COIN of the Realm

COIN of the Realm

Quote:

---

Colin Kahl, an Assistant Professor in the Security Studies Program at Georgetown University and a Fellow at the Center for a New American Security, writes in the November / December issue of Foreign Affairs on while Counterinsurgency – US Army Field Manual 3-24 / Marine Corps Warfighting Publication 33.3.5 is a long overdue step in the right direction, a look back at the history of counterinsurgency offers a sobering reminder of how low the odds of success are regardless of doctrinal excellence...

---

It is a simple fact, but one that is most often overlooked, that insurgencies generally fail. Most fail before they ever become full blown insurgencies or soon thereafter. Consider the American case of the Symbionese Liberation Army which we saw as a police problem and defeated as such. Or consider as well the case of Che Guevara in Bolivia. I happened to arrive in country the day Che ambushed his first Bolivian Army patrol. It was evident, even before anyone knew it was Che, that this insurgency was going to fail, as indeed, it did in a matter of months. ( I should add that the short time span is anomalous for any insrgency that gets through the early organizational phase - witness Malaya.) In the 43 cases of insurgency between 1945 and 1980 that Max Manwaring and I studied, the majority (23) were victories for the COIN side. In Latin America, according to Timothy Wickham-Crowley (who is sympathetic to the insurgents) nearly all insurgencies were defeated. Indeed, the only successful insurgencies in the region were Cuba, Nicaragua, and Bolivia in 1952. (The Contra insurgency had an ambiguous outcome but the Sandinistas were overthrown when forced into free elections.) What Max and I demonstrateed in the quantitative study ("Insurgency & counterinsurgency: Toward a New Analytical Approach" in Small Wars & Insurgencies Winter 1992) and subsequent books including our recent Uncomfortable Wars Revisited (U. of Oklahoma Press, 2006) is that the approach advocated by the new COIN manual is likely to be successful. Neither we, nor the COIN manual eschew targeted violence. Indeed, we argue very strongly that an organized insurgent enemy must be destroyed by violent means whether by military, police, or paramilitary power (most likely a combination). But we insist on legitimacy and what Kilcullen calls a population centric strategy. So, I would conclude that successful COIN is best conducted before the insurgency gets effectively organized with a strong population centric strategy to make sure that what was killed stays dead. (Otherwise, we will see the Phoenix rising from the ashes again and again as it did in Guatemala for 3 generations of insurgents.) If COIN begins after the insurgency is organized, then the COIN operators need to be prepared for a long war.

Cheers

JohnT

Mr. Fishel - interesting post. My personal contention is that the conditions, situations, and relevant social, political, and economic histories of each insurgency are so different (or at least different enough) that it's hard to assess the chances of an insurgency or counterinsurgency effort based on aggregate numbers. It's correlation vs causality, like that study mechanized forces being less successful in COIN operations that Cavguy and other tankers (rightly) did not agree with. The correlation does not prove causality - there are too many circumstances that make each instance unique.

Perhaps in the "counterinsurgent usually prevails" theory, a number of cases where an insurgency never really got off the ground or was easy prey to the government (bad conditions, poor leadership, etc.) causes the aggregate to favor the counterinsurgent? Or in Mr. Kahl's piece, the odds of success against an insurgency as developed, deadly, and in such a political situation are historically small. Either could be right or wrong, given the inherent differences in each case.

What I think Mr. Kahl does a good job of is cautioning the military about adopting the "cookie cutter" approach - essentially hammering the doctrine in and then thinking we have the magic bullet. Each case has been and will be fundamentally different from what preceded it, and while the principles of insurgency or COIN may remain "permanent and unchanging," adopting a standard and predictable response into as fluid an environment as an insurgency would be a mistake.

Matt

From this "armchair" Northern Ireland is too close to ignore, even if peace reigns with a few discontented on the fringes using violence.

Predating 'The Troubles' was the 'Border Campaign' that was contained and defeated (rarely commented upon). Then in 1969 'The Troubles' began, after a few months of disorder the British Army were called in to support the probincial government. All manner of local options and those the Army had learnt in the twlight of Empire were tried - notably detention without trial aka internment. Violence went up, along with a growth in minority support for the Nationalist / IRA insurgency. Out went provincial government, in 1972 and only restored recently - with two opponents in coalition, Ian Paisley's DUP and Gerry Adam's Provisional Sein Fein.

Yes, eventually we learnt what did not work and what did work (alas rarely codified). What a price to pay, principally in Northern Ireland.

Once the "men of violence" are let out they rarely go back. Political violence becomes more like criminal gangs, gangsters and intimidation widespread.

Each insurgency is different. All I'd say is stop it before the violence starts. Step forward brave souls with the intelligence and politicians who listen.

davidbfpo

Matt--

I agree completely that all insurgencies are different. Max and I stress that what we call the SWORD Model is NOT a cookie cutter but rather should be thought of in terms of certain factors that play out in all insurgencies (often in different ways depending on the various antagonists). The short version is that he who ignores any of the 7 dimensions we identify does so at his peril.

In terms of your argument regarding theory: It is not theoretical that most insurgencies fail; that is a matter of fact. Theory involves the rationale for that fact. In this case, you hypothesize that the explanation for the fact that most insurgencies fail is that they are defeated in their early stages. I suspect that your hypothesis would add significantly to the numbers of failed insurgencies but my assertion comes from our sample of 43 cases which are mainly cases of relatively long wars. But a definitive answer requires more empirical research - not a bad project for any student interested in insurgency.;)

David--

N. Ireland is a particularly interesting case. I was the US Army's current intel analyst for W. Europe in 1969 when the "Troubles" broke out. My reaction then was that I was glad that it was a Brit problem rather than ours!:wry: If the peace that has finally been worked out holds, then it will support the argument that most insurgencies fail as well as arguments in favor of a population centric approach. I would also argue that the British government and military ended up giving due attention to all of the 7 seven dimensions that Max and I identified in our article (and it still took over 30 years to terminate the insurgency on terms favorable to the counterinsurgents).

