Monday, November 30, 2009

Levin still banging the "Afghan Surge" drum...

...despite noises out of the White House that the expansion of the ANSF will not be as extensive as GEN McChrystal would like. From the WSJ:

The Obama administration has soured on a call from its top commander to double the size of the Afghan police and army, reflecting the White House's continued skepticism about the Afghan government even as the U.S. prepares a surge of troops into the country, people familiar with the matter say. ...

But the administration seems prepared to reject another of Gen. McChrystal's top priorities: his call to double the size of the Afghan police and army over the next few years.

The administration now favors an alternative plan that would seek to build a larger Afghan security force, but one that would be considerably smaller than what Gen. McChrystal had wanted, these people said. The president is likely to talk about Afghan troops Tuesday, without specifying a growth target for expanding their ranks.

"The president has a realistic view of how successful the training regimen can be, and that has helped inform his decision," a senior administration official said Sunday.

And then there's Sen. Levin, saying, well, pretty much the exact same thing he has been for the last several months, basically verbatim, and in complete defiance of the reasoned arguments that have been made against his recommended course of action by reasonable and educated people (but then, that's politics, innit?):

President Barack Obama must show how more U.S. combat troops will speed the build-up of the Afghan army to generate Democratic support for his new war strategy in Afghanistan, Senator Carl Levin said.

“The key here is an Afghan surge, not an American surge,” Levin, chairman of the Armed Services Committee and a Michigan Democrat, said on CBS’s “Face the Nation” program yesterday.

“If the president lays out the case for why our combat forces that are going particularly to the south will increase the speed-up of the Afghan army, it seems to me that that would be very, very important,” he said. ...

Levin said showing that the mission of more U.S. troops is to “very quickly build up the Afghan army,” and “give them the capacity to take on the Taliban” is important to winning Democratic support.

Everyone's telling me that a funding bill for escalation is going to be a tough vote, especially in the House, and particularly for Democrats. I can't see the rank and file deserting the president and the Speaker on this one, particularly with such a controversial issue as "not supporting the troops," and Democrats in contested districts -- as if they hadn't already faced enough tough votes this year -- will have no choice but to support funding for additional forces.

All of which is to say that I don't think Levin has any damn idea what he's talking about here. Those Democrats who oppose escalation of the war aren't going to feel ok about additional troops so long as the president just promises we're Afghanizing the war, and those who recognize the impossibility of voting against a war supplemental only have so much leverage in demanding that the money go to a specific mission. (This is leaving aside altogether the fact that rapid Afghanization will fail even more spectacularly than will, I expect, whatever alternative President Obama presents tomorrow.) So ok, Senator Levin, we hear you. Now please just stop talking.

Let's be clear: the ANSF need to get bigger, and they need to get better. And perhaps doing that rapidly and Afghanizing the war was a workable solution several years ago, before the insurgency was thriving. But right now, under the current circumstances, it's going to be really, really difficult for them to get bigger and better at the same time. Calls for more training and mentoring of host nation forces at the expense of more extensive, security-oriented counterinsurgency operations are misguided, and strike me as disingenuous. This IS NOT an "easy way out," and it's not going to create conditions under which the U.S. can credibly claim "victory" (at least not any more than we could now); this will not bring peace with honor.

No matter which policy course is (well, was, really, at this point) chosen in Afghanistan, the training and mentoring of ANSF is (/was) probably going to be a part of it. But to imagine that we can just train up the good guys and get the hell out in the middle of a raging insurgency is just not particularly sane.


  1. Just want to say, this is a great blog and I hope you can take up some of AM's slack as he semi-retires from blogging

  2. Thanks for the kind words, Ryan. We'll do our best.

  3. I'm reminded of how Miami PD tried to rapidly expand in the 1980s in response to drug violence. Corruption became an even bigger problem.

  4. I'm reminded of how Miami PD tried to rapidly expand in the 1980s in response to drug violence. Corruption became an even bigger problem.

