Is ‘counterinsurgency’ an empty concept? Can we do better?

A recent conversation with a colleague about the nature and definition of counterinsurgency got me thinking, ‘what is it, really’? Readers will be familiar with the old doctrinal ‘definition’: ‘those military, paramilitary, political, economic, psychological, and civic actions taken by a government to defeat insurgency’. This definition is certainly broad enough, failing in fact to exclude any action ostensibly taken to counter an insurgency from its remit. By definition, therefore, counterinsurgency has no inherent form or substance; it is simply the label used to describe any operation intended to defeat irregular forces. Even the effectiveness of such action is irrelevant to the use of the term: by definition, ‘counterinsurgency’ is about the intention to defeat the insurgency, not the actual success in doing so.

This begs the question of whether the notional distinction between counter-terrorism and counterinsurgency has any merit. It also makes the ‘conventional’ versus ‘ counterinsurgency’ distinction dubious as either approach, if ‘taken by a government to defeat insurgency’, is by definition ‘counterinsurgency’.

Maybe all of this is unnecessary: we all know what we mean when we say ‘counterinsurgency’, right? The reason there is no better definition is because counterinsurgency is a fluid concept that is dependent on circumstance. This seems to be the line taken by David Kilcullen in The Accidental Guerrilla, a book that sets out splendidly what counterinsurgency is. Kilcullen writes:

As we all know, there is no such thing as a “standard” counterinsurgency… the set of counterinsurgency measures adopted depends on the character of the insurgency: the nature of counterinsurgency is not fixed, but shifting; it evolves in response to the changes in the form of insurgency. This means that there is no standard set of metrics, benchmarks, or operational techniques that apply to all insurgencies or remain valid for any single insurgency through its life-cycle. And there are no fixed “laws” of counterinsurgency, except for the sole and simple but difficult requirement to first understand the environment, then diagnose the problem, in detail and in its own terms, and then build a tailored set of situation-specific techniques to deal with it.

There are a few things to comment on there. First, there is this initial formulation ‘as we all know’. Possibly nothing is meant of it here, but it does points to a central problem with the way we use the term ‘counterinsurgency’. We use the term loosely to describe something that we all recognise as such, more or less. So the term is a useful shorthand term to describe something so difficult to define that we prefer not to try — and until proved wrong, we assume we all mean the same thing.

That heuristic consensus, the use of shorthand to avoid talking details, is fine among members of this community, but how to respond to the legitimate questions of outsiders, asking for a definition or for a clearer sense of what this counterinsurgency thing is really about?

This brings us to a second point: are there really no constant, defining characteristics of counterinsurgency? Kilcullen talks of the ‘laws’ of counterinsurgency, but mentions only the importance of understanding and responding to local circumstances. If that is the sole ‘law of counterinsurgency’, does any behaviour prompted by such an understanding invariably constitute counterinsurgency? Were Wehrmacht operations in occupied France counterinsurgency? Was the strategy of annihilation adopted by Syria’s former President Hafez Assad against the Muslim Brotherhood in 1982 ‘counterinsurgency’, and if not, why not? Would it be because they did not, in our estimation, have a good enough understanding of local circumstances, or are there other implicit (and more important) distinctions at play?

This conundrum is what gets us into terms like ‘population-centered counterinsurgency’ or ‘enemy-centered counterinsurgency’, to distinguish what we want counterinsurgency to mean from the meaning that others may ascribe to it. Analytically speaking, these are awkward and highly unsatisfying. Similarly, there is a temptation to talk of ‘good counterinsurgency’ or ‘bad counterinsurgency’ so as to differentiate between those who ‘get it’ and those who do not. The vagueness inherent to these statements is testament to the lack of specificity in our understanding of what counterinsurgency really is.

Of course no one owns the definition of ‘counterinsurgency’ and no one ever will. But because the term may legitimately mean completely different things to different people, be operationalised in a wide variety of ways and still be ‘counterinsurgency’, perhaps it is a term with little utility. Its utility, as noted above, lies in its use as shorthand for something we think we all agree on, but once we get down to specifics, I fear it may be a vacuous concept, devoid of any substance or intrinsic meaning.

If this is a problem, one potential solution may be to laden the definition of ‘counterinsurgency’ with more content. One way of achieving greater precision may be to focus on the concept of ‘legitimacy’. I thought that the old definition of irregular warfare as ‘a violent struggle among state and nonstate actors for legitimacy and influence over the relevant populations’ was quite good and may in fact be particularly relevant to counterinsurgency. Ah, but what’s ‘legitimacy’ — well it would need to be defined but there is also no need to be overly churlish.

Or is it better to drop ‘counterinsurgency’ altogether. I suspect this may be the case, but find myself unable to abandon what is such a convenient shorthand. Still, the term is clearly is not necessary to the discussion of war-to-peace transitions: witness Mats Berdal’s recent and excellent book on Building Peace After War, which mentions counterinsurgency but once and nonetheless says more good things on this topic than many books with ‘counterinsurgency’ in their title.

Then again, his book also features a fairly lengthy bit on ‘scope and definitions’, necessary because of its use of the term ‘post-conflict peace-buildling’, which is not without its own foibles. So it would seem whatever term we resort to presents its own ambiguity and imperfections. It is Friday afternoon so I will need to resume these idle musings at a later stage. In the meantime, I take some comfort from the final paragraph in Mats Berdal’s introductory chapter:

It should be evident already that the use of the term ‘post-conflict’ to describe the kinds of operational settings and challenges explored in this book is strictly misleading. The term appears destined to stay, however, and this is in part why no attempt has been made to replace it here. That said, a degree of terminological inexactitude is unavoidable in dealing with this subject, and historian Hugh Seton-Watson’s exculpatory plea in the introduction to one of his works seems appropriate here as well. Acknowledging that the effort to make sense of his chosen subject ‘undoubtedly lacks neatness’, he adds that this is ‘inevitable because the subject itself is not neat’.

Originally published at kingsofwar.org.uk on March 26, 2010.