Risk-Transfer Militarism and the Legitimacy of War After Iraq

A renaissance of warfare is one of the most striking features of the early twenty-first century. War, it seems, is not the prerogative of international criminals, but the first resort of the righteous. After September 11, 2001, it was widely believed that might could indeed enforce right: President George W. Bush was quick to proclaim his response to the terrorist massacre a “war” rather than a law-enforcement operation. Indeed the Global War on Terrorism (GWoT) quickly became an overarching framework for all politics and any military action, in the eyes of its supporters. And initially, at least, it had widespread support: as Polly Toynbee, one of Britain’s foremost liberal commentators, put it during the campaign in Afghanistan, “bombing works.”1 The confidence in this position, especially but not only in the United States, involves a striking reversal of the pacifistic sentiments that largely prevailed in Western democracies during much of the last century. It is a veritable relegitimation of war. And yet after the Iraq War of 2003, this confidence has been weakened and some of the support for the GWoT has slipped away: most obviously because of the thinness of its manifest rationale, Saddam Hussein’s “weapons of mass destruction”; but also, this paper will argue, because of the contradictions of the new Western way of war, which I characterize as one of risk-transfer.

I write of the relegitimation of war because warfare had been comprehensively – if obviously not finally – delegitimized during the course of the twentieth century. In 1914-18, the trenches of Flanders gave us the paradigm of “senseless slaughter” that helped frame a “structure of feeling” about war that remained influential throughout the century.2 So the new resort to war in 1939-45 in Western democracies was heavier-hearted, accompanied by less jingoism, and motivated as much by anti-fascism as by nationalism. True, this did seem to many like a good war, a perception that has been greatly accentuated in more recent times by the misrepresentation of the war almost as a crusade to halt the Holocaust. But this increasingly appeared very much as an exception. The threat of nuclear extermination created an overwhelming perception, during most of the second half of the twentieth century, that major war was to be prevented at almost all costs. Vietnam reinforced the antiwar structure of feeling by showing how even the limited kind of war that could be fought despite nuclear weapons would also involve senseless slaughter. The importance of this experience was that it affected the most powerful Western state, the only one (apart perhaps from Britain and France) in which the use of war was not already delegitimized by the horrors of 1939-45.

The twenty-first century relegitimation of war is not an entirely new phenomenon. One element of it derives from a similar source as the understanding of the Second World War as “good.” This is the role of war, or at least of organized military force, in halting genocide and other violence against civilians. This new “positive” was already emerging at the time of the last great peace movement in Europe, the campaign against nuclear weapons in the 1980s. At that time, however, the examples of good war came from Third World states like Vietnam (in Cambodia) and Tanzania (in Uganda). More recently, of course, this has come to be called “humanitarian intervention” and has become a declared aim of much Western-sponsored military action.3 This was always, however, only one strand of the new Western willingness to resort to war. Margaret Thatcher pioneered a different mode, 20 years ago in the Falklands, and with the end of the Cold War the United States also began to fight real wars with success again. The first President Bush “kicked the Vietnam syndrome” in the Gulf War of 1991 and NATO successfully concluded its first ever war over Kosovo in 1999.

So while the last century’s pacific lessons still seemed powerful, the ground was well prepared for President George W. Bush to pronounce his “war against terrorism,” and with the U.S. campaigns in Afghanistan and Iraq it is difficult to avoid the conclusion that the trend is now the other way. A series of almost unprecedented successes in the use of arms, by the most powerful forces on the planet, threatens to give warfare a strong new momentum. There is however a growing awareness of the contradictions of this development – no longer opposed mainly by those who confuse it with earlier manifestations of Western military power – although the troubling consequences that it could bring to world society in the coming decades are weakly understood.

In this paper, I will first consider the wars in Afghanistan and Iraq as the latest examples of the new Western way of war, and analyse their casualties alongside those of previous campaigns in the Gulf and Kosovo.4 I shall identify the new type as “risk-transfer war,” a central feature of which is a “militarism of small massacres.” I shall argue that this new type thus offers only a partial answer to the problems, for the legitimacy of warfare, caused by the systematic targeting of civilians in earlier “degenerate war.” Despite a closer approximation to “just war” criteria, the application of which the new mode I shall discuss, inequalities of risk between Western military personnel and civilians in the zone of war revive the question of legitimacy in a new form. The paper then suggests that in our concern for relatively small numbers of civilian casualties, we may be applying to war standards from which it has historically been exempt. In this context, I shall conclude by proposing that the contradictions of the new Western way of war reinforce a “historical pacifist” position toward the general legitimacy of warfare.

