December 23, 2009

Rough times in the Arghandab

The Americans fighting in the Arghandab valley this summer had a rough go. Sean "Not a Good Day to Die" Naylor takes the first hit at explaining why. There appears to be a lot of masked discontent with the first bite of the Kandahar apple by the 5 BCT (and the lead unit, 1/17 Infantry). Naylor relates how first the US battalion commander apparently fires the lead company commander, and then the whole brigade got reassigned.

There's a lot of debate about whether large wheeled APCs like the US LAVs and their Canadian equivalents are good vehicles for the Arghandab terrain, which is very similar to the Zhari terrain, that frustrated Canadians for three full years (possibly even more restrictive, in fact). You've got to go dismounted. A lot. Going with tracked instead of wheeled APCs wouldn't help much, either: the real problem is that moving large vehicles past heavily irrigated orchards and fields (two words: foot bridges) on very narrow high walled roads means inevitably either destroying the irrigation, the fields, or the walls to some degree every time you do it, even if you weren't worried about the IED threat. In the Arghandab, as mentioned below, the main FOB (Frontenac) is farther away from the key terrain areas, on a canalized line of march, giving ample early warning and interdiction abilities, as well.

Combine that with trying to keep on side a population that until recently was strongly pro-government and you have a devil of a pickle. The article mentions some American officers were focussed on creating smaller platoon bases, presumably so they could get troops down into the valley at night; I know we had a devil of a time with that ourselves. If you wish to avoid expropriation, competing land claims (there is no Afghan land registry, so you tend to end up paying everyone and adding months of delay each time you try to buy something to put up a new base), and general unwillingness to have a FOB as a neighbour (you'd be a NIMBY too if there was the possibility of direct fire attacks every night) often leads to ISAF forces taking over the local school or district centre or some other public property location. Which is obviously not a great way to advance your development aims.

Note Naylor's article doesn't mention an ANSF contribution to the Arghandab fight at all: I suspect they were as limited in terms of support in this regard as the forces in Helmand, where a shortage of ANSF has often been remarked upon, were. I also found interesting the US focus on clearance operations (Canadian forces, in part to avoid alienating the government's last friends in the area, always insisted on Afghan troops for any sweeps in the Arghandab), and the belief of 1/17 going in that direct fire attacks would be a greater threat in Kandahar Province than IEDs. I honestly don't know how you could believe that, knowing anything about what we were dealing with a year ago.

Note also that 1/17 is being reassigned to RRS (Ring Road South) security, which is almost certainly an effort to get more ANSF off that task, which they were very effective at, but was demoralizing and kept them from participating effectively in operations that engage the local population. I've commented in the past that in a year when for supply chain reasons, only half-a-dozen or so new kandaks (battalions) and those at 40% of their assigned vehicle strength, were created, assigning the new battalions to highway sweeps (without vehicles?) seemed unwise, if necessary.

Keeping Ring Road South (aka Highway 1) open and IED free between Kandahar and Helmand, and up to Kabul and Herat in either direction is obviously key for our own logistics and Afghan government credibility. And by early 2009 it was being done more or less effectively. What was working was a combination of nighttime aerial surveillance by ISAF and early morning route sweeps by the ANSF, which would more or less prove the road until the next day.

IED layers can't work on the main highways in the daytime: these are busy roads except at night, and digging in an IED to be even reasonably detection-free takes time. Even if ISAF or the ANSF or private security didn't come along, you could never count on every person in every vehicle in the dozens that would pass every hour not pulling out a cellphone and trying to claim a reward. So you have to do everything at night; often over a couple nights.

Because the surveillance track is linear (you're just following a road), and IED activity fairly distinguishable on a main highway (unlike a route through a farmland area, there's little chance of confusion with farmers tinkering with local irrigation or foot bridges at night on the highway), in theory so long as your minimum time between night overflights (X) is less than the minimum time required to lay an IED (Y) you have near-perfect security. And for once we turn the tables on the dictum about insurgents, the one about how we have to be perfect every time and they only have to be lucky once... with a guarantee of good nighttime surveillance, now it's the IED layer who has to evade detection every single time he's out, and the requirement for luck shifts to our side.

