Tweet
To their credit, the folks over at the Brookings Institution have become one of the first mainstream think tanks to recognize the horrendously indiscriminate nature of drone attacks in Pakistan. Brookings Institute scholar Daniel Byman wrote last Monday:
Critics correctly find many problems with this program, most of all the number of civilian casualties the strikes have incurred. Sourcing on civilian deaths is weak and the numbers are often exaggerated, but more than 600 civilians are likely to have died from the attacks. That number suggests that for every militant killed, 10 or so civilians also died.
I’ve been citing numbers that show a worse civilian-combatant ratio (15-1), but the Brookings citation makes the same point: drones kill far more civilians than suspected militants. Good for Brookings for bringing this to folks’ attention.
Unfortunately, though, Byman fails to really get into the details of what causes the high ratio, preferring instead to attribute them to the Evil Taliban:
To reduce casualties, superb intelligence is necessary. Operators must know not only where the terrorists are, but also who is with them and who might be within the blast radius. This level of surveillance may often be lacking, and terrorists’ deliberate use of children and other civilians as shields make civilian deaths even more likely.
The preceding paragraph demonstrates an amazing Fareed-Zakaria-like ability to take the vile and the shocking and transform it into a passive-voice bromide. Translation: "We need good intel to avoid killing noncombatants. We don’t have good intelligence. We don’t let details like that get in the way of firing the weapons, so we kill 10 civilians for every one suspected terrorist. Oh yeah the Taliban are bad."
Americans should be terrified and horrified that CIA operators use a weapons system whose ability to avoid killing innocent men, women and children depends on "superb intelligence" when such intel does not exist. Essentially, what the CIA is doing is analagous to a police sniper aiming into a bank crowded with hostages with a sniper rifle whose barrel lacks rifling, pointing at a suspected robber and pulling the trigger. When the bullet goes astray due to the lack of a key feature that makes the sniper rifle accurate–the rifling– and kills a hostage, the police officer shrugs. "The robber used human shields." If the public found out that our hypothetical police sniper knew in advance that he had, oh, say, a 90-percent chance of killing a hostage rather than a robber and he pulled the trigger anyway, they’d be howling for his head on a platter. But this kind of vile nonsense is exactly what the administration asks the American people to accept through further escalations of the CIA’s undeclared war on the Pakistanis unlucky enough to be living near our national enemies.
I repeat:
The strikes have caused such carnage that leading British legal experts “said the aircraft could follow other weapons considered ’so cruel as to be beyond the pale of human tolerance’ in being consigned to the history books,” likening them to “cluster bombs and landmines.”
Byman’s analysis of the problem, though, ultimately misses the point. It may be true that the high civilian death rate is bad because it undermines our counterinsurgency efforts to win hearts and minds. However, the real problem is not the political consequences of these deaths, but rather the deaths themselves. Even if the 10-1 civilian-combatant death rate had zero political consequences, it would still be immoral to continue the use of drones. As I said on July 14,
"The worst effect of all this talk about counterinsurgency is that it has reduced the civilian populations of countries like Iraq and Afghanistan to mere means to the end of our strategy. They’re not. Drones may be awful in part because their use leads to more terrorism, but the worst effect of their use is the slaughter of people whose right to life exists independent from our goals for the region."
Get those drones on the ground, now.
UPDATE: Despite its problems, the Brookings article shows that the CIA is lying to the American people about the drones. Here’s Leon Panetta in a May 2009 speech:
"[Drone] operations have been very effective because they have been very precise in terms of the targeting and it involved a minimum of collateral damage."
Very simply, Panetta lied.
UPDATE II: The Long War Journal just published an analysis of drone strike activity in 2009 compared to 2008 [h/t/ Noah Schactman at Danger Room]. Their study shows that compared to last year, drone strikes have been more frequent and have killed more people, with the total number of deaths for 2009 already exceeding the 2008 total :
…In 2009, the frequency of Predator strikes in Pakistan has continued to trend upwards. There have already been 31 Predator strikes in Pakistan this year (as of July 18) – nearly matching the total of 36 strikes for all of 2008.
If airstrikes continue at the current rate, the number of strikes in 2009 could more than double the dramatic increase in Predator activity seen in 2008.
…Using low-end estimates of casualties (including Taliban, al Qaeda, and civilian) from US strikes inside Pakistan, we have determined that airstrikes resulted in 317 deaths during 2008. Already, the airstrikes in 2009 have surpassed that total, with 365 killed in 2009 as of July 18. [see Chart 2, Deaths]
…Another indicator of the increasing lethality of US airstrikes inside Pakistan is the rising average number killed per attack. So far in 2009, the average casualty rate has been 11.77 killed per strike, compared to 8.81 in 2008. [see Chart 3, Lethality]
So, to summarize:
- CIA drone operators lack the "superb intel" needed to prevent civilian casualties, but are firing their weapons anyway, causing them to kill ten times as many civilians as suspected terrorists.
