Sparing Civilians

Many works in applied moral and political philosophy start from an intuitive underlying moral principle and then argue that it leads to radical implications for our norms and practices. (For example, that a commitment to equality entails that we should abolish immigration restrictions, or confiscate one another's organs, or refuse to read our children bedtime stories.) Seth Lazar's Sparing Civilians works in the other direction. It aims to provide new moral foundations for a touchstone of commonsense morality and law: the idea that there is an extremely stringent moral constraint on killing noncombatants in war. More precisely, Lazar defends a comparative claim:Moral Distinction: In war, with rare exceptions, killing noncombatants is worse than killing combatants. (2)The uninitiated reader might wonder why a defense of Moral Distinction is needed in the first place. As Lazar points out, “If any moral principle commands near universal assent, this one does” (1). But despite widespread endorsement, civilian immunity has been philosophically eroded over the last two decades. This is because contemporary just war theorists primarily attempt to justify war in terms of liability justifications for killing. These hold that individuals can come to lack their normal right not to be killed in virtue of their responsibility for threats of unjustified harm. This focus on liability in war has two important upshots. First, it implies that combatants whose wars are unjust have no moral reason to target enemy soldiers rather than civilians, since all of their targets are innocent (none are morally responsible for threats of unjustified harm). Second, it raises a dilemma when it comes to waging intuitively just wars. In order to show that most enemy combatants lack a right not to be killed, we require a fairly minimal account of liability. The problem, Lazar argues, is that any such account will also render many civilians legitimate targets, because a significant proportion of noncombatants are as morally responsible as combatants for threats of unjustified harm (8–17).In light of these worries, Lazar proposes an alternative picture of moral constraints in war. On this view, we should adopt a much more demanding standard for liability, and therefore accept that many combatants, as well as the vast majority of noncombatants, retain their rights not to be killed. Whether just or unjust, war involves the intentional mass killing of the innocent. For Lazar, the intuitive moral asymmetry between killing soldiers and killing civilians is grounded in asymmetries between different kinds of rights-transgressive killing, and so applies to parties in both just and unjust wars.Over five substantive chapters, Lazar argues for two claims: (i) that there are multiple factors which exacerbate the moral wrongness of rights-transgressive killing (and of harm more generally), and (ii) that these factors typically (though not uniformly) inhere in the killing of noncombatants but not combatants. The first argument (chap. 2) is partly instrumental: killing civilians is worse than killing soldiers because it is more likely to be ineffective and therefore gratuitous. Lazar concludes that the argument is only partly successful: the empirical evidence is mixed and, most important, the protection it affords civilians is unsatisfyingly contingent. The following four chapters offer more principled arguments in defense of Moral Distinction. Chapter 3 draws on the idea that opportunistic killings, which use a person to achieve an end, are morally worse than eliminative killings, which merely remove an obstacle or problem that the target poses. Chapter 4 focuses on moral risk. Lazar argues that killings are objectively worse the higher the epistemic probability (from the killer's perspective) that the target is not liable, and that civilians are on average less likely to be liable than soldiers. Chapter 5 appeals to the idea that civilians are more vulnerable and defenseless in war than soldiers are. Chapter 6 sets out a range of ways in which combatants diminish the stringency of their right not to be killed, such as exposing themselves to risk (often in order to draw fire away from civilians), engaging in morally risky behavior, and incurring role-based duties to prevent their state's wrongdoing.Throughout, Lazar's arguments are novel, ingenious, and frequently convincing (not to mention numerous). In fact, I think Lazar somewhat undersells the book's contribution. Though billed as a focused defense of civilian immunity, the book provides some of the key materials for a markedly different way of approaching the justification of war (and the use of force more generally). Whereas the field is dominated by liability-centric accounts of permissible war, Sparing Civilians sets out an alternative research program, in which lesser-evil justifications for overriding individual rights do the majority of the heavy lifting. More generally still, by exploring the terrain of lesser-evil justifications, and by showing that the topography is more varied than usually assumed, Sparing Civilians has much to offer readers interested in more abstract questions in normative ethics. I have learned a lot from thinking about Lazar's arguments, and anyone engaging with the book will be rewarded. In that spirit, let me offer two critical comments, focusing on chapters 3 and 4.First, as mentioned above, Lazar defends Moral Distinction by appeal to the differential moral gravity of eliminative and opportunistic killing. One important objection concerns whether