Small Wars Journal

What Sri Lanka Can Teach Us About COIN

Fri, 08/27/2010 - 8:05am
What Sri Lanka Can Teach Us About COIN

by Lionel Beehner

Download the Full Article: What Sri Lanka Can Teach Us About COIN

It has become a truism to say there are no military solutions to defeat an insurgency. That was the thrust of the U.S. military's 2006 counterinsurgency (COIN) manual as well as the mantras repeated by CENTCOM Commander David Petraeus, the manual's coauthor, and his "warrior intellectual" offspring. Conventional wisdom also holds that COIN takes years, if not decades, to complete and emphasizes a population-centric strategy to avoid civilian casualties and win locals' hearts and minds.

But Sri Lanka's successful victory one year ago stands all this conventional wisdom on its head. It was brute military force, not political dialogue or population control, which ended its brutal decades-long war with the Liberation Tigers of Tamil Eelam (LTTE), or Tamil Tigers, a separatist group perhaps most notorious for popularizing the suicide bomb. The final military campaign lasted months, not years or decades. It was a gruesome finale, to be sure. The Sri Lankan government paid little heed to outside calls for preventing collateral damage. While humanitarian workers and journalists were barred from entering the war zone, as many as 20,000 civilians were killed in the crossfire and hundreds of thousands of internally displaced Tamils were corralled into camps after war ended . It was, as one journalist I spoke to in Colombo put it, "a war without witnesses." Hearts and minds took a backseat to shock and awe.

Still, the lesson from Sri Lanka's COIN experiment is that overwhelming force can defeat insurgents, terrorists and other irregular armed groups in relatively short order, but at a steep cost. Its model disproves the notion that counterinsurgencies must be drawn-out, Vietnam-like campaigns. With U.S. forces bogged down in Iraq and Afghanistan, it also provides states fighting small wars with a different counterinsurgency template. Not without reason did Pakistan and Thailand, which both face insurgencies on their peripheries, seek out Sri Lanka for military training and advice in recent months.

So do America's warrior intellectuals and COIN theorists have it all backwards? Should we be emphasizing military solutions over political compromises and accommodation, overwhelming force over clear-hold-and-build campaigns, defeating the enemy over winning locals' "hearts and minds"? Does Sri Lanka's COIN strategy provide any lessons for Washington as it escalates the war in Afghanistan, or for other countries facing violent insurgencies along their unruly peripheries?

Or does the fallout from the use of massive force—the high death toll, the lost hearts and minds, the accusations of war crimes, the unresolved grievances of ethnic minorities—negate whatever victory is achieved on the battlefield or goodwill that comes from a peaceful settlement? It is a perplexing question for military strategists. "The end of the Sri Lankan civil war," wrote Robert Haddick, a managing editor at the Small Wars Journal, "most especially the way it ended, with a clear military solution -- will cause many sleepless nights for Western counterinsurgency theorists."

Download the Full Article: What Sri Lanka Can Teach Us About COIN

Lionel Beehner is a Ph.D. candidate at Yale University and formerly a senior writer at the Council on Foreign Relations, where he is also a term member.

About the Author(s)


Neil Smith your points immediately above are accurate. And as with any defeat of insurgency it appears a combination of factors lead to success. I agree though that when you have zero care for human life and have shut out the eyes of the world - well you have leveled the playing field with the insurgents. No hearts and minds to care about here.

I witnessed first hand the breakaway by Karuna. Unfortunately the gunman stuck his AK in my face after the hijakcing of the van and assaninating the LTTE dudes who were taking Karuna to Kilinochi. (that was going to be a one way trip for Karuna)

Karuna established the TMVP that moved from being its GOSL back guerilla movement against the LTTE to being a political force along the Eastern Province.

It was a politically smart move of the then new Eastern Chief Minister in 2008 to visit the Kattakundy mosque to apologise for the massacre that occurred there 12 years ago(instigated by Karuna)

They then carried out systematic assaninations / Kiddnappings of Tamil business people and academics. A close friend of mine dissappeared in 2006.

The GOSL was very effective at putting the screws on the LTTE funding and resources at a multilateral level. They made top rate use of global tools and the Diaspora to lobby Western governments to slowly but surely close off support. Post 9/11 this became easier with the world wide legislative changes to cut off funding channels to organisations linked to insurgency groups. This seriously depleted the LTTE of funding to run their military campaign, deliver food and supplies to its cadres and population.

The increasing sophistication of the weapons and armour being used by the GOSL from around 2006 onwards it was obvious external parties had injected support in funding, equipment or both.

