Flit

The most important piece on the Afghan army you'll ever read

The same day I ding his blog, Dave Anderson at Newshoggers points me to something absolutely brilliant.

I cannot recommend this piece, by USMC Col (retd.) Haynes on how to fix Afghan army mentoring, highly enough. It's a brilliant recap of everything the good mentors have been saying in public and in private about the ANA. Everything he says, pro and con, could have been said about the Afghans I worked with in a different corps in a different part of the country. It's pieces like this that keep my hope alive for this mission: for we really are learning how to do this right, bit by precious bit.

I sincerely hope that when people are looking at operationalizing the President's undoubtedly fine words to come tomorrow night about building up Afghan military capacity, in order that the West can leave, they think of this piece. Like so many other good proposals, it assumes a degree of Afghan government cooperation with the plan that has not often been in evidence to date, but other than that, I only wish I'd written it. If we all fail in the end, it will be because advice this good wasn't followed in the end:

To get more out of the ANA and decrease the requirement for additional US troops, we urgently need a construct that creates a sense of ANA responsibility and accountability of the security situation in a given area. ANA culture is such that appearance and face saving are strong factors in a leader’s decision-making process. The desire to not “look bad” permeates everything they do. It is pointless to try to change this aspect of their culture, so we must exploit it. Couple their face-saving tendencies with the lack of a clearly articulated ISAF security transition plan and you get the lack of initiative and failure to take risks that we often see in the ANA. Overlay the Soviet background of many of the officers, and you have today’s ANA leadership deficiencies.

Multi-National Force-West’s 2007 Memorandum of Understanding in Al Anbar province with the Iraqi Army (IA) is a successful security transition model worthy of close examination. This agreement defined relationships between coalition and Iraqi units with an eye always toward handing over areas of operational responsibility to the Iraqi Army. Once capable, an IA unit was released from the Tactical Control of the coalition and given responsibility for a specific area of operations (AO) while MNF-W units remained in overwatch as required by the threat and the competence of the Iraqi unit. The purpose was always to take the “training wheels” off as early as possible while maintaining a watchful eye to prevent catastrophe. Embedded advisors provided the linkage to in-extremis US support. Further, MNF-W and the Iraqi Army had the option to establish a Combined Operating Area “when both 1st IA Division and MNF-W ground forces are conducting combined tactical operations within the same area of operations.” A MNF-W like model must be developed and implemented in Afghanistan now, with an eye toward Provincial Afghan control as soon as possible.

Exactly, exactly right. In Afghanistan when I was there, there was not even a formalization of a command relationship ("Tactical Control" being the loosest defined in NATO doctrine) as soldiers understand the term with regard to the ANA. They did their thing, and we did ours, in overlapping battlespace. We accordingly violated that key military principle of "Unity of Command" in nearly every last thing we did. It seemed to get us, and them, almost exactly nowhere. That seems the biggest difference right now between how the comparatively successful ETT effort in Iraq, where command relationships between the US and the Iraqi Army, albeit loose ones and still supplanted by a mentor-liaison framework, were established, and the questionable parallel OMLT efforts in Afghanistan, where command relationships between the Afghan forces and ISAF, at least while I was there, remained largely undefined.

David Ray dis-association

For the record, I have no time or patience for certifiable Truther nutballs like David Ray Griffin. For a blog like Newshoggers to link to him and this blog in near-successive posts diminishes them and this site by association (yes, I recognize it's a group blog, with the two posts coming from different authors), whether Griffin turns out to be right about Osama's fate or not. There are better sources on the question. David Aaronovitch's latest book is worth a read in this regard. (I also thought the Fifth Estate's recent doc on the Truthers that didn't "take sides" was yet another low point for Canadian journalism, that diminishes everything else they've ever done, as well.)

UPDATE: Newshoggers has responded by removing their link to this site.

Afghan reading catchup

Back from a little business trip where I was too busy to blog... a couple random thoughts.

I'm hereby coining the phrase "Going Gant", a combination of "Going Galt" and "Going Kurtz," to describe the new emphasis on Special Forces-type disrupt operations using tribal auxiliaries in uncontrolled areas of Afghanistan. Maj. Gant, who is the exemplar of what this would look like, has since been sent to Iraq, and many have raised perfectly valid criticisms about how this wouldn't work, at least as a comprehensive Afghan security strategy, that don't need reiteration here.