I. Military and civilian deaths in Afghanistan and Iraq compared with Kosovo and the Gulf War

The two main U.S.-led military campaigns of the Global War on Terrorism have been significantly different from each other, as well as from those in Kosovo and in the Gulf. The two GWoT wars, although justified (in the Iraqi case without any serious justification) by the manifest threat of al-Qaeda terrorism, have principally targeted the repressive governments of the Taliban and Saddam Hussein, and their most concrete results have been the overthrow of these regimes and their attempted replacement by U.S.-sponsored local administrations (although at the time of writing this process is still unstable in Afghanistan and has hardly begun in Iraq). The Afghan campaign was from the start very much an anti-guerrilla war, as al-Qaeda and the Taliban dissolved into the remoter regions of the country. The Iraq campaign, relatively a more conventional conflict with a more serious armed enemy, nevertheless had a guerrilla aspect, which has become more prominent in the aftermath as the remnants of the defeated regime have similarly gone underground. In both cases, while airpower played a very large role, there was considerable reliance on special forces and in Iraq on artillery and tank formations on the ground, as well as on local allies (in Afghanistan the Northern Alliance-United Front, and in Iraq the militia of the two Kurdish parties in the autonomous northern region). Both wars differed, in the greater significance of ground forces, especially from Kosovo, where NATO committed none until the Serbian regime conceded its occupation of the disputed province, and from the 1991 Gulf War, where a six-week bombing campaign preceded the final push on the ground, although in the former case of course the Kosovo Liberation Army provided a surrogate armed force.

However, in one key respect they exhibit fundamental similarities. These wars were all fought in ways that resulted in very small numbers of casualties to U.S. and “coalition” (i.e., allied Western) troops. Western troops fight, for the most part, in the relative security of heavy armor: in planes that fly mostly at high altitudes where they are almost invulnerable to enemy airpower (if it exists at all, this is usually destroyed in the first few days of war) and anti-aircraft fire; and in tanks and other armored vehicles that are also heavily protected. When more exposed on the ground, in light vehicles or on the street, they at least benefit from body armor, armed back-up, and training that enables them to minimize their vulnerability; they usually have the means to pre-empt or at least return fire on any actual or suspected attackers. In the limiting case of the Kosovo War, exclusive reliance on airpower actually resulted in a complete avoidance of any NATO deaths due to enemy action (although of course some soldiers died in accidents during the war, and a few were killed during the subsequent years of occupation). The GWoT wars, like the Gulf War, have seen U.S. and Western personnel exposed on the ground, and therefore casualties have occurred, but still in very small numbers compared to historic wars.

On the other hand, these wars have all involved imposing much more extensive casualties on the U.S.’s and the West’s armed enemies and also, albeit “unintentionally,” on civilian non-combatants. How extensive is often difficult to establish: numbers of U.S. and Western casualties can be very precisely established, from lists of the individual dead published by the U.S. Department of Defense, UK Ministry of Defence and other national governmental bodies (although these lists do not always distinguish deaths due to enemy action from those resulting from accident and illness). There is considerable difficulty in estimating all other casualties. In particular, there are virtually no institutional sources for figures of enemy military dead. However, we do know from many media reports that with the extraordinary firepower that is available to Western forces, the killing of enemy combatants is often extremely systematic and efficient, and that many battles have been extremely concentrated. However, information is generally sketchy and key episodes, like the systematic bombing of Taliban fighters in 2001-2 and of Iraqi troops in the early days of the 2003 war (as well as in the first six weeks of the 1991 war during which probably the largest number of soldiers were killed of any recent Western war), still lack comprehensive, detailed accounts. Figures that have been estimated by scholars like Conetta for the 2003 Iraq War5 , like those proposed by authors like Daponte6 for the 1991 Gulf War, suggest that vastly more enemy fighters have been killed than Western soldiers. Indeed, they also suggest that many more fighters have been killed by Western firepower than civilians, a result one would expect given the enhanced precision of the weaponry used compared to earlier periods of war, as well as the greater care professed for avoiding civilian casualties.

Nevertheless, we also know from many reports of incidents in which bombs, missiles and artillery fire have landed on civilians, often in their homes, as well as cases in which civilians have been shot on the street. Altogether the direct killing of civilians by Western forces has been a normal feature of recent campaigns. Although official Western sources produce no figures, enemy authorities sometimes do, and there is an increasingly prolific industry of “body-counting” by critical writers. The latter has become increasingly sophisticated in its accounting, although there is still wide divergence in data collection (reliance on published reports versus on-the-ground collection), methods of analysis (e.g. who counts as non-combatants), and hence in results. Nevertheless, what is indisputable is that even on the lowest estimates, such as those of Conetta for Afghanistan and Iraq7 , the numbers of civilians killed has been a troubling feature of all recent campaigns, and it is likely that more civilians were killed (primarily because of the attack on Baghdad, due to the aim of regime change) in the recent Iraq War than in 1991.8

Table 1 below summarizes estimates that have been produced by serious studies for death tolls among different groups in recent wars. Even the figures for Western troops killed by enemy action are estimates, rounded down from published totals of all Western dead. All other figures are open to much more question, and are given here to indicate orders of magnitude.

Table 1. Estimated death tolls in the Western wars of the global era
(rounded figures)

U.S. and Other Western Military Killed by Enemy

Local Allies’ Military Deaths

Enemy Military Deaths

Civilians Killed by Enemy in Precipitating Events

Civilians Killed by West

Indirectly Caused Civilian Deaths as Result of Western Action

Gulf War 1991


hundreds? (Kuwaiti Resistance)

20,000 – 56,000

low thousands? (in Kuwait )


60,000 – 111,000 (health effects of infrastructure damage)

Kosovo War 1999


hundreds? (KLA)


12,000 (Albanians)



Afghanistan 2001-03


few? (Northern Alliance/UF)

thousands/tens of thousands?

