison principle applies. Thus, AP I allows states to restrict the movements of relief personnel ‘in case of imperative military necessity’.Footnote 82 But most rules provide no such explicit exception, and they represent absolute limits on what war fighters may do regardless of military necessity. For example, AP I states that ‘[t]he civilian population as such, as well as individual civilians, shall not be the object of attack’.Footnote 83 This rule carves out no exception for military necessity. In theory, at any rate, the Hostages formulation rejects Hobbes's view of war as a law-free zone; now, law governs war everywhere, by decreeing which rules will bend to military necessity and which will not.
Momentous as this change is, the Hostages formulation remains overwhelmingly slanted in favour of militaries, and grants them enormous latitude. Read literally, military necessity includes any lawful act that saves a dollar or a day in the pursuit of military victory. These are claims of military convenience, not military necessity. Conceiving necessity in this remarkably broad fashion is even more permissive than the Lieber Code's definition of military necessity as ‘those measures which are indispensable for securing the ends of the war, and which are lawful according to the modern law and usages of war’ – although Lieber himself held a very expansive notion of military necessity.Footnote 84
As Michael Walzer perceptively comments on the Hostages formulation, ‘In fact, this is not about necessity at all; it is a way of speaking in code, or a hyperbolical way of speaking, about probability and risk.’Footnote 85 It is worthwhile spelling out why. One argument is that anything that reduces risk to soldiers’ lives in the slightest degree represents military necessity. Armies must place exceptional value on the life of each of their soldiers, and that is why steps taken to diminish risk to their soldiers’ lives, even very slightly, are militarily necessary.Footnote 86 But time and money get equal billing with human life in the Hostages formulation, and it is impossible to justify cost-cutting and quickness as transcendent values on a par with human life. Here, the argument would have to run differently: saving money and time might conceivably spell the difference between defeat and victory. For want of a nail, the shoe was lost, so the horse was lost, so the rider was lost, so the battle was lost, so the kingdom was lost – ‘and all for the want of a horseshoe nail’. You can't know; therefore don't take the risk.
Calling this military necessity is, indeed, a hyperbolical way of speaking about remote risks. Under the Hostages formulation, an air force can use non-precision-guided bombs rather than more discriminating precision-guided munitions merely because they are cheaper, or merely because commanders do not want to wait for a shipment of PGMs to arrive, and describe the choice as military necessity; that is because there is always a remote probability that the extra money or time might spell the difference between victory and defeat. This way of thinking has devastating implications in interpreting treaty-based protections of civilians against ‘incidental’ damage of attacks. Those protections require militaries to take ‘feasible’ steps to protect civilians, and a plea of military necessity suggests that requiring the use of PGMs is infeasible. The Hostages formulation would allow states to declare precautions that impose even trivial added risk, expense, or delay ‘infeasible’ and therefore wholly optional. Hostages sells the brand-name ‘military necessity’ at the cheapest possible price.
It may be that the Hostages tribunal formulated its definition of necessity in such a permissive way because the judges understood that they were in no position to set standards about how much risk and expense militaries must assume, and therefore left it to the discretion of states and commanders. Understood this way, the Hostages formula is not so much a redefinition of military necessity to mean military convenience as it is a doctrine of deference to military judgement about what really is militarily necessary. If so, however, it is a doctrine of nearly absolute deference, and that too is hyperbolical.
Essentially, the Hostages formula makes only a half-break from the doctrine of Kriegsraison. The distinctive feature in Kriegsraison is that it places the interests of the military beyond all regulation, in effect discounting the interests of civilians down to zero, the same as the suffering of enemy soldiers, which also counts for nothing or even less than nothing. The Hostages formula subordinates Kriegsraison to law, but simultaneously it inflates the concept of necessity to include anything the military finds helpful. Except to the extent built into explicit regulations, the interests of civilians and the suffering of the enemy are still discounted to zero.
Packaging anything the military finds helpful under the heading ‘necessity’ is not only dangerous but dishonest; but it flows naturally from an outlook in which interests other than military interests are simply invisible.
