Blog post

665: Calculating the Lesser Evil

Eyal Weizman 6 April 2016

Image for blog post entitled 665: Calculating the Lesser Evil

Following the recent escalation and justification of US drone strikes in the name of national security, Eval Weizman reminds us how the demand for calculation paves the way for a continued justification of violence. The following excerpt from his book charts the philosophical underpinnings of Western military and humanitarian intervention from the late twentieth century to the present. 

Drone pilots at work, via Wikimedia Commons


If, as a friend recently suggested, we ought to construct a monument to our present political culture as an homage to the principle of the ‘lesser evil’, it should be made in the form of the digits 6-6-5 built of concrete blocks, and installed like the Hollywood sign on hillsides or other high points overlooking city centers. This number, one less than the number of the beast — that of the devil and of total evil — might capture the essence of our humanitarian present obsessed with the calculations and calibrations that seek to moderate, ever so slightly, the evils that it has largely caused itself.

The principle of the lesser evil is often presented as a dilemma between two or more bad choices in situations where available options are, or seem to be, limited. The choice made justifies the pursuit of harmful actions that would be otherwise deemed unacceptable in the hope of averting even greater suffering. Sometimes the principle is presented as the optimal result of a general field of calculations that seeks to compare, measure and evaluate different bad consequences in relation to necessary acts, and then to minimize those consequences. Both aspects of the principle are understood as taking place within a closed system in which those posing the dilemma, the options available for choice, the factors to be calculated and the very parameters of calculation are unchallenged. Each calculation is undertaken anew, as if the previous accumulation of events has not taken place, and the future implications are out of bounds.

Those who seek to justify necessary evils as ‘lesser’ ones, especially when searching for a rationale to explain recent wars and military expeditions, like to appeal to the work of the fourth-century North African philosopher—theologian St Augustine. Augustine’s rejection of the principle of Manichaeism — a world divided into equally powerful good and evil — meant that he no longer saw evil as the perfect mirror image of the good; rather, in platonic terms, as a measure of its absence. Since evil, unlike good, is not perfect and absolute, it is forever measured and calibrated on a differential scale of more and less, greater and lesser. Augustine taught that it is not permissible to practice lesser evils, because to do so violates the Pauline principle ‘do no evil that good may come’. But — and here lies its appeal — lesser evils might be tolerated when they are deemed necessary and unavoidable, or when perpetrating an evil results in the reduction of the overall amount of evil in the world.

In relation to the ‘war on terror’, the terms of the lesser evil were most clearly and prominently articulated by former human rights scholar and leader of Canada’s Liberal Party Michael Ignatieff. In his book The Lesser Evil, Ignatieff suggested that in ‘balancing liberty against security’ liberal states establish mechanisms to regulate the breach of some human rights and legal norms, and allow their security services to engage in forms of extrajuridical violence — which he saw as lesser evils — in order to fend off or minimize potential greater evils, such as terror attacks on civilians of western states. If governments need to violate rights in a terrorist emergency, this should be done, he thought, only as an exception and according to a process of adversarial scrutiny. ‘Exceptions’, Ignatieff states, ‘do not destroy the rule but save it, provided that they are temporary, publicly justified, and deployed as a last resort.’  The lesser evil emerges here as a pragmatic compromise, a ‘tolerated sin’ that functions as the very justification for the notion of exception. State violence in this model takes part in a necro-economy in which various types of destructive measure are weighed in a utilitarian fashion, not only in relation to the damage they produce, but to the harm they purportedly prevent and even in relation to the more brutal measures they may help restrain. In this logic, the problem of contemporary state violence resembles indeed an all-too-human version of the mathematical minimum problem of the divine calculations previously mentioned, one tasked with determining the smallest level of violence necessary to avert the greatest harm. For the architects of contemporary war this balance is trapped between two poles: keeping violence at a low enough level to limit civilian suffering, and at a level high enough to bring a decisive end to the war and bring peace.