The only remaining problem as of last spring, when nighttime surveillance assets were finally available in sufficient quantity in RC-South to achieve this sort of thing over large parts of the highway system, was the possibility of ANSF-ISAF disconnects between the night surveillers and the morning sweepers... obviously sub-optimal. Given the lack of any shared command structure, if the aerial surveiller saw something unusual but not necessarily definitive, it was very difficult to vector in an ANSF asset to that precise location to investigate, or even urge greater caution upon them, before they set out in the a.m. to check out the road... and vice versa. So it's likely a well-resourced ISAF lead on road security, in addition to freeing up more ANSF to deal with the Afghan people rather than search Afghan culverts, could be even more effective and cost fewer lives all around, than using Afghans was, purely because it will be easier for aerial and ground-based sweeps to compare notes every day. I wish them luck.

Posted by BruceR at 02:34 PM

December 16, 2009

On Forsberg's first draft of Kandahar history

A comprehensive review of available open-source reporting of the fight in Kandahar Province, by Carl Forsberg. It's definitely worth a war-follower's time. His summary of my own roto's achievement in 2008-09 regrettably does ring rather true:

ISAF operations in late 2008 and early 2009 did not have lasting effects on the enemy system in Kandahar. Given the short duration of ISAF operations, Taliban fighters could easily move to safe havens several kilometers from where ISAF operated and return to their original location on the same day that the operation concluded. ISAF’s failure to protect the population in these areas left local villagers under the control of Taliban intimidation and governance. Unable to establish a sufficient presence to separate the local population from Taliban intimidation, ISAF suffered from limited intelligence about insurgent activities. In addition, ISAF’s focus on Zhari, Panjwai, and Maiwand during this period targeted areas of secondary importance to the enemy campaign of taking Kandahar City.

Yep, that pretty much sums up how we're going to be remembered by the historians, I'm afraid.

I would certainly recommend it over Michael Yon's lesser effort at a Kandahar synopsis, which drops a few clunkers:

...there are indications that the enemy is today in disarray. The enemy became afraid to sleep indoors where they might be killed by an airstrike—or by U.S. soldiers, who have a tendency to burst in during periods of maximum REM sleep. The Taliban were terrorized and began sleeping in the orchards at night, rigging homes with explosives, which they arm at night.

As said previously on Captain's Journal, this isn't evidence of any new terror on the enemy's part. They do it every summer. Orchards and tall fields provide near perfect cover from UAVs and ground-based night vision, are perfectly comfortable to sleep in in summer, and they know Westerners tend to overfocus on structures in our kinetic operations: the stories of Taliban watching from the fields as the SOF guys take down the house they were supposed to be in are legion. Stringing a grenade on the entranceway if you can on your way out is only logical. Similar behaviour was noted just as much in both summer 2007 and 2008.

What's new to Michael Yon and his U.S. sources about the war is not necessarily new, I'd say. I'll leave his comments on the Canadian Forces being "militarily defeated" in Kandahar for others. But certainly Forsberg's more measured assessment -- that we have been ineffective in much of what we tried to do there -- is also the more strongly argued. (A good example is Forsberg's comments about the suboptimal placement of FOB Frontenac, which he suggests came about because of a Canadian fixation with its Dahla Dam signature development project at the expense of establishing actual provincial security. I don't know that I agree, but it's certainly at least debatable.)

Or if you've given up on attempts at serious analysis, you can always watch this. Special Forces working in Zabul appear to have created a new crime: DWA. Driving While Afghan. "Bad guys or bad stuff?" Kee-ryst.

Posted by BruceR at 10:45 PM

Canadian quarterly report on Afghanistan: security highlights

Not a lot of progress to show on the security side, this time over the last report from three months ago. (Always nice to see realism replace optimism, though.):

**A significant drop in ANA personnel numbers (or possibly more accurate reporting)... three months ago, 1/205 Brigade had 3 kandaks rated at 70% strength or higher, now that's down to one.
**No change in ANA readiness measures... still 1 kandak (battalion) rated at CM1.
**A significant drop in ANA popularity... three months ago opinion polling in 5 of 6 districts had the ANA above 85% popularity, now it's down to 1, apparently.
**No change to opinion polling on whether security is improving in general.

The one stat where there has been improvement, unsurprisingly, was the percentage of "ANSF-led" operations. As discussed previously, this is not a particularly revealing statistic, and very difficult to objectively quantify over time in the Afghan circumstances.

Link to previous report.

Posted by BruceR at 03:24 PM

Um, so, we found them, did we?

Canadian Press, yesterday:

Foreign Affairs Minister Lawrence Cannon has denied a published report that Canada lost track of war prisoners its soldiers transferred to Afghan authorities.

Cannon says Canadian authorities can account for all detainees transferred to Afghanistan, but he admits there have been delays in tracking some of them down.

