How should humanists feel about the ‘drone strikes’ on Taliban or al-Qa’ida leaders? Here are my thoughts.
Military assassination is, more than any other phenomenon, my greatest area of academic interest and expertise. If you check out my CV, you’ll see I have two forthcoming articles on it, and for someone of my very junior academic status, this fact speaks more about my focus on the topic than any other statement I might make. This blog post is my attempt to address the subject from a specifically humanist perspective. It’s a bit hastily put together, broad and therefore also shallow, but as its target audience are minimally informed members of the Humanist community, I don’t think this is a flaw. If anyone wants greater detail, particularly on the Israel-Palestine conflict stuff, please let me know.
Generally speaking, a military assassination is when a military organisation, which could be part of a state or part of some kind of rebel or insurgent group, carries out a premeditated killing of a specific person who is perceived to be an enemy combatant.  Many of the UAV ‘drone strikes’ carried out by the US government – usually by the CIA – in places such as Pakistan and the Yemen are examples of military assassination.
Assassination is a very contentious activity. The frequency with which non-combatant bystanders are killed alongside combatant targets is troubling even when it is low. The ethical and legal criteria for distinguishing combatant from non-combatant are difficult to develop in the abstract and more difficult still to apply in practice, and as I will explain later, this causes problems for military actions where people are targeted in their homes or during their routine lives, and not in places that look much like battlefields. There’s also a general unease, I think, for the idea of ‘named killings’, whereby someone is hunted down and killed regardless of whether or not they are on the battlefield and fighting at the time.
But I still think that assassination has the potential to be among the most ‘humanist’ – that is, the least traumatic and harmful-to-wellbeing – uses of military force. By ‘military force’, I mean acts of organised violence carried out in an environment of armed conflict according to the strategies and guidance of some kind of command structure. Now, regardless of your opinion of killing on an essential level – and I’ve given mine – the fact is, war is and will be a feature of the human condition. It might decrease in prevalence, but it won’t disappear. And it will happen that our community may be a party to a war. So if military assassination seems to offer a way to wage war while remaining as true as possible to our humanist convictions, we should be amenable to using it.
Before going any further, I do need to say a few things. First of all, assassination is unlikely to ever ‘win’ a war. With only rare exceptions, killing a single person, or even a small set of people, is unlikely to end a conflict according to favourable terms, and so regardless of its potential, it should not be viewed as a panacea or in strategic isolation. Second, any method of violence, any application of the military instrument, is open to misuse and abuse, and aligning actualisation with potential is often a very challenging task.
I’ve written already on the morality of war from a humanist perspective, so I won’t go into much detail here. But to review, a humanist morality of war should treat all life as valuable, combatant or noncombatant, and should seek pragmatic, heuristic principles in the use of force that minimise the overall human cost of war. As is already laid out in current international laws, the use of force in war should adhere to the principles of necessity, discrimination, and proportionality. That is, any use of force must be strategically necessary, discriminate between combatant and noncombatant to a reasonable degree (a good example of a heuristic), and be proportional to the military importance of the goals achieved. Violence that harms people not involved in the fighting, that harms more people than is necessary, or that is unnecessary entirely, is bad.
Where a humanist perspective will be a bit controversial is in denying both the analytical separation of justice in war (jus in bello) and the justice of a war (jus ad bellum), and in rejecting the Doctrine of Double Effect . To a humanist, no use of force is sufficiently discriminate or proportional if it is used in the prosecution of an unjust war. Even though it is practically useful, in limiting the trauma of war, to require that violence in war only adhere to those two conditions, this should not influence our overall moral judgements. And whether or not a method of violence specifically targets nomcombatants, if it will predictably kill a large number of them, it should probably be avoided, all things being equal. But again, prudence and pragmatism should be our guiding lights.
It isn’t hard to imagine how assassination might be ‘justified’ according to the above criteria. Particularly in conflicts where combatants only emerge from the noncombatant population to mount specific attacks, then return to their homes and their noncombatant lives, within their larger communities. In such cases it can seem almost impossible to isolate combatants from noncombatants. Theoretically, an assassination could be among the most discriminate possible uses of force, as it requires only one death and therefore can be accomplished with potentially very little force. And given the bloody nature of war and important role that military specialists or commanders play in making war, targeting such individuals is certainly very likely to be proportionate to the threat they pose both to human wellbeing in general and to a war effort in particular. But this is all still in the realm of the abstract.
A bit less abstract are the criteria for determining who is and is not a combatant. The Geneva Conventions, to which most states are signatories and to which even non-signatories such as insurgent groups can be held accountable, require that anyone not openly displaying insignia demonstrating affiliation with a party to combat be treated as a noncombatant, which grants protection from military force but at the same time renders one subject to criminal law. The Additional Protocols to the Geneva Conventions further develop the criteria distinguishing combatant and noncombatant in a clearer way for wars involving non-state parties (such as, say, al-Qa’ida or Hamas), though they do not have the same customary and legal status; the US and Israel are not signatories, for example. But these are still often hard to apply in practice. The Israeli government assassinated numerous leaders of Palestinian nationalist groups despite those leaders arguably not qualifying as combatants under international law. The Israeli government justified targeting these leaders on the basis of their role in developing and ordering military action, comparing them to high-ranking military officers. On the surface, this seems like it should be a persuasive argument, if true, but international law is at best vague about the status of such leaders when they’re not carrying weapons and either in the lead-up to or execution of a military operation in which they are taking part. Furthermore, they were often targeted in their homes or during routine non-military activities such as visiting a shop (or a mistress, as happened in one case). Another even more difficult case is that of al-Qa’ida leader Anwar al-Awlaki,who was killed in a bombing by a state in which he held citizenship – the United States – and by all accounts was unlikely to ever personally become a combatant according to the criteria specified by the Conventions.
