The Volokh Conspiracy

The Wall Street Journal has an article in its Thursday, March 15, 2012 edition titled “The Robots Are Coming to Hospitals.”  Reporter Timothy Hay explores ways in which robots are being deployed to transport linens, laundry, and other things around hospital complexes – which are, of course, often enormous facilities.  (I believe this is an open link.)  I have remarked several times here at Volokh that robotics has a natural place in health care – and, as I’ve said, even more so in nursing than in the operating theater.  (I regard it as the new “plastics.”)  The Journal article seems to agree:

In the next few years, thousands of “service robots” are expected to enter the health-care sector—picture R2D2 from “Star Wars” carrying a tray of medications or a load of laundry down hospital corridors. Fewer than 1,000 of these blue-collar robots currently roam about hospitals, but those numbers are expected to grow quickly. As America’s elderly population grows, the country’s health-care system is facing cost pressures and a shortage of doctors and nurses. Many administrators are hoping to foist some of the less glamorous work onto robots.

This could create a potential bonanza for software and application developers to write new programs for them, investors and industry watchers say. “My guess is that in five years, there will be 10 times the number of robots deployed in hospitals that there are today,” said Donald Jones, a managing director at Draper Triangle Ventures, who is backing privately held robotics company Aethon Inc. “We are just not going to have enough human hands to do all the work.”

These technologies will piggy-back off of many existing and emerging technologies; some of their first and most important roles will be adaptations of technologies developed for warehouse fulfillment centers such as those used by Amazon.  The hospital environment is chaotic – but it is also a defined space, in the way that a warehouse is.  Hospitals are already designed to work with wheeled machines – smooth floors, elevators, etc. The transition to use machines in place of orderlies is a natural step; more interesting will be as other tasks in the nurse’s repertoire also become highly assisted, if not entirely done, by machine.

The direction of robot design in these kind of consumer uses depends in part, as leading commentator Ryan Calo has noted in several important papers, on the requirements of products liability law.  He suggests in particular that a highly stringent products liability law will push in the direction of smarter appliances – but nonetheless appliances aimed at highly specific tasks.  A less stringent products liability law will enable the development of more flexible robots that gradually become multifunctional – closer to Robbie the Robot of science fiction.

These questions and many other related matters will be taken up in an important conference taking place April 21-22, 2012 at University of Miami law school, “We Robot.”  (I am very sorry I did not get off my bottom and propose a paper for the conference – in my case, probably something by Matthew Waxman and me on autonomous weapons systems.)  This will be the place where all the cool people who are “interested in the social life of things” will be.  I don’t know that there will ever be a field of “robot law”; I have doubts that it is needed as a legal speciality as such – but I do think there are, and will be, many legal questions of how robots will be adapted into broader social life.  The Miami conference (current program here), organized by Michael Froomkin, is a place to begin.

Wired’s Chris Anderson points to a new report by the Hizook robotics portal on the relatively limited levels of venture capital funding for robotics.  I was surprised – I assumed there was more.  Information is hard to come by, Hizook says, but the post has a chart with annotations on various VC investments in robotics for 2011.  The comparison for 2011 is on the order of $160 million for robotics versus $6.9 billion for web-based companies.  The robotics figure almost certainly undercounts and probably by a lot, and the category of web-based companies in some ways is too inclusive to mean very much.  It’s an order of magnitude comparison, really, but I am struck at the relative paucity of VC funding for robotics.  That means, of course, that much of the robotics development takes place within large existing technology corporations.  Hizook points to some of the barriers for VC funding in robotics (comments are open):

[G]etting VC funding for robotics is a decidedly tough nut to crack. Robotics companies have large capital requirements for robot hardware, few potential acquirers, and almost no “Google-scale” breakout success stories (ie. IPOs). I mean, c’mon… one of the best known robotics companies, iRobot, has a market cap of just $700 Million. This makes robotics a difficult sell to your typical VC firm. My hope is that this list can give others courage to pursue “swing for the fences” type projects along with a source for robotics-friendly VC firms.

China has been moving to catch up with the US and Israel in production of military UAV drones, reports the Wall Street Journal today, in an article by Jeremy Page (Friday, Nov. 19, 2010, A11). The article says that the uptick in drone output has surprised the West:

Western defense officials and experts were surprised to see more than 25 different Chinese models of the unmanned aircraft, known as UAVs, on display at this week’s Zhuhai air show in this southern Chinese city. It was a record number for a country that unveiled its first concept UAVs at the same air show only four years ago, and put a handful on display at the last one in 2008. The apparent progress in UAVs is a stark sign of China’s ambition to upgrade its massive military as its global political and economic clout grows.

I don’t think Western experts should have been all that surprised, at least looking to the long term. As has been noted repeatedly here at VC, drones are not some fantastically advanced technology, beyond the reach of all but DARPA. On the contrary, the avionics and flight control mechanisms have been around for a long time, with tweaks to the basic concept of remote controlled flight provided by advanced in communications and computers.

Sometimes journalists and others make dire predictions about the US or Israel having set off an “arms race” over drone deployments – the US and Israel, then China and Russia, then India and Pakistan …. but this misses the point. Drones will spread because they will take over significant parts of civil aviation in coming decades, no matter what, and that will be so in any industrialized economy. The technology is widely available and represents a vast cost savings – military aviation has many additional reasons why drones are useful, but this is part of a broader wave for all aviation.

It is not really all that different from DARPA subsidizing research into self-driving vehicles. This has obvious applications to urban warfighting, which is why DARPA has funded it for years – but winning researchers from the DARPA competitions for self-driving vehicles have now moved over to work with Google, finally deploying self-driving vehicles on the streets of the Bay Area this year. It’s not an arms race; it is the future of parts of vehicle automation for both civilian and military vehicles.

The real areas of technical competition in UAVs are not in avionics, nor in the weaponry – though improvements there will make them smaller and more discriminating as well – but in the sensors deployed on the drones. Sensors are hard – even today, the drone sensors, as far as we know publicly, are still in the range of video. There’s a lot of room for sophistication. The Economist had a good article recently on both the difficulties and the gradual improvements in the abilities of robotic “eyes” to “see” things. That’s the future technical competition in robotics, or at least an important part of it – much less so the avionics.

As for arms races, the true arms race in the military UAV world will not be a race to deploy – everyone who wants UAVs will have them, in various sizes. The race that matters will be the technological counters to drones – the counter-technologies that will bring them down out of the sky. That, we have yet to see deployed, but it will arrive very soon.

The idea of robotic cars that drive themselves is a good one, I think, and one whose time is rapidly coming.  The New York Times reports on Google undertaking test drives of cars in the San Francisco area with robotically controlled cars, including a drive down Lombard Street, a famously hilly and difficult street.  The engineers are those who took honors at recent DARPA contests for creating vehicles able to self-navigate urban settings; these are the top people in the business:

During a half-hour drive beginning on Google’s campus 35 miles south of San Francisco last Wednesday, a Prius equipped with a variety of sensors and following a route programmed into the GPS navigation system nimbly accelerated in the entrance lane and merged into fast-moving traffic on Highway 101, the freeway through Silicon Valley.  It drove at the speed limit, which it knew because the limit for every road is included in its database, and left the freeway several exits later. The device atop the car produced a detailed map of the environment.

The car then drove in city traffic through Mountain View, stopping for lights and stop signs, as well as making announcements like “approaching a crosswalk” (to warn the human at the wheel) or “turn ahead” in a pleasant female voice. This same pleasant voice would, engineers said, alert the driver if a master control system detected anything amiss with the various sensors.

The test drives have a human navigator in the car as well as an expert human driver at the wheel to take control if something went wrong; the Times article says that assuming human control is no more difficult than ending cruise control.  None of which I doubt at all.  The article added, however, that Google had carefully examined the California vehicle code to determine that the experimental cars were legal to drive on the road:

But the advent of autonomous vehicles poses thorny legal issues, the Google researchers acknowledged. Under current law, a human must be in control of a car at all times, but what does that mean if the human is not really paying attention as the car crosses through, say, a school zone, figuring that the robot is driving more safely than he would?  And in the event of an accident, who would be liable — the person behind the wheel or the maker of the software?

“The technology is ahead of the law in many areas,” said Bernard Lu, senior staff counsel for the California Department of Motor Vehicles. “If you look at the vehicle code, there are dozens of laws pertaining to the driver of a vehicle, and they all presume to have a human being operating the vehicle.”

The Google researchers said they had carefully examined California’s motor vehicle regulations and determined that because a human driver can override any error, the experimental cars are legal. Mr. Lu agreed.

I am particularly curious whether Google had, or perhaps ought to have in the future, an obligation to let appropriate California authorities know that it was test-driving experimental cars with at least some question as to whether the configuration of a human poised to take over is safe and effective.  In fact, I wonder if it did so in advance – the article gives a California Department of Motor Vehicles counsel’s view, but I wonder whether that came before or after the fact?  Is there some obligation to warn the local law enforcement before undertaking something like this?  And to be perfectly blunt, suppose that it were Toyota that had been doing this?  Would the reaction have been quite so agreeable?

I am delighted to see this kind of technology moving forward, and definitely agree that the technology is ahead of the law in some of these areas.  As a lawyer interested in robotics and the law, I do have some questions as to the appropriate protocols in place for testing new technologies, and whether there are obligations to let the public know in advance, or authorities know in advance, what one is doing.  I have no doubt that Google carefully checked its legal position beforehand.  Query whether this is exactly the right set of legal rules for testing new technologies that, in order to see whether they can function as they are intended, require that they be tested in, among, and with the public.  I do not have a settled view on how these rules should work.

Update:  Alert commenter points to an article at Jalopnik on the question of whether Google’s cars are legal on California roads; the article says that Google alerted local authorities.  But my question is still, what should be the rules for deciding whether the car is safe to take in traffic among people like you and me?  Who should be able to decide that?

In a certain sense, after all, what Google is doing is a form of human subject experimentation, of the kind that were it a university, would likely at a minimum require discussion with a human subjects committee, informing the subjects, etc.  Of course, one can say that this is not the proper analogy – and I would agree as a lawyer looking at California’s codes – but I would still ask the question, without having at this point a fixed answer, should it be?  The nature of Google’s testing requires that it be done among the driving public, by assumption among people who don’t know there is a robot among them; should this be regarded as ethical and should the legal rules allow for it in dealing with robots?

From a technology and society standpoint, the long term gains from this kind of technology will presumably be once we mandate that all cars be driven by robots.  Among other things, if things work as hoped, it will allow for much, much smaller spacings between the cars and higher road speeds that will extract greater efficiencies from roads and highways, and help manage congestion problems.  Being able to read the Volokh Conspiracy on your way to work is vitally important, of course, but not the only potential gain of the new technologies.

Update 2:  Several of the comments suggest that the issue isn’t really a big one, since the “driver” that Google has placed behind the wheel is ultimately legal responsible for the operation of the vehicle under the law, as it currently stands.  I’m sure that Google has researched this question as a matter of current law thoroughly and I don’t doubt the conclusion.  Query once again, however, whether that would be a stable legal rule of liability in the case of ordinary people for whom the vehicle has come equipped with this technology; if something goes wrong, you are responsible and then you turn and sue the manufacturer?

But we’ve been down these paths before; it seems to me it would be hard in some future where the technology is widespread and beyond the ability of a human operator to have much sense of whether the machine is properly operational, to have any very direct form of liability, either as a matter of tort or a matter of the criminal law of the vehicle code.  People simply will conclude, with good reason, that it would be unfair and inefficient to hold them responsible for the operation of technology without any clear way of knowing whether it works or not.  The responsibility then moves backwards to the manufacturer, and other points of expert contact, but it is hard not to conclude that in terms of the actual operation of the vehicle, that point of accountability is lost.  It might well – I am pretty certain that it would – be made up by safety and other gains from the widesrpead use of the technology, but I would not at all believe that the current liability rules in either tort or the criminal aspects of the vehicle code would remain legally stable.

And the idea of the “system” acting as a “driver” when everyone is mandated to use the system – well, I assume at that point, we shift entirely to a different arrangement for liability.  The idea of the “system” as “driver” as “liable” is interesting, but I would assume that at that point, it morphs into some insurance system of liability and compensation.

(No more updates for me – I’m allowing myself to be distracted from writing about the operational and strategic uses of UAVs.  Can I get a robotic system to write that for me?)

Adam Entous, Julan E. Barnes, and Siobhan Gorman have an outstanding piece of national security reporting on the front page of the weekend Wall Street Journal, “CIA Escalates Campaign in Pakistan: Pentagon Diverts Drones to Afghanistan to Bolster Campaign Next Door.” This is a fine piece of journalism that integrates reporting from AfPak and Washington to present findings that are new to the public, and more than merely a deliberate leak to a leading reporter from a government official or a magazine story rather than hard news.  My congrats to what is emerging as a leading national security affairs reporting team at the news pages of the Journal.  (Here is Greg Miller’s account in the Washington Post, Sunday, front page.)

September 2010 saw another sharp uptick in the number of drone attack missions in Pakistan.  The question behind the raw numbers is what strategic purpose they aim at.  One strategic mission of drone missions in Pakistan is counterterrorism aimed at Al Qaeda leadership.  This uptick in September 2010 points to a quite distinct function – rather than counterterrorism as its own mission, the purpose is, as article says in a telling quote, “force protection” for the US counterinsurgency troops in Afghanistan.  The articles details an increasing impatience of the US military and political leadership with Pakistan’s government, and an increased willingness both to strike overtly using NATO military assets quite openly across the border, as happened in the last week, as well as to use CIA Predator attacks in the border regions.