Richard Colvin, today:

As of early October, 2007, when I left Afghanistan, we had not been able to locate the remaining three [missing Canadian] detainees.* However, according to good sources, they were likely in Kabul, but at an NDS ‘black site’ to which we were not given access.

Also of note in Colvin's letter today, with regard to other alternatives his office had proposed at the time to the handover of all local detainees to the Kandahar NDS:

The second option was to transfer our detainees to Kabul. This was the Dutch solution, which they implemented from early 2006. Away from the front‐line violence and pressures of Kandahar, detainees could be monitored and kept free from harm. The Canadian embassy proposed this solution, but it was rejected on the basis that detainees would take up too much space on C‐130 flights to Kabul.

*Short story: Colvin's referring to 4 detainees in Kabul the NDS had offered to Canadian officials in June 2007 as possibly corresponding to four that had been reported to Canada as having been transferred there by the NDS. Colvin goes on to report that only one actually matched up with an existing photo of a previous Canadian detainee.

Posted by BruceR at 02:28 PM

December 14, 2009

Blatchford: time to put down the pen/mouse/stylus on detainees

I have a lot of respect for Christie Blatchford, but she's officially too close to this detainee story. Paul Wells has previously observed this, too, but that was before her story this weekend.

On June 14, 2006, a Canadian Military Police officer who was working with the Afghan National Police was on the scene when the ANP stopped a van leaving a battle. The ANP said one of the three men inside was definitely a Taliban. The MP photographed the man and wrote his name down, but agreed to let him travel with the ANP back to Patrol Base Wilson. It was a 15-minute trip. Back at the base, the MP dutifully checked on the fellow and found the ANP beating him with their shoes. The MP then took the man back and made him an official detainee.

Nice story, but it's contradicted by some facts previously in evidence. Specifically the section commander's report given to the CDS last week. The section commander, as far as we could tell, was not military police. If there was military police present at the initial custody, he certainly wouldn't have referred to sending the man back to the FOB so that "watchdog" (the MPs) could have access to him. That's what an infantry soldier would say. His callsign, Orion 22, also suggests this was a regular infantry section. The section commander did not just "ask questions and take a picture," he conducted a detailed personal search and a vehicle search and catalogued the man's effects. His statement as read by the CDS also seems to indicate the traffic stop was made by Canadian Forces and an interpreter, not the ANP as in Blatchford's telling.

Now, these are fairly minor quibbles that may not indicate anything much (I have yet to hear anything REMOTELY negative about how Canadian Forces personnel handled this particular incident), but these are facts that were in the public documentary record as of the CDS's press conference Wednesday. Blatchford, writing a couple days later, but having received her own account of the story from a military contact, did not evidently take the time or feel the need to in any way deconflict that source's story and what was already known, and instead just repeated her own anonymous hearsay version verbatim to readers. That kind of "too good to check" stuff should not really be seen as acceptable journalistic practice, and more than likely indicates that Blatchford is letting her bias interfere with her news judgment. Which, in the same piece, she admits being prone to on this kind of story, by the way.

In case all the staff in the Globe missed that day in J-school, this is how it's supposed to work. When a writer admits to a personal bias in her stories, stories that at the same time contradict the known facts and favour those biases, well, it's probably time for them to write about something else.

UPDATE, Dec. 16: The actual handwritten document, which confirms the individual was not an military policeman, and that Canadian soldiers were directly involved in the initial traffic stop, is also linked from here.

Posted by BruceR at 06:38 PM

Detainees: who wants ginger snaps?

There's a Simpsons ep where Homer is an astronaut, and long-suffering wife Marge is watching his spaceship's re-entry on TV with the kids. Contact is lost, and Marge says, "Don't worry, I'm sure everything will be fine." When Lisa asks, "what are you basing that on, mom?" she pauses for a moment, and then shouts, "who wants ginger snaps?"

As of this weekend, I'd say media supporters of the government's handling of the Afghan mission are in decidedly ginger-snappish territory.

Let's start with Susan Riley in the Citizen.

But the problem is fixed now. That wouldn't satisfy everyone, but many Canadians -- who have scant sympathy for Taliban suspects and abiding respect for the troops -- would have accepted it. Crisis defused.

What are you basing that on, Susan?

Look, here's the facts. The NDS is still the investigative arm of the police. We don't watch them closely, but they likely haven't changed. All Afghan insurgent-related detainees to this day still are at their disposal eventually. Also, the vast majority of detainees taken in Kandahar Province up to eight months ago were taken by the Afghan security forces without documentation, whether ISAF forces were also involved in their capture or not. No one, Afghans included, has any real idea where any of those people are now. The protocols we have with the Afghan government only govern the smaller number who spent time in the Canadian detainee cage in KAF before the Afghans took them, because those are the only detainees we have followup responsibilities to under current interpretations of our Geneva obligations.