    As Kilcullen's been saying for a while, connecting people to their government is only productive if contact with the government isn't making their lives worse, or casting doubt on its legitimacy, or otherwise inflaming the insurgency.

  5. Gulliver hit the nail on the head, and I'd just add (to make some of the sub-text explicit) that assuming we can rapidly build the ANSF while they're in contact and such pressure (not just military pressure, either) is absurd. Pushing them into the lead before they're ready is just going to have make everything worse, through both the obvious first-order and less obvious second- and third-order effects. Doing it right is going to take time, resources, and not a small bit of luck.*

    * Where luck is the intersection of opportunity and preparedness

  6. I'd like to add unicorns. Opportunity, preparedness, and unicorns.

  7. I'm sort of embarrassed to admit that MK's comment made me think of this:

    But it's gonna take money/ It's gonna take plenty/ To do it right child/ It's gonna take time/ A whole lot of precious time/ It's gonna take patience and time, um/ To do it, to do it, to do it, to do it, to do it/ To do it right child

    Just try to tell me you weren't thinking of George Harrison.

  8. Uh-oh. Between the unicorns and the Harrison, not sure whether to be more worried about you, Gulliver ol' pal, or the rest of us.

  9. To do it, to do it, to do it, to do it...

    You've ruined my whole day with this, MK.

  10. That's not really where the debate has been going within the military, Gulliver.

    What has everyone concerned is two-fold: 1) A lack of legitimacy beheld by the target Pasthun population of the ANA, which is seen (somewhat rightfully) as over-represented with Tajiks and worthless Karzai flunkies; and, 2) as in Vietnam, a concern that the Thieu (errr, I mean Karzai) government has become tethered solely to the occupation for money and force, just as the ANA and police depend too much on our CAS, etc.

    This crossed my transom today. It was written by someone with a great deal of experience in these issues, part of a larger debate within the military and it provides a partial answer to "how good should the ANA be?"


    The standard that should be applied is simple to name but apparently impossible to reach: The Afghan security forces should be considered by the target populations 1) Tougher than their past, present and future enemies, most prominently the Afghan Taliban militias; 2) Capable of aggressively closing with and defeating these guerillas on the battlefield, no matter where it is; 3) Legitimate enough to withstand the taint of being linked to the US-led occupation; and, 4) Self-sufficient to the point that they won’t need to use as crutches US logistical lift, air and indirect fire support, funding, etc, etc, etc.

    The application of force by this putative army takes two forms: Puissant power, which is applied directly against Taliban militias on the COIN battlefield, and should that be successful, the psychological ability to deter future aggression and convert the “people” to believing that the official Afghan army and constabulary might last for awhile.

    Since a goodly minority of the target populations we’re trying to sway apparently continues to believe that the Afghan military is illegitimate (tied as it is to Tajik officers, disputative warlords and Karzai kleptocrats) and, moreover, weak, we’re unlikely to build our way out of the current mess, just as we can’t kill our way out of it.

    Some might argue that ARVN in Vietnam eventually became “good enough” to go toe-to-toe with the North Vietnamese, but I don’t believe that this was so and, more importantly, neither did the people of rural South Vietnam or the North Vietnamese. Just as we didn’t need 15 years to make an ARVN First Sergeant, we apparently needed more years than we were politically willing to give to stand as a crutch to variously worthless Saigon regimes and the military that suckled off of our taxpayers.

    At this point, I’m not exactly sure that the proper combat leaders in the ANA will exist in 15 or 150 years to slug it out with the Haqqani network or the Quetta Shura.

    Perhaps a way around this would be to articulate US force to help certain militias that already have some legitimacy within their own communities. Rather than creating mixed units led by a Tajik, for example, why not simply tie US help to a Tajik force that will keep the Taliban out of their areas? Forget the Karzai kleptocracy and focus on what will serve US interests best – more warlords, only this time the warlords we and the people can agree on supporting.

