The pseudo-kindness of wartime lawbreakers

Abstract

It is sometimes said that cruel yet short wars are better for humanity than restrained yet lengthy ones. The idea finds sympathy among Francis Lieber and his Prussian contemporaries, as well as some modern writers who back selective non-compliance with international humanitarian law (IHL) on act-utilitarian grounds. This article refutes three underlying claims and reaffirms that IHL progressively narrows room for crude interest-balancing by its duty-bearers. First, it is claimed that toughening wars quickens them, whereas moderating wars prolongs them. This empirical claim overlooks how actions of the party resorting to brutality – the “brutalizer”, for short – interact with the intention of its adversary. Although the brutalizer clearly controls the amount of violence that it chooses to inflict on its opponent, it does not control the opponent's will to resist and, consequently, the length of the war it fights. History abounds with instances where adding cruelty has stiffened the enemy's resolve rather than accelerating surrender. Second, it is claimed that ruthless but swift wars lessen net inhumanity. On this act-utilitarian view, it is normatively superior to hasten wars through barbarity than to lengthen them through moderation; it is therefore the brutalizer's responsibility to toughen fighting and the brutalized party's responsibility to refrain from resisting the brutalizer. Problematically, the brutalizer usurps authority by imposing its own utilitarian considerations upon the brutalized party. Moreover, the brutalizer blames its disobliging adversary for the extra bloodshed to which it resorts in the name of maximum utility. Third, it is claimed that IHL does or should permit non-conformity when non-conformity stands a reasonable chance of increasing net humanity. This position is inconsistent with IHL's functions, however. IHL does aim to reduce net wartime harm, but it would be a mistake to assume that utilitarian ends necessarily justify, let alone require, utilitarian means. When IHL enacts unqualified rules, it predetermines their conformity or non-conformity through processes that are distinctly not act-utilitarian. Nowhere in these processes do lesser-evil justifications naturally belong.

One cannot … make war in a sentimental fashion. The more pitiless the conduct of war, the more humane it is in reality, for it will run its course all the sooner. The war which of all wars is and must be the most humane is that which leads to peace with as little delay as possible.

Paul von Hindenburg¹

Introduction

There is something enduringly attractive about the notion that cruel but short wars lead to greater net humanity than restrained but protracted ones. It found support in influential late modern texts, such as Article 29 of the 1863 Lieber Code: “The more vigorously wars are pursued, the better it is for humanity. Sharp wars are brief.”² J. C. Bluntschli, a close friend of Francis Lieber's who played an instrumental part in the 1874 Brussels Conference, approvingly summarized the underlying thinking.³

The idea also resonated with senior Prussian Army leaders fighting in the Austro-Prussian War (1866) and the Franco-Prussian War (1870–71). Helmuth von Moltke, chief of staff of the Prussian Army during the latter war, credited with his kingdom's decisive battlefield victory over France, went so far as to call a war's speedy conclusion its “greatest kindness”.⁴ According to the Kriegsbrauch im Landkriege, published by the German General Staff in 1902,⁵ military history will teach an officer “that certain severities are indispensable in war, nay more, that the only true humanity very often lies in a ruthless application of them”.⁶

This sentiment was not limited to Germanic martial tradition.⁷ In the late nineteenth and early twentieth centuries, several Anglo-American military practitioners⁸ and writers⁹ expressed similar views. The same was true of those advocating strategic aerial bombardment during World War II.¹⁰ In 1999, although in a slightly different context of peacekeeping operations and relief activities, Edward N. Luttwak appealed to the world community to “give war a chance” rather than imposing ceasefires, because

[u]ninterrupted war would certainly have caused further suffering and led to an unjust outcome from one perspective or another, but it would also have led to a more stable situation that would have let the postwar era truly begin.¹¹

The post-9/11 era is no exception. The ensuing debates surrounding counterterrorism warfare and the permissibility of torture rekindled the deep-rooted frustration that some felt within Western policy-making circles about the bloc's unwillingness to inflict a swift and decisive blow on their enemies through unconventional methods.¹² Jeremy Rabkin, writing in 2012, stated: “Perhaps, for example, lives would be saved, overall, by pursuing a speedy victory, even if doing so risks more casualties in the short run.”¹³ As recently as 2021, Samuel Moyn mused:

The American way of war is more and more defined by a near complete immunity from harm for one side and unprecedented care when it comes to killing people on the other. It is informed by the standards of international law that constrain fighting. More remarkably, America's operations have become more expansive in scope and perpetual in time by virtue of these very facts.¹⁴

Two related yet distinct claims are at stake here. First, to say that toughening wars quickens them and softening wars prolongs them is to advance an empirical claim that causally links war's brutality to its brevity. Second, by suggesting that harsh yet brief wars are better for humanity overall than subdued yet drawn-out ones, the idea's proponents espouse a normative claim that justifies – indeed, demands – the infliction of maximum violence in war.

Also implicit herein is a third, legal claim. International humanitarian law (IHL) concerns itself, inter alia, with moderating wartime brutality.¹⁵ Insofar as IHL exerts a restraining influence on belligerent conduct, this body of rules inevitably becomes entangled in the causal and normative assertions just noted. In other words, odd as it may seem, obeying IHL arguably lengthens wars and diminishes net humanity as a result.¹⁶ If so, IHL should permit non-conformity when non-conformity stands a reasonable chance of increasing net humanity.¹⁷

This piece refutes each of these claims. It reaffirms that IHL progressively narrows room for rudimentary act-utilitarian calculations by its duty-bearers.

Empirical claim

Let us begin with the empirical claim that toughening wars quickens them, whereas moderating wars prolongs them. Is there any credence to it?

Does sharpening wars shorten them?

In his 1839 Manual of Political Ethics, Lieber noted:

I am not only allowed … but it is my duty to injure my enemy, as enemy, the most seriously I can, in order to obtain my end, whether this be protection, or whatever else. The more actively this rule is followed out the better for humanity, because intense wars are of short duration.¹⁸

It is not clear whether Lieber was advancing this last point merely as “one of his favourite theories”¹⁹ or as an empirical contention backed up with historical data.²⁰

Moltke himself saw the idea vindicated in the Franco-Prussian War. Reported acts of Prussian forces above and beyond what might be considered standard wartime behaviour at the time included:²¹

• • requiring the cooperation of inhabitants of occupied territory to help prevent persons joining French forces;

• • burning French houses and villages and killing male residents punitively for aiding franc-tireur ambushes;

• • placing notable French locals on trains for protection against hostilities;

• • decreeing severe punishment for inhabitants of Alsace and Lorraine attempting to join the French army; and

• • levying extensive requisitions and contributions.