Moreover, the “force protection” use of drones described in these articles is distinct from still another strategic use of drones, one recounted in earlier articles in the last two weeks, talking about their use to disrupt the planning of attacks against European targets by groups such as the Haqqanis, regional groups thought to be seeking to use people with European or American passports to strike from Pakistan against Western targets; Mumbai shifted further west, so to speak.  As Woodward quotes someone in his new book, “Mumbai changed everything.”  It is because of these overlapping but also separate and shifting roles for drones that it seems to me worthwhile to analytically distinguish them, as I do below.

Thus, the CIA attacks are now on safe havens for Taliban who are part of the fight in Afghanistan but taking refuge in Pakistan.  Rather than simply a raiding strategy against terrorist leadership in Pakistan as an exercise in counterterrorism, the drones are now used a raiding strategy against the safe havens as part of the Afghanistan counterinsurgency surge.  Hence the desire of the Pentagon to divert drone aircraft – which are in demand in Afghanistan for a variety of missions – from Afghanistan to attacks in Pakistan on bases that are seen as links for attacks on US forces.

This is an important shift, or addition, to the role of drones in Pakistan.  (Of course this has always been part of the use of drones; I’ve hardened the analytic categories to make them clearer, but really it is a question not of something new, but of scaling it up.)  The article makes note of something else, too – that drone aircraft are not being produced fast enough to meet demand for them in AfPak.  The article has excellent graphics, including a chart on numbers of attacks on a month by month basis, and maps.

As it happens, this article is timely for me, as I am completing this weekend the draft of an essay for the Hoover Institution on a roster of strategic uses of drones.  In bullet point form, here is an analytic breakdown of categories, as I see them, of drone use.  (I’m not providing more than the bullet title, even though the result is overly-cryptic; the full essay will be available once finished and edited at Hoover’s website or SSRN.  Also, if anyone is interested in my earlier published writing on drone warfare and the law, at SSRN’s free downloads, see this book chapter, this lengthy piece in the Weekly Standard, and two pieces of Congressional testimony, here and here.)

• Surveillance (sometimes policy and legal people forget this in all the controversy over weaponized drones and targeted killing).
• Drones used by the military or CIA in Afghanistan as part of combat operations and counterinsurgency.
• CIA drones in Pakistan used in counterterrorism against high value terrorist leadership (ie, against senior leadership such as Mullah Omar or Bin Laden or others in AQ).
• CIA drones in Pakistan used in counterterrorism against high and lower level operatives, including the current strikes against locations and camps of terrorist groups apparently planning new strikes in Europe or the US (as some of the CIA drone strikes undertaken now are apparently intended to do, such as those against the Haqqanis).
• CIA drones used in Pakistan in support of Afghanistan counterinsurgency operations against safe havens for Afghanistan Taliban.
• CIA (or even military) drones used in Pakistan in support of Pakistan government counterinsurgency operations against the Pakistan Taliban, separate from the fight in Afghanistan.
• CIA drones used in counterterrorism against AUMF targets (ie Al Qaeda or associated forces, in the context of participants in the current non-international armed conflict (NIAC), as legally defined), whether in Yemen, Somalia, or conceivably elsewhere.
• CIA drones used either as part of the AUMF-NIAC or else as (legally independent) self-defense operations, against US citizens who have taken up terrorist participation and made themselves subject to targeting (operationally not really different, but legally potentially raising different law and policy questions because of the US citizen status).
• CIA drones used somewhere down the road against terrorists unrelated to anything going today, whether in an ongoing armed conflict or as an operation in self-defense.

The list proceeds more or less according to an expanding political geography starting with Afghanistan; it deliberately leaves aside Iraq for these purposes.  The main internal analytic axes are political geography; who uses the weapon and on whose behalf; and who is targeted by the weapon.

The long-term question of drones is whether they are going to remain a remarkably useful weapon in support of a large variety of missions in different ways, or whether instead the US decides to try and leverage them into something much more strategically radical – the new strategic air power.  In other words, the latest iteration of a very old dream, the ability to win wars from the air.  But this time with a twist.

Strategic air power in the past both promised to win because it could deliver apparently huge amounts of kinetic energy, but then tended to lose (or at least not be decisive)  because the huge amounts could not be targeted in order to achieve the required strategic aims.  (For a discussion of recent, Kosovo war-era debates over the ability to win wars from the air, look at this excellent article from Byman and Waxman.  Moreover, Israel went down the air power road in Lebanon, hoping to avoid serious ground engagement, and quickly ran into major problems.)

Drone attacks and targeted killing, in the full realized sense with a range of vehicles from very large to tiny, with weapons ranging from large to small, and with improved sensors arrays and processing integration, hold out the hope of being able finally to deliver the blows precisely where one wants.  We think of the virtue of targeted killing as being discrimination for its own sake, limiting the damage of war.  In pursuit of strategic air power victory, however, we might do better to think of discrimination in targeting for the sake of (finally) being able to put power precisely where we want it.

It might work out that way; it might not.  Drones have not yet been met in the field with counters – surface to air missiles, for example, of the kind that dealt a crippling blow to Soviet air superiority in Afghanistan.  Or technological counters to the remote-control communications systems that allow drones to be directed from near or far away. What concerns me from a strategic standpoint is that the US might decide that drones are cheap, reduce risks to but also need for manpower, and that it finally has the ability to achieve its aims (as Vice President Biden has implied in his early-on stance in favor of drone counterterrorism rather than counterinsurgency war in AfPak), not through the burden and cost of boots on the ground, but over the horizon using this new iteration of the dream of strategic air power.

It seems unlikely to me that this will work.  I understand the attraction.  And I also understand that it does represent an advance on an earlier version of strategic air power – the post WWII idea that strategic nuclear air delivered weapons meant we no longer needed a large standing army, because of the nuclear deterrent in an over the horizon way.  That version of strategic air power could not work to the end of reducing the need for a conventional army, because the threat of nuclear war was too monumental to be useful at the margins.  You could never actually pull the nuclear trigger over something important but not so important to go to nuclear war over.  So it turned out that you either ceded ground incrementally at the margin, or else you went back to having a conventional army that could respond at the margins.

Drones potentially make that less of a problem, precisely because they allow calibrated responses in a discriminating way.  But it seems doubtful to me that a technology will not rapidly develop technological responses, and that in any case, it is a huge advance, but not a truly decisive one; it seems to me likely to remain a useful tool if regarded that way, not a panacea for having a sizable military.  In any case, history is littered with instances of believing that at last we had found the successful path to strategic air power victory.

Update: Woodward’s book, which I’m reading now, discusses many issues relevant here.  Four stand out for this topic.

• First, he notes that the number of drone strikes under Bush was tiny, in large part on account of an enormous fear of the consequences of civilian casualties, even in numbers that the administration believed were entirely justifiable – fears, in other words, of accusations of atrocities, war crimes, etc., from the fear of a de-legitimizing activist campaign.  The Obama administration, believing correctly that it was immune to such campaigns, did not have to worry about such repercussions.
• Second, the earliest drone strikes in Pakistan under Bush had only limited effectiveness, because the US, out of concerns for Pakistani sovereignty, advised the government of impending strikes; elements of Pakistan’s intelligence service warned Al Qaeda, and they dispersed from the target zone prior to the strike.
• Third, he notes that Michael Hayden, while embracing the use of drones as counterterrorism against high value targets, did not believe that it could “win” the struggle against Al Qaeda or the jihadist extremists, because the pinpricks were not enough to root out the movement even if leaders were killed.  Hayden thought the drones essentially tactical rather than strategic.
• Fourth, he says that the reason the drone were, and are, effective is because of a strong effort over five years to create a ground level network of intelligence of critical value – developed it seems out of the CIA from human intelligence.  That, integrated with massive advances in signal and communication surveillance, has enabled drones equipped with still not much more than tactical video surveillance in the air to be directed to the crucial targets.  It is not the surveillance gathered by a weaponized drone that matters so much as the intelligence gathered in a combination of on the ground human intelligence and communications monitoring that allows an effective strike.

One of the biggest implications for drone strikes by the CIA in Pakistan and Afghanistan, as I read the Woodward book, then, is that the CIA (under Hayden and surely the same under Panetta) regards its human intelligence and a certain on the ground presence, as well as communications monitoring outside of the tactical use of the drones themselves, as crucial to their success.  The drones are effective – insofar as the ground is prepared by other kinds of intelligence activities and technologies.

Behold, the personal drone, controlled by Ipad, Itouch, Iphone, available on pre-order for later this year from Amazon.

The Parrot Helicopter Drone

Although officially offline and on vacation, I thought I would share one conversation with one of the rangers here in the national park. She remarked that the ranger services – national parks, national forest, etc. – had been watching with great interest the growth of unmanned aerial vehicles (UAVs) in civilian use. So far this includes things like crop dusting and surveillance. Because the air bases that command some of the UAVs are located in Nevada, relatively nearby (in the Empty Quarter, that might mean 300 miles of desert driving, of course), there is a lot of awareness of UAVs and their potential – University of Nevada, Las Vegas has just begun a program to graduate UAV controllers, for example.

When people talk about surveillance UAVs, they are typically thinking about border patrol, but here, the park services are thinking about fire patrols – an immensely expensive task from aircraft now, because of the vast areas to be surveyed in real time – but worth it because the faster the fire is spotted, the better the chance of containing it before it spreads. LIkewise, search and rescue for lost and injured back country hikers. That one is somewhat ahead of existing technology, for what the park services would ideally like, because flying in the steep valleys and canyons is difficult and hazardous now, but UAV technology is not sufficiently up to speed to take over those tasks. But it will happen soon, as smaller UAVs that are more like large birds can be deployed in difficult, deep, or narrow spaces. Likewise, as the sensor technology gets better, cheaper, and more available, it will be easier to find a single lost hiker using not just things like infrared signatures, but sensor arrays that are … well, if they exist, they are still only available to the military.

Point being that UAVs are going to spread rapidly and widely across a huge array of tasks and functions currently carried out by manned aircraft. It will happen because UAVs will be so much cheaper, efficient, and in many functional aspects superior to using people in airplanes. The impetus will rapidly turn from being military, as it still is now, to civilian. Everybody, everywhere in the world will shift that direction.

I raise this because there is a meme that still circulates with some velocity in the international law community, journalists, and others, that the US is risking setting off some kind of UAV arms race by its increasing roboticization of conflict – not just UAVs, but ground vehicles, and so on. I don’t think that’s right; the meme fundamentally misunderstands the technology and its application. Rather, UAVs are going to spread across a very wide range of aviation in any case, in which military uses will just be one of them. The same technology, cost, safety, efficiency, and so on, drivers that push for fire surveillance in the Sierra Nevada will be exactly the same ones that drive the military to use the technology. One can call it an arms race, I suppose, but only if one imagines that it is all about military use, otherwise it is a misleading way of thinking about the technology.

A better way to think about this is to go back to what make robots robots. In general, there are three conceptual pieces: A locomotion function or means of gross movement or action in the world; computing and central processing power to be able to analyze; and sensors to bring in streams of data which, being analyzed, result in some form of gross mechanical action. (In the case of US military UAVs, we can add an additional piece that brings them into an intersection loosely with ‘cyber’ – the communications net that allows them to be piloted over Afghanistan from the US.) Focusing on the UAV’s gross locomotion part, the flying part, and saying that it will lead to an arms race in which everyone will want one and arm it with a missile misses the point. There is no arms race about that – the technology for flying remotely has been around for decades; anyone who wants to build one can do so at a hobby shop. Putting a missile on it is child’s play, literally – presumably no one would be so politically incorrect as to propose building a Predator with a missile as the next high school robotics competition for high school teams, but apart from political sensitivities aside, one reason is that it’s just too darn easy. Flying is easy; making a machine that walks up stairs is hard.

Everyone will have UAVs because everyone will want them for so many, many things, mostly unrelated to military or police missions. Any government that wants to arm one with a missile will have no difficulty doing so. The real technology issues are not with flying, or with weaponization – or even with computing power. That’s all off the hobby kit shelf. No, the real technology issues arise with sensors. One robotics scientist in Silicon Valley told me last year that it was largely unrecognized, but the real advances in technology of the past decade had not been in computers as such, but in sensors and controllers, ranging from new ways and kinds of bringing data streams online to direct neurological, direct brain control of robotic limbs for amputees, and so on.

But now, note the issue. Some of this technology is classified for military R&D; other parts are not. The importance of robots outside of the UAV context are immense in large part because the Baby Boom generation does not have sufficient children to see us off to our reward; we are going to slide into dementia and be cared for and comforted by cuddly robotic dolls that we will think are human, to judge by where things are going in Japan. In the US, we are not so aware of this, yet, although it is striking that the Times and the WSJ have both moved on in their robotics coverage from targeted killing via UAVs to much more friendly news stories about Alzheimer’s patients in Japan being soothed by robot plush dolphins. Dolphins that will be smart enough to monitor medical conditions and call 911 if needed, to take obvious examples, or monitor whether a patient has taken the meds, or any number of things. What lies behind this is sensor technologies.

