This piece was published as part of The Future of Afghanistan and U.S. Foreign Policy, a collaboration between the Chicago Project on Security and Terrorism and

The range of achievable outcomes in Afghanistan is narrowing as Western effort wanes. The ambitious goals of the Bush administration were probably never attainable and are certainly not now. But even minimally democratic accountability may soon be beyond reach. If so, some form of delimited warlord rule will be the outer bound of the achievable. If a new set of bargains between Kabul and provincial powerbrokers can be reached and enforced, such a system could still be tolerable in the limited sense that it could preserve the United States' essential security interests in Afghanistan. But it would be far from ideal. And even this option could slip away if some critical reforms are not instituted soon.

Many Americans see Afghanistan as hopeless and ungovernable -- a chronically violent "graveyard of empires." It is not. For most of the twentieth century, Afghanistan was internally stable and at peace with its neighbors -- in fact, it was a tourist destination for backpacking Westerners in the 1960s. And the Taliban of today are hardly the invincible warriors or authentic vox populi some Westerners assume. A series of coalition offensives since 2009 has driven the Taliban from most of their southern strongholds in Helmand and Kandahar. Taliban counterattacks this summer failed to retake any of the districts they lost. And recent polls show declining Afghan public support for an already unpopular insurgency, as the Taliban have responded to military setbacks by striking civilian targets instead. Afghans know what Taliban government looks like, and in multiple polls over years of war, they have consistently rejected it. While the current government's corruption is unpopular, too, the coalition enjoys the great advantage of an enemy whose ideology is unwelcome. Of course, retaking Taliban-controlled areas is time consuming and costly; the Taliban remain a significant force in the east, and their assassination campaign continues. But in areas where coalition forces have deployed enough strength to protect Afghans from Taliban violence and stayed long enough to build public confidence, it is now clear that government control can be restored and maintained.

So Afghanistan is not hopeless. But neither will it reach nirvana anytime soon. In practical terms, there are now five plausible midterm futures for Afghanistan.

One involves a still democratic but weaker and less centralized Afghan government. The original blueprint for a post-2001 Afghanistan envisioned a strong, modern, centralized, bureaucratic state built around a powerful presidency. Such a system would have had important advantages, including limiting the danger of warlordism and renewed civil war, facilitating decisive action against terrorists, and empowering a modernizing center over a more conservative rural periphery with less interest in Western human rights agendas.

But that system was a poor fit for Afghanistan, where legitimacy is mostly local and personal, rather than national and institutional, and illiberal values remain influential. The mismatch promoted official corruption, weakened popular support for the state, and enabled a still unpopular Taliban to make headway.

The most attractive practical alternative available today would be a weaker, less centralized state with power shared across a wider range of stakeholders and with a larger role for local and tribal authorities. This decentralized system would probably be less decisive, less technocratic, more open to conservative rural influences, and probably more beholden to the country's neighbors. But it would also offer a closer match to the real distribution of political power in Afghanistan, so could be realized with less heroic exertions. And it would retain a fundamentally democratic system of government.

A less attractive possibility would be a modified version of warlord rule. In this scenario, real authority would reside chiefly with subnational powerbrokers and their associated patronage networks in a form of predatory rentier governance. As they increasingly do today, these networks would exert control by extracting cash from the governed and using it to buy off courts, police, officials, and prominent businesses. If allowed to expand unchecked, this system is unsustainable. It would eventually corrupt the army and build broader tolerance for an insurgency that, although unpopular, is nevertheless seen as honest.

The only stable version of a warlord system would thus be one with an enforceable limit on the take. This would require a renegotiation of the tacit deals between Kabul and the periphery that now allow these networks to function. Today, such arrangements generally trade the warlords' political support of Kabul for the capital's acquiescence in their depredations. The renegotiated ones would change the terms to exclude the most damaging forms of corruption while tolerating the rest; Kabul would look the other way for most graft but draw the line at intolerably severe predation.
Limits could be set in different ways. One strategy would be to prohibit land taking, which is perhaps the most damaging official predation in Afghanistan today. In an agrarian society, loss of one's land threatens destitution; when corrupt officials confiscate private land for "economic development" that mostly lines officials' pockets, they victimize groups that have little choice but to seek succor from the Taliban. By dissuading provincial powerbrokers from land grabs, the government's long-term viability would significantly improve, even without eliminating corruption altogether or reforming government service delivery in any comprehensive way.