In an 1880 letter to Bluntschli, Moltke recalled:

Thus energetically, and yet with a moderation previously unknown, was the late war against France conducted. The issue of the campaign was decided in two months, and the fighting did not become embittered till a revolutionary Government, unfortunately for the country, prolonged the war for four more months.²²

Moltke advocated the employment of “all methods” except “those which are absolutely objectionable”.²³ A degree of uncertainty remains as to what exactly Moltke's reservation entailed. In September 1870, he ordered the investment of Paris but refused to bombard the city. He made the latter decision, however, not on humanitarian grounds but because he believed that starvation alone would induce the city's surrender long before bombardment – a costly and logistically challenging method – might become necessary.²⁴ Perhaps somewhat more telling is how, while bombarding Paris subsequently on Kaiser Wilhelm I's orders,²⁵ Moltke responded to accusations that hospitals were being hit with artillery shells by noting that Prussian batteries would soon move close enough to recognize Red Cross flags.²⁶

Despite Moltke's insistence to the contrary, heightening severity did not expedite victory in the Franco-Prussian War. Although relentless fighting by Prussian forces from July 1870 culminated in Napoleon III's capture at Sedan in September, it only prompted the formation of a Government of National Defence in Paris and the rise of francs-tireurs across France.²⁷ The thoroughness with which the Prussians carried out reprisals against the latter tactics and other acts of sabotage “did nothing to endear them to the inhabitants”²⁸ and sent the campaigns “spiral[ling] down to new depths of savagery” instead.²⁹ As noted by Michael Howard, “[i]f the French refused to admit military defeat, then other means [had to] be found to break their will”.³⁰

Moltke also discovered that Parisians were prepared to endure the hardship of investment long after their milk supplies had run out.³¹ Not even the city's January 1871 bombardment catalyzed its fall.³² By all accounts, it is the Battle of Buzenval (one last failed offensive by besieged French forces to break through the investment), combined with a botched uprising and an imminent food shortage inside the city, that did.³³

One may argue that Hiroshima and Nagasaki offer a better example of how sharpening wars shortens them. Thus, it is often claimed in the United States and elsewhere that, brutal though they admittedly were, the two atomic bombs used during the end phase of World War II hastened Japan's decision to surrender.³⁴ On 9 August 1945, three days after Hiroshima and the day on which the United States reduced Nagasaki to ashes, Harry Truman stated:

The world will note that the first atomic bomb was dropped on Hiroshima, a military base. That was because we wish in this first attack to avoid, insofar as possible, the killing of civilians. But that attack is only a warning of things to come. If Japan does not surrender, bombs will have to be dropped on her war industries and, unfortunately, thousands of civilian lives will be lost. … I realize the tragic significance of the atomic bomb. … We have used it in order to shorten the agony of war, in order to save the lives of thousands and thousands of young Americans. We shall continue to use it until we completely destroy Japan's power to make war. Only a Japanese surrender will stop us.³⁵

Two 1946 US Strategic Bombing Surveys portray how Hiroshima helped break the pre-existing deadlock within the Japanese government in favour of those inclined to accept the terms of the Potsdam Declaration against those determined to continue fighting.³⁶ In 1947, Henry L. Stimson, US secretary of war during World War II, recalled: “The destruction of Hiroshima and Nagasaki put an end to the Japanese war.”³⁷ A decisive causal link between the atomic bombings and the war's accelerated end has since become a widely accepted narrative among historians³⁸ and in the US nuclear deterrence doctrine.³⁹

There is nevertheless reason to be sceptical about ascribing such singular weight to Hiroshima and Nagasaki in expediting Japan's decision to surrender. As early as 1955, Ernst R. May argued that the 9 August 1945 declaration of war by the Soviet Union is what forced Japan's ruling elites to accept Potsdam.⁴⁰ Some have even suggested that “the two atomic bombs on Hiroshima and Nagasaki alone were not decisive in inducing Japan to surrender. … The Soviet invasion was.”⁴¹

As is the case with historical debates in general, the truth of the matter most likely lies somewhere in the middle. Considerable authority exists for the view that it is in fact the combination of the US atomic attacks and the Soviet entry that compelled Japanese leaders to surrender by mid-August 1945.⁴² In other words, neither of the two alone did so. This undercuts the idea that the nuclear strikes themselves shortened World War II's Pacific theatre.

Controlling the sharpness of one's warfighting versus controlling the opponent's will to resist

What these examples reveal is twofold. The party resorting to brutality – let us henceforth refer to this party as the “brutalizer” – clearly controls the amount of violence that it chooses to inflict on its opponent. Just as clearly, however, the brutalizer does not control the brutalized party's will to resist and, consequently, the length of the war it fights.⁴³

Carl von Clausewitz said as much in On War, published posthumously in 1832. Clausewitz defined war as “an act of force to compel our enemy to do our will”,⁴⁴ and in his view, “force … is thus the means of war; to impose our will on the enemy is its object”.⁴⁵ He acknowledged that not even the most effective application of force, qua means, necessarily breaks the enemy's will to resist:

The fighting forces must be destroyed: that is, they must be put in such a condition that they can no longer carry on the fight. … The country must be occupied; otherwise the enemy could raise fresh military forces. Yet both these things may be done and the war, that is the animosity and the reciprocal effects of hostile elements, cannot be considered to have ended so long as the enemy's will has not been broken: in other words, so long as the enemy government and its allies have not been driven to ask for peace, or the population made to submit.⁴⁶

Recall how Moltke, a soldier with well-known Clausewitzian leanings,⁴⁷ lamented that France's Government of National Defence “prolonged” the Franco-Prussian War “for four more months” after the issue of the campaign was decided “in two months”.⁴⁸ Here we see Moltke effectively conceding that, far from accelerating that war's conclusion, the severity which Prussia initially brought to the battlefield in fact failed to enable it to impose its will on France.

To be sure, some recent wars do seem to suggest that fierce fighting by one party arguably weakens the other party's will to resist. Consider, for example, the Second Nagorno-Karabakh War of 2020, resulting in Azerbaijan's decisive win after forty-four days of active hostilities; and Azerbaijan's September 2023 offensive resulting in the capture of Nagorno-Karabakh in its entirety after one day of heavy fighting. Conversely, the increasing restraints under which Western forces found themselves in Iraq and Afghanistan have sometimes been blamed for the ineffectiveness of their military operations⁴⁹ and, by implication, their prolonged and ultimately inconclusive involvement in those countries.

These examples are, of course, only one part of the picture; other conflicts show that subjecting an enemy to brutality may not dent its will or ability to continue fighting. On the contrary, adding cruelty has in some cases been shown to harden rather than weaken the enemy's resolve, and to prompt reciprocation in kind instead. During World War II, news of the Malmédy massacre – an incident where elements of Joachim Peiper's Waffen SS unit killed more than eighty US prisoners of war⁵⁰ – “undoubtedly stiffened the will of the American combatants”.⁵¹ Hugh M. Cole noted: “There were American commanders who orally expressed the opinion that all SS troops should be killed on sight and there is some indication that in isolated cases express orders for this were given.”⁵² In March 2022, Ukraine's Special Operations Forces reportedly warned that they would “not spare Russian artillerymen in response to their ‘brutal killing’ of civilians and cities”.⁵³

The claim's stronger and weaker variations

As the foregoing observations make clear, Lieber's blunt assertion that “[s]harp wars are brief”⁵⁴ is empirically untenable. Indeed, just one sharp yet protracted war would be enough to invalidate it.

Perhaps not all is lost, though – for it is possible that the claim inadvertently overstates its case. What if the case was merely that some sharp wars are brief, or that sharp wars can be brief?⁵⁵ All that the claim would need, then, is at least one actual or potential example where a war's sharpness reduces its duration. Replacing the claim's strong sharpness–brevity causal link with a weak one would make it evidentially much easier to defend.

Normative claim

The weak empirical claim also makes it reasonable for a party to expect that, at least in some cases, brutality could sufficiently demoralize its enemy and hasten victory as a result. Might brutalizing tactics be justified in such circumstances? And if so, would the brutalized party resisting capitulation be to blame for the war's prolongation?

Act utilitarianism

In its most elementary form, act utilitarianism holds that “[a]n act is right if and only if it results in at least as much overall well-being as any act the agent could have performed”.⁵⁶ Consequently, “an agent acts rightly if she maximizes overall well-being, and wrongly if she does not”.⁵⁷ If tough yet brief wars serve greater net humanity than tamed yet lengthy ones, then toughening warfighting is morally superior to taming it.