the eliminative/opportunistic distinction reliably tracks the distinction between killing combatants and killing noncombatants. As Lazar himself points out, almost all killing of combatants has an opportunistic aspect (they are killed in order to change the behavior of their comrades, commanders, and civilian leaders). Moreover, some killing of noncombatants can have an eliminative purpose (killing may prevent noncombatants from contributing to threats) (69–73). In response, Lazar makes an intriguing proposal: that actions can manifest multiple morally relevant modes of agency, and to different degrees. As he puts it:“True, all killing in war is also opportunistic. . . . But still, intentionally killing civilians is more opportunistic than intentionally killing soldiers; and killing soldiers almost always involves an eliminative dimension. Each kind of killing involves mixed kinds of agency. But the mixture in anti-civilian violence is more opportunistic than it is for counterforce attacks. So the generalisation that killing civilians is worse than killing soldiers is borne out.” (72)I confess that I just don't have a clear enough understanding of the idea to evaluate it. Most obviously, what determines an action's relative “quantity” of a particular mode of agency? How do we derive a scalar notion of opportunistic agency from Lazar's earlier counter-factual test for opportunism (65–66)? More generally, what are we comparing the mixtures of? Different token acts of killing? Or broader types? Given the importance of the mixed agency argument (as well as its philosophical interestingness), it's disappointing that Lazar doesn't give us a more developed account of how it works.1My second comment concerns Lazar's argument that the moral gravity of violating an individual's right not to be harmed varies according to the violator's evidence about whether the victim is liable to that harm. More precisely, Lazar defendsRisky Killing: When A kills B, and B is not liable to be killed, other things equal A's act is pro tanto more seriously fact-relative wrongful the higher the epistemic probability, when she acted, that B was not liable to be killed. (77)To support this idea, Lazar appeals to contrasting two-agent cases, in which innocent victims are harmed by rights violators in different epistemic situations. Though these cases provide some intuitive support for Risky Killing, the principle has some striking implications once we introduce third parties. This is because, by linking the fact-relative (i.e., perspective-independent) gravity of rights violations to the violator's evidence, Risky Killing seems to make everybody's moral situation highly sensitive to the violator's epistemic situation, regardless of their own epistemic perspective. Consider:Rescue: A is about to shoot B, because he believes that B will otherwise kill many innocent people. A's evidence is that B is 80 percent likely to be liable to be killed. C is about to shoot D because C believes that D will otherwise kill many innocent people. C's evidence is that D is 20 percent likely to be liable. B and D are in fact entirely innocent. E knows all these facts and can prevent B or D from being killed, but not both.According to Risky Killing, C's killing D would constitute a more serious rights violation than would A's killing B. But this implies that E has stronger moral reasons to save D rather than B. This doesn't seem at all obvious. B and D seem to have equally strong claims to E's assistance, and so E should give each an equal chance of being saved (by tossing a coin, for example). A and C's epistemic situation just doesn't seem especially relevant to what E ought to do. The same seems true in intrapersonal cases:Rescue 2: A and B each independently threaten C with a broken leg, because both believe that C will otherwise harm many innocent people. A's evidence is that C is 80 percent likely to be liable. B's evidence is that C is 20 percent likely to be liable. C is in fact entirely innocent. E knows all these facts and can either prevent A's attack or B's attack, but not both.Risky Killing suggests that E ought to prevent C's leg being broken by B rather than A. But this still seems hard to believe, even though there are no competing claims to rescue in this case. Third-party cases thus cast doubt on Risky Killing.In response, Lazar might hold that the evidence-sensitive dimension of rights transgressions is fully agent-relative and so only affects the moral situation of the rights violator, but not third parties.2 But this modification risks limiting the support that Risky Killing gives to civilian immunity (as well as raising an obvious worry of ad hocery). While it would still give soldiers reason to target combatants rather than noncombatants in war, it would not give third parties reason to bring it about that they do so. One implication is that Risky Killing less obviously supports using the law as a means of preventing the killing of civilians. More generally, Lazar's multipronged defense of Moral Distinction would have been strengthened if it had shown that the asymmetries identified robustly affect the normative situation of third parties.But none of these worries undermines the fact that Sparing Civilians is a superb book that deserves to be taken very seriously

Adversary · Harm · Just war theory · Moral responsibility · Morality · Political philosophy · Political science · Politics · Sociology · Spanish Civil War · Virtue · Global Peace and Security Dynamics · Law · Philosophy · Torture, Ethics, and Law · War, Ethics, and Justification