I did not predict Afghanistan’s rapid fall to the Taliban. I had hopes that money without troops might give a fighting chance but sadly pessimistic takes were borne out. I knew many of the factors that drove it, but underestimated their magnitude. The challenges of corruption, dysfunction, and a lack of legitimacy raised in the Afghanistan papers were ones that were well known in the field. That said, stealing the salaries for non-existent “ghost soldiers” is a large problem but systematically failing to equip or provision actual forces is a precursor to collapse.
This failure is not the same as the Afghan National Army being unwilling to fight; as the Costs of War project reports, they’ve suffered 66,000 lost. However, it is a stark demonstration of no confidence in the fallen Afghan government and its ability to operate without U.S. military support. Kori Schake writes an apologia for this failure after $83 billion invested in training, and she concludes “We shouldn’t be surprised that many think the situation is hopeless after our abandonment and are surrendering. We should be amazed and respectful that any have volunteered to fight.” Even though I disagree with her on conditionality, I’d concur that we should not be cavalier regarding others’ willingness to face grave danger collectively for the chance at a better life. But the legitimate complaints of the Afghan Army underline our failure - after 20 years and 2 trillion spent - to have built a state with a political chance of sustaining itself or a military trained and structured to operate independently. My colleagues Anthony Cordesman and Grace Hwang put it plainly:
There are substantial official sources that show that Afghanistan was not winning against the Taliban before the [Trump] peace agreements, even with massive U.S. combat air and intelligence support as well as with the extensive support from allied forces and cadres of U.S. special forces, elite units, train and assist forces, and intelligence operators.
President Biden is correct the status quo was not sustainable. There had been a lull in U.S. casualties due to the Trump deal with the Taliban that promised withdrawal. While the Taliban held back as part of the deal, casualties had been rising in the years beforehand (though far below surge levels). Civilian casualties have been above 10,000 per year, with over 3,000 killed, for the last six years. A reliance on air strikes can result in a grim trade off between those two figures. Colleagues have suggested that approaches liked armed overwatch may have been able to enforce a stalemate while minimizing risks to U.S. troops, but I fear the humanitarian costs of such such an approach based on its antecedents in Somalia, Syria, Libya, and Yemen.
I do not think the war was doomed from the start, although there are many reasons it was inherently challenging. More important is that many of the worst and most lasting mistakes were made early on. Steve Saideman briefly outlines some of the bigger ones:
Who is responsible? Everyone. The US made big mistakes at the outset---relying on warlords, having too small of a footprint, sponsoring a constitution that was a very bad fit, distracted by Iraq--and other mistakes along the way--cycling generals and strategies, for example. Obama made mistakes, Trump didn't help. Biden's team has handled this endgame poorly. The allies could have done better (see our book for some reasons why they didn't). Pakistan did so much to undermine the effort, and Iran and eventually Russia did some damage. The Afghans were served poorly by their own politicians.
I focus most on the constitution because it is foundational to the challenges of building a state that can develop an independent legitimacy. While I disagree with some parts of his take, Shadi Hamid does a good job of elaborating on why the Afghan constitution, a centralized presidential system, was such a terrible fit: it alienated local and regional actors, it failed to support the rise of political parties or any checks on the President, and it raised the stakes on competition in a society riven with divides. In this early period the U.S. did have popular support within Afghanistan and far more diplomatic opportunities, but the political situation degraded over time especially as attention was distracted by the war in Iraq. No constitution will guarantee that power brokers like Abdullah Abdullah would not question the validity of the elections or force brokered deals such as the one with Ashraf Ghani after the 2014 election. However, there would have been a chance to channel more politics into coalition building and also allow for greater political variation between provinces if the governors were not all centrally appointed. The flaws in the constitution are a core reason I reject blaming America for an absence of strategic patience. When the fundamentals are flawed and many indicators are trending in the wrong direction, the limitations of perseverance are shown. We may be able to maintain a stalemate at a higher cost, but as Saiderman notes in his piece, it’s easier to break than to build.
I think a related mistake is that our often incoherent strategy, the oft lamented fighting twenty one year wars rather than one twenty year war, was frequently shaped by different U.S. factions compromising over means, enabled by in more recent years keeping costs in blood and troop counts low. I think our internal divisions contribute to our failure to apply incentives that both Mara Karlin and Rachel Tecott argue are key to successfully building a partner force. The limitations of security assistance is a longstanding topic of concern . I think we often fail to apply what my then-colleague Melissa Dalton called smart conditions. This is partially because conditions are hard for reasons Kori Schake outlined in her above piece, but also because I think our partners’ assessments of donor vulnerability are often based on the faction in the United States that considers recipient the most important. Condition based withdrawal might send the right message to adversaries but can also undermine our local partners’ incentives to address fundamental problems. Moreover, as Christine Fair notes in her harsh critiques suggesting betrayal by the U.S. and Pakistan, the actions of U.S. partners are often highly incompatible.