Three key aspects may be noted here. First, act utilitarianism calculates the expected utility of an act, rather than its actual utility.⁵⁸ William H. Shaw notes:

Consequentialism instructs the agent to do what is likely to have the best result as judged by what a reasonable and conscientious person in the agent's circumstances could be expected to know. It might turn out, however, that because of untoward circumstances, the action with the greatest expected value ends up producing poor results – worse results, in fact, than several other things the agent could have done instead. Assuming that the agent's original estimate of expected value was correct (or, at least, the most accurate estimate we could reasonably expect one to have arrived at in the circumstances), then this action remains the right thing to have done.⁵⁹

This introduces a shift in the focus of our discussion. Unlike the strong empirical claim examined earlier, act utilitarianism resonates with the weak empirical claim and therefore permits its underlying normative claim to be less causally dependent as well. Under this view, it would no longer be so damaging for Lieber to concede that sharpening wars might not always shorten them; or for Moltke to admit that Prussia's initial campaigns failed to break France's will to resist; or for Truman to accept the possibility that Hiroshima and Nagasaki did not single-handedly hasten the end of World War II. What would matter to them here is rather that, under the circumstances, it was reasonable for them to expect that brutality would end their wars sooner and that severe yet brief wars would embody greater net humanity than mild yet protracted ones.

Second, only that kind of act which is likely to maximize overall welfare is right. Theoretically, the mere fact that act utilitarianism considers overall utility-maximizing acts “right” does not mean that it always praises their performance or blames every failure to perform them. Rather, appraising conduct is the subject of a separate utility calculation.⁶⁰

For committed practitioners of act utilitarianism, however, whether their own conduct maximizes utility matters. Wherever harshness stands reasonable chances of forcing the enemy into submission quickly and, with it, of maximizing net humanity, one must harshen the way one fights. The end not only justifies but also obligates the means. Unsurprisingly, Lieber, Moltke and Truman spoke of what they deemed net humanity-maximizing acts in mandatory terms.⁶¹

The third, often overlooked aspect is also the most significant one for our present purposes. Act utilitarianism holds that one acts wrongly by behaving in a manner that is inconducive to maximum overall welfare.⁶² Thus, in wars where brutality generates reasonable prospects of maximizing net humanity through swift victory, it is wrong not to resort to it.

That includes the party on the receiving end of humanity-maximizing brutality. By refusing to accept the inevitability of a looming defeat or by prolonging a war that is already lost, the brutalized party acts wrongly because it frustrates the brutalizer's net humanity-maximizing endeavour. This remains the case even when the brutalized party's refusal to submit would improve its own well-being, for at its core, act utilitarianism is “impartial”, or “agent-neutral”,⁶³ in the sense that “the good of any one individual is of no more importance, from the point of view … of the Universe, than the good of any other”⁶⁴ – including the agent's own good.⁶⁵ Applied to our scenario, it is wrong for the brutalized party to refuse to sacrifice itself on the altar of greater net humanity.⁶⁶ Accordingly, if act-utilitarian brutalizers are right – and there is no reason to doubt that Moltke and Truman sincerely believed in the rightness of their net humanity-maximizing efforts –  then it follows that the parties brutalized by them are wrong to resist them.⁶⁷

Here, too, it is not an integral part of act utilitarianism to fault under-maximizers. Much depends on “whether it will best promote expected well-being to criticize someone for failing to maximize expected well-being”.⁶⁸ Utility-maximizing or not, however, act-utilitarian brutalizers have not shied away from criticizing their disobliging adversaries. Moltke decried France's Government of National Defence for exacerbating acrimony, working against its own interests and, if tacitly, reducing net humanity, by protracting a lost war. France's refusal to lay down its arms “when the issue of the campaign was decided” in September 1870 subverted what for Moltke was “[t]he greatest kindness in war” – i.e., “to bring it to a speedy conclusion”.⁶⁹

Similarly, the United States counted on Japan's compliant behaviour in its net humanity-maximizing endeavours. If Truman was not explicit in this regard, others are. For instance, according to Herbert P. Bix,

it was not so much the Allied policy of unconditional surrender that prolonged the Pacific war, as it was the unrealistic and incompetent actions of Japan's highest leaders. Blinded by their preoccupation with the fate of the imperial house, those leaders let pass every opportunity to end the lost war until it was too late.⁷⁰

Wilson D. Miscamble is blunter:

In moral terms surely the Japanese leadership had a responsibility to surrender at least by June 1945, when there existed no reasonable prospect of success and when their civilian population suffered so greatly. Instead, the twisted neo-samurai who led the Japanese military geared up with true banzai spirit to engage the whole population in a kind of national kamikaze campaign. Their stupidity and perfidy in perpetrating and prolonging the war should not be ignored.⁷¹

In effect, the brutalizer assimilates the brutalized party in its own net humanity-maximizing exercise and accuses the latter party of the exercise's underperformance.

Arrogation of authority and questionable utility of victim-blaming

There are two major problems here. One concerns how the brutalizer groundlessly co-opts the brutalized party into its vision of what maximizes war's net humanity and what each belligerent ought to do. The other involves the brutalizer effectively blaming the brutalized party for the cruelty that the former adds in the name of greater net humanity.⁷²

That neither the brutalizer nor the brutalized party acts in isolation from each other goes without saying. The brutalizer takes the matter further, though. It presupposes that, albeit as adversaries, the brutalizer and the brutalized party are in a relationship of distributed responsibilities in pursuit of maximum net humanity. Robert E. Goodin calls such an arrangement a model of “task-responsibility”,⁷³ stating that

assigning responsibility amounts essentially to assigning duties and jobs. Different people have different responsibilities, ex ante, because they are allocated different duties and tasks. And people bear differential ex post responsibilities for outcomes, on this account, depending on the role that they played or should have played, pursuant to those ex ante task-responsibilities, in producing or averting those outcomes.⁷⁴

One schematic example Goodin offers concerns how various adults are responsible ex ante for raising a child or avoiding its death and are held responsible ex post for the parts they played or failed to play in the child's fate.⁷⁵

This model needs a broader societal set-up. For instance, a mutual benefit society⁷⁶ helps its members agree on certain utilities for which responsibilities should be assigned among them. Absent such a set-up, the individuals involved – even if they are all act utilitarians – risk failing to agree as to what outcome would embody their maximum welfare, let alone what they should each do to achieve it.⁷⁷ This risk is real because, left on their own, not even act-utilitarian philosophers can agree amongst themselves on such basic matters.⁷⁸

Imagine how the brutalizer would describe wartime task-responsibilities. It would most likely run like this. For a given war that the brutalizer fights,

1. a. that war's speediest possible conclusion embodies its net humanity-maximizing outcome;

2. b. the brutalizer's ex ante responsibility is to toughen fighting; and

3. c. the brutalized party's ex ante responsibility is to refrain from resisting the brutalizer.

If the war lengthens, the brutalizer is responsible ex post for not fighting vigorously enough and the brutalized party for resisting swift capitulation.

