Here’s an interesting overview from Steve Coll about the later years of the Soviet mission in Afghanistan:
I mentioned in the previous post in this series that the Soviet Union’s effort to “Afghan-ize” their misadventure, and to build up Afghan security forces as an exit strategy, succeeded partially for a few years, although it ultimately ended in the collapse of the Afghan Army, just after the Soviet Union itself dissolved, an event that orphaned Kabul materially and politically. Before then, however, the late Soviet and immediate post-Soviet period in Afghanistan—from 1986 until 1992—did produce an Afghanistan that resembled in some respects what U.S. policy may well create over the next few years, if it is fortunate.
During this period, the Kabul government, run by President Najibullah, a former secret police chief who became a politically adept strongman, controlled an archipelago of Afghan cities. These included Kabul, Kandahar, Jalalabad, Khost, Herat, Mazar-i-Sharif, Kunduz, and a number of smaller provincial capitals, each of which was ringed by layered defenses, with Afghan forces increasingly in the lead. The insurgents they faced—the U.S.-backed Islamist rebels then known as the mujaheddin, some of whom, such as Gulbuddin Hekmatyar and the Haqqanni clan, are still in the field, now under the Taliban banner and fighting the United States—controlled virtually all of the mountains and countryside. In fortress Kabul, Soviet civilian advisers shuttled among ministries and apartment compounds deep in the center of the ringed defenses, secure from kidnapping. Their client Afghan forces also tried to secure the main roads between the cities, but this proved difficult. At best the uniformed Afghan Army could move down some roads in daylight some of the time without being attacked or ambushed. Even securing roads between exurban airports and city centers proved to be a struggle. The only way government officials could move reliably between their island cities was by air; the mujaheddin began to acquire Stinger anti-aircraft missiles late in 1986, however, so even that method was not foolproof.
I suppose one question to ask ourselves would be: How long could this status quo have been maintain had not the Soviet Union collapsed? And also how different would this have looked if the mujahedeen didn’t have assistance from the United States? Or to put it another way, if we tried to do what the Soviets semi-successfully did, isn’t there good reason to think we would succeed? The United States isn’t going to suffer a domestic collapse. And the United States in 2009 is much wealthier than the Soviet Union in 1989, we can sustain financial aid to the Afghan government much more easily than the Soviets could. Our adversaries are also receiving much less in the way of external support than the Soviets’ adversaries were.
What’s more, there’s also all kinds of help we can give the Afghan government that the Soviets weren’t able to offer and that falls far, far, far short of having 100,000 western soldiers conduct a comprehensive counterinsurgency campaign. It would, for example, be cheap, easy, and uncontroversial in terms of US domestic politics for us to provide our Afghan allies with detailed satellite surveillance that puts to shame anything the Soviet military was capable of.
Or maybe another way of putting is that it might be useful to ask American military commanders something like the following: Suppose I arbitrarily capped the resources available for Afghanistan at a total of 5,000 people in the country and $20 billion a year in total spending, what strategy would you adopt given those resource constraints? Then ask how well that strategy might work. Whenever I think about Afghanistan, I just can’t get past the fact that it seems obvious that there’s something wrong with a strategy that has us spending five times Afghanistan’s GDP and deploying tens of thousands of soldiers to beat what amounts to a highly motivated militia operating on a shoestring budget.