Tuesday, January 11, 2011

Sri Lanka: "a grisly test case for success in modern warfare"? (UPDATED)

You've heard it before: the success of the brutal Sri Lankan campaign against the Tamil Tigers proves that Western approaches to counterinsurgency are fundamentally flawed: they're not brutal enough, not bloody enough, not enough torture or summary execution or firepower or ultimatums. Jon Lee Anderson, biographer of Che, opens his New Yorker piece "Death of the Tiger" with tacit acceptance of that argument, couched in generalized descriptions of the future conflict environment.
In many respects--its entrenched religious and ethnic conflicts, its festering guerrilla warfare and suicide bombings, its seamless interchange between civilians and combatants--the war prefigured any number of later conflicts. Where it differed was in the government's brutal effectiveness in putting down the insurgency. To the extent that a counter-insurgency campaign can be successful, Sri Lanka is a grisly test cast for success in modern warfare.
To the extent that a COIN campaign can be successful, of course. One wouldn't want readers to think that we're  labeling the Iraq war a "success" or holding out hopes for a favorable outcome in Afghanistan. Heavens no! It's far more fashionable to insist that counterinsurgencies never work... unless, that is, they're executed in some way that's completely unreproducible -- whether for circumstantial, practical, or moral reasons -- by the U.S.

Anderson's basic summary of the campaign is useful for its brutal honesty about the tactical measures employed by the Sri Lankan military. I'm summarizing here:
  1. Direct civilians to assemble in safe zones where they'll be free of military reprisals
  2. Heavily shell these areas
  3. Force remaining insurgents and associated (?) civilians into a geographically isolated space
  4. Cut off all transportation, resupply, and communications to this space
  5. Heavily shell this area 
  6. Kill everyone who doesn't succeed in fleeing and/or is not killed by insurgents during efforts to flee
  7. Kill insurgent leader
  8. Declare victory
Each of these steps is to be concluded in an atmosphere of media blindness, and with a firm and steadfast resistance to the corrosive effects of international opinion. Here's how Anderson describes it: 
In military circles around the world, the "Sri Lanka option" for counter-insurgency was discussed with admiration. Its basic tenets were: deny access to the media, the United Nations, and human rights groups; isolate your opponents, and kill them as quickly as possible; and separate and terrify the survivors--or, ideally, leave no witnesses at all.
All of which leads one to wonder, if one is a member of those "military circles" discussing the Sri Lankan job with "admiration" (and I'm not quite sure which circles those are, if I'm honest): how in the hell is all that supposed to work?

The Sri Lankan "insurgency" differed in character so significantly from the Iraqi and Afghan insurgencies as to scarcely justify the same label (just as, to be fair, the Afghan and Iraqi insurgencies differ/ed from those in Colombia, El Salvador, Vietnam, Malaya, Kenya, Algeria, and from one another). Only its character as a geographically isolated ethnic separatist insurgency made such tactics both possible and largely effective. One can draw parallels with Malaya, where forced relocation and isolation of an ethnic minority-based insurgency were also successful, but the process of forcing both guerrilla and civilian alike into successively smaller boxes and killing everyone who doesn't surrender is simply not a viable approach to countering a broad-based national insurgency. And that's leaving aside, of course, the fact that such tactics are FUCKING REPREHENSIBLE.

But that seems to be the point that a number of commentators are trying to make about counterinsurgency these days: you can't really do it well, and if you want to do it as well as possible, you're going to have to kill a shitload of people. Well, true and false. Violence is inarguably a part of effective COIN, just as it's a part of effective military operations of nearly all kinds (and certainly all kinds of war). But sustaining a high rate of fires and an aggressive, patrol-heavy force posture is not the same thing as packing all the living humans into a free-fire area and letting slip the dogs of war. 3-24 might soft-pedal the firepower bit (though not quite so much as its critics on either side suggest), but it's not covering for a Rajapaksa reality.