3,000 (NY/DC)


3,200 (to end Jan. 2002)

Iraq War 2003


few? (Kurdish militias)


no such events (“preemptive”)

3,750 – 9,900


Notes on sources

Figures for all deaths from U.S. websites (Dept. of Defense, etc.) rounded down to exclude accidents.

Figures not generally available; guesstimates.

Figures for Gulf War from Daponte, “A Case Study” and Conetta, Wages of War

Figures for Kosovo from International Independent Commission on Kosovo, The Kosovo Report, (Oxford: Oxford University Press 2000).

Figures for Afghanistan from Conetta, Operation Enduring Freedom, and Benini and Moulton, “The Distribution of Civilian Victims” (see n.11)

Figures for Iraq War from Conetta, Wages of War, and iraqbodycount.net [Jan. 26, 2004]

General notes

“Civilians killed by enemy in precipitating events” refers to deaths during the acts of aggression that initiated the immediate conflict. It does not include the longer-term death tolls of earlier campaigns (e.g. Saddam Hussein’s genocides, Serbian campaigns in Croatia and Bosnia , the Afghan civil war, and Taliban repression) or subsequent war (e.g. Iraqi civil wars after the Gulf War, the elimination of the Marsh Arabs, etc.).

“Indirectly caused civilian deaths as result of Western action” refers to estimates of civilian deaths in the immediate aftermath of the war (e.g. civilian deaths as result of the U.S. destruction of infrastructure in Iraq ; and as a close consequence of recent bombing of Afghanistan ). It does not include longer-term deaths in which the war may have been influential (e.g. as a result of later sanctions against Iraq , or possible long-term consequences of the Afghan war).

What is certain is that there is a broad pattern in which Western military dead are fewer in number, by at least an order of magnitude, than both enemy and civilian dead. (This pattern also seems to exist in the distribution of other casualties, e.g. injuries, when studies have addressed this.) 9 There are divergences in the general levels of killing between wars: the Gulf War was generally the most and Kosovo the least lethal of recent major Western campaigns, and the numbers of civilians dying from indirect (mainly health) effects of bombing was probably uniquely high in the Gulf. Nevertheless, the pattern remains across the wars, and it raises important questions about the ways in which the West fights its wars today. Kosovo was not so exceptional: in the first three months of the Afghan campaign, even Conetta’s conservative estimates put the number of direct civilian casualties in the range 1,000-1,300 10 (while Benini and Moulton using ground-based studies suggest a far higher toll 11 ). Only one U.S. (and no other Western) official, a CIA operative murdered by Taliban fighters in the Mazar-e-Sharif prison rebellion, died in the same period. There are phases during which the ratio is not quite so stark: but a gross disproportion between Western military and local civilian casualties remains constant.

When indirect casualties are added in, this disproportion is amplified. In Afghanistan, Conetta concluded (in a companion study), “A minimum of 3,000 civilian deaths attributable to the impact of the bombing campaign and war on the nation’s refugee and famine crises.” His report uses “an estimate of 8,000-18,000 Afghani deaths occurring during the mid-September to mid-January period and due to starvation, exposure, associated illnesses, or injury sustained while in flight from war zones. Of this total, at least 40 percent of the deaths (3,200+) are attributed to the effects of the crisis and war.”12 It will be noted that because there were multiple causes of these categories of civilian deaths, Conetta was reduced to ascribing a percentage of them to the U.S. campaign. Clearly this procedure is fraught with methodological difficulties, but something like this may be necessary if we wish to put a figure to the deaths caused by the U.S. bombing. To these, Conetta argued, must be further added “800+ troop deaths due to post-war reprisals and mismanagement of prisoners” (the inclusion of this figure is justified because once captured Taliban/al-Qaeda fighters are no longer combatants).13 Therefore, using what he regards as conservative estimates, his report leads to the conclusion that the total of non-combatant deaths as a result of the American military campaign, up to January 2001, was 5,000 or more. Reports from Iraq have suggested less drastic health effects than in Afghanistan (and certainly fewer than in Iraq in 1991), but a wide range of indirect effects on life, from lawlessness and insecurity as well as the destruction of water and electricity supplies, and, of course, from unexploded ordinance.

Recent Western wars have been seen by their supporters as “targeted” violence, by opponents as “indiscriminate” slaughter.14 The discussion so far suggests that both of these claims are too simple. Certainly, the bombing has been quite successfully targeted. The likelihood is that as in the Gulf and (less markedly) Kosovo, the numbers of enemy combatants directly killed in Iraq and Afghanistan is greater than the number of civilian deaths similarly caused. The absolute number of civilians directly killed in these wars, as in the previous two, is very small by comparison with historic U.S. campaigns (e.g. Vietnam and Korea, as well as the world wars). To this extent, the charge of “indiscriminate” killing of civilians appears inappropriate. However, all killing of those who are not directly targeted clearly shows definite limits to discrimination. This kind of killing, notoriously called “collateral damage,” cannot be avoided entirely in any long-distance use of powerful weapons and is inherently disturbing however much it has been scaled down from the historic pattern. Although civilian casualties are routinely described as accidents, this outcome is hardly accidental. It is the product of political choices in the refinement of Western military power, at three main levels: strategy, weaponry and media management. The combination of these elements enables the West to fight wars at relatively little human cost to itself. And since the risk to human lives, pictured on television, has been since Vietnam the major political risk of war, this also means that the West is able to fight wars with a great reduction in the political costs.

