Two failed strategies from the past, combined into one new strategy?
Is this negative × negative = positive? Or negative + negative = negative? And have past failures (i.e., in the title of this blog entry) been ignored? First a brief survey of what may be coming.
If you already know this stuff pretty well and would like to skip the re-read, scroll down to the picture of the underpants gnome diagram below (blatantly stolen from Registan).
I’ll start with the NYT, from October 28th:
President Obama’s advisers are focusing on a strategy for Afghanistan aimed at protecting about 10 top population centers, administration officials said Tuesday, describing an approach that would stop short of an all-out assault on the Taliban while still seeking to nurture long-term stability. [...]
At the moment, the administration is looking at protecting Kabul, Kandahar, Mazar-i-Sharif, Kunduz, Herat, Jalalabad and a few other village clusters, officials said. The first of any new troops sent to Afghanistan would be assigned to Kandahar [...]
But military planners are also pressing for enough troops to safeguard major agricultural areas, like the hotly contested Helmand River valley, as well as regional highways essential to the economy… [...]
“We are not talking about surrendering the rest of the country to the Taliban,” a senior administration official said. Military officers said that they would maintain pressure on insurgents in remote regions by using surveillance drones and reports from people in the field to find pockets of Taliban fighters and to guide attacks, in particular by Special Operations forces.
Next up is an opinion piece from the Weekly Standard:
In its continuing search for an alternative to General Stanley McChrystal’s comprehensive counterinsurgency approach to the war in Afghanistan, and with President Obama having eliminated the minimalist counterterrorism plan of Vice President Joe Biden, the White House has lately been floating a split-the-difference trial balloon: “McChrystal Lite” or, to give the veep his due, “McChrystal for the cities, Biden for the countryside.”
The authors (Tom Donnelly and Tim Sullivan from AEI) then get into technical details and posit probabilities. And they seem to want a full surge, as one would expect from people who work at the American Enterprise Institute.
Back to the newspapers, from WaPo on October 31st:
The military chiefs have been largely supportive of a resource request by Gen. Stanley A. McChrystal, the top U.S. and NATO commander in Afghanistan, that would by one Pentagon estimate require the deployment of 44,000 additional troops. But opinion among members of Obama’s national security team is divided, and he now appears to be seeking a compromise solution that would satisfy both his military and civilian advisers. [..]
Before he can determine troop levels, his advisers have said, he must decide whether to embrace a strategy focused heavily on counterinsurgency, which would require additional forces to protect population centers, or one that makes counterterrorism the main focus of U.S. efforts in the country, which would rely on relatively fewer American troops. One option under review involves a blend of the two approaches, featuring an emphasis on counterterrorism in the north and some parts of western Afghanistan as well as an expanded counterinsurgency effort in the south and east, one of the officials said.
Staying with WaPo, two days later David Ignatius wrote:
As Obama has deliberated Afghanistan strategy, the debate has tended to polarize between “CI” and “CT” advocates. But this is a false argument. What the United States actually has in Afghanistan is a mixture. Obama must now decide whether to provide the resources — and take the risks — to test whether this combined strategy can succeed.
Speaking with Farid Zakaria and Newsweek, Tom Ricks commented:
In fact, the crucial judgments that have to be made involve what the troops will do and how much of Afghanistan to cover. Ricks said to me, “Why not do the Petraeus plan [counterinsurgency] for the major population centers and the Biden plan [counterterrorism] for the rest of the country?” That sounds like a middle course that is smart and practical, which might need some more forces or perhaps can make do with the almost 100,000 already there. Obama should carefully consider these and other options before racing out to demonstrate how tough he is.
And once again, back to WaPo with some behind the scenes info:
The Pentagon’s top military officer oversaw a secret war game this month to evaluate the two primary military options that have been put forward by the Pentagon and are being weighed by the Obama administration as part of a broad-based review of the faltering Afghanistan war, senior military officials said.