6.3. The marginal view of necessity
6.3.1. Necessity as proportionality: Beit Sourik
Once civilian interests enter our moral deliberations – as they must – the Hostages formula becomes unacceptable. Instead, what counts as military necessity must be determined by weighing military importance against civilian damage. We already find this way of thinking in Vitoria:
[C]are must be taken to ensure that the evil effects of the war do not outweigh the possible benefits sought by waging it. If the storming of a fortress or town garrisoned by the enemy but full of innocent inhabitants is not of great importance for eventual victory in the war, it does not seem to me permissible to kill a large number of innocent people by indiscriminate bombardment in order to defeat a small number of enemy combatants.Footnote 87
This would be true, presumably, even if bombarding the fortress or town would save time and money in pursuit of victory, or even if it slightly diminished risk to soldier lives.
Proportionality reasoning of this sort in fact follows the most recent law on military necessity, represented by the Israeli Supreme Court's analysis of the separation barrier Israel constructed during the second Intifada.Footnote 88 Representatives of Palestinians harmed by the barrier brought lawsuits challenging the legality of land seizures for the barrier as well as specific routing decisions. The Supreme Court found that under IHL, civilian property could be seized (with compensation paid) ‘[t]o the extent that construction of the fence is a military necessity’.Footnote 89 One disputed section of the fence was slated for a hill called Jebel Muktam. The Court deferred to the military commander's judgement that alternative routes proposed by the plaintiffs offered less security than the Jebel Muktam route. On the Hostages formulation, that would have sufficed to demonstrate military necessity. Instead, the Court rejected Jebel Muktam because ‘the security advantage reaped from the route as determined by the military commander, in comparison to the . . . route [proposed by the plaintiffs], does not stand in any reasonable proportion to the injury to the local inhabitants caused by this route’.Footnote 90
Before examining the Israeli Supreme Court's reasoning, it is worth noting the most important point: the form of analysis represents a step away from Kriegsraison just as momentous as the taming of military necessity by law in Hostages. Now, soldiers are no longer the only ones whose interests matter. A tactic that better meets the army's security needs than any available alternatives is not militarily necessary if one of the alternatives is much better for civilians and only slightly worse for the army. Civilians, including the enemy's civilians, count. To see how they count, we must delve a bit deeper.
The Israeli Court analyses necessity claims in three steps, which it labels proportionality tests.Footnote 91 First, it asks whether the chosen military tactic bears a rational relationship to its goal, and second, whether that goal might be attained by an alternative that inflicts less damage on civilians. The first is uncontroversial, while the second seems like mere common sense. (Common sense or not, it is incompatible with the Hostages formula, which permits ‘any amount and kind of force’ to compel the enemy's submission.)
Lastly, the Israeli Court weighs the gains to the military against the injury to civilians in marginal terms:
the . . . act is tested vis-à-vis an alternate act, whose benefit will be somewhat smaller than that of the former one. The original . . . act is disproportionate . . . if a certain reduction in the advantage gained by the original act – by employing alternate means, for example – ensures a substantial reduction in the injury caused by the administrative act.Footnote 92
If so, the original act is not militarily necessary.
The idea behind this ‘marginal’ view of necessity – analysed and defended in recent work by Seth Lazar – is that the true military significance of an act is the gain it provides over the next-best alternative; it is ‘necessary’ only in the sense that it is necessary to achieve this marginal gain.Footnote 93 Necessity judgements are inherently comparative. If the advantage over less harmful alternatives is too small, the claim of military necessity lacks normative force.
Why should we accept the marginal view of military necessity? The reason follows from the normative point of appeals to military necessity: they are used to justify inflicting harms on others. To serve that purpose, they must answer the question, ‘Why won't less damaging alternatives do?’ The answer must take the form ‘The other alternatives don't accomplish X’, pointing to some tangible advantage, ‘and X is necessary’. That raises the obvious question ‘Necessary for what?’ The answer ‘Necessary to avert very small risks, delays, or expenses’ simply does not do the job. Or rather, it does not do the job unless the injuries inflicted are even slighter. That is why, in the end, claims of military necessity are not only disguised comparative judgements (with other alternatives), but also disguised claims of proportionality.Footnote 94
Although the Israeli Court's analysis is sophisticated, it is not idiosyncratic. In an important sense, denying the claim of necessity to small military advantages represents a return to the term's traditional meaning: in Vitoria's words, of ‘great importance for eventual victory in the war’, and in Lieber's words, ‘indispensable for securing the ends of the war’. Necessary, one might say, means necessary. It is the Hostages formula that should be regarded as idiosyncratic.
