Distributing the Harm of Just Wars

In Defence of an Egalitarian Baseline Approach

Sara Van Goozen

• Editorial: Routledge
• Año: 2021
• ISBN:
  9781000364569
• Edición: 1
• Categoría: Philosophy
• Categoría: Ethics & Moral Philosophy

This book argues that the risk of harm in armed conflict should be divided equally between combatants and enemy non-combatants.

International law requires that combatants in war take 'all feasible precautions' to minimise damage to civilian objects, injury to civilians, and incidental loss of civilian life. However, there is no clear explanation of what 'feasible precautions' means in this context, or what would count as sufficiently minimised incidental harm. As a result, it is difficult to judge whether a particular war or offensive actually satisfies this requirement. Just war theorists often consider it common sense that merely not intending to harm innocent civilians is not sufficient, but there is little clarity in the literature regarding what this means. One crucial question that is almost always overlooked is that of what the appropriate baseline distribution of risk should be.

This book defends the Minimal Harm Requirement (MHR), which states that combatants should make an effort to reduce merely foreseen harm to enemy non-combatants to the lowest reasonable level. In order to assess which risk impositions are reasonable, and which are not, an egalitarian baseline should be adopted, suggesting that other things being equal risk of harm should be distributed equally between just combatants and unjust non-combatants.

This book will be of much interest to students of just war theory, ethics, security studies, and international relations.

1 Introduction

Introduction

In several of the wars which Western states have been involved in since the end of the Cold War, the number of non-combatant casualties considerably outnumbered Western combatant casualties. For instance, in the NATO intervention in Kosovo, no NATO personnel were killed, but several hundreds of Serbian and Kosovar civilians were killed, and even more were injured or saw their homes destroyed. In the First Gulf War, 250 Western combatants were killed, compared to over 3,000 Iraqi non-combatants.¹ In the 2014 conflict in Gaza, Operation Protective Edge, 66 Israeli soldiers were killed by attacks from Hamas and other terrorist organisations, whereas according to the UN at least 1,462 Gazan civilians died as a result of Israeli Defense Force (IDF) attacks.² In Afghanistan, in 2008, 522 civilians were killed in airstrikes conducted by the NATO-led International Security Assistance Force (ISAF), and 45% of all civilian casualties were attributed to pro-government forces (largely international forces).³ The number of civilian deaths attributed to the US and her allies in Iraq in the period between the end of the first stage of combat (the invasion) in May 2003 and the end of 2006 has been estimated to be in the region of 5,000, though estimates go as high as 8,685.⁴

Some of these civilian casualties may have been the result of deliberate anti-civilian attacks. However, as Colin H. Kahl shows, US compliance with the rule of non-combatant immunity, which prohibits the direct targeting of non-combatants, in Iraq did, in fact, increase considerably over time, and multiple attempts were made to ensure even better compliance.⁵ Thus, it can be assumed that the majority of the 5,000 or so civilian casualties caused between May 2003 and the end of 2006 were killed unintentionally. Marcus Schulzke similarly shows that considerable efforts were made by the US and British Armed Forces to improve compliance with the requirement of discrimination. He suggests that it is rather the inherent tension between soldiers’ indefeasible right to self-defence on the one hand, and the difficulties associated with identifying legitimate targets on the other, that is a significant cause of civilian casualties.⁶ Similar conclusions can be drawn from Lisa Hultman’s research. Analysing civilian deaths caused by pro-government forces in Afghanistan, she finds that the numbers suggest that the counterinsurgency (COIN) forces are especially likely to cause the deaths of civilians after they have suffered losses in combat.⁷ This suggests that when the counterinsurgents have suffered losses, they are more likely to intensify their efforts and perform operations with greater side-effect risk to civilians. In general, as weapons developed and battlefields changed throughout the 20th century, harm to civilians was increasingly incidental (or collateral). This is in contrast to previous centuries when the more limited range of weapons meant that attacks on civilians by and large had to be intentional: sieges and the destruction of crops explicitly target civilians, and typical close-range weapons such as swords or even bows required combatants to get within a distance of their targets, which would make distinguishing between combatants and non-combatants relatively easy.⁸

These unintended but foreseen civilian casualties are, to some extent, the result of deliberate tactical choices by Western military and political elites. According to Martin Shaw, the 1990s saw the rise of a ‘new Western way of war’, which is characterised by great inequalities of risk between Western combatants on the one hand, and the non-combatants in the countries they were fighting in on the other hand. Shaw calls this style of fighting ‘risk-transfer warfare’.⁹ The tactics employed in such wars aim to reduce the risk to Western combatants, but as a result often increase the risks faced by enemy (and neutral) non-combatants significantly. In Kosovo, for instance, fear of NATO casualties – which were thought to erode public support for the intervention – led NATO to only use air power, and to order their pilots to fly high and out of range from Serbian anti-aircraft missiles.¹⁰ This tendency to reduce risks to combatants, even though doing so foreseeably increases risk to non-combatants, was also visible in the early stages of the US invasion of Iraq.¹¹ The extensive use of armed Uninhabited Aerial Vehicles (UAVs, or drones) in, for instance, Pakistan, Afghanistan, and Yemen is one extreme example of this tactic – here, combatants face no immediate risks at all, because they are far away from the area where the drones are deployed. All the immediate risks are imposed on the people living in the area where the drones are operating. In fact, Bradley J. Strawser, among others, argues that the fact that drones enable pilots to avoid all harm, without prejudicing their mission, can be considered their main ethical benefit.¹² The development of fully autonomous weapons systems (AWS) would push this imbalance to its logical extreme. When employing AWS to engage enemy combatants, nobody (on the side using AWS) is, in principle, exposed to any risk at all, whereas drone operators may still suffer stress, burnout, and PTSD associated with their job.¹³

Of course, non-combatant casualties are often very difficult to avoid completely.¹⁴ Moreover, although the laws of war forbid the direct targeting of non-combatants, they permit unintentional, though foreseeable, injury to non-combatants provided it is a proportionate and necessary side effect of legitimate military activity. However, international law is not silent on foreseeable but unintended harm to non-combatants. Article 51 of the First Additional Protocol to the Geneva Conventions forbids indiscriminate attacks, which are attacks which do not discriminate sufficiently between military and civilian targets, or which have effects which cannot be limited to military targets, as well as the intentional targeting of non-combatants.