    If the various cantons of Afghanistan should wish to scab into something resembling a federal government, so be it. But that’s not necessarily in the best interests of America, nor should our utility of force be hitched solely to the (unlikely) enterprise.

    If the Karzai kleptocracy implodes, we should have friends who can replace it and serve our interests of hunting al Qaeda, by bounty if necessary.


  11. SNLII -- How is it exactly that I misunderstand "the way the debate has been going within the military"?

    I'm not sure I disagree with much that you've written here, though I'm not certain we even require the services of whatever assorted warlords are willing to make promises about the inviolability of their territory.

    Why is al-Qaeda currently in Pakistan and not Afghanistan? Because of the 100K coalition troops? How much would their freedom of movement and action change if it were 20K? They seem content in Waziristan, so why even come back?

    Sad as it sounds, Afghanistan becomes a billiard ball again once U.S. troops are gone: opaque and entirely obscure -- one could even say irrelevant -- to us.

    This is pretty scattered, but here's the point: I don't think the ANSF ever get to the level of capability that you've described, or really anywhere even close, and certainly not within the next 10-15 years. All the resources in the world won't change that. So can anyone? Can any party maintain the monopoly of violence inside the fiction that is Afghanistan? Should it matter if they can, considering that 1) terrorist sanctuaries already exist, most notably just across the border in Pakistan, and 2) we're not sure that said sanctuaries are even particularly relevant?

  12. SNL II, the ANA isn't quite like you describe. They have Pashtun generals, not a proportionate number, but some, and senior officers. So did the Northern Alliance, their immediate precursor; so did the Soviet-era ANA. There's an unbroken leadership line going all the way back through those organizations to the coup plotters of the late 1970s. The leadership of the ANA are a group of centralist nationalists, who consider the Taliban to have been a bunch of Pakistani-proxy hicks who got lucky. That's really how they see all this.

    If we left tomorrow, in their minds they'd form a ring around Kabul and prepare to fight whatever proxy force the Pakistanis throw at them, and fighting-withdraw back to the Panshir if they had to. (Massoud is still their #1 hero and subject of emulation.) But they really don't mind Kabuli or Herati Pashtuns... it's the hicks from Helmand or Kandahar rural areas that they have trouble identifying with (and who have trouble identifying with them) because they feel they're dupes of the foreigners, and a little uncivilized and potentially extremist.

    There are no "Tajik warlords" or militias anymore; they all have commissioned rank now. Not many Hazara warlords either. Those two subcultures in particular, along with urban Kabulis, have poured their community's efforts into this joint national army which they sincerely hope will hold off the Pak proxies next time.

    Again, I'm just describing how they see the world, from spending a fair bit of time over chessboards with ANA majors and colonels. Their legitimacy as such is unquestioned already outside the insurgent areas, which is why I think your prescription above is not a valid one.

    Yes, they are likely to remain ineffective. Yes, without us they cannot hold the whole country. And yes, they're not likely to win many friends among rural Pashtuns; their interests, their program of Kabul-based centralized authority is not something the rural Pashtuns can subscribe to.

    But the most plausible scenario when Western influence inevitably wanes will not be an immediate insurgent takeover, but a resumption of a civil war , with neither side gaining the monopoly of violence Gulliver refers to, that could go on for a very long time indeed, with the front line somewhere between Kabul and Kandahar (or really, Panshir and Helmand) depending on whether the Kabuli-centrist ANA or the insurgents are winning. As with the last couple times, how that war goes will largely depend on the external support both sides continue to receive, and how long before the two sides begin to fracture on ethnic or other lines.

  13. Bruce -- This is a great post, and a reminder to me of why it's so dangerous to try to be brief and glib.

  14. The proposed numbers for the Afghan Army and security forces are a lie.

    I worked for the unit responsible for training and equiping the ANA in 05-06 (CSTC-A? Can't remember the exact acronym). Our target end strength for the ANA at that point was roughly 40K. In 3 years, we've gotten the number up to 80K (and I think that number is more in theory than actual.) Even with another 30K troops training, and only a portion of them will be trainers, how in the world would the force ever get to 400k? Especially in the time frames acceptable to the American people?