The question is whether this is also how the brutalized party would describe wartime task-responsibilities. Three considerations militate against such a prospect. First, the brutalized party has no compelling reason to agree that shortening the war it is fighting with the brutalizer maximizes net humanity, for success need not always mean defeating one's opponent – in some cases, war aims may be accomplished by wearing the enemy down and making its efforts more costly.⁷⁹ In Clausewitz's words,

[i]f we intend to hold out longer than our opponent we must be content with the smallest possible objects, for obviously a major object requires more effort than a minor one. The minimum object is pure self-defense; in other words, fighting without a positive purpose. With such a policy our relative strength will be at its height, and thus the prospects for a favourable outcome will be greatest.⁸⁰

Clausewitz also adds that defence is a stronger form of war than attack⁸¹ and that the object of a defensive war may entail “holding one's own until things take a better turn”.⁸² One distinct possibility that Clausewitz notes in this regard is where “[e]ither additional allies come to the defender's help or allies begin to desert his enemy”.⁸³ Put differently, it may sometimes be reasonable for the would-be brutalized party to deem entrenching its resistance no less net humanity-maximizing than capitulating immediately.⁸⁴

While under Prussian investment, Paris sent Adolphe Thiers out on “a tour of Europe to enlist the sympathy and, if possible, the aid of the neutral Powers for the new régime”.⁸⁵ Thiers’ efforts, though fruitless in the end, reportedly left Otto von Bismarck increasingly concerned:

[W]ith one eye cocked on Thiers’ progress round the courts of Europe and his finger on the languid pulse of the South German States, [Bismarck] was anxious for a rapid end to the war. If the French could prolong their resistance for long enough, not only might the Powers intervene but the whole German alliance might collapse. Only the fall of Paris could bring about the end of the war, and only bombardment, he was convinced, could hasten the fall of Paris.⁸⁶

Writing in late December 1870 just before the city's bombardment, Frederick William, then Crown Prince of Prussia, also worried that

[t]he longer this struggle lasts, the better for the enemy and the worse for us. The public opinion of Europe has not remained unaffected by this spectacle. We are no longer looked upon as the innocent sufferers of wrong, but rather as the arrogant victors, no longer content with the conquest of the foe, but fain to bring about his utter ruin. No more do the French appear in the eyes of neutrals as a mendacious, contemptible nation, but as the heroic-hearted people that against overwhelming odds is defending its dearest possessions in honourable fight.⁸⁷

How much of Prussia's growing anxiety was known to France at the time is unclear. It would nevertheless have been reasonable for France to consider that prolonging its war with Prussia would have been at least as likely to maximize net humanity as shortening it. In other words, two act-utilitarian adversaries may reasonably disagree as to whether concluding their war speedily will maximize net humanity.⁸⁸

Second, the brutalized party has no reason to accept ex ante responsibilities to refrain from resisting the brutalizer. While international society may be seen as a rudimentary mutual benefit society, “states will have conflicting aims; and this would be inevitable even if they were all consciously committed to the same utilitarian ideal, for they would differ about how to realize it”.⁸⁹ International society itself privileges no one view about what maximizes welfare in war over any other view. Nor does international society help States agree that shortening wars through brutality is in fact a utility for which belligerents should be assigned different ex ante responsibilities.

Here, John Stuart Mill's discussion of “feeling of justice” as a “consideration of general utility” is instructive.⁹⁰ “[A] people inferior in strength should fight to the death against the attempt of a foreign despot to reduce it to slavery”, Mill contended, adding:

For such iniquitous attempts, even by powers strong enough to succeed in them, are very much discouraged by the prospects of meeting with a desperate though unsuccessful resistance. The weak may not be able finally to withstand the strong if these persist in their tyranny, but they can make the tyranny cost the tyrant something, & that is much better than letting him indulge it gratis.⁹¹

Similarly,

[t]here would be a great deal more tyrannical aggression by the strong against the weak, if those who knew they were not strong enough to succeed in the struggle, gave way at once and allowed the aggressors to carry their point without its costing them anything. A big boy will think twice before tyrannizing over a little one if he expects that the little fellow will fight to the last and make him pay for his victory. Spirit and obstinacy themselves count for much, and for how much can never be known till they are tried.⁹²

Promoting such a feeling of justice may well maximize overall welfare. If so, fighting to the death may well be the brutalized party's ex ante responsibility in the promotion of such a feeling.⁹³

Given its agnosticism on the matter, international society gives the brutalizer no authority to have the brutalized party's act-utilitarian view set aside and to substitute that view with its own. Consequently, international society gives the brutalizer no authority to have any ex ante responsibility imposed on the brutalized party to refrain from resisting it, either.⁹⁴ Nor, for the reasons noted earlier, is it self-evident that rough yet brief wars are in fact more humanity-maximizing than restrained yet protracted wars. The brutalizer has no monopoly of opinions over other agents, including the brutalized party, in the matter.

Third, the brutalized party has no reason to hold itself responsible ex post for prolonging its war with the brutalizer. Recall here that the wrongness of an agent's action and its blameworthiness involve separate utility calculations. Even if, arguendo, the brutalized party is wrong to prolong its war with the brutalizer, it still does not necessarily follow that it is also right to blame that party for it.

One can easily imagine how, when blaming the brutalized party for delaying surrender, the brutalizer most likely considers it also utility-maximizing to blame that party.⁹⁵ What is unclear is whether the brutalizer is actually correct. For one thing, blaming the brutalized party for resisting the brutalizer is likely to make what Mill called “tyranny”⁹⁶ less costly and more commonplace in the future. In act-utilitarian terms, this also means that the greater frequency of unopposed attacks may well offset any reduction made in the frequency of asymmetric wars. The appearance of extra humanity gained through tougher but quicker wars may well prove illusory, with net disutility simply transferred, if not amplified, from one set of outcomes (i.e., milder but lengthier wars) to another (i.e., more pervasive Millian tyrannies). Much of today's geostrategic nervousness surrounding Eastern Europe and East Asia is no doubt shaped by such prospects.

Furthermore, the wrongness in the brutalized party's refusal to capitulate may nevertheless be “blameless”.⁹⁷ Blamelessness arises where, “while [an agent] did act wrongly, we do not blame her for acting wrongly since she did so out of a good motive”.⁹⁸ We observed earlier that act utilitarianism is “impartial” in the sense that an agent's well-being has no greater weight than anyone else's. Blameless wrongdoing holds that we may occasionally decline to blame someone for acting wrongly when he or she exhibits dispositions that we consider utility-maximizing to promote. Though it may be wrong, ex hypothesi, to prolong its war with the brutalizer, in doing so the brutalized party arguably demonstrates worthy dispositions such as commitment to the protection of its citizenry and perseverance in the face of adversity.

By accusing the brutalized party of refusing to surrender, the brutalizer does two things. First, it arrogates to itself the authority to decide what the brutalized party's role is in maximizing net humanity. Second, it effectively blames the brutalized party for forcing its hand in cruelty for the sake of greater net humanity.⁹⁹ Act utilitarianism warrants neither.

Legal claim

Act utilitarianism holds that, where toughening a war is likely to shorten it and therefore maximize net humanity, it is not only right but also mandatory to toughen warfighting. Now, hardening belligerent conduct for maximum utility may well include acting in non-conformity with IHL, but should IHL permit such non-conformity? Is act utilitarianism compatible with IHL, all things considered?

IHL conformity as utility

Though united in their ultimate concern for humanity in warfare, act utilitarianism and IHL could not be further apart from each other in the approaches they each take. IHL deems conformity with its rules “a compulsory minimum” for wartime humanity and “an invitation to exceed that minimum”.¹⁰⁰ Fundamentally, IHL's core obligations embody elements of rule utilitarianism¹⁰¹ or what Robert Nozik calls “side constraints” upon action which “reflect the underlying Kantian principle that individuals are ends and not merely means; they may not be sacrificed or used for the achieving of other ends without their consent”.¹⁰² In contrast, act utilitarianism attaches no intrinsic value to norm conformity; it is merely an instrument that is either conducive or inconducive to overall welfare. Act utilitarians cannot rule out a priori all possible circumstances where IHL non-conformity serves overall humanity better than conformity and consequently becomes justified, if not obligatory.

