September 18, 2009

So how did it come to this?

(Another attempt at a longer explanation of why we're where we are in Afghanistan. Thoughts welcome.)

Military mentoring, as practiced both by NATO OMLTs or their American equivalents, the Embedded Transition Teams (ETTs) remains poorly understood states-side. The mentor experience also varies greatly depending on where in the country and military hierarchy your Afghans are.

At an embattled outpost in Zhari District, it could feel like a bond akin to brotherhood, formed under fire; at the brigade-level or higher, where the job consists of poking into every detail an Afghan headquarters, looking for inefficiencies, communications breakdowns, and opportunities for improvement, mentor teams in some ways more closely resemble a determined if friendly group of management consultants.

Like consultants, our ability to effect change was tightly circumscribed, and our ability to influence limited both by our own personal diplomatic skills, and our counterparts' willingness to change. Often times, we found it wasn't the Afghans who needed to change, either: our own countrymen and their NATO colleagues would often prove the more inflexible party... like when security force development was set aside to action some short-term objective or opportunity. Again and again, explaining the situation to the Afghans, gaining their consent, or leaving them to chase the opportunity in question themselves on their own, would be evaluated, considered, and rejected.

Sometimes operational security was cited as a reason, but it wasn't really: you don't have to spend much time with Afghan soldiers to have faith in their personal commitment to winning. One suspects the real reason all too often was that Afghan soldiers themselves really didn't add much to our own forces in the way of added capability. And to tell the truth, they really didn't, if all we're talking about is the kinetic fight: when hunting for an enemy that primarily uses IEDs to attack you, adding a few dozen more riflemen with dubious marksmanship skills and no reliable logistics of their own is hardly guaranteed to simplify your problems. And once you got ANA majors and lieutenant colonels into the mix around the planning table, any given operation had introduced into it a level of... uncertainty, to be polite about it, that any Western-trained military planners would be likely to have issues with. This is why, in Afghanistan right now, there are almost no districts outside of the capital that are recognized as having an Afghan lead in their security.

We've all seen those tidy military diagrams with boundaries between battalions, regiments and the like, from World War 2, or Vietnam. One of the most crucial things any officer wants to understand is where the boundary line is between his unit and the ones on his flanks. But one complication of the operating environment in Afghanistan right now, there are at least two such diagrams for any area of operations, with 100% overlap. We have two separate "overlays" or "traces", back in the days of physical maps: the Western military laydown, and the Afghan laydown, superimposed. An Afghan battalion (kandak) does not take up a position to the flank of a Western counterpart. Any given provincial district or group of districts will have both a Western company-sized element responsible for it, and an Afghan kandak responsible for it as well.

Drawing a line down the middle, and saying, "you be responsible for this half" has simply not been seen as possible. First, very few if any ANA kandaks have the organizational ability to take on that kind of independent responsibility, even with Western mentors, artillery support, etc. tacked on. Two, the Western half would be completely non-functional as well, because there would be no "Afghan face:" no compound searches, no vehicle searches, no realistic prospect of tactical intelligence. But it's chicken-and-egg: the lack of opportunity to develop any real independent security responsibility means that those Afghan leaders can never really improve, either.

This whole military approach seems both dysfunctional and unsustainable. We start by spending our own money to the tune of some multiple of Afghanistan's entire Gross Domestic Product to fund an Afghan army (left on its own, Afghanistan couldn't support an army a tenth the planned size). We then spend another astronomical sum to put our own armies in the combat zone, incurring our own casualties, because the native army is manifestly unable of doing the job on its own. Meanwhile, the insurgency, spending a dollar for every $200 or so we are spending, continues to thrive and exact a steady toll in casualties, all the while keeping the population effectively cowed. How did it come to this?

Looking back (see graph, above), one can see now mid-2005 was the inflection point.