Just the same, read the whole New Yorker piece if you have access. Anderson is an excellent reporter and an easy storyteller; you can't blame him for the zeitgeist.

UPDATE: Niel Smith helpfully links to his piece on the subject in the 4th Quarter 2010 issue of JFQ. It's worth repeating his conclusion here:
Those who wish to use the LTTE’s [Liberation Tigers of Tamil Eelam, or the Tamil Tigers] defeat as a foil for criticizing U.S. COIN doctrine have adopted an overly simplistic narrative of the LTTE’s defeat. These critics have missed the larger picture of what occurred in Sri Lanka. Appropriate and legitimate debate continues as to the significance of population centric tactics practiced by the U.S. military during the surge to the successful reduction of violence [in Iraq]. Without doubt, numerous changes in the wider internal and external dynamics of the conflict coincided with the tactical shift and accelerated the turnaround in Iraq. Likewise, by 2009, the LTTE was a shadow of its former self, bankrupt, isolated, illegitimate, divided, and unable to meet an invigorated government offensive of any kind. At almost every turn, the LTTE made profound strategic miscalculations in the post-9/11 environment by continuing its use of terror tactics despite a fundamentally changed global environment. Failing to realize this shift, [LTTE leader] Prabhakaran made poor strategic and tactical choices that doomed his movement long before the government began its final offensive. Taken together, these conditions proved essential to the collapse of the LTTE after nearly 30 years of conflict.
N.b.: the Tigers didn't do themselves any favors while becoming one of the most well-known terrorist outfits on the globe. Things might not have shaken out the same way had they not so effectively worn out their welcome with patron and undecided alike.

18 comments:

  1. Might want to read my JFQ Article:

    http://www.ndu.edu/press/understanding-sri-lanka.html

    Adds some perspective.

    ReplyDelete
  2. Niel -- I remember having read your piece when you first published, and I should've linked it here. I was trying to track down the Indian Defense Review article you cited but the name was failing me. Thanks for your comment.

    ReplyDelete
  3. All the bulleted points is similar to what Israel used against entrenched Hamas network in Gaza.So why pick only on Israel? You got to do what you got to do.

    ReplyDelete
  4. All the bulleted points is similar to what Israel used against entrenched Hamas network in Gaza.

    I'm no reflexive defender of Israeli CT tactics, but I feel less than completely confident that this is true. Evidence/details, please?

    ReplyDelete
  5. "But that seems to be the point that a number of commentators are trying to make about counterinsurgency these days: you can't really do it well, and if you want to do it as well as possible, you're going to have to kill a shitload of people."

    There's another side to this argument; namely that those who believe you can fight effective counter-insurgencies by focusing on protecting civilians rather than killing insurgents have a hard historical argument to make. I suppose that's part of the reason that in Afghanistan the emphasis has shifted toward killing insurgents.

    FWIW, I think the most pertinent lesson from Sri Lanka is not that brutality is necessary to wage a successful one . . but that coercion and violence (generally against civilians) is an intrinsic element of counter-insurgency. Indeed, in the all examples you cited above that is perhaps the one consistent variable.

    ReplyDelete
  6. FWIW, I think the most pertinent lesson from Sri Lanka is not that brutality is necessary to wage a successful one . . but that coercion and violence (generally against civilians) is an intrinsic element of counter-insurgency.

    Coercion yes, probably, assuming you're using a generous definition. I'm also willing to accept "violence," less so the "(generally against civilians)" part.

    namely that those who believe you can fight effective counter-insurgencies by focusing on protecting civilians rather than killing insurgents have a hard historical argument to make

    I think you're creating a false duality here. Killing some insurgents is always necessary, and any serious student of counterinsurgency will concede as much. Some commentators have been too eager to accept the "protecting civilians" rhetoric (and here I think there's a fair bit of blame for the self-styled "progressives"), but that shouldn't be taken as evidence that civilian welfare is a sufficient or determinative condition. As I've often tried to point out, counterinsurgency is fundamentally concerned with population control, particularly in its earliest stages (when control over territory is in fact the most important thing). Earning the allegiance of undecideds may be crucial to total "victory" and eventual reintegration, but it will almost always follow sequentially on control. Counterinsurgents who fail to appreciate that collaboration follows control (and is followed by political reconciliation) are working at a significant disadvantage.