The exercise, led by Adm. Mike Mullen, chairman of the Joint Chiefs of Staff, examined the likely outcome of inserting 44,000 more troops into the country to conduct a full-scale counterinsurgency effort aimed at building a stable Afghan government that can control most of the country. It also examined adding 10,000 to 15,000 more soldiers and Marines as part of an approach that the military has dubbed “counterterrorism plus.”
And most importantly, WaPo gives a possible time line for a decision:
Administration officials say Obama might settle on a plan but delay announcing it until after a runoff in the Afghan national elections, scheduled for Nov. 7. The president is to begin a 10-day trip to Asia on Nov. 11.
And recently, from the CS Monitor (h/t Ink Spots):
Put simply, there are two poles in Washington: the counterinsurgency experts, or COIN-istas [sic], who believe Afghanistan’s deteriorating security can only be reversed by adding tens of thousands of troops – perhaps as many as 80,000; and those who believe US interests in Afghanistan are few, and the best way to keep it on a low simmer is to employ a counterterrorism-like model – using drones, bombs, and special forces teams to keep Al Qaeda at bay. The debate has become protracted, with military commanders like Gen. Stanley McChrystal politely urging the commander in chief to make a decision soon.
So yeah. What about that intermediary phase/step? What happens there?
What sayeth the blogosphere? Gulliver at the group blog Ink Spots ( a must read from the moment it started, and honi soit qui mal y pense to anyone that thinks otherwise) – a blog that has lately been taking a hiatus from copious use of French in America – comments in regards to the NYT article above:
Might this be the best possible middle road? It sure seems to beat the hell out of “muddling through,” the other non-drawdown, non-escalation path currently being advocated by some people, in which we’d, uh, just keep doing what we’re doing (because that seems effective). The best line in the whole story cites one official calling the plan “McChrystal for the city, Biden for the country.”
Of course, if this approach is adopted, there will be critics who object to the “abandonment” of parts of the country, and others who question the effectiveness of a sequential, oil spots approach. Bernard Finel has raised some reasonable questions on this point here, and I hope to discuss them a bit more in the near future.
So what does Bernard Finel say? A lot. The friendly back and forth that start at taches d’huile stirs no mauvais quart d’heure. Finel starts with:
But while you’re consolidating government control inside the ink spots, they are consolidating insurgent control in their “ink spots.” Why assume our ink spots will spread, but theirs will shrink?
And once you have moved on, why can’t the insurgents resume operations in their ink spots? And how do you deal with the massive legitimacy loss from the inital abandonment?
…and ends with:
If that is the case, isn’t it obvious that the choice isn’t between well-crafted COIN and imperfect (due to intel and logistical limitations) CT, but between a genuinely half-assed COIN campaign and an imperfect CT approach? Which is why I worry that we are stacking ill-considered assumptions about “rolling COIN” and ink spots on top of the already heroic assumptions embedded in 3-24.
I am really try to understand all of this… but I just don’t get it. What am I missing?
Finel says a lot that leads up to that, mostly to do with the problems of implementation, sustained effort and resourcing. The article is essential reading and I encourage you to give it a once-over. Gulliver, of course, responds. The exchange is the best public debate I can find, and is thankfully free of the overly-politicized language found elsewhere.
That’s the present and the future. But what about the past? The strategies discussed sound familiar. The focus on counter-terrorism in rural areas sounds like Donald Rumsfeld’s “light footprint strategy” and the focus on cities sounds like the later Soviet strategy. First, a graphic illustration of the later Soviet “strategy.”
The shaded areas (the cities and the roads) are the level of control as the Soviets were heading out the door. Click here for a larger, better detailed, look at the map. The map is from page 186 of Gilles Dorronsoro’s Revolution Unending (it’s in my top 5 books). I use strategy in quotation marks above as this is strategy as much as any desperate last resort is a strategy. I’ll get back to Dorronsoro. But first, from a 1987 article in the LA Times:
Kabul remains a beautiful and wild city. [...] But it is also a deeply troubled place. Since the war began 7 1/2 years ago, its population has jumped from 500,000 to more than 2 million. Since it is one of the few relatively safe places left in the country, people have flocked here for protection. So much of the countryside has been devastated by the war that Afghanistan has folded its population into this city-state– Kabulistan , the natives call it. With more than 4 million Afghans living as refugees in Pakistan and Iran, and up to 1 million killed in the war, Kabul’s current population may amount to one-third of Afghanistan’s remaining people.