    Besides that, there are all sorts of reports, and my personal experience working with the ANA confirms them, that the ANA has very high desertion rates. In 05-06, we ended up hiring the same guys multiple times.

    Also, I really enjoy your blog. I think it is better than AbuM. I always thought Ex came across a little to self-assured that CNAS had the answers for Afghanistan.

    You guys come across much better. Anyway, keep up the good work.

  15. 'There are no "Tajik warlords" or militias anymore; they all have commissioned rank now. '

    That was my point.


  16. "Their legitimacy as such is unquestioned already outside the insurgent areas, which is why I think your prescription above is not a valid one."

    This, again, is meaningless. It does NOT matter how legitimate they are amongst the populations they already control, but rather the target populations currently in rebellion.

    This is so obvious that apparently is must be said twice.


  17. "SNL II, the ANA isn't quite like you describe. They have Pashtun generals, not a proportionate number, but some, and senior officers"

    The long, sad history of the creation of the ANA is well known. I could begin with the Inter-Agency Operations Group that initially sought to change the packing of the defense forces with Tajik officers and troops in 2002 (Eikenberry actually promulgated guidelines seeking ethnic balance in the force).

    DoD (how does one put this charitably?) lied about the mix for years. In October, however, OIG for Afghan Reconstruction actually put out a hard count and found that Tajiks (about a quarter of the population) comprise 41 percent of the trained and equipped ANA. It's not likely to get better, because the recruit pool currently is only 30 percent Pasthun and, as Keith indicated, suffers from widespread desertion.

    Another report (I would post it but I don't know if it's classified) recent promulgated suggests that we've reached a ceiling on the numbers we can get because of the law of diminishing returns affecting the quality of officer leadership within the ANA.

    In certain critical areas such as Zabul, almost the entire ANA force deployed there is led by Tajiks. UN officials continue to estimate that 70 percent of the kandaks in the field are commanded by Tajiks, which representsed NA leadership rates, too.

    Our problems aren't simply the facts on the ground: It's the perception. Even if we could radically improve the mix and competence and willingness to serve amongst the ANA, we still would be left with the task of changing perceptions in the target populations of Pathans in the south and east about these issues.

    I'm not sure if that's possible.


  18. Our problems aren't simply the facts on the ground: It's the perception. Even if we could radically improve the mix and competence and willingness to serve amongst the ANA, we still would be left with the task of changing perceptions in the target populations of Pathans in the south and east about these issues.

    I disagree that this is necessary. Armies have exerted control over populations from which they differed ethnically, linguistically, ideologically, politically, and so on for ages; why should this instance be any different?

    The long term goal of developing national unity, of course, is much more difficult. But in the meantime we're not concerned with national unity so much as territorial control, or rather the capability to credibly deny territorial control to another faction.

  19. Because we can't deny the 30 or 40 percent of Afghanistan currently owned at night by the Taliban (and other militias) so long as the population supports them.

    This could change should the public perceive the central army mustered the puissant power (as my buddy put it) to beat the Taliban and would stick around long enough to do it, even without US help, but I don't believe that the target population believes this to be so.

    Also, when you say that "we" are not concerned with national unity, then you've missed what we have instructed our CAGs and other CA units to accomplish in Afghanistan, ironically on behalf of the Afghans.

    Shades of Thieu.


  20. Yeah, but there's a correlation/causation question here, too. The ethnic balance of the ANA roughly corresponds to the ethnic balance of the country less those who don't accept its legitimacy, passively or actively, and those under their sway.

    Hence the rural Pashtun is underrepresented, and the Uzbeks (who are more passive about their separatism), but Hazara, Tajiks and Kabulis are there in the appropriate proportions.