    I think a great deal of your resistance to the "COINdinista" set comes from what you see as a disconnect between rhetoric and reality, and I won't argue that it doesn't exist. But I think it's more a matter of spinning things in a favorable light while executing with different priorities, not the dramatic misreading of history that you fear. I'm not saying this is cool, or that we should be sanguine about it, and I'm certainly not. But it's probably better (from the perspective of policy outcomes) than trying vainly to wage a love n' hugs war that's built on a false reading of history. There's a whole other CMR/IO on the body politic conversation that springs out of this, but I'm not going to go there right now.

    ReplyDelete
  7. Gulliver,

    Just a few points here that have little to do with Sri Lanka:

    - Population control is important. But because of geographical features and a largely rural population, population control doesn't work as well in Afghanistan as it did in Iraq. It is very difficult, and even more so given the #s of available troops, to control every rural village in Afghanistan. Even if you limit yourself to just the "strategically important" villages, there are still too many to cover. Then there is the long, mountainous land border with Pakistan, also, by the Army's own admission, uncontrollable.

    -- Population control can be very coercive. The French force-ably moved entire rural villages to the cities in Algeria. I doubt that's a tactic we'd be willing to use in Afghanistan.

    So here's the thing on population control: We aren't willing to use highly coercive/violent measures, rightly so as you point out, to control the population. Our "softer" measures, at least so far, haven't been terribly successful, as much of Afghanistan is still not under any sort of government control and the most important border is completely out of control.

    Where does that leave us? We might be able to design a work around, that is, if other parts of the effort were going well...like say, we could eliminate Pakistani safe havens (though I think there are also internal safe havens we can't reach, but that's more of a hunch) or the Afghan government was a lot more legitimate or the ANSF wasn't a complete disaster.

    I guess my point here is that critics, like me, of our Afghan COIN effort are rejecting, quite specifically, the Sri Lanka model. But the critique doesn't end there. There are plenty of reasons, and I've tried to lay them out above, that one could be anti-Afghan COIN while stipulating against the use Sri Lanka type tactics.

    * I haven't footnoted this, but I can provide links to back up the assertions on control, legitimacy & the ANSF.

    ReplyDelete
  8. Keith -- I don't dispute any of this, and in fact spend a lot of time making these very points.

    Just the same, thanks for commenting.

    ReplyDelete
  9. "As I've often tried to point out, counterinsurgency is fundamentally concerned with population control, particularly in its earliest stages (when control over territory is in fact the most important thing)."

    I of course agree, but my only point is that population control has almost always been maintained via coercion. The Malaya experience is a good example; if you read Nagl he would argue control came because of civic action, but in reality it was done via resettlement.

    Because pop control is generally done via coercion it leads, in my view, to the conclusion that the US carrot-based approach won't work - and in many respects undermines the entire premise of FM 3-24. This to my mind is a crucial point.

    "Some commentators have been too eager to accept the "protecting civilians" rhetoric (and here I think there's a fair bit of blame for the self-styled "progressives"), but that shouldn't be taken as evidence that civilian welfare is a sufficient or determinative condition."

    Agree on apportioning blame here - but I would just make the point that this was more or less the USM approach in 2009 and the first half of 2010. I mean M4 said this repeatedly - in his testimony and general orders to the troops. You're right about seeing a disconnect btw rhetoric and reality and i think there are serious policy implications of that i.e. in encouraging more not less COIN operations because they are seen as less intrusive to civilians. But at the same time it's not jut rhetoric; it did shape US policy - less so today.

    ReplyDelete
  10. Weren't the Tigers more like the government in an insurgency vs. counterinsurgency point of view? They controlled territory and aimed to hold it. They had an administration operating openly in it. They had a capital and an army and a navy and an air force. The Sri Lankan army's primary aim was to kill people, smash stuff, and bring about mayhem until they cracked and gave up.