Further, Article 57 of the First Additional Protocol requires in §2(a)(ii) that combatants ‘take all feasible precautions in the choice of means and methods of attack with a view to avoiding, and in any event to minimizing, incidental loss of civilian life, injury to civilians and damage to civilian objects’.¹⁵ Thus, the Additional Protocol recognises that it is not enough to simply not intend to cause harm to civilians, but that some precautions must be taken to limit the unintended but foreseen effects of military activities. It fails, however, to specify what it counts as feasible precautions. As such, ‘one of the problems of IHL [International Humanitarian Law] is its inability to provide a clear answer to the question of whether everything has been done to spare civilian lives’.¹⁶

The requirement to reduce merely foreseen harm

The rule expressed in Articles 51 and 57 of the First Additional Protocol, which distinguishes between intended and foreseen harm to non-combatants, reflects the moral principle known as the Doctrine of Double Effect (DDE). The DDE states that, although it is impermissible to cause harm to another person intentionally, it may be permissible to do so as a merely foreseen side effect of an otherwise permissible action. Interpretations of the DDE vary wildly from author to author, but the crucial elements can be summarised, following Michael Walzer, as follows:

DDE: It is permissible to perform an act known to have bad consequences provided

• 1The act is good in itself or at least indifferent.
• 2The direct effect is morally acceptable.
• 3The intention of the agent is good, that is, he aims only at the acceptable effect; the bad effect is not one of his ends, nor is it a means to his ends.
• 4The good effect is sufficiently good to compensate for allowing the bad effect.¹⁷

The first two clauses specify the cases in which the DDE can be applied. The crucial clause is the third one. The idea is that it is worse to intend a bad effect (for instance, to kill an innocent person) than it is to pursue an act which has the same bad effect as a merely foreseen side effect. This is not to say that all merely foreseen side effects are permissible; the fourth clause prohibits overall disproportionate bad effects.

However, like the authors of the First Additional Protocol, most people accept a further requirement that we should take care to limit the merely foreseen harms we cause. It is not enough to merely not intend harm; we should also take steps to reduce any merely foreseen harm. But this raises a problem: in order to limit merely foreseen harm to enemy non-combatants, combatants will often have to take measures that will increase their own risk of suffering harm, decrease their chances of mission success, or both. For instance, they may have to forego using certain tactics, such as high-altitude aerial bombing, and will have to undertake measures in order to ensure that enemy civilians are not present (or are not present in too large a number), which can put the combatants themselves at risk. Thus, combatants face a trilemma: they have to choose between reducing merely foreseen harm to non-combatants, protecting themselves, and achieving the mission they set out to achieve.¹⁸ Because it is likely that attempting to fully achieve one of these will interfere with the others we need to have a good understanding of what the minimum is that we can require of combatants. The question of what can be considered a reasonable amount of risk, and so when combatants can be required to take measures to reduce merely foreseen harm to enemy non-combatants, is the question I address in this book.¹⁹

In this book, I approach this question as one about the fair distribution of the harm which arises from pursuing a just war. The trilemma highlighted earlier is essentially a distributive trilemma – different choices about how much risk will be accepted in the three relevant areas will result in different distributions of harm, and the question thus is which is the fairest distribution of harm. Accordingly, my focus is on distributive fairness as opposed to liability and, as such, the problem I address is merely a subset of the issues raised by the transfer of risk in contemporary warfare. However, a substantive account of what a fair distribution of (risks of) harm between combatants and non-combatants would look like is currently lacking in both the philosophical literature and international law.

The approach which I argue for will be demanding for combatants, though not prohibitively so. After all, just combatants, like enemy non-combatants, have to be treated with respect. The account I present here is, however, largely different from military practice (though perhaps with the exception of modern COIN tactics, which I discuss in some more detail in Chapter 5) where often the burdens of a just war or humanitarian intervention are disproportionately borne by civilians. Of course, even in the context of this type of risk-transfer warfare, combatants can make an effort to reduce the merely foreseen harm they will cause to enemy non-combatants. As will become clear, however, as a strategy, transferring risk from combatants to enemy non-combatants is impermissible insofar as it wrongly prioritises one arm of the trilemma (protecting the just combatants) over others.²⁰

The account that I offer is the most extensive investigation of this particular dimension of just war theory currently available. A few authors have recently discussed problems surrounding the distribution of burdens in a just war (in particular in the context of humanitarian interventions) or proposed to approach these problems as ones of distributive fairness.²¹ Yet there has not been a detailed, comprehensive account of how to establish a fair distribution of risk between combatants and non-combatants. Though I cannot promise to provide prescriptions for every situation, this book provides the necessary first steps for developing a comprehensive account of how to distribute harm in war.

The scope of the argument

Before outlining the argument presented in this book in more detail in the next section,...