II. Risk-transfer militarism: the new Western way of war

It is the argument of this paper that these wars are examples of a new Western way of war. This “new” war is not, of course, entirely new. It is a reinvention of the reliance on airpower that has been central to Anglo-American military thought and practice since the 1920s. The new mode relies on bombing – by both manned bombers and cruise missiles – as well as long-range artillery, even more than before. However, it uses the enhanced precision that computer electronics brings to targeting (the so-called “Revolution in Military Affairs”) to avoid the large-scale and widespread massacres of enemy civilians that occurred in the Second World War and Vietnam. It uses local allies to carry out as much as possible of the actual fighting on the ground. And it uses media management to mitigate the effects of the “small” massacres that “accidentally” happen, as well as of the (sometimes larger) numbers of indirect casualties.

In this way, the new way of war appears to transcend the fundamental degeneracy of earlier bombing, manifested in indiscriminate targeting of civilians and huge numbers of non-combatant deaths. But it introduces new contradictions, through the multiple transfers of risk, particularly to civilian populations, which result in the distribution of death that I discussed above. The comprehensive transfer of risks away from Western military personnel appears to be a major aim of the new way of war. And the significance of this aim shows how closely the new way of fighting is linked to the new way of managing news media and public opinion, and designed to maintain the legitimacy of war in Western societies. And so we can call this way of war risk-transfer militarism, not just risk-transfer war.

To summarize, we can identify 5 major elements in this new militarism:

1. Killing the enemy

The principal risks of being killed (as a direct consequence of military action) are actually applied to enemy armed forces, rather than to civilians. In a historic sense, this is a transfer of risk back from enemy civilians toward the enemy military as such, and appears to reverse (at least for Western campaigns) the long twentieth-century trend toward overwhelmingly civilian casualties. This is clearly of great significance for arguments about the legitimacy of war.

2. Local allies take the risks on the ground

To the extent that Western troops take direct combat risks on the ground, Special Forces increasingly monopolize this function. However, the risks of ground combat are transferred wherever possible to local allies in the zone of conflict. The increasing interdependence between high-tech Western armor (air, artillery and tank formations) and relatively low-tech local armies on the ground (the Croatian and Bosnian armies, the Kosovo Liberation Army, the Northern Alliance-United Front in Afghanistan, the Iraqi Kurdish militia) enabled the West to transfer a greater share of battle casualties to them. However, this transfer may bring military and political costs: an over-reliance on local allies was criticized for weakening the campaign to capture Osama bin Laden in Afghanistan.15

In Iraq, local support for the 2003 war was inhibited by memories of the U.S.’s abandonment of the Shi’ite and Kurdish insurgents in 1991, a crucial error of that war: Shi’ites declined to produce the scripted insurrection in Basra. In the subsequent occupation, moreover, U.S. and other Western troops on policing duties found themselves on the receiving end of suicide attacks: the administration’s response was to work for a fast transfer of responsibilities to newly trained Iraqi soldiers and police, so that the risks would increasingly fall on them rather than on Americans.

3. Small “accidental” civilian massacres

The risk of repeated small massacres of civilians is an understood feature of the way the West fights its wars. Small massacres are “accidental” in the sense that they are not specifically intended and that efforts are made to avoid them. But they are simultaneously programmed into the risk analysis of war. Each of the West’s wars has been marked by numerous massacres, most commonly of a handful of people, but in numerous cases of 50-100 civilians at a time, with the largest single incident being the Amirya shelter bombing in Baghdad in 1991, in which around 400 died. The risk of massacres is not only known and understood by Western military planners, it is a completely predictable consequence of the protection provided to Western military personnel. Reliance on high-altitude and long-range bombardment keeps aircrew and soldiers safe; but it inevitably leads to errors of targeting in which hundreds or thousands of civilians die in each campaign. So the transfer to civilians of the risks of being directly killed is deliberate and systematic.

4. Media management

Because direct civilian killing on a large scale could threaten the mediated legitimacy of the war, media management is a key element in risk-transfer militarism. Mediation and surveillance have become intrinsic to this refined mode of post-total war, but they can make it particularly problematic. Western governments want no more TV pictures of direct victims than absolutely unavoidable: the largest massacres, like the Amirya bombing in the Gulf War and the bombings of a train in Serbia and of a refugee convoy in Kosovo, threaten legitimacy and are subject to intensive “spin” to reduce their effects. Likewise, Western governments want no threateningly large direct casualty numbers. Thus Robin Cook, Leader of the British House of Commons, indicated at the beginning of the U.S. war in Afghanistan that it would be acceptable because the numbers of its civilian casualties would be fewer than the death toll from September 11.16 That the eventual number of direct civilian deaths has been estimated by some at under half that total could be seen as supportive by Western politicians, even if the total number of deaths was greater. Although the words of General Tommy Franks, U.S. commander in Iraq in 2003, “we don’t do body counts,” which have achieved iconic status, have been taken a little out of context17 , a significant point for Western governments and militaries is indeed to avoid any large casualty figures from gaining currency. Conversely, opponents have made “body counting” into a virtual industry, with the aim of making the totals count against Western policies.18 However, despite civilian death tolls of many thousands in Iraq, they have generally failed to make these figures stick as significant points of criticism.