There are some larger issues I’m still grappling with, such as Jacqueline Hazelton’s argument that counter-insurgency involves truly ugly choices that are bad enough that the U.S. should largely stay out of them. I think we can do better on conditions, but this argument does at least help explain my surprise at the relative ruthless and criminal but ultimately more winning Syrian government counterinsurgency approach, which is in no way a model to emulate. However, I find myself more convinced by those that argue that we should have tried harder on corruption, from what Sarah Chayes argues in Foreign Affairs there was much we chose not try. I know at least one contractor facility we visited seemed disproportionately ornate for its minimal level of activity.
I also wonder about our approach to women in Afghanistan in general. In a Smart Women Smart Power podcasts, Lyla Kohinstany argues that we undercut efforts to give Afghan women a chance to participate in building security. On the peacebuilding side, evidence shows that women’s participation results in agreements that are more likely to hold. So I am skeptical about Shadi Hamid’s point on culture. I think backlash is a real phenomenon but simply trying to avoid it ties our own hands. Instead, I think the challenge is weighing backlash against both the breadth and depth of support in among Afghan women themselves. That said, in the present circumstances as shockingly courageous women protest against the Taliban and are suppressed, offering asylum must be a priority.
As for my own mistakes, three stand out:
- First, I failed to face the extent to which the U.S. had lost popular support. By 2009 the favorability in polling towards the U.S. had switched to slightly negative. I was able for some years to find related questions in the Asia Foundation polls but in the ones I’ve looked at in the past month they did not ask directly about some of the core questions of views of Coalition forces that I’d like to know. I don’t think the evidence supports the claim that “the vast majority of Afghans have always viewed the Taliban as the lesser of two evils.” However, I feel I underestimate how much support continued.
- A core finding from my 2011 trip to Afghanistan was that many of the Afghan vendors were happy, even eager to work with the United States or coalition forces. There stories were often inspiring, though of course that was a self-selected group. Those interviews supported the idea that using local vendors helped build the economy and built up the country in other ways, such as the entrepreneur that funneled her profits into a school. However, one of the hopes of the host nation first program was the eventual transition to a more independent state, and here the news was far less optimistic. They largely did not trust the Afghan government and, in at least one case, cited direct experience with corruption. I did not then and do not now know how to overcome that problem, but if I had paid more attention to it, I would have been more prepared for the difficulties of sustaining the Afghan government via financing.
- Building partner militaries is a naturally appealing middle ground in theory, but even as Obama embraced it there were many warning about its limitations. Robert Farley’s summary piece concludes that we have had some success with special forces, but that in general the U.S. is good at building relationships but bad at building independent forces. I believe the U.S. can do more with conditions, especially because I think our strategic interests, especially in the Middle East, are often overstated and thus our donor vulnerability less of a problem than conventional wisdom allows. However, when faced with repeated failure, analysts should be humble about their hopes for better results from better implementation not backed by structural changes. Tobias Switzer reaches a similar conclusion on humility for building air forces in particular, though as he notes our provision of ill suited equipment was a clear unforced error.
i think the biggest consequence of my own mistake was that my own attention stayed on vague hopes of a middle path rather than being one small voice trying to build up the Plan B. The number of Afghans withdrawn in August was a remarkable achievement and, as Gordon Adams notes, the U.S. was ill positioned, in part due to the timing of the Trump withdrawal agreement and an utter lack of prioritization of Afghan allies in planning or transition, to accomplish an evacuation sufficient to the need. The possibility of a longer delay was undercut by mistrust within the United States, but also would have meant both making a then uncertain Afghan collapse seem more likely and accepting even greater risks of U.S. casualties than occurred in August. It may have been worth doing anyway. Moreover, while I fear Charli Carpenter’s U.N. peacekeeping force was a longshot, I think it may have been one of our best shots for an Afghanistan not dominated by either the Taliban or years or decades more of civil conflict, especially if considered as part of an Obama administration endgame when the U.S. had more leverage.
My one hope is that for now, American popular support for Afghan refugees is holding. Future departures will depend on diplomacy than military force, but for those of us in the foreign policy community that feel we have both failed and that honor and humanitarian drives hold further obligations, there is work yet to do.
Images: A few selections from my trip. I would like to center the people of Afghanistan more but I fear showing those we interviewed or their hardworking employees would only bring them danger.
Recent Comments