In an armed conflict context, however, it is questionable how many of the fighting forces in the world, state or non-state, will feel any great obligation to minimize collateral damage or attempt to more and more affirmatively id a target before striking. If you don’t feel that obligation – I would estimate that the countries involved will be the US and Israel, and the rest of NATO only insofar as it ever intends to do any more fighting, but in any case, they will simply acquire US technology. China will likely do so, because it would at least want every capability, and because it can most likely steal the technology and reverse engineer any missing parts. But either sensor technology will spread across civilian uses, such as elder care robots, so as to make the concept of an ‘arms race’ moot, or else the number of countries that will be “racing” to have such technologies will be almost entirely limited to countries that (a) fight and (b) care about the rules. That makes the list frankly pretty short. It is possible that India might join that list, along with Taiwan, South Korea, and a handful of others in Asia. But there will not be an “arms race” around sensors, because they are useful primarily for reasons related to more discriminating targeting, and the militaries in the world interested in that is not a long list.

Will there be an evolution of arms around UAVs, then? Yes, but not likely along those parameters. The likely arms race is along a quite different one. Predators are slow and noisy for targeted killing; it will not take long before some party – Iran – begins doing what the US did via the CIA in Afghanistan against the Soviets, and supplies rudimentary surface to air missiles to attack the drones. The arms race will get underway in the classic evolution of protecting air dominance. The Predator, for example, might launch not a missile, but instead a still smaller drone with a single-person weapon, specifically designed for up close use. That will be a function not of flying technology or weapons technology, however, but, once again, sensors. But an arms race over air superiority is not one that has the implications for the supposed dangerous spread of this new military technology – introducing dangerous new dynamics between India and Pakistan, for example – that numbers of commentators seem (still) to imagine.

I am returning to the solitude, which is to say, the off-lineness of the mountains.

This interesting essay, by Vik Kanwar of Jindal Global Law School (India), reviews four new books that examine how the laws of warfare may or should change in response to the development of autonomous weapons a/k/a “warbots.”

As many of our readers know, I have long been fascinated by robotics, and have a particular interest in battlefield robotics and related questions of law.  I felt I was late to the cyberwarfare field – and don’t know enough about it – and so have left it for others.  But robotics … well!  Robotics and the law, well, well!  However, one of the important features about Predator drones and UAVs as the US has developed them is that they involve important overlaps between robotics and cyber fields, because the UAV has to be controlled somehow from halfway around the world.  If the classic conceptual parts of a robot are

• gross locomotion and its ability to move and act in the physical world;
• the brain and computing and processing power; and
• sensors to bring data streams into the computational resources, so as to figure out how to move and what gross physical world actions to take …

then, in the case of how the US uses UAVs, we need to add a fourth, the cyber component of communication and control over long distances.  At that point, questions of cyberattack on the robotic system become live.

This brings me to a movie I just watched last night on Netflix, Surrogates – from the comic book series of the same name to the Bruce Willis movie.  It manages to combine robotics with cyber.  Not bad – I thought the critics were overly tough, frankly, but then I have both low standards and low taste in movies.  I liked it.  I think it is a movie that Jack Goldsmith and anyone else working on cyber and robotics issues should see (I will assume that Glenn Reynolds has already watched it … twice).  With popcorn.

[youtube]http://www.youtube.com/watch?v=Zl_h9RaL0es[/youtube]

(Robots as caregivers have suddenly been surging to the front pages of the newspapers – the Wall Street Journal, followed by the New York Times.  I’ll say more about the implications of that later.)

The House Subcommittee on National Security and Foreign Affairs, chaired by Rep. John Tierney, held a second hearing (Wednesday, April 28, 2010), once again on the topic of drone warfare. The first, held back on March 23, was focused on the technology of UAVs; this follow-up hearing was specifically about targeted killing and drone warfare. All of the witnesses this time around were senior international law professors: Mary-Ellen O’Connell (Notre Dame), William Banks (Syracuse), David Glazier (Loyola Los Angeles), and KA.

(I’m traveling, so I’ll post the hearing information when I get home; the AP has a reasonably good summary (considering the difficulty of reporting on four law professors with quite different views politely agreeing and disagreeing with each other …!). The ACLU submitted a written statement (just downloaded but not yet read).)

We four professors represented a pretty good range of the available expert views, and in that regard particularly I think this was a useful hearing. In part, it sought to figure out what this debate meant following DOS Legal Adviser Harold Koh’s statement on drones at the ASIL meeting on March 25. Having been a sharp critic of the administration, or anyway its senior lawyers, including in the first drone hearing, for not stepping up to the plate on this topic, I applaud both the fact of the Legal Adviser’s public statement as well as the substance of his statement.

The pivotal issue – not surprisingly, for those following these debates – is the role of the CIA, whether it takes a role at all and, in particular, where it takes a role. The positions presented ranged from:

• Professor O’Connell’s well-known view that the participation of the CIA is and always is an international crime;
• Professor Glazier’s position that the targets who fit within the AUMF are targetable as part of an armed conflict and that the conflict goes where they go;
• Professor Banks’ position that the participation of the CIA is lawful under both domestic and international law, but that the law in each instance needs to be updated and brought to fit contemporary circumstances; and
• my own view, closest to Professor Banks’, that the CIA participation is lawful, but that there is a separation of the legal rationales and bases, depending upon whether it is a matter of armed conflict or the separate ground of self-defense.

Chairman Tierney did an excellent job of disentangling the highly entangled and conflicting positions among the witnesses. Not easy, because the disagreements are very sharp in some areas, less so in others.

(For my own part, I’m going to have to work on delivering a very clear message about my own views. Like many professors, including our President, I have a tendency to bury the lede in a bunch of hypothetical questions. I framed a series of questions about the role of the CIA – not because I think that its participation is unlawful, but because I thought it would be useful to frame the positions among which I was asserting one. When I read a number of the news stories afterwards, I discovered I had been cited for the propositions in the hypotheticals, not for my actual view that the CIA’s participation is, yes, lawful. My bad – there are reasons why I would never be a good litigator.)

The most striking take-away for me from the hearing was just how much the position of the professional international law community, as ably represented by Professor O’Connell, starts from fundamentally different premises than that of the US government, of any administration, at any time. In this particular instance, Professor Banks and I were, in various ways, channeling the administration – or, better said, the way in which the US government has long approached these questions as a matter of pragmatic law.

These two whole approaches to international law – that represented by Professor O’Connell, and that represented by me, and by Professor Banks and, in this regard, Professor Glazier – the approaches to international, and not just the substantive positions derived from them, are really like ships passing in the night. I tried to say this at the end of my comments: Professor O’Connell and I were arguing past each other, ships passing in the night. But the consequences in the real world are not ships passing; eventually they collide and one or the other prevails.  They can’t really meld into some mixed paradigm; the underlying assumptions are too different.

The problem?  If you talk about international crimes being committed by the CIA, each and every time it carries out a strike, those are words with consequences. It is hard to talk about merely “correcting” state practices to conform to international law in some general sense of the government of the United States, when the paradigm is that there are individuals who under principles of international criminal law, should not be undertaking a course correction by the administration. If that’s what you mean – and, please to observe, I don’t think this is the case at all – but if that’s what you mean, well, It should be indictment, arrest, and prosecution. For murder and extrajudicial execution and assassination. That’s what you mean if you invoke about international crimes; the whole paradigm is designed, by its nature, to reach to individuals and not simply “sides” in international law and politics.

Just to be crystal clear this time, I don’t think these CIA actions are international or domestic crimes. But the view of the administration, which I think generally legally correct, and that of its critics such as Professor O’Connell, really can’t be reconciled. And if you think it is a matter of international crimes (as, I stress, I don’t), then it doesn’t go away even if the Obama administration decided tomorrow to change course. It has undertaken hundreds of these strikes now, over a series of years; give it a couple of years more until it’s no longer President Obama, and then call it “crimes against humanity.”  After all, why not, once the political costs of attacking Democrats, with the genuine intent of finding criminal liability, is gone?

If the stakes are raised, as is currently underway by the soft-law community, to the level of serious allegations of criminality, then the Obama administration now finds itself at the beginning of the process that the Bush administration found itself in regarding detention, interrogation, and rendition. The ACLU is like a dog with a bone; following the relentless logic of collective action coordination failures, it will remain relentlessly focused, gnawing away, and those it seeks to play will be diverted and disorganized in their responses. Death by a thousand paper cuts, I think I wrote in the Weekly Standard. I am as ever stunned by the ability of government officials to think that the advocacy soft law community does not have the ability to set the agenda.

(Side note:  My general sense of the politics associated with this, then? Nasty. Not as nasty as I thought before Dean Koh’s speech, so perhaps I am simply inclined to see nastiness, but still … The table is being set now by the NGOs to turn the agenda into a genuinely international criminal one against individual CIA officials – once a Republican administration is in office. I find it hard to resist the sense that if you are, say, the ACLU or the human rights groups, this is win-win politically. The joys of public choice theory and defection game theory.

The game played by human rights organizations of what I have sometimes called “serial absolutism” is a special case in game theory of serially “moving the goalposts,” which in turn is a special case of serial insincere promising (these are the absolute, unshifting standards, we really, really promise this time!!), serial insincere promising about the procedural and constitutive rules of the game, followed by serial defection (whoops! it’s really crimes against humanity, now that the Republicans are in).  ’Moving the goalposts’ does not get enough attention in the iterated game theory literature, I think.  Unstable coalitions in domestic politics – leftwing Democrats aligned with various foreign and international constituencies, “gaming Spain,” as I’ve sometimes called it – prevent a unified American government from recognizing the pattern of repeat defection and trust-breaking and therefore refusing to take the promise seriously.

Thus, the advocacy groups start with the Obama administration, establishing their bona fides so no one can say that they weren’t even-handed, even though they know that it will only become real, a campaign that ripens into a real chase against individual lawyers, officers, and agents (probably starting with any easy-to-cut-out-from-the-herd military contractors), undertaken and underwritten by a concerted NGO campaign, ratcheting up to a crescendo a couple of years into the next Republican presidency. The only way to slow this down is by creating a policy and considered legal views in the Obama administration that can set a stable precedent for those administrations, Republican and Democrat, that follow.)

Director Panetta has been doing a remarkable job of push back in the press, and certainly with Congress, on the efficacy and focus of these strikes. (Of course, I regard Legal Adviser Koh’s statement as an outstanding beginning for the American government to state its long-term position on the law.) I suspect Panetta has convinced some important, and not otherwise friendly, reporters that the collateral damage is far closer to what the CIA says than what the NGOs say. But I’m not sure even he understands how much the ground is shifting even from the argument over collateral damage to a much more basic one – why does the CIA even engage in the use of force at all? Shouldn’t all uses of force be either military (in an armed conflict) or law enforcement?

Why does the CIA have any function of covert action? That question was plainly on the table in the House subcommittee hearing, and I thought there was a twinge of hesitation on the part of some lawmakers. I remarked – seemingly off topic but in fact at the center of things – the United States, like other leading states, has long believed that its national security and self-defense required a covert civilian service. We thought about ending it in the Church hearings but didn’t. We have had many opportunities to end it, but we’ve never taken them.

In my view, we are right to keep that ability. But there are plenty of people who think it a bad idea on both policy and legal grounds. (For that matter, in my experience, the Defense Department has many serious military people, both operational and legal, who think the use of force by the CIA is always a mistake and indeed illegal. I’ll discuss that view in another post.  Professor Glazier seemed to reflect this view in part, although I don’t want to speak for him on so fundamental a question.)

We are at one of those periodic moments, in other words, when the fundamental question of CIA covert use of force is on the table of public debate again. I hope that Director Panetta pays close attention to the tenor of the hearing last week and understands that among the multiple layers of issues – drones, drones beyond AfPak, the role of the CIA in an armed conflict, the role of the CIA in self-defense – the question of whether the CIA should ever use force is on the table.

The Washington Post has an editorial today praising State Department Legal Adviser Harold Koh’s March 25 statement defending the legality of drone warfare (part of a long speech on several international law topics).  The editorial specifically endorses the Legal Adviser’s invocation of self-defense as a separate ground for using force apart from armed conflict in a technical sense.  (Needless to say, I’m pleased by the editorial.)

… Mr. Koh, an unflinching critic of Bush administration anti-terrorism tactics during his years in academia, cited domestic and international law as foundations for the program. The United States is engaged in an “armed conflict” with al-Qaeda and its affiliates, Mr. Koh asserted, and “individuals who are part of such an armed group are belligerents and, therefore, lawful targets under international law.”

He rightly rejected the absurd notion that enemy targets must be provided “adequate process” before the strike occurs. “A state that is engaged in an armed conflict or in legitimate self-defense is not required to provide targets with legal process before the state may use lethal force,” he concluded.

Mr. Koh’s reaffirmation of the right to self-defense — even outside the confines of an existing armed conflict — is particularly important. The Authorization for the Use of Military Force (AUMF) after Sept. 11, 2001, empowered the president to pursue those responsible for the attacks, including al-Qaeda and the Taliban. That authority may wane with time. But the right of self-defense is inherent and may be exercised against current and future enemies that pose an imminent threat, including those operating outside of traditional combat zones.

I’ve noticed a number of posts and comments around the blogosphere on the spread of UAV technology.  Which indeed is happening; many states are developing and deploying UAVs of various kinds.  The WCL National Security Law Brief blog, for example, notes that India is now acquiring weaponized UAVs:

India is reportedly preparing to have “killer” unmanned aerial vehicles (UAVs) in response to possible threats from Pakistan and China. Until now India has denied the use of armed UAVs, but they did use UAVs that can detect incoming missile attacks or border incursions.

The importance of obtaining armed UAVs grew enormously after the recent attack on paramilitary forces in Chhattisgarh that killed 75 security personnel. Sources reveal that the Indian Air Force (IAF) has been in contact with Israeli arms suppliers in New Delhi recently. The IAF is looking to operate Israeli Harop armed UAVs from 2011 onwards, and other units of the armed forces will follow.