More recent works by legal scholars and legal advisers to states and militaries have sought to extend the inherent elasticity of the system of legal exception proposed by Ignatieff into ways of rewriting the laws of armed conflict themselves.  Lesser evil arguments are now used to defend anything from targeted assassinations and mercy killings, house demolitions, deportation, torture, to the use of (sometimes) non-lethal chemical weapons, the use of human shields, and even ‘the intentional targeting of some civilians if it could save more innocent lives than they cost.’ In one of its more macabre moments it was suggested that the atomic bombings of Hiroshima might also be tolerated under the defense of the lesser evil. Faced with a humanitarian A-bomb, one might wonder what, in fact, might come under the definition of a greater evil. Perhaps it is time for the differential accounting of the lesser evil to replace the mechanical bureaucracy of the ‘banality of evil’ as the idiom to describe the most extreme manifestations of violence. Indeed, it is through this use of the lesser evil that societies that see themselves as democratic can maintain regimes of occupation and neo-colonization.


The diffuse body of customs and conventions that make up jus in bello, the laws of war otherwise known as international humanitarian law (IHL), have since the end of the Cold War increasingly become the frame within which the calculation and application of military violence takes place. In recent decades IHL has also become an important part of global political culture. Debates about conflicts and occupations from Kosovo to Afghanistan to Iraq tend to use the terminology of IHL. The juridical categories of ‘necessity’ and ‘proportionality’ seem to be among the most popular terms employed in designing and monitoring state violence. This applies also to the antiwar movements: it is now not uncommon to see demonstrators carrying banners with slogans bearing references to these clauses in the law (as in the oft-invoked recrimination of ‘war crimes’) or to specific legal principles such as proportionality (‘disproportional attack’).

Those protesting in the name of the law must remember, though, that IHL does not seek to end wars but rather to ‘regulate’ and ‘shape’ the way militaries wage them; and that western militaries, increasingly bogged down by a raft of urban insurgencies in various global arenas, are also keen to change the way they fight wars and to minimize civilian casualties. Western militaries tend to believe that by moderating the violence they perpetrate, they might be able to govern populations more efficiently and even finally win over the hearts and minds that have continuously eluded them since the British Malayan counterinsurgency and the Vietnam War.

Within the frame of international humanitarian law the clearest manifestation of the lesser evil principle is the principle of proportionality. This is, of course, embedded in almost every civil legal code. Different versions of it have been used to describe different types of balancing acts, most often in situations where some rights contradict others, or when individual rights are weighed against public interests, or against administrative or economic policies. Within the context of IHL, however, proportionality is a moderating principle that seeks to constrain the use of force. The principle was implicit in most international conventions on the use of force but was formally codified only in 1977, in Protocol I of the Geneva Conventions. The protocol’s wording prohibits ‘an attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated’ (my emphasis). Proportionality thus demands the establishment of a ‘proper relation’ between ‘unavoidable means’ and ‘necessary ends’. While considering the choice of military means, the principle calls for a balance to be established between military objectives and anticipated damage to civilian life and property. Proportionality is thus not about clear lines of prohibition but rather about calculating and determining balances and degrees. In incessantly calculating for the least of all means possible, it embodies something of Pangloss’s principle.

The purpose of proportionality is not to strike a perfect balance, but rather to ensure that there is no excessive imbalance. Nevertheless it is about the ‘too much’ — but how much is too much? Although violence is in constant need of measurement, the principle of proportionality provides no scale, no formulas and no numerical thresholds. Instead, it demands assessment on a case-by-case basis, within parameters that are always relative, situational and immanent. It demands the estimation of aims, impacts and side effects, intended and unintended consequences — which is also to say, the measurement of lesser and greater evils, their exchange and sometimes even transfer — in an economy of a real or imaginary ‘worst-case scenario’ and in an attempt to keep the overall level of violence to the minimum necessary. By opening a field of equivalence, in which different forms of potential and actual violence, risk, and damage become exchangeable, proportionality approximates an algorithmic logic of computation — although, still, in practice, it is rarely computed.

