Portal de Periódicos da CAPES

Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles

2010; Taylor & Francis; Volume: 9; Issue: 4 Linguagem: Inglês

10.1080/15027570.2010.536403

1502-7589

Bradley Jay Strawser,

Free Will and Agency

Abstract A variety of ethical objections have been raised against the military employment of uninhabited aerial vehicles (UAVs, drones). Some of these objections are technological concerns over UAVs abilities’ to function on par with their inhabited counterparts. This paper sets such concerns aside and instead focuses on supposed objections to the use of UAVs in principle. I examine several such objections currently on offer and show them all to be wanting. Indeed, I argue that we have a duty to protect an agent engaged in a justified act from harm to the greatest extent possible, so long as that protection does not interfere with the agent's ability to act justly. UAVs afford precisely such protection. Therefore, we are obligated to employ UAV weapon systems if it can be shown that their use does not significantly reduce a warfighter's operational capability. Of course, if a given military action is unjustified to begin with, then carrying out that act via UAVs is wrong, just as it would be with any weapon. But the point of this paper is to show that there is nothing wrong in principle with using a UAV and that, other things being equal, using such technology is, in fact, obligatory. Keywords: RiskUAVsdronesprotectionjust war theory Acknowledgements The author would like to thank Jeff McMahan, Martin Cook, Uwe Steinhoff, Avery Plaw, Paul Bloomfield, Shawn Kaplan, Andrew Ely, Alexis Elder, Donald Joy, Matt Hallgarth, John Sherman, Casey Johnson, Abbilynn Strawser, and Bill Rhodes for helpful discussions, comments, and editorial advice on this paper throughout its various stages. Special thanks are owed to Stephen Kershnar, in particular, for his thorough criticism and help. Thanks as well to an anonymous reviewer for the Journal of Military Ethics for his or her very helpful comments. Thanks are also due to the participants at the 2010 International Society of Military Ethics Conference and the participants of the 7th Global Conference on War and Peace where some early sections of this paper were presented. Finally, the author is indebted to the members of the UConn Ethics Summer Reading Group for a thorough discussion of the paper. Notes 1. And there are other locations where it is presumed (although not verified) that lethal UAV employment has taken place, such as in Gaza by the Israel Defense Forces. In this paper I will refer to these remotely controlled weapon systems primarily as UAVs (Uninhabited Aerial Vehicles) and occasionally as drones. UAVs that are used for lethal combat purposes are sometimes referred UCAVs (Uninhabited Combat Aerial Vehicles), but I will not use that locution here. See below for further clarification of these terms and some important distinctions. For a far more in-depth classification of various types and kinds of remote weapon systems see Sparrow (Citation2009). 2. Singer (Citation2009) gained wide press and much of his work discusses these ethical concerns over drones. Recent events such as the potential jus in bello violations wrought by Predator drones in Afghanistan – have received international media attention. Various human rights watchdog groups raised alarm over recent Israeli strikes in Gaza using the Predator platform supposedly against noncombatants. In the past year alone, publications such as The New Yorker, The Atlantic, the Washington Post, Scientific American, the New York Times, and media outlets such as National Public Radio and Public Broadcasting Systems, have all had substantial reports and several highly critical opinion pieces on the use of UAVs. Additionally, there are many in the United States military community itself who do not question the efficacy of UAV usage but rather have principled worries concerning their use such as those mentioned above. It is currently a ‘hot topic’ at professional military ethics and development conferences. 3. Occasionally, such aircraft are instead referred to as Remotely Piloted Vehicles (RPVs). This is especially the case in present US Air Force usage which could be due to public relations concerns over worries regarding autonomous weapon systems. That is, ‘RPV’ emphasizes that these vehicles are still controlled by human pilots. That we see a move away from the UAV moniker back to the RPV idiom in common discourse is telling of the felt need by some in the military community to emphasize that these aircraft still require human pilots (particularly the pilot community in the US Air Force). For more on this phenomenon see Fitzsimonds and Mahnken (Citation2007). See Sparrow (2009) for more fine-grained distinctions on the kinds of uninhabited weapon systems and their classifications. Also see Sparrow (2007). 4. See all of the following for good expositions of the historical trail leading to present day lethal UAVs: Singer (2009), Card (Citation2007), Mustin (Citation2002). For a good overview of the planned future of UAVs, see Office of the Under Secretary of Defense (Citation2006). Note that technically speaking, UAVs are not individual aircraft but weapon systems involving several aircraft and ground control stations. 5. Such as the land-based Foster-Miller's Special Weapons Observation Remote Direct-Action System (SWORDS) or Qiniteq's Modular Advanced Armed Robotic System (MAARS) weapons. See Arkin (Citation2009) for a good overview of such weapons. 