- CIA Director Panetta, however, continues to lie and/or propagandize about the drones’ accuracy and "minimal collateral damage."
- Despite their indiscriminate and inhumane nature, the U.S. has doubled the rate of drone strikes and is killing more people per attack in 2009 compared to 2008, which has caused the death toll from these weapons so far in 2009 to exceed the death toll for all of 2008.
History will not be kind to us if we continue to use these indiscriminate weapons that kill ten times as many civilians as suspected combatants.
(Cross-posted at Return Good for Evil)
18 Comments
Hasn’t our intelligence improved and do you have any accurate idea of the ratio of civilian/militant deaths in 2009. All I see in the English-language Pakistani newspapers are quotations of Pakistanis saying that the drones are killing militants with few civilian deaths.
I don’t buy into this any more than I tend to believe that your 10:1 number is current.
As far as I know, this is the best information available, and the 10-1 ratio is a low-end estimate. I haven’t seen anything that shows our intel improving. In fact, the report from the Long War Journal listed above shows we’re using Pakistani news reports as major sources of information for targeting–so you can’t simultaneously discount the accuracy of reports in the Pakistani press while defending our targeting intel. Also, see here for more on our “superb” intel: http://www.wired.com/dangerroo…..more-14503.
I think that you’ve misread the Long War piece. I think that it doesn’t say that we’re using Pakistani newspapers for targeting info, but rather using them for info on the casualties induced.
Possibly. The wording is a bit ambiguous:
Read more: http://www.longwarjournal.org/…..z0LvPATsTS
Not merely possibly incorrect, Derrick. Think about how absurd it is to assert that we could plot people’s movements via newspaper.
macaquerman:
Fair point. However, I’d also note that, related to your larger point, the information linked to in my original post also details some of the absurdities of our so-called “intelligence.” On that larger point, however–do you have some links re: our intelligence “improving?” Or suggested sources of casualty data you feel are more reliable?
No, Derrick, I don’t have any solid info on how precise our bombing is currently. You’ll note that this is what I was asking for in my first comment. All I was seeing were the reports in the Pakistani press, which had become suspiciously laudatory in stating that nearly all the slain were foreign fighters and Talibani.
Your Long War link shows that nearly all of the “high value” targets have been killed over the past year but doesn’t
have any information about the number of civilians.
The reason I’m turning the question around on you is that I’ve cited a pretty staid organization whose estimate is on the low end of the estimates I’ve seen, and you’re response was an assertion that the information isn’t current, but you’re not giving me anything to work with.
If you note the first link, Brookings cites a low-end estimate of 600 people. Other estimates go into the high 700s. As to where Brookings gets their estimate, it comes from analysis of the Pakistani press, which agrees with that of Amir Mar. Both of those guys are heavy hitters re: counterterrorism / Pakistan expertise. I’m fine with relying on their numbers. As I said, if you’ve got contrary information, let’s hear it, but gut feeling “I don’t buy it” doesn’t cut it.
Important thanks for your focus on this critical issue. Folks just do not want to think about the innocent people beings killed in our names
Thanks much. Spread the word!
(deleted and moved to the proper slot for a reply to a previous comment)
The drones are much more accurate than the mass bombing of Dresden, but this is WAR. Civilians do get killed in war. It’s called collateral damage. The terrorists fight their battle in the midst of civilians to attract more collateral damage.
mboutot:
Wrong. We’re not officially at war with Pakistan…right?
Plus, re: Dresden: I don’t think I’ll be using past atrocities to defend current atrocities. The remarks of Churchill et. al. make it clear that the purpose of firebombing Dresden was to “drive home” to the population the error of their ways in consenting to their government. We have a word for violence directed at civilian populations aimed at achieving political ends: terrorism.
Lets just invite the terrorists over here and ask them to bring their weapons of mass destruction with them. Let’s make it easy for them to kill us instead of us killing them.
That sounds pretty progressive to me.
That the drones are not carrying out the longed for “surgical strikes” is obvious. Why is not so obvious. So lets consider what is known about the weapon and then draw conclusions about what is happening in Afghanistan and Pakistan.
First, there can be no doubt that drone strikes cause fewer casualties than many other weapons. The Hellfire–the missile used–carries a warhead of about 20 lbs (9 kg). By comparison a 500-lb aircraft bomb carries about 200 lbs and has a significantly larger lethal radius. Hellfires are guided very precisely by the reflection of a laser that the drone or a ground observer shines on the target–the same mechanism as used with laser-guided bombs. Hellfires should be more accurate than a bomb, because they have a motor and ought to be less affected by crosswinds and such.
So what is going on in Afghanistan? While atmospheric conditions can cause misleading reflections and while missiles can lose their lock on the correct reflection and go ballistic, I doubt that either can explain more than rare cases of mistargeting. It is much more likely that CIA is simply aiming at the wrong houses.