6.3.2. Koskenniemi's objection
Some might object that proportionality judgements are no more determinate than the issues they are meant to resolve. Criticizing Beit Sourik, Martti Koskenniemi complains that ‘[i]f people were able to agree on what is reasonable or proportionate, no courts, or law, would ever be needed’.Footnote 95 In his view, talk about proportionality ‘tests’ invokes pretensions of science that lawyers cannot possibly deliver on. Proportionality raises questions of whose interests count and how much weight to give them. Why balance only the military's interest in ‘military security’ against the interests of Palestinian civilians? Why not include the interests of Jewish settlers, or for that matter the justice or injustice of the entire occupation?Footnote 96 What makes the security of settlements a benefit rather than a cost?Footnote 97 The Beit Sourik Court simply assumes there are answers to those questions – and so, in Koskenniemi's eyes, the Court ‘pulls itself from the quicksand of ever receding argumentative chains onto firm ground like Munchhausen, by his own hair’.Footnote 98
Koskenniemi's imagery is striking, but matters are not as dire as he believes. I said above that claims of military necessity take on their sense as part of a practice of justifying acts of violence – to borrow Wittgenstein's familiar phrase, a ‘language game’ of challenge and response. In response to the challenge ‘Why did you do this dreadful thing?’ the military answers ‘It was necessary to achieve our military goals’. The challenge-and-response language game presupposes an audience that does not reject the military's goals in limine. Koskenniemi is right that that will not be every audience.Footnote 99 Notably, ‘revisionist’ military ethicists who deny that morality permits acts of violence by the unjust side will not be a potential audience. For example, Thomas Hurka denies all claims of military necessity to the unjust side because ‘no act by soldiers on a side without just cause can satisfy proportionality’.Footnote 100 Similarly, the audience for claims of military necessity may not include the enemy, for whom the only thing ‘necessary’ is that their adversary not achieve its goals – although even the enemy might accept necessity claims out of reciprocity, because they too use violence and make necessity claims.
But threshold unwillingness to entertain military necessity claims does not make the meaning of military necessity indeterminate; it merely takes the concept off the table. If the concept of military necessity gets meaning from its use in a language game of normative challenge and response, the assumptions built into that language game may provide the determinacy we need. Here is how:
The initial challenge ‘Why did you do this dreadful thing?’ assumes damage to outsiders (typically civilians) that resulted from the military's violence. That gives us one term of the balance: civilian damage. The necessity justification, linked to a military goal, gives us the second term: military advantage.
Furthermore, in the same way that the practice of challenge and response presupposes an audience willing to entertain the response, it also presupposes an audience sceptical enough of the military to raise the challenge. It assumes, in other words, a more or less neutral audience. As such, the audience gives equal weight to the lives and other interests of all sides – another assumption that helps make proportionality judgements manageable.
Bit by bit, as we flesh out the practice of normative challenge and response that gives military necessity claims their meaning, the interests to be compared become clearer. Of course, Koskenniemi is right that nothing in principle limits us to just two terms – nothing except the fact that once the inquiry broadens to include indirect costs and benefits or larger issues of justice and injustice, the language game becomes unmanageably speculative. Thus another assumption built into the challenge-and-response practice is that it focuses on local and direct rather than global or indirect costs and benefits. In Beit Sourik, for example, the local and direct terms are the marginal security advantage of Jebel Muktam over alternative routes versus marginal damage to civilians in these particular villages. Of course, there is a far more important question about whether the separation barrier as a whole illegally impedes Palestinian self-determination as a whole; but that is simply a different question than the military necessity of a particular route.
6.3.3. Necessity, law, and civilianization
Kriegsraison, the Hostages formulation, and proportionality represent three different views of military necessity. The first sees necessity as a pursuit of victory in which a commander is unbound by law and the interests of other people. The second subordinates necessity to law, but within the limits of law it permits the commander to discount civilian interests completely. The third requires the commander to take civilian interests into account – and, I have argued, this is the best view of military necessity.
Returning to the two-cultures problem, this final step means that, like it or not, the LOAC view of the world must incorporate the interests that the IHL world-view emphasizes. The possibility of sealing itself off from the IHL version of the law of war no longer exists, for a concept of military necessity that considers only military advantage does no justificatory work.
