Against Indiscriminate Killing Even in Supreme Emergencies

For the past few weeks, people on- and offline have spoken up to question Winston Churchill’s legacy. They generally highlight his racism, his support for the use of concentration camps, his treatment of Ireland, his complicity in the Bengal famine, and more. Some protested in a Churchill-themed café. In response, others argue that he nevertheless deserves to be remembered for his role in fighting off the Nazis and inspiring the British public in dark times. There are, however, important questions to ask even about Churchill’s role in fighting the Nazis. Churchill authorised the indiscriminate killing of civilians by bombing German cities. In justifying this tactic, Churchill appealed to the extraordinarily dangerous nature of the situation. But does this justify indiscriminate killing? This question still has relevance today. US drone strikes in the Middle East and Afghanistan in many respects resemble a campaign of indiscriminate violence, and so it is necessary to ask if this campaign can be justified. I will here argue that the logic of Churchill’s defence does not, and indeed cannot, justify the use of indiscriminate violence.

The standard kind of argument deployed in order to defend actions such as the intentional bombing of civilian targets I call the Supreme Emergency Doctrine (SED), following Michael Walzer’s discussion in his seminal Just and Unjust Wars. When the threat to a community is very great and imminent, Walzer claims, it ‘brings us under the rule of necessity’ and ‘necessity knows no rules’. Thus, it becomes permissible do whatever necessary to stave off defeat. Walzer suggests a case could be made that prior to the entry of the US and the USSR into the war it was militarily necessary, that is to say, there were no other tactics available which were more likely to be effective, to attack Germany’s cities. In Churchill’s words, ‘the bombers alone provide the means of victory’. (Though, as Walzer is also aware, when Bomber Command was instructed to focus on attacking German civilians in February 1942, both the US and the USSR had entered the war.)

Although it does perhaps not quite compare to Hamburg or Dresden, the US drone war in Afghanistan and the Middle East bears more than a passing resemblance to indiscriminate bombing. Though drones, when used properly, are infinitely more precise than the equipment available to Churchill, drone operators are entirely dependent on intelligence, gathered often halfway across the world. This intelligence is often out-dated or even plain false, provided to the Americans by informants with an axe to grind. A recent New York Times investigation in Iraq found that the US routinely underestimates the civilian casualties caused by drone strikes. Elsewhere, the US has been known to count all adult men as (suspected) militants, thereby artificially reducing the number of civilians killed with each strike. Even if the drone war isn’t intended to be indiscriminate, it often is in its effect – and it is certainly perceived by many on the receiving end as such.

When we know that a particular tactic is likely to be indiscriminate in effect – and as the Trump administration is getting rid of Obama-era controls over the use of drones, the strikes are likely to become more, rather than less indiscriminate – we need to explain why we nevertheless stick with it. Even though many proponents of drone strikes don’t acknowledge their indiscriminate nature, the existential threat of ISIS or terrorism is often implicit in their justification. But, as I will show, the SED is false – and so we need to do better.

First of all, we need to know what counts as a Supreme Emergency. According to Walzer, a Supreme Emergency is a threat to a community that is both immediate and of an ‘unusual and horrifying kind’. But what does that mean? If it is simply a matter of counting the number of lives lost or ruined as a result of the supposed Supreme Emergency, how can we do this before the threat has actually materialised? (Walzer, incidentally, thinks only political communities can face Supreme Emergencies, which might lead us to wonder about different kinds of communities, such as minority ethnic communities, which are, historically speaking, considerably more at risk of destruction than states).

The second reason follows on from this. There isn’t only guesswork involved in deciding whether a particular community faces a Supreme Emergency, but also in deciding if all other options have been exhausted. How can we know whether or not there are alternatives to indiscriminate violence available? Before the Second World War even started, many in Bomber Command, and the political elite, considered terror bombing a legitimate military strategy. Several influential works had been written in the 1920s advocating it as an element of modern warfare. Given this, we need to question whether other options were ignored when the decision was made to attack the German cities. We need to know whether those deciding to use indiscriminate violence were appropriately positioned to make this judgement.

The final issue relates to indiscriminate violence’s effectiveness. Based on a review of (some of) the research, it seems to me that it is unclear whether killing civilians in war has the desired effect of causing demoralisation and decrease in support for the war, in the kinds of scenarios in which a Supreme Emergency justification would be available.

There is some evidence that suggests indiscriminate killing and terrorism do not work. Max Abrahms and Stathis Kalyvas, who focus on terrorist attacks and indiscriminate reprisals respectively, suggest that attacks on civilians are rarely effective. Others suggest that aerial bombing is more likely to be effective when it targets military, rather than civilian, targets. But sometimes indiscriminate violence does appear to work. Jason Lyall finds that during the Chechen insurgence against Russia, indiscriminate shelling of Chechen villages was effective in reducing insurgent attacks against Russian targets.However, the conflict in Chechnya was of a very different nature than for instance Kalyvas’ primary case-study, the Greek resistance against Nazi occupation, in terms of civilian support for the insurgency. Lyall notes that in Chechnya, there was a lack of support for the insurgency and the insurgency leadership was divided.

Whether attacking civilians helps achieve victory appears to depend to some extent on the nature of the adversary and the level of support for the fighters. If the adversary is credibly able to present itself as being able to protect the targeted civilian populations, indiscriminate anti-civilian attacks will do little to persuade the people that they should side with the ones who are killing their compatriots.

These findings seriously undermine the SED. In order for a party to pose an existential threat, they must be able to militarily overwhelm the threatened community. However, it seems that these kinds of situations are the ones in which indiscriminate violence is least likely to work. When a community is sufficiently powerful to pose an existential threat to another, it is much more likely to be able to survive anti-civilian attacks, for instance because it has a strong base of support and/or has control of media allowing it to spin the attacks for propaganda purposes. If, on the other hand, the adversary is weak and lacks support from its citizens, attacks on civilians may be effective. But if the adversary is weak, they are unlikely to be able to pose a threat sufficiently grave to qualify as a Supreme Emergency. Killing civilians may therefore be effective, in these situations, but soldiers cannot justify it by appealing to the SED.

The SED does not permit soldiers to do just do anything when the stakes are high: it permits them to them to use indiscriminate violence because doing so is thought to be the most effective, or the only, way in which victory can be still be achieved. However, the analyses of the effectiveness of civilian harm discussed here suggest that in the paradigmatic Supreme Emergency situations, killing civilians will not be effective, and therefore it cannot be necessary.