How America Failed in Afghanistan


“To blame Afghans for not getting their act together in light of [the] history is just wrong,” Steve Coll says.

Steve Coll on the humanitarian catastrophe that is now likely to engulf Afghan civilians, and how Joe Biden is shifting the blame.

On Sunday, as the Taliban entered Kabul—the last remaining major Afghan city not under the group’s control—the President of the country, Ashraf Ghani, fled to Tajikistan, making clear that the U.S.-backed Afghan government had collapsed. Five months ago, in April, President Joe Biden announced that all U.S. and nato troops would be withdrawn from Afghanistan by the twentieth anniversary of the 9/11 attacks. Critics have accused the Administration of conducting a rushed, poorly planned, and chaotic withdrawal since then. On Thursday, the U.S. government announced that it would be sending in marines and soldiers to help evacuate embassy personnel. But the speed of the Taliban advance has stunned American officials and left desperate Afghans trying to flee the country. Responding to criticism about his plan, Biden has sought to shift blame to the Afghan government and its people, saying, “They have got to fight for themselves.”

I spoke by phone with my colleague, the New Yorker staff writer Steve Coll, about the situation in Afghanistan. The dean of Columbia Journalism School, Coll is the author of “Ghost Wars” and “Directorate S,” which together chronicle much of the history of the past several decades in Afghanistan and Pakistan. During our conversation, which has been edited for length and clarity, we discussed why it has been so hard for the United States to train the Afghan army, the different humanitarian crises facing the country, and the Biden Administration’s “outrageous” callousness toward a situation America played a role in creating.

What about the events of the past few weeks has surprised you, and what was the predictable result of Biden’s policy announcement in April?

I think the speed of the political collapse in Afghanistan surprised a lot of people. The pathway of the collapse was predicted and predictable. This has happened in Afghan political and military history a couple of times before. But there was a speed and momentum of people recalculating where their interests lay, and switching sides, and capitulating without violence that I don’t think the Biden Administration had expected when it announced its timetable in the spring.

You could argue that this shows the Biden Administration’s policy was a mistake, but you could also argue that, if this was going to happen so quickly after two decades of American troops in Afghanistan, there was no way to make this work without pledging to stay forever. How do you think about those two ways of looking at the situation, or do you think that dichotomy isn’t helpful?

I think that dichotomy describes two poles that represent the range of choices that the Biden Administration faced, and in between those poles had been, more or less, the policy going back to the second term of the Barack Obama Administration—which was a smaller, sustained deployment. There were twenty-five hundred troops there when the Biden Administration came to office. The rate of casualties incurred by nato forces was almost at the level of traffic accidents for much of the past couple of years. So a sustained, smaller deployment—not free, but nothing like the expenditures of the past—linked to a search for some more sustainable political outcome had been visible. The Trump Administration followed that path, too, picking it up from the Obama Administration, and the Trump White House had become quite ambitious about it. It had negotiated with the Taliban an agreement that had a timetable, including regarding American withdrawal. But, until the Trump Administration got to that point, it had been following the same pathway as its predecessor.

I think in between was this question of whether the benefits of a messy degree of stability justified having the small-to-medium deployment that America has in other parts of the world. That is what you are going to hear in Washington. The counter-argument to the Biden Administration’s policy is not going to be forever war and the defeat of the Taliban; it is going to be a critique of the haste with which it pulled the plug on what was not a large deployment, and one that was not incurring a lot of casualties.

Why, ultimately, was it so hard to stand up the Afghan military to a greater extent than America did? Was it some lack of political legitimacy? Some problem with the actual training?

I don’t know what proportion of the factors, including the ones you listed, to credit. But I think that the one additional reason it didn’t work was the sheer scale of the ambition. And this was visible in Iraq as well. Building a standing army of three hundred thousand in a country that has been shattered by more than forty consecutive years of war and whose economy is almost entirely dependent on external aid—that just doesn’t work. What did work was what at various stages people thought might be possible, which was to build a stronger, more coherent, better-trained force, which has effectively been the only real fighting force on behalf of the Kabul government over the past few years. This force is referred to as commandos or Special Forces, but it is basically twenty or thirty thousand people. That you can build with a lot of investment and hands-on training. But you can’t just create an army of three hundred thousand. I remember talking to the Pakistani generals about this circa 2012. And they all said, “You just can’t do that. It won’t work.” They turned out to be right.

The writer Anand Gopal, who has reported extensively from Afghanistan, wrote, “The US designed the Afghan state to meet Washington’s counterterrorism interests, not the interests of Afghans, and what we see today is the result.” Do you agree?