As the graph shows quite clearly, after three years of near-quiescence, the insurgency in Afghanistan returned in strength, following the successful Western-supervised election confirming the Karzai presidency. In NATO, there was growing concern, however, that their underresourced effort to support the relegimitized government ended in practice at the outskirts of the capital, Kabul, and was having little effect beyond it. At the same time, an American military conducting its own terror-suppression, economy-of-force efforts in the South and East of the country, was eager to have Kabul's writ, and that of the International Security Assistance Force that supported the Afghan government, extended, particularly to the south of the country. In part, this has to be seen as an effort to free up some of its own resources, then desperately needed in the war in Iraq, as well. But there was also strong Afghan and international pressure to push the sphere of development beyond the capital.

The then-novel tool used by both Americans and their ISAF allies to address this issue was the Provincial Reconstruction Team, a node of military expertise emplaced in a provincial capital, and intended to develop governance and security in their immediate area with support from civilian and non-governmental agencies. PRTs were drawn from multiple nations, and often had different approaches to the task at hand. But it became clear almost immediately that, particularly in the south, the PRT as an organization would be barely capable of defending itself without help, let alone extend any kind of security umbrella to anyone else. That meant that army battalions would need to be deployed to support that work, as well. And the Afghan battalions weren't ready yet. Which meant Western battalions started pushing out to help their own PRTs. So what you saw in Helmand and Kandahar provinces, still the insurgents' battlefield of choice, was a British or Canadian battalion AND a PRT pushing out to relieve a harried U.S. battalion and taking over lead security responsibility for that province.

There was little if any discussion at this point about the road that was NOT being followed, that being only pushing development as far out as the Afghan security forces themselves could defend the "ink spot" (which to be fair, at this point would not have been very far). There were a couple reasons for this: the first being that the Afghan government itself was entirely unwilling to see Kandahar City or the Helmand Valley fall overtly under insurgent sway, which was certainly the most likely consequence of relying on their own resources at this point. And the fight the Canadians and British subsequently found themselves in the south in the summer of 2006, faced with literally thousands of dedicated insurgents, showed without doubt that the insurgency, had Western battalions tried to withdraw completely from this area, easily would have overrun the country's second-largest city at that point. No, the alternative to Western forces in 2006 would have been outright insurgent control over a significant swath of the country: a full-on insurrection rather than an insurgency, in other words.

But that meant that, starting at the end of 2005 and the start of 2006, military mentoring of the Afghan army began playing a deadly game of catch-up, in the middle of what was already a medium-tempo war in much of the country: pushing the local troops out, regardless of readiness or capability, to be the "Afghan face," with the hope that "on-the-job" training and colocation with vastly more powerful and capable Western forces would enable them to correct their shortcomings. In practice, of course, it largely obscured, or exacerbated them.

Up to this point, the military advisory efforts had actually been something of a qualified success judged within their own lights, as opposed to, say, the efforts at rebuilding the police, now recognized as wholly disastrous. The army at the time was having little difficulty meeting its recruiting quotas, and the first Afghan soldiers deployed to actual battlefields, after being trained in the Kabul area, performed creditably enough. It needs to be recognized, also, that the military efforts up to this point had been quite successful in the DDR (Disarmament, Demobilization and Reintegration) role: providing sinecures and stability to the military leadership of the Northern Alliance and other non-Taliban warrior types within a new model army had undoubtedly contributed to the previous three years of relative peace and governmental stability.

The onset of a robust and active anti-government resistance, however, meant that the terms of reference of all our efforts to help Afghans had profoundly changed, and it's fair to say everyone involved was slow in seeing how much. Hindsight shows that the Afghan army and nation were now in a fight for their literal lives: a lot of the "nice-to-haves" in Afghan development that had been tolerable before 2006 probably needed to be jettisoned at this point, but there was no one in a position to make that case. For instance, substantial efforts continued to be poured into training local police forces as a criminal investigatory arm, rather than what they now would have to become in a warzone, a supportive and flexible paramilitary, whose efforts and the Afghan army's would have to be closely interconnected: hundreds of poorly armed police in shoddily defended police stations have since paid the price for those distractions with their lives.

