July 26, 2010

On the road

Out of comms for a couple weeks. Much as I might love to provide context to the Wikileaks stuff, that'll have to wait.

Given that the Secret-classified account of what really happened for nearly every single Coalition fatality and serious injury is available to be reviewed, information that in many cases might not have been available to the families before now, I expect this will percolate for some time, though. It's certainly going to be a treasure trove for future historians.

But people should be mindful this is still just largely a summary of first-reports, based on relayed messages from headquarters at least two levels lower in many cases, however, and is often going to be inaccurate. For instance, these three reports (1 - 2 - 3) are all of the same incident on Dec. 26, 2008. (I was sort of in the vicinity at the time.) So the Wikileaks count of 4 Afghan KIA and 5 Afghan MIA for the three incidents together that day is exaggerating by a factor of three. Wikileaks has added its own sometimes erroneous interpretations, too: just looking through the incidents I was involved in, I've also noted a couple undoubtedly insurgent fatalities in the Canadian AO during my roto classed as civilian deaths, a civilian encounter with an old Soviet minefield classified as an IED strike, and so on.

A lot of this is pretty unavoidable fog of war, first-reports-are-always-wrong stuff. Just as one for instance, here's one report in the database based on information I was personally involved in passing on.

ANA (2/1/205 [Kandak]) with CAN OMLT [that's us] conducting a clearance patrol ISO [in support of] OP[eration] ATAL 47 found a cache consisting of: medical supplies, DIShK [12.7mm] AMMO (approx 1000Rnds), AUP/ANA uniforms, various IED making materials and yellow smoke grenades. EOD [Explosive Ordnance Disposal] is on site and is exploiting. FF [friendly forces: ie, the Afghans] believe this to be a INS C2 center [Command and Control, ie, a headquarters].

Here's what I actually wrote later in my (unclassified) event summary after I'd straightened out the facts.

15. (U) 29 Dec 08: At 1405, elms [elements] of 2 Kdk [Kandak] found a weapons cache in a grapehut at QQ 3946 9986. The grapehut was later rubbled by CF eng[inee]rs in the process of destroying the cache. Nearby in a field at QQ 3971 0007, two 18L jugs of HME [home-made explosive] were also found in separate holes, with a roll of white wire nearby, in a way that suggested the items had been hidden hastily. Found at the first location (the grapehut) were:
a. 4 kg HME (ammonium nitrate with aluminum powder);
b. 7x 82mm recoilless rifle rounds, not all still functional;
c. estimated 1500 x small arms rds (mostly 12.7mm, loose);
d. 1x case, yellow smoke grenades (US; half full; kept by ANA);
e. 2x improvised pressure plates;
f. A pair of AUP-issue pants, an ANA tunic and a load bearing vest;
g. A pair of black boots;
h. A small quantity of medical supplies, including bandages and syringes;
i. A small quantity of potential IED components, comprising wire, batteries
and ball bearings;
j. Small quantities of marijuana and opium;
k. Documents, later ascertained to not be of intelligence significance;
l. Some pieces of casings for mortar and RPG rounds.
G2 Mentor Comment: It was at this point in the op that the G2 mentor [yours truly] learned that the ANSF uses the same word for “cache,” “depot” and “headquarters”. Misunderstandings due to this initially led to some confusion in reporting.

So in short: in Wikileaks it's reported the ANA found an insurgent headquarters that day, largely because when we passed on the information my colleagues and I either weren't catching or effectively relaying the nuances of what my Afghan colleagues were telling us. In reality 2 Kandak found a kishmishkhana (a drying hut used for preparing raisins, a common feature of south Afghan farmfields) with a couple ready-to-use IEDs and a lot of other leftover crap inside; the Afghans took what little they could use, and then the Canadians blew the rest up. Still a good day's COIN work regardless (in that it may have saved lives and didn't cost any), but hopefully it shows (if it needed showing) to never trust first reports from a battlefield. Wellington's comment about battles and balls comes to mind here, too.

PS: Brace yourself for another round of detainee reporting, btw: this remark in the NYTimes version of this story should pique a few Canadian reporters' interests from the perspective of what it means for the detainee issue: "From 2001 to 2008, the C.I.A. paid the budget of Afghanistan’s spy agency (the NDS) and ran it as a virtual subsidiary." Hmm.

Posted by BruceR at 01:59 AM

July 23, 2010

Today's essential Afghan reading: Tim Lynch again

Baba Tim encapsulates the key problem with any kind of "arm the tribes/auxiliary police" approach.