The nationalist Buddahist movement in Sri Lanka is enormously powerful and would never allow any negotiated peace. They applied significant political pressure on any member of the GOSL who was in favour of peace. In fact if you were a proponent of peace in Sri Lanka you were considered "unpatriotic". One of my favourite new words invented by the GOSL is "peacemonger" used to describe those who were publically calling for negotiations to peace.

The GOSL was effective at evicting anyone from Sri Lanka who even smelt like they were in favour of peace.

The GOSL effectively shut off development assistance to the North from international organisations. It was a test of Machiavellian tenacity with the GOSL to be able to continue operating in these areas.

I was one of the first Westerners allowed into the high security zone just north of Vavuniya within two weeks of the military offensive.
The humanitarian situation was unimaginable.

The situation has barely changed on the ground for the population. Many of the military and police took land and houses vacated by the civilians when they were fleeing the offensive.

What the GOSL must be worried about is how it will deal with the mass of humanity seething in the north and in no better situation than before. While the GOSL may have crushed this generation of Tamil leadership Im not sure that the next generation will be happy to forgive and forget.


Yes and no. Karuna's defection mentioned in the article had much more to do with power struggles within the LTTE than him reacting to pressure from the government. Karuna left well before the recent 2007 offensives started.

My larger point is insurgency survival is basically predicated on its ability to generate/regenerate combat power. Countries with long, rough borders will fare badly when neighboring countries actively or tacitly support the insurgents. Countries which can isolate their borders will have an easier time.

Building on Mckinlay's excellent book thesis, they also have to be cut from external support in the information and financial realms as well - Canada's 2004 ban on LTTE funding deprived the LTTE of over $1m a month. The LTTE misread the change of attitude post-9/11 by failing to distance from ruthless terrorism, resulting in the dismantling of its expatriate funding networks by the US, Canada, and EU. At the same time China became the Sri Lankan government's benefactor, enabling the Army expansion and blocking any UN action against Sri Lanaka for the conduct of its campaign.

Hate to hold back, but my article in JFQ will cover most all of this.

saintsimon (not verified)

Fri, 08/27/2010 - 7:33pm

Seems to me most of you have missed the point of this essay - the point was not to compare Sri Lanka to Afghanistan - the point was to say that if your goal is to win a war than do what you feel is necessary to win it - I think what the author is trying to say here is that America is trying to hedge on winning the war in Afghanistan, thinking rather to manage some kind of resolution that sort of looks like victory but not at the expense of having to kill or piss off too many people - I think what the author is saying is that that is a phony compromise - either do what you have to do to win, which after all is the point of war, or go small ie stop trying to turn Afghanistan into a functioning democracy [an insane notion] and instead form tactical alliances with anti-Taliban tribes in the north, hold the Taliban to a stalemate in the south and then try and separate the malleable Islamists [the author's point was that you can indeed fracture an extremist movement if you apply force effectively] from the hardcore ones - and then once you've isolated the extremists, kill them all... if you can of course.

Bob's World

Fri, 08/27/2010 - 3:35pm


Not to quibble, but the Philippines have been in a constant state of insurgency since at least as long as the Spanish occupation. Sure, various groups have been effectively "defeated", but the conditions of insurgency have only been temporarily suppressed until some new group emerged to pick up once again and ran with it.

Seeing insurgency as being caused by insurgents clouds rational thinking; it is once one realizes it is caused by the governance that one can begin to identify true problems and solutions to the same.

Maybe if we dropped the "Insurgency / Counterinsurgency" tags and replaced them with "Poor Governance / CounterPoor Governance" it would all become more clear as to what is the "action" and what is the "reaction." in this dynamic.

But just as Diogenes wandered about Greece with his lantern in search of an honest man, he would be on just as long a journey if he went in search a government that takes responsibilities for it's short-comings...

Also, as several point out above, whether another Tamil group springs up to replace the LTTE is very much an open question. Add to that the problem with insurgencies on islands (statistically not very successful, because they are easy to isolate) and the Sri Lankan "model" loses its influence.

There's a reason, for example, the Philippines have had great success defeating their insurgent groups, yet a great deal of trouble defeating the insurgency over time. Sri Lanka could be the same.

Unless I missed it I didn't see it noted that Sri Lanka was an internal conflict (the only outside combatants, peacekeepers, were Indians for a short period of time). The U.S. was quite brutal also when it needed to be (our Civil War and the war against the Indians and WWII) when we faced threats to our existance. Dropping a nuclear weapon on Japan didn't address the root causes of the war, but it did stop the fighting. First things first.