This was also true with the military advisory effort. It's not a complete oversimplification to suggest that up to 2005 our efforts were focussed on training the army how NOT to fight, in the sense of limiting their activities to what was requested of them by the central government. The goal at that point had been to restrain warlordism, and impose a system of restraints, using the army's structure, on all those in society who were adept at resolving their problems with violence. But having done that, we found ourselves with a recreated national armed force with a high sense of self-entitlement and institutional pride, but no clear idea at any level how to defeat the insurgents in Kandahar on its own. Suddenly we needed the old Northern Alliance army back, the one that had rolled across the country with the help of a few embedded Special Forces personnel with B-1 bombers at their disposal, but neither we nor the Afghans themselves had any idea how to recreate it.

Military advisory efforts are part and parcel of counterinsurgency, and will likely remain with us for some time. Our counterinsurgency manuals define victory as the point where the fight can be handed off to reliable indigenous forces; that implies a significant effort in their creation each time this kind of war is fought. But the experience in both Afghanistan and Iraq to date can hardly instill great confidence in anyone that the process is a simple or well-understood one.

Posted by BruceR at 12:14 PM

More on Afghan army literacy

Everything Tom Ricks says here about the Afghan army is right. It is also beside the point. Afghan privates don't need to be literate: agreed. But at some level if you're colocated on the same battlefield you need to be able to communicate with each other. That requires some commonality of language and training at the leadership level.

Historically, armies have done this with their auxiliary corps, like the British in India, or even the Americans in Korea, by embedding Western officers as the leaders of the auxiliary units. In recent history, that's been replaced with the idea of mentors, who ride along with the native leaders, but don't exercise any control. It is, whatever its other virtues, a less effective way of doing the same thing.

Do we, as Ricks says, "lack the imagination and historical knowledge to train him to be an Afghan, instead of an imitation American soldier?" No denying it: it's a valid criticism, sometimes even an accurate one.

But the best quote honours go to one of the commenters, talking about why we have yet to cede security responsibility of any part of the country to Afghans themselves: "Nor are we going to give the ANSF an ops box and say 'right - go commando and knock heads your way,' because culturally, western forces can't abide that kind of uncertainty in the plan. Not only do we not have any real Lawrences around today, we're pretty short on Allenbys too."

UPDATE: By the way, the reason Afghans can't read maps isn't illiteracy, primarily. It's because until very recently there have been no topo maps of Afghanistan in a language they can read, and ours are inscrutable to them. It's one of those things we forgot to do early on and we're playing catch-up on, now. The first NGA-produced maps started coming off the line in early 2007 and even five months ago almost none had made it through the Afghan logistics chain to their soldiers and officers.

UPDATE #2: Ricks' historical examples are, of course, ludicrous, btw. The 300 Spartiates at Thermopylae were the elite of their society: they could almost certainly read better than the average Greek of their time (and, it should go without saying, they *lost*). Actually, the real lesson from the Greek wars is probably the opposite, showing the dangers of not having interoperability: the Persians, who had a polyglot, multilingual army, were consistently less effective than the Greeks, who more or less all understood each other. And the Gurkhas, as noted in the same comment above, had British officers, and in their early decades, British NCOs. Makes kind of a difference in interoperability terms.

Posted by BruceR at 09:36 AM

Quote of the day

Ackerman comes closer than I have yet to articulating my personal current position, which I hope will be seen as the responsible one, on foreign interventions by Western powers:

The right move in Somalia and Yemen is not to get into Somalia or Yemen. The right move in Afghanistan is to create the conditions for responsible extrication from the security piece.

Here's how I'd have said it: neither "cut and run" nor "as long as it takes" are responsible international policy positions for the Afghan problem (with respectful apologies in the Canadian context to Sen. Kenny and Gen. (retd.) Mackenzie, respectively). What else have you got?

Posted by BruceR at 09:09 AM