Reconstruction is not hard, establishing credibility is and that takes time in countries like Afghanistan. It also takes people who can operate on their own on the economy and not just survive but continue to function if attacked. That kind of thinking is not found inside the closed loop of the classified crowd. They do not know what they do not know. They can’t leave the FOB’s so they don’t have an accurate read on anything except what comes through the classified loop. Anyone who has dealt with that sort of information understands how limited it is.

Which bring us back to the Local Security Forces. This “inspired” idea of using locals to provide security will fail because nobody responsible for it will get off the FOB to provide daily detailed supervision. I can’t stress enough the importance of daily, full time, supervision. The Skipper’s EOD program works because he provides daily, detailed supervision, while EOD programs elsewhere in the country languish.

This is the real problem with these sorts of approach. If you don't have at least as much Western mentoring and liaison (I don't think we'll ever achieve "supervision" of security forces) as you do with the police now (and mentoring of police even a year ago in Afghanistan was in many parts of the country essentially drive-by already) then any effort and money will be wasted. But if you do have that kind of partnering of Western and Afghan forces, then why not give them the added imprimatur that comes with a formal place in, and paying them as if they were, in the "national" security forces?

Supporters see the auxiliary police or "arm the tribes" approaches as a new form of force augmentation where Western support and awareness can be limited, Afghan government and ISAF deniability can be preserved, and the newly armed locals can assert they don't have to take other people's orders. But any force you establish that carries those limitations is highly unlikely to contribute to anyone's security, ever.

(The Karzai family and other Afghans support new "arm the tribes" plans because they figure, rightly, that their militias and their supporters can skim off a lot of the proceeds. But it's hard to see what ISAF gets out of it strategically, other than possibly to drive up the price that the insurgents would have to pay to neutralize or coopt local would-be warlords.)

Posted by BruceR at 02:13 PM

Howz-e hasn't changed much, I see

Just another day on Ring Road South, courtesy of CBS News.

Howz-e Madad was always a hardship post. What you had during our tour was significant unappreciated population movement, as people vacated the green belt south of the highway, between the road and the river, to avoid the fighting, establishing themselves in temporary dwellings in the desert north of the road. It had the tendency of making our aerial photographs and maps of this area a little outdated. Attacks on Western and ANA facilities in this area were regular, and the displaced population and those still living in the green belt both entirely unconvinced of the merits of a Western/ANA presence. In the video, you can see the Afghans are firing south from the strong point across the road, towards the green belt, which is likely where the attack came from, and where it withdrew into. One imagines it was more blowing off steam on the ANA's part than anything else. Getting hit and doing IED sweeps daily without any way to strike back can do that to you.

Indiscriminate firing by Compass Security personnel is also a standard feature of life in Kandahar Province. I suspect those RPG rounds the reporter finds are theirs, as well. The military and police have no monopoly on heavy weapons.

(Impressive (or lucky) attack by the insurgents, btw. Assuming it was an RPG, they either managed to make a difficult shot from distance (you'd normally want to be within 50 m to be sure of hitting an SUV) or they exposed themselves in daylight in broad open, flat, terrain only 400m from a "rapidly expanding U.S. military compound" and then managed to leg it back to cover without incurring casualties, apparently. Almost makes me wonder if a recoilless rifle, man-portable and more accurate at range, was used, instead. But the area around Howz-e, where several canals parallel and south of the road allow close approaches to the road with easy exfiltration back into the green belt, has been a historic location for direct-fire ambushes on highway traffic since the Soviet days.)

Posted by BruceR at 01:58 PM

July 21, 2010

Unfortunate names

Chad Vader's stepfather is in the news, apparently.

Posted by BruceR at 12:45 AM

July 19, 2010

Today's essential Afghan reading: Panjwaii Tim and retelling Semrau

Two good pieces on Afghanistan in Canadian papers over the weekend.

First, the Star's Mitch Potter with Tim Lynch's buddies running a stealth development crew on the streets of Kandahar. Good for Potter for going the extra mile.

The Canadian military today requires reporters to fill out 47 pages of forms before embedding. Panjwaii Tim required only a handshake — and a solemn promise on the ground rules.

“We're proud of the work we do. But you understand the stakes: this is life or death for us. No last names, no naming our NGO. No precise description of where we live. The danger is real. Do not make me regret this.”

Back home, Andrew Duffy gives another of the better accounts I've read so far of the circumstances surrounding the Semrau court martial.

Under a strict interpretation of the rules of engagement that govern the conduct of Canadian soldiers in Afghanistan, Semrau should have offered the insurgent first aid and called in a helicopter or field ambulance to evacuate him. It was also open to Semrau to leave the wounded man at a battlefield collection point to be picked up later.