6. Some argue that PUR as presented here is false due to the possibility of someone entering into a foolish agreement with another. Imagine if X signs a contract with Y to follow his commands no matter how stupid or irrational they may be. In such a case, if Y orders X to incur potentially lethal risk for no good reason, then X cannot claim that his right is infringed. Assuming X entered into the agreement with informed consent and was not under coercion or exploited, one could argue that Y's order is not morally impermissible. (Thanks to Stephen Kershnar for this objection.) If one wish to grant this is possibility, then PUR can be amended to reflect those relationships where X enters under the authority of Y on the assumption that Y will not order him to take on risk for no good reason; that is, X assumes Y will follow PUR. At any rate, I think most military members in Western militaries implicitly expect their commanders not to risk their lives unnecessarily. 7. That is, in my view there may very well be a self-regarding duty to oneself that is entailed by PUR, but I set aside the possibility here to avoid libertarian objections and paternalism concerns. I am strongly inclined to think, however, that there is a self-regarding form of PUR that could hold up against many libertarian objections so several initially apparent counter-examples to PUR would not actually hold. For example, some may offer the activity of skydiving as a morally permissible act to undertake even though it involves incurring potentially lethal risk. But PUR would allow for the moral permissibility of this action for, presumably, the lethal risk involved in skydiving is actually a necessary part of the good sought by the action. In this case, the good is the thrill and excitement of the act of jumping out of an airplane. Thus, the ‘rush’ sought after by such skydivers (among other possible goods they seek when undertaking the activity) requires taking on the risk. So, according to a self-regarding PUR, it would be morally permissible for an agent to take on the risk of skydiving because that risk directly contributes to the good sought. Whether or not seeking out death-defying activities for the sake of an adrenaline rush is a good that should be sought (or a good at all) is another matter. The PUR does not resolve disputes over what is and is not a good to seek, but rather only the moral demand not to unnecessarily incur lethal risk in seeking an established putative good. 8. Thanks to an anonymous reviewer for suggesting such a principle. 9. Indeed, I think there are strong arguments that can be made against resource expenditures for military ventures in general when it is shown the plethora of other good ways such resources could be alternatively spent. It is an empirical question, certainly, but it is not implausible in the least to imagine that, for example, the roughly $750 billion dollars the United States has spent thus far on the war in Iraq could not have been spent in other ways that would have done far more good in the world. But that is a debate for another paper. For an interesting presentation of the various trade-offs military expenditures impose on a populace, see the National Priorities Project (Citation2010) for the literal financial cost of war. 10. Notice that, if true, then OP carries with it a corollary to pursue the development of and transition to an all-UAV military force: (OPT) For any given state, if it is possible to transition its entire military inventory of inhabited aerial vehicles to UAVs without a significant loss of capability, then that state has an ethical obligation to do so. 11. Although it certainly appears to be technologically possible since there are already UAVs in operation. See below for some discussion on this. 12. Although, granting the ethical obligation to protect the just warfighter I lay out below, this would have to be a truly astronomical cost – particularly in relative comparison to the amount presently spent on defense budgets. Of course, it's entirely possible that morality demands resources be spent on other things entirely outside of defense costs, such as education, development, and the like. But that is another matter. 13. Notice that although this paper is focused on lethal UAVs, the corollary claim of OPT (see note 10) would hold for all aircraft. That is, even cargo planes and the like (even those used to transport soldiers), should be transitioned to UAVs. The idea is simple: risking one less person on the flight (the pilot or pilots) is better than risking them if not necessary. However, of course, it is quite possible that troops would refuse to fly on a plane without a present pilot. If that is the case, then that would be a ‘significant loss of capability’ and so, perhaps, UAVs are not equally capable as inhabited aircraft in the case of troop transporting cargo planes. An example of such platforms for small cargo loads is the Mist Mobility Integrated Systems Technology (MMIST) CQ-10A SnowGoose which is already operationally capable. 14. Singer (2009) also discusses the use of bomb-disarming robots and the connection between their use and the use of remote robots and weaponry more broadly by military forces. 15. Thus, while it is true that UAV usage is ethically impermissible in such instances, so too would any kind of strike via any kind of weapon system (inhabited or uninhabited) be impermissible; the impermissibility does not derive from UAV-specific employment. That is, it is not the UAV qua UAV that makes such a strike impermissible. 16. I will note briefly that I am not even airing objections that claim UAV employment is ‘weak’ or somehow not ‘tough enough’ or ‘cowardly.’ These responses against UAVs derived from some type of wrong-headed machismo are certainly common but, one hopes, are not taken seriously by any military policy decisionmakers. See below for discussion on how the perception of the UAV by enemy forces, however, could have an impact on determining its capability. 