Soviet armor and artillery, the so-called “rings of steel” defense, are positioned in concentric circles around the city and its immediate suburbs. Trips to other cities by road take days and are very dangerous. So Kabul has become a fortress town, connected to the rest of the world by infrequent commercial airline flights and to the Soviet Union by the military air umbilical cord. Diplomats have counted more than 50 Soviet military cargo flights in one day. Kabul has become like blockaded Berlin during the 1948 airlift.
Clear BS on the population guess. But what about those “concentric circles”? A US government publication writes:
Western diplomatic sources reported in early 1985 that guerrilla rocket attacks into Soviet military camps in and near Kabul reportedly had killed 60 and wounded 34. The Soviet response to these attacks included setting up a security network of eight concentric circles around Kabul: at each post an estimated 100 soldiers, plus eight to 10 tanks or APCs, were posted.
And in War in Afghanistan, Mark Urban notes that once the rocketing started increasing in range (up to 26km by 1988), defending Kabul with these rings became unfeasible with the number of troops available. However, he does note that Kabul experienced rocketing since 1984. [pp.64-5, 248-50]
Now back to Dorronsoro. On page 189 he notes that
Between 1980 and 1983 the Soviets adopted a defensive position, avoiding direct involvement and relying largely on the Afghan forces. [in 1982]… the Soviets attempted to improve the defenses of major towns.
That first sentence sounds like Rumsfeld’s Light Footprint strategy (but minus his reliance on shady local militia types in favor of the shady old ANA), doesn’t it?
Dorronsoro continues and describes the second phase (1984 and 1985) that resulted from the failure of the first phase (1980 to 1983). In the second phase the Soviets undertook large offensives throughout the country while again attempting to improve the defenses of urban areas. Basically, a dual surge that was stymied by the inability to hold territory due to lack of troop numbers. By 1986 the Soviets “were getting ready to pull out and were launching fewer offensives.” (p. 190).
It was this later phase that resulted in the map you see above.
Mark Urban describes a similar time frame. He writes (pp. 287-8):
From early 1986 the Soviet Army switched to a more defensive strategy. Rural operations were reduced and defences around towns increased while Mikhail Gorbachev prepared the army for withdrawal. During 1984-5 there were probably six offensives involving more that 5000 Soviet troops each. In 1986-9, if one excludes the actual withdrawals of Soviet garrisons, there was just one – Operation Highway.
Going off of formerly secret archives, during the Soviet Afghan War there was a strategy that also sounds very familiar. From page 123 of Mitrokhin:
The Soviet nomenklatura was forced to change its tactics in the war and to abandon the idea of conquering the whole country at once. A plan was devised to keep firm control over the regions which could be effectively controlled and to introduce national, social and economic changes in them. They reckoned that this would allow them gradually to gain control of the whole country. The aim was to achieve a decisive victory in the northern zones bordering the Soviet Union first. Territorial committees for the defense of the revolution were set up in villages, streets and groups of houses. Surveillance of the population and its mood was stepped up and any movements and new people were noted.
Some translated Soviet documents very recently released by the National Security Archive have resulted in some media attention by various authors who noted the parallels (i.e. in The Guardian and the NYT). Nothing very original in those two articles. As far as lessons learned from the Soviet experience… well, Artemy Kalinovsky says it best while writing in Foreign Policy:
It’s a failure the United States apparently has no intention of repeating — to the extent that it doesn’t even seem to study it. The U.S. Army/Marine Corps Counterinsurgency Field Manual does not mention the Soviet experience once. One analyst told me that when she suggested including the conflict as a way to inform current policy, Pentagon officials seemed to have little awareness about what Moscow had been trying to do there or for how long.