That said, there's a bit of a category error here. You can't rebuild Afghanistan using Gant's methods, true, but I think it's really being looked into more as another method of keeping the enemy hopping in areas where a sustained government or ISAF presence is uneconomical at present. And that's not a totally crazy idea. If the alternatives when all you're trying to do is "sustain the disrupt" are Predator strikes and SOF kinetic ops involving black-painted helicopters and door-kicking, Gant-type teams could even be the more economical way of keeping the Taliban from settling down in an uncontrolled area. (Pat Lang, who if nothing else has danced this dance before, is one of those who thinks it has promise in that more limited context.)

There's two requirements for that limited-Gant-type approach to work that haven't been clearly identified yet, though. One is a clear delineation by ISAF and the Karzai government of what parts of the country are "outside the inkspot" at present. Without calling them "free-fire zones" or the like, both Afghans and westerners need to be clear what districts of the country are the preserve of SOF, SF, etc., and which parts are the preserve of ANSF, their mentors, and PRTs. Because otherwise you'll be into a friendly-fire situation before the week's out.

And for that division between the "hold" and the "clear" you'd need the full cooperation of the Afghan government, which is always loathe to admit it doesn't control a district any longer. Because otherwise they'll continue to send fresh police and poorly protected soldiers into areas outside the inkspot to fly flags at district centres, and desert, be killed, or be suborned (and, assuming there's a Gant team around, inevitably shoot at or be shot by them). Of course, it's fair to say a LOT of other things could have been going better if we'd had the "full cooperation of the Karzai government", too... that's hardly a trivial objection to even a limited Gant-type approach.

Other points: with regard to the fallout from the Kunduz attack: keep it coming. There's more people that need firing over this one yet. It's hard to count the number of rules and best-practices that were broken or disregarded here if this had happened in a Canadian AO, but I'll start: single-source humint, no eyes-on from the ground, no apparent PID, no apparent "green PA" or CDA checks, on a truck stuck in a river and not going anywhere fast, practically within sight of the base, and within easy Quick Reaction Force range. In the KAF context, it'd be like calling in an air strike because someone had reported sounds of firing at Tarnak Farms. (Oh, yeah... right. Question: is there a USAF investigation on Kunduz ongoing? If not, why not?) Watching the leaked aerial footage in Bild is worth your time.

And as far as ANA desertion rates, you may well have read it here first, but good work to Gareth Porter for exploring the inconsistencies in U.S. government reporting. Any story with the phrase "questionable accounting methods" is also worth the time spent. I would agree with the Mason and Jefferson study mentioned by Porter, that 100,000 seems a pretty realistic maximum size for the ANA alone. Any plans to grow much beyond that are not likely to succeed in real-world timescales.

Finally, from the Star:

But what concerns [photographer Louie Palu] more is the increasing number of so-called "in-field" detainees – Afghan men taken into custody in Afghan-led operations, with Canadian or U.S. soldiers (or both) present as advisors.

"When the blended mentoring teams capture a detainee, who does he belong to? They become Afghan detainees because the Afghans are in the lead."

Buried in a weekend edition, Palu has pinpointed the shoe that hasn't dropped yet on the Colvin detainee allegations ("in-field transfers" is the more common term). It isn't just those taken today in mentored operations, although I'm sure that number is increasing. Over the last couple years, dozens of Afghans have been taken on ISAF-led ops in Kandahar Province and handed over at the point of capture by ISAF personnel to participating Afghan security forces. Because they were never technically in Western hands for any extended period, there is no reporting requirement imposed on the Afghans under our existing protocols. Unavoidable? Probably. But supporters of the mission might want to continue to keep their arms braced on the seat-back in front of them, because at some point the media may connect those dots.

Oh no no no

Canadian soldiers will recall that once upon a time we had an unarmored jeep-type runabout called an Iltis. The Germans used it for base MP duties. The Canadians, because we had nothing better at the time, used it as a primary tactical reconnaissance platform briefly, even if it wasn't big enough to fit a machine gun on.

