A U.S. Army crew chief in a CH-47F Chinook helicopter during a training flight in Afghanistan, March 2018 
U.S. Air Force Tech. Sgt. Gregory Brook / via Reuters

For nearly 20 years, the U.S. intervention in Afghanistan has been sustained by a single, vital national interest: the clear and present danger of another September 11–like attack emerging from this region of the world, absent constant efforts to thwart it. To this end, U.S. strategy has been threefold: deploying American and allied forces to Afghanistan to conduct sensitive counterterrorism missions there and in neighboring parts of Pakistan; training and enabling Afghan partner forces to assume the bulk of responsibility for security inside their country; and backing a friendly government in Kabul that has permitted international forces to operate from its territory against Islamist extremism.

This strategy has been costly and unsatisfying—but also reasonably successful. It enabled the United States to eliminate the al Qaeda camps that flourished in Afghanistan under the Taliban prior to its ouster from power in late 2001, and equally important, it has kept that extremist infrastructure from being reestablished. When terrorists attempted to rebuild their networks in the nearby tribal areas of northwest Pakistan in the mid-2000s, the United States was able to smash them there, too, from its Afghan bases. And it was out of Afghanistan that the operation against Osama bin Laden in his Abbottabad hideout was launched in May 2011. More recently, the U.S. footprint in Afghanistan again proved its value when an Islamic State (ISIS) affiliate emerged on the Afghanistan-Pakistan frontier and attempted to raise the black flag of the caliphate there. It likewise has been pummeled.

Rather than a safe haven for extremists to plot devastating strikes on the United States and its allies, in other words, Afghanistan over the last two decades became an outpost from which the United States and its allies could project power against the terrorists.

The U.S.-Taliban agreement, signed on February 29, now proposes to jettison this approach. It is, in essence, a wager that the United States can still achieve the same strategic ends in Afghanistan, but by radically different means. Rather than sustaining an American presence in coalition with like-minded Afghans, the deal seeks to make the Taliban itself into a principal guarantor of U.S. counterterrorism interests. In doing so, the agreement holds out the tantalizing prospect of transforming Afghanistan from a problem that will require the perpetual military management of the United States into one that can be solved politically, once and for all. But the risks presented by this gamble are huge, and the signs from the deal’s early aftermath—continued Taliban attacks and an Afghan government in disarray—are not encouraging. Most worrying, there is a fundamental asymmetry at the heart of the agreement, such that the more Washington implements its obligations under the deal, the less constrained the Taliban will be to keep its own.


Analysis of the U.S.-Taliban deal so far has focused mostly on the immediate obstacles to its implementation. Already, key aspects of the accord have stalled—prompting Secretary of State Mike Pompeo to embark on a diplomatic rescue mission last week to restore momentum to the peace process, flying first to Kabul and then to Doha, Qatar, where the Taliban leadership is headquartered.

Among the challenges that have arisen, the agreement decreed March 10 as the date by which an “intra-Afghan dialogue” was supposed to kick off—meant to bring together for the first time the Taliban, the internationally recognized Afghan government, and other representatives of Afghan society in direct negotiations to end the conflict. That deadline has since come and gone, with the Taliban and the Afghan leadership in Kabul locked in a dispute over the terms of a prisoner release that the U.S.-Taliban agreement established should precede the intra-Afghan meeting but that was never agreed to by the Afghan government itself—having been excluded from the U.S.-Taliban talks at the insurgents’ insistence. Levels of violence, meanwhile, have begun to creep up across Afghanistan, following the expiration of a partial cease-fire that provided a brief but hopeful respite from fighting in the week before the U.S.-Taliban deal was unveiled.

A bitter power struggle is playing out inside the Afghan state, threatening its collapse from within.

Even more vexing for Washington—and the principal focus of Pompeo’s trip—is the bitter power struggle playing out inside the Afghan state, threatening its collapse from within. Incumbent President Ashraf Ghani and his top opponent in last September’s presidential election, Abdullah Abdullah, have both declared themselves the winner of that contest and, after holding dueling inauguration ceremonies earlier this month, are now moving to form parallel administrations. Despite Pompeo’s personal efforts to broker a compromise between the two rivals, the top U.S. diplomat left the region empty-handed and is now threatening coercive measures to force a resolution, including the withholding of $1 billion in aid to Afghanistan. Although the feud is ultimately likely to prove surmountable, it is a preview of just how challenging it will be to achieve any kind of reconciliation between the Taliban and its opponents, when the latter can barely reconcile among themselves.

For the United States, however, the strategic implications of the Taliban agreement extend far beyond the internal Afghan intrigues it must overcome.

For all its complexities, the U.S.-Taliban agreement is at its core a simple tradeoff. On the one hand, the United States has promised that all foreign forces, including its own, will withdraw from Afghanistan within 14 months. As an initial down payment toward that commitment, approximately one-third of the coalition troops presently deployed there will depart over the next 135 days—a drawdown that has already begun. The Taliban, on the other hand, has sworn to prevent “any group or individual” from using Afghan territory to threaten the security of the United States and its allies. In exchange for the U.S. military and its international partners leaving Afghanistan, in other words, the Taliban will guard against the conditions that compelled those foreign forces to deploy there in the first place.