In Iraq and Afghanistan our existence is not being challenged, so we are compelled to follow gentlemen's rules of war (although the governments in Afghanistan and Iraq are threatened). I often wonder if our involvement simply makes worse, because we also compell the host nation to take half measures based on the vague concept of legitimacy?

Unfortunately it is not reasonable to think that nations can address all underlying causes of their internal conflicts. One key aspect of legitimacy for a government is the ability to apply force in greater measure and to greater effect than any other internal group.

Joint Force Quarterly # 59, due to be published next week, will contain an article by myself covering some of the same ground as Mr. Beehner's piece above - but with a very different take.

From my reading of events, Sri Lanka's victory had much, much more to do with the economic and physical isolation of the LTTE between 2001 and 2007 than the shift in tactics (brutality) emphasized in most of the literature and papers published thus far (including this one)

The LTTEs collapse was the result of cumulative external and internal forces, not simply the employment of ruthless new tactics. Indeed, there is little beside the ability to disregard Western criticism that distinguishes Sri Lankan tactics or brutality post-2005 from earlier eras, as the conflict was already one of the most violent and ruthless in the world. Critical blows from internal defections, loss of external funding, a global antiterrorist mindset after 9/11, and second order effects of the 2004 tsunami crippled the LTTE. At the same time, foreign aid, domestic politics, and external political cover from China enabled the Sri Lankan government to resume its COIN campaign from a position of strength. The combination of these factors proved decisive in the defeat of the LTTE.

Hopefully I will be able to link the JFQ paper on SWJ in the next week.


Dylan McFadyen

Fri, 08/27/2010 - 1:04pm

The article represents an interesting comparative effort, but there are a number of problems with it. First, as has been pointed out, it's based on a false premise; namely, that someone is of the impression that it's impossible to defeat an insurgency militarily. As Nygdan said it would would only be impossible (or nearly so) for us to do it, because it would involve too much brutality and ignorance of human rights to the extent that would be untenable in our society. At the domestic level, no government could do it without being brought down, and in Afghanistan, the U.S. couldn't do it without losing the support of NATO etc. etc.

Second, there are substantive errors beyond that. An analogue between Sri Lanka and Afghanistan is tenuous in the best of cases. The author references the fact that it's a jungle island where a population 2/3rd as large as Afghanistan's is squeezed into an area one tenth the size - but then practically dismisses these incredibly important differences. More importantly in my mind, he completely fails to point out that while in Sri Lanka, the government contended with a minority population predominantly located in a single part of the country, in Afghanistan NATO is dealing with a plethora of tribal and ethnic groups spread out accross the country, all of which the Taliban can conceivably draw support from. This also means that there isn't a convenient group of people we can toss into camps. That is a monumentally more difficult situation, which goes unaddressed in this article.

Thirdly, the author ignores the historical precedent of Malaya, a situation far more analagous to Afghanistan (albeit with the same minority population issue) where similar solutions were employed, but far more elegantly and sixty years ago. He also fails to emphasize that the Tamil Tigers were obviously far, far better organized, funded, and more centralized than the Taliban, more like the Viet Minh than the Taliban or AQI. This also makes them far more vulnerable to the kind of leadership strikes the government employed (which, of course, we're trying to do as well; if we're not doing them properly, it's still different from not doing them at all).

Another problem is, as CWOT pointed out, the length of the campaign; the author claims that the military solution took only a couple of years when the war itself has been going on for 30. That the final campagin only lasted a couple years is still a far different thing.

There are also a number of contradicting statements in the paper, such as the assertion that holding territory is basically unimportant, followed by the acknowledgement that the climactic campaign took place once the Tigers and their leaders had been squeezed into a stretch of coastline the size of Central Park.

Finally, there were a number of political/non-military elements to the government's victory. Some of these were even mentioned by the author, but were not acknowledged as such, including the turning of rival factions against the primary leadership.

Again, not disagreeing that force can be very effective in combatting insurgents - just look at the Romans - but we can't afford to do stuff like that. If anything the lesson of Sri Lanka in that case would be "don't get involved."

Speaking of Romans: Nygdan, unless I'm mistaken I believe decimation was actually invented as a form of punishment for Roman military units who had displayed cowardice or insubordination.

Swatkind (not verified)

Fri, 08/27/2010 - 11:49am

What shouldn't be left out of the analysis of Sri Lanka's COIN is that Sri Lankan leaders had already reversed and ameliorated many of the political issues that sparked the ethnic conflict in the first place.