    Does it even make sense to consider it a counterinsurgency? It was once, but that was a long time ago. Rather than trying to eventually bring about the revolution, the Tigers had carved out their separatist state and the war was about (on their side) holding ground and imposing pain on the government in the hope of forcing them to accept it, and on the government's side, trying to reconquer the Tiger zone.

    ReplyDelete
  11. I of course agree, but my only point is that population control has almost always been maintained via coercion. The Malaya experience is a good example; if you read Nagl he would argue control came because of civic action, but in reality it was done via resettlement.

    Agree that Nagl overemphasizes certain factors to the detriment of his credibility, but I think you mischaracterize his argument slightly. I'd need to go back and read it again to be sure.

    Because pop control is generally done via coercion it leads, in my view, to the conclusion that the US carrot-based approach won't work - and in many respects undermines the entire premise of FM 3-24.

    This is the crux of our general disagreement: I don't think that "the entire premise of FM 3-24" is that love n' hugs COIN paves the road to victory. I think, again, that this is a mischaracterization. 3-24 emphasizes soft-power-y, persuasive tactics at times, yes; but the public narrative about the doctrine has emphasized this far more than the doctrine itself. (This is actually the subject of the much more extensive critique of your Nation piece that I was working on a month ago but set aside. Will try to take it up again some time soon.)

    but I would just make the point that this was more or less the USM approach in 2009 and the first half of 2010. I mean M4 said this repeatedly - in his testimony and general orders to the troops.

    Again, not sure repeated public insistence is necessarily the whole truth here. Obviously ROE played a visible part, but it's tough to know the overall context without a more granular picture of both operations and TTPs during the McChrystal period.

    ReplyDelete
  12. I'm going to side with Michael on this - especially the nature of FM 3-24. The verbiage itself in the doctrine is somewhat vague, but it's almost all carrots and few sticks. It was worthless to me as a planning tool in 07/08. And the GO corps generally (no pun intended) interpreted it as the hugs, etc. How many division warfighting handbooks had a "Love Bank" graphic in them? I know of quite a few.

    As for Sri Lanka and it's relationship to the body of knowledge on COIN, the lessons from that conflict do fit into a box of its own. 1) The final stages were more conventional fights that civilians were caught in. 2) The Tamils moved from Ph 2 to Ph 3 before they were ready - which often leads to disaster. 3) The Tamils were becoming as oppressive as the state government. 4) There's little conflict now, but it seems that is not likely to last as the government hasn't learned the carrots part of what they're doing. Oh, and it's on an island. I can't say we can glean too many lessons from it.

    ReplyDelete
  13. A couple more points about the Sri Lankan case that make it, IMHO, so different as to be nearly useless for Western COIN practitioners:

    1. The Sri Lankan government was acting on its own behalf; it wasn't fighting the Tamils on behalf of and constrained by a "legitimate" proxy. This gave them a unity of effort and freedom of action that can't be matched in the current iteration of Western COIN.

    2. The LTTE held Sri Lankan territory. The Sri Lankan government fought to regain what it considered important pieces of its state, held by an organization that used terrorism to strike at the heart of their country. A much more coherent narrative of existential threat and vital national interest than anything Western COIN advocates can manage. Not to mention the logistical advantages to not having to sustain your force 6500 miles from home (obviously not outweighed by fighting on your own territory, but worth noting).

    And these aside from the morally unacceptable tactics, the fact it's an island, and other differentiators already noted.