5. Indirect civilian casualties

A corollary of this is that indirect, less visible and less quantifiable casualties are more acceptable than direct, visible and quantifiable ones, and less decisive efforts may be made to minimize the former. Where other causes of death – enemy policies, civil war, drought, etc. – are possible, responsibility is less easy to pin down and therefore the West finds the risks more acceptable. The fact that experts can ascribe a figure to indirect bombing deaths only by making an arbitrary assumption about the proportion of them caused by U.S. action, indicates the lesser political danger that indirect deaths cause for Western leaders. Of course, ever since the Kurdish refugee crisis after the Gulf War, Western leaders understand that even indirect casualties can rebound badly, where responsibility can be established.19 Hence Tony Blair tried to insert a “humanitarian” dimension into the “war against terrorism” from the start. However, Western strategists’ awareness of the relationships between human risk and political risk may lead as much to efforts to deflect political responsibility as to real efforts to minimize indirect harm. This was the balance of effort in Western responses to the long-running immiseration of the Iraqi people, through the combination of UN sanctions and Iraqi regime policies.

III. “Just war” theory and risk-transfer militarism

How do we evaluate the legitimacy of this new way of war? The principal moral tools that we have available derive from the just war tradition. According to this tradition, as is well known, both ends and means have to be just. However, most of the debates on recent wars have focussed on the justice of the ends: whether the Kosovo campaign really had “humanitarian” objectives, and whether it achieved them; whether the Afghan campaign was a proportionate and effective response to al-Qaeda’s terrorism; and whether Iraq really possessed “weapons of mass destruction” and military action was necessary to forestall any dangers that they posed. However, these issues concerning the justice of the ends of recent wars, or of the “war on terrorism” in general, are not really the subject of this paper, since by concerning myself with the “way of war,” I have focused on the means that are more or less common to a number of different campaigns. For just war theory, this raises the question of jus in bello rather than ad bellum.

Let us evaluate risk-transfer militarism in just war terms. As Michael Walzer points out, it is axiomatic that the destruction of enemy is justified: “‘Soldiers are made to be killed,’ as Napoleon once said; that is why war is hell.”20 There can be little argument it would seem, with the apparently sharper focus, in the new way of war, on killing the enemy. Of course, if killing can be shown to be superfluous to the goal of destroying the enemy’s power, then its legitimacy can be put in question. Violence inflicted must be proportional, not only to the goals of one’s own side but to the severity of the initial aggression. On first inspection, maybe the intensive bombing of Taliban, al-Qaeda and Iraqi fighters fits this bill, and the awesome weaponry used – such as the “daisy-cutter” bombs – was merely an efficient means to this end. But if “bombing worked” to defeat these enemies, it did so surely by slaughtering them. There are legitimate concerns about these victims. Since the slaughter in the trenches, we have learned to attach more significance to the lives of soldiers. When one side can minimize the risk to its own soldiers to very low levels, is it moral to practice industrial killing on a hapless enemy? The image of Iraqi conscripts bulldozed (literally) into the sand, at the end of the Gulf War, is emblematic of this concern. Certainly, as we are moved to contemplate these inequalities of means, we should recall the slaughter inflicted on helpless office workers by terrorists using civilian airliners. But if the U.S.’s own killing is almost as one-sided, does the fact that Taliban or Iraqi soldiers were carrying guns make it so much more tolerable?

If risk-transfer warfare raises questions for just war thinking even around the treatment of enemy soldiers, the issues concerning civilians go to the heart of the tradition. As Walzer continues, “even if we take our standpoint in hell, we can still say that no one else [i.e., other than soldiers] is made to be killed. This distinction is the basis of the rules of war.”21 True, Walzer is prepared to countenance the extension of combatant status to civilian munitions workers in their workplaces, while they are actually making weapons, and he is also prepared to say that this “plausible line … may be too finely drawn.”22 Walzer also notes that the doctrine of “double effect” provided a way in which “it is permitted to perform an act likely to have evil consequences,” such as the killing of innocent civilians. The key condition is that “The good act is sufficiently good to compensate for the evil effect; it must be justified under [the] proportionality rule.”23 This was little justification for the atom bomb on Hiroshima, although U.S. apologists had used, and still use a similar argument. However, it appears plausible as an account of the “accidental” killing of civilians in Afghanistan, Kosovo and Iraq, if only because the numbers of direct victims are much smaller and so the evil might conceivably be outweighed. And Walzer perfected a rationale that has been widely applied in these instances. “Double effect is defensible,” he argues, “only when the two outcomes are the product of a double intention: first that the ‘good’ be achieved; second, that the foreseeable evil be reduced as far as possible.”24 The latter is exactly what the West now routinely claims to be doing in all its campaigns, with more plausibility than in the days before “smarter” bombing.