This structure presents an obvious asymmetry. Determining if thousands of uniformed military personnel have exited Afghanistan is relatively straightforward. Assessing the Taliban’s compliance with its counterterrorism commitments, by contrast, entails much greater nuance and complexity.

To begin with, do the United States and the Taliban have a common definition of exactly what “groups and individuals” constitute a threat to the security of the United States and its allies? The text makes a fleeting reference to al Qaeda but otherwise does not explicitly name the entities that the Taliban is now obligated to keep off Afghan soil. Nor does it lay out a process by which the two parties are to reach consensus about the identity of such bad actors.

Does the restriction, for instance, extend to other U.S.-designated foreign terrorist organizations? What about UN-sanctioned extremists? Given the multiplicity of jihadist factions operating in the Afghanistan-Pakistan theater, as well as the Taliban’s long-standing ties to most of them, glossing over these details now would seem like an invitation to crisis later.

There is ample evidence the Taliban interprets “threats” to the United States in ways that diverge from Washington’s own security perceptions.

Compounding this danger, there is ample evidence the Taliban interprets “threats” to the United States in ways that diverge from Washington’s own security perceptions. In an interview last autumn, for example, the Taliban spokesman refused to acknowledge that al Qaeda was, in fact, the perpetrator of the September 11 attacks. More recently, deputy Taliban leader Sirajuddin Haqqani—whose eponymous branch of the insurgency is especially notorious for close cooperation with transnational jihadists—argued in a New York Times opinion piece that fears of Afghanistan once again becoming a safe haven for precisely such groups were “inflated” and the “politically motivated exaggerations by the warmongering players.”


There are also elementary reasons to doubt the Taliban’s good faith in signing the accord. As recently as January, UN investigators assessed that al Qaeda’s relations with the Taliban continued to be “close and mutually beneficial,” and there has been no reporting to suggest this has changed. Even in announcing the deal with the United States, the Taliban notably refrained from renouncing al Qaeda, much less committing to cooperate with Washington against the international terror network. Reportedly, the insurgent leadership has no intention of actually sticking to its counterterrorism promises.

How, then, will the United States know if the Taliban is cheating? Since insurgent interactions with al Qaeda are almost certain to be clandestine, at least at first, detecting violations of the deal will be a test of U.S. intelligence gathering. Yet it is precisely these capabilities that are likely to degrade as the United States leaves Afghanistan. In this respect, the more completely Washington fulfills its part of the Taliban agreement, the less certain it may become as to whether the insurgents are upholding theirs.

The more completely Washington fulfills its part of the Taliban agreement, the less certain it may become as to whether the insurgents are upholding theirs.

Then there is the matter of how violations of the deal, when they inevitably arise, are to be dealt with. Here, too, the text of the agreement is silent. Will Washington notify the Taliban leadership that one of its members has been detected aiding or otherwise working with a banned individual or group? To share such intelligence with any degree of precision could risk compromising the sources and methods that were used to collect it. Calling out the insurgents for violations of the agreement might thus end up instructing them on how to subvert it more effectively. Yet absent some sort of reporting mechanism, the Taliban can reasonably protest that it has no way of answering the allegation, much less trying to correct it.

When indications of Taliban misconduct do occur, furthermore, they are unlikely to be clear-cut. On the contrary, intelligence in such cases is invariably ambiguous and susceptible to competing interpretations. How will the administration of U.S. President Donald Trump react, for instance, if it receives reports of ongoing Taliban collusion with al Qaeda on Afghan soil—but only with low or moderate confidence?

Getting such assessments right is also invariably a time-intensive process, making the rapid timetable for withdrawal codified by the agreement difficult to square with the methodical analysis required to justify it. For a preview of how vexing and drawn-out these determinations can prove, consider the years of wrenching debate within the U.S. government over the extent and nature of dealings between the Pakistani intelligence service and the Taliban. It seems a reasonable bet that conclusions about the nature of the Taliban’s relationship with al Qaeda in the wake of its agreement will likewise resist swift or definitive resolution.

The United States’ tortured experience with Islamabad suggests another cautionary lesson relevant to the Taliban deal: namely, the tendency of U.S. decision-makers, once invested in a relationship or policy, to downplay or disregard information that threatens to upend it. In this respect, it is all too easy to imagine reports of Taliban breaches of the peace accord being written off as the work of rogue factions—an interpretation the Taliban’s own leadership will presumably encourage.


In the final calculus, the most important incentive to compel Taliban compliance with the agreement is likely to be the cost its leadership concludes Washington is prepared to impose if the group is caught double-dealing. The United States’ strongest leverage in this respect is the threat that it will suspend or reverse the drawdown of troops and go back on the battlefield offensive.

Yet the Taliban may calculate the Trump administration is unlikely to take such a draconian and politically contentious step, especially in response to minor or ambiguous violations of the deal. Moreover, the closer the United States gets to having no boots on the ground, the less plausible an American return in force becomes—and therefore the less reason the Taliban will have to play by the rules.