I’ve also read comments various places suggesting that increased use of drone technologies by the United States causes other countries to follow suit, or to develop or acquire similar technologies.  In some cases, the dangling implication is that if the US would not get involved in such technologies, others would not follow suit.

In some relatively rare cases of weapons technologies, the US refraining from undertaking the R&D, or stopping short of a deployable weapon, might induce others not to build the same weapon.  Perhaps the best example is the US stopping its development of blinding laser antipersonnel weapons in the 1990s; if others, particularly the Chinese, have developed them to a deployable weapon, I’m not aware of it.  The US stopped partly in relation to a developing international campaign, modeled on the landmines ban campaign, but mostly because of a strong sense of revulsion and pushback by US line officers.  Moreover, there was a strong sense that such a weapon (somewhat like chemical weapons) would be not deeply useful on a battlefield – but would be tremendously threatening as a pure terrorism weapon against civilians.  In any case, the technologies involved would be advanced for R&D, construction, maintenance, and deployment, at least for a while.

The situation is altogether different in the case of UAVs.  The biggest reason is that the flying-around part of UAVs – the avionics and control of a drone aircraft in flight – is not particularly high technology  at all.  It is in range of pretty much any functioning state military that flies anything at all.  The same for the weaponry, if all you’re looking to do is fire a missile, such as an anti-tank missile like the Hellfire.  It’s not high technology, it is well within the reach of pretty much any state military.  Iran?  Without thinking twice.  Burma?  Sure.  Zimbabwe?  If it really wanted to, probably.

So it doesn’t make any substantial difference whether or not the US deploys UAVs, not in relation to a decision by other states to deploy their own.  The US decision to use and deploy UAVs does not drive others’ decisions one way or the other.  They make that decision in nearly all cases – Iran perhaps being an exception in wanting to be able to show that they can use them in or over the Iraqi border – in relation to their particular security perceptions.  Many states have reasons to want to have UAVs, for surveillance as well as use of force.  It is not as a counter or defense to the US use of UAVs.

The real issue is not flying the plane or putting a missile on it.  The question is the sensor technology (and related communication links) – for two reasons.  One is the ability to identify the target; the other is to determine the level, acceptable or not, of collateral damage in relation to the target. That’s the technologically difficult part.  And yet it is not something important to very many of the militaries that might want to use UAVs, because not that many are going to be worried about the use of UAVs for discrete, targeted killing.  Not so discrete and not so targeted will be just fine – and that does not require super-advanced technology.   China might decide that it wants an advanced assassination platform that would depend on such sensors, and in any case be interested in investing in such technology for many reasons – but that is not going to describe Iran or very many other places that are capable of deploying and using weaponized UAVs.  Iran, for example, won’t have super advanced sensor technology (unless China sells it to them), but they will have UAVs.

(The attached weaponry follows the same pattern.  Most countries will find a Hellfire type missile just fine.  The US will continue to develop smaller weapons finally capable of a single person hit.  Few others will develop it, partly because they don’t care and partly because its effectiveness depends on advanced sensors that they are not likely to have.)

Robots are broadly defined by three characteristics – computation, sensor inputs, and gross movement.  Movement in the case of a weaponized robot includes both movement and the use of its weapon – meaning, flying the UAV and firing a weapon.  The first of those, flying the UAV, is available widely; primitive weapons are available widely as well, and so is the fundamental computational power.  Sensors are much, much more difficult – but only to the extent that a party cares about discretion in targeting.  But it is not the case that they are making these decisions on account of US decisions about UAVs; UAVs are useful for many other reasons for many other parties, all on their own.

It’s not true that everything I do is about drone warfare, but it has taken a lot of my time lately and, of course, a lot of stuff is happening, both on the operational side as well as legal side.  So here’s a little round-up of links, more or less at random.

First, the New York Times has a front page article today, looking at the impact of drones on terrorist and militant activities in Waziristan.  It is an interesting piece, not least because it acknowledges not just the effectiveness of the campaign, but the effectiveness of the Obama administration’s sharply ramped-up CIA campaign.  (HT Instapundit.)

A stepped-up campaign of American drone strikes over the past three months has battered Al Qaeda and its Pakistani and Afghan brethren in the tribal area of North Waziristan, according to a mid-ranking militant and supporters of the government there.

The strikes have cast a pall of fear over an area that was once a free zone for Al Qaeda and the Taliban, forcing militants to abandon satellite phones and large gatherings in favor of communicating by courier and moving stealthily in small groups, they said.

The drones, operated by the C.I.A., fly overhead sometimes four at a time, emitting a beelike hum virtually 24 hours a day, observing and tracking targets, then unleashing missiles on their quarry, they said.

The strikes have sharpened tensions between the local tribesmen and the militants, who have dumped bodies with signs accusing the victims of being American spies in Miram Shah, the main town in North Waziristan, they said.

The impact of the drone strikes on the militants’ operations — on freedom of movement, ability to communicate and the ease of importing new recruits to replace those who have been killed — has been difficult to divine because North Waziristan, at the nether reaches of the tribal area, is virtually sealed from the outside world.

None of those interviewed would allow their names to be used for fear for their safety, and all were interviewed separately in a city outside the tribal areas. The supporters of the government worked in positions where they had access to information about the effects of the drone campaign.

Along with that of the militant, the accounts provided a rare window on how the drones have transformed life for all in the region.

By all reports, the bombardment of North Waziristan, and to a lesser extent South Waziristan, has become fast and furious since a combined Taliban and Qaeda suicide attack on a C.I.A. base in Khost, in southern Afghanistan, in late December.

On the legal side, responses trickle in to Harold Koh’s statement about drone warfare in his American Society of International Law address.  As I’ve indicated, as someone who had been highly critical of the long wait for the US government to offer a defense of its lawfulness, I’m very pleased with the statement.  For precisely the reasons I’m pleased, of course, numbers of others are not so pleased, as Ari Shapiro, who was at the speech, noted in his story for NPR.  For a good, even-handed discussion of the Legal Adviser’s statement, see this analysis by Anthony Dworkin of Crimes of War, including the several links provided at the end.

At first sight, Koh’s justification appears to be based on the idea, familiar from the Bush administration, that the United States is engaged in a worldwide armed conflict with al-Qaeda and the Taliban and can use lethal force against anyone fighting on the other side. However a closer reading of Koh’s remarks shows that his position is not so clear-cut. Although Koh clearly refers to an ongoing armed conflict, he also offers the broader notion of self-defence as an alternative justification. This is clear when he answers the possible objection that drone strikes away from a battlefield constitute unlawful extrajudicial killing. Not so, Koh replies—a state “that is engaged in an armed conflict or in legitimate self-defence is not required to provide targets with legal process before the state may use lethal force.”

The suggestion here is that a state that is the subject of sustained threat from an armed group may use lethal force when necessary to defend the lives of its citizens, even outside the context of a recognisable armed conflict. And furthermore that this right of self-defence extends not just to targeting those individuals engaged in an imminent attack against the United States, but those members of the armed group who are integral to the group’s broader campaign of violence against US citizens.

What is striking in Koh’s speech is that the existence of an armed conflict and the broader right of self-defence are both offered as possible justifications without any attempt to delineate the boundary between the two: it is not clear how far Koh is claiming that the purported armed conflict against al-Qaeda and the Taliban extends. Does it cover military actions in Pakistan? Somalia and Yemen? The ambiguity here is consistent with the continued lack of a definitive statement from the administration about the precise legal contours of its fight against al-Qaeda, in a way that is visible as regards detention policy as well.

Koh’s suggestion that drone strikes might be justifiable as self-defence even outside a recognisable armed conflict is in line with the position of earlier US administrations, as the legal scholar Ken Anderson recently argued in an article for the Weekly Standard. But what is missing in the administration’s justification for the drone attacks is any sense of what the limits are on the use of lethal force against individuals who do not pose an immediate threat. The restrictions in the laws of armed conflict, which concern only whether the target is engaged in hostilities against the United States and the degree of harm to other civilians, do not seem adequate here.

I should add that although I do take the view that non-international armed conflict is defined by the facts of where sustained, intense, hostilities are underway – which is one reason I think the applicable legal ground in some cases is self-defense – and in that sense “geographically” limited, it is important to recognize that many in law and policy in the United States, at least, do not accept that at all.  The Wall Street Journal was one of the few major newspapers that said much editorially or as news about the speech; its editorial is here, overall praising the Koh speech:

Count us among those applauding last week for the Obama Administration’s robust defense of the use of unmanned drones for targeted strikes against al Qaeda in Iraq, Afghanistan and elsewhere. In a speech to the American Society of International Law, State Department Legal Advisor Harold Koh presented a broad assertion of the U.S. right to pursue and kill terrorists overseas, on or off an active battlefield.  In laying out the legal argument for the strikes, the former Yale Law School dean was, to put it mildly, meticulous, even fastidious.

Pushback was found, among other places, in some commentary at Huffington Post, for example, by human rights lawyer Chris Rogers:

Koh failed to address serious concerns over the U.S.’s use of drones to kill al-Qaeda and Taliban militants, and in particular the debate over strikes in Pakistan and other areas outside Afghanistan. Hopefully Koh’s remarks indicate that a fuller account of the U.S.’s legal position is forthcoming. But for now, the program remains shrouded in secrecy and Koh’s mere assertions of the program’s legality fail to provide the kind of accountability that is urgently needed.

My own assessment of the fall-out is that there is a general sense of satisfaction within the intelligence community that Dean Koh specifically distinguished armed conflict from self defense, although many in it would have preferred to see a direct reference to the CIA and the lawfulness of its role.  Those who were already critical are no more satisfied – partly, I don’t think it is snarky to say, because the issue is not really about knowing more about it, but fundamental objection to it.  Knowing more about it isn’t really the issue for the ACLU or the human rights groups or various UN officials.  Ben Wittes is right in saying that for the international soft-law community, it remains the Next Big Thing.

Over at Opinio Juris, Julian Ku points to today’s front page NYT story by Charlie Savage detailing secret memos and meetings among the leading lawyers seeking to address national security and terrorism issues, including detentions, drone strikes, the question of armed conflict, and other things.  In my posts, I have been focused on the question of drones and targeted killing, but of course that issue and its legal authorities are intertwined with other questions, such as detention.  It’s a long piece by Savage, introduced by saying:

The discussions, which shaped classified court briefs filed this month, have centered on how broadly to define the types of terrorism suspects who may be detained without trials as wartime prisoners. The outcome of the yearlong debate could reverberate through national security policies, ranging from the number of people the United States ultimately detains to decisions about who may be lawfully selected for killing using drones.

(Note: I started composing draft posts about Harold Koh’s important ASIL speech, particularly its discussion of drones and targeted killing, and concluded it would be better not to do a post that would turn into an online article, but instead some shorter posts on particular issues, even if they are somewhat random.)

Toward the end of the section on drone warfare in Legal Adviser Koh’s March 25, 2010 speech to ASIL is a discussion that runs to US domestic law and regulation.  For the first time in a very long time – it might be since Koh’s predecessor Abe Sofaer addressed the question in a 1989 speech – the meaning of “assassination” in US law and regulation has been addressed in an authoritative and considered way by a senior US government lawyer.  Legal Adviser Koh said with respect to “assassination”:

[S]ome have argued that our targeting practices violate domestic law, in particular, the long-standing domestic ban on assassinations. But under domestic law, the use of lawful weapons systems—consistent with the applicable laws of war—for precision targeting of specific high-level belligerent leaders when acting in self-defense or during an armed conflict is not unlawful, and hence does not constitute “assassination.”

To refresh on the background to this.  Apart from questions of international law, the US has had a domestic ban on “assassination” in the form of an executive order that has been in place (or renewed in slightly different language) since the presidency of Gerald Ford.  In 1976, in the wake of revelations of CIA activities in the Church Committee hearings, President Ford issued EO 11905, the single relevant sentence of which read:  ”No employee of the United States Government shall engage in, or conspire to engage in, political assassination.”  President Carter reissued the order in 1978 in slightly different language in EO 12036, which read in relevant part:  ”No person employed by or acting on behalf of the United States Government shall engage in, or conspire to engage in, assassination.”  President Reagan reissued the order in 1981 using identical language in EO 12333:  ”No person employed by or acting on behalf of the United States Government shall engage in, or conspire to engage in, assassination.”  EO 12333 was amended by subsequent EO’s, but the specific assassination ban text remains unchanged.

That said, the term “assassination” is never defined.  Whole forests have fallen as commentators, in law reviews and elsewhere, have debated its meaning over decades, however.  Does it refer to political leaders?  To whom does it apply or not apply?  Non-state actors?  Terrorist groups?  Political leaders of states with which the United States is at war?  Military-political leaders of such states (given how frequently that is the case)?  There is little material in the record as to what was intended – and perhaps not surprisingly.  What little anecdotal information exists from the EO’s issuance in the 1970s suggests that it was intended as a way of placating Congress, and avoiding an actual statutory ban.  The EO was apparently intended to be vague and undefined, and subsequent presidents – and, note, Congresses – have found that to be a useful ambiguity in which to leave it.  It has the status of a binding executive order in domestic law – and amendable, alterable, and revocable should the President want to do it.

I’ve never understood, to be frank, the scholarly agonizing around a single sentence with a wholly undefined term in an order not, and never, codified as a statute.  It defies interpretive settlement, I would have thought, precisely because it was not designed to bear any real legal weight.  It was instead merely a declaration in vague and general terms that whatever wicked killings the CIA did that were revealed in the Church hearings, those would not happen any more, a mea culpa and promise not to do some ill-defined bad things any more.  Well and good – the CIA did some bad and wicked things – but beyond that, one is not really going to get by textual interpretation.  Unsurprisingly, much of the commentary uses the interpretive discussion as a way of launching a normative view of when and what should be acceptable in the way of targeted killing.