I assume what that means is that the state-building project, such as it was—and about which there were varying degrees of commitment, including very little at the very beginning, after the fall of the last Taliban government—was undermined by the dependence on independent militias and commanders whose role in security was seen as necessary, especially early on, because the main U.S.-led nato agenda in Afghanistan and the region was counterterrorism. The men under arms—the power brokers or warlords—were seen as essential to that agenda, and it was very difficult to build a normal state when the militias were beyond political accountability (never mind the rule of law) and dominating so many regions of the country.

Over time, there was a recognition that this was not sustainable, and there were efforts to try to fold them into a more normal-looking state and constitutional military, but that project was never accompanied by a push for accountability or an end to the effective independence and corruption associated with those regional militias. I assume you can say that is all the fault of the Western design, but I am not sure I buy that. Afghanistan had these fighting forces on its soil on 9/11 because of the continuous war that had been triggered by the Soviet invasion in 1979, and they didn’t require a U.S.-dictated constitutional design to persist. Of course, they persisted. The real complication about the design of the Afghan state that is now collapsing has at least as much to do with Afghans coming into the country from exile—the same dynamic that we saw in Iraq. Often, very talented and committed people who had been forced out of the country by the wars going back to the late nineteen-seventies tried to bargain with the leaders in Afghanistan about what kind of constitutional and power-sharing system should be designed. They were trying to create a system that would accommodate the power of the militias who had never left, in a very centralized constitutional design.

President Biden’s attitude toward Afghanistan of late has seemed to be one of annoyance, while he’s also putting a strong emphasis on the need for Afghans to stand up and fight for their country. How do you feel about an American President putting that forward after the U.S. has been intimately involved in that country for decades?

I try to tamp down my emotions about it, because I think it is an outrageous critique. I can understand the frustration that American decision-makers have had with their partners in the Kabul government for the past twenty years. It has been a very rocky road, and it isn’t all the fault of U.S. Presidents and Vice-Presidents and national-security advisers. But to suggest that the Afghan people haven’t done their bit is a kind of blame-shifting that I think is not only unjustifiable but outrageous. The Afghans now have suffered generation after generation of not just continuous warfare but humanitarian crises, one after the other, and Americans have to remember that this wasn’t a civil war that the Afghans started among themselves that the rest of the world got sucked into. This situation was triggered by an outside invasion, initially by the Soviet Union, during the Cold War, and since then the country has been a battleground for regional and global powers seeking their own security by trying to militarily intervene in Afghanistan, whether it be the United States after 2001, the C.I.A. in the nineteen-eighties, Pakistan through its support first for the mujahideen and later the Taliban, or Iran and its clients. To blame Afghans for not getting their act together in light of that history is just wrong.

Next door to Afghanistan, in what is now Pakistan, the British stayed a very long time and then left so abruptly, in 1947, that it resulted in incredible bloodletting. Do you see a callback to that imperial age, in the sense that the outside power wants to wash its hands of the situation and says that it is done without much care for how it leaves?

I agree with that. There is a lot to learn, and, in fairness to the Biden Administration, it inherited from the Trump Administration a terrible situation, because of the concessions the Trump Administration had made to the Taliban about the timing of a U.S. withdrawal. I said earlier that the U.S. deployment that Biden pulled was small, and the violence faced by them was de minimis, but that was true largely because of the terms of the flawed deal that the Trump Administration had negotiated with the Taliban, in which one of the commitments by the Taliban was not to attack U.S. troops in exchange for a hard deadline of May of this year for the last troops to leave. So, when the Biden Administration, in the pressure-filled first weeks of its term, reviewed the situation, it understandably feared that, if it tried to repudiate or rewrite the agreement that the Trump Administration had reached, it might transform a relatively quiet and stable experience for the U.S. military into another bloody round of combat that would undermine the Administration’s plans and foreign-policy priorities. So it pulled the plug and got exactly that result, which is now all that anyone is talking about.

But, to go back to your original observation, I do think that the haste and indifference, the blaming of the Afghans, the linkage of the decision to narrow U.S. interests and the anniversary of 9/11, all of that did have an air of—maybe “contempt” is not too strong a word for it—about what the consequences of this would be in Afghanistan. The decisions of the Obama Administration, and the Trump Administration in the first couple of years, reflected a rare political consensus in the United States that there was a willingness to sustain a relatively small troop deployment and expenditures in Afghanistan for a path out that would not lead to what we are watching now. And the President himself seems to have personally decided that that was a fool’s errand, and that he would not persist with what he perceived to be the illusions around that kind of search. But there was no crisis in expenditure or war from the American perspective that would have required such a quick decision.

O.K., but doesn’t the speed at which this is now happening suggest that the Taliban would never have negotiated something in good faith for a long-term solution.