From his perspective, Semrau likely believed his options were more limited. The area could not be secured for a helicopter since hundreds of Taliban were rumoured to be in the area and the operation was in its infancy. What's more, the ANA, upon whom the Canadians relied for their battlefield security, had left the insurgent to die and had pushed south. It would be dangerous for two Canadians to remain isolated on the battlefield.

Yep, that pretty much sums it up.

Posted by BruceR at 09:03 AM

July 15, 2010

Doonesbury the mentor

Doonesbury's been looking at the Iraqi army this week; rang a few bells. Worth your time.

Posted by BruceR at 01:57 PM

Today's essential Afghan reading: Forsberg and Kagan

Carl Forsberg and the Kagans sort out the tangle of armed Afghans working in and around Kandahar, and how the Karzai clan continues to tighten their grip independent of official government forces in the area.

The formation of a powerful conglomerate of PSCs under the political control of local powerbrokers like Ahmed Wali Karzai would undermine the long-term stability of southern Afghanistan and the strength of Afghanistan’s legitimate security institutions. There is a very real risk that these institutions will be relied on by the Karzais and their allies as the guarantors of Kandahar’s security. If the Kandahar Security Company were in fact to grow to 2,500 armed men as Ruhullah suggests (and this is certainly feasible) it would be more than twice the current size of the Afghan Uniformed Police in Kandahar, and would exceed the size of the expanded police force that ISAF and the MOI are planning to add to the city.

A must-read to understand the local power dynamics: for instance, note the complete absence of the ex-Canadian provincial governor anywhere in the link-analysis schematic on page 2. (That itself should be a "well, there's your problem" kind of moment.)

The report recommendations don't seem particularly sound, however:

ISAF’s aim should be to disband these armed units and replace them with ANSF. To achieve ANSF primacy, ISAF cannot simply incorporate these PSCs or their members into the formal security forces. It will be necessary to vet their members, retrain them, and disperse them throughout the country via the established national recruiting and assignment procedures. The army would better incorporate these militias than the police, because of its ability to assign forces nationally.

Yeah, that's not gonna happen.

Indeed, things are going the other way entirely, so much so that one has to assume that Karzai's support for the new ISAF plan to expand the AP3 program out of Wardak hinged on his thinking he'll be able to go to that well to arm and deputize those local Kandahar militias that support his family, like the KAU.

You really can't grok Afghan security politics until you grasp that the president really trusts his national army much less than we do.

Posted by BruceR at 09:08 AM

July 13, 2010

Today's essential Afghan counterpoint: Inkspots vs Tim Lynch

Two respected Afghan bloggers, talking past each other. MK at the Inkspots, arguing for focussing on improving local justice systems instead of services:

Despite knowing this, and nearly a decade into the effort, we still struggle to set up even the simplest credible dispute resolution mechanisms. I don't mean an elaborate and fully developed national justice system: I mean local adjudicative bodies that have local legitimacy that need to be backed by our (or where, possible, GIRoA) firepower to enforce their decisions and protect them from being assassinated.

This isn't to suggest that military control of territory and population, building effective local security forces, or tackling corruption aren't just as important (or more, depending phase of operations in a given area). But it seems that as we've come to realize that development assistance is of limited utility in winning Afghans over to our side, we're a bit stymied as to what 'effective governance' means in concrete terms. Seems like solving local land disputes would be an excellent place to start.

In the other corner, Tim Lynch on staying away from dispute resolution and focussing on services instead:

The local people have every right to upset about the performance of the government in Kabul. But they have no interest in seeing any kind of central government which is strong enough to meddle in their affairs. An example, Afghans will go to great lengths to avoid having their problems brought into the legal system. Regardless of the crime be it murder or little boys stealing apples from a neighbor the Afghans know how to handle it and feel personally disgraced when the authorities step in to apply the rule of law. Their family business them becomes public and their problems known to people outside their clan which brings disgrace upon the sons of the family. They are going to bitch about the central government no matter who is in charge and how effective it becomes. The best we can do is concentrate on making regional government functional at basic things like irrigation, sanitation, health care delivery and other municipal services.

Posted by BruceR at 08:39 AM

Today's essential Afghan reading: Condra, et al.