The essentially normative launching pad of which much of the commentary consists is reinforced by the observation that US officials did offer two senior commentaries on the meaning of the ban in the late 1980s.  One was offered by senior law of war Department of Defense lawyer Hays Parks in a 1989 memorandum of law.  The conclusion of that memorandum, with respect to the use of military forces of the Department of Defense:

Assassination constitutes an act of murder that is prohibited by international law and Executive Order 12333. The purpose of Executive Order 1233 and its predecessors was to preclude unilateral actions by individual agents or agencies against selected foreign public officials, and to establish beyond any doubt that the United States does not condone assassination as an instrument of national policy. Its intent was not to limit lawful self defense options against legitimate threats to the national security of the United States or individual U.S. citizens. Acting consistent with the Charter of the United Nations, a decision by the President to employ clandestine, low visibility or overt military force would not constitute assassination if U.S. military forces were employed against the combatant forces of another nation, a guerrilla force, or a terrorist or other organization whose actions pose a threat to the security of the United States.

This analysis (or at least this particular concluding passage) is not explicit as to whether the act in question is murder because it is assassination and therefore unlawful, or whether it is assassination because it is murder and therefore unlawful.  In 1989, then Legal Adviser to the DOS, Abraham Sofaer, made a speech at the US Army JAG school in Charlottesville, Virginia, that spoke to this underlying issue of what is the predicate, independent violation of law, murder as such or assassination on grounds independent of murder:

The meaning of the term “assassination”’ in historical context, and in the light of its usage in the laws of war, is, simply, any unlawful killing of particular individuals for political purposes … virtually all available definitions of “assassination”’ include the word “murder,”’ which in law is a word of art. Murder is a crime, the most serious form of criminal homicide. That element is the most fundamental aspect of the assassination prohibition. All criminal killing is therefore potentially subject to the prohibition.

Under no circumstances, however, should assassination be defined to include any lawful homicide. Assassination is also commonly defined as killing with a political purpose. Murders that have no political purpose or context are criminal and remain subject to punishment, but these too should not be characterized as assassinations. Other elements offered in available definitions seem superfluous or even misleading. Thus, for example, whether a killing is done “secretly”’ or “treacherously”’ and whether the person is “prominent”’ would appear to be of little or no consequence for purposes of the Executive Order. Nor should it matter that the assassin “kills in the belief that he is acting in his own private or public interest”’ or whether the action is “surprising”’ or “secret.”  The pivotal elements in terms of controlling the behavior of government officials would seem to be illegality and political purpose … the historical background of the term casts considerable light on its meaning and strongly supports a definition limited to illegal, politically motivated killing.

The assassination ban, on this reading, argues that in order to be an assassination, the killing must already be unlawful on some independent ground – because it is murder, for example, or cannot be justified under international law of self-defense.  The result is that the ban on assassination becomes coincident with killing that would be unlawful under domestic or international law in any case.  The assassination ban does not include lawful acts of self-defense.

Consider again the Koh speech on this topic.  It too emphasizes that the ban is to be understood as not including lawful self-defense:

[U]nder domestic law, the use of lawful weapons systems—consistent with the applicable laws of war—for precision targeting of specific high-level belligerent leaders when acting in self-defense or during an armed conflict is not unlawful, and hence does not constitute “assassination.” (Emphasis added.)

This is one of several places in which the Koh address specifically distinguishes “self-defense” from “armed conflict.”  It expresses a view that when acting in lawful self-defense (which might or might not be “armed conflict” in a specifically legal sense of that term), targeting specific high-level belligerent leaders is not (independently) unlawful, and because it is not, it thus does not constitute “assassination.”

This is very significant, for at least two reasons.  One is that this reasoning adopts, as premise about international law, that there is a distinction between self-defense and armed conflict.  This is not the only place that the speech does so.  It specifically holds out that the possibility that there can be acts of self-defense separate from acts of armed conflict (I would say, though this particular passage does not, that lawful armed conflict is a subset of lawful self-defense).

Second, more centrally:  It clarifies, by essentially restating and reaffirming a legal view of the US government. one not re-stated (at least not plainly or very often, if at all, so far as I can tell) on both self-defense in international law and assassination in domestic law, dating back to the 1980s.  If the importance of the implication of the international law premise (the self-defense premise) is clear, however, why do I suggest that the domestic law interpretation of the assassination ban is also so significant? After all, it’s merely an executive order, alterable by a president at any time.

The reason I place such significance on this domestic law interpretation is practical.  Outsiders to government like me do not have much information on the thinking during the last two decades about targeted killing and assassination, including 1990s arguments over targeting Bin Laden and others.  Certainly I am not privy to any special information.  However, there is a reference to an OLC opinion from the 1990s, referenced in the 9/11 Commission report, about concerns that in order to avoid possible violation of the assassination ban, it was necessary to conclude that Bin Laden was a combatant, and therefore a target.  Otherwise – quite apart from any international law considerations – targeting him might contravene the assassination ban.

This led, so far as an outsider like me can tell, to push the US government, as it became aware of the Al Qaeda threat in the 1990s, to regard its actions as having to satisfy the requirements of combatancy, and not to assert any genuinely independent grounds of self-defense as a basis for targeting.  As I have argued at great length elsewhere, the collapse of the use of force into a binary consisting of law enforcement or armed conflict deprived the United States of recourse to the most useful and most obvious category of use of force – self-defense at minimal, targeted, discrete levels of violence that would not constitute a genuine legal armed conflict, using covert “intelligence” actors (or at least to accept constraints arising from a legal characterization of such intelligence uses of force that had to be based around the fundamentally armed conflict framing of combatancy).  Not an armed conflict because it would not be, paradoxically, violent, or intense, or sustained enough.  Which is, after all, what covert action is supposed to be, when successfully carried out as a use of force.

I believe that the precautionary legal view that the categories had to be either law enforcement or full-on armed conflict had an inhibiting effect on the effective use of force against Al Qaeda in the 1990s.  I can’t prove it or corroborate it; I’m not an insider in any sense.  However, I was always struck by a conversation with someone who was an insider in the 1990s; when I asked, sometime around 2004, why the insistence on establishing that someone was a “combatant” for purposes of targeting, the answer was immediate – “Oh, well, if we didn’t, we might be in violation of the assassination ban.”  No reference to a violation of international law – presumably because there was no ICC or courts of Spain or Baltasar Garzon in the mid-1990s – but a very real concern about US domestic law.

I won’t try to re-explain here why in my view it matters whether one appeals to combatancy as the legal frame, in an armed conflict, or the broader category of self-defense; it is discussed at length in this article, and for that matter I have raised it on Volokh and Opinio Juris (and the international law scholar Marko Milanovic has gravely disputed the self-defense argument at EJILTalk! blog).  Given the complete acceptance among the three branches of government that the United States is in an armed conflict with Al Qaeda – so that concepts of combatancy do apply – it would have been very easy merely to say, combatants in armed conflict can be targeted under international law, and targeting them does not violate the assassination ban.  Assuming that one takes a global view of the armed conflict (which, I should say, I do not, believing, as I do, that armed conflict with a non-state actor is limited to places of active, sustained, and intense hostilities, but leave that discussion for another post), one need not get into questions of self defense as an independent and broader category of the use of force.

The fact that this speech goes beyond armed conflict to the broader category of self defense speaks to the care that went into preparing this speech – with a view in mind of the needs of future presidents, facing future threats, and responsible for the safety of the United States of America in circumstances having nothing to do with Al Qaeda.  The Legal Adviser might have stopped merely with armed conflict, taking a narrow and presentist view that all that matters is Al Qaeda today, and not presidents tomorrow.  That he did not speaks highly of his long view of the presidency.  Because the law of self defense, while relevant today particularly to the use of force by the CIA, is most relevant to presidents stretching into the future.

But returning, finally, to assassination.  Yes, in my view, it matters a lot in a practical sense that officials of the CIA and other agencies are able to go about their business with considerably less concern that, quite apart from anything else, the one sentence concerning assassination in the EO might come back to haunt them as a matter of domestic law.  It’s not the end of the matter in a lawyerly sense, of course; memos need to be written, legal opinions given, and while I’m doubtful that the text of the EO would be modified to reflect this explicitly, that would be a good thing (it is possible that all of this has already happened inside government).  A single paragraph in a public speech does not take care of things all by itself.

Still, I think the Legal Adviser’s interpretation is correct, and I think it a matter of considerable significance that it so clearly draws upon a long-standing view within the US government, but one that had fallen (if this is not too strong a characterization) into a certain desuetude, at least to those of us outside of government.  Outsiders like me – and most readers – are not fully able to appreciate just how much work, digging, and evaluation of non-public legal memos, opinions, and reasoning dating back decades must necessarily go into what appears to be a simple conclusion in a couple of sentences.  The work behind it was, I’m certain, very substantial.

After getting a number of email requests for this testimony, I’ve posted it up for convenience at SSRN, although it is also available through the hearing page as well.  I submitted along with it my earlier Weekly Standard article, Predators Over Pakistan.  As I have said here earlier, however, this hearing on March 23 came just ahead of Legal Adviser Harold Koh’s March 25, 2010 ASIL speech.  Dean Koh’s speech addressed many of the criticisms made of administration policy on drones, and I should add, in ways that I thought substantively correct as a matter of legal policy.  I thought the Legal Adviser’s statement strong, clear, and substantively excellent.  Overall, I was pleased with the form and direction of its legal framework (if you look at the second half of the Predators Over Pakistan article, the more legally oriented part, you will see that my analysis tracks certain parts of the Koh speech analysis fairly closely).   I was also impressed with its directness – it was not filled with careful dodges and lawyerly hedges that would make it impossible to find a clear takeaway, quite the contrary.

There are many open questions, of course, and ways in which interpretations and legal judgments could go – but the statement on drones itself was impressively direct in what it chose to address.  So I strongly recommend reading this testimony, if you plan to, with the Legal Adviser’s text to hand.  (Below the fold is the SSRN abstract for my testimony.)

Other articles worth reading on the Koh speech?  Shane Harris’s quick take at the Atlantic is a useful one, likewise Ari Shapiro’s take at NPR (including his sense of the audience reaction at the ASIL meeting, which I was not able to attend), and I will post links to some other articles, including Peter Finn’s upcoming Washington Post piece.  In the academic blogosphere, reaction was not so enthused – although my Opinio Juris colleague Julian Ku was pleased that the Koh speech embraced the Bush rationale of armed conflict in addressing Al Qaeda terrorism, my other Opinio Juris colleague Kevin Jon Heller was not so enthused (although KJH was actually commenting on another part of the Koh address, on the ICC and the crime of aggression).  Diane Marie Amann was also not enthused.

In reading the entire speech – which ran over an hour to a packed hotel ballroom – and not just the parts related to drones, I was struck by a general, and in my view deeply important and (especially by academics like me who have never worked in government) underappreciated, question of lawyerly method, sources, methods of interpretation and grounds of authority.  Koh made this point by quoting from Walter Dellinger on his experience at OLC:

[U]nlike an academic lawyer, an executive branch attorney may have an obligation to work within a tradition of reasoned, executive branch precedent, memorialized in formal written opinions. Lawyers in the executive branch have thought and written for decades about the President’s legal authority…When lawyers who are now [in my office] begin to research an issue, they are not expected to turn to what I might have written or said in a floor discussion at a law professors’ convention. They are expected to look to the previous opinions of the Attorneys General and of heads of this office to develop and refine the executive branch’s legal positions.

I think this is important as a question of method and authority in international law as expressed by the United States as its binding obligation.  When the Legal Adviser’s office sat down to come to a view (and with, drawing on an earlier section of the speech, the extraordinarily difficult and complex task of coordinating with other agencies of government and their lawyers), in other words, it presumably looked to its own practices, opinions, internal conclusions and advice to the executive branch, and so on.  Not to outside scholarship, or at least presumably not until well down the path of interpretive authority within the internal structure of interpretation and legitimate legal authorities within the context of US government lawyering.

Like most academics, particularly when dealing with an administration with a sizable number of lawyers who come from academic communities such as my own, I certainly want to believe that Everyone Who Is Anyone is reading … my stuff.  But of course it’s not true and, more importantly, it shouldn’t be true.  Not except as very, very tertiary.  Why?  Because authority in law is partly a question of the logic, compelling or not, of the ideas and argument expressed – but its legitimacy is also partly a question of its provenance, particularly in the law of a democratic government, in which the legitimate authority of law is partly about internal logical consistency but also partly about consent of the governed.  Who said it also matters.

Continue reading ‘Drone Warfare Subcommittee Testimony, Up at SSRN’ »

Update:  I have had a chance to watch the video twice – I strongly recommend watching it, as it adds considerable language to the statements below in the press release.  And welcome Instapundit readers and any others.  Given how much I have pressed publicly for a statement by the administration’s lawyers, I want to say this much even while I’m still doing a careful lawyerly parsing of the text.

First, let me praise Harold Koh for stepping up to the plate.  This is a plain, clear statement of the US view of the law and its application.  It is measured, and yet exceedingly direct.  My thanks and congratulations to the Legal Adviser for something that stands as clear opinio juris of the United States.  As someone who has been calling every more sharply for a public statement by the administration’s lawyers on targeted killing and drone warfare – most recently in a Weekly Standard article on exactly that theme, again this week in a sharply worded statement to a House subcommittee hearing, and an Ari Shapiro interview on NPR this morning recorded a a month ago – this was an enormously positive step.