These differences leave IHL and act utilitarianism in awkward company. Jeoffrey P. Whitman criticizes act utilitarianism for favouring “a campaign of terror bombing against civilian population centers” in order to “secure victory over a brutal aggressor with a minimal loss of life”.¹⁰³ Similarly, according to David Rodin and Henry Shue,

[t]he [consequentialist] argument is at its most plausible when one thinks of a war with a supremely important moral purpose: a war to stop genocide, for example, or to prevent the invasion and destruction of a legitimate state. From a consequentialist perspective, it is difficult to understand why the just side in such a war should always be bound to observe in bello restrictions such as discrimination and proportionality if doing so might imperil victory. It is equally hard to understand why the unjust side should receive any in bello privileges, such as the right to kill enemy combatants, if this would aid their prosecution of evil purposes.¹⁰⁴

Of the relatively few act-utilitarian defenders of civilian immunity, Shaw is perhaps the most outspoken and articulate. He contends that belligerents purporting to rationalize killing civilians on act-utilitarian grounds are “almost certainly wrong” and that there is a “good chance” that their victory through such methods will not maximize expected overall well-being.¹⁰⁵ On this view, “the danger of harming civilians when, on utilitarian grounds, one should not is significantly greater than the risk of missing a chance to harm them in those very rare cases when this would in fact be the correct utilitarian thing to do”.¹⁰⁶ The probability of wartime utility miscalculation is so great that it is net utility-maximizing for States to train their soldiers “to see it as their duty always to respect civilian immunity”.¹⁰⁷

It is doubtful whether Shaw succeeds in his argument. Assume that, at time T, a State is to choose between training its soldiers always to respect civilian immunity and not training them to do so. The high risk of utility miscalculation mid-fighting at time T + 1 may well suggest, as Shaw does, that the former course of action maximizes expected overall well-being at T.¹⁰⁸ In most cases of hostilities at T + 1, respecting civilian immunity, as trained, also maximizes overall well-being.¹⁰⁹ In “those very rare cases” of hostilities at T + 1bis, however, it is attacking civilians that maximizes overall well-being and is therefore also the right thing to do. Conversely, at T + 1bis, respecting civilian immunity, as trained, diminishes overall well-being and is accordingly the wrong thing to do.

This reveals four things. First, the act-utilitarian rightness of civilian immunity training at T does nothing to vitiate the act-utilitarian rightness of attacking civilians at T + 1bis. Second, this is so even though, as noted earlier, praising or dispraising attacks on civilians at T + 1bis is a separate matter from an act-utilitarian point of view. Third, the act-utilitarian rightness of civilian immunity training at T does nothing to repair the act-utilitarian wrongness of respecting civilian immunity at T + 1bis. Fourth, the latter wrongness remains, although it may be made “blameless” on account of the respecter's prior training.¹¹⁰ This blamelessness is arguably the only difference that Shaw's assertion makes.

Since act utilitarians cannot reject possibilities, no matter how remote, where IHL non-conformity maximizes overall well-being, they cannot bring themselves to uphold that such non-conformity is wrong per se, either. Act utilitarians are not alone here.¹¹¹ Even some thinkers who otherwise oppose act utilitarianism as a matter of principle concede that IHL non-conformity may need to be tolerated in truly exceptional situations.¹¹² Michael Walzer, for instance, uses the expression “supreme emergency” to convey and defend the position that “[u]tilitarian calculation can force us to violate the rules of war only when we are face-to-face not merely with defeat but with a defeat likely to bring disaster to a political community”.¹¹³ Igor Primoratz narrows the range of situations to which Walzer's supreme emergency is available down to “moral disasters”¹¹⁴ such as genocide and the ethnic cleansing of an entire community:

I therefore prefer to think of civilian immunity as an almost absolute principle that spells out one of the central and most stringent requirements of justice as it applies to war, and recognizes an almost absolute right of the vast majority of civilians – namely, all those who cannot be considered “currently engaged in the business of war” – not to be targets of deadly violence. The right and the principle trump other moral considerations with which they may come into conflict, with one exception: that of a (narrowly defined) moral disaster.¹¹⁵

Once the morality of norm conformity is relativized in this way, however, attempts to contain its repercussions are likely to be futile.¹¹⁶ “Rather than severely restricting the abrogation of noncombatant immunity”, Stephen Nathanson warns, the supreme emergency doctrine – and, by extension, any other doctrine like it – “opens the floodgates”.¹¹⁷

Selective IHL non-conformity in the name of greater net humanity

When those floodgates are opened, act-utilitarian brutalizers await downstream. For them, any difference between preventing Walzerian-Primoratzian disasters such as genocide, wholesale ethnic cleansing and foreign tyrannical rule, on the one hand, and maximizing wartime net humanity, on the other, is only a matter of degrees. IHL must permit (if not obligate) non-conformity with its rules whenever non-conformity is expected to accelerate a war's end and maximize net humanity as a result.

Some lawyers are quite explicit in this regard. Consider Rabkin, for instance, who asserts that

[s]trict adherence to rules, if it prolonged the conflict and substantially increased the casualty count – let alone if it inhibited a democracy's war effort enough to give victory to a brutal tyrant – might then be judged less humane.¹¹⁸

Elsewhere, Rabkin argues that “[i]f a tactic helps assure a speedier victory, that must matter, especially if a prolonged war will entail more civilian suffering”.¹¹⁹

We may detect two versions of act utilitarianism in Rabkin's reasoning. In one, it is a democracy's victory over a brutal tyrant that maximizes net humanity. On this view, not all warring States enjoy equal normative standing and, accordingly, not all victories carry equal weight. A win for democracies counts more for humanity overall than a win for tyrants. IHL must permit democracies to dispense with its rules where such dispensation improves their prospects of defeating tyrants.¹²⁰

The other version treats the “casualty count”, or the totality of “civilian suffering”, as the benchmark for a given war's net humanity. Sometimes, an unrestrained war that has a short duration saves more lives than a restrained war that lasts for a long time. IHL therefore ought to permit belligerents to dispense with its rules where such dispensation improves their prospects of expediting victory.

Gabriella Blum is more vivid:

[T]he taboo surrounding any mathematical calculation of deliberate killings necessarily detracts our attention from the would-be casualties of Operation Downfall [a land invasion of Japan that the American military would have pursued had the atomic bombs not been dropped and had Japan not surrendered]. Real victims are the only ones we can see and count. Their tragedy is visible and certain. Imaginary victims are, by definition, imaginary. The absolute rules of IHL exclude calculations that would allow us to prefer the welfare of would-be victims. Considerations for the latter would require us to accept, to some extent, the legitimacy of a deliberate infliction of harm on innocent people in order to avoid the infliction – deliberate or foreseen – of harm on still many more. We would have to accept, to some extent, the instrumentalization of innocent people.¹²¹

Blum proposes formalizing the underlying idea into a “humanitarian necessity” defence.¹²² If pleaded successfully, this act-utilitarian defence would justify non-conformities with IHL's unqualified prohibitions,¹²³ including “assassinations of foreign leaders, the Early Warning Procedure, and, in some extreme cases, even the deliberate killing of civilians or combatants who are hors de combat”.¹²⁴ Given the flatness of international society, with no monopoly of State power and no central authority to make lesser-evil choices for States, “we might well imagine a less constraining, more strictly utilitarian paradigm” when designing the humanitarian necessity defence than would be the case for the necessity defence in domestic law.¹²⁵ Consequently, according to Blum,