The closer the United States gets to having no boots on the ground, the less plausible an American return in force becomes.

This in turn points to a final asymmetry in the agreement. After the United States has fulfilled its part of the accord and completely exited Afghanistan, the cost for Washington to reverse itself—which is to say, to reinvade the country—will be extraordinarily high. Consequently, there is no “snapback” that will restore the military status quo ante in response to Taliban malfeasance once the United States’ armed forces are gone.

Moreover, in Yemen, Libya, or Somalia, the United States can launch punishing drone strikes and special operations raids from its bases in neighboring countries or from littoral waters, but landlocked Afghanistan presents a unique set of logistical challenges for Pentagon planners. Simply put, it is much harder to sustain an effective “offshore” counterterrorism strategy in a country that doesn’t have a coastline. Most of Afghanistan’s immediate neighbors are also implausible hosts for U.S. forces. Even if a willing patron were found in the region, the United States will have succeeded only in swapping its long-term military presence from one Central Asian country to another, likely less favorable one.

The Taliban, on the other hand, will enjoy considerably more leeway on its side of the deal. To begin with, the agreement’s principal counterterrorism prohibitions apply to Afghan soil alone; it contains no evident bar on Taliban interactions with extremists in Pakistan, where most such collaboration in recent years has taken place.

More fundamentally, there is nothing in the agreement that compels the Taliban to make an irrevocable break with transnational terrorists. Rather, it is easy to imagine how the insurgents could calibrate their relationship with al Qaeda and other radicals, paring it back over the months ahead as the United States draws down, only to reverse themselves after the American exit is complete. In this respect, there is an obvious danger the framework could provide the Taliban with permanent relief from U.S. military pressure in exchange for what proves to be only temporary restraint in its behavior.

Should the Taliban revert to its old ways, Washington may also find that it doesn’t have much of an Afghan counterterrorism partner to fall back upon. Under the terms of the U.S. agreement with the insurgents, all foreign military troops, including trainers and advisers for the Afghan army, are required to leave the country in 14 months—regardless of whether the intra-Afghan talks have succeeded or stalemated.

Consequently, the Taliban’s deal with Washington would seem to give the group little incentive to bargain seriously with the internationally recognized government in Kabul, since its opponent’s position will grow progressively weaker as the deadline for international withdrawal approaches. Rather than setting the stage for a difficult intra-Afghan compromise, then, the deal implicitly appears to anticipate the endgame the insurgents themselves have consistently articulated since 2001: a Taliban reconquest of the country.


The Trump administration has indicated that the verification and enforcement of the Taliban agreement are governed by a pair of annexes that it negotiated in parallel with the deal itself. These protocols have been classified, so their contents are, somewhat incongruously, known to the Taliban but unavailable for the American public to review. Members of Congress do have the authority to access these documents, however, and must now undertake rigorous oversight of both their substance and the wider strategic concept into which they fit. In particular, what mechanisms do they put in place and how will they function in practice? Do they provide sufficient safeguards for the United States’ indispensable national security requirements in Afghanistan and regionally? 

In addition to scrutinizing the internal workings of the Taliban deal, Congress should consider legislative action in response to it. For instance, it could mandate a National Intelligence Estimate on whether the Taliban has in fact ceased cooperation with al Qaeda, as required by its agreement with Washington, as well as with any other group that the U.S. intelligence community characterizes as a threat to the security of the United States or its allies. Congress could also order quarterly reporting by the intelligence community that assesses the impact of proposed force reductions on its capacity to determine the Taliban’s compliance with the deal.

For some, such technical nuances are beside the point. After almost two decades of entanglement in Afghanistan, they argue, a deal with the Taliban is worth the risk as long as it provides a plausible path for U.S. troops to come home. The alternative policy—endless war for Americans and Afghans alike—is neither militarily nor fiscally sustainable.

That is an understandable instinct. Yet Americans should not delude themselves into thinking that Afghanistan will somehow stabilize itself or vanish from the world stage absent U.S. involvement there or that the United States can disassociate itself from the chaos that its retreat is likely to unleash—with consequences for international security that are as catastrophic as they are foreseeable. More broadly, history suggests that capitulation in the name of peace rarely succeeds in either curbing an adversary’s ambitions or moderating its behavior—at least not for long.

The United States first sent its military to Afghanistan to end the terrorist sanctuary that developed there under the Taliban and culminated in cataclysm. Before it withdraws, Washington must make sure appropriately robust and reliable arrangements are in place to prevent this danger from reemerging. It is, at present, far from clear that the Trump administration’s deal with the Taliban meets that standard.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • DAVID PETRAEUS served as Director of the CIA, Commander of Coalition Forces in Iraq and Afghanistan, and Commander of the U.S. Central Command. He is now Chairman of the KKR Global Institute and a Partner at KKR.
  • VANCE SERCHUK is Executive Director of the KKR Global Institute and an Adjunct Senior Fellow at the Center for a New American Security. He previously served as Foreign Policy Adviser to Senator Joseph Lieberman (I-Conn.).
  • More By David Petraeus
  • More By Vance Serchuk