There is also the problem of bystander casualties. Between October 2000 and July 2007 Israel carried out 134 assassination operations, resulting in 367 deaths, 218 of them of specified targets. The rest were a mix of combatant and noncombatant bystanders. Once one accepts that in this particular conflict, fighting must take place in populated areas, a two-thirds success rate for targeting should be seen as quite discriminate. There are many ways in which the IDF has tried to minimise bystander deaths, such as by preferring to kill targets while they are in vehicles, and by using missiles with unusually small warheads. Or by using snipers. I think that the ongoing threat to Israeli lives posed by Palestinian militants justifies targeting them, but it must be noted that the acuteness of this threat has diminished substantially since the days of the Second Intifada, and that in turn makes it harder to justify less discriminate operations. As for the case of the US ‘drone strikes’, it may be easiest to quote my academic self (or just read this blog post at Duck of Minerva which cites a few more up-to-date sources):
In compiling accurate data on the US drone strikes, one must remain humble in the face of several significant epistemic limitations. As the programme is covert, the details of its particular actions and policies are difficult to discover or verify [and] it is extremely difficult to accurately gauge the true number of fatalities the strikes produce, as casualty figures are largely based upon unverifiable information…. The Jamestown Foundation published in its journal Terrorism Monitor that the most rigorous interrogation of drone strike reports found that, with a total of 144 strikes by19 June 2010, the CIA drone campaign in Pakistan was ‘impressively accurate’: only 68 of the 1,372 confirmed deaths (4.95%) were clearly identifiable as civilian, and 1,098 (80%) were by comparison clearly identifiable as combatant. While as previously mentioned, these numbers should be taken with a great quantity of salt and be viewed with scepticism – though the Jamestown Foundation study appears to have been conducted with as much empirical rigour as one could demand – it is nevertheless possible to see that the US assassination programme is perceived by ‘Western’ publics and by much of the population in its theatres of action to cause significant if not indiscriminate damage to civilian bystanders.
Military assassination also might not work as well as its proponents suggest. One statistical analysis of Palestinian nationalist violence during the Second Intifada showed that assassinations neither increased nor decreased the rate of attacks . Another showed that assassinations neither reduced nor increased fatalities either , despite it being fair to assume that killings of expert bombmakers and veteran commanders should reduce the lethality of attacks, even if not their frequency. It must be noted, though, that isolating the effect of assassinations from the large range of Israeli counterterrorism activities via statistical analyses is difficult if not impossible, and there has been strong support for assassinations amongst Israeli security experts according to my research. Not so for the US drone strikes, though, which are criticised by some experts for alienating local populations due to their perceived indiscriminate nature. The truth of the matter may be that they are quite discriminate, but even a small number of bystander deaths might be enough to completely undermine their strategic value in most cases, and besides, often perceptions matter more than reality. Furthermore, the Terrorism Monitor stats don’t differentiate between assassinations and attacks of opportunity which did not target specific individuals, and there may be a significant difference between the two.
Notably, while Israel has clarified its policies and procedure for determining who may be targeted and when an assassination is legitimate via multiple legal entities including its Supreme Court, the US has not done so at all.
- Military assassinations aren’t essentially bad from a humanist perspective, and are potentially among the least harmful military actions
- Military assassinations should be subject to the same rules of war as other military actions, meaning that they should be used to target combatants and should adhere to rules governing proportionality and necessity. They should therefore be conducted with as much transparency and oversight as possible – something which, even taking into account the need for secrecy, the US has not been particularly good at doing.
- Military assassinations must be studied further and considered carefully on a case-per-case basis to determine whether and how they would be strategically beneficial, for violence that does not avert more harm than it causes should be considered a bad thing, and much of the harm that violence causes can be difficult to measure and lies in the feelings of insecurity or moral outrage it can produce in an audience
 This is loosely based off of the definition I gave for ‘assassination’ in my article ‘Crossing Off Names’, forthcoming in Small Wars and Insurgencies
 Very generally and minimally speaking, it states that if a bad effect is foreseeable but not intended, the act that causes it is not wrong. For example, the ‘terror bomber’ who bombs a factory in order to kill its workers and create fear is committing wrong, but the tactical bomber who bombs the factory to render it inoperable is not doing wrong, even if the deaths of workers are likely. One is in the clear if one can answer ‘yes’ to the question ‘if the bad effect didn’t happen, would my goal still be served?’
 Hafez, Mohammad and Joseph Hatfield, “Do Targeted Assassinations Work? A Multivariate Analysis of Israelʼs Controversial Tactic during Al-Aqsa Uprising,” Studies in Conflict & Terrorism 29, no. 43 (2006)
 Plaw, Avery. Targeting Terrorists – A Licence to Kill? Hampshire: Ashgate Publishing Limited, 2008.