A friend of mine likes to tell of the day he was working with a Bundeswehr colleague, who was telling him how the MPs had the same jeep back on his base back home... but when the German found out Iltises would be the first vehicles the rampaging Soviet hordes would have to deal with, he adopted a look of petrified horror, and started muttering and rocking back and forth, "Oh, no.... no, no, no.... no, no, no..."

After reading Joe Klein this week I know how the German felt.

The military has been shockingly slow when it comes to matching U.S. training companies with Afghan battalions. No such joint units currently exist... And the idea that illiterate and tribal Afghans can be trained into soldiers and police officers remains more a hope than a fact.

An accurate diagnosis. It's his proposed solution that alarms:

When the military, for example, thinks about sending more troops to Kandahar, it thinks about months of planning, new fortifications and so forth--instead of a quick, transitional insertion. There is a huge airbase just outside of Kandahar city that could house substantial numbers of U.S. troops on a temporary basis, in tents if necessary, while joint security stations and the other accoutrements of counterinsurgency warfare are established in the city itself. This is especially necessary since, I"m told, the situation in Kandahar is declining rapidly. "We've lost the surrounding areas," one counterinsurgency expert recently told me, "and I'm not so sure we're in control of the city, especially at night."

Oh no... oh no no no...

Look, experience has proved the only way you keep the Afghan police alive and honest long enough for them to make a difference is by living with them 24/7, not by driving out on alternate mornings to see if they're still breathing. Putting more soldiers into KAF, by contrast, will increase the lineups at the French bistro and the Burger King, but will do absolutely nothing to help Afghans. Nothing in KAF is ever a net-positive for this war: like BAF in the north, at best it's a necessary evil. But it's also a gigantic, Afghan-free soul-destroying vortex of resources and Western-style banality that mostly only succeeds in making its inhabitants weep for humanity's fall. Fewer KAF rats, not more, please. Do what we did when we were in the city*. Live light. Keep out of sight. Take risks. Trust your Afghan hosts to want to stay alive as much as you do.

Or, alternatively, you could always take over Nathan Smith from the Canadian PRT and dedicate it completely to police mentoring (it already does quite a bit). Put the other essential personnel from that organization who need daily contact with the governor, etc. somewhere in the city where that might actually happen at least as reliably, and evacuate the rest back to KAF. Or back home for that matter. Because realistically, if you're in KAF you're over halfway there anyway.

*For the record, less than four weeks total, and none of it at Nathan Smith, FWTW.

The prisoner of Mushan

It's probably fair to say that the latest claims of abuse of former Canadian detainees by the Afghan secret police, the NDS in Kandahar have put perhaps the last nail, if one was even needed, on a post-2011 Canadian military deployment in that country.

The first allegations that the NDS were then abusing the detainees we gave them in 2007 had led to a Canadian freeze on all detainee transfers for a while, and by the time I got there in late 2008, detainee issues were still very much a concern. The one guaranteed way you could make a hard-bitten Canadian duty officer blanch was to tell him that an Afghan had been taken into Canadian custody, as opposed to Afghan custody, with all the extra work THAT entailed.

Whenever humanly possible in late 2008 and early 2009, detainee responsibilities were diverted by Canadians to the closest thing to an Afghan authority figure that they could find nearby: police, Afghan army, a passing civilian district leader... -- right at the point and time of capture. Taking them into our national custody for even a minute was always the suboptimal alternative. Working "jointly" with Afghans in this way did serve to limit our national reporting and followup requirements; it had other potential advantages from an Afghan capacity-building and national sovereignty point of view, too, obviously.* But because there were no Western mentors working with the NDS at the time, we didn't really have any visibility to say one way or the other what was happening to all those undocumented, non-Canadian detainees.

With detainees, we always seemed to be in one of those perfect catch-22s that typified the ISAF mission; in this case we and the ANA were also running perpetually afoul of the Afghan government's own, possibly more effective anti-NDS abuse measure. Afghan law at the time said all detainees had to be brought before a judge (for a sort of "show-cause" hearing) within 72 hours of capture, without fail. In Kandahar, this was actually being fairly rigorously enforced, to NDS officers' chagrin at the time. Unfortunately, that made it kind of difficult, given the IED situation and everything else, for ISAF or Afghan forces to establish any kind of evidentiary linkage with an IED attack or other insurgent activity that would justify their continued detention and then deliver detainee and evidence by road to Kandahar in time (if it was ISAF-collected evidence, the declassification and translation processes each might have taken days if not weeks). So by a literal reading of Afghan law, nearly all detainees, both innocent and guilty as hell, should have been promptly released by the judge at their first court appearance, and certainly many were.