Second, on the substance.  On first read, I think this is a great statement.  It addresses an armed conflict with Al Qaeda, the Taliban, and associated forces.  But it also asserts self-defense several times as an alternative.  I had been greatly concerned that the administration’s lawyers would narrowly confine the justification for targeted killing using drones to situations that would really only cover the military using them on active battlefields. But on first read, this statement does not do that at all. It appears to address situations of safe havens, for example, and indeed reasserts the traditional US view – that sovereignty and territorial integrity are important, but the US preserves its rights to go after its enemies in their safe havens.

I want a chance to go over the written text and say something much more exact.  But given how much, particularly this week on account of the Congressional testimony, my criticisms of the administration’s lawyers have been in the news, I would like to make it known as publicly as possible that, on first read, the Legal Adviser’s statement on targeted killing and drone technology is very positive, very strong, and admirably forthright.  My congratulations and thanks to the Legal Adviser.

*

I was teaching classes and unable to attend, much as I wanted to, DOS Legal Adviser Harold Koh’s keynote address at the American Society of International Law annual meeting tonight.  Shane Harris sent me along the press release from ASIL with these excerpts from Dean Koh’s address specifically on the topic of drones.  I have barely had a chance to skim this and have not seen the full text, so I will refrain from commenting at all until I’ve had a chance to read this through.  But I wanted to post this immediately for interested readers; I will now read it through and watch the video):

Koh’s wide-ranging remarks on the Obama administration’s international law policies included a specific affirmation of the administration’s approach to the use of force, including the use of unmanned aerial vehicles (UAVs), which has recently come into question by some legal experts.  Among Koh’s statements regarding the use of UAVs are the following.

“…[I]t is the considered view of this administration…that targeting practices, including lethal operations conducted with the use of unmanned aerial vehicles (UAVs), comply with all applicable law, including the laws of war….As recent events have shown, Al Qaeda has not abandoned its intent to attack the United States, and indeed continues to attack us.  Thus, in this ongoing armed conflict, the United States has the authority under international law, and the responsibility to its citizens, to use force, including lethal force, to defend itself, including by targeting persons such as high-level al Qaeda leaders who are planning attacks….[T]his administration has carefully reviewed the rules governing targeting operations to ensure that these operations are conducted consistently with law of war principles, including:

-                             First, the principle of distinction, which requires that attacks be limited to military objectives and that civilians or civilian objects shall not be the object of the attack; and

-                             Second, the principle of proportionality, which prohibits attacks that may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, that would be excessive in relation to the concrete and direct military advantage anticipated.

In U.S. operations against al Qaeda and its associated forces – including lethal operations conducted with the use of unmanned aerial vehicles – great care is taken to adhere to these principles in both planning and execution, to ensure that only legitimate objectives are targeted and that collateral damage is kept to a minimum.”

Responding to some recent arguments made against the use of UAVs, Koh defended the administration’s policy saying:

“[S]ome have suggested that the very use of targeting a particular leader of an enemy force in an armed conflict must violate the laws of war.  But individuals who are part of such an armed group are belligerent and, therefore, lawful targets under international law….[S]ome have challenged the very use of advanced weapons systems, such as unmanned aerial vehicles, for lethal operations.  But the rules that govern targeting do not turn on the type of weapon system involved, and there is no prohibition under the laws of war on the use of technologically advanced weapons systems in armed conflict – such as pilotless aircraft or so-called smart bombs – so long as they are employed in conformity with applicable laws of war….[S]ome have argued that the use of lethal force against specific individuals fails to provide adequate process and thus constitutes unlawful extrajudicial killing.  But a state that is engaged in armed conflict or in legitimate self-defense is not required to provide targets with legal process before the state may use lethal force.  Our procedures and practices for identifying lawful targets are extremely robust, and advanced technologies have helped to make our targeting even more precise.  In my experience, the principles of distinction and proportionality that the United States applies are not just recited at meeting.  They are implemented rigorously throughout the planning and execution of lethal operations to ensure that such operations are conducted in accordance with all applicable law….Fourth and finally, some have argued that our targeting practices violate domestic law, in particular, the long-standing domestic ban on assassinations.  But under domestic law, the use of lawful weapons systems – consistent with the applicable laws of wear – for precision targeting of specific high-level belligerent leaders when acting in self-defense or during an armed conflict is not unlawful, and hence does not constitute ‘assassination.’”

To view video of the relevant portion of Koh’s address to the Society, visithttp://fora.tv/v/10561.

Yesterday (Tuesday, March 23, 2010) I was pleased to appear as a witness before the House Subcommittee on National Security and Foreign Affairs, to discuss drone warfare and Unmanned Aerial Vehicles (UAVs).  The list of witnesses is here, along with a link to a webcast (which I can’t seem to open), and links to the witness statements.  I am always fascinated to listen to Brooking’s Peter Singer, author of Wired for War, and the rest of the panel was equally interesting, especially on the technical aspects of UAVs.

I was particularly interested to hear in this session discussion of the many ways in which UAV technology is entering civilian fields – everything from crop dusting to monitoring schools of fish from the sky.  Someone mentioned (but I haven’t verified) that University of Nevada now offers a degree program in UAV control.   My points were fairly removed from much of the rest of the technological or military discussion, being aimed squarely at policy and legal policy concerns.  In my oral statement, I emphasized the following:

• The Obama administration has embraced drone warfare, but its lawyers have not caught up; presumably they have private concluded that the several drone warfare programs underway are legal, but it is not a good idea to keep the rationales secret.  Not, at least, with a program in Pakistan that is only a matter of official denial, not secret.  It leaves the President and operations people on down hanging out there exposed.
• The drone warfare campaign embraced most thoroughly as a strategic matter – correctly, in my view – by the senior administration officials starting with the President is not the “hot battlefield” use of drones, in which they are essentially a substitute air support weapon for a manned system.  It is, instead, the use of drones in a role unique to them, as a targeted killing mechanism for use in places far off of active battlefields.  There are some questions raised about military battlefield use of drones, but the real questions lie off the active hot battlefield – and it is these uses that are the real reason why senior Obama officials have celebrated them.
• The fundamental issue in a legal policy sense, then, is not so much drones as such – it is their use in a campaign by the CIA of targeted killing.  A better way to understand the drone debate is that this emerging technology has forced onto the table discussion of the lawful and appropriate role of the CIA, in the use of force, and in covert operations using force.  Much of the purely military discussion is interesting technically, but legally somewhat beside the point at least at this point.  A millimeter beneath the surface of the drone discussion is the question of the CIA.  This is the point of the attack on the CIA’s role as criminal made by Mary Ellen O’Connell, and also by Gary Solis in his characterization of America’s “own unlawful combatants.”
• CIA director Leon Panetta has been engaged in a vigorous public defense of the drone program of targeted killing by the CIA.  I am all in favor of it, of course – but as a friend of the program, so to speak, and happy as I am to see Panetta speak up, his defense of the CIA program is slightly beside the point.  He defends the success of the program. Leaving aside criticisms of “blowback” – easy to make, hard to prove or disprove, frankly, but leave that aside – for most of us the success of the program is not the issue.  At issue is its legality, not its success.  Or at least, to know on what basis the lawyers think it is legal.  (Which, note to the Al Jazeera interviewer yesterday following the hearing, I took great pains to distinguish – I do not question the legality of the program as such, I question which legal rationale for its legality is the right one).
• Even you believe targeted killing using drones is, as a general proposition, legal under domestic and international law, as I do, it matters what kind of legal rationale one asserts for it.  Not all lawyers would agree, to be sure, and I suspect this includes lawyers inside the administration, but in my view it matters whether one adopts an “armed conflict” justification in which, when the US targets someone in Pakistan or Yemen or Somalia or wherever, that person is determined to be a lawful target – and it doesn’t matter, for the bare legality of targeting, where he is located.  The armed conflict goes where the combatant goes.  Mumbai, Mogadishu, wherever.  I think this is wrong as a question of the laws of armed conflict, because it does not adhere to the threshold requirements of this peculiar non-international conflict with a non-state actor taking place on a global basis that the US has with Al Qaeda.  In my view, the proper legal frame is international law of self-defense – and it is what the US has traditionally viewed the exercise of these discrete uses of force by the CIA, covert or clandestine, as being anyway.
• These two legal rationales eventually lead to different legal conclusions, constraints and authority for the use of force, unless one uses very awkward legal reasoning to force these very different uses of force altogether into the same frame.  My view is that forcing CIA targeted killing in places that might range widely in the world into armed conflict rationales is bad for the CIA’s legal reasoning – and requires ever greater legal contortions to make it fit into a model of armed conflict law that is fundamentally about the overt meeting of armies and state forces, not covert ops and intelligence services using force.  I don’t think it is a stable legal view for the CIA over the long term.  Moreover, as I thought in analyzing the Bush administration’s “global war on terror” legal frame, forcing all uses of force outside of plain law enforcement into the armed conflict box of law for overt war distorts that body of law and its interpretation, too.  It isn’t about regulating covert ops and intelligence services – there was a good reason why, during all those decades before the 1990s, exceptional and extraordinary uses of force by the covert civilian services was regarded as not unlawful under international law, but not part of the law of armed conflict, either.

Part of the “framing” problem of law here is that it is not just a division of the military operations from the CIA ones, overt from covert.  The Pakistan program by the CIA is, obviously, not covert – it is an operation in formal deniability – the ability to say at least officially (US and Pakistani officials), US military forces are not operating on the ground in Pakistan.  But, as Peter Singer pointed out in yesterday’s hearing, the scale of this non-covert program is up at the level of the opening of the Kosovo air campaign in 1999 – it is up there at the level of “real” war.  It might well be that the proper legal characterization of the CIA’s role in that theater of hostilities is not “self defense” operations as I would use to characterize the CIA operations in other places, smaller, more discrete, and perhaps genuinely covert or closer to it.  Quite possibly operations in Pakistan by the CIA have to be treated as part of the armed conflict.

The point is that a “deniability operation” (if it can be called that, to distinguish from truly covert) on a large scale eventually starts to resemble “hostilities” in a legal sense and in this case eventually merges geographically into the classic hostilities situation of non-international armed conflict that spills over a border.  If that is the case, then quite possibly it should be characterized as part of a single AfPak theatre of hostilities.  With, therefore, an armed conflict rationale for activities by the CIA.

But I do not think this rationale will work for the operation that matter into the future as Islamist terror groups morph and change and seek safe haven in other places.  Nor do I think it will work for post-Al Qaeda groups.  For those, I would opt for self-defense as Abe Sofaer characterized it in 1989.  And, to be clear, if that is the rationale, it does have to remain – as Gabriella Blum noted in a Harvard symposium on this a few weeks ago – exceptional measures.

So I think it will turn out to matter quite a lot over the long term that the Obama lawyers eventually opt for the specific law of armed conflict, including its definitions and thresholds under treaty and customary law, or some customary law of self-defense.  My guess is that the administration’s lawyers will finally opt for what appears to them to be both a narrower and a superficially reassuring ground of armed conflict that allows them to say that everyone being targeted is a “combatant” in a legal sense.  Again, my own view is that this won’t work over the long run – it will degrade as a rationale for the CIA under withering attack from the soft-law community, and it will have a tendency to degrade the law of war itself as rationales suited to a body of rules about large scale, or at least overt, conflict modeled on inter-state armed conflict, continues to pressed into service in an area that is mostly about covert or clandestine operations, and which aims to avoid overt war.

Put more bluntly, as a political evaluation, I think this rationale will work so long, but just so long, as there is an Obama administration.  As a legal rationale, armed conflict as a rationale will turn out to be a legally and politically “wasting asset.”  The legal attacks on targeted killing and the use of drones therein will be developing, but won’t ripen into active claims of war crimes, and so on, at least ones that have teeth, until the Obama administration leaves office.  Like many of the Bush-era legal tools in counterterrorism, my guess is that the Obama administration will rely on them in point of fact – but not legally defend them in ways that leave them plainly and cleanly available to the next president or future administrations.

Jonathan Manes, the ACLU lawyer whose work and presentations I have cited a couple of times in posts here at Volokh, and someone for whom, despite some fundamental agreements, I have great respect and like very much, has responded with the following letter.  He has given me permission to post it here.  The most important issue is the question of whether I have correctly represented his or the ACLU’s position on the question whether a problem with drones is that they make resort to force by the US too easy.

I don’t believe that it’s proper for me to editorialize on my views here, when the post is really to allow a “letter to the editor,” so I will simply post the letter below; I’m also not opening it to comments. I encourage you to read the FOIA request that Jonathan drafted on behalf of the ACLU (as well as review, if you are very ambitious, the panel discussion at Harvard – although I agree entirely (as I remarked with a similarly impromptu discussion by Harold Koh) that one should not hold people to the strict words of an impromptu panel discussion, and if Jonathan feels that I’ve played “gotcha” unfairly, then my apologies). (I am stealing time in a meeting – one of the deans will shortly glare at me, with good reason – and I will have to provide a couple of background links later, including a couple in Jonathan’s letter that I can’t reproduce at this moment.)

One thing I should add, though, is that whatever Jonathan and I do disagree about, we agree altogether on the need of the US government to offer a legal rationale for the practice.  As I said in my last post, CIA director Panetta’s assertion that the program is legal does not tell us enough to form a view.

Okay, with that, my thanks to Jonathan for taking the time to respond in so thoughtful a fashion and so advance this discussion.  (Dated March 22, 2010.)