[a] workable definition of a humanitarian necessity justification might read as follows:

A person shall not be criminally responsible if, at the time of that person's conduct:

…

The conduct that is alleged to constitute a crime was designed to minimize harm to individuals other than the defendant's compatriots, the person could reasonably expect that his or her action would be effective as the direct cause of minimizing the harm, and these were no less harmful alternatives under the circumstances to produce a similar humanitarian outcome.¹²⁶

Five key aspects of this defence may be noted. First, according to Blum, humanitarian necessity is a justification located “within IHL, not outside it”.¹²⁷ In other words, it is IHL itself that ought to permit non-conformity with its rules on account of humanitarian necessity. Second, Blum includes the welfare of enemy civilians and soldiers,¹²⁸ but not that of one's own civilians and soldiers,¹²⁹ in her lesser-evil calculus. It may be said that Blum's is a utilitarian argument that is more than impartial or agent-neutral – it is, in fact, altruistic. Third, when calculating the amount of reduced evil that this defence needs in order to work, Blum considers a “strict utilitarian” or act-utilitarian reduction but ultimately opts for a more demanding “significant” reduction.¹³⁰ The latter reflects our need to be “suspicious of the decisionmaker's ability to make lesser-harm determinations” and to counter risks of uncertainties, slippery slope exploitations, spillover effects and moral hazards.¹³¹

Fourth, Blum proposes to impose a “direct” causal link requirement between the violation and the harm abated.¹³² At first, this may appear to exclude from the scope of eligible harm abatement a “military victory” and an “end to the suffering of enemy combatants and civilians”.¹³³ The picture changes, however, when Blum continues:

Undoubtedly designed to achieve victory and end the war, the attacks [on Hiroshima and Nagasaki] might nonetheless be justified if we were to determine that under the prevailing political and military circumstances, winning the war was the only thing left to do. Continuing well after Germany's surrender, the final active front of a six-year-old world conflict had withstood the fire-bombings of Tokyo and the invasion of Okinawa and the hundreds of thousands of casualties that it had left behind. Perhaps this is the sui generis case in which a military victory that ends the war is a warranted exception.¹³⁴

Fifth, Blum adds a condition that, all else being equal, the evil chosen must be “the least possible harmful means that could avert the greater evil”.¹³⁵ Whether this does anything to affect the defence's fundamentally utilitarian rationale is unclear, for, according to act utilitarianism, the right course of action at a given moment is that amongst the available options which is likely to maximize net utility. Vis-à-vis the common greater evil that the agent endeavours to avert, the least harmful means is, by definition, also the most net utility-maximizing means.

If the Franco-Prussian War had been governed by today's IHL, and if today's IHL had contained Blum's humanitarian necessity defence, Moltke would have pleaded that defence on the grounds that:

1. a. those acts committed by Prussian forces in non-conformity with IHL were designed to minimize harm to French combatants and civilians – not by securing their capitulation, but rather by bringing the inevitability of their capitulation forward;

2. b. Prussian commanders reasonably expected that their action would be effective as the direct cause of minimizing harm to French combatants and civilians through their accelerated surrender; and

3. c. there were no less harmful alternatives under the circumstances to produce a similar humanitarian outcome for French combatants and civilians.

This truly encapsulates Moltke's “greatest kindness in war”.¹³⁶

Utilitarian ends versus utilitarian means

Blum's humanitarian necessity defence has faced detailed rebuttals elsewhere.¹³⁷ At stake here is the notion's fitness with IHL, given what IHL is for and how it sets out to accomplish its goal. This should also be a matter of interest to Blum, since, as she claims herself, the defence should make sense “as judged by IHL's own standards”.¹³⁸

It is evident that modern IHL's end state has been essentially utilitarian – i.e., a net reduction in wartime harm and suffering.¹³⁹ As early as 1868, States “by common agreement fixed the technical limits at which the necessities of war ought to yield to the requirements of humanity” and applied such limits to the use of lightweight explosive projectiles in the form of the St Petersburg Declaration.¹⁴⁰ Though somewhat less forcefully, both Hague Convention II of 1899 and Hague Convention IV of 1907 proclaim in their preambles that the wording of their provisions was “inspired by the desire to diminish the evils of war so far as military necessities permit”.¹⁴¹ In its 1996 Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons, the International Court of Justice held that “the principles and rules of law applicable in armed conflict – at the heart of which is the overriding consideration of humanity – make the conduct of armed hostilities subject to a number of strict requirements”.¹⁴²

Commentators agree. Writing in 1952, Hersch Lauterpacht observed:

We shall utterly fail to understand the true character of the law of war unless we realize that its purpose is almost entirely humanitarian in its literal sense of the word, namely, to prevent or mitigate suffering and, in some sense, to rescue life from the savagery of battle and passion.¹⁴³

Theodor Meron, meanwhile, highlights a shift in the weight assigned to military necessity and constraint on the conduct of belligerents: “Humanitarian restraint has been promoted more vigorously in normative developments and in the elaboration of new standards.”¹⁴⁴

It would be a mistake to assume, however, that utilitarian ends necessarily justify, let alone obligate, utilitarian means – including, in particular, act-utilitarian ones.¹⁴⁵ This is arguably where Blum errs, for she writes:

IHL very clearly states its own goal as maximizing humanitarian protections from harms of inevitable wars. In other words, IHL, as it stands, is the epitome of the principle of lesser evil: the taming of warfare at the price of granting a legal imprimatur for all actions not strictly forbidden.¹⁴⁶

If IHL really were act-utilitarian in the way it creates and administers its rules, then two things should follow. First, as noted earlier, act utilitarianism dictates that it is right to maximize general utility and that it is wrong not to do so. IHL would therefore obligate, rather than merely permit, justify or excuse, those acts that are covered by humanitarian necessity.¹⁴⁷ And yet, at no point does Blum suggest that this is so.¹⁴⁸ Her claim may rather be that IHL should permit, justify or excuse actions taken in humanitarian necessity insofar as they are what act utilitarianism morally obligates.¹⁴⁹ This, however, would still not make IHL itself a system of act-utilitarian rules.

Second, a truly act-utilitarian system of rules would keep the agent, as well as his or her choices and actions, at the centre of its evaluative focus.¹⁵⁰ Legislated norms are merely instruments at the agent's disposal. Weighty rules, typically couched in unqualified terms, signal to the agent how important and valuable it is for him or her to internalize them, even non-instrumentally, should that agent find him- or herself acting in a relevant situation.¹⁵¹ What unqualified rules do not signify from an act-utilitarian point of view, however, is that the legislator has predetermined the unlawfulness of their non-conformity in advance or withdrawn the agent's competence to determine the matter him- or herself on the spot.

There is no indication that this is how IHL works. On the contrary, IHL development progressively narrows situations where “the cases not provided for should, for want of a written provision, be left to the arbitrary judgment of the military commanders”.¹⁵² Where unqualified rules are adopted,¹⁵³ modern IHL replaces its duty-bearers acting in specific situations with its legislators deliberating in advance, as the arbiters of competing interests¹⁵⁴ – including utility considerations¹⁵⁵ – and, with it, as the deciders of rule conformity.