    The army is having trouble recruiting people who reject the current political setup and those under the rejectionists' sway. True. But saying the ANA in 2009 is lacking sufficient Pashtuns is a little like McClellan saying in 1862 he has an insufficient number of Carolinians in the Army of the Potomac.

    One can make a strong case for scaling back our expectations of what the ANA is likely to accomplish in terms of defeating this insurgency militarily, without advocating we need to encourage its decay into ethnically homogeneous units, like the Tajik militias you're proposing. It's hard to see how that would help, frankly.

  21. Because we can't deny the 30 or 40 percent of Afghanistan currently owned at night by the Taliban (and other militias) so long as the population supports them.

    No, we can't currently. But we're not talking about "currently," are we? We're talking about the future. I'm asserting that the delta between where they are now and where they'd need to be to exert control is about capability (and of course, relatedly, numbers), not about legitimacy. It's all well and good to say that the subject population will continue to support the insurgency until such time as they feel the army and the government are legitimate, but those institutions must assert control at some point in order for perceptions of legitimacy to change. Again, collaboration follows control, not the other way around.

    Look, I don't think this is going to happen either way, because I don't think Afghanistan is a country that can sustain an army that makes the necessary leap in capability to assert control. I just don't think it's a problem of legitimacy and ethnic representation.

    Also, when you say that "we" are not concerned with national unity, then you've missed what we have instructed our CAGs and other CA units to accomplish in Afghanistan, ironically on behalf of the Afghans.

    I haven't missed anything, I simply think it's misguided. Our military forces are currently operating under the conception that we can help the Afghan government develop sufficient legitimacy to quell the insurgency. I don't think this is likely, so I'm less concerned with national unity and more concerned with capability.

  22. To restate my last point in another way, I think we should be focused on helping the ANSF take control of as much territory as possible, and less interested in ensuring that they develop the perception of legitimacy or ethnically proportional representation. The insurgency ain't gonna quit when the army has enough Pashtuns -- can we agree on that? They're a lot more likely to quit once they get their ass kicked by a Tajik-led force, right?

    So let's worry about building an ass-kicking army, not a representative one. Representation comes later.

  23. Both of you are dancing around the problem.

    Gulliver, you say that numbers and units come before legitimacy or ethnic proportionality. I would suggest, however, that Eikenberry was right the first time: You can't have legitimacy if you don't have proportional representation both in the ranks and amongst the commanders because the target population we're trying to sway is going to see the national military as Tajik or Uzbek or the wrong sort of Pathan, and therefore resist it.

    Bruce, you're also going to find that it's hard to recruit Pathans for the military because they shall have to serve in a less-than-meritocratic institution with a Tajik ceiling on promotions, or at least that will be the perception.

    If the only gun club that then is hiring becomes the Taliban or related insurgencies, you will have young men of surplus employment going there and not to the ANA recruiter.

    The way around this would have been to have stuck with hard quotas, resulting in a smaller but ultimately more professional and competent and inclusive institution rather than lying about the mix and then realizing that one reason it is resisted is because the people realized long ago that it wasn't representative and, therefore, illegitimate.

    If you truly want to prod the nation toward sectarian civil war again, by all means create a military everyone believes is dominated by Tajiks and former warlords.

    With any luck, of course, perhaps desertions will skew toward the Uzbek and Tajik ranks this year, and the US taxpayer won't have to fund their replacements, thereby right-sizing the force AND making it more representative and sustainable.


  24. The other problem, Gulliver, is that there's some concern that we've reached diminishing returns on the units we're creating.

    There's a shortage of qualified, competent commanders, and we seem to have reached it. There's no point in creating units that will be poorly led, cowardly on the battlefield and something of an extortion racket in the villages to which they're posted.

    Nor are they likely sustainable, even in the short-term because of the turnover rate. The desertion rate has become a real issue and it's probably not something we can pretend isn't happening anymore.