Military lawyers and experts in international humanitarian law are the first to accept the fact that the predictions required for proportionality analysis are always contingent, immanent and prone to subjective interpretations. Contemporary military debates about IHL concern precisely the impossibility of bringing together in practice the legal demand that violence be measured, and the impossibility of doing so. Like the finance specialists who acknowledge the impossibility of prediction but do little else than calculate, the economists of violence are incessantly weighing their options and hedging their risks under the assumption of unpredictability and uncertainty. It is the very act of calculation — the very fact calculation took place — that justifies their action. Indeterminacy, the very principle that makes the economies of liberal capitalism generate profit, or burst after a sequence of failures, is also central to the conduct and potential outcomes of the contemporary wars.

But along with the growing capacity of technological means the incalculability of their consequences also grows. Some military lawyers think that indeterminacy will always work in their favor. Others, allergic to the idea of vagueness, see in technology an opportunity to dispel inherent uncertainties and incalculability. Daniel Reisner, former head of the International Law Division in the Israeli military, is of the latter kind. In a conversation he described to me the problems of calculating the economy of violence on the proportionality principle, and later his attempt tried to dispel something of the ethical/legal fog surrounding the question:

Proportionality is a complex logic with many variables — but how do you compare these? There is no choice but to ask the question, compare and calculate. Proportionality does not tell us what to include in the calculation, what is the equation and what is the exchange rate. Should a man of combatant age be counted as a civilian? If so, does he count for more or for less? How do you count women in relation to men? How do you count the death of children? Does one dead child equal one dead grownup, or does he equal five grownups? As a lawyer I need numbers to work with. I need thresholds in order to instruct the soldiers. Any number could become a useful benchmark. But when the ground of the law is shaking I am also unstable.

The legal definition of civilian does not, of course, involve any distinctions by gender and age; civilian life is civilian life and children are legally considered equal to adults. But lawyers’ insistence on the fine details of a necro-economy, and the conversion rates between people of different genders and ages, is explained by the fact that proportionality has become a means to an end: measuring the public legitimacy of an act of violence. In this arena there is indeed a different meaning attached to the killing of children or women.

Lacking any other criterion for measurement, death ratio is one of the gruesome ways in which proportionality is calculated and managed in practice. It has its macabre side effects, too. In a 2002 meeting of a team of experts on law and military ethics, Reisner challenged his colleagues to an experiment. He asked each of them what ratio of ‘collateral civilian death’ — how many civilians killed — they considered to be legitimate in the context of a specific scenario that he recounted, of an armed militant about to be killed by the Israeli military. Each of his colleagues wrote down a number of civilian deaths they’d accept as legitimate under the principle of proportionality. The numbers were then counted and collated, and an average was calculated. It was 3.14 — very approximately the mathematical constant π whose value is the ratio of a circle’s circumference relative to its diameter in Euclidean space.

Another instance of calculation, while not referring directly to proportionality, embodies this grotesque logic of necro-economy in practice. In 2002, while still a general in the Israeli military, Itzhak Ben Israel, now a professor of physics at the Tel Aviv University and chairman of Israel’s space agency, was in charge of the ‘weapons and technological infrastructure research and development directorate’. There he developed an equation based on systems theory in order to predict the necessary number of people the Israeli military must eliminate from a militant organization by arrest or targeted assassinations in order to defeat it.

The formula was Q=1—(q ln q + 1/q ln 1/q). In this equation, which seeks to apply the entropic behavior of molecules in a gaseous state to military and political matters, Q stands for the probability that the organization will collapse and q is the percentage of militants you kill. To put it simply, if you kill (or neutralize in other ways) 20—25 per cent of the members of an organization — any organization — there is an 85 per cent likelihood that the confusion and knowledge loss generated will lead to its collapse. If you kill 50 per cent, the formula has it, the result converges on a 100 per cent probability that it will collapse.