This transfer of agency can be discerned in several ways. At its most structural, it is implicit in the fact that IHL governs warfare, where necessity interests often reign supreme. Thus, the 2001 Articles on Responsibility of States for Internationally Wrongful Acts disqualify a State from invoking necessity as a circumstance precluding wrongfulness if “the international obligation in question excludes the possibility of invoking necessity”.¹⁵⁶ This exclusion, according to the International Law Commission, encompasses “humanitarian conventions applicable to armed conflict”.¹⁵⁷ The transfer may also manifest itself through the operation of IHL provisions themselves. For instance, under the 1954 Hague Cultural Property Convention, States may waive their obligations to respect cultural property “where military necessity imperatively requires such a waiver”.¹⁵⁸ This rule itself contains no further details, leaving imperative military necessity determinations to each belligerent concerned. Cultural Property Protocol II of 1998 significantly tightens the scope and manner of such determinations by limiting when and what kind of cultural property may be subjected to a waiver, as well as which official is competent to take the decision.¹⁵⁹

Perhaps more dramatically, IHL lawmakers may “with a stroke of the pen”¹⁶⁰ eliminate room for discretionary utility-calculation exercises by commanders on the ground. The preamble of the St Petersburg Declaration is a case in point, at least as far as the use of lightweight explosive projectiles is concerned. The preamble reads, in relevant parts:

[T]he Undersigned are authorized by the orders of their Governments to declare as follows:

Considering:

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 …¹⁶¹

Step by step, these stipulations restrict the scope of rule conformity determination on the employment of arms that would remain open to in loco assessment by belligerents. States effectively disempower themselves – and, by extension, their soldiers – from evaluating

1. 1. whether, in a given war, it is legitimate for them to endeavour to accomplish objects other than weakening the military forces of their enemy;

2. 2. whether, in a given military engagement, disabling the greatest number of enemy combatants is sufficient for the aforementioned purpose, i.e., weakening the military forces of their enemy; or

3. 3. whether, in a given case of enemy disablement, the use of lightweight explosive projectiles  

   1. a. aggravates the sufferings of disabled enemy combatants or renders their death inevitable;

   2. b. exceeds the aforementioned sufficiency, i.e., maximum enemy disablement; or

   3. c. is contrary to the laws of humanity;

and, consequently, from justifying a given instance of non-conformity with the ban on the use of lightweight explosive projectiles.

France, Prussia and the North German Confederation were among the States that negotiated the St Petersburg Declaration and became parties to it on 11 December 1868.¹⁶² Accusations of explosive bullets use by France and Prussia alike arose during the Franco-Prussian War.¹⁶³ Let us suppose that the St Petersburg Declaration did apply to that war,¹⁶⁴ and that Prussia did use such munitions against French soldiers.¹⁶⁵ Prussia would have been precluded from pleading, even if true in the particular circumstances of fighting, that its commanders found the wounds which the explosive bullets inflicted on their opponents not uselessly aggravating, that this tactic weakened France's military forces, and so on.

This preclusion is markedly not act-utilitarian. It is little wonder, then, that Moltke – for whom IHL provisions ought to maintain “elasticity”, with situation-sensitive qualifiers, lest they be “broken by inexorable reality”¹⁶⁶ – found the St Petersburg Declaration so disagreeable.¹⁶⁷

Ultima ratio humanitatum?

It may be objected that no law can anticipate all eventualities and have the necessary rules ready for them in advance.¹⁶⁸ There will inevitably come a point where the legislator has not accounted for all relevant interests and where, for all its appearance of certainty, the resulting legal rule fails to provide authoritative behavioural guidance. Warfighting epitomizes situations where this may occur. States, qua IHL lawmakers, should perhaps restore a degree of agency to States and their soldiers, qua belligerents engaged in combat, in the assessment of competing interests and the determination of rule conformity on a case-by-case basis.

Two potential solutions present themselves. One involves affixing exemptions to principal IHL obligations. Indeed, some IHL rules explicitly envision flexibility that is broadly utilitarian. In Gary D. Solis’ words,

[t]hese allowances in the name of military necessity demonstrate that [IHL] remains cognizant that military necessity sometimes requires extreme measures. In these allowances, terms like “if possible,” “as far as possible,” and “if urgent,” introduce elements of uncertainty and risks of arbitrary conduct. Without these concessions, which take reality into account, the allowances could not have been formulated and approved in the first place.¹⁶⁹

Not only military necessity interests but also humanitarian interests are eligible for such allowances, depending on the circumstances and the conduct in question.¹⁷⁰ IHL also contains provisions with clauses that expressly admit deviations on account of military necessity,¹⁷¹ or humanity,¹⁷² or both.¹⁷³ These examples denote instances where IHL lawmakers have specifically and deliberately elected to let the duty-bearer retain an element of agency to evaluate interests and determine rule conformity on their own. In such instances, act-utilitarian considerations may very well inform the duty-bearer's interest evaluations and rule conformity determinations.

The other potential solution concerns IHL rules that are unqualified. Our discussion thus far suggests that the absence of qualifiers indicates the inadmissibility of de novo pleas in defence of non-conformity with such rules. This remains so even when non-conformity proves to be the morally right thing to do – e.g., maximizing net humanity and fulfilling altruistic humanitarian imperatives.

Admitting de novo pleas in such situations somewhat resembles what the present author has called Humanitätsgebot elsewhere.¹⁷⁴ According to this idea, “compliance with [humanitarian imperatives] may be invoked as a ground for non-compliance with unqualified IHL rules”.¹⁷⁵ Humanitätsgebot emanates from two possibilities, namely: (a) that IHL law-making may not have resolved in advance all genuine norm conflicts between its unqualified rules and contrary moral demands¹⁷⁶; and, consequently, (b) that the latter “may operate as additional layers of lawfulness modification over and above positive law for particular acts”.¹⁷⁷ Understood thus, it might appear as though Humanitätsgebot is functionally synonymous with the humanitarian necessity defence. After all, they both purport to facilitate deviations from unqualified IHL rules on account of weighty countervailing considerations of humanity.

Four key differences separate Humanitätsgebot from humanitarian necessity. Firstly, they do not treat the same kind of considerations as eligible. On the one hand, humanitarian necessity justifies the deliberate infliction of evil – such as assassinating rogue leaders, forcibly using residents of occupied territory for the occupier's military needs, or killing enemy civilians or combatants who are hors de combat ¹⁷⁸ – that passes a threshold characterized by altruism,¹⁷⁹ significant and causally proximate harm reduction,¹⁸⁰ and the absence of less evil alternatives.¹⁸¹ On the other hand, Humanitätsgebot deals exclusively with opposing imperatives of which none are evil in themselves. This is evident in the two emblematic cases that Humanitätsgebot examines, namely:

• • declining to repatriate those prisoners of war after the cessation of hostilities for whom non-refoulement concerns exist,¹⁸² despite Article 118 of Geneva Convention III (GC III);¹⁸³ and

• • interning those prisoners of war aboard military vessels at sea who lack access to suitable facilities on land nearby,¹⁸⁴ despite Article 22 of GC III.¹⁸⁵

Secondly, the nature of the norm conflict involved is also different. It is exceedingly unlikely that, for Blum, assassinations, civilian killings and the like are imperatives (e.g., “you must assassinate”) vis-à-vis which IHL's counter-prohibitions should be juxtaposed (e.g., “you must not assassinate”). Rather, humanitarian necessity's normative juxtaposition would be of the kind where an IHL prohibition (e.g., “you must not assassinate”) frustrates the measure that the actor should be entitled to take (e.g., “you may assassinate”). Of concern to Humanitätsgebot are conflicts between mutually incompatible obligations (e.g., “you must repatriate” versus “you must not repatriate”, “you must intern on land” versus “you must intern aboard vessels at sea”).

