Assessing the COMISAF COIN Guidance

I’ve been writing in several blog threads my concerns about McChrystal’s new guidance document, most notably at AM: COMISAF COIN Guidance Released | Center for a New American Security.

But it is probably worthwhile to summarize my views:

(1) The guidance seems like a perfectly reasonable set of principles for any army operating in a foreign country where the goal is trying to help support the local government.  It is a Miss Manners for soldiers.  And I don’t say that dismissively.  The reason we need Miss Manners is precisely because it is quite easy to behave boorishly without knowing it, and then to justify ones conduct as a matter of contingent necessity.  The guidance, in short, provides an important warning on the conditions under with “military necessity” can be invoked.  So far so good.

(2) The problem with the guidance — well, really more with 3-24 than the guidance itself which is just a well-written consequence of adopting 3-24 — is that it makes a ton of assumptions about what constitutes an effective “theory of victory” in counter-insurgency operations.  At the core, the conceptual foundation of discussions of “legitimacy” is thin.  Razor thin.  And indeed, at odds with most historical research on the topic.

(3) The document implies that legitimacy flows from the provision of goods and services — primarily physical security, but also various local projects.  It is, in short, a profoundly materialist mechanic being drafted in service of an inherently ideational dynamic.  In reality there are multiple pathways to state legitimacy — and while the provision of goods and services many play a role, an equally large role is played by shared historical myths, use of force, the existence of external threats, ritual and religion, the role of charismatic leaders, the institutionalization of mechanisms for allocation of resources, and on and on and on.  And while legitimacy may come from any of those, so can illegitimacy.

(4) Indeed, the greater the need for materialist payoffs, the less the actual level of legitimacy.  This issue is complicated to disentangle, and our ability (as Americans) to address it is constrained by the Contractarianism at the core of American political theory.  But think about a contract… does it imply to you any grant of authority or consent to rule?  Contracts are made, essentially, by equal parties, though in exchange for goods one might choose to place oneself in a subordinate position (say, an employment contract).  But by doing so, I am not accepting the “legitimacy” of my boss.  I am simply engaged in a quid pro quo.  The need to make payments, in short, is not a sign of legitimacy, but rather a source of concern.

(5) There is also reason to be skeptical of the security/legitimacy connection, for two reasons.  First, because I think that as a matter of empirics, the correlation is not strong.  Japan 1945… massive insecurity, massive legitimacy.  Southern United States 1861, close to perfect security, strong perception of illegitimacy of the federal government.  How do we explain cases like that?  Second, “security” is conceived too narrowly in the guidance and 3-24.  I think history is replete with examples — including in Rwanda in 1994 for instance — where “insecurity” has been “constructed” by powerful elites in order to maintain control and in some cases initiate strategic dispossessive conflicts. So, in Afghanistan, we’re tending — I think — to define “security” as stuff not going boom. But what are the “security” implications of a powerful Afghan National Army controlled by a non-Pashtun regime in Kabul? How does that goal play into the strategies of Taliban leaders who are clearly playing on fears of being power being institutionalized outside their control? Okay, I’ll stop with the questions. My guess — and I could be very, very wrong — is that in Pashtun communities, insecurity is going to be more durably associated with potential domination from Kabul than fear of immediate violence.

(6) As problematic as is the legitimacy/security link, worse is the legitimacy/insurgency link.  First, it is true that as a definitional matter an insurgency implies some degree of illegitimacy — at least as perceived by the insurgents.  Though, even if this case, we need to be careful of context.  An insurgency that seeks to replace a central government is different from one that seeks to secede.  Why?  Because the former implicitly accepts the legitimacy of the state, but doubts the legitimacy of the leaders, while the latter may actually accept the legitimacy of the leaders in a way, but not accept the legitimacy of the state.  In short, neither insurgency nor legitimacy are dichotomous, nor are they even measured on a single axis.  Second, while it may be true that lack of legitimacy is a permissive or even necessary condition for insurgency, it is clearly not a sufficient condition.  So, if you want to create a grand theory of COIN, then it is inevitably tempting to focus on illegitimacy, although actually if there are a relatively limited number of other intervening conditions that are collectively exhaustive, it might be possible to build an equally compelling typological theory.   But be that as it may.  The focus on restoring legitimacy is convenient because we know — or think we know — that this factor will be present in all insurgencies.  But as a matter of strategy, once we accept that illegitimacy is a necessary but not sufficient cause then it opens up the possible of addressing one of the other causes in a particular case.  In short, yes, fixing the legitimacy gap may work.  But so might other approaches, at least some of which may be more tractable in any given case.

(7) Finally… in both the guidance and 3-24, the insurgency-legitimacy connection forms a closed intellectual loop that allows for neither falsification nor adjustment.  How do we know that illegitimacy is a problem?  Because there is an insurgency.  So if I take steps to increase the legitimacy of the government, and the insurgency persists, that just means I haven’t done enough to make the government legitimate.  There is, in short, no mechanism to test the assumption.  All we can do is plug away until either (a) the insurgency ends or (b) we get tired and go home.  It is, in short, a closed conceptual system, immune to adjustment except at the tactical level.

Anyway, short version — a well-written and thoughtful document, but one that rests uneasily atop a very problematic foundation.

3 comments to Assessing the COMISAF COIN Guidance

  • I hope to deal with this (and your SWJ post) in greater depth later, but briefly: do you have any commentary on Stathis Kalyvas’ The Logic of Violence in Civil War, which basically (as I understand it) asserts the primacy of instrumental violence related to security (or “legitimacy” in your formulation, or the proximity of capable authority) over other forms of identity- and grievance-based violence?

  • I need to read it. I’ve only gotten second hand summaries of the argument. Should be arriving from Amazon tomorrow.

  • This is going to sound a little over-the-top, and I know I’m probably parroting the “dominant narrative of the COIN crowd” (to paraphrase COL Gentile), but I think it’s the most important book I’ve read on the subject of what animates insurgencies. (In light of our recent exchange about conceptual grouping and abstraction, I’ll caveat that with ‘…to the degree that we can generalize about “insurgency”‘).

Leave a Reply