It certainly suggests that the Taliban were not serious about ever sharing power in the way that the United States and its European allies and many sections of the Afghan government and society had hoped they might. But the problem was not a naïve faith that the Taliban had lessened their unwillingness to share power, because they hadn’t. They had been terribly stubborn throughout the negotiations. They declined even to talk to the Afghan government on very ideological grounds of historical legitimacy. So there was no reason to say that the Taliban had passed the test of credible international diplomacy, but they were talking. And the hope was that, over time, they could be drawn gradually to demonstrations of reduced violence in which the conversation about political futures would not be dominated by the violence and revolutionary ambitions of the Taliban’s history. Look at the negotiations with the farc in Colombia. How long did those last? Twenty years? Even now, you have a messy result, but these are not negotiations that typically result quickly in an agreement. You would expect it to be a very slow process.

What was flawed about the Trump Administration’s agreement was not a misunderstanding of the Taliban necessarily, but the timeline it set. The Administration basically used the negotiations as cover to leave before it had achieved any of the things that it said the negotiations were designed to do.

But is there any tension between saying, essentially, on the one hand that the deployment was not costing that many Americans lives and on the other that we got the Taliban to stop killing so many Americans only because we promised to leave? Could we have kept that balance if we weren’t promising to leave?

The American casualty rate was low even before the Taliban agreed to stop attacking Americans. This is not to say that each loss wasn’t painfully felt and meaningful, but coverage on the rate of casualty had disappeared from the news cycle starting around 2015 or 2016, when the deployment was down to the ten-thousand range. The problem with sustainability outside the negotiations was the way in which the war was being prosecuted after nato troops were drawn way down; essentially, it was an air war connected to the Special Forces on the ground. A harder question about this middle-way policy is how long you are going to bomb the Taliban into submission. The reason there was a military stalemate from roughly 2006 to this summer was that the American-led coalition, including the Afghan forces on the ground, had a monopoly on air power, and the Taliban had no answer for air power. But, of course, bombing in a country like Afghanistan is never precise, and there are civilian casualties and a sense of siege in some of the areas where air power was brought to bear. How long was that going to be the answer?

Your books on the region suggest that the Taliban may not have initially come to power, nor survived this long, had it not been for the aid and comfort of the Pakistani security apparatus—its military and intelligence services. How is Pakistan feeling about what’s happening now? I sense maybe there’s a tiny bit more anxiety than usual about what this might mean for Pakistan.

It seems likely that it is partially a case of watching what you wish for. I am sure they did not forecast the speed with which events are unfolding this summer, and they may also have expected that the role of negotiations and the timetable by which political change would occur in Afghanistan would allow them to build a platform for greater international legitimacy and credibility for a potential Taliban government. One of the reasons that I would be anxious if I were them is that this is happening in a way that is already inducing governments such as Germany’s—not usually first out of the box on these things—to say that they won’t provide any aid to a government that imposes Sharia against the will of its people.

Zalmay Khalilzad, Biden’s negotiator, is trying to tell the Taliban that they won’t be recognized by anyone if they take power this way. Well, we’ll see. In the nineteen-nineties, there were only three governments in the world that recognized the Taliban: Pakistan, Saudi Arabia, and the United Arab Emirates. And this time around, too, Pakistan will be one of them, I expect. But things are different. The Saudis and the Emiratis have a new geopolitical outlook. But China is not the same country that it was in the nineties. How will China support Pakistan in trying to manage a second Taliban regime, especially one that may attract sanctions or other kinds of pressure from the United States and its allies? It isn’t the nineties, but Pakistan is still in the same awkward place that it was last time around. And to the extent that the Taliban return to a kind of internationalism of their interpretation of Islam and welcome Al Qaeda types or other forms of radicals, allow the Islamic State to incubate on Afghan soil, or don’t have the interest or the capacity to do something about it, you can be sure that, as it did the last time, all of that will blow back on Pakistan in one way or another, be that in the form of international pressure or instability.

How do you see the next month playing out?

It is important for the international community to recognize that Afghanistan is entering another devastating humanitarian crisis of the kind it has seen too many times before, between refugee flows, insecurity in the areas that the Taliban have seized, uncertainty about how the Taliban will handle their enemies—will there be mass executions or internments? Then, you add to that the covid crisis and the humanitarian challenges in rural Afghanistan that were already in place before this summer, and you are looking at a really dark season for the Afghan population. I don’t expect the Biden Administration to change its policy, and even if it did I don’t expect that it could reverse the Taliban’s momentum without bombing Afghanistan to smithereens. But it can certainly take responsibility for the lion’s share of the response to this unfolding humanitarian crisis, which after all does involve a little bit of self-interest, because, if Afghanistan generates another massive refugee flow toward Europe, that will have political consequences that will wash right through the West.