  25. That's just not my experience, SNL II. The brigade I was with had a Jalalabadi Pashtun BGen as commander, a Kabuli Pashtun G2 and a Pashtun G3. They didn't see their ethnicity alone as a limiting factor in their careers... nepotism was, sure, based in some cases on connections from 30 years of military service that were not exclusively ethnically based. But those kinds of connections would have required years to fully understand and unravel and were certainly not Pashtun/nonPashtun-binary in nature.

    And they had networks of their own, too: they in turn tended to be friendlier with their Pashtun juniors. But they weren't any more overt about it than I'm sure I was when I tended to hang around with the Canadians and other anglophones at NATO meetings, or guys I had been on course with on exercises back home. We're really talking military peer group dynamics there, not equal opportunity hiring.

    Again, I don't see how you can expect absolute proportionality in hiring (as opposed to equal opportunity, which the ANA has in my experience) at the same time as the under-represented minority (or at least, its de facto leadership) is engaged in violently rejecting the very institution you're trying to hire them into.

  26. But that's exactly what the IG reported, Bruce.

    I often have the same temptation to base my assumptions on what I directly saw. But that's not an objective analysis of the problem.

    I was tasked out to what became perhaps the most accomplished IA battalion in Iraq. I assumed that all the battalions had to be as good as mine -- so meritocratic, so well led by promising junior officers and NCOs, so competent in their mastery of arms.

    I was wrong.


  27. SNLII made some of the points I wanted to make, including the one about the IG. You don't want to base policy on lessons learned/good practice which are an anomaly. When you're so close to it, you typically can't be trusted to recognize an aberration when you see one.

    But to get back to the importance of having representative forces and success in building a force I think that one of trickiest parts of all of this is balancing efficient/effective forces with legitimate ones. The tricky part is this: what makes a force legitimate in any given context and second, who gets to decide those factors? I think representation is one of these things because representation is a lot more complicated than it seems when we're talking a stable job, stable money, recognition, prestige etc. These things matter in a society and if you ignore them, even temporarily you'll be in trouble later.

    So say we don't worry about it and just say, we'll take just those guys. Well, what happens when things have calmed down enough that you want to make the force representative? You have a limited budget so you can't keep everyone on and hire a ton of others. So how do you decide which ones to fire, what about the costs of pensions or compensating people after you fire them? How do you mitigate risks from having all these likely pissed off, well-trained people running around?

  28. No, as I read the IG report it pointed to a basic lack of proportionality as regards the Afghan population as a whole. My point is that if you exclude the rejectionist subpopulations, it is not that disproportionate yet.

    This isn't Iraq, where there were a lot of Sunni Army general-officer types excluded from the new IA and sitting around because they were Saddam's men, and this was itself a cause for grievance: those with military experience in accepted institutions are still inside the tent. There's a lot of Pashtuns on both sides of this fight. If they or their fathers fought for the NA or Najibullah, they're inside the pale and competitive for promotion, because those two sides have made up. If they fought for Hekmatyar or Omar, not so much yet. That's not going to change until there's a political reconciliation.

    SNL II points to a limit on qualified people right now... I'd absolutely agree. Forget this is an army we're talking about for just a second: can anyone point to a situation where quota-based hiring (to redress the balance rapidly enough to matter you'd basically have to freeze army enrollment for all non-Pashtuns, at a time we're trying to grow the thing, but never mind) rapidly improved the *quality* of a workforce?

    As to the relevance of my experience, statistics without investigation can be misleading, too. I was also told before I deployed that the army officers would all be Tajiks, and the ANP would all be Pashtun... wildly wrong on both counts.

  29. Ugggggh... Bruce, you can't obtain a healthy, represenatational military that is legitimate to the people until it represents some sort of proportionality.

    If you take a force that's 70 percent Tajik and you make it the "local" force that's occupying and pacifying an overwhelmingly Pasthun province, you're going to have some problems. You might even end up with a sitrep that increasingly looks like a sectarian war to the people you're trying to woo (or coerce) to your side.