For this to work, the West would need to provide Kabul with carrots (through aid that Kabul could redistribute as it wants) and sticks (assisting the Afghan military in enforcing the deals' terms). Western leverage is constrained by its willingness to pay, so the ask must be limited accordingly. If it includes positive incentives for compliance directed against a modest agenda focused on containing but not eliminating predation, then the levers available could suffice.

There are other possibilities. Afghanistan could split up -- most likely with the largely Tajik, Uzbek, and Hazara north and west separating from the predominantly Pashtun south and east. The country could descend into anarchy, with a collapsed central government giving way to atomized civil warfare. Today's insurgency could simply continue, although an unreformed Afghan government's ability to contain the Taliban after a Western withdrawal is doubtful. And in theory at least, Afghanistan could become a dictatorship, whether under the Taliban or some other force. A strongman is unlikely to succeed, however, in consolidating absolute power in a country as divided as Afghanistan; an attempt to do so would likelier devolve into civil war.

Many outcomes are thus possible; only some are acceptable. The acid test is the twofold stake Washington has long articulated: that Afghanistan does not become a base for terrorism against the West, or a haven for destabilizing its neighbors. The United States has many other worthy goals there, from prosperity to human rights and respect for the will of the governed, but the stakes for which the United States normally wages war are narrower.

By this standard, decentralized democracy is clearly acceptable. A weaker, less centralized government might be less able to police its territory than the original post-2001 model, and its local politics would periodically offend many in the West. But such a state would constitute a far smaller threat than the old Taliban regime, and its representative nature would lend it presumptive legitimacy.
Delimited warlord rule is far less appealing. But it, too, would probably pass the acid test of minimal acceptability -- if it were truly delimited and those limits included a prohibition on the use of powerbrokers' territory for cross-border violence.
The other possible outcomes fail to pass the test. Partition would tacitly concede the south and east to the Taliban, enabling terrorism in the process. Anarchy would allow militants to operate across Afghanistan's borders. An attempted dictatorship that merely fueled civil warfare would be no better, whoever the putative dictator. All would violate both of the security interests for which the United States now fights.

If the United States is serious about decentralized democracy, it will have to push major reforms to dismantle antidemocratic patronage networks, create inclusive checks and balances, and ensure honest elections. Doing all this would be time consuming, costly, and difficult, and will pose important trade-offs with short-term security progress. To date, security has mostly taken priority, and many crucial reforms have been postponed; the looming 2014 deadline may have already put this option beyond reach. If it has not, there is no time left to lose.

If the West is unwilling to make these sacrifices, the best achievable outcome will be delimited warlord rule. This is a defensible choice if U.S. citizens are unwilling to do more; it is, after all, minimally sufficient to meet U.S. security aims. But even such a course will require systematic efforts to harness the West's remaining resources to develop a program of conditional pressure to reshape the deals that underlie today's powerbroker governance. And little time is left. After 2014, remaining Western leverage might suffice to maintain a modified equilibrium between Kabul and provincial bosses, but it will be far harder to create it at that point.

The worst of all results, however, would be another two years of combat without the governance reforms needed to create a sustainably acceptable outcome. Reasonable people can differ on whether a tolerable result is worth the sacrifice in Afghanistan. But no one can justify continued sacrifice for an unsustainable result.

Read more from The Future of Afghanistan and U.S. Foreign Policy.

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
  • STEPHEN BIDDLE is Roger Hertog Senior Fellow for Defense Policy at the Council on Foreign Relations.
  • More By Stephen Biddle