In the 50s and 60s the Sri Lankan gov't imposed Sinhalese nationalism on the country and alienated its Tamil minority in the process. The best examples were replacing English with Sinhalese as the official language and effectively disqualifying many Tamils from gov't jobs. SL is officially a "Socialist Democracy", meaning disqualification from gov't jobs was economic disenfranchisement for many Tamils. Then there was also the failed attempt to impose Buddhism as the country's "official religion".

The insurgency and civil war started in the 70s after wide-scale riots and pogroms by Sinhalese nationalists that were tacitly if not openly supported by the SL gov't against minority Tamils.

After the end of the Cold War the SL government made Tamil an official language, liberalized its economy and made restitution to some if not all of Tamil victims. The Tigers only real legitmacy rested on an unending cycle of violence, revenge and distrust connected to the events of the 70s, 80s and 90s but disconnected with the original issues that sparked the civil conflict in the first place....

What will matter is how Tamils and Sinhalese live with each other from here on out. SL is the most educated, and forward looking country in all of South Asia. They have the capacity to become a new Asian Tiger in the way India, Pakistan, Bangladesh and Nepal have consistently failed to...

Haven't we always known this? Insurgencies are almost allways put down by brutual, imperial tactics. In the 21st century, western democracies have utterly rejected those types of actions.
I mean, if we wanted put down the iraqi insurgency as quickly as possibly, we would've kidnapped Sistani and forced him to make fatwas condeming insurgents to permanent hell, captured al-Sadr instead of made truces each time with him and hung him publically, destroyed towns and tribes that resisted, etc. I mean, the Romans came up with the term 'decimate' for when you capture a people, and execute every Tenth man. THAT puts down insurgencies right quick. But we don't WANT to do that, we're not there to CAPTURE Iraq or Afghanistan, and the insurgencies are arguably defeated anyway; they can't mount large attacks, they can't force us from their countries, and their attacks prevent them from participating in the government.

Other things I wonder about is the duration of the Tamil insurgency, which started in the 1970's, the fact that the Tamils became a conventional force and thus became a conventional enemy fixed to terrain, and what the Sri Lankan government learned by decimating the JVP insurrection in the late 1980's.
I'm not sure that the counterinsurgency addressed here was really that short and not just a phase within a 30 year COIN fight.

Bob's World

Fri, 08/27/2010 - 10:05am

Many insurgent movements have been crushed militarily over the centuries. Everyone knows this, and no informed person can really mount a reasonable argument of those facts.

What I don't see, in that same study of history, is where it has ever actually resolved the conditions of insurgency that gave rise to the afore mentioned militarily defeated group. Suppression of insurgency can indeed stop violence for some period of time, so long as one keeps their boot firmly on the throat of the populace. We saw and see this in so many countries. Tito kept his boot firmly in place and stabilized one of the most unstable regions of the world. We all know what happened when his boot came off. Similarly, the Saudis keep a heavy boot on the throat of their populace. That does not mean there is not insurgency or an insurgency is defeated, it just means it has been suppressed.

The question for Sri Lanka is "what will the government do with the breathing room that their military has earned for them?" Will they celebrate the "victory" as this author does, and not address the conditions of insurgency in the Tamil people born of their perceptions of poor governance? Or, will they recognize this for what it is, and make the necessary changes to ensure that the entire populace feels adequately served by the government. Time will tell, but it is far too soon to pop the corks just yet.

John T. Fishel

Fri, 08/27/2010 - 8:38am

What Sri Lanka actually proves is (1)that not all insurgencies are alike; (2) that what Kilcullen calls an "enemy centric" strategy is sometimes wholly appropriate; and (3) a bit farther afield, that any good "population centric" strategy includes significant enemy centric components. Like the American Civil War, the Tamil insurgency in Sri Lanka was secessionist with a clear demarcation between territory controlled by the insurgents and that controlled by the government. This simple fact gave the govt forces a territorial target against which to apply massive force.

Marzipan (not verified)

Fri, 08/27/2010 - 9:06am

The civil war in Sri Lanka was a different for many reasons, but the impact of the terrain should not be underestimated. Tamil Tiger's safe havens were vulnerable because they were withing the geopolitical boundaries of Sri Lanka and lacked the ability to withdraw across the national boarder to safe havens in another country. This, as well as the points Mr Fishel points out above, meant the insurgency was more vulnerable to being resolved via a militarty means.

Having said that though, their were successful efforts made on the part of the Sri Lankan government to recruit break away factions of the Tamil Tigers. Afghanistan is a different kettle of fish and I would hesitate to adopt a more kinetic approach given that the Soviet's killed a million people or so and still failed to win their counter insurgency campaign.