Matt--

Just to add a bit, after the original article/study all our subsequent work has been qualitative rather than quantitative. we have applied the model to counterdrug operations, peace operations, and other contingency operations with quite a bit of success. But, I will stress again that we don't see it (the approach) as being a cookie cutter.

Cheers

JohnT

Quote:

---

Originally Posted by John T. Fishel

But we insist on legitimacy

---

How many wars did you study were regime change and COIN were attempted at the same time? (It seems to me that creating a new government might make it much harder to establish legitimacy.)

Hi Rank--

You make an interesting point. But I'm not sure I would phrase it the same way. First, I'd have to go back to the case list to see how many fit the regime change/insurgency criteria. Clearly, most do not. Second, is there a factor that is common to thses criteria that appears more often? I think there is and that is found in cases where an insurgency takes place against any newly established government. In that case, I would agree with your semi-explicit hypothesis that COIN is harder because, among other reasons, legitimacy must be established where none existed for that government previously while it is in the process of simply learning to govern. A few cases with different results are: Spain in the Napoleonic wars, South Vietnam from 1954 to 1957, El Salvador from 1979 to 1992, and Nicaragua from 1979 to 1992. As these cases show, difficult is hardly impossible.

Cheers

JohnT

Most people call me RA, but you can call me anything you want.;)

Is the "winner" always the side that establishes legitimacy? (The winning insurgencies that I know about seemed to be facing governments that had legitimacy problems: Batista was corrupt and mobbed up, many Vietnamese saw the South Vietnamese government as seen as a US puppet.) When both sides are legitimate, does it always end in a political settlement that accommodates both sides? (N. Ireland.)

If so, that would seem to simplify the strategy and perhaps place limits on our abilities to influence the political situation.

Quote:

---

Originally Posted by Rank amateur

Is the "winner" always the side that establishes legitimacy? (The winning insurgencies that I know about seemed to be facing governments that had legitimacy problems: Batista was corrupt and mobbed up, many Vietnamese saw the South Vietnamese government as seen as a US puppet.) When both sides are legitimate, does it always end in a political settlement that accommodates both sides? (N. Ireland.)

If so, that would seem to simplify the strategy and perhaps place limits on our abilities to influence the political situation.

---

Given that insurgency/COIN is inherently a political contest, legitimacy is a critical objective. Defeating a political force viewed as legitimate will require overwhelming suppression - think of most of the Latin American wars of the 70s and 80s.

When both sides are legitimate, I think that has to make a clear-cut "victory" (as much as such a thing exists in a COIN struggle) of one side nearly impossible to achieve.

Matt

Quote:

---

Originally Posted by John T. Fishel

Matt--

In terms of your argument regarding theory: It is not theoretical that most insurgencies fail; that is a matter of fact. Theory involves the rationale for that fact. In this case, you hypothesize that the explanation for the fact that most insurgencies fail is that they are defeated in their early stages. I suspect that your hypothesis would add significantly to the numbers of failed insurgencies but my assertion comes from our sample of 43 cases which are mainly cases of relatively long wars. But a definitive answer requires more empirical research - not a bad project for any student interested in insurgency.;)


Just to add a bit, after the original article/study all our subsequent work has been qualitative rather than quantitative. we have applied the model to counterdrug operations, peace operations, and other contingency operations with quite a bit of success. But, I will stress again that we don't see it (the approach) as being a cookie cutter.

Cheers

JohnT

---

Mr. Fishel,

Thanks. The more I thought about it, the more I realized I was qualifying your data to the point where I was essentially saying, "well, if you leave out all but the really effective and tenacious insurgencies, you won't get that result."

To which, of course, the answer is, "well, duh."

And I don't mean to suggest YOUR suggestions are in any way a cookie cutter approach. My caution was directed more towards becoming doctrinally rigid and inflexible - a month or so ago, there was a long thread on adaptation, and the need to constantly evolve our tactics as the insurgents adapt to us. In any large organization, doctrine can become gospel to the rank-and-file, and can suppress initiative and adaptation. That was more what I meant.

Occasionally thinking before posting would probably be a decent idea, I guess. :o

Matt

RA, I've always believed in calling people what they wish to be called:D
Matt, I've never been much on titles and you aren't one of my students or a junior officer working for me, so, I mostly go by John (or the JohnT handle).:)

While legitimacy is an extremely important dimension (and one of the 2 that were individually the most statistically significant) the essence of the research was that the whole (all 7 dimensions) were greater than the sum of the parts. Moreover, legitimacy is not an all or nothing dimension. It is made up of multiple variables each of which changes over time. I also break it down into 2 large components: system legitimacy and regime legitimacy. The US has system legitimacy because of the way our Constitution and form of government are perceived by the people. The Bush Administration has gone from relatively low regime legitimacy in January 2001 to extremely high legitmacy between the defeat of the Taliban and the defeat of Saddam to the low in the polls of today. (To complicate it further there is also legitimacy de jure and de facto:eek:) So, in the case of 2 opponents, both having legitimacy, id depends on (1) the other dimensions and (2) the relative legitimacy of the regime, system, de jure, and de facto at that time.

Cheers

JohnT