However, Walzer also provided the fly in the ointment when he pointed out that “Simply not to intend the death of civilians is too easy. … What we look for in such cases is some sign of a positive commitment to save civilian lives. Civilians have a right to something more. And if saving civilian lives means risking soldiers’ lives, that risk must be accepted.”25 In risk-transfer war, this is precisely what is avoided at all costs. Bombing and long-range artillery bombardments are undertaken in the firm knowledge that it will increase the risk to civilians compared to other possible means, military as well as non-military. High-altitude and long-range destruction is inherently indiscriminate.26 Amazingly, Walzer provided a way out for Western strategists in this situation. He immediately qualified his statement by arguing, “But there is a limit to the risks that we require. These are, after all, unintended deaths and legitimate military operations, and the absolute rule against attacking civilians does not apply. War necessarily places civilians in danger; that is another aspect of its hellishness. We can only ask soldiers to minimize the dangers they impose.” Exactly far they must go in doing this, he argues, “is hard to say”; “It is best … to say simply that that civilians have a right that ‘due care’ is taken.”27

This sort of escape clause could have made rough and ready sense in the context of wars where large numbers of Western soldiers were seriously risking their own lives. It is difficult to see how it can be sustained in the context of wars like Afghanistan where, after risk assessments have been carried out, only one American soldier was killed by the enemy in the same period that over one thousand Afghan civilians were (predictably) killed by American bombs. The care taken for civilians is not only less than the care taken for American soldiers, it is undermined by a policy adopted to keep the latter safe. Risk to civilians is reduced not as far as practically possible, but as far as judged necessary to avoid adverse global media coverage. Civilians’ risks are proportional not to the risks to soldiers, as Walzer envisaged, but to the political risks of adverse media coverage. Thus even if there is a limit to the risks we can require of soldiers, it can surely be shown that Western forces in Afghanistan, Kosovo and the Gulf have gone nowhere near to this limit. We could conclude that even if the aims of the “war on terrorism” are just, its methods are insufficiently so.

It could be argued that my argument here vindicates Walzer’s extension of the just war tradition, since I have used his case to provide a basis for criticizing U.S. policy. This is the line taken by Falk, who claims that “The ‘just war’ doctrine provides the most flexible and relevant normative framework. It has roots in the ethics of all the great world religions, it is a vital source of modern international law governing the use of force and it focuses attention on the causes, means and ends of war.”28 However it seems to me that the opposite could just as plausibly be claimed. Wars in which mortal risk to one’s own combatants is reduced to very low levels, but such risk is routinely inflicted on substantial numbers of innocent civilians, require such flexibility that they extend the “double effect” and “proportionality” ideas to the point of absurdity. This may no longer be the most appropriate framework within which to evaluate the legitimacy of war.

The degeneracy of war and the case for historical pacificism

I do not doubt that a consistent exponent of the just war tradition, like Falk, could try to rehabilitate it in the face of this reductio ad absurdam. But this would miss the point. The West is using armed force in a way that kills, directly, more enemy fighters than civilians; it generally doesn’t target civilians except in error; it aims to minimize “collateral damage” and “accidental” massacres. Although civilians are still killed, in historical, especially mid-twentieth century, terms the numbers of victims are small. The new Western way of war thus meets, prima facie, many of the historic demands for just war, even if we may question some excesses. However, if my argument has been accepted, there is still something fundamentally awry. That disparity between over 1000 innocent Afghans killed, to one American (in the first months of the Afghan campaign), says it all. If we are not satisfied with the justifications offered, this suggests that we may be applying different standards. In the rest of this paper, I want to explore what these may be, where they have come from, and where they may be leading. One obvious source of alternative standards is the political ethics of human rights. War has long been protected in Western thought from the norms that apply elsewhere in social life; but we may now be applying to war the standards from which it has previously been exempt. “Thou shalt not kill” has been tightened as a general norm, with fewer and fewer exceptions allowed; many Western states even decline to impose the death penalty. And yet war has remained a huge exception. Could it be that now that exception is being challenged, that tight norms against killing are being extended even into the realm of legitimate organized killing itself?

There is certainly some evidence to suggest this. The enhanced concern to protect Western soldiers’ lives is in itself a historic change: it reflects the outcry over the deaths of GIs in Vietnam, and a rejection of the idea of “cannon fodder” in favor of the notion of soldiers’ rights. As military sociology has shown, (professional) soldiering is now viewed more as an “occupation;”29 one, certainly, with different risks, but still one in which all efforts must be made to reduce risk. This is far from the notion of the heroic warrior. And when things go wrong, officers and governments must be made to take responsibility. During the Gulf War, nine British servicemen were killed when their vehicle was mistakenly attacked by a U.S. plane: the largest loss of British lives in a single incident during this conflict.30 The soldiers’ families took the British Ministry of Defense to court, and tried to get the U.S. airmen brought as witnesses, in an effort to hold the state accountable for the avoidable accident of their sons’ deaths. They did not fully succeed in their aims, but they did make the matter a cause of public debate. It is difficult to imagine any comparable concern over the lives of soldiers in similar incidents in earlier wars, simply because large-scale loss of life was so commonplace. But when numbers of deaths are reduced to small numbers, partly in consequence of fear about political effects of casualties, then individual lives can be made to matter.