Dear Ken,

I very much enjoyed meeting you and speaking with you at the conference last week. In case you hadn’t seen it, I thought I’d pass along this recent news item, which quotes Harold Koh saying he has conducted a legal review of the drone program and that the administration is likely to disclose its legal rationale.  In light of our conversation, I think we’ll both agree that this is a very positive development!

Also, this past weekend I came across the item you posted on Volokh debriefing the HLS conference. I also saw the recent posting about your upcoming Subcommittee testimony. Both of these postings were engaging, as usual. But in both cases you unfortunately misinterpret part of my HLS presentation and the ACLU’s position, prompting this reply, which you should feel free to post on Volokh or elsewhere.

You’re right that at the conference I made the point that drones and other remote-controlled technologies make it easier to project force anywhere in the world. As a factual matter, this seems clearly to be true. As was discussed at the conference, drones are cheap to produce, easy to use (requiring only a few months of training), and less costly to deploy than ordinary weapons systems. But the point of highlighting that fact was emphatically not to suggest that keeping troops out of harm’s way is a bad thing, or that troops should be put in danger in order to “restrain” the U.S. I didn’t make that point, and you’re incorrect to attribute it to me or to the ACLU.

So, why mention the fact that drones are easier to deploy than other weapons? The point of my presentation was transparency, and I was making the case for why basic information about the drone program should be disclosed. Remote-controlled technologies allow the U.S. to wage war without putting troops in harm’s way. Keeping troops safe is clearly a good thing. But when force is deployed primarily by remote-control, people might be less motivated, by their affinity and concern for soldiers at risk, to scrutinize what the government is doing. As such, when it comes to drone warfare, it is especially crucial to be vigilant in demanding transparency, so that the public knows what the government is doing in its name abroad.

The ACLU, the organization that I work for and represent as a lawyer, has filed a FOIA request and lawsuit in order to serve this democratic function. The ACLU is trying to facilitate the kind of public debate and discussion that is necessary and appropriate in a democracy when it comes to using lethal military force abroad. Targeting and killing specific people who are off the battlefield also raises significant civil liberties concerns – especially when the targets can include U.S. citizens, as the government has acknowledged. The public therefore has a right to the kind of basic information we’re seeking in this FOIA: the legal rationale and limits on the program, the internal oversight that ensures these methods are not abused, and the results of the program, in terms of civilian and non-civilians killed. Disclosure of this kind of information is important to satisfy the public that its government is conducting the drone campaign in a manner that is consistent with U.S. law, values, and interests. (It is also not the kind of information that should jeopardize national security if released – if it does, the government will surely withhold the documents and let us know why they can’t be released.)

That, in short, is the rationale for the ACLU’s efforts on the drone issue. As such, it is simply not the case that “the ACLU has decided to campaign on the issue of jus ad bellum, whether the US should be resorting to force and to what extent.” And let me be clear: the ACLU does not believe – nor did I express – the view that “the US needs to be restrained — through direct and personal exposure to death on the part of its soldiers — in order that it have the proper incentives not to over-resort to the use of force.” Perhaps other groups make this argument. But not the ACLU.

I appreciate that in discussing complicated matters like these, it is easy to be misunderstood or to misread a person’s meaning. But because you came away with the wrong impression from my talk, and because in your recent post you again misattributed this argument to me and the ACLU, it’s important that the record be corrected.

Having said all that, I concur with you that we agree on at least two things — i.e., the need for the administration to disclose the legal rationale for the drone program, and that there are geographical limits on the IHL justification for using drone strikes. But in light of what I’ve said above, it should be clear that we agree on more than just the two points you mentioned. Like you, I deplore the loss of soldiers’ lives in combat. I have immense respect for those who choose to serve in the armed forces. Some of the sharpest and most honorable people I have met are current or former servicemembers, including among them Tad Oelstrom and Missy Cummings, our fellow presenters at the HLS conference. As such, the suggestion that I think that U.S. lives should be put at risk unnecessarily – in order to create “the proper incentives not to over-resort to the use of force” or otherwise – is repugnant. Likewise, I imagine that we share a strong concern that the people on the receiving end of U.S. weapons not be civilians, and that the military do everything it can to minimize the loss of innocent life.

In any case, I, too, thought that the conference was extremely interesting and useful. The combination of speakers and topics was very good. I am grateful to the Harvard National Security Journal for putting it on and inviting me to attend. I also enjoyed meeting and chatting with you. And I am hopeful that the next time we meet, we’ll find at least a few more things to agree on!

Best wishes,

Jonathan

–

Jonathan Manes

Legal Fellow | National Security Project

American Civil Liberties Union Foundation

In the midst of the healthcare debate in Congress, other business does go on (sort of), including a subcommittee hearing on drone warfare, at which I’ll be testifying, along with some other experts, including the author of the seminal Wired for War, which put the issue of robotics and war squarely on the table.  I’ll post more later about my written testimony and about the process, but late tonight I’m preparing what I’m told can’t exceed four minutes of oral presentation.  Hmm.  What points can I make in four minutes?  Regular readers of my posts on the blog will not doubt think – none.  Well, subject to revision, but so far:

• CIA director Panetta has been conducting a visible, on the front pages, PR campaign to argue that the Program That Cannot Be Acknowledged in Pakistan (and elsewhere) is every bit as successful as administration officials, from the President on down, have said.  That’s great, but somewhat beside the point.  Most of us are convinced that it is successful; the question is whether, and on what basis, the program is legal.  On that, the administration says, it’s legal but gives no clue on what basis it thinks that.  It’s gotta step up to the plate and declare itself.
• The issue on the surface is drone warfare.  But in fact, drone warfare is a set of heterogeneous activities, conducted sometimes by the military as tactical air support, and sometimes, the other extreme, CIA strikes in far away places.  The use of drones by the military on the conventional battlefield is not really very controversial, not at bottom – it is just another standoff firing platform.  The real question on the table is the role of the CIA in the use of force.  Drones can be thought of as less the issue themselves, than the technological issue that forces the role of the CIA in the use of force onto the table.  This debate is fundamentally about the CIA.
• Defending the legality of drone warfare gets you to different places depending on what legal rationale you think fits the case.  Going down the road of armed conflict will finally rule the CIA out of it, as well as limit the geographical reach of the use of drones, whether by the CIA or the military.  That is an important reason – given that the Obama presidency is foursquare behind drone strikes in places well beyond the active battlefields of Afghanistan – why the administration’s lawyers should not be giving up the US’s traditional, longstanding position that it is legitimate to use force off the hot, active, traditional battlefields in pursuit of terrorists – but that the appropriate way to understand this as a matter of law is self-defense, rather than the narrower law of armed conflict.
• Folks who buy into the currently Jane Mayeresque tropes about drone warfare removing the constraints of geography, the “natural barriers” that US forces would otherwise have to fight their way through, as the ACLU’s lawyer put it at Harvard recently, and thereby reducing the US disincentive to use violence, had better be really clear what they are saying.  Because what they are saying is that the US will have easy incentives to reach to the drone as a weapon because its servicemen and women are not sufficiently at personal risk to deter the US from using force.  That’s what you mean when you talk about “removing natural barriers” or reducing the American disincentives to violence.  You mean that not enough US personnel are at risk of getting killed.  I’d be interested to meet the US Congressperson who’d like to stand up and make that argument.

Well, this needs work, but it’s 1:00 am.

I’m writing from Cambridge, where I just finished a great short conference on targeted killing and drone warfare put on by the Harvard Law School National Security Law Society and Journal. The  presentations should go up as video one of these days I hope!) and I highly recommend them if you are studying this topic. Particularly for us lawyer types, the presentations by engineering professor Missy Cummings and Kennedy School professor Tad Oelstrom – both former military aviators, among many other things – was highly informative about the real world of UAVs in military use. My thanks to all the students who put the conference together – we had a lovely dinner afterwards, and I enjoyed the conversation with them a great deal. My thanks to the law school as well for funding it – times are tough, even at places like Harvard – and I appreciate the dean and the school’s willingness to continue hosting events that allow for important legal and policy discussions to go forward.  (Delayed a couple of days posting this – travel and then internet problems at home.)

I tried to use my time at the conference to think through some of the questions I’ve been raising both in my first published piece on this subject, Targeted Killing in US Counterinsurgency Strategy and Law, and in the long opinion essay in last week’s Weekly Standard, Predators over Pakistan. I’m going to do a series of posts extending some of those discussions.

Probably the most important conclusion for me from discussions following my Predators over Pakistan essay is that I have not been sufficiently clear, with myself or in writing, that the appeal I am making to self-defense law does not preclude the application of the normal laws of war in those places where there is an armed conflict. Discussion of this topic seems a little bit like the blind men and the elephant – the military people responsible for a counterinsurgency ground war in Afghanistan see Predator strikes in their theatre of conflict, quite rightly, as not a big issue. It is not different, really, from the missile fired by a jet 25 or 30 miles away – it’s just another standoff platform. The legal rules of targeting are no different, and it’s just another standoff firing option.

At the other extreme, however, is the CIA using Predators to attack a targeted designated by the President under procedures outlined in US statutes for covert action by the CIA. Is that different, legally? US forces are in an armed conflict with Al Qaeda, and members of AQ, combatants if we find them in Afghanistan, flee to Somalia – so we chase them down there and fire at them with Predators run by the CIA. Why isn’t that just more of the “armed conflict”?   Same rules apply – geography is not really an issue.

I have argued that geography matters, not for its own sake, but rather more precisely, because in establishing whether you have an armed conflict to which the laws of war apply – displacing other regimes of law – you have to meet a threshold of active, sustained, etc., etc. hostilities and fighting (I’m not looking to be strictly precise here) if, as the US has said, particularly in the SCOTUS analysis, it is a non-international armed conflict with a non-state actor. I don’t think SCOTUS got it right. Moreover, it led to the unintended consequences of result-oriented jurisprudence, the Court seeking to find a basis on which to apply the standards of CA3, and leading the Court to conclude that a provision envisioned entirely (read Pictet) about internal civil war could apply by ignoring the territoriality phrase in CA3 to focus instead solely on the non-state actor part. I think the Court should simply have said that CA3 standards apply as some universal baseline standard irrespective of armed conflict, and it should left aside how it intended to get there, but it didn’t. It was not well-briefed on the issue, and waded into deep waters about the law of armed conflict with implications for many other important parties in the world.

Viz., it appears that territoriality is an issue after all. I don’t think the Bush administration got it right by defining the conflict as the “global” war on terror – it is one matter on which I agree with the critics, whether academics such as Mary Ellen O’Connell or organizations such as the ICRC. Nor can the AQ case be assimilated to WWII and chasing combatants around wherever they go – if you take the treaty language seriously, and the customary law intended by the United States and other states when drafting CA3 to keep the threshold of a non-international armed conflict high, then, yes, it matters legally that WWII was an international armed conflict and the AQ war a non-international one with a non-state actor. It seems anomalous, because despite the non-international language, we are following them around the world, but we have legally decided to focus exclusively on the non-state actor part, in order to apply CA3. If that is the case, then the customary law threshold matters. Continue reading ‘Drone Warfare and the Harvard National Security Conference’ »

Well, “polemic” is probably closer to it.  Regular readers have been hearing about this piece for a while, and I have posted various arguments from it (concerning targeted killing and Predator drones and the CIA and armed conflict and self-defense, and my general concern that the Obama administration has embraced a policy that its lawyers have not so far stood up publicly to defend as lawful against its gradually emerging critics in the international “soft law” community) here at Volokh and over at Opinio Juris.  I will post a couple of comments on the piece later, including of couple of things I wish I had clarified or said differently (and open it up for comments then).  Meanwhile, if you are interested, it is the cover in this week’s Weekly Standard (March 8, 2010).  It is also very, very long, at some 8,000 words – for which I am deeply grateful to the WS’s editors but you perhaps will not be – and so you might find it easier to read a pdf of the print edition at SSRN.

Harold Koh, Legal Adviser to the State Department, held an informal public discussion a couple of days ago with his predecessor from the Bush administration, John Bellinger.  This was an American Society of International Law event, held at John’s law firm, Arnold & Porter, and moderated by my old friend and ASIL’s Treasurer, Nancy Perkins.  CSPAN covered it, and the video is now available:  The Obama Administration and International Law, February 17, 2010.  (If I can find a youtube version from ASIL, I’ll see if I can embed it.)

I was teaching and so could not attend in person, but I have now watched the video and it is a terrific event.  My public thanks to Harold Koh, John Bellinger, and Nancy Perkins for doing it. It’s a good thing for an administration’s senior lawyers, who have a difficult task of both setting out legal policies and often highly abstract and complicated legal arguments – and at the same time communicating them to the public, in part the professionals and lawyers and diplomats, but also to a broader public.  While John was adviser, he experimented with entirely new avenues of discussion and communication, including a guest blogging appearance at Opinio Juris international law blog that was very well received.  Harold Koh has also been doing some experimentation with different avenues of communication, and this kind of unscripted, informal discussion is an outstanding example of that.

(One note I would add is that a very great virtue of this kind of unscripted event is that it is informal, and not every word, phrase, and utterance has been vetted and run through the law-machine for alternative interpretations, and so on.  So although I strongly urge everyone to watch the video closely, I believe equally strongly that one has to adopt a charitable interpretation of what the speaker intends, and not focus on individual words or phrases that, in a formal speech or court filing or testimony, might be far more carefully – but less informatively – phrased.  So, for example, when Justices Breyer and Scalia held a discussion at my law school a few years ago on constitutional comparativism, in writing about it, I declined to quote them directly, preferring to paraphrase, precisely because I thought direct quotation was a disservice to the informal spirit of the occasion.  To hammer on precise words in impromptu settings simply causes lawyers to be ever more circumspect and less forthcoming, and to limit their statements to much less useful formal occasions.)