Perhaps Canadians, who've hadn't taken detainees in a counterinsurgency situation before this since, oh, about 1902, weren't the best people to be instructing the Afghans on how to do this right. And once again, really we were falling afoul of that early handover by the West of Afghan sovereignty, to the point where we really had insufficient influence left over what had been essential processes in previous counterinsurgency fights: the courts, the pre-trial prison system, police questioning. But all we could do was carry on, trying not to notice how much that deficit might have been subverting our other efforts.

The title of this piece refers to one of the more amusing detainee moments during our tour. Strongpoint Mushan, deep into Panjwaii, now long closed, once housed a Canadian mentor team and an ANA company. Late in 2008, the Afghan company commander there rousted up someone he claimed was a local insurgent leader, Shasta Gul by name, and took him back to Mushan as his detainee. So... now what? The only way you got to or from Mushan at that time was by helicopter. But at the time an Afghan soldier prisoner escort would not be allowed on ISAF helicopters (safety: couldn't understand in-flight instructions, we were told), and if the detainee had been transferred by an ISAF flight crew, he'd have then become the West's detainee by virtue of us having custody for that period, which was what we strenously wished to avoid. So, having no way to actually do anything with young Shasta, and not wanting him to just walk away and resume attacking them so soon, the Afghans kept him around the strongpoint for a while. The story we heard was the detainee was doing their cooking**... But a couple weeks later, after it was clear the Afghan National Air Force probably wasn't going to be mature enough to pick the guy up any time soon, the Prisoner of Mushan was released, and walked out of the now thoroughly reconned strongpoint a free man. (Even if he had somehow been teleported back to KAF, it wouldn't have mattered anyway, as our 72 hours had long since passed; all we'd have done thereby would have been increase his cab fare to get home.)

It was hard for us not to interpret these sorts of paradoxical catch-and-release events at the time as partly attributable to that earlier Western reaction to ANSF prisoner abuse allegations. But whether that was justifiable of us to say or not, this was still Afghanistan: no matter what you did, there were rarely good choices, but you always had your pick of the bad ones.

*Please note, in case I haven't made it clear enough yet: my tour and detainee experiences related here came 1-2 years after the period of the Colvin allegations in the press this week (and long after the first Afghan prisoner stories began to appear in the papers) allegations on which I have nothing useful to relate. It was in many ways a very different war back in Colvin's time.

**I'm proud to state that the ANA I worked with always treated their detainees with extreme civility. They had learned our lessons well (plus an Afghan's default mode is always hospitality): detainees in the army's charge were invariably treated honorably, transferred to the NDS as soon as possible (that quirky 72-hr rule applied to them, too) and filling out any tiresome intelligence- or custody-related documentation on them was thereby avoided.

Today's essential Afghan viewing

Blogging Heads, in particular David Frum in his guest appearances, has been doing some good Afghan work recently. His discussion with Time's Joe Klein is definitely worth your time. He had another one yesterday with Andrew Bacevich I'm saving for my next deskbound number crunch time.

The Afghan-related dialog between Christopher Preble and Peter Beinart was good, too.

Couple places I've shown up other than here

I was quoted as part of the University's Remembrance Day coverage, here. I think the pic on p. 4 is one of my favourites from the roto.

My speech before the CDAI last month also appears to be playing a small evidentiary role in Tony Cordesman's CSIS group's latest study on the ANA (draft PDF).

Yet more essential stuff

Judah Grunstein efficiently dissects the current Gant/Pressfield "arm the Afghan tribes" meme. Every future OMLT-eer should write this bit on his field message pad cover:

Anyone who has worked in a helping role -- whether in social work, aid and development, and apparently population-centric counterinsurgency -- has witnessed (or lived) the phenomenon of a line worker identifying with the target population, especially when the line worker is subject to extended immersion within the social structure of the targeted population. Inevitably, it leads to a confusion of loyalties and friction with the broader institutional goals. There's a very delicate balance between listening to and empathizing with the people you're trying to help, and maintaining defined boundaries -- to identify their needs, without identifying with them. But it's an essential balance to strike, because the helping relationship is by its very nature vulnerable to manipulation and abuse, on both sides.