An already much-commented on paper on the effect of civilian casualties on the Afghan insurgency:

The evidence shows that the conflicts in Afghanistan and Iraq are quite different in how civilian casualties affect the ability of insurgents to produce violence. In Afghanistan, we find strong evidence of a revenge effect. In Iraq, we find no such effect. This highlights two important differences in these insurgencies. The insurgency in Afghanistan is rural and centralized while that in Iraq was urban and featured a decentralized command. We suspect that the greater population density in Iraq made insurgent activity easier to observe and, combined with higher counterinsurgent force levels, increased insurgents’ reliance on the general populations’ reluctance to cooperate with counterinsurgents. In Afghanistan, the more dispersed population and lower counterinsurgent force levels mean the supply of insurgents is much more likely to be the binding constraint. The centralized structure of the Afghan insurgency also bears directly on their ability to engage in more sophisticated information operations. If an insurgent organization is to capitalize on the activities of the counterinsurgent, it requires a coordinated response targeted to key areas without any other sources of information. This is only possible in a consolidated, non-competitive insurgency.

Posted by BruceR at 08:27 AM

July 09, 2010

Not only no, Colonel, but f___ no

Gen. Mattis of the USMC, who replaces Gen. Petraeus as commander Centcom, has already been immortalized, of course. Playing as Mattis, Robert John Burke in the miniseries Generation Kill gave one of the most memorable cameos of that series, in depicting the chewing out of Col. Dowdy at Nasariyah in 2003, a historically documented incident that would later contribute to Dowdy's controversial relief of command. No idea how close it mirrors reality, but the fictional Dowdy's place is a place every soldier has nightmares about being found in someday: right in the gunsights of a superior commander who sees excuses instead of caution. At 4:27 of this video.

Of course the amusing bit is that Generation Kill itself was based on a series of articles in none other than... Rolling Stone, the same magazine whose current issue has saved Mattis from retirement, as well.

Posted by BruceR at 05:02 PM

Today's essential Afghan reading: Roland Paris

On the Canadian government's assessments of success:

Without at least receiving information on both security incidents and Afghan public perceptions of their security, Canadians will not be in a position to assess the progress of allied and Canadian efforts in Kandahar — or, for that matter, to grasp the urgency of the current situation in Kandahar.

Yet Ottawa routinely withholds information on broad trends in insurgent activity as well as basic data on public opinion of Afghans living in Kandahar.

This is similar to what I wrote here, about how the government's quarterly reports on security are focussed on second-order effects.

Posted by BruceR at 08:06 AM

July 08, 2010

Today's essential Afghan reading: Ethan Kapstein

Canadian and other soldiers who see precious little progress in Kandahar should keep in mind (as I always tried to do) that in Afghanistan, the "flypaper theory" really does work. We fought the bad guys in the south and east so that the north and west (Kabul, Mazar, Herat) could prosper, could make gains in human freedom, could watch their kids grow up... because ultimately what will finally delegitimate the Taliban in Afghan eyes will be when people in non-Taliban areas are visibly more prosperous. Ethan Kapstein documents the peaceful half of the country:

Nor is all this growth dependent on foreign aid. In fact, the regions of the country that are enjoying the most economic activity-like Herat and Balkh (where Mazar is located)-are probably those where the least aid has gone on a per capita basis. In both these provinces, for example, strong governors have made security a priority, giving entrepreneurs the breathing space to exploit existing business opportunities.

Kapstein's recommendations, all of which make a lot of sense to me:

First, the military should focus on its primary mission of providing security and bolstering the Afghan National Army and Police, giving entrepreneurs the "breathing space" to develop their economy.

Second, the United States and its coalition partners should support those regional authorities who are creating a more secure environment by providing them with some transparent and accountable budget support. Too much aid money is going to projects like schools and meeting halls that are not necessarily the local priority.

Third, the United States and European Union should pass free trade agreements with Afghanistan. It is absurd that local entrepreneurs face high tariff barriers when they ship their goods to Afghanistan's coalition partners. As in much of the developing world, it seems that Washington and its European allies prefer providing aid to supporting trade.

Posted by BruceR at 08:17 AM

Today's essential Afghan counterpoint

Robert Blackwell:

Announcing that we will retain an active combat role in Afghanistan for years to come and that we do not accept permanent Taliban control of the south, the United States and its allies could withdraw combat forces from most of Pashtun Afghanistan (about half the country), including Kandahar, over several months.

We would stop fighting and dying in the mountains, valleys and urban areas of southern Afghanistan — where 102 coalition soldiers were killed in June, the most in any month of the war and almost three times as many as a year ago. But we could be ready to assist tribal leaders on the Pashtun periphery, who may decide to resist the Taliban.

We would then focus on defending the northern and western regions — containing roughly 60 percent of the population. These areas, including Kabul, are not Pashtun dominated, and locals are largely sympathetic to U.S. efforts.