This is the weak link in this approach to warfare. We can hit the correct target dead on almost every time and limit the destruction and death to the immediate vicinity of the target–IF we know what the correct target is. But, in practice, we almost never do.
We do not see the enemy in most cases, if at all. Even if we do, they probably all look alike through the lens of small TV or FLIR camera like those on the drones. So we rely on our interpretations of what informants or intercepted communications tell us, usually in languages we still have too little knowledge of. We conclude that so-and-so will be in this particular village. We notice a banquet in a particular house and assume that so-and-so is the honored guest. Or an informant tells us that the biggest house is where so-and-so will be sleeping. But, unfortunately, the banquet is for an unrelated wedding. Or the biggest house happens to be that of our informant’s hated brother-in-law. Or the house belongs to the government representative and our informant is a Taliban agent. Or the wrong house just looks bigger to the operator–he is looking at it from hundreds if not thousands of miles away, through the narrow field of view of the drone camera and possibly through intervening heat haze, mist, or smoke. If a lot of people are in the house for whatever reason, there will be a lot of casualties. The Hellfire warhead is not large as such things go, but 20 lbs of high explosive in a confined space (with or without the new thermobaric warhead design) will not leave many survivors.
We need to understand that the real reason for the use of the drones is their political pallatability, not their lethality or accuracy. To be sure of killing Taliban or Al Qaida members while sparing civilians, we’d do better to send infantryman on the ground to walk up and look at them before shooting (we’d still kill civilians, but not as many). We use drones instead because drones are more politician- and public opinion-friendly. Drones appear to discriminate between friend and foe the way a well-trained foot soldier would. But you do not have to risk the foot soldier, who is someone’s constituent. The people that die are far away and are constituents.
The use of drones is thus symptomatic of our unwillingness to commit to one course of action or an other. We want to eat our cake–by killing our enemies, being strong on something, etc.–and we still want to have it too–by not taking casualties. Technologies like the drones and so-called tactical innovations like surges let us keep up this self-delusion. We have been able to fight seven years in a hopeless war in Iraq while keeping casualties down to a level acceptable to all but the families of the dead and wounded. We have not “won” (whatever that wpould mean in this context), but we have also escaped obvious defeat. Nobody can currently be blamed for “losing” Iraq or Afghanistan.
But this can’t continue. Either now or at some point soon, we have to decide whether we really want to achieve our supposed war objectives enough to pay the price in American lives. And if we do not, as most of us do not, then we have to stop dragging the thing out. We have to recognize that “victory” is not worth the effort and expenditure it would require and just stop.
Robspierre:
Thanks for the thoughtful reply. A couple of points:
While the drones certainly can use the Hellfire missile, they can also use much larger bombs. From Danger Room’s David Hambling:
Also from the Danger Room, on the topic of targeting (which goes a long way to explaining civilian casualties):
You’re right about the political reasons for the use of drones, in my opinion, but also taking into account that we are using drones in Pakistan because Pakistan would not tolerate our boots on the ground.
The IR beacons are probably used to designate the building that the drone will lase when an informant has access to the target. These beacons look like the same sorts of things that were issued to our troops after several needless nocturnal friendly-fire incidents in the first Guld War. The Forward-Looking InfraRed (FLIR) camera will see the hot spot.
I’m sceptical about the ability of the Reaper to carry loads of the magnitude your informant describes. Predators definitely can’t. I’ve seen photos of Reapers with four Hellfires–twice the Predator load–which toals about 800 lbs. Paveway II is nominally a 500-lb bomb, but even the base bomb ways something like 580 lbs. Add more for the wings and guidance kit. Two of these are more lethal but cut the firepower in half (fewer shots). They also push the weight. JDAM is even less likely, because it is GPS-guided. It is less accurate than laser but ahs the advantage of being able to hit targets (more or less) through cloud. This is no advantage to the drones, because they use optics for guidance and do not fly above weather.
Even the Reaper is not very large or powerful. It has to operate at high altitudes and often in high temperatures, both of which reduce performance and load. Also, higher weaqpons loads means less fuel and less range and/or less time on station.
I have read that the Israeli Spike mssile has been tested on our drones. This makes more sense. Spike only weighs about 100 lbs and has an imaging-IR seeker that lets you see what (who?) the missile is seeing. So it would be attractive both for weight and accuracy. Unfortunately, when used as we are using it in Afghanistan, it will only see the same houses.
I checked online and it appears that the loads described are theoretically possible–just. There are photos showing a Reaper with four Hellfires and two 500-lb bombs. But with full fuel, the remaining payload is only 1600 lbs, so you could not have full fuel with this load. Even with reduced fuel, a 2000-lb plus load does not leave a lot of margin for altitude and temperature. I’m only guessing, but four Hellfires seems like a more reasonable, normal load.