The thing that bugged me in reading Maj. Gant, other than the huge difficulties in operationalizing his "arm the tribes" proposal, was that bit about how he fit in so well that the Afghan headman's wife met with him unveiled on several occasions. I'm just not sure that's the sort of thing one puts in a generally distributed academically-inclined paper simply to make a point in one's argument.

Today's essential Afghan PowerPoint

From the CounterInsurgency Center blog, essential info for troops heading to Afghanistan. The presentation at the first link, a pictorial summary of Marine lessons learned, is exactly the kind of thing any predeploying soldier should be thirsting for.

(The second PowerPoint, for the basic COIN education of ANA 205 Corps soldiers, was clearly done by the corps' Western mentors. Anyone who has seen an actual Afghan government PowerPoint will know what I mean. Still, it's great that it's being done, at all.)

Today's essential Afghan reading

Gilles Dorronsoro.

Dorronsoro and, singing in a slightly different register, David Kilcullen are shaping the new Conventional Wisdom on Afghanistan virtually as I write. What's out, in this new CW? PRTs, NATO-style OMLTs, and clear-hold-build, at least as a generally applicable model for Afghanistan.

The growing consensus of Dorronsoro and others seems to be that 2009, despite all the new resources thrown at it by Pres. Obama, was still another year of stepping backwards, primarily due to: a) an overfocus on Helmand, where whatever payoff it has had on the ground has to be balanced with the spike in casualties that came with it starkly eroding U.S. will and nearly driving the U.K. out of the war; and b) the election fiasco. Dorronsoro defines the problem:

A common misperception is that the insurgents are terrorizing the Afghan people and that the insurgents’ level of support among the people is marginal. This has led to the objective of “separating the Taliban from the population” or “protecting the population” from the Taliban. Yet at this stage of the war, and specifically in the Pashtun belt, there is no practical way to separate the insurgency from the population in the villages, and furthermore there is no Afghan state structure to replace the Coalition forces once the Taliban have been removed. In fact, this approach reflects a misunderstanding about just who the Taliban are. Even if it is possible to find examples where the Taliban are not local and oppressive to villagers, the situation in the Pashtun belt is much more complex. The Taliban have successfully exploited local grievances against corrupt officials and the behavior of the foreign forces, framing them as a jihad. Moreover, the Taliban are generally careful not to antagonize the population. They are much more tolerant of music and of beardless men than before 2001, and Mullah Omar has repeatedly made clear that the behavior of the fighters should be respectful (for example, paying for the food they take). Most of the insurgents are local and, especially in case of heavy fighting, the local solidarities tend to work in favor of the Taliban and against foreigners in a mix of religious and nationalist feelings.

How does the Coalition control the supposedly cleared areas? Trust between Coalition forces and the Afghan people (especially the Pashtuns) simply does not exist, and, after eight years in the country, the battle for hearts and minds has been lost. The Coalition forces still have not worked out how to be accepted locally—that it is counterproductive to patrol villages with soldiers who are ill-equipped to overcome linguistic and cultural barriers and whose average stay is six months. This miscalculation has been compounded by the past poor behavior of some Coalition forces— the beating of prisoners, arbitrary imprisonment, aggressive behavior on the road—and the unwitting bombing of civilians.

Finally, given the complexity of the strategy—one that requires a deep understanding of Pashtun society—one must ask whether the Coalition has the bureaucratic agility and competence to implement it and outsmart the Taliban, who are obviously quite good at playing local politics. There is no reason for confidence in this regard, so the Coalition should pursue a simpler strategy in Afghanistan.

On Helmand:

The policy of clearing is plainly not working. The insurgents are woven into the population, and there is no way to distinguish them from ordinary villagers. As a consequence, the area targeted by the Coalition forces remains unsafe, and because the Afghan National Army is too weak to substitute, the troops can’t withdraw without allowing the Taliban to regain control...