7. Conclusion
These arguments about military necessity are not meant as a ‘refutation’ of the LOAC version of the laws of war or anything resembling it. That would be silly. Military necessities are real, and law will not make them go away. The same is true of the other elements of the LOAC vision. States may no longer be the sole sources of international law, but we live in a world of states, which remain the pre-eminent international lawmakers. The laws of war must take the civilian point of view seriously, but it is still a long step from there to human rights. On all these points, humanitarian lawyers who pretend otherwise are fooling themselves both legally and practically. Legally, because the sources of law will not bear so much humanitarian weight, and practically because the only hope for the humanitarian project lies in militaries and military lawyers who believe in it and want to make it happen. Like it or not, the two legal cultures must live with each other, and that requires reasonableness, in the sense defined by John Rawls: a reciprocal desire for principles that could be accepted even by adherents of the other comprehensive view.
To illustrate with an example: Article 57 of AP I requires militaries to take all ‘feasible’ precautions to verify that their targets are legitimately military and to minimize civilian damage. Notoriously, there is no agreement on what ‘feasible’ means. Does it include anything technologically possible, regardless of cost or risk to the attacker? Alternatively, does it exclude anything that might increase military risk, no matter how slightly? Clearly, militaries could not reasonably accept the former, and humanitarians could not reasonably accept the latter – so, on my proposal, neither of these interpretations can be right, and lawyers should not advance them.
This conciliatory approach is not self-evident. In purely scientific pursuits, epistemologists offer powerful arguments that it is more rational both for individual researchers and for the scientific community at large if competing research programmes forcefully press their own agendas, even in cases when one programme is less likely than its rivals to be fruitful.Footnote 101 Lawyers are, for obvious reasons, instinctively drawn to a similarly adversarial, competitive model of truth seeking. Why not let the LOAC and IHL versions of the law of war continue to compete for supremacy? Is that not the most likely way in which truth will out?
The obvious difference is that lawyers arguing about the interpretation of law are not pursuing hidden truths. They are not physicists hunting the Higgs boson or mathematicians vying for the honour of being first to solve a famous problem.Footnote 102 They are trying to give concrete meaning to past lawmakers’ constructions, in order to impose discipline on violence when collectivities go to war. The obvious danger in an adversarial competition over who owns the law of war is one David Kennedy highlights: when legal interpretation turns into a political game, the players’ trust in each other's candour inevitably erodes, so that ‘as we use the discourse more, we believe it less – at least when spoken by others’.Footnote 103 The result (Kennedy adds) is a law of armed combat that undermines itself and casts its own legitimacy into disrepute, even in the eyes of its practitioners. I wholeheartedly agree with this diagnosis, but not with Kennedy's cure, which is to downplay legality in favour of pure choice – to ‘be wary of treating the legal issues as the focal points for our ethics and politics’.Footnote 104 In place of legalism, Kennedy calls for ‘recapturing the human experience of responsibility for the violence of war’ – accepting that ‘those who kill do “decide in the exception”, . . . [and] as men and women, our military, political, and legal experts are, in fact, free – free from the comfortable ethical and political analytics of expertise, but not from responsibility for the havoc they unleash’.Footnote 105 His argument appears to be that debates over the laws of war are irredeemably strategic. Officers and political leaders – and, for that matter, humanitarians – find it all too convenient to fob responsibility onto lawyers and the law when in fact the law is ‘an elaborate discourse of evasion’.Footnote 106
But suppose there were no LOAC or ICL. Do we really believe that more responsible decisions would result, that fewer lives would be lost, or that an alternative and better vocabulary than ‘the analytics of expertise’ would arise for deliberation? I see no reason to think so. Without some vocabulary for deliberation, the pure experience of responsibility floats in a vacuum and goes nowhere. Like it or not, and no matter where we end up, we must start with the vocabulary we have. That is the legal vocabulary of the law of war, heavily inflected with the just-war theory of past centuries. Where else could we start? In Quine's words, ‘We are like sailors who on the open sea must reconstruct their ship but are never able to start afresh from the bottom.’Footnote 107
The two cultures are stuck with each other aboard the same wounded ship. The argument of this article has been that their differing comprehensive views arise from competing premises about the primacy of military necessity and human dignity. Both are reasonable premises, and mutual recognition that they are reasonable – more precisely, willingness to discard one's own interpretations if a similarly willing adherent to the alternative view could not possibly accept them – seems like a plausible canon of interpretation. It is also the most plausible strategy for achieving whatever convergence is humanly possible.