The Afghan government's detainee apparatus I worked with was simply not designed to generate POWs, or intelligence for that matter. It presumably was more successful at generating ransoms, given its revolving-door nature at the time. And sure, maybe things have gotten better that way, but in April 2009, in my last report after watching this through the ANA G2's office for over half a year, I said I was watching them get worse by the month. Hey, T.I.A. (This. Is. Afghanistan), as I've said before: the fact that getting taken prisoner often appeared to be a 72-hour time out pass for those who could pay promptly was nowhere remotely NEAR the biggest problem in Kandahar at the time. But the idea that all Afghanistan's detainee issues were "fixed" two years ago and remains anywhere near fixed today, by any definition you want to use of the word "fixed," could potentially be contradicted by any of a thousand Afghan veterans with stories to tell if the PM or the Defence Minister ever pretended it was all fine and dandy, and I don't believe they are all die-hard Conservative voters. Oh look, here's one now.

In short, Riley's recommendation on the government's best media talking point on this one likely would not have been the best tack for them to take.

UPDATE: I personally think Jeff Simpson is persuasive on the better course of action on this issue.

Posted by BruceR at 05:10 PM

December 10, 2009

People-who-don't-know-what-they're-talking-about Watch, part 2

Adam Rawnsley for Wired:

Taking a filmed “victory” lap around an American facility might make some sense as a propaganda tool. But taking a giggly victory lap on an elliptical in said base – as two Taliban enthusiastically do in the video – is a big infowar fail.

This guy doesn't get it at all, I'm afraid. I can say fairly authoritatively the average Afghan response to this video will be, "here I am without electrical power 5 days out of 7, in winter nights no less, and those American twits had a machine running to simulate walking uphill. These insurgents are looking better and better." More here.

Posted by BruceR at 08:44 AM

December 09, 2009

Detention vs custody

Re the 2006 Zangabad detention referred to by the CDS today:

A good piece here hints at some of the other issues just starting to surface:

British and Dutch forces, who followed the Canadians into southern Afghanistan, were "deeply frustrated" even though their agreements with Kabul allowed them more access to prisoners.

"UK/Dutch pol/mil colleagues lament that they are unable to track their detainees," said a Dec. 4, 2006, memo viewed by The Canadian Press.

"It is unclear whether they are tortured, held beyond legal limits, or (all too frequently) released back to battlefield."

The Allies were worried "the detainee issue could explode at any moment into a political firestorm."

Detainee release "back to the battlefield" by Afghan authorities is the flip side of this debate. Afghans tend to keep very little in the way of records of significance of the disposition of those they've detained, enabling the taking of bribes for release of even well-known insurgents.

At the time, officials were putting a lot of stock into the idea that the Afghan government would take the handling of detainees away from the National Directorate of Security and hand prisoners over to a special camp run by the army.

But Afghan Defence Minister Abdul Rahim Wardak eventually sank the proposal.

In my experience, the Afghan Army at all ranks tried to have as little as possible to do with anything that looked like a detainee. Not their line of country, so sorry. Can't really blame them, either.

Posted by BruceR at 10:04 PM

People-who-don't-know-what-they're-talking-about Watch

Defence analyst Allen Sens says the contract for the surveillance aircraft indicates that Canada’s Afghan military mission is not yet over.

“This seems to support the idea that we will be staying on with a military mission,” said Sens, an analyst with the University of British Columbia. “I was always under the impression we would continue with some kind of military presence such as JTF2 ( the military’s elite force), PRT assets and a headquarters battlegroup.”

Ottawa Citizen. Some analyst: "Headquarters battlegroup" is a nonsense phrase. If he'd said "...and an oatmeal lobster" it would have made about as much sense in that context. So, either UBC's defence analyst knows nothing about defence, or the Ottawa Citizen's defence reporting team of O'Neill and Pugliese doesn't, or both. Your pick. (You should probably judge the rest of Sens' "analysis" accordingly.)

Posted by BruceR at 04:02 PM

...had to stop in my tracks for fear, of walking on the mines I laid...

The CDS "dropped a bombshell" today, says the Star. (It's always jarring to see military-themed figures of speech used by the papers when discussing actual, you know, military people doing their jobs, isn't it?... next they're going to say he "drove a tank" through the committee room and I won't know what to believe...)