In an age of human rights, this concern for individuals is in principle extended to individual civilians, and may even begin to apply to enemy soldiers, too. After all, concern about illegitimate killing has been magnified in the justification of recent wars: where they are not actually proclaimed to halt human rights abuses, as in Kosovo, they may well be designed to punish the perpetrators of killing against innocent civilians, as after 9/11. Alongside these wars there are legal proceedings: the activities of the International Criminal Tribunal for Former Yugoslavia (ICTY) were ratcheted up during the Kosovo War, and some at least of the al-Qaeda captives will face criminal cases (and there is evident inconsistency and embarrassment in the U.S. attempts to keep the Guantanamo Bay prisoners away from the courts). Not surprisingly, the ICTY felt obliged to consider the case against NATO itself for its “accidental” massacres of civilians in Serbia and Kosovo. The report of the Committee it established made a case that there is no prima facie basis for formally investigating NATO’s conduct of that war.31 Whether, in terms of the current law of war, that was a correct conclusion is not my concern here. What is clear is that NATO could be held accountable, in principle at least, for the deaths of the civilian victims. What was driving the demand for justice was not so much the legal norms as the perception that all the individual lives mattered: the three people killed in the Chinese embassy, the 16 killed in the Serbian TV station, the 70 killed when a railway bridge was bombed, and so on. Incidents as small as traffic accidents, in terms of numbers of victims, could be matters for which the world’s most powerful state could be brought to account, and in basically the same way.

No wonder that the U.S. is concerned about the establishment of the International Criminal Court. The laws of war were never intended to be applied in criminal courts in this manner. The drive to do this stems from general trends toward intensified legal regulation, heightened awareness of individual rights and extensive litigation – trends that derive much of their momentum (ironically) from U.S. society. But criminalization involves treating war like any other human activity, no longer de facto or even de jure excluded from norms that apply in all other fields. Taking “Thou shalt not kill” seriously, in the context of war, does indeed threaten to make the practice of war very difficult. If the means of war are generally picked over with a fine toothcomb, in the courts, in the press and (indeed) in academia, then the legitimacy of war will be regularly undermined.

It is difficult, therefore, to resist the conclusion that the door has been fundamentally opened to new kinds of delegitimation of war. Regardless of Western governments’ success in mobilising media and public opinion in particular cases, like the opening phase of the “war on terrorism,” the new Western way of war is generally vulnerable to new criticisms that will, sooner or later, challenge its even its newly refined justifications. The failure of any of the transfers of risk could expose the West to risk rebound. If airpower is insufficient to break the enemy, if the local forces are incapable of carrying out ground operations – or if they commit too many atrocities – or if the fickle media turn away from their would-be masters, the risks of the new mode of war will return to the West.

In the 1960s, 1970s and 1980s, it appeared that war was fundamentally compromised by the tendency of “limited” wars like Vietnam, as well as of large-scale “conventional” and nuclear war, to produce large numbers of civilian casualties. Since the 1990s, Western governments and militaries have developed a new mode that seems to have overcome this problem. What I have tried to show in this paper is that alongside this new mode have come new bases for criticizing war. The legacy of degenerate war can still be identified in disparities of risk: war’s legitimacy can be challenged through even the smallest episodes of killing. Historical pacifism, though challenged by the reinvention of war, is also being renewed, as the tests for justly killing get ever tighter.