So, Coalition forces have failed to clear a significant portion of the province of Taliban fighters. The border with Pakistan remains wide open; it cannot be controlled with only U.S. troops. Predictably, the Pakistanis are not helping, allowing Taliban groups sanctuary when they need it. Only one district—Nad Ali—has been secured by a heavy U.S. military presence, and even then, not totally because IEDs are deployed there. There is no guarantee that the Taliban—in the North and South of the province—will not return.

Today's essential Afghan reading

Why do I suspect Josh Foust had something to do with this? It's excellent.

7 3/8

For some reason I always forget my hat size, so this is a note to self. (The rest of you just go about your business, nothing to see here...)

Honours and awards trivia challenge

That second award on Her Excellency the Gov. Gen., (yellow with light green bars): any guesses?

Hint: There's a matching one from Alberta.*

*Update: previous text corrected thanks to Colby Cosh; apparently the Wikipedia bio as it stands is wrong about Her Excellency's being granted the Alberta honour, and she is not on its list of recipients.

R-Day

Paula, Warrants Wilson, Murphy, and Roberge, and all the rest: I'm sorry you're not here to share this. That's all I have to say.

Why blogs were created, really

The advantage of blogging in the good old days was you could often get information from trustable people close to the action, without the media filter, and early hyped mis-impressions.

In the case of the Fort Hood shooting, if it turns out that the person who actually brought down Maj. Hasan was not, in fact, Sgt. Kimberley Munley but her partner, Senior Sgt. Mark Todd, you'd have read it first the way I did at Kevin Drum's place Friday morning. Nice scoop, Kevin. Both brave officers, regardless, of course, but credit where credit's due, please.

Today's... I don't know what this is, frankly

At serious risk of breaking the stones-glass houses rule, I feel compelled to write something here about another Canadian military online essayist.

The fellow behind this post has more relevant experience, with the military, with Afghanistan, and probably with life in general than I do. So please take my criticism of his writing with that in mind. It's his attempt at a big solution piece on What to Do in Afghanistan. The synopsis:

First, get the world to legalize the consumption of heroin. Check.
Second, get Karzai and Abdullah to form a national unity government. Um, check.
Third, get the Karzai-Abdullah government to introduce conscription and mass-enrol Afghans into the army and police. (...)
Fourth, get Pakistan to "reclaim control" of its FATA territories. (!!!)
Fifth, get Afghan and Western forces to "consider the Durand line irrelevant" and chase Taliban onto Pakistani territory whenever required and thus deny them a safe haven. (!!!!!)

If you do all that, on a "tight and non-negotiable timetable" it will not mean victory, but you will have bought Afghans their "last, best chance," after which we can pull out.

I'm sorry, but reading that in this context is not unlike reading "first we need to breed a race of superponies. Then the superponies need to spawn a new breed of flying superponies. The flying superponies will need to be extremely small, so they can power all our electronic devices with their minds." It's nutty.

Now, obviously there's a couple ways to respond to the "five impossibilities before breakfast" piece, above, besides lamenting the state of Canadian military thinking. One is to assume actual thought went into this and conclude, "oh, that's the throwaway COA*," the planner's real preferred COA being, presumably, getting out. Or another way to put it would be, if this is the minimum requirements for success, then you're basically arguing that success is unachievable. Which is the actual effect of the piece, whatever the author's intent here.

Look, I'm in favour of military writers, and I love a good policy or strategy idea as much as the next guy. But I think we all need to ask ourselves before we stick our ideas out in public, "is what I'm advocating possible in the real world that I live in?" Because an impossible idea, and no idea at all, are functionally equivalent in this context.

UPDATE: Just to show that the Canadian military does not have a monopoly on Afghan-related flying pony fantasies, unnamed diplomats say Karzai now needs to jail his brother, his running mate, and the man who controls all the country's millions of Uzbeks. Best part is the quote: "the diplomat, like other officials, spoke on the condition of anonymity because of the delicacy of the matter." Either that, or because he knows people would call him a "friggin loon": you pick...

*COA=Course of Action. "Throwaway COA" refers to the military staff practice of always presenting the commander with three options, even if the problem is clearly binary: the staff's preferred option, the one they consider suboptimal, and the throwaway. So given the problem of getting from point A to point B with only one road in between, COAs 1 and 2 might be "take the road," and "ask for helicopters," and the throwaway #3 will be "invent hovertanks", or something similar on the crazy scale.