As alluded to at the end of this post here and previously, this is a very deep rabbit-hole, one I fear that we're not even partway down yet. In retrospect, one imagines the "no confirmed case of Canadian detainee torture" line might end up as the most ill-advised military talking point put into a minister's mouth since the Somalia days.

UPDATE: This just keeps getting better. On poor tactics, see also Paul Wells. The smarter approach on this was always the one we've started to hear from the PM and others again today... our own soldiers did nothing wrong. Which, in truth, they certainly appear to have done... every new detainee anecdote is only making them look better and better. It's the relying on a PR attack point that could have been potentially impeached by any credible evidence of Afghan misbehaviour, past, present, or future, that now must just seem... unwise. (EDITING NOTE: Some other ruminations hived off into a later post of their own.)

Posted by BruceR at 02:55 PM

December 07, 2009

Money, meet mouth

Yes, that's me... having a thoughtful discussion about Afghanistan with Robert Wright on A smart guy, that one.

Posted by BruceR at 06:32 PM

Um, what?

Spencer Ackerman:

[Undersecretary of defense for policy Michele Flournoy] said that the administration is in ongoing dialogue with Canada and the Netherlands, two allied countries scheduled to remove their troops from Afghanistan in 2010, about possible continued contributions to the war effort. “All options are still on the table in discussion,” Flournoy said.

Posted by BruceR at 02:33 PM

More essential stuff

The always essential Nir Rosen, on patrol with the Helmand ANP. He ruminates about the biggest difference between Iraq COIN and Afghan COIN:

More fundamentally, COIN helped to control violence in Iraq because sectarian bloodshed—which changed the conflict from an anti-occupation struggle to a civil war, displaced millions, and resulted in the deaths of hundreds of thousands—was already exhausting itself when the Surge started in 2007... In comparison with al Qaeda in Iraq and Shia gangs, the Americans looked good. They could step into the void without escalating the conflict...

In Afghanistan, there is no comparable exhaustion of the population, more than two-thirds of which lives in hard-to-reach rural areas. In addition, population protection—the core of COIN—is more complicated in Afghanistan. The Taliban only attack Afghan civilians who collaborate with the Americans and their puppet government or who are suspected of violating the extremely harsh interpretation of Islamic law that many Afghans accept. And unlike in Iraq, where innocent civilians were targeted only by predatory militias, civilians in Afghanistan are as likely to be targeted by their “own” government as by paramilitary groups. Afghanistan has not fallen into civil war—although tension between Pashtuns and Tajiks is increasing—so the United States cannot be its savior. You can’t build walls around thousands of remote Afghan villages; you can’t punish the entire Pashtun population, the largest group in the country, the way the minority Sunnis of Iraq were punished.

Rosen's observations of his time with the ANP are cutting. Every mentor will empathize with the frustrated American PMT he rides along with:

The sun rose golden over the shrubs as we made our way back to the checkpoint. The police had mentioned seeing a Taliban car. “What was that about a Taliban car?” Sergeant Ryan Killacky, of Prowler, asked. “The ANP think everything is Taliban,” Westby replied, “I don’t think they f---in’ know.”

Some Canadian mentors are telling a similar story in their province. In Kabul, meanwhile, the London Times contrasts a successful Afghan army HA (humanitarian aid) drop with the ANP's problems in that neck of the woods. Short version: improving quality and growing quantity at the same time in a warzone is hard, and a lot of the problems that bedevilled mentoring from prior to this year are only now being fully acknowledged.

Lest anyone think the above puts the ANA on a pedestal, this clip of British mentors with their ANA in Helmand might also look familiar to other Afghan vets.

Posted by BruceR at 08:29 AM

Done its dash?

There has been a very interesting ongoing dialog between influential counter-terrorism researcher Leah Farrall and Al-Qaeda fellow traveller Abu Walid al-Masri recently.

In the process, we have discovered we agree on some things: most notably that al-Qa'ida has done its dash in the Islamic world. Abu Walid believes al-Qa'ida's actions have caused more harm than good. The terrorism war, he tells me, has proven "to be far from the mood of the Muslim people and the result has been popular hatred towards it".

More here.

Posted by BruceR at 08:15 AM

Today's essential Afghan reading

"Old Blue", writing from Kabul:

Many leaders that I’ve met in my travels say, “Oh, yeah! I took the COIN course. Good stuff!” So I suggest that we look at their ASCOPE and see how they’re doing on it, where they are having problems identifying key players, etc.