    You also might end up with true locals in the police force as infiltrated proxies for the Taliban because, like the Taliban, they're more representative of the people than the federal army.

    We had the same problem in Anbar.

    Eikenberry had it right six years ago. Had we properly implemented a proportional test for recruitment and advancement, it would have led to a smaller but more professional and legitimate Afghan army.

    But for our own cynical reasons we chose to push numbers over both legitimacy and quality. Today we have units suffering from almost a 30 percent desertion rate, considered as foreign to the village they've been sent as the US enabling forces bestriding them and we can't understand why the Afghan military isn't helping us pacify the population?

    Come on!

    Here's a statistic that's true and you can't get around: Seven out of every 10 battalion commanders in Afghanistan are Tajiks. How are we supposed to show the Pasthun that we're not favoring the Tajiks by inflating their role in the cabinet and military when they see that?

    They're not stupid.

    The sad thing is that we lied to ourselves about this. We didn't properly monitor the recruitment, promotion and retention of officers so today we have an Afghan military that actually is spurring the revolt instead of pacifying it.

    We've set our advisers up to fail, and that was unfair to them.

    We have brave advisers doing great work in the field, but the more they do the more the revolution grows. Why has the reach of the shadow governments gained on pace with the spread of the Afghan military to Pasthun enclaves?


  30. As to the relevance of my experience, statistics without investigation can be misleading, too. I was also told before I deployed that the army officers would all be Tajiks, and the ANP would all be Pashtun... wildly wrong on both counts.

    Fair enough but if that was your experience, then those who obtained that data and were presenting it to you weren't doing their jobs properly.

  31. Glottal exclamations aside, I don't think you're seriously considering the alternate possibility that if Eikenberry had imposed quota-based hiring, the ANA would be even more filled with incompetents in its higher ranks than it is now. Quota systems tend to do that.

    And again, you've got to see these guys in action. They're not considered "as foreign" as us. Many of them are effortlessly bilingual. The difference between the way locals in Kandahar Province treat an ISAF colonel and an ANA colonel, and the way they treat the populace in return, are night and day. They handle themselves better and get more cooperation in a village on arrival. What they do with that extra goodwill is often less than we'd hope--that's the competence problem--but there's no question they have it to start with. And the MoD seems to do what it can to put Pashtun officers in key positions in the south, for whatever reason... 2 of the 5 battalion S2s I directly worked with in Kandahar Province were Pashtun, which is pretty much your replacement rate.

    Really, it's not an Iraq Sunni-Shia thing... Afghan identity (particularly military identity) is much more fluid than that.

    I think you're also discounting that local hires could very well be more corrupt, and cast a national institution in more disrepute in many cases. Hazara ANP have a rep (deserved or not) as the least corruptible, and Pashtuns seem to like having them around as a result, rather than their own "ANP."

  32. You make the assumption that I haven't seen the ANA in "action" (and by "action" I mean the act of engaging the enemy with some competence and not shaking down the local populace, hobnobbing with a buddy warlord or doing their best imitations of Lazy, Shiftless and Dull).

    While Afghan identity might be somewhat "fluid," there's a distinct difference between a Tajik and a Pathan, and it would be absurd to suggest that there isn't or that there isn't a brewing sectarian war.

    If anything, because of the pan-Iraqi tribal systems, the long custom of intermarriage, the urbanization of the Iraqi state over the past four decades and a shared language (if not necessarily the exact same dialect), Arab Shiites and Sunnis (a very broad brush) have far more in common than a NW Afghan and a SE Afghan (randomly choose where to stick your finger on the map), and a far greater likelihood of having interacted (because of a globalized economy, conscription in the Baathist armed forces, etc).

    Perhaps this is because Iraq was something of a rickety state, but a state nevertheless, whereas Afghanistan for the past three decades has been lost within its own pathologies.

    You keep referring back to your experiences as if this Panglossian reality is shared by others on the training mission. I can assure you that it's not.