Thirdly, dissimilar norm conflicts give rise to dissimilar consequences. The humanitarian necessity defence merely releases the actor from the normative burden of non-conformity with IHL. Humanitätsgebot would operate in such a way that IHL, all things considered, obligates non-conformity with its own rules.

Lastly, by the author's own admission, Humanitätsgebot is “no more than a hypothesis”.¹⁸⁶ That this is so becomes evident given how Humanitätsgebot “arguably offers a more cogent explanation” about the non-conclusiveness of GC III Article 118's normative content, but only “if it does exist”.¹⁸⁷ Also, Humanitätsgebot's hypothetical existence is fundamentally an epistemic one (“does it exist?”), not a prescriptive one (“should it exist?”). A “careful scrutiny” of this hypothesis is advisable “in view of [its] potentially far-reaching ramifications”.¹⁸⁸

In contrast, Blum presents her humanitarian necessity defence as an overtly programmatic remedy to IHL's supposed defects:¹⁸⁹

I argue that the law's current absolutist stance prevents parties in conflict from lawfully pursuing actions that might lessen the harms of war. Specifically, I argue that it is possible to conceive of a humanitarian necessity justification, a variation on the necessity defense in domestic criminal law, which would exempt those who pursue such actions from criminal liability. Such an exemption, I hope and expect, could further the humanitarian goals of IHL without eroding its rules and status. What is required for such an exemption to work is a willingness on our part to shift the focus of care from the immediate, visible victims to would-be, invisible ones who are nonetheless just as real. We must not inertly accept certain victims as the necessary collateral damage of war, but instead make a positive and genuine choice to protect many by harming the few.¹⁹⁰

IHL takes an absolutist stance on the issue of deliberately killing enemy civilians who take no direct part in hostilities. Blum's charge is that, if IHL's absolutism is deontologically driven, then it is:

1. a. a hard position to maintain amidst the prevalence of various kinds of evil in war;¹⁹¹

2. b. inconsistent with the permissibility of incidental harm and the deliberate killing of soldiers;¹⁹²

3. c. a position that makes it difficult to explain unqualified prohibitions that do not appear deontologically grounded;¹⁹³

4. d. unable to explain the existence of provisions subject to military necessity exceptions;¹⁹⁴ and

5. e. irreconcilable with the existence of those deontologists who concede the inevitability of utilitarian solutions in extreme cases.¹⁹⁵

Plainly, none of these deontological defects, even if correct in themselves, shows why IHL should not take an absolutist stance on a particular matter. For the same reason, Blum's exposé of rule utilitarianism's shortcomings fails to explain why IHL should reject the prohibition of deliberate civilian killings in all circumstances and replace it with their occasional permissibility instead.¹⁹⁶

Act utilitarianism is incompatible with IHL because, while both share an ultimate concern for net reduction in wartime suffering, IHL does not treat as justified – let alone obligatory – any means likely to contribute to that end. Incompatibility also emanates from the fact that IHL's unqualified obligations are not merely there to signal their elevated instrumental value to belligerents. On the contrary, these obligations indicate predetermined parameters of rule conformity from which their addressees are no longer entitled to deviate on account of utility considerations.

It is not beyond the realm of theoretical possibility that IHL may obligate non-conformity with its own rules when so demanded by contrary humanitarian imperatives. IHL justifying killing enemy civilians on act-utilitarian grounds – no matter how altruistic and restrictive – is.

Conclusion

We had not been given any warning beforehand that our houses were going to be burned. No one in the whole ridge knew that we were to move. The police just came one day, and drove everybody out of their homes, while the Home Guards burned the houses right behind us. … During the move I got separated from my children, and I could not trace them. They had been in front, leading our remaining cattle, but I failed to find them. During the whole night I could hear a lot of shooting and screaming. I cried the whole night, knowing that my children were gone.

Ruth Ndegwa¹⁹⁷

In 1952, Kenya, then under British rule, found itself engulfed in an insurgency.¹⁹⁸ This protracted, increasingly violent and fast-expanding conflict pitched Britain's colonial forces against a group of national liberation guerrillas popularly known as the Mau Mau. These rebels were funded and coordinated primarily through extensive Kikuyu networks in Nairobi.¹⁹⁹ On 24 April 1954, British forces launched Operation Anvil. After four weeks of intense cordon and search action, 30,000 suspected Mau Mau members in the Kenyan capital were arrested, screened and “repatriated” to Kikuyu reserves in the Central Province.²⁰⁰

Two months later, the colonial War Council mandated mass forced resettlement, called “villagization”, throughout the province.²⁰¹ Ruth Ndegwa, left alone to care for her children and elderly relatives following the arrest and detention of her husband, was removed from her homestead and detained at a “protected village” in Kiamariga, a site that was surrounded by armed guards and had no shelter, food, water, sanitation facilities or medical supplies.²⁰² At Kiamariga, Ndegwa and her fellow detainees were forced to engage in hard labour. They first built huts for themselves and dug trenches around the village perimeters to be barbed-wired, then dug trenches and graves in the forest. Throughout this ordeal, village guards subjected their captives to rampant starvation, filth, disease, beating, humiliation, rape, torture and execution.²⁰³

The sudden and drastic deterioration in the circumstances of Ndegwa's life – as well as those of over 1 million Kikuyu forced into villagization²⁰⁴ – was the result of a shift in the colonial administration's wartime utility calculus. Since the uprising began in 1952, it had taken the British almost three years to gain the upper hand in battle.²⁰⁵ Colonial authorities introduced Kikuyu villagization in order to cut supplies to Mau Mau fighters in the forest²⁰⁶ and break popular support for the group.²⁰⁷ In the event, villagization proved instrumental in the Mau Mau's 1956 military defeat.²⁰⁸

If British colonial officials at the time were to offer any act-utilitarian rationale for villagization programmes, it would be something along the following lines:

All else being equal, the acute yet short-term suffering to which the villagized Kikuyu would now be subjected is less inhumane overall than the moderate yet long-term suffering to which they had been subjected earlier in their Central Province reserves.

Our discussion exposes flaws in the empirical and normative underpinnings of such a claim. The notion that causally links toughening wars to shortening them has an uncertain evidentiary basis at best. It also conflates the control that a belligerent wields over the amount of violence it chooses to inflict on the enemy with the control it seeks to exert over the enemy's will to resist. More importantly, by claiming that brutal but swift wars are better for humanity than restrained but drawn-out ones, the party resorting to brutality usurps authority to decide what the brutalized party's role is in maximizing net humanity. In addition, the brutalizer unwarrantedly blames its opponent for the cruelty to which it resorts in the name of greater net humanity.

Legally, act utilitarianism – the moral framework upon which these empirical and normative claims rest – is incompatible with IHL. While it is indeed IHL's ultimate aim to reduce net wartime harm, when it enacts unqualified rules, it predetermines their conformity or non-conformity through processes that are not act-utilitarian. Despite some assertions to the contrary, nowhere in these processes do lesser-evil justifications naturally belong.

Inhumane though it admittedly was, the scale of hardship and IHL non-conformity that Ndegwa and others in her Kikuyu community had endured early in the insurgency paled in comparison to the scale of hardship and IHL non-conformity that they suffered through villagization. Explaining all that away by reference to act-utilitarian considerations would be bound to sound hollow. Indeed, much like Moltke's “greatest kindness”, Britain's “benevolent trustee[ship]”²⁰⁹ behind mass Kikuyu villagization revealed itself to be little more than the wartime brutalizer's unsubstantiated and self-serving pseudo-kindness.²¹⁰