17. Some worry over this development and others laud it. Arkin (2009), for example, sees the development of UAVs as in-line with the development of IAWs and focuses his work on how to develop such autonomous systems to follow the Laws of Armed Conflict and various Rules of Engagement (how to give the machines an ‘ethics upgrade’). Arkin contends that such developments are moral improvements and should be pursued with vigor. 18. For some helpful efforts to this end, see Sparrow (2007). 19. Indeed, precisely because I do agree with the first premise of objection 1, I argue that now is the time to institute policies that would block IAW development even while we develop UAV and other remote weapon systems that are human controlled. 20. Many thanks to an anonymous reviewer for this objection. 21. This position is controversial and widely discussed. Or, better, the entire notion of who should bear risks in any conflict is greatly contested, but most of the debate hinges on questions of liability, debates over the doctrine of the moral equality of combatants (MEC), and distinction issues. If a just warrior was fighting in a truly just war and the innocents in question were truly innocent and in no way liable, some will argue their moral status (the just combatant and the innocent) is equal. And, thus, while just warriors should do all they can to shield the innocent from harm, they should not treat the noncombatants’ worth as above their own safety. And once the doctrine of the moral equality of combatants is disposed of (see below on McMahan), it can become difficult to sustain my view (that the just warrior ought to bear the burden for shielding innocents from harm) for all cases, particularly once strict boundaries between combatants and noncombatants are questioned. Seth CitationLazar has argued that McMahan's position becomes untenable in precisely this way because liability will extend to far too many noncombatants in ways that should make them justifiable targets in McMahan's rubric. See Lazar (Citation2010). Steinhoff (Citation2008) has challenged McMahan on his rejection of MEC by using claims from within McMahan's own theory. See McMahan (Citation2008) for a thorough response. 22. Nota bene: In this objection I am focusing solely on the technical ability of UAVs to discriminate properly; I will consider the impact on the psychology of the warfighter (and subsequent in bello worries therein) below in objection 2. 23. And see Plaw (2010) cited above for all the various references used in creating the database. Regarding the estimate for the global causality ratio, see Osiel (Citation2009: 143) and Kaldor (Citation1999: 8). 24. Many thanks to several UAV pilots for firsthand accounts and discussion of these phenomena (their identification is withheld by request). See Wallach and Allen (Citation2009) and Singer (2009) where the cavalier attitude of treating UAV operations like a video game is discussed. 25. Militaries could even go so far as to force UAV operators to live in bunkers and set off fake mortar rounds and so forth around the compound in order to make it feel more ‘real’ if such effects were shown to help overcome this supposed problem of cognitive dissonance caused by being too far from the battlefield. 26. See, for example, Mayer (Citation2009). Notice that nowhere in this long article does the author ever discuss the ethical justification of UAVs themselves. As nearly all recent discussions of UAVs in the public square do, it goes into great detail regarding the ethical concerns raised by targeted killings, sovereignty issues regarding operations in Pakistan, worries over military functions being carried out by a non-military department of the government (the CIA), and so forth, without ever discussing the ethical justification of UAVs in principle. 27. For an interesting discussion of the justification of assassination see Kaufman (Citation2007). See also Kershnar (Citation2004). For an argument that many leaders should not be treated with non-combatant immunity but are legitimate targets, see Kershnar (Citation2005). Presumably, the UAV-targeted leaders of Al-Qaeda and the Taliban would fit as such targets under Kershnar's argument. Gross (Citation2006) gives the argument that targeted killings cannot fit into a proper moral category. If they are an extension of law enforcement, they fail due process, and assassination as self-defense seems implausible, or so Gross argues. 28. For a helpful account of the future of this new form of warfare, see Dipert (Citation2010). 29. I will note, however, that such an empirical question would be very difficult to determine in many contexts. That is, the question is not if a given local populace would display resistance to munitions being dropped by UAVs (that is likely). Rather the empirical question that would have to be determined is if that resistance to UAVs is significantly greater than what resistance would be encountered via an inhabited aircraft. And that would be hard to determine, to say the least. 30. This kind of objection is raised by Phythian (Citation2010). 31. For a helpful discussion of this point see Kolff (Citation2003). 32. A further reason to think that the removal of the warrior from the theater of combat in itself is nothing particularly new is the existence of Intercontinental Ballistic Missile operators and the like who are certainly removed from the theater of combat where their munitions would be delivered. (Thanks to an anonymous reviewer for this point.) 