Ann Jones:

You have only to look around in Kabul and elsewhere, as I did this month, to see that the more American military there is, the more insurgents there are; the more insurgent attacks, the more private security contractors; the more barriers and razor wire, the more restrictions on freedom of movement in the capital for Afghans and internationals alike; and the more security, the higher the danger pay for members of the international community who choose to stay and spend their time complaining about the way security prevents them from doing their useful work.

It seems to me there's a potential causality error here. There's less war and more public support for ISAF in the north and west. There is also significantly less ISAF presence in the north and west. Blackwell's argument only works if you assume that the light footprint is a result of the goodwill, and not the other way around. The risk in making Fortress Northern Afghanistan is that the insurgency just follows you there, because the oppressive foreign presence itself produces significant alienation, and delegitimization of local power structures.

The stronger argument would be that the ISAF footprint should be lightened throughout the country to close to northern levels, to avoid the negative impacts Jones is describing, even if that means increased insurgent freedom of movement in other parts of the country.

The necessary second part of any kind of force-lightening strategy, though, is the abandonment of the kinds of societally transformative goals we've heard in the past... accepting Afghans as they are, and merely promoting the interests of those uninterested in resuming the "terrorist haven" aspects of Taliban foreign policy. That also would mean limiting our military presence to enabler support to whatever the Afghan military wants to do... which would likely not be overly congruent what we would want them to do. The historical analogies to the Arab Revolt, or British policy towards American Indians in the early 19th century, or American FID work in Latin America, should be obvious. It's a mindset shift, to be sure. But just moving our conventional forces wholesale to parts of the country where they don't hate us YET doesn't seem to promise to be an improvement on anything.

Posted by BruceR at 07:46 AM

July 06, 2010

Today's essential Afghan reading: Tim Lynch

More of this please:

There is much more American military activity around Jalalabad including flying columns of the varsity Afghan SF with their American advisers who use Toyota trucks just like their Afghan colleges. These small, fast, powerful formations are by far the most effective joint US/Afghan effort of the war and the only example of real embedded (as opposed to co-located) training currently being done with the Afghans.

Look, it's real simple: you don't need to "arm the tribes." You don't need to train the Afghans to fight our way, either. You just need to train those willing to fight for the central government enough to be able to explain what they need from us, and then train a (much smaller) number of our soldiers to fight alongside them, in a manner that doesn't get in their way, bringing all of our nifty technological enablers along to ensure a victory. And that's a lot simpler challenge... if only because more of ours can read the manual, be it 3-24 or Seven Pillars. It worked in 2001, and it can work again. We really seem to be overthinking this thing.

Posted by BruceR at 08:41 AM

Today's essential Afghan reading: Bernard Finel

A must-read for would-be nation builders:

The other issue I wanted to mention is that successful models of state-building include self-sustaining dynamics. Our current plans for Afghanistan do not. Instead, every expansion of central authority increases the demands on the Afghan state. A state-building through largess approach becomes progressively more difficult as you reach out to areas where development and state control are weaker. This is precisely the opposite of successful processes which, once begun, often provide the resources needed for further consolidation.

This is exactly right. In growing the security forces so large, and attempting to extend the central government's reach, we have created a state so entirely dependent on cash transfers from foreign countries to continue that it is clearly unsustainable, even to its strongest supporters among its own people. The Afghan GDP will never, ever rise to the point where it can pay the army we've created for them at the wage levels we've set. Because it is clear the good times cannot last, it is only rational for the key players to be focussed on shorter-term profit-taking.

A year ago, I was writing that the foreign-funded increase in teacher's pay was a good thing... because Afghan teachers were horribly underpaid compared to even junior soldiers. So we ante'd up and increased their pay, too, just as we'd increased that of the police before. When army recruitment starts to flag, we'll undoubtedly raise that salary again. But at some point, this must collapse, and Afghans have to know it. It's led to gold-rush economics, and frontier justice.

Posted by BruceR at 08:08 AM

July 05, 2010

Today's essential Afghan reading: Bing West

Reviewing Kilcullen:

Because they are partnered with our troops, Afghan soldiers are copying our rules of engagement and risk-avoidance procedures. Since they wear our heavy armor, they too cannot pursue the light and mobile Taliban forces. When the enemy initiates contact, the Afghan soldiers are trained to wait alongside our troops until our attack helicopters force the Taliban to flee. The Afghan soldiers will not be able to fight that way as U.S. resources are reduced. The Afghan security forces simply cannot take over the fight anytime soon.

Reminds me of what I wrote (point #7) fresh from the fight last August.

UPDATE: I also addressed West's concern a little in this post last September.

Posted by BruceR at 08:56 AM