So chto delat‘? What is to be done? Dorronsoro, who is very aware of history, suggests in his WaPo article:
To succeed, the coalition must control Afghanistan’s cities, where institution-building can take place and where the population is more neutral or even favorable to the coalition. The Afghan army and, in certain cases, small militias must protect cities, towns and the roads linking them. Fewer casualties and the improvement of the Afghan security forces — Afghanization — will allow the coalition to focus more resources in the north, where the situation is becoming extremely unstable. Stabilizing the country will allow the coalition to focus on al-Qaeda, the enemy that attacked the United States on Sept. 11.
He provides a similar, but more in-depth argument, in this paper.
But how about I posit the same failure as the Soviets? A weak partner in the form of the Afghan government, a dysfunctional local army and police, a loss of goodwill, untrustworthy potential militia clients, and insurgents that are increasingly able to strike the cities and roads. Will those Soviet concentric circle around Kabul be recreated? Can they?
Getting back to the bigger picture, who is influencing the current debate (in a diffuse intellectual manner)? As far as pushing for the strategies in the title of this blog entry, it sounds like Rory Stewart when I hear about how the cities and the north need to be prioritized. He’s been saying that for a while. But he also complains that no one in government listens to him. He may be wrong about being ignored. His ideas are being considered, even if he is not the source. And since Gilles Dorronsoro came to Washington the views that he has expressed above have become much more prevalent. But more likely, one should look to the September Senate testimony of John Nagl to get a small glimpse of what is being told to the big people upstairs.
It seems some of the current proposals being floated are essentially a return to the Rumsfeld days of counter-terrorism mated with the late Soviet strategy that allowed the government in Kabul to survive in a form that never did extend its authority from its more secure areas. So the Soviet strategy actually stabilized/stalemated the situation for a few years. I accept that. Of course, when it attempted to extend central authority with its unsuccessful try to take over of Rashid Dostum’s militia the civil war got fully started. Is this a move towards the least worst option? Or just a bad option?
Are “we” better at counter-terrorism in Afghanistan? Probably, since it was so terrible back in the Rumsfeld years. But it is still bad. And can “we” do a counter-insurgency strategy in Afghanistan that cedes an enemy’s favorite territory? Does that strategy essentially open up a “highway” straight from Pakistan to the “Urban shooting/bombing galleries”? Will we see an endless stream of attacks against coalition forces and the Afghan government as they hunker down in their urban fortresses and attempt to run the gauntlet of the main roads? So yes, more debate is needed. I suppose that is happening behind closed doors as I write. We should find out in a couple of weeks what the president’s decision is.
It took several years to determine that Rumsfeld’s Light Footprint strategy was a failure. It is in a similar time frame that the Soviet’s “Cities, Roads and the North” focus was a failure (after it was inherited). There won’t likely be such a long time frame in the present case.
Is this obsessive analogizing with the past? Maybe. Afghanistan is not condemned to follow the fate of its 1979-2001 trajectory. Foreign forces are not the Anglo-Afghan expeditionary force. As the title of Kalinovsky’s article reads, “Afghanistan is the new Afghanistan.”
So what do I think? I’m on board with Bernard Finel’s comments. I think his points are good enough to warrant a repeat:
But while you’re consolidating government control inside the ink spots, they are consolidating insurgent control in their “ink spots.” Why assume our ink spots will spread, but theirs will shrink?
And once you have moved on, why can’t the insurgents resume operations in their ink spots? And how do you deal with the massive legitimacy loss from the inital abandonment?
[...]
If that is the case, isn’t it obvious that the choice isn’t between well-crafted COIN and imperfect (due to intel and logistical limitations) CT, but between a genuinely half-assed COIN campaign and an imperfect CT approach? Which is why I worry that we are stacking ill-considered assumptions about “rolling COIN” and ink spots on top of the already heroic assumptions embedded in 3-24.
I am really try to understand all of this… but I just don’t get it. What am I missing?
I concur. More debate is needed. More information is needed.