Proportionality’s system of calculations approximates models applied in the insurance industry to assess risk. Risk analysis developed indeed as a means of determining the probabilities of bad things occurring, their potential for damage and their spatial or systemic distribution. For the military, risk is the means of determining the probability of destruction and injury to personnel and equipment and their potential severity. The conception of risk is central to the calculation of proportionality, especially when attempts to minimize civilian casualties is measured against potential harm to soldiers. The ‘trade off’ of risk means that reducing risk to the attacking military tends to increase the risk to civilians. One of the clearest examples for this ‘risk transfer war’ was NATO’s bombing of Kosovo and Belgrade in 1999. This was mainly due to the decision to conduct high altitude aerial attacks that reduced the danger to NATO air force, but dramatically increased it for the civilians on the ground. The result — no combat fatalities among NATO forces compared with five hundred civilians killed by the bombardment — was understood by many international law scholars as an indication of a breach of the proportionality principle. This case also demonstrated that the balance expected in proportionality has a territorial dimension. Different calculations, formulas, balances and death ratios are deemed appropriate to state militaries in different zones of action and across different borders.

Military violence endeavors not only to bring death and destruction to its intended targets but also to communicate with its survivors — those that remain, those not killed. The laws of war have become one of the ways in which military violence is interpreted by those who experience it, as well as by global bystanders. It could thus be said to have a pedagogical pretension. It is a violence that should not only convince but also manufacture the possibility for conviction. In contemporary war, the principle of proportionality has become the main translator of the relation between violence, law and its political meaning.

The communicative dimension of military threats can function only if gaps are maintained between the possible destruction that an army is able to inflict and the actual destruction that it does inflict. It is through the constant demonstration of the existence and size of this gap that the military communicates with the people it fights and occupies. Sometimes the gap opens wide, such as when the military governs the territories it occupies — its violence in a state of potential, existing as a set of threats and possibilities that are not, for the time being, actualized. In a state of war the gap closes — but rarely does it do so completely. Even in the most brutal of wars, something of the gap still exists as the stronger side restrains and moderates its full destructive capacity. Restraint is also what allows for the possibility of further escalation, an invitation for those people receiving violence to make their own cost-benefit calculation and opt for consent. A degree of restraint is thus part of the logic of almost every military operation: however bad military attacks may appear to be, they can always get worse. The gap is measured also against ‘the potentiality of the worst’ — an outburst of performative violence without rules, limits, proportion or measures — which has to be demonstrated from time to time. This necessarily creates a precedent against which all other bad events are understood and measured. With the initial recording of ‘the worst’, its reappearance, as Hannah Arendt commented, becomes ever more likely.

The gap thus communicates the potential for destruction without the need for further violence. When the gap between the possible and the actual application of force closes completely, violence loses its function as a language. War becomes total war — a form of violence stripped of semiotics, in which the enemy is expelled, killed or completely reconstructed as a subject. Degrees in the level of violence are precisely what makes war less than total. Game theory, as applied by military think tanks since the early Cold War days of RAND, is conceived to simulate the enemy’s responses, and help manage the gap between actual and potential violence. This practical form of military restraint is now often presented as the adherence to the laws of war.

While symmetric interstate warfare assumes a language that is well understood by both parties, and a rational basis for calculating the losses and profits of war, colonial violence presupposes that the language itself has to be constructed. Colonial wars have often been total wars, because the people colonized were not perceived to share the same humanity as the colonizers, and were therefore not seen as a party capable of rational behavior and discourse. These wars are not about an enemy that has to be convinced but about an irrational people that has to be either reconstructed or killed. In these pedagogical wars, it is the disproportional violence of the madmen that is reserved. If proportionality stands for the ethical, rational aspect of war, in which the economy of balances functions well, then breaking this economy is intended as a message of a different order and magnitude. ‘We will wield disproportionate power against every village from which shots are fired on Israel, and cause immense damage and destruction . . . This is not a suggestion. This is a plan that has already been authorized.’ Similar statements by members of the Israeli security establishment and politicians, including the prime minister, proliferated in the aftermath of the 2006 Lebanon War with Hezbollah, and in anticipation of the 2008 invasion of Gaza where such violence was actualized. Disproportionality — the breaking of the elastic economy that balances goods and evils — is violence in excess of the law, and one that is directed at the law. Disproportional violence is also the violence of the weak, the governed, those who cannot calculate and are outside the economy of calculations. This violence is disproportional because it cannot be measured and because, ultimately, having its justice not reflected in existing law, comes to restructure its basis altogether.

Adapted from The Least of All Possible Evils