33. An important point to note here is that the operators of UAVs would be considered combatants under the traditional just-war theory rubric. (Thanks to Uwe Stienhoff for raising this point.) I'll remain neutral on this point (if for no other reason than that I reject the moral equality of combatants thesis), but note that whatever one's division of combatants, the UAV pilots would certainly still qualify. This has the (perhaps odd) result of meaning the UAV operators would be legitimate targets under most just-war accounts, even though they would be conducting operations from their office thousands of miles away (at places like Nellis Air Force Base, Nevada.) So be it. 34. Interestingly, McMahan holds that those who fight without a just cause cannot, in principle, ever satisfy the jus in bello principle of proportionality. See McMahan (Citation2004). But a traditional advocate of MEC would disagree, of course. For an attempt to defend some of the traditional elements entailed by MEC, see Benbaji (Citation2008). 35. For a good discussion of some of the complexities and difficulties of symmetrical rights to personal defensive measures enjoyed by soldiers on the traditional just-war theory model (contra the asymmetry of a right to personal defensive measures in individual self-defensive cases where culpability is included in such determinations) see Emerton and Handfield (Citation2009). For a good note on this issue of liability and defense see McMahan (Citation2005: 10) and McMahan (2009). 36. Additionally, there is a different concern with the asymmetry created by advanced technology which worries that the asymmetry makes it too easy to go to war and thus lowers a nation's jus ad bellum threshold too far. I will address this separate concern below. 37. Paul Bloomfield, personal correspondence, 12 July 2010 and discussions held at the University of Connecticut ethics reading group, summer 2010. 38. Steinhoff is quoting from Münkler (Citation2003: 234), Steinhoff's translation. In the original paper Steinhoff is discussing the general lack of bravery involved in attacking the defenseless as part of the debate over torture. The quote is referencing an attitude against any remote weaponry where the warrior has effectively removed his/herself from risk, and, as such, can easily be applied to UAV usage. Note that it further affirms the point above that if there is some kind of dishonorable fight for UAVs due to the asymmetry of the combat, the threshold was crossed long before UAVs (as in Münkler's referencing Tomahawk missiles) and is therefore not endemic to them, in particular, but to modern warfare across the board. The use of the phrase ‘pest control’ to describe what is seen as this particularly non-courageous form of warfare was discussed in my panel at the 7th Global Conference on War and Peace, 2010, Prague, Czech Republic. Many thanks to Uwe Steinhoff for permission to use the remark and directing me toward Münkler's work. 39. Bloomfield argues that the root of the ‘pest control’ worry is our aversion to being treated as pests ourselves were we to be attacked via remote control. If UAVs were used by our enemies against us, we would think they are wronging us in some way to kill us in this manner (wronging us over-and-above the killing itself, that is.) Thus, the thought runs, we should extend this respect for all people into all contexts, even against the unjust enemy: that is, all humans deserve the respect of not being killed via such remotely controlled ‘pest control’ measures. 40. Notice I say the ‘massive reduction in risk’ not total reduction for, presumably, on most just-war theory accounts, the UAV operators would still be considered liable targets for attack since they would most certainly be combatants. For more on UAV operators being legitimate targets, see Singer (2009: 386). 41. Thanks to the audience at the 2010 International Society of Military Ethics Annual Conference in San Diego for the thorough discussion of this objection. 42. It could be objected here that this analogy does not hold because of different obligations that arise from ‘purely’ defensive military technologies (such as bullet-proof vests) as opposed to those offensive weapons that serve to increase defensive capabilities (such as UAVs). The distinction between offensive versus defensive military capabilities is contentious for many reasons. For one, anything (such as a vest) that increases a soldier's defensive abilities will thereby increase that person's va;ue as an offensive force. But, I will not argue this point here. If one is convinced that my story regarding Zandar does not apply to UAVs due to the offensive/defensive distinction, then the entire thought experiment could be re-cast with the use of eighteenth century muskets versus the use of contemporary M-16 rifles as the competing choices. The muskets (clearly offensive weapons) would reduce the troops’ defensive capabilities because they take longer to load, are less accurate, etc. After replacing vests with M-16s, the results of this thought experiment, mutatis mutandis, would be the same. Thanks to Donald Joy for helping to develop this point. 43. I should note, of course, that there is a long history of arguments for disarmament that proceed precisely along these grounds; particularly for nuclear weapons. Notice, however, that the strongest arguments of this type are advanced against particular technologies that are viewed as ethically problematic in principle in isolation from other evils. That is, there is something wrong with (say) nuclear weapons or landmines in principle (the inability to discriminate, etc.) that provides an impetus for banning them in the first place, wholly apart from what future harms they could make more likely. Additionally, by reducing them we reduce the future chance of their unjust use. But this is precisely because any future use of them would be unjust so we can have a certainty that if they were ever used in the future, such use would be unjust. This is entirely different for UAVs, which can be used justly in some circumstances. 44. Although I have not fully argued for this stronger claim here. Again, see Plaw (2010) cited above. Of course, if this stronger claim is true, it would press an even greater ethical obligation to employ UAVs and transition military inventories to all-UAV forces.