“Oh. Well, we don’t have time for that.”

Really? No time for the steppingstone behavior to not only learning about the operational environment… but to actually passing it on to your successor? No time for that. Great. So, the thing that units have been complaining about for years… that they come in with no real understanding of the people and key systems in place in the local communities… will simply continue. Some of the other instructors say that they have run into units who are actually documenting their environments, but I personally have not...

"Old Blue" was previously a police mentor in the Tagab, recently cited by Col. Haynes in that paper I liked as the most prominent ANSF-led security success story in the east. Now he's back at the Afghan COIN Academy as an instructor. The corporate knowledge is building, slowly, but surely. His blog, formerly known as "Bill and Bob's Afghan Adventure," now called "Afghan Quest," is worth the time.

UPDATE: I would be remiss, in a post about the Tagab, not to mention that Josh Foust had a considerably more pessimistic view than Col. Haynes about the ANA efforts in that area in 2008 (see also here and here).

Posted by BruceR at 08:05 AM

December 04, 2009

Looking for some Afghan graphs?

Tony Cordesman's CSIS group has collected, um, all of them, I think.

Posted by BruceR at 05:09 PM

What happened to my doughnut?

The Globe and Mail's Patrick White, who is evidently still getting his chops as a war correspondent, amused me this week by putting forward the headline, "Canadian troops to form doughnut of stability," referring to efforts around Kandahar City. At some point later, sober individuals prevailed, and "doughnut" was changed to "ring."

To be fair, since there's no real approaches to Kandahar City from the Northeast, it really was more like a croissant of stability, anyway...

All the indications are that the effort is being placed where it should be, that being in the semi-rural districts around Kandahar City, rather than in Kandahar City proper. Joe Klein's fever dreams aside, it's hard to see how more Western forces inside the city proper would help.

That's largely because of the vehicles. The city itself is densely packed, and there is no bypassing highway... everything going west to east or vice versa through this part of the country goes through the city centre, on a two-lane road (with a partition, thankfully). Daytime traffic is always congested... night time is much better.

There's already massive numbers of Western vehicles on that road. And, through no fault of the drivers that means they sometimes scrape foot carts, hit bicycles, or run into ditches and stop traffic getting pulled out, for hours. Or get attacked: there's that, too.

(A related problem is remote weapons stations, like on the RG-31. The RG is a great vehicle for highway patrol. But despite the bullet-proof windows, visibility to the side and rear is extremely limited except through the gunner's viewscreen. Unfortunately, unlike the periscope on a tank, which can swivel independent of the gun, a remote weapon station's camera is normally slaved to the gun. Which means if you're trying to watch behind you you have to point a .50 calibre machine gun at everything you look closely at. Which especially given that an rooftop RWS being used effectively looks a little like something out of a Terminator movie, elevating and swivelling jerkily and robotically around, can be unintentionally intimidating. I'm not sure the average Afghan knows there's a person operating it. And whereas a vehicle like a LAV can use humans standing in rooftop hatches as "air sentries" to watch all around it instead, the RG and other mine-resistant vehicles tend to be hermetically sealed against overpressure, so they don't have that option. Overall, RG-31s and similar RWS-dependent vehicles can't help but look hostile to the average person. They're simply not good vehicles for crowded cityscapes, I'm afraid.)


It's all well and good to say there's value in more presence patrols in the city. But the only kind that is likely to have a positive impact is the dismounted kind. Mounted presence patrols exert a presence, all right, but it's an annoying and intimidating one. The trouble is, to get more dismounts into the city will also increase the amount of vehicle traffic getting them in and out and around.

What Kandaharis would really like, other than reliable electricity, is to be able to get through the night without the sound of artillery fire, gunfire, or loud explosions. It's the areas around the city, which can be used by insurgents for their campaigns of assassination, intimidation, and bombing in the city proper, that would benefit the most from a greater Western presence. In the heart of the city, Western troops, and particularly their vehicles, are only ever going to be at best a necessary evil in limiting that. The people don't see the attack we might have prevented from happening, they just see that they missed an appointment, or that we hit somebody's pet goat.

Having read "The Good Soldiers" recently, I'm not totally convinced Western troops were that valuable in Iraq, either, at least when they were mounted. Certainly, there is value in constant contact between Afghan and Western forces... having an infantry section or platoon 24/7 at each major police or army installation in the city to provide a "hard seal" between Western and Afghan efforts would be very helpful, I'm sure. So long as they don't have to drive around in the day too much.