    ReplyDelete
  14. To build on what I believe Stilicho is saying in point 1:
    I think most people would agree that each insurgency (Afghanistan, Iraq, Malaya, Sri Lanka) is unique and different, but the hope is that we can gain some insight and best practices from each to move forward in the future. This motivates much of the discourse on the subject, and is positive, but I think tends to blur the lines of these disparate conflicts by making us think in academic terms of ‘insurgent’, ‘counterinsurgent’ and ‘population’ and fitting in the appropriate actor (US, Taliban, ect). I believe this then divorces the actor from their unique operational environment and leads us to equate one counterinsurgent with another and one insurgent with another. Much has been said about the merits of more coercive actions versus FM 3-24 techniques (sticks and carrots to be simplistic). The missing piece of this debate is that the actors in each of these insurgencies are not equal in terms of their relationship with (or legitimacy within) the population. A Sri Lankan government who (largely) represents the ethnic majority fighting against an ethnic minority (isolated on an island) will face different constraints than the US (a third party force) fighting with, and on behalf of, a home government who doesn’t represent anyone conclusively against an insurgency with diverse representation, but largely made up of the nation’s most populace ethnic group, the Pashtuns. We could get into a long discussion over who and what the government is made up of and represents, and who and what the Taliban is made up of and represents, but it is not that important for this short comment. My point is that the legitimacy of each faction and the power bases they draw from are not particularly clear cut, especially in comparison to the Sri Lankan case. The political environment within which an actor (both counterinsurgent and insurgent) operates is dictated by their legitimacy drawn from the population bases in the country. This environment will dictate greatly the freedom of action an actor has and the tools at their disposal. Now, obviously, the US can ‘do’ whatever it wants, but the reaction to these acts will vary greatly to the reaction experienced by another counterinsurgent, such as the Sri Lankan government. The relationship between the US, the Taliban and the Afghan population is not equal to the relationship between the Sri Lankan government, the Tamil Tigers and the population of Sri Lanka. Debating carrots or sticks without appreciating how these incentives will to construed or responded to seems to miss a big piece of the puzzle.

    ReplyDelete
  15. "I don't think that "the entire premise of FM 3-24" is that love n' hugs COIN paves the road to victory. I think, again, that this is a mischaracterization."

    "The entire" might be too strong, but I find it very hard to argue that the focus of FM 3-24 is not far more on carrots than sticks. If you want to argue that I've mischaracterized what is happening in the field that's one thing; but to say it's not the central feature of FM 3-24's conception of COIN strikes me as a very tall hill to climb. After all the book goes out of its way to de-emphasize the kinetic aspects of COIN. How else do you explain a discussion of CORDS that completely ignores Phoenix?

    "But the public narrative about the doctrine has emphasized this far more than the doctrine itself."

    Just to be clear not only do I agree, but in fact this is basically central to my entire argument - that the public narrative on COIN is deeply misleading about the nature of COIN fights. And as I've argued that the doctrine doesn't even accurately describe what the US did in Iraq, which as Jason knows was highly kinetic.

    As for Nagl, I've read Learning to Eat Soup with Knife I've read a lot of the literature on Malaya and I don't think I mischaracterize his argument at all - but if you think so I'm all ears!

    "Obviously ROE played a visible part, but it's tough to know the overall context without a more granular picture of both operations and TTPs during the McChrystal period."

    This is true; but the stats tell a tale - airstrikes are up 300%, body counts once deemphasized are now being emphasized and of course the ROEs matter too. I think one can fairly accuse me of overstating this argument but I think it's firi to say that the emphasis of the US COIN focus has shifted - which is what I argue in the Nation piece.

    You should read my World Policy Journal piece as well on this.- might help with your critique.

    ReplyDelete
  16. On Sri Lanka I think Jason and t get this right and I don't think many lessons apply . . . except a more general one: namely about the fundamentally coercive and violent nature of COIN.

    ReplyDelete
  17. The main argument I see missing from the critics of COIN is that both recent examples (Iraq and Af/Pak) have become COIN fights many years into the conflict. Its been COIN as a desperate measure to avoid failure, not as a well-designed and carefully calibrated plan. With the danger of sounding like a broken record, those 7 years wasted in AFghanistan were the real window of opportunity for a COIN-focused army. What were doing now is more or less schizophrenic.

    ReplyDelete