End Notes

  1. Polly Toynbee, The Guardian, October 31, 2001.
  2. The concept, “structure of feeling,” originated with Raymond Williams, The Long Revolution, (Harmondsworth: Penguin, 1961).
  3. See Nicholas J. Wheeler, Saving Strangers, (Oxford: Oxford University Press, 2000).
  4. In this paper I shall refer to a “Western way of war,” “Western military personnel,” etc. The West here refers to the U.S. and Canada, Western Europe, Japan, Australia, etc., all of which are linked through a network of military alliances as well as political and economic international institutions, bilateral relations, etc. (See my Theory of the Global State, Cambridge: Cambridge University Press, 2000, for further discussion.) Given the military preponderance of the United States within the West, the “way of war” and the soldiers are both largely American, although the U.S. hardly ever acts alone, but always as representative of a larger Western interest, and usually with a wider coalition of which the West is the core. Thus in the Gulf War, the U.S. led a coalition under United Nations auspices, to which Britain and France were major military contributors and Germany and Japan financial contributors; in Kosovo, the U.S. led a NATO campaign; in Afghanistan there were a wide range of contributors, including German and Japanese military personnel. France and Germany famously refused to support or participate in the 2003 Iraq War, but most other Western states took part in the subsequent U.S. occupation, even if the UK was alone in providing combat troops. “Independent” uses of British and French troops, in Sierra Leone and the Ivory Coast for example, also correspond to the same “way of war.”
  5. Conetta, C. (03) The Wages of War: Iraqi Combatant and Noncombatant Fatalities in the 2003 Conflict. Project on Defense Alternatives Research Monograph #8.
  6. Daponte, B. (1993) “A Case Study in Estimating Casualties from War and Its Aftermath: The 1991 Persian Gulf War,” Medicine and Global Survival 3, 2
  7. Carl Conetta, Operation Enduring Freedom: Why a Higher Rate of Civilian Bombing Casualties, Project on Defense Alternatives Briefing Report #11, January 18, 2002; and The Wages of War: Iraqi Combatant and Noncombatant Fatalities in the 2003 Conflict. Project on Defense Alternatives Research Monograph #8, 2003.
  8. Conetta, The Wages of War.
  9. Compare Iraq Body Count, “Adding indifference to injury,” 2003, www.iraqbodycount.net, with figures for coalition wounded given by Iraq Coalition Casualty Count.
  10. Conetta, Operation Enduring Freedom: Why a Higher Rate of Civilian Bombing Casualties, Project on Defense Alternatives Briefing Report #11, 18 January 2002.
  11. A. A. Benini and L.H. Moulton, “The Distribution of Civilian Victims in An Asymmetrical Conflict: Operation Enduring Freedom, Afghanistan,” Journal of Peace Research, 2004 (forthcoming).
  12. Carl Conetta, Strange Victory: A critical appraisal of Operation Enduring Freedom and the Afghanistan war, January 30, 2002.
  13. Conetta, Strange Victory.
  14. Leading article, Eclipse, University of Sussex antiwar journal, November 2001, p. 1.
  15. The Washington Post quoted U.S. intelligence officials “as saying there was strong evidence that the Saudi fugitive had been in the eastern Afghan highlands of Tora Bora when U.S. forces and their Afghan allies launched an assault on the al-Qaeda mountain hideout. The failure to commit large numbers of U.S. ground troops and the reliance instead on Afghan militias with ambiguous loyalties was the “gravest error in the war”, civilian and military officials said.” Julian Borger and Richard Norton-Taylor, “U.S. blunder “let Bin Laden escape”,” Guardian, April 18 2002.
  16. Newsnight, BBC2 Television, October 2001.
  17. Unfortunately, none of the many critical websites that use this “quotation” – www.iraqbodycount.net emblazons it over its home page – source it. Although assumed to refer to the Iraq War, in which Franks also had a command role, the quote originates in Afghanistan in 2002. “I don’t believe you have heard me or anyone else in our leadership talk about the presence of 1,000 bodies out there, or in fact how many have been recovered,” Gen. Tommy Franks, commander of the Afghanistan operation, said Monday at Bagram Air Base. “You know we don’t do body counts.” The San Francisco Chronicle, March 23, 2002 [http://www.globalsecurity.org/org/news/2002/020323-attack01.htm, January 26, 2004].
  18. See for example www.iraqbodycount.net.
  19. Martin Shaw, Civil Society and Media in Global Crises, (London: Pinter, 1996), Part III.
  20. Michael Walzer, Just and Unjust Wars, Second Edition, (New York: Basic Books, 1992), p. 136.
  21. Walzer, Just and Unjust Wars, p. 136.
  22. Walzer, Just and Unjust Wars, p. 146.
  23. Walzer, Just and Unjust Wars, p. 153.
  24. Walzer, Just and Unjust Wars, p. 153
  25. Walzer, Just and Unjust Wars, pp. 153-54.
  26. Some forms, at least, of action on the ground, especially on the lines of armed policing, offer the opportunity to discriminate more and avoid civilian casualties to a greater extent, although here too armed soldiers have an inherent advantage over civilians, which may lead to self-protective “shoot first, ask questions later” killing of civilians suspected of being insurgents.
  27. Walzer, Just and Unjust Wars, p. 154.
  28. Richard Falk, “In Defense of ‘Just War’ Thinking.” The Nation December 6, 2001.
  29. Charles Moskos and F. Wood, editors, The military: more than just a job?, Oxford: Pergamon-Brassey,1988
  30. The attack on February 26, 1991 was the worst “friendly fire” incident in the war. However the effective protection of troops from enemy fire means that an increasing proportion of all Western casualties are from “friendly fire”: “35 of the 148 American servicemen and women who perished on the battlefield in the Persian Gulf War were killed inadvertently by their comrades, an extraordinary proportion by historical standards.” Barton Gellman, Washington Post, August 14, 1991.
  31. International Criminal Tribunal for former Yugoslavia, Final Report to the Prosecutor by the Committee Established to Review the NATO Bombing Campaign Against the Federal Republic of Yugoslavia.

This paper is based on an earlier article, “Risk-Transfer militarism, small massacres and the historic legitimacy of war,” International Relations, Volume 16 (3), 343-60, 2002. Extensively rewritten in the light of Iraq, the present paper has appeared online on http://www.theglobalsite.ac.uk and is to be published in Paul Eden and Thérèse O’Donnell, eds., 11 September 2001: A Turning Point in International and Domestic Law? (Ardsley, New York: Transnational Publishers, 2004). (Reprinted with author’s permission, all rights reserved.)