The conversation ranged across a wide variety of issues, including something that Julian Ku noted over at Opinio Juris blog (where I’m cross-posting) with respect to DOS international law counsellor Sarah Cleveland’s recent University of Virginia Law School speech on the Obama administration and international law – the pace of treaty exchanges.  John flags Dean Koh on that issue, saying (my summary) that in 2007-2008, the State Department got the Senate to approve more treaties (a record 90 or thereabouts, I believe) than at any point in American history.

On the broad question of whether the Obama administration’s international law policies represent continuity or change, Dean Koh suggested somewhat wryly that to the extent that the old policies were good ones, they were being continued, and to the extent they weren’t, they were being changed.  But Dean Koh also pressed the general theme that the Obama administration inherited policies, practical as well as legal, from the previous administration and turning on a dime wasn’t very easy.

Dean Koh was asked about targeted killing, including a specific question about targeted killing aimed at American citizens.  The exchange takes place at approximately minutes 58-65 in the tape.  The context is a question from someone – I believe, but couldn’t quite hear the feed, from an NGO – unhappy with targeted killings of American citizens, and wanting to know, among other things, whether an American citizen would have the right to contest the evidence against him in a court.  I raise this because Dean Koh’s response was being directed to someone who had strong views, apparently, that the practice, particularly against Americans (who, in the view of the US government, had made themselves liable to attack by joining with groups in hostilities with the United States), was unlawful.  The nuance, in other words, might have been different had it been someone, wanting to know why the State Department wasn’t out in front defending targeted killing as a practice.

Dean Koh’s response was interesting, in that he noted that he had served in both Republican and Democratic administrations, and as a human rights lawyer had sued both Republican and Democratic administrations.  He added that he already a permanent job, and so in effect (I paraphrase) he wasn’t beholden to anyone in formulating his legal views.  He went on to say that he would leave the government if he concluded that targeted killing was illegal.  He did not say that he had concluded that it was legal, but that he would leave if he concluded that it was illegal and, (my) presumably, if that conclusion were not accepted by the administration.  Having said that, however, he added that he was still there. Continue reading ‘Harold Koh Discussion with John Bellinger on International Law and the Obama Administration’ »

Last fall I was on a great panel at Stanford Law School on robotics and the law.  It had great people on it – Dan Siciliano, Paul Saffo, and Ryan Calo.  Great discussion; one of the things it brought home to me, as someone who came to law-and-robotics issues from laws of war questions, was how much those issues have cognates in other areas of emerging robotics, such as elder-care.  The panel discussion is up on video here:

Legal Challenges in an Age of Robotics, November 12, 2009.  One of the things I really liked about this panel was the way that Ryan Calo served as a very active moderator – he’s an expert in these issues himself, and so was able to lead the discussion, including the audience discussion.  The best parts are actually Dan Siciliano and Paul Saffo; I was a little unsure of how much the audience knew about the battlefield issues, and had too much wind-up.

(Ryan has also written very interesting stuff on privacy and technology.)

[youtube]http://www.youtube.com/watch?v=P021944l2LA[/youtube]

The Wall Street Journal has a weekend editorial on the same topic as my National Journal post below, Drone Wars, January 9, 2010.  Interestingly, I find myself believing that the WSJ editorial gets it, if not precisely wrong, at least puts the emphasis exactly backwards.  The WSJ editorial is primarily concerned to defend the “war” frame as a legal proposition for the global war on terror, and through that lens defend Predator drone attacks.  Only secondarily does it offer the idea of customary self-defense law, not necessarily part of an “armed conflict” as defined under IHL – meaning, treaties such as the Geneva Conventions.

I think the WSJ’s priorities here are part of the problem – not precisely wrong, but the wrong order of priority.  You can defend, as I do, the idea of a war on terror as a strategic frame for addressing terrorism – without, however, believing that you get or need all the legal particulars of a legal state of war everywhere on the globe.  The better way to approach this gap between the strategic frame and the law is that even though the strategic “war on terror” is not a war in a legal sense in all places in the world anywhere a target happens to located, the gap is filled by customary law of self-defense.

The Obama, Bush, and Clinton administrations apparently believed that they were on firmer legal ground going after “combatants” in an “armed conflict,” rather than relying on the customary law of self defense as an independent ground for the use of force.  Certainly it sounded better, as a law-PR matter, to say that one was targeting “combatants.”  What these administrations seemingly neglected to consider, as a legal matter, is that law of war treaties and customary law defining armed conflict actually have conditions and thresholds that must be met before all the particulars of the laws of war kick in.  Armed conflict in a legal sense is lex specialis, and you get its particulars only if the circumstances meet either the treaty law, in the case of inter-state conflict, or the customary law standards for armed conflict with a non-state actor, which requires sustained, persistent fighting and, in my view, in a place, even if loosely defined, and not simply the whole planet.  Armed conflict can break out in new places with a nonstate actor, but it does have to meet those thresholds.

(There is a problem here, one that I won’t address fully, but is important and not really much noticed.  Although a new “armed conflict” can get going outside of existing ones, the bias in this way of conceiving of it is that (absent the self-defense rationale, and assuming the territorial contraints and sustained fighting thresholds as I have described them) a “new” theater of armed conflict can be initiated by the non-state actors, where the NSA chooses to fight, but not by the state forces.  But notice that this isn’t really an issue for the legal frame for which non-state actor versus state conflicts were originally conceived, internal conflicts and civil wars in a territory; the territory is the national territory and so the problem of “global” war does not really arise.  It arises only if one accepts that a state can only fight in an armed conflict and armed conflict is territorial and theater-defined as a matter of law, but for that reason a state cannot initiate that use of force in a military sense without violating some human rights norms in revving up, so to speak, the fighting to a level at which an armed conflict exists.  This is a separate issue, however, and I won’t say more about it here.  It isn’t a problem, anyway, under self-defense, which was partly why it was not an issue in the past.  Framed the way the US has been framing its war on terror, however, it introduces yet another legal asymmetry in “asymmetrical warfare.”)

If it does not meet those thresholds, then, so far as the “international community” is concerned, the applicable law is the full weight of human rights law, under a view (never accepted by the United States) that the most important human rights treaty in this matter, the International Covenant on Civil and Political Rights, applies to the acts of US agents outside of the territory of the United States.  That is the basis, for example, on which the claim of “extrajudicial execution” is made against US Predator strikes – particularly in places like Yemen or Somalia, in which, although there are Al Qaeda targets present, it is difficult to make out a state of armed conflict with a non-state actor, because the fighting is not sufficiently persistent or sustained.

To be sure, the customary law standards for armed conflict with a non-state actor do not make a huge amount of sense in a borderless sense – they were conceived as a threshold for the application of Common Article Three of the Geneva Conventions, which covers internal and civil wars with non-state actor, and not for transnational terrorists who migrate around to many different places.  It was an innovation of the US Supreme Court in Hamdan – wanting the substantive standards of CA3 and willing to say whatever seemed necessary to get them – to declare that CA3 could apply to a borderless terrorist group such as Al Qaeda when it had never been conceived as a provision that applied outside of the traditional idea of civil war, internal to a state’s territory, with perhaps some spillover to safe havens, but not something in which territory was irrelevant.  Which is to say, Hamdan implicitly endorsed the idea of a global war on terror, although the idea of a global war in armed conflict terms is exceedingly … innovative is perhaps the kindest word.  It would have been better if the Supreme Court had simply said that there were minimum standards that always applied, and they were best stated in CA3, rather than saying that CA3 standards applied because it was a CA3 conflict.  If you are going after results-jurisprudence, just say so.

The better approach to this, rather than a global war on terror of the kind that the WSJ editorial goes to great, if mistaken, lengths to defend, is the customary law of self-defense.  That’s what the WSJ ought to be calling for first, rather than as an afterthought.  That’s the legal authority that permits the US to strike at its enemies whether in a combat theater or not, in safe havens far away from any regular battlefield, and it is the traditional authority on which the US has always relied.  And it is the authority on which the Obama administration is actually relying, if one looks to how it behaves.  That is, if you asked US government lawyers twenty or twenty-five years ago on what basis the US would strike Al Qaeda targets in Somalia, the answer would likely be, customary law of self-defense – if a state is unable or unwilling to control its territory, the non-state actor safe havens are liable to attack.

The sovereign  ”consent” argument that the WSJ puts first is true, as far as it goes – but it really ought to be the fall-back position.  Do we really think that if Yemen or Somalia refused to do anything about Al Qaeda operating from those territories, and refused to grant consent, the United States should or would refrain from attacking?  Consent is nice, but it is secondary to how the US in fact sees international law on this point, and how in fact it would behave. Ask candidate Obama who, after all, promised the American public that he would go after Al Qaeda in Pakistan, with or without Pakistan’s consent.  He was right, not just as a strategic matter, but as a matter of international law.  None of this is particular to the United States; this is how other great powers in the world that actually address military threats have behaved and would behave in regards to safe havens.

The US needs to plainly and publicly shift back to reliance upon customary law of self-defense in its Predator drone policies.  Not on a “global” war on terror, but on customary self-defense.  One reason this matters is that in the complex intertwining of activist, government, international organization, and academic positions around these debates, perception matters a lot.  One of the perceptions is that CIA officers engaged in these activities are legally much more vulnerable than a uniformed military person doing the identical task.  In making plain its reliance upon self-defense as the underlying legal basis for its targeted killing policy, the Obama administration has an important opportunity to make clear that self-defense law, which is not limited to armed conflict, allows – as US domestic law has since at least 1947 and the legislation creating the CIA – the use of force by civilian agents.  The reasons for using civilian agents versus military personnel are complicated – sometimes involving deniability, and other things.  But an assertion of self-defense as the legal basis would allow the Obama administration to affirm that it is lawful, under both international and domestic law, for CIA agents to engage in these activities, and that legal moves against them in other countries, now or in the future, would be distinctly unfriendlly.

The middle officials of the CIA in my estimation have doubts about that now – they see the whole activity moving the way that detention and interrogation did after 9/11, in part because targeting and detention, as Columbia Law professor Matthew Waxman has pointed out, have distinct similarities.  It is a point that both Waxman and Benjamin Wittes make in Shane Harris’ National Journal article.  The advocacy international community, for its part, seeks to move the debate that way.  Interrogation and detention have been made very legally uncertain activities for CIA personnel, and one part of targeted killing – only a part, however – is an incentive to kill rather than try to detain.  The advocacy folks would like to make Predator drone attacks as legally uncertain as detention and interrogation – and indications, at least in my experience, are that it does not take much uncertainty about what the future might hold in the way of Holder-style announcements, the possibility of indictments or arrest in Spain or other jurisdictions, and all that stretching into a long and unknown future, to induce changes in behavior by US personnel.  This is all very bad for the Obama administration, and it remains to be seen if it will step up to the plate and make clear that it stands behind its people, including its civilian officials in the CIA who are most vulnerable, on targeted killing.

It also remains to be seen whether Obama administration would actually be believed, as well.  It is an administration at risk of becoming known for saying merely whatever it thinks expedient in the moment, to whatever audience, for whatever ends.  You might disagree and it might indeed be an unfair characterization – but, again in my experience, that is what people in the relevant positions think.  The perception matters.  This is a legal position, however, and if the administration wants to reduce the legal uncertainty, it needs to have its legal officials state its legal position plainly and put themselves personally on the record as a matter of the US view of the law.  It matters if Harold Koh has reaffirmed his predecessors’ view on self-defense.

To that end, as well as the United States putting itself plainly on the record as to its legal rationales in the international community for targeted killing, the Obama administration needs to carry itself both beyond the narrow, inadequate, and not actually accurate legal position pressed by the WSJ editorial – and back to its customary law roots to the position that the US actually intends when it engages in targeted killing through Predators or other means, which is to say, self-defense.  The easiest way for the United States to make clear that this is its position as to what international law actually is, and that it is not an innovation on the US part nor a novelty, would be for it to return to Abraham Sofaer’s 1989 address on self-defense, a quite comprehensive speech on this topic regarding non-state actor terrorists, sovereignty and territory, the assassination ban, and other important topics in this area.  The WSJ editorial board needs to go back and re-read this speech, and press for it to be reaffirmed as US legal policy, and get away from its mistaken view that the way to approach targeted killing is as yet another way of reaffirming that this is a global war.  Self-defense gives the discretionary ability to attack anywhere in the world where a target is located, without having to make claims about a state of armed conflict everywhere and always across the world.

And the Obama administration, for its part, needs to reach beyond its indubitable state practice of targeted killing, and offer some opinio juris on the subject – i.e., its public and declared view of what international law on this topic is, as far as the US is concerned.  It should offer an affirmative rationale based in the law as the US has always understood it, and never disavowed – but not exactly reaffirmed in the last twenty years, either.  Let’s reduce the legal uncertainty and draw some firm lines.  A joint letter, signed by (oh, I don’t know, let’s dream) the general counsels or equivalents of DOS, DOD, DOJ, NSC, CIA, DNI, DHS, White House counsel, and perhaps a couple of others, reaffirming in toto the Sofaer address, and adding that any move by foreign states, courts, prosecutors, or international tribunals to go after American personnel, whether uniformed, civilian, or civilian contractors working to US government specifications, would be regarded as a distinctly unfriendly act.

(I’ll probably go back and fix some things later, but don’t have time now.)