Posted by BruceR at 09:03 AM


Kevin Drum asks "what's the plan?" in Afghanistan. I don't feel the same disconnect. I thought Obama, when read with Gen. McChrystal's previous staff work, doesn't leave much in the way of ambiguity, actually.

The key date is July 2011, and the deliverable the "beginning of the transfer of security responsibilities to Afghan forces. Not winning the war, or fixing the country, or something else equally ephemeral.

It's always been a key deliverable, long before the President's speech. During my tour, the phrase was "Transfer of Lead Security Responsibility," or TLSR. The question was always how far away it really was: it was always a real moving target of a deadline. Previously in this space I've tried to indicate some reasons why. Col Haynes' paper gives another good summary of the problems.

To fully get this, you have to understand that as of a year ago there were no military boundaries between Coalition and Afghan-controlled areas. There was near total overlap. No distinct Canadian or Afghan AO (area of operations). The reason being that Afghans weren't considered ready enough to take on the insurgents, or even know roughly who and where those insurgents were within a given geographic area, even with ISAF mentors in overwatch.

But the fact that Afghan soldiers and police were still necessary under ISAF and Afghan rules for door-kicking and detainee collection duties on all operations, but believed to be incapable of surviving long on the battlefield alone, left ISAF/ANSF with little choice other than this sort of bifurcation of the "monopoly of force" an armed force normally enjoys, if anyone from either army was going to ever leave a FOB.

Also completely missing, as Col. Haynes pointed out, was any kind of memorandum of agreement between the two forces similar to what there was in Iraq, which allowed the one army's commanders tactical control of the other under the right circumstances. Instead, what you had was an atmosphere of total ad hockery. They had their plans, and we had ours, and sometimes if the mentors were on the ball we'd show up at the same spot for the same reason. At best, they were our adjuncts, there to support our operations by providing that interpository layer between us and the Afghan people. But unity of effort or shared intent were completely theoretical.

It is the belief of many military mentors that the sum total of these effects has been the perpetuation of tendencies towards passivity and dependence within the ANSF. Gen. McChrystal's review concurred with this, saying that the forces must actually start to work closely together now, not just continue the lip service: enabling Afghan operations, not just dragging them along on ours. This change in attitude towards ANSF development is the most important part of his new strategy, separate from any change in troop levels. As McChrystal himself said, in the absence of the attitudinal change, any troop increase is irrelevant.

The end state here, at least within the parts of the country the government is trying to hold, is akin to what was supposed to happen in the early 1970s with Vietnamization... Afghan ground forces, supported and backed up by Western enablers... ISR drones, helicopters, fast air, casevac, highly accurate artillery, delivered to them through control relationships or imbedded mentors. Everyone agrees that gradually the part of the country patrolled on foot and in vehicles by Western soldiers needs to decrease... this is the way to get there. First, drop the bureaucratic and operational barriers to full cooperation. Then start handing back the country, district by district and province by province, with only Western mentors or their military equivalents (albeit with radios that can call down whatever's needed) remaining. The areas where a Western battalion has the lead will correspondingly shrink progressively, removing them as a presence from populated areas.

That means there will still be lots of Americans and ISAF troops post July '11. But they'll be less visible... and in a real sense working for and with the Afghans, not around them. Much of the sense of being an occupying army will hopefully be lifted without losing too much actual fighting capability. So that's the plan, Kevin.

(The plan will likely be combined with an expanded effort in the more remote areas outside the growing ANSF ambit, involving SOF, Special Forces Gant-type tribal engagement teams and the like, to keep the enemy from getting comfortable there, as well. The two are complementary, not contradictory: the one approach suited for where there's a Afghan population to defend, the other where there isn't. These more remote areas could remain with coalition forces having the security lead in that sense for a very long time yet.)

One of those enablers we're going to continue to provide for a long time is ISR (intelligence, surveillance and reconnaissance). Now, there's one complication with that plan, and for that I need to show a picture.

This is the 1/205 ANA Brigade's intelligence desk on deployed operations. Seriously, this is all of it. There is no more. 1/205 was responsible for fighting the insurgency in Kandahar Province, which has roughly a million people. What you are seeing is all the tools, all of them, the Afghan army had at the time to find their enemy and evaluate the effect they were having on them in that province. Consider that when we're evaluating how far we are from TLSR.

Now, there's a reason for that... the Afghans aren't stupid. They don't feel a need for huge intelligence capabilities in their army brigades, because those brigades are expected to get all their intelligence support from somewhere else. No, not us... when I was there we weren't allowed to share squat. From their own National Directorate of Security, or NDS. The NDS guys actually know what's going on, actually keep files, and the like. At least, we think they do... during my time we had very little interaction with them, and they were too good as secret police and intelligence agents to volunteer much to us, either. But they did seem to find the bad people reliably, and know when we'd captured a couple, using undoubtedly a combination of humint, interrogation, and decades of accumulated street sense.

The trouble is, the NDS ARE a secret police in every sense, with an unsavoury reputation, as the recent Colvin allegations in Canada are reminding us. How, exactly, do you interact with such an organization, without bringing guilt on yourself by association? If you don't, however, we can never really provide the kind of ongoing support from our much more complex intelligence resources to the Afghan security effort... if the army has no capacity to ingest those kind of inputs, and we're shunning the Afghans who do, who, exactly, would we support? It was a problem we never found a solution to during my time there, sadly.

Posted by BruceR at 08:19 AM

December 02, 2009

Today's essential Afghan reading

NY Times:

But the far more worrying prospect is the quality of the Afghan troops and officers. While many Afghans have demonstrated an eagerness to fight the Taliban, the Afghan Army and police have shown themselves unable to maintain themselves in the field, to purge their ranks of corruption, to mount operations at night or to operate any weapon more complicated than a rifle.

One example often cited by American trainers: the bureaucratic skills and literacy levels necessary to administer a large force have not materialized, even after years of mentoring. When it comes to paying their soldiers, keeping them fed, providing them with ammunition and equipment, tracking who is on leave and who is injured, most Afghan units perform very poorly. These tasks — essential to the readiness of any army — are almost invariably performed by American or NATO soldiers.

Indeed, American trainers often spend large amounts of time verifying that Afghan rosters are accurate — that they are not padded with "ghosts" being "paid" by Afghan commanders who quietly collect the bogus wages.

"The focus of the training program has always been 'more soldiers' at the expense of quality training," said an American involved in training Afghan forces, who demanded that his name be withheld because he was still working with Afghan soldiers. "There are no 'tests.' A soldier does not have to master any task prior to graduating. Attendance equals graduation."

All too, too true, sadly. Without a competent Afghan army, the President's plans last night cannot succeed, and he will likely be held to one term. And that depends on a marked improvement in the effectiveness of Western military mentoring, which is now officially all I ever talk about. It's interesting, being courtside on history being made, like this. Anyway, stay tuned; this ride is just getting interesting.

Posted by BruceR at 08:57 AM

December 01, 2009

Calm down, Fred

I've called Slate's Fred Kaplan "hysterical" before this. So this doesn't come as much of a surprise. Discussing the worst case in Afghanistan:

As with confronting most messes in life, the initial impulse is to flee. But if we simply pulled out, it's a near-certain bet that the Taliban would march into Kabul, and most other Afghan towns they'd care to, in a matter of weeks.

I don't know anyone who really believes that. A lot of people think the place would return to a state of civil war in a matter of weeks or not days. I've previously said the army would rapidly revert to its Northern Alliance roots and the ANP in places it was unpopular would likely dissolve. But it'd take a while until the Taliban were back in Kabul in any scenario.

The piece Kaplan cites as his evidence for the quote above, by the much more level-headed Steve Coll, doesn't back him up either:

It is inconceivable that the Taliban could triumph in the country completely and provide a regime (however perverse) of stability... [withdrawal] would likely produce a long-running civil war between northern, Tajik-dominated ethnic militias and the Pashtun-dominated Taliban...

For Kaplan to take that statement and translate it to "a near-certain bet" that the Taliban would conquer Kabul again in "a matter of weeks" is the sign of a hysterical mind.

The trouble is that if you go around with that kind of overwrought scenario as your worst-case, you're not in a good position to make sound cost-benefit decisions about the current options.

Not considered is how little possibility the Taliban would have had to go anywhere in the 1990s if they had been opposed by the Northern Alliance and U.S. air power together, even in the absence of ground troops. Ironically, while Vietnamization (local forces backed by Western air power) was wrong in South Vietnam, it could conceivably work much more effectively in Afghanistan, where the friendly forces do not have a much more powerful neighboring nation to fight off as well. A new Northern Alliance supported by special forces advisers, air support and diplomatic checks on Pakistani interference might not ever defeat a Pashtun insurgency on their own, but its hard to see how they wouldn't greatly slow any Taliban resurgence, possibly even reverse it. The worst case in Afghanistan is not as bad as some are making out at all.

Posted by BruceR at 08:25 PM